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 S1755379AbdLTODS (ORCPT <rfc822;w@1wt.eu>);
        Wed, 20 Dec 2017 09:03:18 -0500
Received: from bombadil.infradead.org ([65.50.211.133]:43942 "EHLO
        bombadil.infradead.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
        with ESMTP id S1755362AbdLTODP (ORCPT
        <rfc822;linux-kernel@vger.kernel.org>);
        Wed, 20 Dec 2017 09:03:15 -0500
Date: Wed, 20 Dec 2017 15:03:06 +0100
From: Peter Zijlstra <peterz@infradead.org>
To: Brendan Jackman <brendan.jackman@arm.com>
Cc: Vincent Guittot <vincent.guittot@linaro.org>,
        Dietmar Eggemann <dietmar.eggemann@arm.com>,
        Ingo Molnar <mingo@kernel.org>, linux-kernel@vger.kernel.org,
        Ingo Molnar <mingo@redhat.com>,
        Morten Rasmussen <morten.rasmussen@arm.com>
Subject: Re: [PATCH v2 1/2] sched: force update of blocked load of idle cpus
Message-ID: <20171220140306.egxub7sog4il2plm@hirez.programming.kicks-ass.net>
References: <20171201180157.18937-1-brendan.jackman@arm.com>
 <20171201180157.18937-2-brendan.jackman@arm.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20171201180157.18937-2-brendan.jackman@arm.com>
User-Agent: NeoMutt/20170609 (1.8.3)
Sender: linux-kernel-owner@vger.kernel.org
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org

On Fri, Dec 01, 2017 at 06:01:56PM +0000, Brendan Jackman wrote:
> @@ -8955,8 +8964,20 @@ static void nohz_balancer_kick(void)
>  	if (ilb_cpu >= nr_cpu_ids)
>  		return;
>  
> -	if (test_and_set_bit(NOHZ_BALANCE_KICK, nohz_flags(ilb_cpu)))
> +	if (test_and_set_bit(NOHZ_BALANCE_KICK, nohz_flags(ilb_cpu))) {
> +		if (!only_update) {
> +			/*
> +			 * There's a pending/ongoing nohz kick/balance. If it's
> +			 * just for stats, convert it to a proper load balance.
> +			 */
> +			clear_bit(NOHZ_STATS_KICK, nohz_flags(ilb_cpu));
> +		}
>  		return;
> +	}
> +
> +	if (only_update)
> +		set_bit(NOHZ_STATS_KICK, nohz_flags(ilb_cpu));
> +
>  	/*
>  	 * Use smp_send_reschedule() instead of resched_cpu().
>  	 * This way we generate a sched IPI on the target cpu which

This looks racy.. if its not we don't need atomic ops, if it is but is
still fine it needs a comment.