)]}'
{
  "commit": "4ed023ce2a39ab5812d33cf4d819def168965a7f",
  "tree": "27c97cac67d481e985e52f3b745ecae8cdb9f379",
  "parents": [
    "99e314ebca8c7b3450e4beaa95117c63d8f2f393"
  ],
  "author": {
    "name": "Dr. David Alan Gilbert",
    "email": "dgilbert@redhat.com",
    "time": "Thu Nov 05 18:11:16 2015 +0000"
  },
  "committer": {
    "name": "Juan Quintela",
    "email": "quintela@redhat.com",
    "time": "Tue Nov 10 15:00:28 2015 +0100"
  },
  "message": "Round up RAMBlock sizes to host page sizes\n\nRAMBlocks that are not a multiple of host pages in length\ncause problems for postcopy (I\u0027ve seen an ACPI table on aarch64\nbe 5k in length - i.e. 5x target-page), so round RAMBlock sizes\nup to a host-page.\n\nThis potentially breaks migration compatibility due to changes\nin RAMBlock sizes; however:\n   1) x86 and s390 I think always have host\u003dtarget page size\n   2) When I\u0027ve tried on Power the block sizes already seem aligned.\n   3) I don\u0027t think there\u0027s anything else that maintains per-version\n      machine-types for compatibility.\n\nSigned-off-by: Dr. David Alan Gilbert \u003cdgilbert@redhat.com\u003e\nReviewed-by: Juan Quintela \u003cquintela@redhat.com\u003e\nSigned-off-by: Juan Quintela \u003cquintela@redhat.com\u003e\n",
  "tree_diff": [
    {
      "type": "modify",
      "old_id": "36886eeba24ca4d6acc79fb561874fd68800a545",
      "old_mode": 33188,
      "old_path": "exec.c",
      "new_id": "b09f18b2a4d9e063929486cc700fad8075784d25",
      "new_mode": 33188,
      "new_path": "exec.c"
    }
  ]
}