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 S1751606AbcETVYM (ORCPT <rfc822;w@1wt.eu>);
	Fri, 20 May 2016 17:24:12 -0400
Received: from mail-wm0-f46.google.com ([74.125.82.46]:35384 "EHLO
	mail-wm0-f46.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
	with ESMTP id S1750836AbcETVYJ (ORCPT
	<rfc822;linux-kernel@vger.kernel.org>);
	Fri, 20 May 2016 17:24:09 -0400
Date: Fri, 20 May 2016 22:24:05 +0100
From: Matt Fleming <matt@codeblueprint.co.uk>
To: Peter Zijlstra <peterz@infradead.org>
Cc: mingo@kernel.org, linux-kernel@vger.kernel.org,
        Pavan Kondeti <pkondeti@codeaurora.org>,
        Ben Segall <bsegall@google.com>,
        Morten Rasmussen <morten.rasmussen@arm.com>,
        Paul Turner <pjt@google.com>, Thomas Gleixner <tglx@linutronix.de>,
        byungchul.park@lge.com, Andrew Hunter <ahh@google.com>,
        Mike Galbraith <mgalbraith@suse.de>,
        Mel Gorman <mgorman@techsingularity.net>
Subject: Re: [PATCH 2/3] sched,fair: Fix local starvation
Message-ID: <20160520212405.GL21993@codeblueprint.co.uk>
References: <20160510174314.355953085@infradead.org>
 <20160510174613.902178264@infradead.org>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20160510174613.902178264@infradead.org>
User-Agent: Mutt/1.5.24+41 (02bc14ed1569) (2015-08-30)
Sender: linux-kernel-owner@vger.kernel.org
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org

On Tue, 10 May, at 07:43:16PM, Peter Zijlstra wrote:
> Mike reported that the recent commit 3a47d5124a95 ("sched/fair: Fix
> fairness issue on migration") broke interactivity and the signal
> starve test.
> 
> The problem is that I assumed ENQUEUE_WAKING was only set when we do a
> cross-cpu wakeup (migration), which isn't true. This means we now
> destroy the vruntime history of tasks and wakeup-preemption suffers.
> 
> Cure this by making my assumption true, only call
> sched_class::task_waking() when we do a cross-cpu wakeup. This avoids
> the indirect call in the case we do a local wakeup.
> 
> Cc: Pavan Kondeti <pkondeti@codeaurora.org>
> Cc: Ben Segall <bsegall@google.com>
> Cc: Matt Fleming <matt@codeblueprint.co.uk>
> Cc: Morten Rasmussen <morten.rasmussen@arm.com>
> Cc: Paul Turner <pjt@google.com>
> Cc: Thomas Gleixner <tglx@linutronix.de>
> Cc: byungchul.park@lge.com
> Cc: Andrew Hunter <ahh@google.com>
> Fixes: 3a47d5124a95 ("sched/fair: Fix fairness issue on migration")
> Reported-by: Mike Galbraith <mgalbraith@suse.de>
> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
> ---
>  kernel/sched/core.c |   29 +++++++++++++++++++++--------
>  kernel/sched/fair.c |   41 ++++++++++++++++++++++++++++++++++++++++-
>  2 files changed, 61 insertions(+), 9 deletions(-)
 
This patch appears to cause a regression for hackbench -pipe of
between ~8% and ~10% with groups >= NR_CPU.

I haven't probed much yet, but it looks like the vruntime of tasks has
gone nuts.