From: Jens Axboe <axboe-tSWWG44O7X1aa/9Udqfwiw@public.gmane.org>
To: Tejun Heo <tj-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org>
Cc: Joseph Glanville
<joseph.glanville-2MxvZkOi9dvvnOemgxGiVw@public.gmane.org>,
cgroups <cgroups-u79uwXL29TY76Z2rM5mHXA@public.gmane.org>,
Vivek Goyal <vgoyal-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>,
linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
Subject: Re: [PATCH 1/2] block: lift the initial queue bypass mode on blk_register_queue() instead of blk_init_allocated_queue()
Date: Fri, 21 Sep 2012 15:25:40 +0200 [thread overview]
Message-ID: <505C6AD4.6030206@kernel.dk> (raw)
In-Reply-To: <20120920210852.GC7264-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org>
On 09/20/2012 11:08 PM, Tejun Heo wrote:
> b82d4b197c ("blkcg: make request_queue bypassing on allocation") made
> request_queues bypassed on allocation to avoid switching on and off
> bypass mode on a queue being initialized. Some drivers allocate and
> then destroy a lot of queues without fully initializing them and
> incurring bypass latency overhead on each of them could add upto
> significant overhead.
>
> Unfortunately, blk_init_allocated_queue() is never used by queues of
> bio-based drivers, which means that all bio-based driver queues are in
> bypass mode even after initialization and registration complete
> successfully.
>
> Due to the limited way request_queues are used by bio drivers, this
> problem is hidden pretty well but it shows up when blk-throttle is
> used in combination with a bio-based driver. Trying to configure
> (echoing to cgroupfs file) blk-throttle for a bio-based driver hangs
> indefinitely in blkg_conf_prep() waiting for bypass mode to end.
>
> This patch moves the initial blk_queue_bypass_end() call from
> blk_init_allocated_queue() to blk_register_queue() which is called for
> any userland-visible queues regardless of its type.
>
> I believe this is correct because I don't think there is any block
> driver which needs or wants working elevator and blk-cgroup on a queue
> which isn't visible to userland. If there are such users, we need a
> different solution.
>
> Signed-off-by: Tejun Heo <tj-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org>
> Reported-by: Joseph Glanville <joseph.glanville-2MxvZkOi9dvvnOemgxGiVw@public.gmane.org>
> Cc: Vivek Goyal <vgoyal-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
> Cc: stable-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
> ---
> Jens, while these are fixes, I think it isn't extremely urgent and
> routing these through 3.7-rc1 should be enough.
Agree, I'll shove them into for-3.7/core
--
Jens Axboe
next prev parent reply other threads:[~2012-09-21 13:25 UTC|newest]
Thread overview: 19+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-09-16 17:31 3.6-rc5 cgroups blkio throttle + md regression Joseph Glanville
[not found] ` <CAOzFzEhf2LfT0BCNPAgPgxZ3=pj2KvJ4Z6kP7XF8nnxag1dfvg-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2012-09-19 16:28 ` Joseph Glanville
[not found] ` <CAOzFzEiC4313K4H9393ffzNyBo398BPYSxTk7ZEmuH4GfW5qtw-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2012-09-19 18:20 ` Joseph Glanville
[not found] ` <CAOzFzEhWk_7oQFOyW6Ri_9Cvsshj2s3pa=Oo-p8uL6r340MoTw-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2012-09-19 19:42 ` Vivek Goyal
[not found] ` <20120919194231.GF31860-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2012-09-20 18:31 ` Tejun Heo
[not found] ` <20120920183153.GI28934-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org>
2012-09-20 18:42 ` Vivek Goyal
[not found] ` <20120920184219.GH4681-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2012-09-20 19:17 ` Vivek Goyal
[not found] ` <20120920191716.GI4681-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2012-09-20 19:20 ` Tejun Heo
[not found] ` <20120920192038.GJ28934-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org>
2012-09-20 19:32 ` Vivek Goyal
[not found] ` <20120920193227.GJ4681-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2012-09-20 20:17 ` Tejun Heo
2012-09-20 19:57 ` Vivek Goyal
[not found] ` <20120920195759.GK4681-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2012-09-20 20:18 ` Tejun Heo
[not found] ` <20120920201815.GB7264-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org>
2012-09-20 21:08 ` [PATCH 1/2] block: lift the initial queue bypass mode on blk_register_queue() instead of blk_init_allocated_queue() Tejun Heo
[not found] ` <20120920210852.GC7264-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org>
2012-09-20 21:09 ` [PATCH 2/2] block: fix request_queue->flags initialization Tejun Heo
[not found] ` <20120920210930.GD7264-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org>
2012-09-21 13:25 ` Vivek Goyal
2012-09-21 13:25 ` [PATCH 1/2] block: lift the initial queue bypass mode on blk_register_queue() instead of blk_init_allocated_queue() Vivek Goyal
2012-09-21 13:25 ` Jens Axboe [this message]
[not found] ` <505C6AD4.6030206-tSWWG44O7X1aa/9Udqfwiw@public.gmane.org>
2012-10-16 10:00 ` Joseph Glanville
[not found] ` <CAOzFzEiCWazLEfjo=w8c+7qCce98Q6faW1uwGm-tRmCNPJUztw-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2012-10-16 19:11 ` Tejun Heo
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=505C6AD4.6030206@kernel.dk \
--to=axboe-tswwg44o7x1aa/9udqfwiw@public.gmane.org \
--cc=cgroups-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
--cc=joseph.glanville-2MxvZkOi9dvvnOemgxGiVw@public.gmane.org \
--cc=linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
--cc=tj-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org \
--cc=vgoyal-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.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 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).