)]}'
{
  "commit": "1187cf0a3c8b647d08bc86e043563c8d2a327adc",
  "tree": "759515a3fdcbf962f35255f90ea6c4d7c708391a",
  "parents": [
    "eea1bf384e05b5ab747f8530c4fba9e9e6907fff"
  ],
  "author": {
    "name": "NeilBrown",
    "email": "neilb@suse.de",
    "time": "Tue Mar 31 14:27:02 2009 +1100"
  },
  "committer": {
    "name": "NeilBrown",
    "email": "neilb@suse.de",
    "time": "Tue Mar 31 14:27:02 2009 +1100"
  },
  "message": "md: Allow write-intent bitmaps to have chunksize \u003c PAGE_SIZE\n\nmd currently insists that the chunk size used for write-intent\nbitmaps (the amount of data that corresponds to one chunk)\nbe at least one page.\n\nThe reason for this restriction is lost in the mists of time,\nbut a review of the code (and a vague memory) suggests that the only\nproblem would be related to resync.  Resync tries very hard to\nwork in multiples of a page, but also needs to sync with units\nof a bitmap_chunk too.\n\nThis connection comes out in the bitmap_start_sync call.\n\nSo change bitmap_start_sync to always work in multiples of a page.\nIf the bitmap chunk size is less that one page, we flag multiple\nchunks as \u0027syncing\u0027 and generally make them all appear to the\nresync routines like one chunk.\n\nAll other code either already works with data ranges that could\nspan multiple chunks, or explicitly only cares about a single chunk.\n\nSigned-off-by: Neil Brown \u003cneilb@suse.de\u003e\n",
  "tree_diff": [
    {
      "type": "modify",
      "old_id": "719943763391263c7a4ab98ad0201244ce1d3d62",
      "old_mode": 33188,
      "old_path": "drivers/md/bitmap.c",
      "new_id": "8fa3277f72dcf7af665042c88334e1d07f579cae",
      "new_mode": 33188,
      "new_path": "drivers/md/bitmap.c"
    }
  ]
}