)]}'
{
  "commit": "f307b2558f61e068ce514f2dde2cad74c62036d6",
  "tree": "a507b7ba20578c1af38cc4778b0cdf7b797c0e41",
  "parents": [
    "001c158defb65e88e6c50c85d6f20501f7149ddd"
  ],
  "author": {
    "name": "Max Reitz",
    "email": "mreitz@redhat.com",
    "time": "Wed Oct 22 14:09:39 2014 +0200"
  },
  "committer": {
    "name": "Kevin Wolf",
    "email": "kwolf@redhat.com",
    "time": "Thu Oct 23 15:34:01 2014 +0200"
  },
  "message": "qcow2: Do not perform potentially damaging repairs\n\nIf a referenced cluster has a refcount of 0, increasing its refcount may\nresult in clusters being allocated for the refcount structures. This may\noverwrite the referenced cluster, therefore we cannot simply increase\nthe refcount then.\n\nIn such cases, we can either try to replicate all the refcount\noperations solely for the check operation, basing the allocations on the\nin-memory refcount table; or we can simply rebuild the whole refcount\nstructure based on the in-memory refcount table. Since the latter will\nbe much easier, do that.\n\nTo prepare for this, introduce a \"rebuild\" boolean which should be set\nto true whenever a fix is rather dangerous or too complicated using the\ncurrent refcount structures. Another example for this is refcount blocks\nbeing referenced more than once.\n\nSigned-off-by: Max Reitz \u003cmreitz@redhat.com\u003e\nSigned-off-by: Kevin Wolf \u003ckwolf@redhat.com\u003e\n",
  "tree_diff": [
    {
      "type": "modify",
      "old_id": "b87eafcc315938879b88fd9689858c29615ffcbb",
      "old_mode": 33188,
      "old_path": "block/qcow2-refcount.c",
      "new_id": "e96466635a820dd6b68ad78bec7a47723e6d58a8",
      "new_mode": 33188,
      "new_path": "block/qcow2-refcount.c"
    }
  ]
}