Bug 83446 - PulseAudio crash caused by automatic deactivation of dedicated GPU's soundcard via vgaswicheroo
Summary: PulseAudio crash caused by automatic deactivation of dedicated GPU's soundca...
Status: RESOLVED NOTOURBUG
Alias: None
Product: PulseAudio
Classification: Unclassified
Component: alsa (show other bugs)
Version: unspecified
Hardware: x86-64 (AMD64) Linux (All)
: medium normal
Assignee: pulseaudio-bugs
QA Contact: pulseaudio-bugs
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2014-09-03 16:27 UTC by Paul
Modified: 2014-09-04 05:46 UTC (History)
1 user (show)

See Also:
i915 platform:
i915 features:

Attachments

Description Paul 2014-09-03 16:27:56 UTC 
Hi, 

I'm using a HP EliteBook 850 G1 with Intel HD 4400 / AMD Radeon HD 8750M hybrid graphics on the mesa drivers respectively. 

After using Pulseaudio with a Linux Kernel version greater than 3.12 the sound system hangs as it tries to access the dedicated GPU's integrated soundchip which is deactivated alongside the GPU by the means of run-time power management ( http://lists.freedesktop.org/archives/dri-devel/2013-September/045650.html )
This feature was enabled by default as of Kernel 3.13. 
It seems that Pulseaudio does not recognize that the soundcard on the radeon is no longer available and tries to access it, which it can't. 

I already filed a bug at launchpad which can be found here: https://bugs.launchpad.net/ubuntu/+source/pulseaudio/+bug/1310260
Four other people are affected but unfortunately Canonical didn't pick up this bug report. Raymond already did some basic analysis on the logs.

This problem still persists on Arch Linux, with the latest updates applied so I don't believe it to be a distribution specific problem. 

Dominik on the launchpad bugreport proposed a functioning workaround by inserting "options snd-hda-intel enable=0,1,0" in /etc/modprobe.d . As this effectively kills my HDMI Audio it is obviously not an optimal solution. 

If you need anything please let me know. 
I can set up a testing system on a dedicated SSD to tackle this problem. Please let me know which system you would prefer.
I hope I did everything right as this is my first bug report on freedeskttop.org

Best regards and thanks in advance!
Comment 1 David Henningsson 2014-09-04 05:46:29 UTC 
From your log:

(   0.118|   0.000) D: [pulseaudio] alsa-mixer.c: Checking for playback on Digital Stereo (HDMI) (hdmi-stereo)
(   0.118|   0.000) D: [pulseaudio] alsa-util.c: Trying hdmi:2 with SND_PCM_NO_AUTO_FORMAT ...
(  53.036|   0.011) D: [pulseaudio] alsa-util.c: Managed to open hdmi:2
(  53.036|   0.000) D: [pulseaudio] alsa-util.c: Maximum hw buffer size is 371 ms
(  53.041|   0.004) D: [pulseaudio] alsa-util.c: Set buffer size first (to 3528 samples), period size second (to 441 samples).

It takes 52 seconds to just open the ALSA device "hdmi:2". I assume this is what you mean with "the sound system hangs". And it is most definitely a kernel problem, nothing we can do about that in PulseAudio.

Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.