Merge git://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf
Daniel Borkmann says:
====================
pull-request: bpf 2021-09-14
The following pull-request contains BPF updates for your *net* tree.
We've added 7 non-merge commits during the last 13 day(s) which contain
a total of 18 files changed, 334 insertions(+), 193 deletions(-).
The main changes are:
1) Fix mmap_lock lockdep splat in BPF stack map's build_id lookup, from Yonghong Song.
2) Fix BPF cgroup v2 program bypass upon net_cls/prio activation, from Daniel Borkmann.
3) Fix kvcalloc() BTF line info splat on oversized allocation attempts, from Bixuan Cui.
4) Fix BPF selftest build of task_pt_regs test for arm64/s390, from Jean-Philippe Brucker.
5) Fix BPF's disasm.{c,h} to dual-license so that it is aligned with bpftool given the former
is a build dependency for the latter, from Daniel Borkmann with ACKs from contributors.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
diff --git a/include/linux/cgroup-defs.h b/include/linux/cgroup-defs.h
index e1c705f..db2e147 100644
--- a/include/linux/cgroup-defs.h
+++ b/include/linux/cgroup-defs.h
@@ -752,107 +752,54 @@
* sock_cgroup_data is embedded at sock->sk_cgrp_data and contains
* per-socket cgroup information except for memcg association.
*
- * On legacy hierarchies, net_prio and net_cls controllers directly set
- * attributes on each sock which can then be tested by the network layer.
- * On the default hierarchy, each sock is associated with the cgroup it was
- * created in and the networking layer can match the cgroup directly.
- *
- * To avoid carrying all three cgroup related fields separately in sock,
- * sock_cgroup_data overloads (prioidx, classid) and the cgroup pointer.
- * On boot, sock_cgroup_data records the cgroup that the sock was created
- * in so that cgroup2 matches can be made; however, once either net_prio or
- * net_cls starts being used, the area is overridden to carry prioidx and/or
- * classid. The two modes are distinguished by whether the lowest bit is
- * set. Clear bit indicates cgroup pointer while set bit prioidx and
- * classid.
- *
- * While userland may start using net_prio or net_cls at any time, once
- * either is used, cgroup2 matching no longer works. There is no reason to
- * mix the two and this is in line with how legacy and v2 compatibility is
- * handled. On mode switch, cgroup references which are already being
- * pointed to by socks may be leaked. While this can be remedied by adding
- * synchronization around sock_cgroup_data, given that the number of leaked
- * cgroups is bound and highly unlikely to be high, this seems to be the
- * better trade-off.
+ * On legacy hierarchies, net_prio and net_cls controllers directly
+ * set attributes on each sock which can then be tested by the network
+ * layer. On the default hierarchy, each sock is associated with the
+ * cgroup it was created in and the networking layer can match the
+ * cgroup directly.
*/
struct sock_cgroup_data {
- union {
-#ifdef __LITTLE_ENDIAN
- struct {
- u8 is_data : 1;
- u8 no_refcnt : 1;
- u8 unused : 6;
- u8 padding;
- u16 prioidx;
- u32 classid;
- } __packed;
-#else
- struct {
- u32 classid;
- u16 prioidx;
- u8 padding;
- u8 unused : 6;
- u8 no_refcnt : 1;
- u8 is_data : 1;
- } __packed;
+ struct cgroup *cgroup; /* v2 */
+#ifdef CONFIG_CGROUP_NET_CLASSID
+ u32 classid; /* v1 */
#endif
- u64 val;
- };
+#ifdef CONFIG_CGROUP_NET_PRIO
+ u16 prioidx; /* v1 */
+#endif
};
-/*
- * There's a theoretical window where the following accessors race with
- * updaters and return part of the previous pointer as the prioidx or
- * classid. Such races are short-lived and the result isn't critical.
- */
static inline u16 sock_cgroup_prioidx(const struct sock_cgroup_data *skcd)
{
- /* fallback to 1 which is always the ID of the root cgroup */
- return (skcd->is_data & 1) ? skcd->prioidx : 1;
+#ifdef CONFIG_CGROUP_NET_PRIO
+ return READ_ONCE(skcd->prioidx);
+#else
+ return 1;
+#endif
}
static inline u32 sock_cgroup_classid(const struct sock_cgroup_data *skcd)
{
- /* fallback to 0 which is the unconfigured default classid */
- return (skcd->is_data & 1) ? skcd->classid : 0;
+#ifdef CONFIG_CGROUP_NET_CLASSID
+ return READ_ONCE(skcd->classid);
+#else
+ return 0;
+#endif
}
-/*
- * If invoked concurrently, the updaters may clobber each other. The
- * caller is responsible for synchronization.
- */
static inline void sock_cgroup_set_prioidx(struct sock_cgroup_data *skcd,
u16 prioidx)
{
- struct sock_cgroup_data skcd_buf = {{ .val = READ_ONCE(skcd->val) }};
-
- if (sock_cgroup_prioidx(&skcd_buf) == prioidx)
- return;
-
- if (!(skcd_buf.is_data & 1)) {
- skcd_buf.val = 0;
- skcd_buf.is_data = 1;
- }
-
- skcd_buf.prioidx = prioidx;
- WRITE_ONCE(skcd->val, skcd_buf.val); /* see sock_cgroup_ptr() */
+#ifdef CONFIG_CGROUP_NET_PRIO
+ WRITE_ONCE(skcd->prioidx, prioidx);
+#endif
}
static inline void sock_cgroup_set_classid(struct sock_cgroup_data *skcd,
u32 classid)
{
- struct sock_cgroup_data skcd_buf = {{ .val = READ_ONCE(skcd->val) }};
-
- if (sock_cgroup_classid(&skcd_buf) == classid)
- return;
-
- if (!(skcd_buf.is_data & 1)) {
- skcd_buf.val = 0;
- skcd_buf.is_data = 1;
- }
-
- skcd_buf.classid = classid;
- WRITE_ONCE(skcd->val, skcd_buf.val); /* see sock_cgroup_ptr() */
+#ifdef CONFIG_CGROUP_NET_CLASSID
+ WRITE_ONCE(skcd->classid, classid);
+#endif
}
#else /* CONFIG_SOCK_CGROUP_DATA */
diff --git a/include/linux/cgroup.h b/include/linux/cgroup.h
index 7bf6045..75c1514 100644
--- a/include/linux/cgroup.h
+++ b/include/linux/cgroup.h
@@ -829,33 +829,13 @@
*/
#ifdef CONFIG_SOCK_CGROUP_DATA
-#if defined(CONFIG_CGROUP_NET_PRIO) || defined(CONFIG_CGROUP_NET_CLASSID)
-extern spinlock_t cgroup_sk_update_lock;
-#endif
-
-void cgroup_sk_alloc_disable(void);
void cgroup_sk_alloc(struct sock_cgroup_data *skcd);
void cgroup_sk_clone(struct sock_cgroup_data *skcd);
void cgroup_sk_free(struct sock_cgroup_data *skcd);
static inline struct cgroup *sock_cgroup_ptr(struct sock_cgroup_data *skcd)
{
-#if defined(CONFIG_CGROUP_NET_PRIO) || defined(CONFIG_CGROUP_NET_CLASSID)
- unsigned long v;
-
- /*
- * @skcd->val is 64bit but the following is safe on 32bit too as we
- * just need the lower ulong to be written and read atomically.
- */
- v = READ_ONCE(skcd->val);
-
- if (v & 3)
- return &cgrp_dfl_root.cgrp;
-
- return (struct cgroup *)(unsigned long)v ?: &cgrp_dfl_root.cgrp;
-#else
- return (struct cgroup *)(unsigned long)skcd->val;
-#endif
+ return skcd->cgroup;
}
#else /* CONFIG_CGROUP_DATA */
diff --git a/include/linux/mmap_lock.h b/include/linux/mmap_lock.h
index 0540f01..3af8f7f 100644
--- a/include/linux/mmap_lock.h
+++ b/include/linux/mmap_lock.h
@@ -144,15 +144,6 @@
__mmap_lock_trace_released(mm, false);
}
-static inline bool mmap_read_trylock_non_owner(struct mm_struct *mm)
-{
- if (mmap_read_trylock(mm)) {
- rwsem_release(&mm->mmap_lock.dep_map, _RET_IP_);
- return true;
- }
- return false;
-}
-
static inline void mmap_read_unlock_non_owner(struct mm_struct *mm)
{
up_read_non_owner(&mm->mmap_lock);
diff --git a/kernel/bpf/disasm.c b/kernel/bpf/disasm.c
index ca3cd9a..7b4afb7 100644
--- a/kernel/bpf/disasm.c
+++ b/kernel/bpf/disasm.c
@@ -1,4 +1,4 @@
-// SPDX-License-Identifier: GPL-2.0-only
+// SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
/* Copyright (c) 2011-2014 PLUMgrid, http://plumgrid.com
* Copyright (c) 2016 Facebook
*/
diff --git a/kernel/bpf/disasm.h b/kernel/bpf/disasm.h
index e546b18..a4b0407 100644
--- a/kernel/bpf/disasm.h
+++ b/kernel/bpf/disasm.h
@@ -1,4 +1,4 @@
-/* SPDX-License-Identifier: GPL-2.0-only */
+/* SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause) */
/* Copyright (c) 2011-2014 PLUMgrid, http://plumgrid.com
* Copyright (c) 2016 Facebook
*/
diff --git a/kernel/bpf/stackmap.c b/kernel/bpf/stackmap.c
index e8eefdf..09a3fd9 100644
--- a/kernel/bpf/stackmap.c
+++ b/kernel/bpf/stackmap.c
@@ -179,7 +179,7 @@
* with build_id.
*/
if (!user || !current || !current->mm || irq_work_busy ||
- !mmap_read_trylock_non_owner(current->mm)) {
+ !mmap_read_trylock(current->mm)) {
/* cannot access current->mm, fall back to ips */
for (i = 0; i < trace_nr; i++) {
id_offs[i].status = BPF_STACK_BUILD_ID_IP;
@@ -204,9 +204,15 @@
}
if (!work) {
- mmap_read_unlock_non_owner(current->mm);
+ mmap_read_unlock(current->mm);
} else {
work->mm = current->mm;
+
+ /* The lock will be released once we're out of interrupt
+ * context. Tell lockdep that we've released it now so
+ * it doesn't complain that we forgot to release it.
+ */
+ rwsem_release(¤t->mm->mmap_lock.dep_map, _RET_IP_);
irq_work_queue(&work->irq_work);
}
}
diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index 047ac4b..e76b559 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -9912,6 +9912,8 @@
nr_linfo = attr->line_info_cnt;
if (!nr_linfo)
return 0;
+ if (nr_linfo > INT_MAX / sizeof(struct bpf_line_info))
+ return -EINVAL;
rec_size = attr->line_info_rec_size;
if (rec_size < MIN_BPF_LINEINFO_SIZE ||
diff --git a/kernel/cgroup/cgroup.c b/kernel/cgroup/cgroup.c
index 881ce14..8afa869 100644
--- a/kernel/cgroup/cgroup.c
+++ b/kernel/cgroup/cgroup.c
@@ -6572,74 +6572,44 @@
*/
#ifdef CONFIG_SOCK_CGROUP_DATA
-#if defined(CONFIG_CGROUP_NET_PRIO) || defined(CONFIG_CGROUP_NET_CLASSID)
-
-DEFINE_SPINLOCK(cgroup_sk_update_lock);
-static bool cgroup_sk_alloc_disabled __read_mostly;
-
-void cgroup_sk_alloc_disable(void)
-{
- if (cgroup_sk_alloc_disabled)
- return;
- pr_info("cgroup: disabling cgroup2 socket matching due to net_prio or net_cls activation\n");
- cgroup_sk_alloc_disabled = true;
-}
-
-#else
-
-#define cgroup_sk_alloc_disabled false
-
-#endif
-
void cgroup_sk_alloc(struct sock_cgroup_data *skcd)
{
- if (cgroup_sk_alloc_disabled) {
- skcd->no_refcnt = 1;
- return;
- }
-
/* Don't associate the sock with unrelated interrupted task's cgroup. */
if (in_interrupt())
return;
rcu_read_lock();
-
while (true) {
struct css_set *cset;
cset = task_css_set(current);
if (likely(cgroup_tryget(cset->dfl_cgrp))) {
- skcd->val = (unsigned long)cset->dfl_cgrp;
+ skcd->cgroup = cset->dfl_cgrp;
cgroup_bpf_get(cset->dfl_cgrp);
break;
}
cpu_relax();
}
-
rcu_read_unlock();
}
void cgroup_sk_clone(struct sock_cgroup_data *skcd)
{
- if (skcd->val) {
- if (skcd->no_refcnt)
- return;
- /*
- * We might be cloning a socket which is left in an empty
- * cgroup and the cgroup might have already been rmdir'd.
- * Don't use cgroup_get_live().
- */
- cgroup_get(sock_cgroup_ptr(skcd));
- cgroup_bpf_get(sock_cgroup_ptr(skcd));
- }
+ struct cgroup *cgrp = sock_cgroup_ptr(skcd);
+
+ /*
+ * We might be cloning a socket which is left in an empty
+ * cgroup and the cgroup might have already been rmdir'd.
+ * Don't use cgroup_get_live().
+ */
+ cgroup_get(cgrp);
+ cgroup_bpf_get(cgrp);
}
void cgroup_sk_free(struct sock_cgroup_data *skcd)
{
struct cgroup *cgrp = sock_cgroup_ptr(skcd);
- if (skcd->no_refcnt)
- return;
cgroup_bpf_put(cgrp);
cgroup_put(cgrp);
}
diff --git a/net/core/netclassid_cgroup.c b/net/core/netclassid_cgroup.c
index b49c57d..1a6a866 100644
--- a/net/core/netclassid_cgroup.c
+++ b/net/core/netclassid_cgroup.c
@@ -71,11 +71,8 @@
struct update_classid_context *ctx = (void *)v;
struct socket *sock = sock_from_file(file);
- if (sock) {
- spin_lock(&cgroup_sk_update_lock);
+ if (sock)
sock_cgroup_set_classid(&sock->sk->sk_cgrp_data, ctx->classid);
- spin_unlock(&cgroup_sk_update_lock);
- }
if (--ctx->batch == 0) {
ctx->batch = UPDATE_CLASSID_BATCH;
return n + 1;
@@ -121,8 +118,6 @@
struct css_task_iter it;
struct task_struct *p;
- cgroup_sk_alloc_disable();
-
cs->classid = (u32)value;
css_task_iter_start(css, 0, &it);
diff --git a/net/core/netprio_cgroup.c b/net/core/netprio_cgroup.c
index 99a431c..8456dfb 100644
--- a/net/core/netprio_cgroup.c
+++ b/net/core/netprio_cgroup.c
@@ -207,8 +207,6 @@
if (!dev)
return -ENODEV;
- cgroup_sk_alloc_disable();
-
rtnl_lock();
ret = netprio_set_prio(of_css(of), dev, prio);
@@ -221,12 +219,10 @@
static int update_netprio(const void *v, struct file *file, unsigned n)
{
struct socket *sock = sock_from_file(file);
- if (sock) {
- spin_lock(&cgroup_sk_update_lock);
+
+ if (sock)
sock_cgroup_set_prioidx(&sock->sk->sk_cgrp_data,
(unsigned long)v);
- spin_unlock(&cgroup_sk_update_lock);
- }
return 0;
}
@@ -235,8 +231,6 @@
struct task_struct *p;
struct cgroup_subsys_state *css;
- cgroup_sk_alloc_disable();
-
cgroup_taskset_for_each(p, css, tset) {
void *v = (void *)(unsigned long)css->id;
diff --git a/tools/testing/selftests/bpf/cgroup_helpers.c b/tools/testing/selftests/bpf/cgroup_helpers.c
index 0330517..f3daa44 100644
--- a/tools/testing/selftests/bpf/cgroup_helpers.c
+++ b/tools/testing/selftests/bpf/cgroup_helpers.c
@@ -12,27 +12,36 @@
#include <unistd.h>
#include <ftw.h>
-
#include "cgroup_helpers.h"
/*
* To avoid relying on the system setup, when setup_cgroup_env is called
- * we create a new mount namespace, and cgroup namespace. The cgroup2
- * root is mounted at CGROUP_MOUNT_PATH
+ * we create a new mount namespace, and cgroup namespace. The cgroupv2
+ * root is mounted at CGROUP_MOUNT_PATH. Unfortunately, most people don't
+ * have cgroupv2 enabled at this point in time. It's easier to create our
+ * own mount namespace and manage it ourselves. We assume /mnt exists.
*
- * Unfortunately, most people don't have cgroupv2 enabled at this point in time.
- * It's easier to create our own mount namespace and manage it ourselves.
- *
- * We assume /mnt exists.
+ * Related cgroupv1 helpers are named *classid*(), since we only use the
+ * net_cls controller for tagging net_cls.classid. We assume the default
+ * mount under /sys/fs/cgroup/net_cls, which should be the case for the
+ * vast majority of users.
*/
#define WALK_FD_LIMIT 16
+
#define CGROUP_MOUNT_PATH "/mnt"
+#define CGROUP_MOUNT_DFLT "/sys/fs/cgroup"
+#define NETCLS_MOUNT_PATH CGROUP_MOUNT_DFLT "/net_cls"
#define CGROUP_WORK_DIR "/cgroup-test-work-dir"
+
#define format_cgroup_path(buf, path) \
snprintf(buf, sizeof(buf), "%s%s%s", CGROUP_MOUNT_PATH, \
CGROUP_WORK_DIR, path)
+#define format_classid_path(buf) \
+ snprintf(buf, sizeof(buf), "%s%s", NETCLS_MOUNT_PATH, \
+ CGROUP_WORK_DIR)
+
/**
* enable_all_controllers() - Enable all available cgroup v2 controllers
*
@@ -139,8 +148,7 @@
return 0;
}
-
-static int join_cgroup_from_top(char *cgroup_path)
+static int join_cgroup_from_top(const char *cgroup_path)
{
char cgroup_procs_path[PATH_MAX + 1];
pid_t pid = getpid();
@@ -313,3 +321,114 @@
}
return cg_fd;
}
+
+/**
+ * setup_classid_environment() - Setup the cgroupv1 net_cls environment
+ *
+ * After calling this function, cleanup_classid_environment should be called
+ * once testing is complete.
+ *
+ * This function will print an error to stderr and return 1 if it is unable
+ * to setup the cgroup environment. If setup is successful, 0 is returned.
+ */
+int setup_classid_environment(void)
+{
+ char cgroup_workdir[PATH_MAX + 1];
+
+ format_classid_path(cgroup_workdir);
+
+ if (mount("tmpfs", CGROUP_MOUNT_DFLT, "tmpfs", 0, NULL) &&
+ errno != EBUSY) {
+ log_err("mount cgroup base");
+ return 1;
+ }
+
+ if (mkdir(NETCLS_MOUNT_PATH, 0777) && errno != EEXIST) {
+ log_err("mkdir cgroup net_cls");
+ return 1;
+ }
+
+ if (mount("net_cls", NETCLS_MOUNT_PATH, "cgroup", 0, "net_cls") &&
+ errno != EBUSY) {
+ log_err("mount cgroup net_cls");
+ return 1;
+ }
+
+ cleanup_classid_environment();
+
+ if (mkdir(cgroup_workdir, 0777) && errno != EEXIST) {
+ log_err("mkdir cgroup work dir");
+ return 1;
+ }
+
+ return 0;
+}
+
+/**
+ * set_classid() - Set a cgroupv1 net_cls classid
+ * @id: the numeric classid
+ *
+ * Writes the passed classid into the cgroup work dir's net_cls.classid
+ * file in order to later on trigger socket tagging.
+ *
+ * On success, it returns 0, otherwise on failure it returns 1. If there
+ * is a failure, it prints the error to stderr.
+ */
+int set_classid(unsigned int id)
+{
+ char cgroup_workdir[PATH_MAX - 42];
+ char cgroup_classid_path[PATH_MAX + 1];
+ int fd, rc = 0;
+
+ format_classid_path(cgroup_workdir);
+ snprintf(cgroup_classid_path, sizeof(cgroup_classid_path),
+ "%s/net_cls.classid", cgroup_workdir);
+
+ fd = open(cgroup_classid_path, O_WRONLY);
+ if (fd < 0) {
+ log_err("Opening cgroup classid: %s", cgroup_classid_path);
+ return 1;
+ }
+
+ if (dprintf(fd, "%u\n", id) < 0) {
+ log_err("Setting cgroup classid");
+ rc = 1;
+ }
+
+ close(fd);
+ return rc;
+}
+
+/**
+ * join_classid() - Join a cgroupv1 net_cls classid
+ *
+ * This function expects the cgroup work dir to be already created, as we
+ * join it here. This causes the process sockets to be tagged with the given
+ * net_cls classid.
+ *
+ * On success, it returns 0, otherwise on failure it returns 1.
+ */
+int join_classid(void)
+{
+ char cgroup_workdir[PATH_MAX + 1];
+
+ format_classid_path(cgroup_workdir);
+ return join_cgroup_from_top(cgroup_workdir);
+}
+
+/**
+ * cleanup_classid_environment() - Cleanup the cgroupv1 net_cls environment
+ *
+ * At call time, it moves the calling process to the root cgroup, and then
+ * runs the deletion process.
+ *
+ * On failure, it will print an error to stderr, and try to continue.
+ */
+void cleanup_classid_environment(void)
+{
+ char cgroup_workdir[PATH_MAX + 1];
+
+ format_classid_path(cgroup_workdir);
+ join_cgroup_from_top(NETCLS_MOUNT_PATH);
+ nftw(cgroup_workdir, nftwfunc, WALK_FD_LIMIT, FTW_DEPTH | FTW_MOUNT);
+}
diff --git a/tools/testing/selftests/bpf/cgroup_helpers.h b/tools/testing/selftests/bpf/cgroup_helpers.h
index 5fe3d88..629da38 100644
--- a/tools/testing/selftests/bpf/cgroup_helpers.h
+++ b/tools/testing/selftests/bpf/cgroup_helpers.h
@@ -1,6 +1,7 @@
/* SPDX-License-Identifier: GPL-2.0 */
#ifndef __CGROUP_HELPERS_H
#define __CGROUP_HELPERS_H
+
#include <errno.h>
#include <string.h>
@@ -8,12 +9,21 @@
#define log_err(MSG, ...) fprintf(stderr, "(%s:%d: errno: %s) " MSG "\n", \
__FILE__, __LINE__, clean_errno(), ##__VA_ARGS__)
-
+/* cgroupv2 related */
int cgroup_setup_and_join(const char *path);
int create_and_get_cgroup(const char *path);
-int join_cgroup(const char *path);
-int setup_cgroup_environment(void);
-void cleanup_cgroup_environment(void);
unsigned long long get_cgroup_id(const char *path);
-#endif
+int join_cgroup(const char *path);
+
+int setup_cgroup_environment(void);
+void cleanup_cgroup_environment(void);
+
+/* cgroupv1 related */
+int set_classid(unsigned int id);
+int join_classid(void);
+
+int setup_classid_environment(void);
+void cleanup_classid_environment(void);
+
+#endif /* __CGROUP_HELPERS_H */
diff --git a/tools/testing/selftests/bpf/network_helpers.c b/tools/testing/selftests/bpf/network_helpers.c
index 7e9f637..6db1af8 100644
--- a/tools/testing/selftests/bpf/network_helpers.c
+++ b/tools/testing/selftests/bpf/network_helpers.c
@@ -208,11 +208,26 @@
static int connect_fd_to_addr(int fd,
const struct sockaddr_storage *addr,
- socklen_t addrlen)
+ socklen_t addrlen, const bool must_fail)
{
- if (connect(fd, (const struct sockaddr *)addr, addrlen)) {
- log_err("Failed to connect to server");
- return -1;
+ int ret;
+
+ errno = 0;
+ ret = connect(fd, (const struct sockaddr *)addr, addrlen);
+ if (must_fail) {
+ if (!ret) {
+ log_err("Unexpected success to connect to server");
+ return -1;
+ }
+ if (errno != EPERM) {
+ log_err("Unexpected error from connect to server");
+ return -1;
+ }
+ } else {
+ if (ret) {
+ log_err("Failed to connect to server");
+ return -1;
+ }
}
return 0;
@@ -257,7 +272,7 @@
strlen(opts->cc) + 1))
goto error_close;
- if (connect_fd_to_addr(fd, &addr, addrlen))
+ if (connect_fd_to_addr(fd, &addr, addrlen, opts->must_fail))
goto error_close;
return fd;
@@ -289,7 +304,7 @@
return -1;
}
- if (connect_fd_to_addr(client_fd, &addr, len))
+ if (connect_fd_to_addr(client_fd, &addr, len, false))
return -1;
return 0;
diff --git a/tools/testing/selftests/bpf/network_helpers.h b/tools/testing/selftests/bpf/network_helpers.h
index da7e132..d198181 100644
--- a/tools/testing/selftests/bpf/network_helpers.h
+++ b/tools/testing/selftests/bpf/network_helpers.h
@@ -20,6 +20,7 @@
struct network_helper_opts {
const char *cc;
int timeout_ms;
+ bool must_fail;
};
/* ipv4 test vector */
diff --git a/tools/testing/selftests/bpf/prog_tests/cgroup_v1v2.c b/tools/testing/selftests/bpf/prog_tests/cgroup_v1v2.c
new file mode 100644
index 0000000..ab3b9bc
--- /dev/null
+++ b/tools/testing/selftests/bpf/prog_tests/cgroup_v1v2.c
@@ -0,0 +1,79 @@
+// SPDX-License-Identifier: GPL-2.0
+
+#include <test_progs.h>
+
+#include "connect4_dropper.skel.h"
+
+#include "cgroup_helpers.h"
+#include "network_helpers.h"
+
+static int run_test(int cgroup_fd, int server_fd, bool classid)
+{
+ struct network_helper_opts opts = {
+ .must_fail = true,
+ };
+ struct connect4_dropper *skel;
+ int fd, err = 0;
+
+ skel = connect4_dropper__open_and_load();
+ if (!ASSERT_OK_PTR(skel, "skel_open"))
+ return -1;
+
+ skel->links.connect_v4_dropper =
+ bpf_program__attach_cgroup(skel->progs.connect_v4_dropper,
+ cgroup_fd);
+ if (!ASSERT_OK_PTR(skel->links.connect_v4_dropper, "prog_attach")) {
+ err = -1;
+ goto out;
+ }
+
+ if (classid && !ASSERT_OK(join_classid(), "join_classid")) {
+ err = -1;
+ goto out;
+ }
+
+ fd = connect_to_fd_opts(server_fd, &opts);
+ if (fd < 0)
+ err = -1;
+ else
+ close(fd);
+out:
+ connect4_dropper__destroy(skel);
+ return err;
+}
+
+void test_cgroup_v1v2(void)
+{
+ struct network_helper_opts opts = {};
+ int server_fd, client_fd, cgroup_fd;
+ static const int port = 60123;
+
+ /* Step 1: Check base connectivity works without any BPF. */
+ server_fd = start_server(AF_INET, SOCK_STREAM, NULL, port, 0);
+ if (!ASSERT_GE(server_fd, 0, "server_fd"))
+ return;
+ client_fd = connect_to_fd_opts(server_fd, &opts);
+ if (!ASSERT_GE(client_fd, 0, "client_fd")) {
+ close(server_fd);
+ return;
+ }
+ close(client_fd);
+ close(server_fd);
+
+ /* Step 2: Check BPF policy prog attached to cgroups drops connectivity. */
+ cgroup_fd = test__join_cgroup("/connect_dropper");
+ if (!ASSERT_GE(cgroup_fd, 0, "cgroup_fd"))
+ return;
+ server_fd = start_server(AF_INET, SOCK_STREAM, NULL, port, 0);
+ if (!ASSERT_GE(server_fd, 0, "server_fd")) {
+ close(cgroup_fd);
+ return;
+ }
+ ASSERT_OK(run_test(cgroup_fd, server_fd, false), "cgroup-v2-only");
+ setup_classid_environment();
+ set_classid(42);
+ ASSERT_OK(run_test(cgroup_fd, server_fd, true), "cgroup-v1v2");
+ cleanup_classid_environment();
+ close(server_fd);
+ close(cgroup_fd);
+}
diff --git a/tools/testing/selftests/bpf/prog_tests/task_pt_regs.c b/tools/testing/selftests/bpf/prog_tests/task_pt_regs.c
index 53f0e0f..37c20b5 100644
--- a/tools/testing/selftests/bpf/prog_tests/task_pt_regs.c
+++ b/tools/testing/selftests/bpf/prog_tests/task_pt_regs.c
@@ -1,7 +1,6 @@
// SPDX-License-Identifier: GPL-2.0
#define _GNU_SOURCE
#include <test_progs.h>
-#include <linux/ptrace.h>
#include "test_task_pt_regs.skel.h"
void test_task_pt_regs(void)
diff --git a/tools/testing/selftests/bpf/progs/connect4_dropper.c b/tools/testing/selftests/bpf/progs/connect4_dropper.c
new file mode 100644
index 0000000..b565d99
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/connect4_dropper.c
@@ -0,0 +1,26 @@
+// SPDX-License-Identifier: GPL-2.0
+
+#include <string.h>
+
+#include <linux/stddef.h>
+#include <linux/bpf.h>
+
+#include <sys/socket.h>
+
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_endian.h>
+
+#define VERDICT_REJECT 0
+#define VERDICT_PROCEED 1
+
+SEC("cgroup/connect4")
+int connect_v4_dropper(struct bpf_sock_addr *ctx)
+{
+ if (ctx->type != SOCK_STREAM)
+ return VERDICT_PROCEED;
+ if (ctx->user_port == bpf_htons(60123))
+ return VERDICT_REJECT;
+ return VERDICT_PROCEED;
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/progs/test_task_pt_regs.c b/tools/testing/selftests/bpf/progs/test_task_pt_regs.c
index 6c059f1..e6cb092 100644
--- a/tools/testing/selftests/bpf/progs/test_task_pt_regs.c
+++ b/tools/testing/selftests/bpf/progs/test_task_pt_regs.c
@@ -1,12 +1,17 @@
// SPDX-License-Identifier: GPL-2.0
-#include <linux/ptrace.h>
-#include <linux/bpf.h>
+#include "vmlinux.h"
#include <bpf/bpf_helpers.h>
#include <bpf/bpf_tracing.h>
-struct pt_regs current_regs = {};
-struct pt_regs ctx_regs = {};
+#define PT_REGS_SIZE sizeof(struct pt_regs)
+
+/*
+ * The kernel struct pt_regs isn't exported in its entirety to userspace.
+ * Pass it as an array to task_pt_regs.c
+ */
+char current_regs[PT_REGS_SIZE] = {};
+char ctx_regs[PT_REGS_SIZE] = {};
int uprobe_res = 0;
SEC("uprobe/trigger_func")
@@ -17,8 +22,10 @@
current = bpf_get_current_task_btf();
regs = (struct pt_regs *) bpf_task_pt_regs(current);
- __builtin_memcpy(¤t_regs, regs, sizeof(*regs));
- __builtin_memcpy(&ctx_regs, ctx, sizeof(*ctx));
+ if (bpf_probe_read_kernel(current_regs, PT_REGS_SIZE, regs))
+ return 0;
+ if (bpf_probe_read_kernel(ctx_regs, PT_REGS_SIZE, ctx))
+ return 0;
/* Prove that uprobe was run */
uprobe_res = 1;