Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Date:   Fri, 26 Apr 2019 14:40:40 +1000
From:   Nicholas Piggin <npiggin@gmail.com>
Subject: Re: [PATCH] kernel/sched: run nohz idle load balancer on HK_FLAG_MISC
 CPUs
To:     Peter Zijlstra <peterz@infradead.org>
Cc:     Frederic Weisbecker <fweisbec@gmail.com>,
        linux-kernel@vger.kernel.org, Ingo Molnar <mingo@redhat.com>,
        Thomas Gleixner <tglx@linutronix.de>
References: <20190412042613.28930-1-npiggin@gmail.com>
        <20190425115620.GQ4038@hirez.programming.kicks-ass.net>
In-Reply-To: <20190425115620.GQ4038@hirez.programming.kicks-ass.net>
MIME-Version: 1.0
User-Agent: astroid/0.14.0 (https://github.com/astroidmail/astroid)
Message-Id: <1556253219.geomfr4bo2.astroid@bobo.none>
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

Peter Zijlstra's on April 25, 2019 9:56 pm:
> On Fri, Apr 12, 2019 at 02:26:13PM +1000, Nicholas Piggin wrote:
>> The nohz idle balancer runs on the lowest idle CPU. This can
>> interfere with isolated CPUs, so confine it to HK_FLAG_MISC
>> housekeeping CPUs.
>>=20
>> HK_FLAG_SCHED is not used for this because it is not set anywhere
>> at the moment. This could be folded into HK_FLAG_SCHED once that
>> option is fixed.
>=20
> Frederic? Anyway, I thnk I'll take this patch as is.

That would be great, thanks. We've been testing it in a staging
environment (this is where they noticed the noise in the first
place), and results have been as expected:

  I've been able to test Nick's idle-loop load balancer (ILB) patch,=20
  with and without the TEO cpuidle governor. With the ILB patch (and=20
  nohz_full) I get a very quiet noise profile with either cpuidle=20
  governor (menu or teo). For my tests, I don't see a meaningful=20
  difference between the two governors.

  [...]

  Bottom line: Nick's patch that constrains the ILB to run on non-nohz=20
  cores has a noticeable noise-reduction effect. For this type of=20
  workload, the choice of cpuidle governor, menu or teo, is immaterial.

This is against a slightly backported RHEL kernel they are using, but
no significant differences from upstream in these areas.

Thanks,
Nick

=