From: Andy Whitcroft <apw@canonical.com>
To: David Howells <dhowells@redhat.com>
Cc: tytso@mit.edu, Dave Airlie <airlied@gmail.com>,
Dave Airlie <airlied@redhat.com>,
linux-kernel@vger.kernel.org
Subject: Re: Why is kslowd accumulating so much CPU time?
Date: Wed, 16 Jun 2010 14:33:31 +0100 [thread overview]
Message-ID: <AANLkTimTAqERLRx-eSpbeXXwMh3K9f8FaZOeNY-oNmdL@mail.gmail.com> (raw)
In-Reply-To: <AANLkTilZElLEBJDjfpZVx72uDR-x_3t14Jkga5eHy3wf@mail.gmail.com>
On Wed, Jun 16, 2010 at 12:37 PM, Andy Whitcroft <apw@canonical.com> wrote:
>> Can you see what they're doing?
>>
>> watch -n0 cat /sys/kernel/debug/slow_work/runqueue
>
> Turned on the debugging and applied the patch from Ted, and when
> things are bad I see constant cycling of all four threads in the
> output showing similar to the below, note only one thread shows at a
> time:
>
> Every 0.1s: cat /sys/kernel/debug/slow_work/runqueue Wed Jun 16 12:34:52 2010
>
> THR PID ITEM ADDR FL MARK DESC
> === ===== ================ == ===== ==========
> 0 897 ffff88012bb07510 12 20ms DRM_CRTC_HELPER: i915@pci:0000:00:02.0
>
> When things are working well I see the same output appearing for one
> update, about every 10s.
Ok I've added some additional debug and these appear to be being
thrown a loop by the HPD interrupt which is firing continiusly, here
is a sample from some 4 minutes after boot:
[ 284.862670] APW: drm_helper_hpd_irq_event: cancel/submit
[ 284.882803] APW: drm_helper_hpd_irq_event: cancel/submit
[ 284.902691] APW: drm_helper_hpd_irq_event: cancel/submit
[ 284.922814] APW: drm_helper_hpd_irq_event: cancel/submit
[ 284.942620] APW: drm_helper_hpd_irq_event: cancel/submit
[ 284.962707] APW: drm_helper_hpd_irq_event: cancel/submit
[ 284.982937] APW: drm_helper_hpd_irq_event: cancel/submit
[ 285.004181] APW: drm_helper_hpd_irq_event: cancel/submit
[ 285.022622] APW: drm_helper_hpd_irq_event: cancel/submit
[ 285.042569] APW: drm_helper_hpd_irq_event: cancel/submit
[ 285.062593] APW: drm_helper_hpd_irq_event: cancel/submit
[ 285.082683] APW: drm_helper_hpd_irq_event: cancel/submit
Later when things calm down we see them but much much less often:
[ 379.038239] APW: drm_helper_hpd_irq_event: cancel/submit
[ 379.044904] APW: drm_helper_hpd_irq_event: cancel/submit
[ 382.438106] APW: drm_helper_hpd_irq_event: cancel/submit
[ 382.444754] APW: drm_helper_hpd_irq_event: cancel/submit
[ 385.838072] APW: drm_helper_hpd_irq_event: cancel/submit
[ 385.844735] APW: drm_helper_hpd_irq_event: cancel/submit
[ 389.237876] APW: drm_helper_hpd_irq_event: cancel/submit
[ 389.244523] APW: drm_helper_hpd_irq_event: cancel/submit
[ 392.638048] APW: drm_helper_hpd_irq_event: cancel/submit
[ 392.644740] APW: drm_helper_hpd_irq_event: cancel/submit
-apw
next prev parent reply other threads:[~2010-06-16 13:33 UTC|newest]
Thread overview: 15+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-06-09 18:36 Why is kslowd accumulating so much CPU time? Theodore Ts'o
2010-06-09 18:56 ` David Howells
2010-06-09 19:00 ` David Howells
2010-06-13 8:23 ` Dave Airlie
2010-06-13 19:49 ` tytso
2010-06-13 20:00 ` Dave Airlie
2010-06-14 18:42 ` Nick Bowler
2010-06-14 21:46 ` tytso
2010-06-15 17:25 ` Andy Whitcroft
2010-06-15 18:07 ` David Howells
2010-06-16 11:37 ` Andy Whitcroft
2010-06-16 13:33 ` Andy Whitcroft [this message]
2010-06-16 14:36 ` Nick Bowler
2010-06-16 14:55 ` Andy Whitcroft
2010-06-17 13:25 ` Nick Bowler
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=AANLkTimTAqERLRx-eSpbeXXwMh3K9f8FaZOeNY-oNmdL@mail.gmail.com \
--to=apw@canonical.com \
--cc=airlied@gmail.com \
--cc=airlied@redhat.com \
--cc=dhowells@redhat.com \
--cc=linux-kernel@vger.kernel.org \
--cc=tytso@mit.edu \
/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).