)]}'
{
  "commit": "51170840fe91dfca10fd533b303ea39b2524782a",
  "tree": "ecc5a035c11c0ce1f7a5891c6d0331a39a5293d1",
  "parents": [
    "525705d15e63b7455977408e4601e76e6bc41524"
  ],
  "author": {
    "name": "Rik van Riel",
    "email": "riel@redhat.com",
    "time": "Thu Nov 05 15:56:23 2015 -0500"
  },
  "committer": {
    "name": "Ingo Molnar",
    "email": "mingo@kernel.org",
    "time": "Mon Nov 23 09:37:54 2015 +0100"
  },
  "message": "sched/numa: Cap PTE scanning overhead to 3% of run time\n\nThere is a fundamental mismatch between the runtime based NUMA scanning\nat the task level, and the wall clock time NUMA scanning at the mm level.\nOn a severely overloaded system, with very large processes, this mismatch\ncan cause the system to spend all of its time in change_prot_numa().\n\nThis can happen if the task spends at least two ticks in change_prot_numa(),\nand only gets two ticks of CPU time in the real time between two scan\nintervals of the mm.\n\nThis patch ensures that a task never spends more than 3% of run\ntime scanning PTEs. It does that by ensuring that in-between\ntask_numa_work() runs, the task spends at least 32x as much time on\nother things than it did on task_numa_work().\n\nThis is done stochastically: if a timer tick happens, or the task\ngets rescheduled during task_numa_work(), we delay a future run of\ntask_numa_work() until the task has spent at least 32x the amount of\nCPU time doing something else, as it spent inside task_numa_work().\nThe longer task_numa_work() takes, the more likely it is this happens.\n\nIf task_numa_work() takes very little time, chances are low that that\ncode will do anything, but we will not care.\n\nReported-and-tested-by: Jan Stancek \u003cjstancek@redhat.com\u003e\nSigned-off-by: Rik van Riel \u003criel@redhat.com\u003e\nSigned-off-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\nCc: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\nCc: Mike Galbraith \u003cefault@gmx.de\u003e\nCc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\nCc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nCc: mgorman@suse.de\nLink: http://lkml.kernel.org/r/1446756983-28173-3-git-send-email-riel@redhat.com\nSigned-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n",
  "tree_diff": [
    {
      "type": "modify",
      "old_id": "309b1d551f25a0ea568ce38b589b9632e6648ec2",
      "old_mode": 33188,
      "old_path": "kernel/sched/fair.c",
      "new_id": "95b944ecf7e4c90cb8d4721a234e6f8566bb6b8e",
      "new_mode": 33188,
      "new_path": "kernel/sched/fair.c"
    }
  ]
}