[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-Id: <20180401.205644.1807066350495572745.davem@davemloft.net>
Date: Sun, 01 Apr 2018 20:56:44 -0400 (EDT)
From: David Miller <davem@...emloft.net>
To: saeedm@...lanox.com
Cc: netdev@...r.kernel.org, talgi@...lanox.com
Subject: Re: [PATCH net 1/1] net/mlx5e: Set EQE based as default TX
interrupt moderation mode
From: Saeed Mahameed <saeedm@...lanox.com>
Date: Fri, 30 Mar 2018 15:50:08 -0700
> From: Tal Gilboa <talgi@...lanox.com>
>
> The default TX moderation mode was mistakenly set to CQE based. The
> intention was to add a control ability in order to improve some specific
> use-cases. In general, we prefer to use EQE based moderation as it gives
> much better numbers for the common cases.
>
> CQE based causes a degradation in the common case since it resets the
> moderation timer on CQE generation. This causes an issue when TSO is
> well utilized (large TSO sessions). The timer is set to 16us so traffic
> of ~64KB TSO sessions per second would mean timer reset (CQE per TSO
> session -> long time between CQEs). In this case we quickly reach the
> tcp_limit_output_bytes (256KB by default) and cause a halt in TX traffic.
>
> By setting EQE based moderation we make sure timer would expire after
> 16us regardless of the packet rate.
> This fixes an up to 40% packet rate and up to 23% bandwidth degradtions.
>
> Fixes: 0088cbbc4b66 ("net/mlx5e: Enable CQE based moderation on TX CQ")
> Signed-off-by: Tal Gilboa <talgi@...lanox.com>
> Signed-off-by: Saeed Mahameed <saeedm@...lanox.com>
Applied and queued up for -stable.
Powered by blists - more mailing lists