)]}'
{
  "commit": "79c537998d143b127c8c662a403c3356cb885f1c",
  "tree": "ed358cd2f78741b17f87cd74517250f434a34e19",
  "parents": [
    "543cf4cb3fe6f6cae3651ba918b9c56200b257d0"
  ],
  "author": {
    "name": "Dmitry Adamushko",
    "email": "dmitry.adamushko@gmail.com",
    "time": "Sun Jun 29 00:16:56 2008 +0200"
  },
  "committer": {
    "name": "Ingo Molnar",
    "email": "mingo@elte.hu",
    "time": "Sun Jun 29 08:50:21 2008 +0200"
  },
  "message": "sched: fix cpu hotplug\n\nthe CPU hotplug problems (crashes under high-volume unplug+replug\ntests) seem to be related to migrate_dead_tasks().\n\nFirstly I added traces to see all tasks being migrated with\nmigrate_live_tasks() and migrate_dead_tasks(). On my setup the problem\npops up (the one with \"se \u003d\u003d NULL\" in the loop of\npick_next_task_fair()) shortly after the traces indicate that some has\nbeen migrated with migrate_dead_tasks()). btw., I can reproduce it\nmuch faster now with just a plain cpu down/up loop.\n\n[disclaimer] Well, unless I\u0027m really missing something important in\nthis late hour [/desclaimer] pick_next_task() is not something\nappropriate for migrate_dead_tasks() :-)\n\nthe following change seems to eliminate the problem on my setup\n(although, I kept it running only for a few minutes to get a few\nmessages indicating migrate_dead_tasks() does move tasks and the\nsystem is still ok)\n\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n",
  "tree_diff": [
    {
      "type": "modify",
      "old_id": "3aaa5c8cb4214586bd283eedc2375baa3dc739c0",
      "old_mode": 33188,
      "old_path": "kernel/sched.c",
      "new_id": "a66e85639de2e541dd9726245cd344d3ea2d9f3b",
      "new_mode": 33188,
      "new_path": "kernel/sched.c"
    }
  ]
}