)]}'
{
  "commit": "ca8d9ea30bc79b2965a1d169dcb2f48f02af4d2d",
  "tree": "f4bc15ecbd4a2604621e8c9d6bb2241bd206c6a6",
  "parents": [
    "d160439e7753aa19092bb12d8f4c6c918f8267e0"
  ],
  "author": {
    "name": "Andy Gospodarek",
    "email": "andy@greyhouse.net",
    "time": "Wed Jan 06 12:56:37 2010 +0000"
  },
  "committer": {
    "name": "David S. Miller",
    "email": "davem@davemloft.net",
    "time": "Thu Jan 07 00:46:39 2010 -0800"
  },
  "message": "fix bonding: allow arp_ip_targets on separate vlans to use arp validation\n\nOn Wed, Jan 06, 2010 at 10:10:03PM +0100, Eric Dumazet wrote:\n\u003e Le 06/01/2010 19:38, Eric Dumazet a écrit :\n\u003e \u003e\n\u003e \u003e (net-next-2.6 doesnt work well on my bond/vlan setup, I suspect I need a bisection)\n\u003e\n\u003e David, I had to revert 1f3c8804acba841b5573b953f5560d2683d2db0d\n\u003e (bonding: allow arp_ip_targets on separate vlans to use arp validation)\n\u003e\n\u003e Or else, my vlan devices dont work (unfortunatly I dont have much time\n\u003e these days to debug the thing)\n\u003e\n\u003e My config :\n\u003e\n\u003e               +---------+\n\u003e vlan.103 -----+ bond0   +--- eth1 (bnx2)\n\u003e               |         +\n\u003e vlan.825 -----+         +--- eth2 (tg3)\n\u003e               +---------+\n\u003e\n\u003e $ cat /proc/net/bonding/bond0\n\u003e Ethernet Channel Bonding Driver: v3.6.0 (September 26, 2009)\n\u003e\n\u003e Bonding Mode: fault-tolerance (active-backup)\n\u003e Primary Slave: None\n\u003e Currently Active Slave: eth2\n\u003e MII Status: up\n\u003e MII Polling Interval (ms): 100\n\u003e Up Delay (ms): 0\n\u003e Down Delay (ms): 0\n\u003e\n\u003e Slave Interface: eth1  (bnx2)\n\u003e MII Status: down\n\u003e Link Failure Count: 1\n\u003e Permanent HW addr: 00:1e:0b:ec:d3:d2\n\u003e\n\u003e Slave Interface: eth2   (tg3)\n\u003e MII Status: up\n\u003e Link Failure Count: 0\n\u003e Permanent HW addr: 00:1e:0b:92:78:50\n\u003e\n\nThis patch fixes up a problem with found with commit\n1f3c8804acba841b5573b953f5560d2683d2db0d.  The original change\noverloaded null_or_orig, but doing that prevented any packet handlers\nthat were not tied to a specific device (i.e. ptype-\u003edev \u003d\u003d NULL) from\never receiving any frames.\n\nThe null_or_orig variable cannot be overloaded, and must be kept as NULL\nto prevent the frame from being ignored by packet handlers designed to\naccept frames on any interface.\n\nSigned-off-by: Andy Gospodarek \u003candy@greyhouse.net\u003e\nSigned-off-by: Jay Vosburgh \u003cfubar@us.ibm.com\u003e\nAcked-by: Eric Dumazet \u003ceric.dumazet@gmail.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n",
  "tree_diff": [
    {
      "type": "modify",
      "old_id": "f9aa699ab6cb3aefae3248062a7259e9f3689819",
      "old_mode": 33188,
      "old_path": "net/core/dev.c",
      "new_id": "d9ab9be0c323ca653bfadb159a6f1c96c2d47623",
      "new_mode": 33188,
      "new_path": "net/core/dev.c"
    }
  ]
}