Bug 454712 - pulseaudio daemon crashes when enabling bluetooth audio device
pulseaudio daemon crashes when enabling bluetooth audio device
Status: CLOSED WONTFIX
Product: Fedora
Classification: Fedora
Component: pulseaudio (Show other bugs)
9
x86_64 Linux
low Severity high
: ---
: ---
Assigned To: Lennart Poettering
Fedora Extras Quality Assurance
:
Depends On:
Blocks:
  Show dependency treegraph
 
Reported: 2008-07-09 16:49 EDT by Carl Roth
Modified: 2008-09-09 15:59 EDT (History)
2 users (show)

See Also:
Fixed In Version:
Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of:
Environment:
Last Closed: 2008-09-09 15:59:47 EDT
Type: ---
Regression: ---
Mount Type: ---
Documentation: ---
CRM:
Verified Versions:
Category: ---
oVirt Team: ---
RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: ---


Attachments (Terms of Use)
gdb session trace from pulseaudio (11.15 KB, text/plain)
2008-07-09 16:52 EDT, Carl Roth
no flags Details

  None (edit)
Description Carl Roth 2008-07-09 16:49:35 EDT
Description of problem:

I was trying to figure out a workaround to BUG454672, where I have a
multi-device pulseaudio setup with not all devices active at the same time...  I
was playing around with pacmd, seeing if I could interactively load pulseaudio
modules.

With my setup, if I enable a bluetooth audio device at startup (in default.pa or
thereabouts) then the device works correctly.

If I start up pulseaudio without the BT device defined, but then go in with
pacmd and try to load the bluetooth support, it crashes pulseaudio.

With all of the devices defined, here is a snippet from my default.pa:

  load-module module-alsa-sink device=intelHDA sink_name=Spkr rate=192000
  set-default-sink Spkr

  .nofail
  load-module module-alsa-sink device=plantronicsUSB sink_name=HeadPh rate=44100
  set-default-sink HeadPh
  .fail

  .nofail
  load-module module-alsa-sink device=bluetooth sink_name=bluetooth rate=44100
  set-default-sink bluetooth
  .fail

This configuration works (as long as the BT device is powered on, of course --
if the BT is powered off then pulseaudio stalls for a long while and also
sometimes fails to start at all)

If I comment out the last section and feed it line-by-line to a running
pulseaudio daemon using 'pacmd', the daemon crashes on the 'load-module' line.

Version-Release number of selected component (if applicable):

pulseaudio-0.9.10-1.fc9.x86_64
bluez-utils-3.35-2.fc9.x86_64
alsa-lib-1.0.16-3.fc9.x86_64

How reproducible:


Steps to Reproduce:
1.
2.
3.
  
Actual results:


Expected results:


Additional info:
Comment 1 Carl Roth 2008-07-09 16:52:33 EDT
Created attachment 311418 [details]
gdb session trace from pulseaudio
Comment 2 Carl Roth 2008-07-09 16:53:47 EDT
I attached a sample gdb session from the pulseaudio failure to see if it would
reveal anything.  Note from the trace that 1) many of the stack parameters are
redacted ("value optimized out") and 1) gdb eventually fails with an internal
error.  Ouch.
Comment 3 Lennart Poettering 2008-09-09 15:57:10 EDT
This seems to be a bug in the BT module for ALSA, not in PA. It is known to be buggy. We now have patches that link PA directly to BlueZ, which will avoid this mess. 

I will close this bug now, since it is not a PA bug. Please try the bluezified PA when we have this in rawhide.

Note You need to log in before you can comment on or make changes to this bug.