)]}'
{
  "commit": "ddbd3b61708483f73dbcc62a94d16cc7db928cba",
  "tree": "7600caa289d1517e8642d6be21d0a0cfe88c40d5",
  "parents": [
    "95c9c1da79e59fd10ec5da3aeba22981383f7040"
  ],
  "author": {
    "name": "Artem Bityutskiy",
    "email": "Artem.Bityutskiy@nokia.com",
    "time": "Sat May 23 13:44:09 2009 +0300"
  },
  "committer": {
    "name": "Artem Bityutskiy",
    "email": "Artem.Bityutskiy@nokia.com",
    "time": "Tue May 26 12:47:47 2009 +0300"
  },
  "message": "UBI: fix race condition\n\nThis patch fixes a minor problem where we may fail to wake\nupe the UBI background thread. This is not fatal at all,\nit may just result at sligtly worse performace for a short\nperiod of time, just because the thread will be woken up\nwhen real I/O on the UBI starts.\n\nAnywey, the issue is the race condition between\n\u0027ubi_attach_mtd_dev()\u0027 and \u0027ubi_thread()\u0027. If we do not\nserialize them, the \u0027wake_up_process()\u0027 call may be done\nbefore \u0027ubi_thread()\u0027 went seep, but after it checked\n\u0027ubi-\u003ethread_enabled\u0027.\n\nThis issue was spotted by Shin Hong \u003chongshin@gmail.com\u003e\n\nSigned-off-by: Artem Bityutskiy \u003cArtem.Bityutskiy@nokia.com\u003e\n",
  "tree_diff": [
    {
      "type": "modify",
      "old_id": "2c3269ea133d906c70eb46458afaa73b7664db0a",
      "old_mode": 33188,
      "old_path": "drivers/mtd/ubi/build.c",
      "new_id": "1405b556c65a7c6246c6095ca770d0d2e404ae18",
      "new_mode": 33188,
      "new_path": "drivers/mtd/ubi/build.c"
    }
  ]
}