diff options
| author | Eric Dumazet <edumazet@google.com> | 2013-02-21 12:18:52 +0000 |
|---|---|---|
| committer | Ben Hutchings <ben@decadent.org.uk> | 2013-03-06 03:24:21 +0000 |
| commit | 0fd0ff7e1fcc4b4bc5d17ab1d200f23dea7c681d (patch) | |
| tree | b0633b26fe93a7d93385ce2744fcb19cecda82cd /include/net/inet_sock.h | |
| parent | 52430c06469c05c36dd688c8daff25e5bcfde8e9 (diff) | |
ipv6: use a stronger hash for tcp
[ Upstream commit 08dcdbf6a7b9d14c2302c5bd0c5390ddf122f664 ]
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>
Signed-off-by: Ben Hutchings <ben@decadent.org.uk>
Diffstat (limited to 'include/net/inet_sock.h')
| -rw-r--r-- | include/net/inet_sock.h | 1 |
1 files changed, 1 insertions, 0 deletions
diff --git a/include/net/inet_sock.h b/include/net/inet_sock.h index f941964a9931..ee4ee918de6c 100644 --- a/include/net/inet_sock.h +++ b/include/net/inet_sock.h @@ -199,6 +199,7 @@ static inline void inet_sk_copy_descendant(struct sock *sk_to, extern int inet_sk_rebuild_header(struct sock *sk); extern u32 inet_ehash_secret; +extern u32 ipv6_hash_secret; extern void build_ehash_secret(void); static inline unsigned int inet_ehashfn(struct net *net, |