[pulseaudio-tickets] [PulseAudio] #329: CPU time limit stops deamon

[PulseAudio]

#329: CPU time limit stops deamon
---------------------+------------------------------------------------------
  Reporter:  sladi   |       Owner:  lennart              
      Type:  defect  |      Status:  new                  
  Priority:  normal  |   Milestone:                       
 Component:  daemon  |    Severity:  normal               
Resolution:          |    Keywords:  cpu time limit daemon
---------------------+------------------------------------------------------
Comment (by boi):

 I'm experiencing the same problem on my machine. I'm running Fedora 9
 x86_64, which currently uses PulseAudio 0.9.8. I've experienced this
 problem since this summer (not sure exactly when).

 The machine is an Athon X2 3800+, with AMD690 chipset ("shared memory" - I
 don't play games, so I don't care about blazing graphics performance) and
 the ATI SB600. I believe the codec is Realtek ALC883.

 {{{
 $ uname -r -v -m -o
 2.6.26.6-49.fc8 #1 SMP Fri Oct 17 15:33:32 EDT 2008 x86_64 GNU/Linux

 $ rpm -q pulseaudio
 pulseaudio-0.9.8-5.fc8

 $ cat /proc/asound/cards [[BR]]

  0 [SB             ]: HDA-Intel - HDA ATI SB
                       HDA ATI SB at 0xfe020000 irq 16
 }}}

 The only message from the driver I see at all occurs always right after
 starting PulseAudio:
 Nov 20 09:11:37 freebie kernel: hda-intel: Invalid position buffer, using
 LPIB read method instead.

 I've tried using the position_fix parameter, but to no avail. To be
 honest, I didn't try much with kernel module parameters. I don't have
 issues with my sound not working at all or excessive skipping, which is
 what usually pops up when you look for snd-hda-intel parameters.

 I can actually somewhat reproduce the problems: when I scroll a page in
 Firefox, either by keypresses or mouse, music sometimes skips or even
 hangs completely. If I restart PulseAudio by hand, the problem will
 usually pop up again, and I'll get the dreaded CPU limit exhausted
 messages. Other events seem to trigger this as well (eg. screen
 unblanking, switching desktops), but not nearly as reliably.

 I've also seen and heard it stop when I wasn't even close to my computer,
 so user input is not the only trigger. Nor does my system have to be
 really busy when sound goes dead.

 Applications don't seem to matter. I have the same issues with Amarok,
 vlc, mplayer and Flash.


 I hope this is useful. Feel free to contact me if you need more info - it
 is really annoying. Last few weeks, I sometimes had to restart the bugger
 10 times a day.

-- 
Ticket URL: <http://www.pulseaudio.org/ticket/329#comment:9>
PulseAudio <http://pulseaudio.org/>
The PulseAudio Sound Server