Received-SPF: pass (google.com: domain of linux-kernel+bounces-8661-linux.lists.archive=gmail.com@vger.kernel.org designates 147.75.199.223 as permitted sender) client-ip=147.75.199.223;
Precedence: bulk
MIME-Version: 1.0
References: <20231215052652.917741-1-imran.f.khan@oracle.com>
 <CAKfTPtD85OSem=7RMquLWokVp7gffvDaY3mtwevkxp1mSSVVqQ@mail.gmail.com>
 <d8507f86-2458-4b01-a774-5102473e657e@oracle.com> <1c13b0cc-098c-4f75-9f37-a53a66965409@oracle.com>
 <CAKfTPtDqqAYNCFb6US-4Yd=-xvj8mhMuO0PkUhWsqekxp__M4g@mail.gmail.com>
 <a9993f8b-6449-4452-a168-db966d9bae49@oracle.com> <CAKfTPtD1Q7R0ySc=a_5Aw9M0McQiom80DWEB6FRv7vYoPRVebg@mail.gmail.com>
 <20231221082105.GA600162@ziqianlu-desk2>
In-Reply-To: <20231221082105.GA600162@ziqianlu-desk2>
From: Vincent Guittot <vincent.guittot@linaro.org>
Date: Thu, 21 Dec 2023 16:26:30 +0100
Message-ID: <CAKfTPtDP6FXKa0ifpXd+jh5DBBBXmz7CcgaB7w8R5in1iQbTiw@mail.gmail.com>
Subject: Re: [RFC PATCH] sched: fair: reset task_group.load_avg when there are
 no running tasks.
To: Aaron Lu <aaron.lu@intel.com>
Cc: Imran Khan <imran.f.khan@oracle.com>, mingo@redhat.com, peterz@infradead.org, 
	juri.lelli@redhat.com, dietmar.eggemann@arm.com, rostedt@goodmis.org, 
	bsegall@google.com, mgorman@suse.de, bristot@redhat.com, vschneid@redhat.com, 
	linux-kernel@vger.kernel.org
Content-Type: text/plain; charset="UTF-8"

Hi Aaron,

On Thu, 21 Dec 2023 at 09:21, Aaron Lu <aaron.lu@intel.com> wrote:
>
> Hi Vincent,
>
> Thanks for the heads up.
>
> On Wed, Dec 20, 2023 at 06:31:08PM +0100, Vincent Guittot wrote:
> > Aaron,
> > Could you run the tests that you run for testing "Ratelimit update to
> > tg->load_avg" and check if this patch impacts your performance ?
>
> I run hackbench/netperf/postgres_sysbench with nr_thread=nr_cpu on a 2
> sockets/120cores/240cpus Intel server and didn't notice any performance
> change after applying your diff so I think it's not a problem.

Thanks for the tests.

I'm going to prepare a proper patch

>
> Regards,
> Aaron