freebsd stable - Mar 2005 - consistent problem with pcm0 audio

Hello,

I am using FreeBSD 5-STABLE from Feb 23rd with an onboard VIA 
VT8237/Avance Logic ALC200 sound controller. Periodically, 
sound will stop working. I will get the last note (sound
packet?) repeated over and over until I close the program. Any
further attempt to play a sound will cause the same note to be 
repeated. I am not able to reload the driver. Correct sound is 
restored when I reboot the computer.

This has been happening for awhile and I am not certain if it
was introduced with any particular update.

I am using ACPI. I don't know if this also happens without ACPI.

I also get a message on the console like this when the error 
occurs:

	pcm0:play:0: play interrupt timeout, channel dead

uname: 

FreeBSD tyr.housing.ufl.edu 5.3-STABLE FreeBSD 5.3-STABLE #2: Wed Feb 23 18:53:5
7 EST 2005     wills@tyr.housing.ufl.edu:/usr/obj/usr/src/sys/kernel  amd64

pciconf:

pcm0@pci0:17:5: class=0x040100 card=0x13001462 chip=0x30591106 rev=0x60 hdr=0x00
    vendor   = 'VIA Technologies Inc'
    device   = 'VT8233/33A/8235/8237 AC97 Enhanced Audio Controller'
    class    = multimedia
    subclass = audio

-Will

--
Will Saxon
Systems Programmer, Network Services
University of Florida Department of Housing
(352) 392-2171 x10148
wills@housing.ufl.edu

On 2005-03-08, Will Saxon scribbled these
curious markings:
> I am using FreeBSD 5-STABLE from Feb 23rd with an onboard VIA 
> VT8237/Avance Logic ALC200 sound controller. Periodically, 
> sound will stop working. I will get the last note (sound
> packet?) repeated over and over until I close the program. Any
> further attempt to play a sound will cause the same note to be 
> repeated. I am not able to reload the driver. Correct sound is 
> restored when I reboot the computer.
Couple of things that I've found will muck up your sound:

o ACPI, as you mentioned. Be glad that you can even get your system to
boot with ACPI. Mine crashes within 10 seconds of hitting Enter at the
boot menu.

o Threading libraries. games/doomlegacy exhibits behaviour endemic of a
sound buffering / channel issue (sound stuttering and playing slowly
during periods of high activity, working better during periods of less
activity) for me when it runs with libpthread. I spent *hours* (okay,
I'm a Doom addict, I'll admit it) trying to diagnose the problem in the
mindset that it was a buffering / channel problem, when all I had to do
was set it to use libc_r in libmap.conf.

Best Regards,
Christopher Nehren
-- 
I abhor a system designed for the "user", if that word is a coded
pejorative meaning "stupid and unsophisticated". -- Ken Thompson
If you ask the wrong questions, you get answers like "42" and
"God".
Unix is user friendly. However, it isn't idiot friendly.