From: Christoph Hellwig <hch@infradead.org>
To: Jens Axboe <axboe@fb.com>
Cc: Christoph Hellwig <hch@infradead.org>,
axboe@kernel.dk, linux-block@vger.kernel.org, jbacik@fb.com,
Ming Lei <ming.lei@canonical.com>
Subject: Re: [PATCH 2/2] blk-mq: add flag for drivers wanting blocking ->queue_rq()
Date: Thu, 22 Sep 2016 08:12:22 -0700 [thread overview]
Message-ID: <20160922151222.GA515@infradead.org> (raw)
In-Reply-To: <a4162d4e-dc64-0e76-6f0c-3ff12eb0231c@fb.com>
On Thu, Sep 22, 2016 at 09:03:56AM -0600, Jens Axboe wrote:
> Having to grab a mutex, for instance. We invoke ->queue_rq() with
> preemption disabled, so I'd hope that would not be the case... What
> drivers block in ->queue_rq?
I though I had converted a lot of them to GFP_NOIO instead of GFP_ATOMIC
allocations, but I can't find any evidence of that. Maybe it was just
my imagination, or an unsubmitted patch series. Sorry for the
confusion.
> Loop was another case that was on my radar to get rid of the queue_work
> it currently has to do. Josef is currently testing the nbd driver using
> this approach, so we should get some numbers there too. If we have to,
> we can always bump up the concurrency to mimic more of the behavior of
> having multiple workers running on the hardware queue. I'd prefer to
> still do that in blk-mq, instead of having drivers reinventing their own
> work offload functionality.
There should be a lot of numbers in the list archives from the time
that Ming did the loop conversion, as I've been trying to steer him
that way, and he actually implemented and benchmarked it.
We can't just increase the concurrency as a single work_struct item
can't be queued multiple times even on a high concurreny workqueue.
next prev parent reply other threads:[~2016-09-22 15:12 UTC|newest]
Thread overview: 19+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-09-22 14:52 [PATCH 0/2]: Add option for async ->queue_rq Jens Axboe
2016-09-22 14:52 ` [PATCH 1/2] blk-mq: get rid of manual run of queue with __blk_mq_run_hw_queue() Jens Axboe
2016-09-22 14:56 ` Christoph Hellwig
2016-09-22 18:56 ` Christoph Hellwig
2016-09-22 20:29 ` Jens Axboe
2016-09-23 21:59 ` Sagi Grimberg
2016-09-23 21:56 ` Sagi Grimberg
2016-09-22 14:53 ` [PATCH 2/2] blk-mq: add flag for drivers wanting blocking ->queue_rq() Jens Axboe
2016-09-22 14:59 ` Christoph Hellwig
2016-09-22 15:03 ` Jens Axboe
2016-09-22 15:12 ` Christoph Hellwig [this message]
2016-09-22 15:17 ` Jens Axboe
2016-09-23 1:43 ` Ming Lei
2016-09-23 2:44 ` Josef Bacik
2016-09-23 1:35 ` Ming Lei
2016-09-22 15:04 ` Josef Bacik
2016-09-22 18:01 ` [PATCH 0/2]: Add option for async ->queue_rq Josef Bacik
2016-09-28 0:25 ` Bart Van Assche
2016-09-28 16:43 ` Omar Sandoval
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=20160922151222.GA515@infradead.org \
--to=hch@infradead.org \
--cc=axboe@fb.com \
--cc=axboe@kernel.dk \
--cc=jbacik@fb.com \
--cc=linux-block@vger.kernel.org \
--cc=ming.lei@canonical.com \
/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.