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 S966712AbcHBNou (ORCPT <rfc822;w@1wt.eu>);
	Tue, 2 Aug 2016 09:44:50 -0400
Received: from merlin.infradead.org ([205.233.59.134]:45151 "EHLO
	merlin.infradead.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
	with ESMTP id S966624AbcHBNlh (ORCPT
	<rfc822;linux-kernel@vger.kernel.org>);
	Tue, 2 Aug 2016 09:41:37 -0400
Date: Tue, 2 Aug 2016 13:24:17 +0200
From: Peter Zijlstra <peterz@infradead.org>
To: bsegall@google.com
Cc: linux-kernel@vger.kernel.org, stable@vger.kernel.org,
        Andrey Ryabinin <aryabinin@virtuozzo.com>,
        Chris Wilson <chris@chris-wilson.co.uk>,
        Linus Torvalds <torvalds@linux-foundation.org>,
        Mike Galbraith <efault@gmx.de>, Thomas Gleixner <tglx@linutronix.de>,
        Yuyang Du <yuyang.du@intel.com>, kernel@kyup.com,
        morten.rasmussen@arm.com, pjt@google.com, steve.muckle@linaro.org,
        Ingo Molnar <mingo@kernel.org>,
        Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Subject: Re: [PATCH 4.4 043/146] sched/fair: Fix cfs_rq avg tracking underflow
Message-ID: <20160802112417.GH6862@twins.programming.kicks-ass.net>
References: <20160725203521.340401316@linuxfoundation.org>
 <20160725203523.152875464@linuxfoundation.org>
 <xm26fuqxgyev.fsf@bsegall-linux.mtv.corp.google.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <xm26fuqxgyev.fsf@bsegall-linux.mtv.corp.google.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 Mon, Jul 25, 2016 at 03:12:24PM -0700, bsegall@google.com wrote:
> > @@ -2690,15 +2707,15 @@ static inline int update_cfs_rq_load_avg
> >  
> >  	if (atomic_long_read(&cfs_rq->removed_load_avg)) {
> >  		s64 r = atomic_long_xchg(&cfs_rq->removed_load_avg, 0);
> > -		sa->load_avg = max_t(long, sa->load_avg - r, 0);
> > -		sa->load_sum = max_t(s64, sa->load_sum - r * LOAD_AVG_MAX, 0);
> > +		sub_positive(&sa->load_avg, r);
> > +		sub_positive(&sa->load_sum, r * LOAD_AVG_MAX);
> >  		removed = 1;
> >  	}
> >  
> >  	if (atomic_long_read(&cfs_rq->removed_util_avg)) {
> >  		long r = atomic_long_xchg(&cfs_rq->removed_util_avg, 0);
> > -		sa->util_avg = max_t(long, sa->util_avg - r, 0);
> > -		sa->util_sum = max_t(s32, sa->util_sum - r * LOAD_AVG_MAX, 0);
> > +		sub_positive(&sa->util_avg, r);
> > +		sub_positive(&sa->util_sum, r * LOAD_AVG_MAX);
> >  	}
> >  
> >  	decayed = __update_load_avg(now, cpu_of(rq_of(cfs_rq)), sa,

> I missed this the first time around, and I have no problem with this
> backport, but it's not remotely obvious that update_cfs_rq_h_load() is
> intended to be something approaching racing-safe when not under
> rq->lock. (And given only somewhat-adversarial compilers I agree that it
> probably won't do any worse than skip updates, though I certainly won't
> swear to it)

Right, this is somewhat 'new' because we now use the avg load value for
the load-balancer, which is entirely unserialized.