From: Ming Lei <ming.lei@redhat.com>
To: Bart Van Assche <bvanassche@acm.org>
Cc: Christoph Hellwig <hch@lst.de>, Jens Axboe <axboe@kernel.dk>,
linux-block@vger.kernel.org, John Garry <john.garry@huawei.com>,
Hannes Reinecke <hare@suse.com>,
Thomas Gleixner <tglx@linutronix.de>
Subject: Re: [PATCH V11 11/12] blk-mq: re-submit IO in case that hctx is inactive
Date: Thu, 14 May 2020 08:45:03 +0800 [thread overview]
Message-ID: <20200514004503.GD2073570@T590> (raw)
In-Reply-To: <837d3c51-5a14-8c91-7e4a-9ef9b83359b9@acm.org>
On Wed, May 13, 2020 at 08:03:13AM -0700, Bart Van Assche wrote:
> On 2020-05-13 05:21, Christoph Hellwig wrote:
> > Use of the BLK_MQ_REQ_FORCE is pretty bogus here..
> >
> >> + if (rq->rq_flags & RQF_PREEMPT)
> >> + flags |= BLK_MQ_REQ_PREEMPT;
> >> + if (reserved)
> >> + flags |= BLK_MQ_REQ_RESERVED;
> >> + /*
> >> + * Queue freezing might be in-progress, and wait freeze can't be
> >> + * done now because we have request not completed yet, so mark this
> >> + * allocation as BLK_MQ_REQ_FORCE for avoiding this allocation &
> >> + * freeze hung forever.
> >> + */
> >> + flags |= BLK_MQ_REQ_FORCE;
> >> +
> >> + /* avoid allocation failure by clearing NOWAIT */
> >> + nrq = blk_get_request(rq->q, rq->cmd_flags & ~REQ_NOWAIT, flags);
> >> + if (!nrq)
> >> + return;
> >
> > blk_get_request returns an ERR_PTR.
> >
> > But I'd rather avoid the magic new BLK_MQ_REQ_FORCE hack when we can
> > just open code it and document what is going on:
> >
> > static struct blk_mq_tags *blk_mq_rq_tags(struct request *rq)
> > {
> > struct blk_mq_hw_ctx *hctx = rq->mq_hctx;
> >
> > if (rq->q->elevator)
> > return hctx->sched_tags;
> > return hctx->tags;
> > }
> >
> > static void blk_mq_resubmit_rq(struct request *rq)
> > {
> > struct blk_mq_alloc_data alloc_data = {
> > .cmd_flags = rq->cmd_flags & ~REQ_NOWAIT;
> > };
> > struct request *nrq;
> >
> > if (rq->rq_flags & RQF_PREEMPT)
> > alloc_data.flags |= BLK_MQ_REQ_PREEMPT;
> > if (blk_mq_tag_is_reserved(blk_mq_rq_tags(rq), rq->internal_tag))
> > alloc_data.flags |= BLK_MQ_REQ_RESERVED;
> >
> > /*
> > * We must still be able to finish a resubmission due to a hotplug
> > * even even if a queue freeze is in progress.
> > */
> > percpu_ref_get(&q->q_usage_counter);
> > nrq = blk_mq_get_request(rq->q, NULL, &alloc_data);
> > blk_queue_exit(q);
> >
> > if (!nrq)
> > return; // XXX: warn?
> > if (nrq->q->mq_ops->initialize_rq_fn)
> > rq->mq_ops->initialize_rq_fn(nrq);
> >
> > blk_rq_copy_request(nrq, rq);
> > ...
>
> I don't like this because the above code allows allocation of requests
> and tags while a request queue is frozen. I'm concerned that this will
> break code that assumes that no tags are allocated while a request queue
> is frozen. If a request queue has a single hardware queue with 64 tags,
The above code path will never be called for single hw queue.
> if the above code allocates tag 40 and if blk_mq_tag_update_depth()
> reduces the queue depth to 32, will nrq become a dangling pointer?
allocation for nrq is just like other normal allocation, and if
it doesn't work with blk_mq_tag_update_depth(), it must be a more
generic issue instead of relating with this specific use case.
The only difference is that 'nrq' will be allocated from a new active
hctx, so the two requests can co-exist and we needn't to worry deadlock.
thanks,
Ming
next prev parent reply other threads:[~2020-05-14 0:45 UTC|newest]
Thread overview: 32+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-05-13 3:47 [PATCH V11 00/12] blk-mq: improvement CPU hotplug Ming Lei
2020-05-13 3:47 ` [PATCH V11 01/12] block: clone nr_integrity_segments and write_hint in blk_rq_prep_clone Ming Lei
2020-05-13 3:47 ` [PATCH V11 02/12] block: add helper for copying request Ming Lei
2020-05-13 3:47 ` [PATCH V11 03/12] blk-mq: mark blk_mq_get_driver_tag as static Ming Lei
2020-05-13 3:47 ` [PATCH V11 04/12] blk-mq: assign rq->tag in blk_mq_get_driver_tag Ming Lei
2020-05-13 3:47 ` [PATCH V11 05/12] blk-mq: add blk_mq_all_tag_iter Ming Lei
2020-05-13 11:56 ` Christoph Hellwig
2020-05-13 3:47 ` [PATCH V11 06/12] blk-mq: prepare for draining IO when hctx's all CPUs are offline Ming Lei
2020-05-13 6:35 ` Hannes Reinecke
2020-05-13 11:58 ` Christoph Hellwig
2020-05-14 0:33 ` Ming Lei
2020-05-13 3:47 ` [PATCH V11 07/12] blk-mq: stop to handle IO and drain IO before hctx becomes inactive Ming Lei
2020-05-13 11:59 ` Christoph Hellwig
2020-05-14 0:36 ` Ming Lei
2020-05-14 1:12 ` Bart Van Assche
2020-05-14 3:10 ` Ming Lei
2020-05-13 3:47 ` [PATCH V11 08/12] block: add blk_end_flush_machinery Ming Lei
2020-05-13 12:00 ` Christoph Hellwig
2020-05-13 3:48 ` [PATCH V11 09/12] blk-mq: add blk_mq_hctx_handle_dead_cpu for handling cpu dead Ming Lei
2020-05-13 12:06 ` Christoph Hellwig
2020-05-13 3:48 ` [PATCH V11 10/12] block: add request allocation flag of BLK_MQ_REQ_FORCE Ming Lei
2020-05-13 10:34 ` [PATCH V12 " Ming Lei
2020-05-13 3:48 ` [PATCH V11 11/12] blk-mq: re-submit IO in case that hctx is inactive Ming Lei
2020-05-13 9:21 ` John Garry
2020-05-13 12:21 ` Christoph Hellwig
2020-05-13 15:03 ` Bart Van Assche
2020-05-14 0:45 ` Ming Lei [this message]
2020-05-14 0:40 ` Ming Lei
2020-05-13 3:48 ` [PATCH V11 12/12] block: deactivate hctx when the hctx is actually inactive Ming Lei
2020-05-13 7:34 ` [PATCH V11 00/12] blk-mq: improvement CPU hotplug John Garry
2020-05-13 10:37 ` Ming Lei
2020-05-13 11:33 ` John Garry
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=20200514004503.GD2073570@T590 \
--to=ming.lei@redhat.com \
--cc=axboe@kernel.dk \
--cc=bvanassche@acm.org \
--cc=hare@suse.com \
--cc=hch@lst.de \
--cc=john.garry@huawei.com \
--cc=linux-block@vger.kernel.org \
--cc=tglx@linutronix.de \
/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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).