[alsa-devel] via82xx: unreliable audio, strange mixer behavior

[Takashi Iwai]

At Mon, 24 Nov 2008 15:03:06 -0500,
Forest Bond wrote:
> 
> Hi,
> 
> On Fri, Nov 21, 2008 at 05:36:49PM +0100, Takashi Iwai wrote:
> > At Fri, 21 Nov 2008 11:32:56 -0500,
> > Forest Bond wrote:
> > > 
> > > On Fri, Nov 21, 2008 at 01:54:43PM +0100, Takashi Iwai wrote:
> > > > At Fri, 21 Nov 2008 07:08:23 -0500,
> > > > Forest Bond wrote:
> > > > > On Fri, Nov 21, 2008 at 12:59:10PM +0100, Takashi Iwai wrote:
> > > > > > At Thu, 20 Nov 2008 12:04:09 -0500,
> > > > > > Forest Bond wrote:
> > > > > >> 
> > > > > >> Hi,
> > > > > >> 
> > > > > >> I'm using snd-via82xx on an Ubuntu 8.04 system.  alsa-info.sh output is here:
> > > > > >> 
> > > > > >> http://www.alsa-project.org/db/?f=1c3d5570e7b57d4d2f79b964a323dd7b4efd7e78
> > > > > >> 
> > > > > >> Here are the symptoms:
> > > > > >> 
> > > > > >> Audio goes away after a semi-random interval.  Sometimes it returns after
> > > > > >> `modprobe -r snd-via82xx' followed by `modprobe snd-via82xx', but sometimes it
> > > > > >> seems to take a full reboot.
> > > > > >> 
> > > > > >> This looks similar to:
> > > > > >> 
> > > > > >> https://bugtrack.alsa-project.org/alsa-bug/view.php?id=3D3194
> > > > > >> 
> > > > > >> Some folks think that this kind of issue is triggered by audio volume being
> > > > > >> above a certain level.  I'm not entirely sure if this is true, but I think it
> > > > > >> is.  One time, I apparently triggered the issue by unplugging and replugging the
> > > > > >> audio cable from the mainboard during playback.
> > > > > >> 
> > > > > >> Audio can be heard even when the PCM volume level is 0 (as told by alsamixer).
> > > > > >> The volume does get louder when I turn up that mixer control, but it never goes
> > > > > >> away entirely when I turn the volume to zero.  Mute works as expected.
> > > > > > 
> > > > > > The volume 0 doesn't always mean the mute on several hardwares.
> > > > > > So this can be irrelevant if it happens always.
> > > > > > 
> > > > > > I vaguely remember about the low-power mode of some codec chips.
> > > > > > Which codec chip does your system have?  Please run alsa-info.sh with
> > > > > > --no-upload option, and attach the output file.
> > > > > 
> > > > > The output would be the same as what was already uploaded, right?  I've attached
> > > > > that output here for your convenience.
> > > > 
> > > > Thanks, it's always better to post the content than a URL.
> > > > 
> > > > Any chance to try the later version of alsa-driver?
> > > > Yours looks pretty old.
> > > 
> > > I can try this a bit later today.  I'll report the results ASAP.
> > 
> > BTW, try the one from the URL below instead of 1.0.18a.
> > This contains the newest  alsa-driver snapshot:
> >     ftp://ftp.kernel.org/pub/linux/kernel/people/tiwai/snapshot/alsa-driver-snapshot.tar.gz
> 
> Brief testing indicates that 1.0.18a seems to fix the issue, actually.  Sorry, I
> forgot that you had asked me to test the snapshot.
> 
> Would it be helpful for you for me to try the snapshot as well (maybe to check
> for a regression)?

Yes, it'd be helpful.

>  Any idea as to what change may have fixed this?

It's about the low-power mode of VIA ac97 codec chip.


Takashi