)]}'
{
  "commit": "1d41b0c1ec66d38355a1e76c29dd2200433335f6",
  "tree": "0b088fd96cff908d0441d19823437333a784504b",
  "parents": [
    "f5436dd96aea2ec937964230831f241ebd3b658b"
  ],
  "author": {
    "name": "Anthony Liguori",
    "email": "aliguori@us.ibm.com",
    "time": "Thu Oct 22 17:43:48 2009 +0100"
  },
  "committer": {
    "name": "Anthony Liguori",
    "email": "aliguori@us.ibm.com",
    "time": "Tue Oct 27 12:29:04 2009 -0500"
  },
  "message": "Work around dhclient brokenness\n\nWith the latest GSO/csum offload patches, any guest using an unpatched version\nof dhclient (any Ubuntu guest, for instance), will no longer be able to get\na DHCP address.\n\ndhclient is actually at fault here.  It uses AF_PACKET to receive DHCP responses\nbut does not check auxdata to see if the packet has a valid csum.  This causes\nit to throw out the DHCP responses it gets from the virtio interface as there\nis not a valid checksum.\n\nFedora has carried a patch to fix their dhclient (it\u0027s needed for Xen too) but\nthis patch has not made it into a release of dhclient.  AFAIK, the patch is in\nthe dhclient CVS but I cannot confirm since their CVS is not public.\n\nThis patch, suggested by Rusty, looks for UDP packets (of a normal MTU) and\nexplicitly adds a checksum to them if they are missing one.\n\nThis allows unpatched dhclients to continue to work without needing to update\nthe guest kernels.\n\nSigned-off-by: Anthony Liguori \u003caliguori@us.ibm.com\u003e\nSigned-off-by: Mark McLoughlin \u003cmarkmc@redhat.com\u003e\nSigned-off-by: Anthony Liguori \u003caliguori@us.ibm.com\u003e\n",
  "tree_diff": [
    {
      "type": "modify",
      "old_id": "5e9520de7fd5d9412b38a41a9d2f95e3968e8be7",
      "old_mode": 33188,
      "old_path": "hw/virtio-net.c",
      "new_id": "064ec2e609ab9c89198458a2f0b5904198b3ffca",
      "new_mode": 33188,
      "new_path": "hw/virtio-net.c"
    }
  ]
}