)]}'
{
  "commit": "140dfc9299c33bbfc9350fa061f5ab65cb83df13",
  "tree": "09508691964e277f4835d30f7b9c3962e8cac596",
  "parents": [
    "f94784bdb114439eb3a5e62343826887bbf3f37c",
    "1a71d6ffe18c0d0f03fc8531949cc8ed41d702ee"
  ],
  "author": {
    "name": "Linus Torvalds",
    "email": "torvalds@linux-foundation.org",
    "time": "Mon Dec 08 21:10:03 2014 -0800"
  },
  "committer": {
    "name": "Linus Torvalds",
    "email": "torvalds@linux-foundation.org",
    "time": "Mon Dec 08 21:10:03 2014 -0800"
  },
  "message": "Merge tag \u0027dm-3.19-changes\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm\n\nPull device mapper updates from Mike Snitzer:\n\n - Significant DM thin-provisioning performance improvements to meet\n   performance requirements that were requested by the Gluster\n   distributed filesystem.\n\n   Specifically, dm-thinp now takes care to aggregate IO that will be\n   issued to the same thinp block before issuing IO to the underlying\n   devices.  This really helps improve performance on HW RAID6 devices\n   that have a writeback cache because it avoids RMW in the HW RAID\n   controller.\n\n - Some stable fixes: fix leak in DM bufio if integrity profiles were\n   enabled, use memzero_explicit in DM crypt to avoid any potential for\n   information leak, and a DM cache fix to properly mark a cache block\n   dirty if it was promoted to the cache via the overwrite optimization.\n\n - A few simple DM persistent data library fixes\n\n - DM cache multiqueue policy block promotion improvements.\n\n - DM cache discard improvements that take advantage of range\n   (multiblock) discard support in the DM bio-prison.  This allows for\n   much more efficient bulk discard processing (e.g.  when mkfs.xfs\n   discards the entire device).\n\n - Some small optimizations in DM core and RCU deference cleanups\n\n - DM core changes to suspend/resume code to introduce the new internal\n   suspend/resume interface that the DM thin-pool target now uses to\n   suspend/resume active thin devices when the thin-pool must\n   suspend/resume.\n\n   This avoids forcing userspace to track all active thin volumes in a\n   thin-pool when the thin-pool is suspended for the purposes of\n   metadata or data space resize.\n\n* tag \u0027dm-3.19-changes\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm: (49 commits)\n  dm crypt: use memzero_explicit for on-stack buffer\n  dm space map metadata: fix sm_bootstrap_get_count()\n  dm space map metadata: fix sm_bootstrap_get_nr_blocks()\n  dm bufio: fix memleak when using a dm_buffer\u0027s inline bio\n  dm cache: fix spurious cell_defer when dealing with partial block at end of device\n  dm cache: dirty flag was mistakenly being cleared when promoting via overwrite\n  dm cache: only use overwrite optimisation for promotion when in writeback mode\n  dm cache: discard block size must be a multiple of cache block size\n  dm cache: fix a harmless race when working out if a block is discarded\n  dm cache: when reloading a discard bitset allow for a different discard block size\n  dm cache: fix some issues with the new discard range support\n  dm array: if resizing the array is a noop set the new root to the old one\n  dm: use rcu_dereference_protected instead of rcu_dereference\n  dm thin: fix pool_io_hints to avoid looking at max_hw_sectors\n  dm thin: suspend/resume active thin devices when reloading thin-pool\n  dm: enhance internal suspend and resume interface\n  dm thin: do not allow thin device activation while pool is suspended\n  dm: add presuspend_undo hook to target_type\n  dm: return earlier from dm_blk_ioctl if target doesn\u0027t implement .ioctl\n  dm thin: remove stale \u0027trim\u0027 message in block comment above pool_message\n  ...\n",
  "tree_diff": []
}