Re: [PATCH] kthread: Fix race condition between kthread_parkme() and kthread_unpark()

Path	csiph.com!aioe.org!bofh.it!news.nic.it!robomod
From	Junaid Shahid <junaids@google.com>
Newsgroups	linux.kernel
Subject	Re: [PATCH] kthread: Fix race condition between kthread_parkme() and kthread_unpark()
Date	Wed, 31 May 2017 00:40:01 +0200
Message-ID	<tMXXX-4Lf-5@gated-at.bofh.it> (permalink)
References	<tBqsF-8mn-1@gated-at.bofh.it>
Dkim-Signature	v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20161025; h=from:to:cc:subject:date:message-id:organization:user-agent :in-reply-to:references:mime-version:content-transfer-encoding; bh=D7hxMQRI3fDQHn3HrDPkDjbXEShhn303w4RBFprE7ts=; b=KuNSl5PpQ51K1llBA86plZTlIJygiB4rkbM9jF61svfLEMISfz9t3I0q0vGHzOWe9N DQut0KCF1o/N4N84EEC6rPPtKjI+M+U/HInrbO10xjhIKc+uFVH3bgAVNUHDMQJpDL8L vakR0Kgr0dyLEJmjqkrfhJv6UyHCxNQCk2wPn1vUExEiIctTM7qg9hA9HzAdWgNFd1zu iEIy6zHDuOz2a7I8n9FZ473JS2jsw/Rrg07hrdIeDDYVmmJJWTISkHDUtADBZIOHl4HT trThkjf854nrx7GFSci4dyYFbxoPg+CpPaeOO815v/vh9s6/gCxyp0G3pwGH4AOIdDRQ Gc1g==
X-Google-Dkim-Signature	v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:from:to:cc:subject:date:message-id:organization :user-agent:in-reply-to:references:mime-version :content-transfer-encoding; bh=D7hxMQRI3fDQHn3HrDPkDjbXEShhn303w4RBFprE7ts=; b=uhWoeDRQwXn3M5A34vnzbfmYA8jcExasBXADPiN9EKZW9MHTkPR5q+W4G+trHM+TcR 74meTuG1wZRQ4PXeNaZpooLSfKPCpF4tho6lyIjze9Xu5PsKcwC/osE8ZvTzWovIbvpC gz1BFF9JHsZTuIKFxqGz3Q3MYfM0Xzebn4ePX+GHCnQNG628BFs9Bwa+KzTU+e9WhaX2 ZzvykFbh8kjORRHT6YZVpz/InTaZ6HFBKe6As6B0Ynf3yPS2EgXEuTgts+CUVrAwVzXB hX/OlpVPR/HVgjczNfv4P+6si6F2l/orGWmjSjAx5H/wLYN5G13mMDQU1ZZnsul0OZ2C ZbEw==
X-Gm-Message-State	AODbwcDtSr3OjvuA+jSpoV7icvcLVJb52i6xb7T54xT1H1KMLj689X74 5jaiUbRb5SFcCrPx
X-Received	by 10.84.233.205 with SMTP id m13mr85137282pln.72.1496183844660; Tue, 30 May 2017 15:37:24 -0700 (PDT)
Organization	Google
User-Agent	KMail/4.13.3 (Linux/4.4.0-75-generic; KDE/4.13.3; x86_64; ; )
MIME-Version	1.0
Content-Transfer-Encoding	7Bit
Content-Type	text/plain; charset="us-ascii"
Sender	robomod@news.nic.it
List-ID	<linux-kernel.vger.kernel.org>
X-Mailing-List	linux-kernel@vger.kernel.org
Approved	robomod@news.nic.it
Lines	52
X-Original-Cc	andreslc@google.com, gthelen@google.com, rusty@rustcorp.com.au
X-Original-Date	Tue, 30 May 2017 15:37:23 -0700
X-Original-Message-ID	<4713763.QsSBOgQyOh@js-desktop.svl.corp.google.com>
X-Original-References	<20170429023236.60452-1-junaids@google.com>
X-Original-Sender	linux-kernel-owner@vger.kernel.org
Xref	csiph.com linux.kernel:1653687

Show key headers only | View raw

(Resending)

On Friday, April 28, 2017 07:32:36 PM Junaid Shahid wrote:
> In general, if kthread_unpark() and kthread_parkme() execute together,
> the kthread is supposed to be in an unparked state. This is because
> kthread_unpark() either wakes up the thread if it already got parked,
> or it cancels a prior kthread_park() call and hence renders the
> kthread_parkme() moot.
> 
> However, if kthread_unpark() happens to execute between the time that
> kthread_parkme() checks the KTHREAD_SHOULD_STOP flag and sets the
> KTHREAD_IS_PARKED flag, then kthread_unpark() will not wake up the
> thread and it will remain in a parked state.
> 
> This is fixed by making the checking of KTHREAD_SHOULD_STOP and the
> setting of KTHREAD_IS_PARKED atomic via a cmpxchg inside kthread_parkme.
> 
> Signed-off-by: Junaid Shahid <junaids@google.com>
> ---
>  kernel/kthread.c | 16 ++++++++++++----
>  1 file changed, 12 insertions(+), 4 deletions(-)
> 
> diff --git a/kernel/kthread.c b/kernel/kthread.c
> index 26db528c1d88..651f03baf62f 100644
> --- a/kernel/kthread.c
> +++ b/kernel/kthread.c
> @@ -169,12 +169,20 @@ void *kthread_probe_data(struct task_struct *task)
>  
>  static void __kthread_parkme(struct kthread *self)
>  {
> +	ulong flags;
> +
>  	__set_current_state(TASK_PARKED);
> -	while (test_bit(KTHREAD_SHOULD_PARK, &self->flags)) {
> -		if (!test_and_set_bit(KTHREAD_IS_PARKED, &self->flags))
> -			complete(&self->parked);
> -		schedule();
> +	flags = self->flags;
> +
> +	while (test_bit(KTHREAD_SHOULD_PARK, &flags)) {
> +		if (cmpxchg(&self->flags, flags,
> +			    flags | (1 << KTHREAD_IS_PARKED)) == flags) {
> +			if (!test_bit(KTHREAD_IS_PARKED, &flags))
> +				complete(&self->parked);
> +			schedule();
> +		}
>  		__set_current_state(TASK_PARKED);
> +		flags = self->flags;
>  	}
>  	clear_bit(KTHREAD_IS_PARKED, &self->flags);
>  	__set_current_state(TASK_RUNNING);
>

Back to linux.kernel | Previous | Next — Next in thread | Find similar | Unroll thread

Thread

Re: [PATCH] kthread: Fix race condition between kthread_parkme() and kthread_unpark() Junaid Shahid <junaids@google.com> - 2017-05-31 00:40 +0200
  Re: [PATCH] kthread: Fix race condition between kthread_parkme()  and kthread_unpark() Andrew Morton <akpm@linux-foundation.org> - 2017-05-31 00:50 +0200

csiph-web