)]}'
{
  "commit": "37355bdc5a129899f6b245900a8eb944a092f7fd",
  "tree": "de0ef30ac3ec4d63f5937bacb34c029fa30c8e60",
  "parents": [
    "efaffc5e40aeced0bcb497ed7a0a5b8c14abfcdf"
  ],
  "author": {
    "name": "Mel Gorman",
    "email": "mgorman@techsingularity.net",
    "time": "Mon Oct 01 11:05:25 2018 +0100"
  },
  "committer": {
    "name": "Ingo Molnar",
    "email": "mingo@kernel.org",
    "time": "Tue Oct 02 11:31:33 2018 +0200"
  },
  "message": "sched/numa: Migrate pages to local nodes quicker early in the lifetime of a task\n\nAutomatic NUMA Balancing uses a multi-stage pass to decide whether a page\nshould migrate to a local node. This filter avoids excessive ping-ponging\nif a page is shared or used by threads that migrate cross-node frequently.\n\nThreads inherit both page tables and the preferred node ID from the\nparent. This means that threads can trigger hinting faults earlier than\na new task which delays scanning for a number of seconds. As it can be\nload balanced very early in its lifetime there can be an unnecessary delay\nbefore it starts migrating thread-local data. This patch migrates private\npages faster early in the lifetime of a thread using the sequence counter\nas an identifier of new tasks.\n\nWith this patch applied, STREAM performance is the same as 4.17 even though\nprocesses are not spread cross-node prematurely. Other workloads showed\na mix of minor gains and losses. This is somewhat expected most workloads\nare not very sensitive to the starting conditions of a process.\n\n                         4.19.0-rc5             4.19.0-rc5                 4.17.0\n                         numab-v1r1       fastmigrate-v1r1                vanilla\nMB/sec copy     43298.52 (   0.00%)    47335.46 (   9.32%)    47219.24 (   9.06%)\nMB/sec scale    30115.06 (   0.00%)    32568.12 (   8.15%)    32527.56 (   8.01%)\nMB/sec add      32825.12 (   0.00%)    36078.94 (   9.91%)    35928.02 (   9.45%)\nMB/sec triad    32549.52 (   0.00%)    35935.94 (  10.40%)    35969.88 (  10.51%)\n\nSigned-off-by: Mel Gorman \u003cmgorman@techsingularity.net\u003e\nReviewed-by: Rik van Riel \u003criel@surriel.com\u003e\nAcked-by: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nCc: Jirka Hladky \u003cjhladky@redhat.com\u003e\nCc: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\nCc: Linux-MM \u003clinux-mm@kvack.org\u003e\nCc: Srikar Dronamraju \u003csrikar@linux.vnet.ibm.com\u003e\nCc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nLink: http://lkml.kernel.org/r/20181001100525.29789-3-mgorman@techsingularity.net\nSigned-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n",
  "tree_diff": [
    {
      "type": "modify",
      "old_id": "25c7c7e09cbdd7af90c35a9c4887fa60208dbfdb",
      "old_mode": 33188,
      "old_path": "kernel/sched/fair.c",
      "new_id": "7fc4a371bdd248ee94ffb7f5088bb86b5fa42262",
      "new_mode": 33188,
      "new_path": "kernel/sched/fair.c"
    }
  ]
}