Re: snd-intel8x0 buggy on Nvidia CK804, TYAN, AMD Opteron board?

[Takashi Iwai <tiwai@suse.de>]

At Sun, 22 May 2005 15:59:51 +0200,
Christian Parpart wrote:
> 
> On Friday, 20. May 2005 10:58, Takashi Iwai wrote:
> > At Fri, 20 May 2005 09:38:15 +0200,
> >
> > Christian Parpart wrote:
> > > Hi all,
> > >
> > > well, I'm just trying to listen to my music, however, it's either a no-go
> > > (using ALSA-OSS emulation) or just a plain pain to listen to (via ALSA
> > > directly).
> > >
> > > In first case, I just hear nothing. A `dd if=/dev/urandom of=/dev/dsp`
> > > stops at a certain byte and in my headset I hear a very high beep tone.
> > >
> > > in second case, the music seems very deformed and the output is very
> > > buggy at all (meaning, that it played just for a few minutes).
> > > deformed means, that the foreground singer has been somewhat in the very
> > > background and it overall has been very unfunny to listen to.
> > >
> > > I was trying different players and versions anyway.
> > >
> > > So, is this supposed to be a bug in the kernel sound driver for my
> > > certain hardware?
> >
> > Do you run cpufreq or something related with that?
> 
> Nope, sorry, I even didnt install any of these cpufreq-alike tools on my host:
> AMD Opteron on a TYAN board with nForce4 chipset.

Well, I don't mean to activate cpufreq.  Rather cpufreq may be
sometimes the cause of output noises.

Does the noise also on the analog loopback, e.g. CD or line-in?
If yes, it's likely a problem of hardware, either the broken h/w or
the bad plug/cable.  Or, it's an inproper mixer configuration.
To check the mixer configuration, show the contents of
/etc/asound.state (it's generated via "alsactl store"), and
/proc/asound/card0/codec97#0/* files.

OTOH, if this happens only on PCM, it's often an interrupt problem
(typically ACPI) or the influence of cpufreq...


Takashi