Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
References: <20201116091054.GL3371@techsingularity.net> <20201116131102.GA29992@willie-the-truck> <20201116133721.GQ3371@techsingularity.net> <20201116142005.GE3121392@hirez.programming.kicks-ass.net> <20201116193149.GW3371@techsingularity.net> <20201117083016.GK3121392@hirez.programming.kicks-ass.net> <20201117091545.GA31837@willie-the-truck> <20201117092936.GA3121406@hirez.programming.kicks-ass.net> <20201117094621.GE3121429@hirez.programming.kicks-ass.net>
User-agent: mu4e 0.9.17; emacs 26.3
From:   Valentin Schneider <valentin.schneider@arm.com>
To:     Peter Zijlstra <peterz@infradead.org>
Cc:     Will Deacon <will@kernel.org>,
        Mel Gorman <mgorman@techsingularity.net>,
        Davidlohr Bueso <dave@stgolabs.net>,
        linux-arm-kernel@lists.infradead.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH] sched: Fix data-race in wakeup
In-reply-to: <20201117094621.GE3121429@hirez.programming.kicks-ass.net>
Date:   Tue, 17 Nov 2020 12:52:12 +0000
Message-ID: <jhjv9e4w3gj.mognet@arm.com>
MIME-Version: 1.0
Content-Type: text/plain
Precedence: bulk


On 17/11/20 09:46, Peter Zijlstra wrote:
> How's this then? It still doesn't explicitly call out the specific race,
> but does mention the more fundamental issue that wakelist queueing
> doesn't respect the regular rules anymore.
>
> --- a/include/linux/sched.h
> +++ b/include/linux/sched.h
> @@ -775,7 +775,6 @@ struct task_struct {
>  	unsigned			sched_reset_on_fork:1;
>  	unsigned			sched_contributes_to_load:1;
>  	unsigned			sched_migrated:1;
> -	unsigned			sched_remote_wakeup:1;
>  #ifdef CONFIG_PSI
>  	unsigned			sched_psi_wake_requeue:1;
>  #endif
> @@ -785,6 +784,21 @@ struct task_struct {
>  
>  	/* Unserialized, strictly 'current' */
>  
> +	/*
> +	 * This field must not be in the scheduler word above due to wakelist
> +	 * queueing no longer being serialized by p->on_cpu. However:
> +	 *
> +	 * p->XXX = X;			ttwu()
> +	 * schedule()			  if (p->on_rq && ..) // false
> +	 *   smp_mb__after_spinlock();	  if (smp_load_acquire(&p->on_cpu) && //true
> +	 *   deactivate_task()		      ttwu_queue_wakelist())
> +	 *     p->on_rq = 0;			p->sched_remote_wakeup = Y;
> +	 *
> +	 * guarantees all stores of 'current' are visible before
> +	 * ->sched_remote_wakeup gets used, so it can be in this word.
> +	 */

Isn't the control dep between that ttwu() p->on_rq read and
p->sched_remote_wakeup write "sufficient"? That should be giving the right
ordering for the rest of ttwu() wrt. those 'current' bits, considering they
are written before that smp_mb__after_spinlock().

In any case, consider me convinced:

Reviewed-by: Valentin Schneider <valentin.schneider@arm.com>

> +	unsigned			sched_remote_wakeup:1;
> +
>  	/* Bit to tell LSMs we're in execve(): */
>  	unsigned			in_execve:1;
>  	unsigned			in_iowait:1;