[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-ID: <20200514100814.GH3758@techsingularity.net>
Date: Thu, 14 May 2020 11:08:14 +0100
From: Mel Gorman <mgorman@...hsingularity.net>
To: Jirka Hladky <jhladky@...hat.com>
Cc: Phil Auld <pauld@...hat.com>,
Peter Zijlstra <peterz@...radead.org>,
Ingo Molnar <mingo@...nel.org>,
Vincent Guittot <vincent.guittot@...aro.org>,
Juri Lelli <juri.lelli@...hat.com>,
Dietmar Eggemann <dietmar.eggemann@....com>,
Steven Rostedt <rostedt@...dmis.org>,
Ben Segall <bsegall@...gle.com>,
Valentin Schneider <valentin.schneider@....com>,
Hillf Danton <hdanton@...a.com>,
LKML <linux-kernel@...r.kernel.org>,
Douglas Shakshober <dshaks@...hat.com>,
Waiman Long <longman@...hat.com>,
Joe Mario <jmario@...hat.com>, Bill Gray <bgray@...hat.com>
Subject: Re: [PATCH 00/13] Reconcile NUMA balancing decisions with the load
balancer v6
On Thu, May 14, 2020 at 11:58:36AM +0200, Jirka Hladky wrote:
> Thank you, Mel!
>
> We are using netperf as well, but AFAIK it's running on two different
> hosts. I will check with colleagues, if they can
> add network-netperf-unbound run on the localhost.
>
> Is this the right config?
> https://github.com/gormanm/mmtests/blob/345f82bee77cbf09ba57f470a1cfc1ae413c97df/bin/generate-generic-configs
> sed -e 's/NETPERF_BUFFER_SIZES=.*/NETPERF_BUFFER_SIZES=64/'
> config-network-netperf-unbound > config-network-netperf-unbound-small
>
That's one I was using at the moment to have a quick test after
the reconciliation series was completed. It has since changed to
config-network-netperf-cstate-small-cross-socket to limit cstates, bind
the client and server to two local CPUs and using one buffer size. It
was necessary to get an ftrace function graph of the wakeup path that
was readable and not too noisy due to migrations, cpuidle exit costs etc.
--
Mel Gorman
SUSE Labs
Powered by blists - more mailing lists