[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CANn89iLBvwmRbbgw=U3z8k+i_S-ycSf7K-ow4rH5ZQP9CFJuWQ@mail.gmail.com>
Date: Fri, 22 Jan 2021 21:41:27 +0100
From: Eric Dumazet <edumazet@...gle.com>
To: Enke Chen <enkechen2020@...il.com>
Cc: "David S. Miller" <davem@...emloft.net>,
Jakub Kicinski <kuba@...nel.org>,
Alexey Kuznetsov <kuznet@....inr.ac.ru>,
Hideaki YOSHIFUJI <yoshfuji@...ux-ipv6.org>,
netdev <netdev@...r.kernel.org>,
LKML <linux-kernel@...r.kernel.org>,
Neal Cardwell <ncardwell@...gle.com>
Subject: Re: [PATCH net] tcp: make TCP_USER_TIMEOUT accurate for zero window probes
On Fri, Jan 22, 2021 at 8:13 PM Enke Chen <enkechen2020@...il.com> wrote:
>
> From: Enke Chen <enchen@...oaltonetworks.com>
>
> The TCP_USER_TIMEOUT is checked by the 0-window probe timer. As the
> timer has backoff with a max interval of about two minutes, the
> actual timeout for TCP_USER_TIMEOUT can be off by up to two minutes.
>
> In this patch the TCP_USER_TIMEOUT is made more accurate by taking it
> into account when computing the timer value for the 0-window probes.
>
> This patch is similar to the one that made TCP_USER_TIMEOUT accurate for
> RTOs in commit b701a99e431d ("tcp: Add tcp_clamp_rto_to_user_timeout()
> helper to improve accuracy").
>
> Signed-off-by: Enke Chen <enchen@...oaltonetworks.com>
> Reviewed-by: Neal Cardwell <ncardwell@...gle.com>
> ---
SGTM, thanks !
Signed-off-by: Eric Dumazet <edumazet@...gle.com>
Powered by blists - more mailing lists