)]}'
{
  "commit": "ab61766635328afe458ef61fa953a72bd765e559",
  "tree": "56b6f0f4b14f48c3f24f249f3b72daacb3767702",
  "parents": [
    "179a6d5b2265964b1353838256b4839471bd6fa9",
    "7ca3702d40d9ba2de7876abc03b116ca0e9f5264"
  ],
  "author": {
    "name": "Jakub Kicinski",
    "email": "kuba@kernel.org",
    "time": "Fri Sep 25 17:45:35 2026 -0700"
  },
  "committer": {
    "name": "Jakub Kicinski",
    "email": "kuba@kernel.org",
    "time": "Fri Sep 25 17:45:35 2026 -0700"
  },
  "message": "Merge branch \u0027net-macb-rework-hardware-timestamp-configuration\u0027\n\nKim Wooseok via says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nnet: macb: Rework hardware timestamp configuration\n\nThis started with two problems I found while using PTP hardware\ntimestamping on a Raspberry Pi 5. A rejected RX filter could still\nchange the TX one-step setting, and the PTPv1 Sync and Delay_Req filters\nreturned success without enabling RX timestamps.\n\nThis series reworks gem_set_hwtst() to validate the whole request before\nprogramming the hardware. It calculates the settings locally, then\nupdates the registers and saved configuration under bp-\u003elock. With the\nregister writes now in the setter, the two helpers are no longer needed.\n\nThe remaining patches use the existing ALL fallback for the specific\nPTPv1 filters and clear SRTSM when switching away from a PTPv2 filter.\nEach change is kept in its own patch.\n\nFor v1, I compared the old and new behavior on the Pi 5. When I\nrequested a TX mode change together with an invalid RX filter, the\npatched driver rejected the request without changing the saved settings\nor NCR. I also checked transitions between PTPv2 RX and the other\nreceive filters. SRTSM followed the selected RX filter without changing\nunrelated NCR bits.\n\nFor the PTPv1 filters, I checked the effect by sending Sync and Delay_Req\npackets over Ethernet. Before the change, the packets arrived without\nhardware timestamps. With the patches, the driver returned ALL and I\ncould read the hardware timestamps from the received packets. PTPv2\nmulticast reception continued to provide hardware timestamps, and\nselecting NONE disabled timestamping. I ran these tests with the changes\nbackported to the Pi\u0027s existing Linux 6.18.46 RT kernel.\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n\nLink: https://patch.msgid.link/20260923-macb-hwtstamp-rework-v2-0-75c909b2676d@khu.ac.kr\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n",
  "tree_diff": []
}