[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <ZK6pJ8Z3VZOKKMyo@corigine.com>
Date: Wed, 12 Jul 2023 14:22:47 +0100
From: Simon Horman <simon.horman@...igine.com>
To: Jian Wen <wenjianhn@...il.com>
Cc: edumazet@...gle.com, davem@...emloft.net,
Jian Wen <wenjian1@...omi.com>, netdev@...r.kernel.org
Subject: Re: [PATCH v2 net-next] tcp: add a scheduling point in
established_get_first()
On Tue, Jul 11, 2023 at 11:24:05AM +0800, Jian Wen wrote:
> Kubernetes[1] is going to stick with /proc/net/tcp for a while.
>
> This commit reduces the scheduling latency introduced by
> established_get_first(), similar to commit acffb584cda7 ("net: diag:
> add a scheduling point in inet_diag_dump_icsk()").
>
> In our environment, the scheduling latency affects the performance of
> latency-sensitive services like Redis.
>
> Changes in V2 :
> - call cond_resched() before checking if a bucket is empty as
> suggested by Eric Dumazet
> - removed the delay of synchronize_net() from the commit message
>
> [1] https://github.com/google/cadvisor/blob/v0.47.2/container/libcontainer/handler.go#L130
>
> Signed-off-by: Jian Wen <wenjian1@...omi.com>
Reviewed-by: Simon Horman <simon.horman@...igine.com>
Powered by blists - more mailing lists