)]}'
{
  "commit": "1eddceadb0d6441cd39b2c38705a8f5fec86e770",
  "tree": "db2c9a019a74d71cf2aab0268a1d561d6c5bbd84",
  "parents": [
    "cab758ef30e0e40f783627abc4b66d1b48fecd49"
  ],
  "author": {
    "name": "Eric Dumazet",
    "email": "eric.dumazet@gmail.com",
    "time": "Fri Jun 17 03:45:15 2011 +0000"
  },
  "committer": {
    "name": "David S. Miller",
    "email": "davem@conan.davemloft.net",
    "time": "Fri Jun 17 15:27:31 2011 -0400"
  },
  "message": "net: rfs: enable RFS before first data packet is received\n\nLe jeudi 16 juin 2011 à 23:38 -0400, David Miller a écrit :\n\u003e From: Ben Hutchings \u003cbhutchings@solarflare.com\u003e\n\u003e Date: Fri, 17 Jun 2011 00:50:46 +0100\n\u003e\n\u003e \u003e On Wed, 2011-06-15 at 04:15 +0200, Eric Dumazet wrote:\n\u003e \u003e\u003e @@ -1594,6 +1594,7 @@ int tcp_v4_do_rcv(struct sock *sk, struct sk_buff *skb)\n\u003e \u003e\u003e  \t\t\tgoto discard;\n\u003e \u003e\u003e\n\u003e \u003e\u003e  \t\tif (nsk !\u003d sk) {\n\u003e \u003e\u003e +\t\t\tsock_rps_save_rxhash(nsk, skb-\u003erxhash);\n\u003e \u003e\u003e  \t\t\tif (tcp_child_process(sk, nsk, skb)) {\n\u003e \u003e\u003e  \t\t\t\trsk \u003d nsk;\n\u003e \u003e\u003e  \t\t\t\tgoto reset;\n\u003e \u003e\u003e\n\u003e \u003e\n\u003e \u003e I haven\u0027t tried this, but it looks reasonable to me.\n\u003e \u003e\n\u003e \u003e What about IPv6?  The logic in tcp_v6_do_rcv() looks very similar.\n\u003e\n\u003e Indeed ipv6 side needs the same fix.\n\u003e\n\u003e Eric please add that part and resubmit.  And in fact I might stick\n\u003e this into net-2.6 instead of net-next-2.6\n\u003e\n\nOK, here is the net-2.6 based one then, thanks !\n\n[PATCH v2] net: rfs: enable RFS before first data packet is received\n\nFirst packet received on a passive tcp flow is not correctly RFS\nsteered.\n\nOne sock_rps_record_flow() call is missing in inet_accept()\n\nBut before that, we also must record rxhash when child socket is setup.\n\nSigned-off-by: Eric Dumazet \u003ceric.dumazet@gmail.com\u003e\nCC: Tom Herbert \u003ctherbert@google.com\u003e\nCC: Ben Hutchings \u003cbhutchings@solarflare.com\u003e\nCC: Jamal Hadi Salim \u003chadi@cyberus.ca\u003e\nSigned-off-by: David S. Miller \u003cdavem@conan.davemloft.net\u003e\n",
  "tree_diff": [
    {
      "type": "modify",
      "old_id": "9c1926027a268162a3c37374a5f3397661c6dba6",
      "old_mode": 33188,
      "old_path": "net/ipv4/af_inet.c",
      "new_id": "eae1f676f870a8e219bcb3ee4a28453961604def",
      "new_mode": 33188,
      "new_path": "net/ipv4/af_inet.c"
    },
    {
      "type": "modify",
      "old_id": "a7d6671e33b8a6e46aba0b352ad083736d1b421b",
      "old_mode": 33188,
      "old_path": "net/ipv4/tcp_ipv4.c",
      "new_id": "708dc203b0348c3365b6ac3cd2e36559a069d0cc",
      "new_mode": 33188,
      "new_path": "net/ipv4/tcp_ipv4.c"
    },
    {
      "type": "modify",
      "old_id": "d1fd28711ba50e1ad8ac90a0c32df360ec911d70",
      "old_mode": 33188,
      "old_path": "net/ipv6/tcp_ipv6.c",
      "new_id": "87551ca568cd80e35671153b80acb5c6f6986fe8",
      "new_mode": 33188,
      "new_path": "net/ipv6/tcp_ipv6.c"
    }
  ]
}