)]}'
{
  "commit": "e91defa26c527ceeaff6266c55cdc7e17c9081a2",
  "tree": "03554f76949e5b1bcd4e640a10846d9540b97640",
  "parents": [
    "acae05156551fd7528fbb616271e672789388e3c"
  ],
  "author": {
    "name": "Rusty Russell",
    "email": "rusty@rustcorp.com.au",
    "time": "Tue Mar 31 13:05:35 2009 -0600"
  },
  "committer": {
    "name": "Rusty Russell",
    "email": "rusty@rustcorp.com.au",
    "time": "Tue Mar 31 13:05:35 2009 +1030"
  },
  "message": "module: don\u0027t use stop_machine on module load\n\nKay Sievers \u003ckay.sievers@vrfy.org\u003e discovered that boot times are slowed\nby about half a second because all the stop_machine_create() calls,\nand he only probes about 40 modules (I have 125 loaded on this laptop).\n\nWe only do stop_machine_create() so we can unlink the module if\nsomething goes wrong, but it\u0027s overkill (and buggy anyway: if\nstop_machine_create() fails we still call stop_machine_destroy()).\n\nSince we are only protecting against kallsyms (esp. oops) walking the\nlist, synchronize_sched() is sufficient (synchronize_rcu() is probably\nsufficient, but we\u0027re not in a hurry).\n\nKay says of this patch:\n\t... no module takes more than 40 millisecs to link now, most of\n\tthem are between 3 and 8 millisecs.\n\n\tThat looks very different to the numbers without this patch\n\tand the otherwise same setup, where we get heavy noise in the\n\ttraces and many delays of up to 200 millisecs until linking,\n\tmost of them taking 30+ millisecs.\n\nTested-by: Kay Sievers \u003ckay.sievers@vrfy.org\u003e\nSigned-off-by: Rusty Russell \u003crusty@rustcorp.com.au\u003e\n",
  "tree_diff": [
    {
      "type": "modify",
      "old_id": "e8cf636e614bafbd773397fcae32830291ae44bd",
      "old_mode": 33188,
      "old_path": "kernel/module.c",
      "new_id": "1a9a3986b136a47ae68d5255abaed6ed032d4bfe",
      "new_mode": 33188,
      "new_path": "kernel/module.c"
    }
  ]
}