[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20201023191348.7a6003f4@kicinski-fedora-PC1C0HJN.hsd1.ca.comcast.net>
Date: Fri, 23 Oct 2020 19:13:52 -0700
From: Jakub Kicinski <kuba@...nel.org>
To: Arjun Roy <arjunroy.kdev@...il.com>
Cc: davem@...emloft.net, netdev@...r.kernel.org, arjunroy@...gle.com,
edumazet@...gle.com, soheil@...gle.com, ncardwell@...gle.com
Subject: Re: [net v2] tcp: Prevent low rmem stalls with SO_RCVLOWAT.
On Fri, 23 Oct 2020 11:47:09 -0700 Arjun Roy wrote:
> From: Arjun Roy <arjunroy@...gle.com>
>
> With SO_RCVLOWAT, under memory pressure,
> it is possible to enter a state where:
>
> 1. We have not received enough bytes to satisfy SO_RCVLOWAT.
> 2. We have not entered buffer pressure (see tcp_rmem_pressure()).
> 3. But, we do not have enough buffer space to accept more packets.
>
> In this case, we advertise 0 rwnd (due to #3) but the application does
> not drain the receive queue (no wakeup because of #1 and #2) so the
> flow stalls.
>
> Modify the heuristic for SO_RCVLOWAT so that, if we are advertising
> rwnd<=rcv_mss, force a wakeup to prevent a stall.
>
> Without this patch, setting tcp_rmem to 6143 and disabling TCP
> autotune causes a stalled flow. With this patch, no stall occurs. This
> is with RPC-style traffic with large messages.
>
> Fixes: 03f45c883c6f ("tcp: avoid extra wakeups for SO_RCVLOWAT users")
> Signed-off-by: Arjun Roy <arjunroy@...gle.com>
> Acked-by: Soheil Hassas Yeganeh <soheil@...gle.com>
> Acked-by: Neal Cardwell <ncardwell@...gle.com>
> Signed-off-by: Eric Dumazet <edumazet@...gle.com>
Applied, thank you!
Powered by blists - more mailing lists