ipv6: use a stronger hash for tcp
authorEric Dumazet <edumazet@google.com>
Thu, 21 Feb 2013 12:18:52 +0000 (12:18 +0000)
committerDavid S. Miller <davem@davemloft.net>
Thu, 21 Feb 2013 23:15:58 +0000 (18:15 -0500)
commit08dcdbf6a7b9d14c2302c5bd0c5390ddf122f664
treefc1d00005fda323296b93064416d255b81fdab73
parent0ab8a9f5fa94ac625c1f19acc48db299416c6d49
ipv6: use a stronger hash for tcp

It looks like its possible to open thousands of TCP IPv6
sessions on a server, all landing in a single slot of TCP hash
table. Incoming packets have to lookup sockets in a very
long list.

We should hash all bits from foreign IPv6 addresses, using
a salt and hash mix, not a simple XOR.

inet6_ehashfn() can also separately use the ports, instead
of xoring them.

Reported-by: Neal Cardwell <ncardwell@google.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Yuchung Cheng <ycheng@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
include/net/inet6_hashtables.h
include/net/inet_sock.h
include/net/ipv6.h
net/ipv4/af_inet.c