All of lore.kernel.org
 help / color / mirror / Atom feed
From: Jens Axboe <jens.axboe@oracle.com>
To: Matthew <jackdachef@gmail.com>
Cc: Kasper Sandberg <lkml@metanurb.dk>,
	Daniel J Blueman <daniel.blueman@gmail.com>,
	Linux Kernel <linux-kernel@vger.kernel.org>
Subject: Re: performance "regression" in cfq compared to anticipatory, deadline and noop
Date: Tue, 13 May 2008 15:05:09 +0200	[thread overview]
Message-ID: <20080513130508.GQ16217@kernel.dk> (raw)
In-Reply-To: <e85b9d30805130558l502addbcq21996b92fdbfdc5b@mail.gmail.com>

On Tue, May 13 2008, Matthew wrote:
> On Tue, May 13, 2008 at 2:20 PM, Jens Axboe <jens.axboe@oracle.com> wrote:
> >
> > On Sun, May 11 2008, Kasper Sandberg wrote:
> >  > On Sun, 2008-05-11 at 14:14 +0100, Daniel J Blueman wrote:
> >  > > I've been experiencing this for a while also; an almost 50% regression
> >  > > is seen for single-process reads (ie sync) if slice_idle is 1ms or
> >  > > more (eg default of 8) [1], which seems phenomenal.
> >  > >
> >  > > Jens, is this the expected price to pay for optimal busy-spindle
> >  > > scheduling, a design issue, bug or am I missing something totally?
> >  > >
> >  > > Thanks,
> >  > >   Daniel
> [snip]
> ...
> [snip]
> >  >
> >  > Thisd would appear to be quite a considerable performance difference.
> >
> >  Indeed, that is of course a bug. The initial mail here mentions this as
> >  a regression - which kernel was the last that worked ok?
> >
> >  If someone would send me a blktrace of such a slow run, that would be
> >  nice. Basically just do a blktrace /dev/sda (or whatever device) while
> >  doing the hdparm, preferably storing output files on a difference
> >  device. Then send the raw sda.blktrace.* files to me. Thanks!
> >
> >  --
> >  Jens Axboe
> >
> >
> 
> Hi Jens,
> 
> I called this a "regression" since I wasn't sure if this is a real bug
> or just something introduced recently, I just started to use cfq as
> main io-scheduler so I can't tell ...
> 
> testing 2.6.17 unfortunately is somewhat impossible for me (reiser4;
> too new hardware - problems with jmicron)
> 
> google "says" that it seemingly already existed since at least 2.6.18
> (Ubuntu DapperDrake) [see:
> http://ubuntuforums.org/showpost.php?p=1484633&postcount=12]

Funky :/

> well - back to topic:
> 
> for a blktrace one need to enable  CONFIG_BLK_DEV_IO_TRACE , right ?
> blktrace can be obtained from your git-repo ?

Yes on both accounts, or just grab a blktrace snapshot from:

http://brick.kernel.dk/snaps/blktrace-git-latest.tar.gz

if you don't use git.

-- 
Jens Axboe


  reply	other threads:[~2008-05-13 13:05 UTC|newest]

Thread overview: 35+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2008-05-11 13:14 performance "regression" in cfq compared to anticipatory, deadline and noop Daniel J Blueman
2008-05-11 14:02 ` Kasper Sandberg
2008-05-13 12:20   ` Jens Axboe
2008-05-13 12:58     ` Matthew
2008-05-13 13:05       ` Jens Axboe [this message]
     [not found]         ` <e85b9d30805130842p3a34305l4ab1e7926e4b0dba@mail.gmail.com>
2008-05-13 18:03           ` Jens Axboe
2008-05-13 18:40             ` Jens Axboe
2008-05-13 19:23               ` Matthew
2008-05-13 19:30                 ` Jens Axboe
2008-05-14  8:05               ` Daniel J Blueman
2008-05-14  8:26                 ` Jens Axboe
2008-05-14 20:52                   ` Daniel J Blueman
2008-05-14 21:37                     ` Matthew
2008-05-15  7:01                       ` Jens Axboe
2008-05-15 12:21                         ` Fabio Checconi
2008-05-16  6:40                           ` Jens Axboe
2008-05-16  7:46                             ` Fabio Checconi
2008-05-16  7:49                               ` Jens Axboe
2008-05-16  7:57                                 ` Jens Axboe
2008-05-16  8:53                                   ` Daniel J Blueman
2008-05-16  8:57                                     ` Jens Axboe
2008-05-16 15:23                                       ` Matthew
2008-05-16 18:39                                         ` Fabio Checconi
2008-08-24 20:24                           ` Daniel J Blueman
2008-08-25 20:29                             ` Fabio Checconi
2008-08-25 15:39                               ` Daniel J Blueman
2008-08-25 17:06                                 ` Fabio Checconi
2008-12-09 15:14                                   ` Daniel J Blueman
     [not found]                   ` <e85b9d30805140332r3311b2d6r6831d37421ced757@mail.gmail.com>
     [not found]                     ` <e85b9d30805140334q69cb5eacued9a719414e73d53@mail.gmail.com>
     [not found]                       ` <20080514103956.GD16217@kernel.dk>
     [not found]                         ` <e85b9d30805141239g5df9abc6i666b1f621d632b44@mail.gmail.com>
     [not found]                           ` <e85b9d30805161549o7c8f065do24b6567e2ade0afa@mail.gmail.com>
2008-05-19 10:39                             ` Matthew
2008-05-13 13:51     ` Kasper Sandberg
2008-05-14  0:33       ` Kasper Sandberg
  -- strict thread matches above, loose matches on Subject: below --
2008-05-10 19:18 Matthew
     [not found] ` <20080510200053.GA78555@gandalf.sssup.it>
2008-05-10 20:39   ` Matthew
2008-05-10 21:56     ` Fabio Checconi
2008-05-11  0:00     ` Aaron Carroll

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=20080513130508.GQ16217@kernel.dk \
    --to=jens.axboe@oracle.com \
    --cc=daniel.blueman@gmail.com \
    --cc=jackdachef@gmail.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=lkml@metanurb.dk \
    /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.