All of lore.kernel.org
 help / color / mirror / Atom feed
From: Arthur Marsh <arthur.marsh@internode.on.net>
To: Andrea Arcangeli <aarcange@redhat.com>
Cc: Clemens Ladisch <cladisch@googlemail.com>,
	alsa-user@lists.sourceforge.net, linux-kernel@vger.kernel.org,
	Mel Gorman <mel@csn.ul.ie>
Subject: Re: [Alsa-user] new source of MIDI playback slow-down identified - 5a03b051ed87e72b959f32a86054e1142ac4cf55 thp: use compaction in kswapd for GFP_ATOMIC order > 0
Date: Thu, 24 Feb 2011 12:10:45 +1030	[thread overview]
Message-ID: <4D65B71D.5030109@internode.on.net> (raw)
In-Reply-To: <20110223235946.GW31195@random.random>



Andrea Arcangeli wrote, on 24/02/11 10:29:
> On Thu, Feb 24, 2011 at 08:25:43AM +1030, Arthur Marsh wrote:
>> One more combination I tried:
>>
>> Mel Gorman's mm/compaction.c patch with Andrea Archangeli's
>> kswapd-high_wmark + compaction-no-kswapd-3 patches - kswapd0 CPU less
>> than 2 percent and no noticable slowdown of MIDI playback.
>
> Applying Mel's patch on top should decrease latency more.
>
>> If you can send me an updated patch compaction-no-kswapd-3 (I presume
>> that kswapd-high_wmark is still needed) it would be easier for me to apply.
>
> It's a compaction-kswapd-3. It's likely going to work the same as the
> previous compaction-kswapd-2 (not as good as
> compaction-no-kswapd). It's better to apply both the kswapd-high_wmark
> and Mel's patch too (not only this one) during testing.

OK, with kswapd-high_wmark + compaction-kswapd-3 + Mel's patch (with the 
compaction initialisation fix), MIDI playback is fine.
kswapd0 CPU can very occasionally hit equal highest (17 percent was the 
highest I noticed, but is generally below the top 4-5 processes and less 
than 10 percent when working, dropping to around 0.3 percent when swap 
activity has subsided). This was with loading KDE 3.5.10, konversation, 
aptitude -u, icedove and iceweasel with several dozen tabs in addition 
to aplaymidi.

Regards,

Arthur.

  reply	other threads:[~2011-02-24  1:40 UTC|newest]

Thread overview: 31+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <g0ia38-jj6.ln1@ppp121-45-136-118.lns11.adl6.internode.on.net>
2011-02-22  7:37 ` [Alsa-user] new source of MIDI playback slow-down identified - 5a03b051ed87e72b959f32a86054e1142ac4cf55 thp: use compaction in kswapd for GFP_ATOMIC order > 0 Clemens Ladisch
2011-02-22  7:46   ` Arthur Marsh
2011-02-22 13:40   ` Andrea Arcangeli
2011-02-22 16:15     ` Andrea Arcangeli
2011-02-22 16:59       ` Mel Gorman
2011-02-22 17:08         ` Andrea Arcangeli
2011-02-22 17:37           ` Mel Gorman
2011-02-22 17:47       ` Arthur Marsh
2011-02-22 19:43         ` Andrea Arcangeli
2011-02-23  9:15           ` Mel Gorman
2011-02-23 11:41             ` Arthur Marsh
2011-02-23 13:50               ` Clemens Ladisch
2011-02-23 17:01               ` Mel Gorman
2011-02-23 17:40                 ` Andrea Arcangeli
2011-02-23 16:24         ` Andrea Arcangeli
2011-02-23 16:36           ` Andrea Arcangeli
2011-02-23 16:40             ` Andrea Arcangeli
2011-02-23 16:47               ` Andrea Arcangeli
2011-02-23 16:55           ` Andrea Arcangeli
2011-02-23 20:07             ` Arthur Marsh
2011-02-23 21:25               ` Andrea Arcangeli
2011-02-23 21:55                 ` Arthur Marsh
2011-02-23 23:59                   ` Andrea Arcangeli
2011-02-24  1:40                     ` Arthur Marsh [this message]
2011-02-24  1:54                       ` Andrea Arcangeli
2011-02-26  6:43                         ` Andrea Arcangeli
2011-02-27  8:48                           ` Arthur Marsh
2011-02-23 17:10           ` Mel Gorman
2011-02-23 17:27             ` Andrea Arcangeli
2011-02-23 17:44               ` Mel Gorman
2011-02-23 18:14                 ` Andrea Arcangeli

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=4D65B71D.5030109@internode.on.net \
    --to=arthur.marsh@internode.on.net \
    --cc=aarcange@redhat.com \
    --cc=alsa-user@lists.sourceforge.net \
    --cc=cladisch@googlemail.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mel@csn.ul.ie \
    /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.