Video / audio troubles with RX 570 / stretch

Everything about X, Gnome, KDE, ... and everything running on it

Re: Video / audio troubles with RX 570 / stretch

Postby llivv » 2018-12-19 00:48

oddhack wrote:I think there's a known BIOS issue that's causing my timeouts - checking with AsRock to see if the latest BIOS might address this, I'd rather not do a BIOS upgrade just on the off chance it might help. .
I wouldn't BIOSupdate either at this point - unless the update specifically listed your issue as fixed. You might want to see what happens by uninstalling WindowMaker...It's one more variable you can remove from the list of possibilities.
oddhack wrote:Run as root, it shows 'Card: HDA ATI HDMI'. F6 says 'Select sound card' but doesn't show anything. alsamixer shows 6 S/PDIF channels but doesn't let me actually adjust the volume on any of them.
Run as my user, it shows 'Card: PulseAudio'. F6 shows two options, '0: HDA ATI HDMI' and '1: HD-Audio Generic'. I am unable to toggle between them because of what looks like some weird ncurses UI bug - hitting F6 shows the menu very very briefly and then returns to the channel control UI before I can select anything.
It seems like /proc/asound/card0 maps onto the 'HDA ATI HDMI' device and /proc/asound/card1 onto the 'HD-Audio Generic'. It is peculiar that pulseaudio only shows the first device, though. I've tried to look for errors from pulseaudio but don't see anything relevant in syslog/user.log. N.b. where does pulseaudio even get started? I don't see anything under /etc/systemd or /etc/init.d that appears to fire off the daemon.

https://wiki.debian.org/PulseAudio

I suggest you create a new user
than add audio,video groups

Are the repeating Call Traces Locking Up the system or does it just take longer to shutdown?
I see between the Call Traces that systemd is still shutting down processes.



Code: Select all
[ /proc/asound/devices ]
  2: [ 0- 3]: digital audio playback
  3: [ 0- 7]: digital audio playback
  4: [ 0- 8]: digital audio playback
  5: [ 0- 9]: digital audio playback
  6: [ 0-10]: digital audio playback
  7: [ 0-11]: digital audio playback
  8: [ 0- 0]: hardware dependent
  9: [ 0]   : control
 10: [ 1- 0]: digital audio playback
 11: [ 1- 0]: digital audio capture
 12: [ 1- 2]: digital audio capture
 13: [ 1- 0]: hardware dependent
 14: [ 1]   : control
 33:        : timer

Code: Select all
[ /proc/asound ]
/proc/asound
/proc/asound/oss
/proc/asound/oss/devices
/proc/asound/oss/sndstat
/proc/asound/pcm
/proc/asound/seq
/proc/asound/HDMI
/proc/asound/card0
/proc/asound/card0/id
/proc/asound/card0/pcm3p
/proc/asound/card0/pcm3p/info
/proc/asound/card0/pcm3p/sub0
/proc/asound/card0/pcm3p/sub0/info
/proc/asound/card0/pcm3p/sub0/status
/proc/asound/card0/pcm3p/sub0/prealloc
/proc/asound/card0/pcm3p/sub0/hw_params
/proc/asound/card0/pcm3p/sub0/sw_params
/proc/asound/card0/pcm3p/sub0/prealloc_max
/proc/asound/card0/pcm7p
/proc/asound/card0/pcm7p/info
/proc/asound/card0/pcm7p/sub0
/proc/asound/card0/pcm7p/sub0/info
/proc/asound/card0/pcm7p/sub0/status
/proc/asound/card0/pcm7p/sub0/prealloc
/proc/asound/card0/pcm7p/sub0/hw_params
/proc/asound/card0/pcm7p/sub0/sw_params
/proc/asound/card0/pcm7p/sub0/prealloc_max
/proc/asound/card0/pcm8p
/proc/asound/card0/pcm8p/info
/proc/asound/card0/pcm8p/sub0
/proc/asound/card0/pcm8p/sub0/info
/proc/asound/card0/pcm8p/sub0/status
/proc/asound/card0/pcm8p/sub0/prealloc
/proc/asound/card0/pcm8p/sub0/hw_params
/proc/asound/card0/pcm8p/sub0/sw_params
/proc/asound/card0/pcm8p/sub0/prealloc_max
/proc/asound/card0/pcm9p
/proc/asound/card0/pcm9p/info
/proc/asound/card0/pcm9p/sub0
/proc/asound/card0/pcm9p/sub0/info
/proc/asound/card0/pcm9p/sub0/status
/proc/asound/card0/pcm9p/sub0/prealloc
/proc/asound/card0/pcm9p/sub0/hw_params
/proc/asound/card0/pcm9p/sub0/sw_params
/proc/asound/card0/pcm9p/sub0/prealloc_max
/proc/asound/card0/pcm10p
/proc/asound/card0/pcm10p/info
/proc/asound/card0/pcm10p/sub0
/proc/asound/card0/pcm10p/sub0/info
/proc/asound/card0/pcm10p/sub0/status
/proc/asound/card0/pcm10p/sub0/prealloc
/proc/asound/card0/pcm10p/sub0/hw_params
/proc/asound/card0/pcm10p/sub0/sw_params
/proc/asound/card0/pcm10p/sub0/prealloc_max
/proc/asound/card0/pcm11p
/proc/asound/card0/pcm11p/info
/proc/asound/card0/pcm11p/sub0
/proc/asound/card0/pcm11p/sub0/info
/proc/asound/card0/pcm11p/sub0/status
/proc/asound/card0/pcm11p/sub0/prealloc
/proc/asound/card0/pcm11p/sub0/hw_params
/proc/asound/card0/pcm11p/sub0/sw_params
/proc/asound/card0/pcm11p/sub0/prealloc_max
/proc/asound/card0/codec#0
/proc/asound/card0/eld#0.0
/proc/asound/card0/eld#0.1
/proc/asound/card0/eld#0.2
/proc/asound/card0/eld#0.3
/proc/asound/card0/eld#0.4
/proc/asound/card0/eld#0.5
/proc/asound/card0/oss_mixer
/proc/asound/card1
/proc/asound/card1/id
/proc/asound/card1/pcm0c
/proc/asound/card1/pcm0c/oss
/proc/asound/card1/pcm0c/info
/proc/asound/card1/pcm0c/sub0
/proc/asound/card1/pcm0c/sub0/info
/proc/asound/card1/pcm0c/sub0/status
/proc/asound/card1/pcm0c/sub0/prealloc
/proc/asound/card1/pcm0c/sub0/hw_params
/proc/asound/card1/pcm0c/sub0/sw_params
/proc/asound/card1/pcm0c/sub0/prealloc_max
/proc/asound/card1/pcm0p
/proc/asound/card1/pcm0p/oss
/proc/asound/card1/pcm0p/info
/proc/asound/card1/pcm0p/sub0
/proc/asound/card1/pcm0p/sub0/info
/proc/asound/card1/pcm0p/sub0/status
/proc/asound/card1/pcm0p/sub0/prealloc
/proc/asound/card1/pcm0p/sub0/hw_params
/proc/asound/card1/pcm0p/sub0/sw_params
/proc/asound/card1/pcm0p/sub0/prealloc_max
/proc/asound/card1/pcm2c
/proc/asound/card1/pcm2c/info
/proc/asound/card1/pcm2c/sub0
/proc/asound/card1/pcm2c/sub0/info
/proc/asound/card1/pcm2c/sub0/status
/proc/asound/card1/pcm2c/sub0/prealloc
/proc/asound/card1/pcm2c/sub0/hw_params
/proc/asound/card1/pcm2c/sub0/sw_params
/proc/asound/card1/pcm2c/sub0/prealloc_max
/proc/asound/card1/codec#0
/proc/asound/card1/oss_mixer
/proc/asound/cards
/proc/asound/hwdep
/proc/asound/timers
/proc/asound/Generic
/proc/asound/devices
/proc/asound/modules
/proc/asound/version


I'll look at the kernel traces again
A related? thread at Arch Forum is listed as Solved
from what I've read in the thread so far I'm not sure it's gonna be much help for your issue - we'll see
In memory of Ian Ashley Murdock (1973 - 2015) founder of the Debian project.
User avatar
llivv
 
Posts: 5488
Joined: 2007-02-14 18:10
Location: cold storage

Re: Video / audio troubles with RX 570 / stretch

Postby oddhack » 2018-12-19 01:32

I suggest you create a new user
than add audio,video groups


In order to see if audio behavior is different than when I'm logged in as my normal userid?

Are the repeating Call Traces Locking Up the system or does it just take longer to shutdown?


It only seems to impact shutdown, but I'm not sure I'd necc. notice if systemd-udevd was timing out and restarting during the course of normal operation. I use a pretty primitive X environment, just WindowMaker for WM as you noticed, with occasional GNOME utilities and controls as needed - I usually access pulseaudio controls from the GNOME sound control panel, though I've tried pavucontrol with similar effect.
oddhack
 
Posts: 13
Joined: 2018-07-21 23:35

Re: Video / audio troubles with RX 570 / stretch

Postby oddhack » 2018-12-19 11:26

I checked my login and added the 'video' group - already in 'audio'. After rebooting again (what is this, Windows? But it seems to make a difference), and then restarting pulseaudio, I am getting close to done. pavucontrol does show my analog line out / headphone status now, and I can play music through the headphones and HDMI both. So I think this is basically solved, although it's hard to be sure which exact steps achieved it. I'm unclear why being in the 'video' group helps, but it does seem to. The one odd thing I'm still noticing is that the 'Test Speakers' window in the GNOME sound controls doesn't produce any sound; but that's something I can probably track down on my own, and not essential.

Summary of things that seemed to help:

* Updated kernel from stretch-backports, and make sure to enable non-free repo too
* Ensure all necc. free and non-free firmware packages for my video / audio hardware have been updated as well, which they are't, automatically.
* Use amdgpu driver, not radeon
* Make sure to resolve all relevant error messages from syslog
* Ensure user is in both 'audio' and 'video' groups
* Don't get misled by unrelated errors like the systemd-udevd timeout tracebacks seen in syslog
* Reboot to ensure all proper firmware is getting applied
* Might need to restart pulseaudio

@llivv, thank you very much for your patient help with this. I think I'd be struggling for quite a while more without your suggestions.
oddhack
 
Posts: 13
Joined: 2018-07-21 23:35

Re: Video / audio troubles with RX 570 / stretch

Postby llivv » 2018-12-19 13:10

backtracking to this post
oddhack wrote:(Edit: the call trace problem is almost certainly unrelated to my audio issues - it looks like a BIOS problem with this motherboard vs. my Ryzen 2600, see https://bugzilla.redhat.com/show_bug.cgi?id=1608242)
that issue may take a bit longer for the kernel or DRM DRI devs to get around patching,,,https://bugzilla.redhat.com/show_bug.cgi?id=1427340
oddhack wrote:Boot actually isn't hanging, but shutdowns hang for several minutes with the messages:
Good that it's not freezing up
oddhack wrote:I suspect this is a bit of a red herring and could be unrelated to audio issues, though. After several reboot cycles, I'm still seeing these panic messages in syslog, but now I'm actually getting audio over HDMI and can control it with pulseaudio control apps. This is mystifying since nothing changed in the system configuration between reboots. If it persists, the only remaining problem will be that the realtek driver isn't showing up in pulseaudio (cross fingers).
Good that you can control audio

In your OP () original post - I see this desktop mobo has a vga-switcheroo (like laptops with multipule gpu's).The switcharoo might be the reason the integrated audio is not available while the discrete GPU is providing the audio .... - since the discrete GPU's audio is part of the switcharoo
https://www.kernel.org/doc/html/v4.14/gpu/vga-switcheroo.html
see also :vga_switcheroo: Deduplicate power state tracking
and also ALSA: hda - Silence PM ops build warning edit:fixed the URL title
Don't expect to get any clues how to fix the no audio from the integrated chip from the links above
They are just explaining what the switcheroo is and some of the issues being worked on currently.

I'd be stoked if I could find out why all 12 cpu's are hanging multiple times during shutdown.
Last edited by llivv on 2018-12-20 03:10, edited 1 time in total.
In memory of Ian Ashley Murdock (1973 - 2015) founder of the Debian project.
User avatar
llivv
 
Posts: 5488
Joined: 2007-02-14 18:10
Location: cold storage

Re: Video / audio troubles with RX 570 / stretch

Postby llivv » 2018-12-19 13:22

oddhack wrote:
I suggest you create a new user

In order to see if audio behavior is different than when I'm logged in as my normal userid?

Well from the groups you posted above that your normal user was a member of it was clear
that your normal user was actually a superuser. it kinda defeats the purpose of having a normal user.
You can safely add the plugdev group to your new users group affiliations ...
please refrain from adding root and adm groups to a normal user in the future.. :wink:
In memory of Ian Ashley Murdock (1973 - 2015) founder of the Debian project.
User avatar
llivv
 
Posts: 5488
Joined: 2007-02-14 18:10
Location: cold storage

Re: Video / audio troubles with RX 570 / stretch

Postby llivv » 2018-12-19 13:38

oddhack wrote:Summary of things that seemed to help:

oddhack wrote:* Updated kernel from stretch-backports, and make sure to enable non-free repo too
v/
oddhack wrote:* Ensure all necc. free and non-free firmware packages for my video / audio hardware have been updated as well, which they are't, automatically.
v/
oddhack wrote:* Use amdgpu driver, not radeon
v/
oddhack wrote:* Make sure to resolve all relevant error messages from syslog
v/
oddhack wrote:* Ensure user is in both 'audio' and 'video' groups
v/ v/ v/
oddhack wrote:* Don't get misled by unrelated errors like the systemd-udevd timeout tracebacks seen in syslog
some of those messages are almost perment residents in the log - the trick is knowing which ones are relevant and which are hints...
oddhack wrote:* Reboot to ensure all proper firmware is getting applied
also reboot in some circumstances so that systemd can take control of the new process(es) getting more like Windows every day ..... :wink:
oddhack wrote:* Might need to restart pulseaudio
v/
oddhack wrote:@llivv, thank you very much for your patient help with this. I think I'd be struggling for quite a while more without your suggestions.
You're very welcome. And thank you for being patient with me as well, I learned many new things too...

v/ = check :wink:
In memory of Ian Ashley Murdock (1973 - 2015) founder of the Debian project.
User avatar
llivv
 
Posts: 5488
Joined: 2007-02-14 18:10
Location: cold storage

Previous

Return to Desktop & Multimedia

Who is online

Users browsing this forum: No registered users and 6 guests

fashionable