All of lore.kernel.org
 help / color / mirror / Atom feed
From: Ming Lei <ming.lei@redhat.com>
To: Keith Busch <kbusch@kernel.org>
Cc: Jens Axboe <axboe@kernel.dk>,
	linux-block@vger.kernel.org, David Jeffery <djeffery@redhat.com>,
	Stefan Hajnoczi <stefanha@redhat.com>,
	virtualization@lists.linux-foundation.org,
	Bart Van Assche <bvanassche@acm.org>
Subject: Re: [PATCH V2 1/1] blk-mq: avoid double ->queue_rq() because of early timeout
Date: Wed, 26 Oct 2022 11:29:48 +0800	[thread overview]
Message-ID: <Y1iprFhtHLSESDdJ@T590> (raw)
In-Reply-To: <Y1ilaQV3hz6kudH3@kbusch-mbp.dhcp.thefacebook.com>

On Tue, Oct 25, 2022 at 09:11:37PM -0600, Keith Busch wrote:
> On Wed, Oct 26, 2022 at 09:55:21AM +0800, Ming Lei wrote:
> > @@ -1564,8 +1571,13 @@ static bool blk_mq_check_expired(struct request *rq, void *priv)
> >  	 * it was completed and reallocated as a new request after returning
> >  	 * from blk_mq_check_expired().
> >  	 */
> > -	if (blk_mq_req_expired(rq, next))
> > +	if (blk_mq_req_expired(rq, expired)) {
> > +		if (expired->check_only) {
> > +			expired->has_timedout_rq = true;
> > +			return false;
> > +		}
> >  		blk_mq_rq_timed_out(rq);
> > +	}
> >  	return true;
> >  }
> >  
> > @@ -1573,7 +1585,10 @@ static void blk_mq_timeout_work(struct work_struct *work)
> >  {
> >  	struct request_queue *q =
> >  		container_of(work, struct request_queue, timeout_work);
> > -	unsigned long next = 0;
> > +	struct blk_expired_data expired = {
> > +		.check_only = true,
> > +		.timeout_start = jiffies,
> > +	};
> >  	struct blk_mq_hw_ctx *hctx;
> >  	unsigned long i;
> >  
> > @@ -1593,10 +1608,24 @@ static void blk_mq_timeout_work(struct work_struct *work)
> >  	if (!percpu_ref_tryget(&q->q_usage_counter))
> >  		return;
> >  
> > -	blk_mq_queue_tag_busy_iter(q, blk_mq_check_expired, &next);
> > +	/* check if there is any timed-out request */
> > +	blk_mq_queue_tag_busy_iter(q, blk_mq_check_expired, &expired);
> > +	if (expired.has_timedout_rq) {
> > +		/*
> > +		 * Before walking tags, we must ensure any submit started
> > +		 * before the current time has finished. Since the submit
> > +		 * uses srcu or rcu, wait for a synchronization point to
> > +		 * ensure all running submits have finished
> > +		 */
> > +		blk_mq_wait_quiesce_done(q);
> > +
> > +		expired.check_only = false;
> > +		expired.next = 0;
> > +		blk_mq_queue_tag_busy_iter(q, blk_mq_check_expired, &expired);
> 
> I think it would be easier to follow with separate callbacks instead of
> special casing for 'check_only'. One callback for checking timeouts, and
> a different one for handling them?

Both two are basically same, with two callbacks, just .check_only is saved,
nothing else, meantime with one extra similar callback added.

If you or anyone think it is one big deal, I can switch to two callback version.


Thanks,
Ming


WARNING: multiple messages have this Message-ID (diff)
From: Ming Lei <ming.lei@redhat.com>
To: Keith Busch <kbusch@kernel.org>
Cc: Jens Axboe <axboe@kernel.dk>, David Jeffery <djeffery@redhat.com>,
	Bart Van Assche <bvanassche@acm.org>,
	virtualization@lists.linux-foundation.org,
	linux-block@vger.kernel.org,
	Stefan Hajnoczi <stefanha@redhat.com>
Subject: Re: [PATCH V2 1/1] blk-mq: avoid double ->queue_rq() because of early timeout
Date: Wed, 26 Oct 2022 11:29:48 +0800	[thread overview]
Message-ID: <Y1iprFhtHLSESDdJ@T590> (raw)
In-Reply-To: <Y1ilaQV3hz6kudH3@kbusch-mbp.dhcp.thefacebook.com>

On Tue, Oct 25, 2022 at 09:11:37PM -0600, Keith Busch wrote:
> On Wed, Oct 26, 2022 at 09:55:21AM +0800, Ming Lei wrote:
> > @@ -1564,8 +1571,13 @@ static bool blk_mq_check_expired(struct request *rq, void *priv)
> >  	 * it was completed and reallocated as a new request after returning
> >  	 * from blk_mq_check_expired().
> >  	 */
> > -	if (blk_mq_req_expired(rq, next))
> > +	if (blk_mq_req_expired(rq, expired)) {
> > +		if (expired->check_only) {
> > +			expired->has_timedout_rq = true;
> > +			return false;
> > +		}
> >  		blk_mq_rq_timed_out(rq);
> > +	}
> >  	return true;
> >  }
> >  
> > @@ -1573,7 +1585,10 @@ static void blk_mq_timeout_work(struct work_struct *work)
> >  {
> >  	struct request_queue *q =
> >  		container_of(work, struct request_queue, timeout_work);
> > -	unsigned long next = 0;
> > +	struct blk_expired_data expired = {
> > +		.check_only = true,
> > +		.timeout_start = jiffies,
> > +	};
> >  	struct blk_mq_hw_ctx *hctx;
> >  	unsigned long i;
> >  
> > @@ -1593,10 +1608,24 @@ static void blk_mq_timeout_work(struct work_struct *work)
> >  	if (!percpu_ref_tryget(&q->q_usage_counter))
> >  		return;
> >  
> > -	blk_mq_queue_tag_busy_iter(q, blk_mq_check_expired, &next);
> > +	/* check if there is any timed-out request */
> > +	blk_mq_queue_tag_busy_iter(q, blk_mq_check_expired, &expired);
> > +	if (expired.has_timedout_rq) {
> > +		/*
> > +		 * Before walking tags, we must ensure any submit started
> > +		 * before the current time has finished. Since the submit
> > +		 * uses srcu or rcu, wait for a synchronization point to
> > +		 * ensure all running submits have finished
> > +		 */
> > +		blk_mq_wait_quiesce_done(q);
> > +
> > +		expired.check_only = false;
> > +		expired.next = 0;
> > +		blk_mq_queue_tag_busy_iter(q, blk_mq_check_expired, &expired);
> 
> I think it would be easier to follow with separate callbacks instead of
> special casing for 'check_only'. One callback for checking timeouts, and
> a different one for handling them?

Both two are basically same, with two callbacks, just .check_only is saved,
nothing else, meantime with one extra similar callback added.

If you or anyone think it is one big deal, I can switch to two callback version.


Thanks,
Ming

_______________________________________________
Virtualization mailing list
Virtualization@lists.linux-foundation.org
https://lists.linuxfoundation.org/mailman/listinfo/virtualization

  reply	other threads:[~2022-10-26  3:31 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-10-26  1:55 [PATCH V2 1/1] blk-mq: avoid double ->queue_rq() because of early timeout Ming Lei
2022-10-26  1:55 ` Ming Lei
2022-10-26  3:11 ` Keith Busch
2022-10-26  3:29   ` Ming Lei [this message]
2022-10-26  3:29     ` Ming Lei
2022-10-26  3:44     ` Keith Busch

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=Y1iprFhtHLSESDdJ@T590 \
    --to=ming.lei@redhat.com \
    --cc=axboe@kernel.dk \
    --cc=bvanassche@acm.org \
    --cc=djeffery@redhat.com \
    --cc=kbusch@kernel.org \
    --cc=linux-block@vger.kernel.org \
    --cc=stefanha@redhat.com \
    --cc=virtualization@lists.linux-foundation.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body. 
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.