)]}'
{
  "commit": "cd0aca2d550f238d80ba58e7dcade4ea3d0a3aa7",
  "tree": "9581a77ce54247a18963c9d827063923a667add7",
  "parents": [
    "25636e282fe95508cae96bb27f86407aef935817"
  ],
  "author": {
    "name": "Tejun Heo",
    "email": "tj@kernel.org",
    "time": "Wed Apr 15 22:10:25 2009 +0900"
  },
  "committer": {
    "name": "Jens Axboe",
    "email": "jens.axboe@oracle.com",
    "time": "Wed Apr 22 08:35:09 2009 +0200"
  },
  "message": "block: fix queue bounce limit setting\n\nImpact: don\u0027t set GFP_DMA in q-\u003ebounce_gfp unnecessarily\n\nAll DMA address limits are expressed in terms of the last addressable\nunit (byte or page) instead of one plus that.  However, when\ndetermining bounce_gfp for 64bit machines in blk_queue_bounce_limit(),\nit compares the specified limit against 0x100000000UL to determine\nwhether it\u0027s below 4G ending up falsely setting GFP_DMA in\nq-\u003ebounce_gfp.\n\nAs DMA zone is very small on x86_64, this makes larger SG_IO transfers\nvery eager to trigger OOM killer.  Fix it.  While at it, rename the\nparameter to @dma_mask for clarity and convert comment to proper\nwinged style.\n\nSigned-off-by: Tejun Heo \u003ctj@kernel.org\u003e\nSigned-off-by: Jens Axboe \u003cjens.axboe@oracle.com\u003e\n",
  "tree_diff": [
    {
      "type": "modify",
      "old_id": "69c42adde52bafb61913d127b9d867e69e45a0c4",
      "old_mode": 33188,
      "old_path": "block/blk-settings.c",
      "new_id": "57af728d94bb08732e9d217a94dafe9bae9ef992",
      "new_mode": 33188,
      "new_path": "block/blk-settings.c"
    }
  ]
}