Wayland Pipewire Audio Crash and Restart Problem

General system Gaming, sound and video
1

Hello! I switched to Wayland after the Nvidia 555 drivers were released, as they solve the flickering issue that I was having for XWayland apps. The problem is, Pipewire randomly restarts or something happens which causes the entire audio system to restart. Moreover, YouTube videos do not play audio sometimes. Clicking the play button twice makes the audio return to the videos.

Is anyone else having this problem as well? What information do I have to give to get an idea of where this problem might lie?

1 Like
2

I have something similar but i am using combine-sink and it happens to freeze pipewire when I remove USB headset.

Try checking out dmesg or journalctl when the issue happens.

1 Like
3

You using something like easyeffects? it could be causing this.

1 Like
4

Similar sound problem after updating June 29th. Seems to be pipewire v1.2.0 is causing the issue. I had to revert pipewire to v1.0.7 to correct the problem. Still no resolve yet that I can see.

1 Like
5

pipewire 1:1.2.1-1 had been just released. Maybe you want to try that version?

1 Like
6

I am using Easy Effects. However, the problem appears for other people as well. So you think that easy effects is causing the problem?

Easy Effects works fine. Do you believe I should go to the Easy Effects developer and ask them about it?

7

I see. I was told that Easy Effects might be the thing that is causing the problem. I am not sure yet. There is an update that was released right now. Probably it has an update that will fix this issue.

8

first figure out if it is what is causing the problem, try not using easyeffects, see if the audio still crashes then.

If it doesn’t, then easyeffects is the problem, so what you should try to do then is remove all your effects except 1, and then test 1 effect at a time to see if a specific effect is causing the issue.

And if it is THEN you go report the bug to the easyeffect dev assuming nobody else beat you to it.

9

I stopped using Easy Effects completely. That doesn’t seem to be the problem. The YouTube bug is still present. The video audio doesn’t play, unless I stop the video using the play button and then click the play button again.

After a while, I think KWin crashes entirely. It isn’t just the audio that crashes.

10

Does it happen on X11 too?

11

No. It does not happen on X11. X11 is working fine. It only happen on Wayland. I think it is a pipewire issue.

Even OBS is quite glitchy on Wayland right now. An entire recording of mine got corrupted due to it.

12

I don’t think it’s a pipewire issue if it only happens on wayland (You are still using pipewire on X too), it’s possible but unlikely, you said you’re on nvidia, are you using the nvidia hdmi audio or something like that? If you are, does the isssue occur still when you just use headphones plugged into the jack or something and leave the hdim audio off?

13

I think it might be an Nvidia HDMI audio issue. However, I am using USB Headphones. One of my screens uses Nvidia HDMI audio. Ever since putting ‘nvidia.NVreg_EnableGpuFirmware=0’ inside of the grub.cfg file on ‘GRUB_CMDLINE_LINUX_DEFAULT=’ the crash doesn’t seem to happen. However, the YouTube audio problem still exists.

I have even told the KDE folks about this. I am quite sure it is an Nvidia problem, with the 555 drivers. Possibly due to explicit sync.

14

Well, the crash happened again. The only thing that I know that might be causing it is the NVIDIA driver. The crash does not happen whatsoever on X11. It might be an explicit sync issue. I wish I could find out through a log what goes wrong. I can then help other people who might have similar problem.

15

For some reason, the crash does not happen when I use the Linux Zen kernel. This is slightly strange!

16

I have found something! In Journalctl this is the error that I get. Where should I post this, and how can I help resolve this issue?

Jul 24 12:59:02 attish-archLinux kernel: 
Jul 24 12:59:07 attish-archLinux kernel: usb 3-4: 3:1: cannot get freq at ep 0x84
Jul 24 12:59:17 attish-archLinux sddm-helper[974]: gkr-pam: unable to locate daemon control file
Jul 24 12:59:20 attish-archLinux kernel: usb 3-4: 3:1: cannot get freq at ep 0x84
Jul 24 12:59:20 attish-archLinux kernel: usb 3-4: 3:1: cannot get freq at ep 0x84
Jul 24 12:59:22 attish-archLinux org_kde_powerdevil[1257]: busno=4, sleep-multiplier =  2.00. Testing for supported feature 0x10 returned Error_Info[DDCRC_RETRIES in ddc_write_read_with_retry, causes: DDCRC_DDC_DATA(10)]
Jul 24 13:19:26 attish-archLinux systemd-coredump[4584]: Process 1108 (pipewire) of user 1000 dumped core.
                                                         
                                                         Stack trace of thread 1111:
                                                         #0  0x00007f8edfe691c6 n/a (libspa-audioconvert.so + 0x271c6)
                                                         #1  0x00007f8edfe53377 n/a (libspa-audioconvert.so + 0x11377)
                                                         #2  0x00007f8ee1198758 n/a (libpipewire-0.3.so.0 + 0x76758)
                                                         #3  0x00007f8ee116a8a2 n/a (libpipewire-0.3.so.0 + 0x488a2)
                                                         #4  0x00007f8ee11ec176 n/a (libspa-support.so + 0x7176)
                                                         #5  0x00007f8ee11423f3 n/a (libpipewire-0.3.so.0 + 0x203f3)
                                                         #6  0x00007f8ee0fc539d n/a (libc.so.6 + 0x9439d)
                                                         #7  0x00007f8ee104a2a4 __clone (libc.so.6 + 0x1192a4)
                                                         
                                                         Stack trace of thread 1108:
                                                         #0  0x00007f8ee104a8b2 epoll_wait (libc.so.6 + 0x1198b2)
                                                         #1  0x00007f8ee11fb3e7 n/a (libspa-support.so + 0x163e7)
                                                         #2  0x00007f8ee11ec0b1 n/a (libspa-support.so + 0x70b1)
                                                         #3  0x00007f8ee11609a0 pw_main_loop_run (libpipewire-0.3.so.0 + 0x3e9a0)
                                                         #4  0x00005f5eb4921522 n/a (pipewire + 0x1522)
                                                         #5  0x00007f8ee0f56e08 n/a (libc.so.6 + 0x25e08)
                                                         #6  0x00007f8ee0f56ecc __libc_start_main (libc.so.6 + 0x25ecc)
                                                         #7  0x00005f5eb49216b5 n/a (pipewire + 0x16b5)
                                                         
                                                         Stack trace of thread 1110:
                                                         #0  0x00007f8ee104a8b2 epoll_wait (libc.so.6 + 0x1198b2)
                                                         #1  0x00007f8ee11fb3e7 n/a (libspa-support.so + 0x163e7)
                                                         #2  0x00007f8ee11ec0b1 n/a (libspa-support.so + 0x70b1)
                                                         #3  0x00007f8ee118dc07 n/a (libpipewire-0.3.so.0 + 0x6bc07)
                                                         #4  0x00007f8ee0fc539d n/a (libc.so.6 + 0x9439d)
                                                         #5  0x00007f8ee104a2a4 __clone (libc.so.6 + 0x1192a4)
                                                         ELF object binary architecture: AMD x86-64
░░ Subject: Process 1108 (pipewire) dumped core
░░ Defined-By: systemd
░░ Support: https://lists.freedesktop.org/mailman/listinfo/systemd-devel
░░ Documentation: man:core(5)
░░ 
░░ Process 1108 (pipewire) crashed and dumped core.
░░ 
░░ This usually indicates a programming error in the crashing program and
░░ should be reported to its vendor as a bug.
17

Please report an issue on the pipewire bug tracker.

18

For anyone experiencing this issue, consider using Linux Zen kernel, which does not have this problem. There might be an effect in Easy Effects that I am using which does not work well on Wayland and is possibly related to Output Streams not being managed properly. Consider disabling the Easy Effects ‘service’ on startup and you should not have the crashes any longer.

19

My 2c

I am experiencing similar issues for months now.

  • regardless wether i use X11 or wayland
  • no matter what kernel i use
  • i use a “driverless” class compliant external audio interface
  • the issue does not occur without easyeffects loaded
  • disabling session suspend in wireplumber mitigated the issue but did not solve it

Still, i don´t consider disabling easyeffects a viable solution. :stuck_out_tongue:

1 Like
20

This topic was automatically closed 2 days after the last reply. New replies are no longer allowed.