From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <linux-kernel-owner@vger.kernel.org>
Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand
	id S1751718AbcFXNJk (ORCPT <rfc822;w@1wt.eu>);
	Fri, 24 Jun 2016 09:09:40 -0400
Received: from bombadil.infradead.org ([198.137.202.9]:60413 "EHLO
	bombadil.infradead.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
	with ESMTP id S1751552AbcFXNJi (ORCPT
	<rfc822;linux-kernel@vger.kernel.org>);
	Fri, 24 Jun 2016 09:09:38 -0400
Date: Fri, 24 Jun 2016 15:09:32 +0200
From: Peter Zijlstra <peterz@infradead.org>
To: Vincent Guittot <vincent.guittot@linaro.org>
Cc: Jirka Hladky <jhladky@redhat.com>,
        linux-kernel <linux-kernel@vger.kernel.org>,
        Ingo Molnar <mingo@redhat.com>, Kamil Kolakowski <kkolakow@redhat.com>,
        Morten Rasmussen <morten.rasmussen@arm.com>,
        Yuyang Du <yuyang.du@intel.com>,
        Dietmar Eggemann <dietmar.eggemann@arm.com>,
        Mike Galbraith <umgwanakikbuti@gmail.com>,
        Benjamin Segall <bsegall@google.com>, Paul Turner <pjt@google.com>,
        Matt Fleming <matt@codeblueprint.co.uk>
Subject: Re: Kernel 4.7rc3 - Performance drop 30-40% for SPECjbb2005 and
 SPECjvm2008 benchmarks against 4.6 kernel
Message-ID: <20160624130932.GA30909@twins.programming.kicks-ass.net>
References: <CAE4VaGDmCugUDS-fn9jz9UvBgGCZCp2nhENszrSVKr6t+0GnoQ@mail.gmail.com>
 <20160616172256.GJ30921@twins.programming.kicks-ass.net>
 <CAE4VaGCYvAbvNRD65SPfLPGVCspRfUYDAtfPt-VmkukPbt-L4Q@mail.gmail.com>
 <20160623183318.GN30154@twins.programming.kicks-ass.net>
 <20160623184318.GX30927@twins.programming.kicks-ass.net>
 <CAE4VaGBzNyhw58cASkQofCVaUHPUYrJ97_U8ymHMw7A9_0mjsg@mail.gmail.com>
 <20160624120250.GY30154@twins.programming.kicks-ass.net>
 <CAKfTPtBvCuWfzjn2wQPvbaGn786nL1t4_h1DA8XzO3J51_F+zQ@mail.gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <CAKfTPtBvCuWfzjn2wQPvbaGn786nL1t4_h1DA8XzO3J51_F+zQ@mail.gmail.com>
User-Agent: Mutt/1.5.23.1 (2014-03-12)
Sender: linux-kernel-owner@vger.kernel.org
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org

On Fri, Jun 24, 2016 at 02:44:07PM +0200, Vincent Guittot wrote:

> > diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
> > index 22d64b3f5876..d4f6fb2f3057 100644
> > --- a/kernel/sched/fair.c
> > +++ b/kernel/sched/fair.c
> > @@ -2484,7 +2484,7 @@ static inline long calc_tg_weight(struct task_group *tg, struct cfs_rq *cfs_rq)
> >          */
> >         tg_weight = atomic_long_read(&tg->load_avg);
> >         tg_weight -= cfs_rq->tg_load_avg_contrib;
> > -       tg_weight += cfs_rq->load.weight;
> > +       tg_weight += cfs_rq->avg.load_avg;
> 
> IIUC, you are reverting
> commit  fde7d22e01aa (sched/fair: Fix overly small weight for
> interactive group entities)

Ah!, I hadn't yet done a git-blame on this. Right you are, we should
have put a comment there.

So the problem here is that since commit:

  2159197d6677 ("sched/core: Enable increased load resolution on 64-bit kernels")

load.weight and avg.load_avg are in different metrics. Which completely
wrecked things.

The obvious alternative is using:

	scale_load_down(cfs_rq->load.weight);

Let me go run that through the benchmark.

> I have one question regarding the use of cfs_rq->avg.load_avg
> cfs_rq->tg_load_avg_contrib is the sampling of cfs_rq->avg.load_avg so
> I'm curious to understand why you use cfs_rq->avg.load_avg instead of
> keeping cfs_rq->tg_load_avg_contrib. Do you think that the sampling is
> not accurate enough to prevent any significant difference between both
> when we use tg->load_avg ?

I'm not entirely sure I understand your question; is it to the existence
of calc_tg_weight()? That is, why use calc_tg_weight() and not use
tg->load_avg as is?

It seemed like a simple and cheap way to increase accuracy, nothing more
behind it until the commit you referred to.