[pulseaudio-discuss] alsa value < min_avail

[Arun Raghavan]

On Fri, 2012-02-10 at 22:01 +0100, Peter Meerwald wrote:
> Hello,
> 
> I hit the following (Ubuntu natty):
> 
> pmeerw at pmeerw-pc2:~/Desktop$ pulseaudio 
> E: alsa-sink.c: ALSA woke us up to write new data to the device, but there was actually nothing to write!
> E: alsa-sink.c: Most likely this is a bug in the ALSA driver 'snd_hda_intel'. Please report this issue to the ALSA developers.
> E: alsa-sink.c: We were woken up with POLLOUT set -- however a subsequent 
> snd_pcm_avail() returned 0 or another value < min_avail.
> 
> alsa-time-test hw:0 outputs the following 
> ...
> 477445	477433	137324	6253	4220	196	4	1	3
> 477450	477433	137324	6254	4219	197	4	1	3
> 496587	496584	156371	6255	5058	-642	4	1	3
> alsa-time-test: alsa-time-test.c:220: main: Assertion `(unsigned) avail <= 
> buffer_size' failed.
> Aborted
> 
> 00:1b.0 Audio device: Intel Corporation 5 Series/3400 Series Chipset High 
> Definition Audio (rev 05)
> 
> is this still relevant?
> http://www.pulseaudio.org/wiki/BrokenSoundDrivers

How are you running the test? If you run it on a terminal, or even write
to disk (at least on a rotating disk), there will be an underrun and you
will get this assert. Or did you run it as suggested with:
alsa-time-test hw:0 | tail -n 50

Following the mail thread linked to in the wiki link you pointed to --
the jump in the system time and the negative delay signal an underrun.
Given the message in your PA logs though, there's likely a problem
somewhere.

-- Arun