public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Ingo Molnar <mingo@elte.hu>
To: Mark Knecht <markknecht@gmail.com>
Cc: lkml <linux-kernel@vger.kernel.org>, Lee Revell <rlrevell@joe-job.com>
Subject: Re: 2.6.14-rt1 - xruns in a certain circumstance
Date: Mon, 31 Oct 2005 15:22:04 +0100	[thread overview]
Message-ID: <20051031142204.GA6136@elte.hu> (raw)
In-Reply-To: <5bdc1c8b0510301828p29ea517ew467a5f6503435314@mail.gmail.com>


* Mark Knecht <markknecht@gmail.com> wrote:

> What I'm seeing is that when using basic menus, or when watching 
> videos I get no xruns. However, if I'm in the preview menu I get an 
> xrun every few minutes. [...]

this could be some sort of hardware latency, as Lee suspects. Videocards 
are known to be pretty agressively holding the system bus, for the last 
few percentiles of Quake performance ... Also, mainboard chipsets are 
sometimes not that good at enforcing fairness between DMA agents - 
possibly starving the CPU itself for lengthly amounts of time. We have 
seen such incidents before, and latency tracing ought to be able to show 
this with reasonable certainty. If it's some sort of generic hardware 
latency then you ought to see weird traces when enabling WAKEUP_TIMING 
and LATENCY_TRACING in the .config. No need for any other debug options 
or Jack level hackery at this point - just enable these and do a:

	echo 0 > /proc/sys/kernel/preempt_max_latency

and try to trigger as large latencies as possible via MythTV. (you wont 
necessarily see a large latency reported by the kernel when you see an 
xrun. We can trace xruns too, but that needs Jackd changes and is more 
effort to set up.)

	Ingo

  parent reply	other threads:[~2005-10-31 14:21 UTC|newest]

Thread overview: 13+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2005-10-31  2:28 2.6.14-rt1 - xruns in a certain circumstance Mark Knecht
2005-10-31  6:49 ` Lee Revell
2005-10-31 14:22 ` Ingo Molnar [this message]
2005-10-31 16:18   ` Lee Revell
2005-10-31 15:10 ` K.R. Foley
2005-10-31 15:26   ` Mark Knecht
2005-10-31 15:35     ` K.R. Foley
2005-10-31 16:39     ` Lee Revell
2005-10-31 17:38       ` Mark Knecht
2005-10-31 23:22       ` Mark Knecht
2005-11-01  1:48         ` Nuno Silva
2005-11-01  7:37           ` Lee Revell
2005-11-01  7:54           ` Lee Revell

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=20051031142204.GA6136@elte.hu \
    --to=mingo@elte.hu \
    --cc=linux-kernel@vger.kernel.org \
    --cc=markknecht@gmail.com \
    --cc=rlrevell@joe-job.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