Re: [LAU] MOTU AVB discussion from LAC

From: Goli4thus <goli@email-addr-hidden>
Date: Sun Jun 02 2019 - 16:40:25 EEST

Hi,

I've read through the following threads:
- https://linuxmusicians.com/viewtopic.php?f=6&t=18046
-
http://linux-audio.4202.n7.nabble.com/Hardware-Soundcard-MOTU-624-AVB-Working-with-Gnu-Linux-Debian-8-7-td103759i60.html

Especially in the latter thread georgnk replied with the following:
http://linux-audio.4202.n7.nabble.com/Hardware-Soundcard-MOTU-624-AVB-Working-with-Gnu-Linux-Debian-8-7-tp103759p105745.html

I'm curious regarding those period clicks/xruns.
Reason being that I own a Motu 624 AVB and have been troubleshooting xruns
for a couple weeks now.

I've ended up filling a bug report:
https://bugzilla.kernel.org/show_bug.cgi?id=203659

It's basically about xruns happening within suspiciously regular time
intervals depending on
alsa settings.
Various system optimizations do not make a difference. (RT kernel, realtime
priorities, cpu max performance, ...)
The attachment within https://bugzilla.kernel.org/show_bug.cgi?id=203659#c9
shows an analysis of
of a ftrace-trace, which shows a clear pattern in the interaction between
xhci and jackdbus whenever
such an xrun occurs.

By any chance, does someone else experience "sporadic" xruns even while the
system is idle,
that are actually of pretty regular nature?
Easiest way to verify this is to run cadence in verbose mode and have a look
at the message log;
specifically xrun messages and their timestamps.
Running at higher sample rates and low buffer size speeds up the occurrence
of these xruns in my case.

-- Goli4thus

--
Sent from: http://linux-audio.4202.n7.nabble.com/linux-audio-user-f5.html
_______________________________________________
Linux-audio-user mailing list
Linux-audio-user@lists.linuxaudio.org
https://lists.linuxaudio.org/listinfo/linux-audio-user
Received on Sun Jun 2 20:15:01 2019

This archive was generated by hypermail 2.1.8 : Sun Jun 02 2019 - 20:15:02 EEST