From: Jens Axboe <axboe@fb.com>
To: "Paolo Valente" <paolo.valente@unimore.it>,
"Matias Bjørling" <m@bjorling.me>,
"Arianna Avanzini" <avanzini@google.com>
Cc: Akinobu Mita <akinobu.mita@gmail.com>,
"Luis R. Rodriguez" <mcgrof@suse.com>,
Ming Lei <ming.lei@canonical.com>,
Mike Krinkin <krinkin.m.u@gmail.com>,
<linux-kernel@vger.kernel.org>
Subject: Re: [PATCH BUGFIX 2/3] null_blk: guarantee device restart in all irq modes
Date: Mon, 2 Nov 2015 09:25:23 -0700 [thread overview]
Message-ID: <56378E73.2040703@fb.com> (raw)
In-Reply-To: <1446474673-2566-3-git-send-email-paolo.valente@unimore.it>
On 11/02/2015 07:31 AM, Paolo Valente wrote:
> From: Arianna Avanzini <avanzini@google.com>
>
> In single-queue (block layer) mode,the function null_rq_prep_fn stops
> the device if alloc_cmd fails. Then, once stopped, the device must be
> restarted on the next command completion, so that the request(s) for
> which alloc_cmd failed can be requeued. Otherwise the device hangs.
>
> Unfortunately, device restart is currently performed only for delayed
> completions, i.e., in irqmode==2. This fact causes hangs, for the
> above reasons, with the other irqmodes in combination with single-queue
> block layer.
>
> This commits addresses this issue by making sure that, if stopped, the
> device is properly restarted for all irqmodes on completions.
This looks good. I did a double take at the removal of the q->mq_ops
check before blk_queue_stopped(), but we can't get there in MQ mode.
Perhaps a comment would be warranted for that, the ->mq_ops check served
as a bit of documentation for that before.
--
Jens Axboe
next prev parent reply other threads:[~2015-11-02 16:25 UTC|newest]
Thread overview: 15+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-11-02 14:31 [PATCH BUGFIX 0/3] null_blk: fix throughout losses and hangs Paolo Valente
2015-11-02 14:31 ` [PATCH BUGFIX 1/3] null_blk: set a separate timer for each command Paolo Valente
2015-11-02 16:14 ` Jens Axboe
2015-11-03 9:01 ` Paolo Valente
2015-11-29 17:27 ` Paolo Valente
2015-11-30 15:55 ` Jens Axboe
2015-12-01 10:48 ` [PATCH BUGFIX V2 0/3] null_blk: fix throughput losses and hangs Paolo Valente
2015-12-01 10:48 ` [PATCH BUGFIX V2 1/3] null_blk: set a separate timer for each command Paolo Valente
2015-12-01 10:48 ` [PATCH BUGFIX V2 2/3] null_blk: guarantee device restart in all irq modes Paolo Valente
2015-12-01 10:48 ` [PATCH BUGFIX V2 3/3] null_blk: change type of completion_nsec to unsigned long Paolo Valente
2015-12-01 17:52 ` [PATCH BUGFIX V2 0/3] null_blk: fix throughput losses and hangs Jens Axboe
2015-11-02 14:31 ` [PATCH BUGFIX 2/3] null_blk: guarantee device restart in all irq modes Paolo Valente
2015-11-02 16:25 ` Jens Axboe [this message]
2015-11-03 9:02 ` Paolo Valente
2015-11-02 14:31 ` [PATCH BUGFIX 3/3] null_blk: change type of completion_nsec to unsigned long Paolo Valente
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=56378E73.2040703@fb.com \
--to=axboe@fb.com \
--cc=akinobu.mita@gmail.com \
--cc=avanzini@google.com \
--cc=krinkin.m.u@gmail.com \
--cc=linux-kernel@vger.kernel.org \
--cc=m@bjorling.me \
--cc=mcgrof@suse.com \
--cc=ming.lei@canonical.com \
--cc=paolo.valente@unimore.it \
/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).