diff options
| author | Eric Dumazet <edumazet@google.com> | 2014-06-27 08:36:16 -0700 |
|---|---|---|
| committer | David S. Miller <davem@davemloft.net> | 2014-07-01 23:46:21 -0700 |
| commit | 9fe516ba3fb29b6f6a752ffd93342fdee500ec01 (patch) | |
| tree | 4254ea02ed5d51d92fd7109d33d28bc95b8d3fb1 /net/ipv6/af_inet6.c | |
| parent | 090cce42639c2c82bce2cc26a902e4c674c62882 (diff) | |
| download | linux-9fe516ba3fb29b6f6a752ffd93342fdee500ec01.tar.gz linux-9fe516ba3fb29b6f6a752ffd93342fdee500ec01.tar.bz2 linux-9fe516ba3fb29b6f6a752ffd93342fdee500ec01.zip | |
inet: move ipv6only in sock_common
When an UDP application switches from AF_INET to AF_INET6 sockets, we
have a small performance degradation for IPv4 communications because of
extra cache line misses to access ipv6only information.
This can also be noticed for TCP listeners, as ipv6_only_sock() is also
used from __inet_lookup_listener()->compute_score()
This is magnified when SO_REUSEPORT is used.
Move ipv6only into struct sock_common so that it is available at
no extra cost in lookups.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Diffstat (limited to 'net/ipv6/af_inet6.c')
| -rw-r--r-- | net/ipv6/af_inet6.c | 6 |
1 files changed, 3 insertions, 3 deletions
diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c index 7cb4392690dd..a426cd7099bb 100644 --- a/net/ipv6/af_inet6.c +++ b/net/ipv6/af_inet6.c @@ -197,7 +197,7 @@ lookup_protocol: np->mcast_hops = IPV6_DEFAULT_MCASTHOPS; np->mc_loop = 1; np->pmtudisc = IPV6_PMTUDISC_WANT; - np->ipv6only = net->ipv6.sysctl.bindv6only; + sk->sk_ipv6only = net->ipv6.sysctl.bindv6only; /* Init the ipv4 part of the socket since we can have sockets * using v6 API for ipv4. @@ -294,7 +294,7 @@ int inet6_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len) /* Binding to v4-mapped address on a v6-only socket * makes no sense */ - if (np->ipv6only) { + if (sk->sk_ipv6only) { err = -EINVAL; goto out; } @@ -371,7 +371,7 @@ int inet6_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len) if (addr_type != IPV6_ADDR_ANY) { sk->sk_userlocks |= SOCK_BINDADDR_LOCK; if (addr_type != IPV6_ADDR_MAPPED) - np->ipv6only = 1; + sk->sk_ipv6only = 1; } if (snum) sk->sk_userlocks |= SOCK_BINDPORT_LOCK; |