linux-kernel - Re: [PATCH v1 3/3] blk-mq: start to freeze queue just after setting dying

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Sat, 18 Mar 2017 09:01:00 +0800
From:   Ming Lei <tom.leiming@...il.com>
To:     Bart Van Assche <Bart.VanAssche@...disk.com>
Cc:     "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
        "hch@...radead.org" <hch@...radead.org>,
        "linux-block@...r.kernel.org" <linux-block@...r.kernel.org>,
        "axboe@...com" <axboe@...com>,
        "yizhan@...hat.com" <yizhan@...hat.com>,
        "tj@...nel.org" <tj@...nel.org>
Subject: Re: [PATCH v1 3/3] blk-mq: start to freeze queue just after setting
 dying

On Fri, Mar 17, 2017 at 11:26:26PM +0000, Bart Van Assche wrote:
> On Fri, 2017-03-17 at 17:57 +0800, Ming Lei wrote:
> > Given blk_set_queue_dying() is always called in remove path
> > of block device, and queue will be cleaned up later, we don't
> > need to worry about undoing the counter.
> > 
> > diff --git a/block/blk-core.c b/block/blk-core.c
> > index d772c221cc17..62d4967c369f 100644
> > --- a/block/blk-core.c
> > +++ b/block/blk-core.c
> > @@ -500,9 +500,12 @@ void blk_set_queue_dying(struct request_queue *q)
> >  	queue_flag_set(QUEUE_FLAG_DYING, q);
> >  	spin_unlock_irq(q->queue_lock);
> >  
> > -	if (q->mq_ops)
> > +	if (q->mq_ops) {
> >  		blk_mq_wake_waiters(q);
> > -	else {
> > +
> > +		/* block new I/O coming */
> > +		blk_mq_freeze_queue_start(q);
> > +	} else {
> >  		struct request_list *rl;
> >  
> >  		spin_lock_irq(q->queue_lock);
> 
> Hello Ming,
> 
> The blk_freeze_queue() call in blk_cleanup_queue() waits until q_usage_counter
> drops to zero. Since the above blk_mq_freeze_queue_start() call increases that
> counter by one, how is blk_freeze_queue() expected to finish ever?

It is q->mq_freeze_depth which is increased by blk_mq_freeze_queue_start(), not
q->q_usage_counter, otherwise blk_freeze_queue() would never return, :-)

Thanks,
Ming