[pulseaudio-tickets] [Bug 95055] pulseaudio blocked in kernel

[bugzilla-daemon at freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=95055

--- Comment #23 from Brian J. Murrell <brian at interlinx.bc.ca> ---
(In reply to Raymond from comment #19)
> ftp://download.nvidia.com/XFree86/gpu-hdmi-audio-document/gpu-hdmi-audio.
> html#_verify_your_eld_is_validPorts
> 
> 
> 		hdmi-output-0: HDMI / DisplayPort (priority: 5900, latency offset: 0 usec,
> available)
> 			Properties:
> 				device.icon_name = "video-display"
> 				device.product.name = "24MP76"
> 			Part of profile(s): output:hdmi-stereo
> 
> 
> 
> did your graphic driver get the correct EDID of your LG 24MP76 and pass ELD
> to audio driver ?

I don't (usually) have any monitors connected to the HDMI.  That must have been
a/the one time I did to see if having a monitor on the HDMI port made the
problem any better or worse.

(In reply to Alexander E. Patrakov from comment #20)
> 
> ...and there are no monitors connected to the NVidia GPU. So maybe it is
> powered down. Try to disable this by adding this parameter to the kernel
> command line:
> 
> nouveau.runpm=0

OK.  That will take effect on my next reboot.

> Anyway, this is a kernel bug, not something that can be fixed in PulseAudio.
> All we can do here is to gather some evidence.

Happy to help with that in any way I can.

> Please verify (without the above parameter) whether this command also gets
> blocked in the kernel. Run it repeatedly just in case.
> 
> amixer -c2

I only had to run it once:
$ amixer -c2
[ didn't return]
$ ps axf | grep amixer
27202 pts/1    D+     0:00  |   \_ amixer -c2

Here's the stack trace of it:

 task                        PC stack   pid father
amixer          D ffff8802f995bb98     0 27202   3824 0x00000000
ffff8802f995bb98 ffff88049d6cda00 ffff88046a3b9e00 ffff8802f995c000
ffff8802f995bbd0 ffff8804afb0e080 ffff8804afb0e080 ffff88049c17a000
ffff8802f995bbb0 ffffffff8179cf95 000000010b2a9d8b ffff8802f995bc60
Call Trace:
[<ffffffff8179cf95>] schedule+0x35/0x80
[<ffffffff8179fdb3>] schedule_timeout+0x123/0x270
[<ffffffff8110d350>] ? trace_event_raw_event_tick_stop+0x120/0x120
[<ffffffff817a0b5e>] ? _raw_spin_unlock_irqrestore+0xe/0x10
[<ffffffffa05dbf05>] snd_power_wait+0xb5/0x110 [snd]
[<ffffffff810cddd0>] ? wake_up_q+0x70/0x70
[<ffffffffa05dd901>] snd_ctl_elem_info_user+0x61/0xf0 [snd]
[<ffffffffa05e034c>] snd_ctl_ioctl+0x5ec/0x6c0 [snd]
[<ffffffff813407fc>] ? selinux_file_ioctl+0x10c/0x1c0
[<ffffffff81241428>] do_vfs_ioctl+0x298/0x480
[<ffffffff813380b3>] ? security_file_ioctl+0x43/0x60
[<ffffffff81241689>] SyS_ioctl+0x79/0x90
[<ffffffff817a10ee>] entry_SYSCALL_64_fastpath+0x12/0x71

> Also try:
> 
> alsamixer -c2
> 
> and mute/unmute various spdifs repeatedly.
> 
> Finally:
> 
> time pasuspender -- aplay -d 5 -D hdmi:2 -f dat /dev/zero

Do you still want the above, given that amixer has blocked in the kernel?  If
so, I assume you want those if/when amixer does finally return?

(In reply to Raymond from comment #22)
> https://git.kernel.org/cgit/linux/kernel/git/tiwai/sound.git/tree/sound/pci/
> hda/hda_eld.c
> 
> do you have any output of snd_hdmi_print_eld_info

How/where would I get that?

-- 
You are receiving this mail because:
You are the QA Contact for the bug.
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.freedesktop.org/archives/pulseaudio-bugs/attachments/20160430/14a184b1/attachment.html>