Commit e3f2c4a3 authored Jan 03, 2018 by Soheil Hassas Yeganeh Committed by David S. Miller Jan 05, 2018

ip: do not set RFS core on error queue reads



We should only record RPS on normal reads and writes.
In single threaded processes, all calls record the same state. In
multi-threaded processes where a separate thread processes
errors, the RFS table mispredicts.

Note that, when CONFIG_RPS is disabled, sock_rps_record_flow
is a noop and no branch is added as a result of this patch.

Signed-off-by: Soheil Hassas Yeganeh <soheil@google.com>
Signed-off-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Neal Cardwell <ncardwell@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

parent 2e40b823

net/ipv4/af_inet.c

+2 −1

Original line number	Original line	Diff line number	Diff line
	@@ -790,6 +790,7 @@ int inet_recvmsg(struct socket sock, struct msghdr msg, size_t size,
	int addr_len = 0;		int addr_len = 0;
	int err;		int err;

			if (likely(!(flags & MSG_ERRQUEUE)))
	sock_rps_record_flow(sk);		sock_rps_record_flow(sk);

	err = sk->sk_prot->recvmsg(sk, msg, size, flags & MSG_DONTWAIT,		err = sk->sk_prot->recvmsg(sk, msg, size, flags & MSG_DONTWAIT,

Admin message