Commit 8ba3c9d1 authored Sep 14, 2020 by Soheil Hassas Yeganeh Committed by David S. Miller Sep 14, 2020

tcp: return EPOLLOUT from tcp_poll only when notsent_bytes is half the limit



If there was any event available on the TCP socket, tcp_poll()
will be called to retrieve all the events.  In tcp_poll(), we call
sk_stream_is_writeable() which returns true as long as we are at least
one byte below notsent_lowat.  This will result in quite a few
spurious EPLLOUT and frequent tiny sendmsg() calls as a result.

Similar to sk_stream_write_space(), use __sk_stream_is_writeable
with a wake value of 1, so that we set EPOLLOUT only if half the
space is available for write.

Signed-off-by: Soheil Hassas Yeganeh <soheil@google.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

parent ed6d9b02

net/ipv4/tcp.c

+2 −2

Original line number	Diff line number	Diff line
		@@ -564,7 +564,7 @@ __poll_t tcp_poll(struct file file, struct socket sock, poll_table *wait)
		mask \|= EPOLLIN \| EPOLLRDNORM;

		if (!(sk->sk_shutdown & SEND_SHUTDOWN)) {
		if (sk_stream_is_writeable(sk)) {
		if (__sk_stream_is_writeable(sk, 1)) {
		mask \|= EPOLLOUT \| EPOLLWRNORM;
		} else { /* send SIGIO later */
		sk_set_bit(SOCKWQ_ASYNC_NOSPACE, sk);
		@@ -576,7 +576,7 @@ __poll_t tcp_poll(struct file file, struct socket sock, poll_table *wait)
		* pairs with the input side.
		*/
		smp_mb__after_atomic();
		if (sk_stream_is_writeable(sk))
		if (__sk_stream_is_writeable(sk, 1))
		mask \|= EPOLLOUT \| EPOLLWRNORM;
		}
		} else

Admin message