)]}'
{
  "commit": "a237c1c5bc5dc5c76a21be922dca4826f3eca8ca",
  "tree": "a216c9a6d9e870b84424938e9e0b4722dc8634cd",
  "parents": [
    "5853b4f06f7b9b56f37f457d7923f7b96496074e"
  ],
  "author": {
    "name": "Jens Axboe",
    "email": "jaxboe@fusionio.com",
    "time": "Sat Apr 16 13:27:55 2011 +0200"
  },
  "committer": {
    "name": "Jens Axboe",
    "email": "jaxboe@fusionio.com",
    "time": "Sat Apr 16 13:27:55 2011 +0200"
  },
  "message": "block: let io_schedule() flush the plug inline\n\nLinus correctly observes that the most important dispatch cases\nare now done from kblockd, this isn\u0027t ideal for latency reasons.\nThe original reason for switching dispatches out-of-line was to\navoid too deep a stack, so by _only_ letting the \"accidental\"\nflush directly in schedule() be guarded by offload to kblockd,\nwe should be able to get the best of both worlds.\n\nSo add a blk_schedule_flush_plug() that offloads to kblockd,\nand only use that from the schedule() path.\n\nSigned-off-by: Jens Axboe \u003cjaxboe@fusionio.com\u003e\n",
  "tree_diff": [
    {
      "type": "modify",
      "old_id": "1c76506fcf11cfe260276475762fadc6d4b09dfc",
      "old_mode": 33188,
      "old_path": "include/linux/blkdev.h",
      "new_id": "ec0357d8c4a58679c96bc3e937b5de97e9a4063c",
      "new_mode": 33188,
      "new_path": "include/linux/blkdev.h"
    },
    {
      "type": "modify",
      "old_id": "a187c3fe027b58debb8c15d6ec98f76d6eb62b75",
      "old_mode": 33188,
      "old_path": "kernel/sched.c",
      "new_id": "312f8b95c2d44fbbc7c7097de04b52d92abc0648",
      "new_mode": 33188,
      "new_path": "kernel/sched.c"
    }
  ]
}