All of lore.kernel.org
 help / color / mirror / Atom feed
From: Ming Lei <tom.leiming@gmail.com>
To: ricardo.ribalda@gmail.com
Cc: linux-block <linux-block@vger.kernel.org>,
	ben@decadent.org.uk, Ming Lei <ming.lei@redhat.com>
Subject: Re: blk-mq 5-8 times slower for bmap-tools
Date: Sat, 18 Aug 2018 07:43:46 +0800	[thread overview]
Message-ID: <CACVXFVNguNEJ_R7XXvC8dG_yu5tghQk+Hdg6Hp6KKr4Ys4EV_A@mail.gmail.com> (raw)
In-Reply-To: <CAPybu_2k-0d=8NxL7kr757QcWOnXyU_TAyV7DMwiMrSDkiXfgA@mail.gmail.com>

On Sat, Aug 18, 2018 at 1:32 AM Ricardo Ribalda Delgado
<ricardo.ribalda@gmail.com> wrote:
>
> Hello
>
> Coming from: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=906328
>
> bmaptools is kind of a "smart dd" tool, that lets you write images
> very fast and securely. Since the last Debian Kernel update it has
> become 5-8 times slower.
>
> After some debugging, we have figured out that the reason for that
> slowness is the Multi-Queue Block IO Queueing Mechanism.
>
> Debian maintainer has pointed out that in the near future the single
> queue will be deprecated.
>
> My question is if we can get a similar perfomance for bmaptools with
> blk-mq and how?

It suppose to be not worse than non-blk-mq, and maybe there is bug
somewhere.

Could you use 'iostat -dx $DISK_PATH 1' to collect io stat when running
bmaptools in both blk-mq and non-blk-mq?

BTW, don't change the default io scheduler as none, which shouldn't
work well for slow disk, such as non-SSD.

Thanks,
Ming Lei

  parent reply	other threads:[~2018-08-18  2:49 UTC|newest]

Thread overview: 21+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-08-17 17:31 blk-mq 5-8 times slower for bmap-tools Ricardo Ribalda Delgado
2018-08-17 17:35 ` Paolo Valente
2018-08-17 17:39   ` Ricardo Ribalda Delgado
2018-08-17 17:41     ` Jens Axboe
2018-08-17 17:49       ` Ricardo Ribalda Delgado
2018-08-17 21:23         ` Jens Axboe
2018-08-17 17:42     ` Paolo Valente
2018-08-17 17:52       ` Ricardo Ribalda Delgado
2018-08-17 23:43 ` Ming Lei [this message]
     [not found]   ` <CAPybu_2ke4usj9Qwr9iSKTymAk1510pzZnZAgLemRB+cK4Se1A@mail.gmail.com>
2018-08-20  8:30     ` Ming Lei
2018-08-20  9:04       ` Ricardo Ribalda Delgado
2018-08-20  9:55         ` Ming Lei
2018-08-20  9:56           ` Ricardo Ribalda Delgado
2018-08-20 10:15             ` Ming Lei
2018-08-20 11:42               ` Ming Lei
2018-08-20 11:45                 ` Ming Lei
2018-08-22 19:22         ` Ben Hutchings
2018-08-22 22:02           ` Ming Lei
2018-08-23  0:51             ` Ben Hutchings
2018-08-23  2:17               ` Ming Lei
2018-08-23  6:29           ` Ricardo Ribalda Delgado

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=CACVXFVNguNEJ_R7XXvC8dG_yu5tghQk+Hdg6Hp6KKr4Ys4EV_A@mail.gmail.com \
    --to=tom.leiming@gmail.com \
    --cc=ben@decadent.org.uk \
    --cc=linux-block@vger.kernel.org \
    --cc=ming.lei@redhat.com \
    --cc=ricardo.ribalda@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 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.