netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* [PATCH] tcp: fix tcp_grow_window() for large incoming frames
@ 2012-04-17  9:28 Eric Dumazet
  2012-04-17 12:40 ` Neal Cardwell
  2012-04-18  2:33 ` David Miller
  0 siblings, 2 replies; 3+ messages in thread
From: Eric Dumazet @ 2012-04-17  9:28 UTC (permalink / raw)
  To: David Miller; +Cc: Neal Cardwell, Tom Herbert, netdev

From: Eric Dumazet <edumazet@google.com>

tcp_grow_window() has to grow rcv_ssthresh up to window_clamp, allowing
sender to increase its window.

tcp_grow_window() still assumes a tcp frame is under MSS, but its no
longer true with LRO/GRO.

This patch fixes one of the performance issue we noticed with GRO on.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Neal Cardwell <ncardwell@google.com>
Cc: Tom Herbert <therbert@google.com>
---
Sorry for the delay, this was discussed a long time ago.

 net/ipv4/tcp_input.c |    1 +
 1 file changed, 1 insertion(+)

diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index 9944c1d..3ff36406 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -335,6 +335,7 @@ static void tcp_grow_window(struct sock *sk, const struct sk_buff *skb)
 			incr = __tcp_grow_window(sk, skb);
 
 		if (incr) {
+			incr = max_t(int, incr, 2 * skb->len);
 			tp->rcv_ssthresh = min(tp->rcv_ssthresh + incr,
 					       tp->window_clamp);
 			inet_csk(sk)->icsk_ack.quick |= 1;

^ permalink raw reply related	[flat|nested] 3+ messages in thread

* Re: [PATCH] tcp: fix tcp_grow_window() for large incoming frames
  2012-04-17  9:28 [PATCH] tcp: fix tcp_grow_window() for large incoming frames Eric Dumazet
@ 2012-04-17 12:40 ` Neal Cardwell
  2012-04-18  2:33 ` David Miller
  1 sibling, 0 replies; 3+ messages in thread
From: Neal Cardwell @ 2012-04-17 12:40 UTC (permalink / raw)
  To: Eric Dumazet; +Cc: David Miller, Tom Herbert, netdev

On Tue, Apr 17, 2012 at 5:28 AM, Eric Dumazet <eric.dumazet@gmail.com> wrote:
> From: Eric Dumazet <edumazet@google.com>
>
> tcp_grow_window() has to grow rcv_ssthresh up to window_clamp, allowing
> sender to increase its window.
>
> tcp_grow_window() still assumes a tcp frame is under MSS, but its no
> longer true with LRO/GRO.
>
> This patch fixes one of the performance issue we noticed with GRO on.
>
> Signed-off-by: Eric Dumazet <edumazet@google.com>
> Cc: Neal Cardwell <ncardwell@google.com>
> Cc: Tom Herbert <therbert@google.com>
> ---
> Sorry for the delay, this was discussed a long time ago.
>
>  net/ipv4/tcp_input.c |    1 +
>  1 file changed, 1 insertion(+)

Acked-by: Neal Cardwell <ncardwell@google.com>

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: [PATCH] tcp: fix tcp_grow_window() for large incoming frames
  2012-04-17  9:28 [PATCH] tcp: fix tcp_grow_window() for large incoming frames Eric Dumazet
  2012-04-17 12:40 ` Neal Cardwell
@ 2012-04-18  2:33 ` David Miller
  1 sibling, 0 replies; 3+ messages in thread
From: David Miller @ 2012-04-18  2:33 UTC (permalink / raw)
  To: eric.dumazet; +Cc: ncardwell, therbert, netdev

From: Eric Dumazet <eric.dumazet@gmail.com>
Date: Tue, 17 Apr 2012 11:28:07 +0200

> From: Eric Dumazet <edumazet@google.com>
> 
> tcp_grow_window() has to grow rcv_ssthresh up to window_clamp, allowing
> sender to increase its window.
> 
> tcp_grow_window() still assumes a tcp frame is under MSS, but its no
> longer true with LRO/GRO.
> 
> This patch fixes one of the performance issue we noticed with GRO on.
> 
> Signed-off-by: Eric Dumazet <edumazet@google.com>

Applied.

^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2012-04-18  2:33 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2012-04-17  9:28 [PATCH] tcp: fix tcp_grow_window() for large incoming frames Eric Dumazet
2012-04-17 12:40 ` Neal Cardwell
2012-04-18  2:33 ` David Miller

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).