Occasional Pop sounds (new in Jaunty Beta)

Changed to pulseaudio, but to be honest I'm now getting confused by where this is coming from.
Today I'm running a KDE session rather than Gnome and I'm not hearing any popping in Amarok2 but also I'm not hearing popping in Rhythmbox (which still uses Pulse).

Same thing here, but on 32-bit:

Intel(R) Core(TM) Solo CPU U1400 @ 1.20GHz

Linux eesti 2.6.28-11-generic #38-Ubuntu SMP Fri Mar 27 09:00:52 UTC 2009 i686 GNU/Linux

$ cat /proc/asound/cards
 0 [Intel ]: HDA-Intel - HDA Intel
                      HDA Intel at 0xefebc000 irq 21

$ cat /proc/asound/devices
  2: : timer
  3: : sequencer
  4: [ 0- 0]: digital audio playback
  5: [ 0- 0]: digital audio capture
  6: [ 0] : control

Sound was running impeccably here until Intrepid, but broke just after the Jaunty repo was created. Using PA on a GNOME desktop, in both cases. Downgrading PA packages to those from Intrepid restores operation.

OK, some more info; I've spent the day in KDE and as per my previous comment - not a pop.
Also, back in Gnome (after a reboot) I have the pops again.

But another issue that may be related is that if I mute the audio (on the volume control) then it's quiet; if I mute the audio and set rhythmbox playing I hear a lot of popping and noise - a lot more than when it's unmuted.
I've tried all the channels I can see with all the different choices for device and I've tried aumix - I can't get the noise
when muted to go away.

$ cat /proc/asound/devices
  2: : timer
  3: : sequencer
  4: [ 0- 6]: digital audio playback
  5: [ 0- 6]: digital audio capture
  6: [ 0- 0]: digital audio playback
  7: [ 0- 0]: digital audio capture
  8: [ 0] : control

I'm not convinced these symtoms aren't due to ALSA-kernel, in which case of course PulseAudio would be adversely affected. Please download and run http://www.alsa-project.org/alsa-info.sh explicitly using bash.

Also, I need to know what audiosinks GSt is configured to use ("gconftool -g /system/gstreamer/0.10/default/audiosink ; gconftool -g /system/gstreamer/0.10/default/musicaudiosink"), and I also need output from "pgrep -f /usr/bin/pulseaudio ; sudo fuser -v /dev/dsp* /dev/snd/*" when you're in KDE using Rhythmbox.

OK 1st part; attached is my output of alsa-info.sh (also at http://www.alsa-project.org/db/?f=b5c6516c7cac4ed98fb2848716669f2f5dbb3cbf ) - this was captured after a reboot with a gnome session running rhythmbox, firefox and a few shells.

dg@davros:~$ gconftool -g /system/gstreamer/0.10/default/audiosink ; gconftool -g /system/gstreamer/0.10/default/musicaudiosink
autoaudiosink
autoaudiosink

and in gnome the following output:

dg@davros:~$ pgrep -f /usr/bin/pulseaudio ; sudo fuser -v /dev/dsp* /dev/snd/*
4386
                     USER PID ACCESS COMMAND
/dev/snd/controlC0: dg 4386 F.... pulseaudio
                     dg 4537 F.... mixer_applet2
/dev/snd/pcmC0D0p: dg 4386 F...m pulseaudio

KDE follows shortly.

Dave

and now in the land of the green dragon. Clean reboot into KDE; it restored session.

New output of alsa-info.sh attached

dg@davros:~$ gconftool -g /system/gstreamer/0.10/default/audiosink ; gconftool -g /system/gstreamer/0.10/default/musicaudiosink
autoaudiosink
autoaudiosink

dg@davros:~$ pgrep -f /usr/bin/pulseaudio ; sudo fuser -v /dev/dsp* /dev/snd/*
4336
                     USER PID ACCESS COMMAND
/dev/snd/controlC0: dg 4311 F.... knotify4
                     dg 4336 F.... pulseaudio
                     dg 4341 F.... kmix
/dev/snd/pcmC0D0p: dg 4336 F...m pulseaudio

Dave

requested info supplied

Two additional notes:

I notice in my logs:

[ 3379.511209] hda-intel: IRQ timing workaround is activated for card #0. Suggest a bigger bdl_pos_adj.

don't know if that is relevant.

(Current kernel is 2.6.28-11-generic #41-Ubuntu SMP Wed Apr 8 04:39:23 )

also I think many of the clicks may be associated with things that are doing some GUI work - but I'm not sure;certainly nothing heavy, but for example opening a new terminal seems to be a bit more associated with it.

Dave

On Sun, 12 Apr 2009, Dave Gilbert wrote:

> [ 3379.511209] hda-intel: IRQ timing workaround is activated for card
> #0. Suggest a bigger bdl_pos_adj.
...
> also I think many of the clicks may be associated with things that are
> doing some GUI work - but I'm not sure;certainly nothing heavy, but for
> example opening a new terminal seems to be a bit more associated with
> it.

Please try the test kernel mentioned in bug 345627.

Hi Daniel,
  Well there's good news and bad news about that kernel (2.6.28-12-generic #42~crimsun1lp345627 SMP Sat Apr 11 02:00:26 UTC linux-image-2.6.28-12-generic_2.6.28-12.42~crimsun1lp345627_amd64.deb)

The good news is I've not heard any popping.

The bad news is that a bug that I had early in Jaunty (maybe pre-the Beta) that had gone away in the last weeks has come back - and it's a worse bug. This bug is that every so often the audio goes nuts for many seconds, complete static, and in the case of Rhythmbox it actually sees an error (not sure what the error is - but it's put an error mark next to the track). I've been running the kernel for a couple of hours now, and that's happened twice; I'm running Rhythmbox and the first time it did it was when I started a flash/youtube video and the second time was at exactly the same time I started a (tiny) gcc compilation in another window - I'm not sure what I can draw from those two cases.

There is a :
[ 767.162033] pulseaudio[4227]: segfault at 7f9cb637dfd0 ip 00007f9cb637dfd0 sp 00007fffc702c808 error 4 in module-null-sink.so[7f9cb6ba4000+3000]

in the log, I suspect that might coincide to the first (flash video) crash. There is only one such seg in the log.

To be clear, it doesn't happen on every flash video play, or every compilation - and it really seems like the same bug that got fixed a few versions back; I think it was the one that spammed the logs with lots of errors.

Ah, looking at user.log I'm seeing:
Apr 12 21:59:19 davros pulseaudio[5193]: cpulimit.c: Received request to terminate due to CPU overload.
Apr 12 21:59:30 davros pulseaudio[5919]: main.c: Called SUID root and real-time and/or high-priority scheduling was requested in th
e configuration. However, we lack the necessary privileges:
Apr 12 21:59:30 davros pulseaudio[5919]: main.c: We are not in group 'pulse-rt', PolicyKit refuse to grant us the requested privile
ges and we have no increase RLIMIT_NICE/RLIMIT_RTPRIO resource limits.

and I have two of those at about the right times, so I suspect that happened in both cases.

(Core 2 Duo 1.66GHz, so it's hardly CPU heavy)

Dave

On Sun, 12 Apr 2009, Dave Gilbert wrote:

> The good news is I've not heard any popping.

I'm only concerned about this symptom for this bug report.

* Daniel T Chen (<email address hidden>) wrote:
> On Sun, 12 Apr 2009, Dave Gilbert wrote:
>
> > The good news is I've not heard any popping.
>
> I'm only concerned about this symptom for this bug report.

OK, I'll keep listening again tomorrow, but I think this verison isn't doing pop sounds.

Dave

> --
> Occasional Pop sounds (new in Jaunty Beta)
> https://bugs.launchpad.net/bugs/350274
> You received this bug notification because you are a direct subscriber
> of the bug.
>
> Status in ???pulseaudio??? source package in Ubuntu: New
>
> Bug description:
> Hi,
> Just updated to Jaunty Beta (x64) this morning (from Jaunty Alpha as of last weekend) and I'm occasional (every few minutes) *POP* sounds very audible over my rhythmbox audio playback.
>
> I've never had this on this machine - and it's run all the way up from Hoary; it also doesn't sound like the type of thing you get when the app runs out of CPU. Originally I thought it was because of tracker-indexer eating my machine, but it's still doing it with that killed off.
>
> Machine: Toshiba Equium A100-306:
>
> 00:1b.0 Audio device: Intel Corporation 82801G (ICH7 Family) High Definition Audio Controller (rev 02)
> Subsystem: Toshiba America Info Systems Device ff10
> Flags: bus master, fast devsel, latency 0, IRQ 22
> Memory at dc240000 (64-bit, non-prefetchable) [size=16K]
> Capabilities: <access denied>
> Kernel driver in use: HDA Intel
> Kernel modules: snd-hda-intel
>
> /proc/asound/cards:
> 0 [Intel ]: HDA-Intel - HDA Intel
> HDA Intel at 0xdc240000 irq 22
>
> uname -a:
> Linux davros 2.6.28-11-generic #38-Ubuntu SMP Fri Mar 27 10:01:17 UTC 2009 x86_64 GNU/Linux
>
>
> Dave
--
 -----Open up your eyes, open up your mind, open up your code -------
/ Dr. David Alan Gilbert | Running GNU/Linux on Alpha,68K| Happy \
\ gro.gilbert @ treblig.org | MIPS,x86,ARM,SPARC,PPC & HPPA | In Hex /
 \ _________________________|_____ http://www.treblig.org |_______/