From: Jens Axboe <axboe@fb.com>
To: Ming Lei <tom.leiming@gmail.com>
Cc: Jens Axboe <axboe@kernel.dk>,
Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
linux-block <linux-block@vger.kernel.org>,
Christoph Hellwig <hch@lst.de>
Subject: Re: [PATCH 3/4] blk-mq: implement hybrid poll mode for sync O_DIRECT
Date: Thu, 3 Nov 2016 07:41:28 -0600 [thread overview]
Message-ID: <5762c14d-eb4e-b601-a15f-71ef5fbfec3d@fb.com> (raw)
In-Reply-To: <CACVXFVOoUz7dZ9y=-jqLQFWaekYHnoEQk-ao1s8ydsdh5PJbOw@mail.gmail.com>
On 11/03/2016 06:27 AM, Ming Lei wrote:
> On Wed, Nov 2, 2016 at 5:05 AM, Jens Axboe <axboe@fb.com> wrote:
>> This patch enables a hybrid polling mode. Instead of polling after IO
>> submission, we can induce an artificial delay, and then poll after that.
>> For example, if the IO is presumed to complete in 8 usecs from now, we
>> can sleep for 4 usecs, wake up, and then do our polling. This still puts
>
> I guess in reality it isn't easy to figure a perfect poll time:
>
> - for one driver, different CPU and different drive/disk may cause different
> completion time
>
> - for requests with different size, the completion time can be different too
>
> Is there one way to figure out the poll time automatically?
Yes, it's not easy to figure out the perfect time, the point is to try
and make a guess that's a bit better than the current "let's poll the
whole time". I suspect that for most real world cases, you are going to
be polling for smallish IO of roughly the same size. Hence the stats
should be useful.
But we could extend the tracking a bit and make it smarter.
--
Jens Axboe
next prev parent reply other threads:[~2016-11-03 13:41 UTC|newest]
Thread overview: 27+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-11-01 21:05 [PATCHSET] block: IO polling improvements Jens Axboe
2016-11-01 21:05 ` [PATCH 1/4] block: add scalable completion tracking of requests Jens Axboe
2016-11-01 22:25 ` Johannes Thumshirn
2016-11-02 5:37 ` Jens Axboe
2016-11-02 14:52 ` Christoph Hellwig
2016-11-02 14:55 ` Jens Axboe
2016-11-02 14:59 ` Christoph Hellwig
2016-11-03 11:17 ` Ming Lei
2016-11-03 13:38 ` Jens Axboe
2016-11-03 14:57 ` Ming Lei
2016-11-03 16:55 ` Jens Axboe
2016-11-04 23:13 ` Ming Lei
2016-11-05 20:49 ` Jens Axboe
2016-11-05 20:59 ` Jens Axboe
2016-11-03 14:10 ` Bart Van Assche
2016-11-03 14:18 ` Jens Axboe
2016-11-01 21:05 ` [PATCH 2/4] block: move poll code to blk-mq Jens Axboe
2016-11-02 14:54 ` Christoph Hellwig
2016-11-01 21:05 ` [PATCH 3/4] blk-mq: implement hybrid poll mode for sync O_DIRECT Jens Axboe
2016-11-02 14:54 ` Christoph Hellwig
2016-11-03 12:27 ` Ming Lei
2016-11-03 13:41 ` Jens Axboe [this message]
2016-11-03 14:01 ` Bart Van Assche
2016-11-03 14:15 ` Jens Axboe
2016-11-01 21:05 ` [PATCH 4/4] blk-mq: make the polling code adaptive Jens Axboe
2016-11-02 14:51 ` [PATCHSET] block: IO polling improvements Christoph Hellwig
2016-11-02 14:54 ` Jens Axboe
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=5762c14d-eb4e-b601-a15f-71ef5fbfec3d@fb.com \
--to=axboe@fb.com \
--cc=axboe@kernel.dk \
--cc=hch@lst.de \
--cc=linux-block@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=tom.leiming@gmail.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 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).