)]}'
{
  "commit": "1f8f41659c133ab9e19e30850bbc90a3c0a0ac7d",
  "tree": "11c74002bdfaea33e3524e23993544496226b5dd",
  "parents": [
    "be21a3e3f3bbcb165fece05f2e8044112d4f7ac6"
  ],
  "author": {
    "name": "Imran Khan",
    "email": "imran.f.khan@oracle.com",
    "time": "Tue May 09 08:31:24 2023 +1000"
  },
  "committer": {
    "name": "Paul E. McKenney",
    "email": "paulmck@kernel.org",
    "time": "Wed Jun 07 13:37:25 2023 -0700"
  },
  "message": "smp: Reduce NMI traffic from CSD waiters to CSD destination\n\nOn systems with hundreds of CPUs, if most of the CPUs detect a CSD hang,\nthen all of these waiting CPUs send an NMI to the destination CPU in\norder to dump its backtrace.\n\nGiven enough NMIs, the destination CPU will spent much of its time\nproducing backtraces, thus further delaying that CPU\u0027s response to the\noriginal CSD IPI.  In the worst case, by the time destination CPU is\ndone producing all of these backtrace NMIs, the CSD wait timeout will\nhave elapsed so that the waiters resend their backtrace NMIs again,\nfurther delaying forward progress.\n\nTherefore, to avoid these delays, issue the backtrace NMI only from\nthe first waiter.  The destination CPU\u0027s other waiters can make use of\nbacktrace obtained from the first waiter\u0027s NMI.\n\nSigned-off-by: Imran Khan \u003cimran.f.khan@oracle.com\u003e\nCc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\nCc: Juergen Gross \u003cjgross@suse.com\u003e\nCc: Valentin Schneider \u003cvschneid@redhat.com\u003e\nCc: Yury Norov \u003cyury.norov@gmail.com\u003e\nSigned-off-by: Paul E. McKenney \u003cpaulmck@kernel.org\u003e\n",
  "tree_diff": [
    {
      "type": "modify",
      "old_id": "b7ccba677a0a003fce3f3cb2a7d07b10605afd51",
      "old_mode": 33188,
      "old_path": "kernel/smp.c",
      "new_id": "a1cd21ea8b3083c029c14b2b46b1f436d171c25e",
      "new_mode": 33188,
      "new_path": "kernel/smp.c"
    }
  ]
}