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 S1755444AbcFQJ56 (ORCPT <rfc822;w@1wt.eu>);
	Fri, 17 Jun 2016 05:57:58 -0400
Received: from mga09.intel.com ([134.134.136.24]:63659 "EHLO mga09.intel.com"
	rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP
	id S1753362AbcFQJ55 (ORCPT <rfc822;linux-kernel@vger.kernel.org>);
	Fri, 17 Jun 2016 05:57:57 -0400
X-ExtLoop1: 1
X-IronPort-AV: E=Sophos;i="5.26,483,1459839600"; 
   d="scan'208";a="829994597"
Date: Fri, 17 Jun 2016 10:01:09 +0800
From: Yuyang Du <yuyang.du@intel.com>
To: Peter Zijlstra <peterz@infradead.org>
Cc: Chris Wilson <chris@chris-wilson.co.uk>,
        Andrey Ryabinin <aryabinin@virtuozzo.com>,
        Linus Torvalds <torvalds@linux-foundation.org>,
        Mike Galbraith <efault@gmx.de>, Thomas Gleixner <tglx@linutronix.de>,
        bsegall@google.com, morten.rasmussen@arm.com, pjt@google.com,
        steve.muckle@linaro.org, linux-kernel@vger.kernel.org, kernel@kyup.com
Subject: Re: [PATCH] sched/fair: Fix cfs_rq avg tracking underflow
Message-ID: <20160617020109.GQ8105@intel.com>
References: <20160609090142.GS32344@nuc-i3427.alporthouse.com>
 <20160609013324.GH8105@intel.com>
 <20160609130750.GQ30909@twins.programming.kicks-ass.net>
 <20160616085040.GF30927@twins.programming.kicks-ass.net>
 <20160616122504.GG30927@twins.programming.kicks-ass.net>
 <20160617091948.GJ30927@twins.programming.kicks-ass.net>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20160617091948.GJ30927@twins.programming.kicks-ass.net>
User-Agent: Mutt/1.5.21 (2010-09-15)
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 17, 2016 at 11:19:48AM +0200, Peter Zijlstra wrote:
> +/*
> + * Unsigned subtract and clamp on underflow.
> + *
> + * Explicitly do a load-store to ensure the intermediate value never hits
> + * memory. This allows lockless observations without ever seeing the negative
> + * values.
> + */
> +#define sub_positive(_ptr, _val) do {				\
> +	typeof(_ptr) ptr = (_ptr);				\
> +	typeof(*ptr) val = (_val);				\
> +	typeof(*ptr) res, var = READ_ONCE(*ptr);		\
> +	res = var - val;					\
> +	if (res > var)						\
> +		res = 0;					\
> +	WRITE_ONCE(*ptr, res);					\
> +} while (0)
> +

maybe sub_nonnegative() or sub_til_zero() ...

I wonder whether it is the if statement finally allows GCC to 'registerize'
load_avg, and I almost tried it...