)]}'
{
  "commit": "6a51a10decf92a2a0d585d485aada065877625c8",
  "tree": "1fb2dccfd426240bd60e9407120528bb3bc338c0",
  "parents": [
    "f21051b72829aa254ef2cb5b09417506410d1a38"
  ],
  "author": {
    "name": "Nicolas Pitre",
    "email": "npitre@baylibre.com",
    "time": "Mon Feb 13 00:43:47 2023 -0500"
  },
  "committer": {
    "name": "Fabio Baltieri",
    "email": "fabio.baltieri@gmail.com",
    "time": "Mon Feb 20 09:52:53 2023 +0100"
  },
  "message": "kernel/timer: fix period argument clamp handling\n\nCommit 3e729b2b1c52 (\"kernel/timer: Correctly clamp period argument\")\nincreased the lower limit to 1 so that it wouldn\u0027t conflict with a\nK_NO_WAIT. But in doing so it enforced a minimum period of 2 ticks.\nAnd the subtraction must obviously be avoided if the period is zero, etc.\n\nInstead of doing this masquerade in k_timer_start(), let\u0027s move the\nsubtraction and clamping in z_timer_expiration_handler() right before\nregistering a new timeout. It makes the code cleaner, and then it is\npossible to have single-tick periods again.\n\nWhith this, timer_jitter_drift in tests/kernel/timer/timer_behavior does\npass with any CONFIG_SYS_CLOCK_TICKS_PER_SEC value, even when the tick\nperiod is equal or larger than the specified timer period for the test\nwhich failed the test before.\n\nSigned-off-by: Nicolas Pitre \u003cnpitre@baylibre.com\u003e\n",
  "tree_diff": [
    {
      "type": "modify",
      "old_id": "325224d11bd87a98027fa20b149d71222e4e9733",
      "old_mode": 33188,
      "old_path": "kernel/timer.c",
      "new_id": "9de9768c3613fd7673e711ce6b574b2dce0f94a4",
      "new_mode": 33188,
      "new_path": "kernel/timer.c"
    }
  ]
}