From 726587dc2adf3287fcdd02aa39a10f136332503f Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 12 Dec 2019 12:55:31 -0800 Subject: [PATCH] tcp: refine rule to allow EPOLLOUT generation under mem pressure At the time commit dcfac0c9d509 ("tcp: ensure epoll edge trigger wakeup when write queue is empty") was added to the kernel, we still had a single write queue, combining rtx and write queues. Once we moved the rtx queue into a separate rb-tree, testing if sk_write_queue is empty has been suboptimal. Indeed, if we have packets in the rtx queue, we probably want to delay the EPOLLOUT generation at the time incoming packets will free them, making room, but more importantly avoiding flooding application with EPOLLOUT events. Solution is to use tcp_rtx_and_write_queues_empty() helper. Fixes: 42899b27ca39 ("tcp: implement rb-tree based retransmit queue") Signed-off-by: Eric Dumazet Cc: Jason Baron Cc: Neal Cardwell Acked-by: Soheil Hassas Yeganeh Signed-off-by: Jakub Kicinski --- net/ipv4/tcp.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 8a39ee7948919..716938313a325 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -1087,8 +1087,7 @@ do_error: goto out; out_err: /* make sure we wake any epoll edge trigger waiter */ - if (unlikely(skb_queue_len(&sk->sk_write_queue) == 0 && - err == -EAGAIN)) { + if (unlikely(tcp_rtx_and_write_queues_empty(sk) && err == -EAGAIN)) { sk->sk_write_space(sk); tcp_chrono_stop(sk, TCP_CHRONO_SNDBUF_LIMITED); } @@ -1419,8 +1418,7 @@ out_err: sock_zerocopy_put_abort(uarg, true); err = sk_stream_error(sk, flags, err); /* make sure we wake any epoll edge trigger waiter */ - if (unlikely(skb_queue_len(&sk->sk_write_queue) == 0 && - err == -EAGAIN)) { + if (unlikely(tcp_rtx_and_write_queues_empty(sk) && err == -EAGAIN)) { sk->sk_write_space(sk); tcp_chrono_stop(sk, TCP_CHRONO_SNDBUF_LIMITED); } -- 2.39.5