[pulseaudio-discuss] [Alsa-user] Pops/Crackles Messing up my audio...

Nasa nasa01 at comcast.net
Mon Aug 8 05:04:03 PDT 2011



----- Original Message -----
> 'Twas brillig, and Nasa at 07/08/11 21:26 did gyre and gimble:
> > Hi,
> >
> > I was hoping I could get some help troubleshooting some audio
> > quality issues I am running into.
> > Specifically, I am getting a lot of *pops/crackles* when I play
> > audio files. It shows up when I
> > move from 2-channel to 4/5 channel audio (via pauvcontrol) - I am
> > using a USB X-FI sound-card with
> > the Meego IVI.
> >
> > I have tried changing the resample method (going from ffmpeg to
> > high-quality) and setting tsched=0,
> > neither making any noticeable difference.
> >
> > The only things I saw in /var/log/messages that seemed relevant was:
> > messages:Aug 6 16:59:26 localhost pulseaudio[570]: alsa-sink.c: ALSA
> > woke us up to write new data to the device, but there was actually
> > nothing to write!
> > messages:Aug 6 16:59:26 localhost pulseaudio[570]: alsa-sink.c: Most
> > likely this is a bug in the ALSA driver 'snd_usb_audio'. Please
> > report this issue to the ALSA developers.
> > messages:Aug 6 16:59:26 localhost pulseaudio[570]: alsa-sink.c: We
> > were woken up with POLLOUT set -- however a subsequent
> > snd_pcm_avail() returned 0 or another value < min_avail.
> >
> > And this was spout out of dmesg
> > [ 21.857184] ALSA sound/usb/mixer.c:2110: status interrupt: c0 00
> > [ 21.897058] ALSA sound/usb/mixer.c:2110: status interrupt: c0 00
> > [ 21.961062] ALSA sound/usb/mixer.c:2110: status interrupt: c0 00
> >
> > I have already eliminated the speakers and the amp (played music
> > from a different source and
> > they sounded fine).
> >
> > Let me know what else I should post (I could record the sound, if
> > that would be useful).
> >
> > Thanks in advance,
> 
> While this could be an issue in the alsa driver itself, can I ask if
> you're applying any volume changes to your streams or sink?
> 
I personally am not, however, the meego project maybe.  There are over 
250 patches against the base pulseaudio package (9.22) of which some 
seem to have something to do with volume settings...

0249-bluetooth-Fix-HSP-volume-handling.patch     
0248-bluetooth-restore-original-sco_-sink-src-set_volume-.patch  
0247-bluetooth-fix-set_volume_cb-on-sco-over-pcm.patch  
0240-volume-proxy-small-lib-to-allow-communicating-volume.patch
0224-pactl-Accept-more-volume-specification-formats.patch     
0223-sink-input-Add-volume_writable-to-pa_sink_input.patch 
0190-alsa-mixer-Refactoring-merge-element_mute_volume-ele.patch      
0189-alsa-mixer-Implement-constant-volume.patch  
0140-dbus-Always-accept-mono-volumes-when-setting-device-.patch
0133-volume-Add-Orc-based-optimised-volume-scaling.patch      
0132-volume-Fix-sample-array-size-for-tests.patch      
0131-volume-Make-tests-use-only-valid-volumes.patch  
0130-alsa-mixer-Fix-a-git-am-cockup-in-b0f72311.patch  
0129-volume-Add-a-PA_VOLUME_UI_MAX-define-for-the-recomme.patch  
0124-introspect-Client-side-implementation-for-has_volume.patch 
0118-Revert-Add-volume-ramping-feature-envelop-fix.patch      
0117-Revert-Add-volume-ramping-feature-sink-input-modific.patch      
0116-Revert-Add-volume-ramping-feature-sink-modification.patch 
0114-Revert-core-volume-ramping-fix.patch
0107-virtual-sink-Add-a-modarg-for-forcing-flat-volume.patch      
0106-virtual-sink-Add-a-modarg-for-enabling-volume-sharin.patch      
0105-Implement-the-volume-sharing-feature.patch  
0090-Allow-read-only-or-non-existing-sink-input-volume.patch
0042-core-Use-volume_change_safety_margin-when-rewinding-.patch
0022-volume-Trivial-cosmetics-remove-a-space.patch
0011-alsa-sink-take-base-volume-into-account-when-applyin.patch
0001-volume-Add-explicit-checks-for-ARMv6-instructions.patch
0001-fix-the-assumption-that-volume-is-always-positive.patch

So can see all the patches and what's in them here:
https://build.pub.meego.com/package/files?package=Pulseaudio&project=home%3Anasa
(not that I'm expecting and/or requesting you do that -- I just
put this here for reference)

Nasa

BTW: I will try your suggestion when I get home this evening.


> There could be a problem with optimized paths for software volume
> adjustments. You can disable these optimisations via a special
> environment var: PULSE_NO_SIMD=1
> 
> It's worth checking this to see if it's that area that's at fault.
> 
> Col
> 
> 


More information about the pulseaudio-discuss mailing list