From 2276f58ac5890e58d2b6a48b95493faff7347e3a Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Tue, 16 May 2017 11:20:14 +0200 Subject: udp: use a separate rx queue for packet reception under udp flood the sk_receive_queue spinlock is heavily contended. This patch try to reduce the contention on such lock adding a second receive queue to the udp sockets; recvmsg() looks first in such queue and, only if empty, tries to fetch the data from sk_receive_queue. The latter is spliced into the newly added queue every time the receive path has to acquire the sk_receive_queue lock. The accounting of forward allocated memory is still protected with the sk_receive_queue lock, so udp_rmem_release() needs to acquire both locks when the forward deficit is flushed. On specific scenarios we can end up acquiring and releasing the sk_receive_queue lock multiple times; that will be covered by the next patch Suggested-by: Eric Dumazet Signed-off-by: Paolo Abeni Acked-by: Eric Dumazet Signed-off-by: David S. Miller --- include/linux/udp.h | 3 +++ 1 file changed, 3 insertions(+) (limited to 'include/linux/udp.h') diff --git a/include/linux/udp.h b/include/linux/udp.h index 6cb4061a720d..eaea63bc79bb 100644 --- a/include/linux/udp.h +++ b/include/linux/udp.h @@ -80,6 +80,9 @@ struct udp_sock { struct sk_buff *skb, int nhoff); + /* udp_recvmsg try to use this before splicing sk_receive_queue */ + struct sk_buff_head reader_queue ____cacheline_aligned_in_smp; + /* This field is dirtied by udp_recvmsg() */ int forward_deficit; }; -- cgit v1.2.3-70-g09d2