Return-Path: <linux-kernel-owner@vger.kernel.org>
Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand
	id S1753336AbcDBMie (ORCPT <rfc822;w@1wt.eu>);
	Sat, 2 Apr 2016 08:38:34 -0400
Received: from mail-lb0-f179.google.com ([209.85.217.179]:35143 "EHLO
	mail-lb0-f179.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
	with ESMTP id S1752261AbcDBMic (ORCPT
	<rfc822;linux-kernel@vger.kernel.org>);
	Sat, 2 Apr 2016 08:38:32 -0400
Date: Sat, 2 Apr 2016 14:38:29 +0200
From: Frederic Weisbecker <fweisbec@gmail.com>
To: Peter Zijlstra <peterz@infradead.org>
Cc: LKML <linux-kernel@vger.kernel.org>,
        Byungchul Park <byungchul.park@lge.com>,
        Chris Metcalf <cmetcalf@ezchip.com>,
        Thomas Gleixner <tglx@linutronix.de>,
        Luiz Capitulino <lcapitulino@redhat.com>,
        Christoph Lameter <cl@linux.com>,
        "Paul E . McKenney" <paulmck@linux.vnet.ibm.com>,
        Mike Galbraith <efault@gmx.de>, Rik van Riel <riel@redhat.com>,
        Ingo Molnar <mingo@elte.hu>
Subject: Re: [PATCH 3/4] sched: Optimize tick periodic cpu load updates
Message-ID: <20160402123828.GC27395@lerouge>
References: <1459516987-15745-1-git-send-email-fweisbec@gmail.com>
 <1459516987-15745-4-git-send-email-fweisbec@gmail.com>
 <20160402072337.GF2906@worktop>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20160402072337.GF2906@worktop>
User-Agent: Mutt/1.5.24 (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
Content-Length: 1103
Lines: 38

On Sat, Apr 02, 2016 at 09:23:37AM +0200, Peter Zijlstra wrote:
> On Fri, Apr 01, 2016 at 03:23:06PM +0200, Frederic Weisbecker wrote:
> > Don't bother with the whole pending tickless cpu load machinery if
> > we run a tick periodic kernel. That's less job for the CPU on ticks.
> 
> Again, the changelog really could use help. Is this a pure optimization
> patch? If so, do you have numbers?

Well until now we have always tried to keep the nohz code under ifdef.
For optimizations and kernel size. I haven't measured it though, I guess
the gain is hardly visible.

> 
> > +++ b/kernel/sched/sched.h
> > @@ -585,8 +585,10 @@ struct rq {
> >  #endif
> >  	#define CPU_LOAD_IDX_MAX 5
> >  	unsigned long cpu_load[CPU_LOAD_IDX_MAX];
> > +#ifdef CONFIG_NO_HZ_COMMON
> > +# ifdef CONFIG_SMP
> 
> I'm not a fan of this #ifdef indenting and nothing near there uses this
> style, so please don't introduce it here.

Ok.

Thanks.

> 
> >  	unsigned long last_load_update_tick;
> > -#ifdef CONFIG_NO_HZ_COMMON
> > +# endif
> >  	u64 nohz_stamp;
> >  	unsigned long nohz_flags;
> >  #endif
> > -- 
> > 2.7.0
> >