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>

1 files changed, 1 insertions, 3 deletions

diff --git a/net/dccp/minisocks.c b/net/dccp/minisocks.c
index c69eb9c4fbb8..b50dc436db1f 100644
--- a/net/dccp/minisocks.c
+++ b/net/dccp/minisocks.c
@@ -55,11 +55,9 @@ void dccp_time_wait(struct sock *sk, int state, int timeo)
                 const int rto = (icsk->icsk_rto << 2) - (icsk->icsk_rto >> 1);
 #if IS_ENABLED(CONFIG_IPV6)
                 if (tw->tw_family == PF_INET6) {
-                        const struct ipv6_pinfo *np = inet6_sk(sk);
-
                         tw->tw_v6_daddr = sk->sk_v6_daddr;
                         tw->tw_v6_rcv_saddr = sk->sk_v6_rcv_saddr;
-                        tw->tw_ipv6only = np->ipv6only;
+                        tw->tw_ipv6only = sk->sk_ipv6only;
                 }
 #endif
                 /* Linkage updates. */