Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Easyeffects process killed on idle #3699

Open
disposableone opened this issue Mar 4, 2025 · 36 comments
Open

Easyeffects process killed on idle #3699

disposableone opened this issue Mar 4, 2025 · 36 comments

Comments

@disposableone
Copy link

EasyEffects Version

7.1.9

What package are you using?

Fedora

Distribution

Fedora 41

Describe the bug

If I leave Easyeffects on for a little while, it is automatically killed. I even leave the window open, both the window and process get killed.

Expected Behavior

I didn't see this behavior last year, it appeared relatively recently. Before, easyeffects would stay on at all times. That's what I would expect.

Debug Log

No response

Additional Information

No response

@wwmm
Copy link
Owner

wwmm commented Mar 4, 2025

If I leave Easyeffects on for a little while, it is automatically killed. I even leave the window open, both the window and process get killed.

Do you see the message "killed" in your system logs? Run sudo journalctl -b in a terminal. It should show if that is the case.

Have you installed any package with ananicy rules or something similar that could be affecting EasyEffects? In the past some users had problems because of that. But there are other cases where the reason why some users systems decide to kill EasyEffects remain unknown. For some reason I could never reproduce this kind of issue.

@disposableone
Copy link
Author

If I leave Easyeffects on for a little while, it is automatically killed. I even leave the window open, both the window and process get killed.

Do you see the message "killed" in your system logs? Run sudo journalctl -b in a terminal. It should show if that is the case.

Yessir. Bunch of code=killed for easyeffects.

Have you installed any package with ananicy rules or something similar that could be affecting EasyEffects? In the past some users had problems because of that. But there are other cases where the reason why some users systems decide to kill EasyEffects remain unknown. For some reason I could never reproduce this kind of issue.

No ananicy. Really not much changed from a few months ago when easyeffects wasn't getting killed. I noticed something weird though. I have Grok tab open in Firefox oftentimes, and I just leave it open. Then I see easyeffects ramping up some CPU cycles (27%, i.e.) when everything is idle, and easyeffects showing Grok tab from Firefox as the only output source in the Output tab of easyeffects (even though no sound ever comes out of Grok). Kinda weird, maybe that's it? The kills start on Feb 23rd, around when I started using Grok 3 after it came out.

@disposableone
Copy link
Author

Ok, now it "exited" with Soundcloud and Youtube playing. This Firefox bug I have been seeing (https://bugzilla.mozilla.org/show_bug.cgi?id=1951237) may be related to this.

@wwmm
Copy link
Owner

wwmm commented Mar 4, 2025

even though no sound ever comes out of Grok

Firefox sometimes keeps alive audio streams that do not seem to be doing anything. As strange as they are I have never seen they cause issues on my systems. The real source of the problem is probably something else.

Yessir. Bunch of code=killed for easyeffects.

Ok. One more mysterious case where the system kills EasyEffects without saying the reason why it is doing that.

Ok, now it "exited" with Soundcloud and Youtube playing. This Firefox bug I have been seeing (https://bugzilla.mozilla.org/show_bug.cgi?id=1951237) may be related to this.

I am not sure. Firefox crashing should make EasyEffects crash.

@riskyrebase
Copy link

I am having crashes with the flatpak version 7.2.3
From my testing it seems easyeffects only crashes when the system and the app itself is under heavy load (I mean high CPU usage).

A simple way to reproduce this is to install the flatpak version, add a heavy resource using effect to the microphone (haven't checked with speaker) i.e. deep noise remover. Then open obs or something that uses the microphone (easy effects source). Then somehow put the system under heavy CPU load. You can just maximize and unmaximize a browser window. Make sure CPU gets under high usage. And then easyeffects should crash in a couple of seconds.

Here is the command output for RUST_BACKTRACE=full flatpak run com.github.wwmm.easyeffects

2025-03-08T11:39:53.036Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.34). Processing too slow!
2025-03-08T11:39:55.387Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.53). Processing too slow!
2025-03-08T11:39:55.455Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.72). Processing too slow!
2025-03-08T11:39:55.490Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.05). Processing too slow!
2025-03-08T11:39:55.534Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.34). Processing too slow!
2025-03-08T11:39:57.447Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.06). Processing too slow!
2025-03-08T11:39:57.497Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.43). Processing too slow!
2025-03-08T11:39:57.528Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.44). Processing too slow!
2025-03-08T11:40:01.777Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.06). Processing too slow!
2025-03-08T11:40:01.800Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.06). Processing too slow!
2025-03-08T11:40:01.864Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 2.11). Processing too slow!
2025-03-08T11:40:02.014Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.14). Processing too slow!
2025-03-08T11:40:02.481Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.06). Processing too slow!
2025-03-08T11:40:02.526Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 2.12). Processing too slow!
2025-03-08T11:40:02.588Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.93). Processing too slow!
2025-03-08T11:40:02.611Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.06). Processing too slow!
2025-03-08T11:40:02.662Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.54). Processing too slow!
2025-03-08T11:40:02.699Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.73). Processing too slow!
2025-03-08T11:40:02.732Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.54). Processing too slow!
2025-03-08T11:40:02.755Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.06). Processing too slow!
2025-03-08T11:40:02.780Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.17). Processing too slow!
2025-03-08T11:40:02.835Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.55). Processing too slow!
2025-03-08T11:40:02.888Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.49). Processing too slow!
2025-03-08T11:40:02.962Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 2.61). Processing too slow!
2025-03-08T11:40:03.012Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.35). Processing too slow!
2025-03-08T11:40:03.087Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.54). Processing too slow!
2025-03-08T11:40:03.158Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 2.30). Processing too slow!
2025-03-08T11:40:03.199Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.15). Processing too slow!
2025-03-08T11:40:03.257Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 2.01). Processing too slow!
2025-03-08T11:40:03.340Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 3.01). Processing too slow!
2025-03-08T11:40:03.381Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.24). Processing too slow!
2025-03-08T11:40:03.492Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.45). Processing too slow!
2025-03-08T11:40:03.514Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.05). Processing too slow!
2025-03-08T11:40:03.558Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 2.02). Processing too slow!
2025-03-08T11:40:03.595Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.06). Processing too slow!
2025-03-08T11:40:03.647Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.74). Processing too slow!
2025-03-08T11:40:03.718Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 2.32). Processing too slow!
2025-03-08T11:40:03.749Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.46). Processing too slow!
2025-03-08T11:40:03.795Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 2.13). Processing too slow!
2025-03-08T11:40:03.871Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 2.90). Processing too slow!
2025-03-08T11:40:03.907Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.66). Processing too slow!
2025-03-08T11:40:03.936Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.34). Processing too slow!
2025-03-08T11:40:03.989Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.53). Processing too slow!
2025-03-08T11:40:04.024Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.63). Processing too slow!
2025-03-08T11:40:04.073Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.55). Processing too slow!
2025-03-08T11:40:04.099Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.24). Processing too slow!
2025-03-08T11:40:04.168Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 2.33). Processing too slow!
2025-03-08T11:40:04.197Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.36). Processing too slow!
2025-03-08T11:40:04.222Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.16). Processing too slow!
2025-03-08T11:40:04.258Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.64). Processing too slow!
2025-03-08T11:40:04.287Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.35). Processing too slow!
2025-03-08T11:40:04.329Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.34). Processing too slow!
2025-03-08T11:40:04.371Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.15). Processing too slow!
2025-03-08T11:40:04.451Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 3.18). Processing too slow!
2025-03-08T11:40:04.501Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.54). Processing too slow!
2025-03-08T11:40:04.524Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.06). Processing too slow!
2025-03-08T11:40:04.549Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.15). Processing too slow!
2025-03-08T11:40:04.582Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.55). Processing too slow!
2025-03-08T11:40:04.632Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.45). Processing too slow!
2025-03-08T11:40:04.662Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.35). Processing too slow!
2025-03-08T11:40:04.703Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.16). Processing too slow!
2025-03-08T11:40:04.749Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.35). Processing too slow!
2025-03-08T11:40:04.799Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.34). Processing too slow!
2025-03-08T11:40:04.830Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.44). Processing too slow!
2025-03-08T11:40:04.964Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.45). Processing too slow!
2025-03-08T11:40:05.018Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.92). Processing too slow!
2025-03-08T11:40:05.061Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.34). Processing too slow!
2025-03-08T11:40:05.128Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.15). Processing too slow!
2025-03-08T11:40:05.190Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 2.03). Processing too slow!
2025-03-08T11:40:05.264Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 2.52). Processing too slow!
2025-03-08T11:40:05.329Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 2.21). Processing too slow!
2025-03-08T11:40:05.399Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.44). Processing too slow!
2025-03-08T11:40:05.445Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 2.12). Processing too slow!
2025-03-08T11:40:05.509Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 2.41). Processing too slow!
2025-03-08T11:40:05.532Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.06). Processing too slow!
2025-03-08T11:40:05.586Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.53). Processing too slow!
2025-03-08T11:40:05.615Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.35). Processing too slow!
2025-03-08T11:40:05.671Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 2.60). Processing too slow!
2025-03-08T11:40:05.698Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.26). Processing too slow!
2025-03-08T11:40:05.774Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 2.52). Processing too slow!
2025-03-08T11:40:05.820Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.45). Processing too slow!
2025-03-08T11:40:05.874Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.74). Processing too slow!
2025-03-08T11:40:05.918Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.16). Processing too slow!
2025-03-08T11:40:06.001Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 2.02). Processing too slow!
2025-03-08T11:40:06.050Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.34). Processing too slow!
2025-03-08T11:40:06.081Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.44). Processing too slow!
2025-03-08T11:40:06.121Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.24). Processing too slow!
2025-03-08T11:40:06.177Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.74). Processing too slow!
2025-03-08T11:40:06.200Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.06). Processing too slow!
2025-03-08T11:40:06.256Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.74). Processing too slow!
2025-03-08T11:40:06.292Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.65). Processing too slow!
2025-03-08T11:40:06.316Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.07). Processing too slow!
2025-03-08T11:40:06.365Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.54). Processing too slow!
2025-03-08T11:40:06.442Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.74). Processing too slow!
2025-03-08T11:40:06.467Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.15). Processing too slow!
2025-03-08T11:40:06.523Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.64). Processing too slow!
2025-03-08T11:40:06.562Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.34). Processing too slow!
2025-03-08T11:40:06.604Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.25). Processing too slow!
2025-03-08T11:40:06.685Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 2.79). Processing too slow!
2025-03-08T11:40:06.714Z | WARN |  deep_filter_ladspa | DF c883907d03b6 | Underrun detected (RTF: 1.34). Processing too slow!
thread '<unnamed>' panicked at 'DF c883907d03b6 | Processing too slow! Please upgrade your CPU. Try to decrease 'Max DF processing threshold (dB)'.', ladspa/src/lib.rs:444:17
stack backtrace:
   0:     0x7fa668d19fa4 - <unknown>
   1:     0x7fa668b7ed1f - <unknown>
   2:     0x7fa668ced036 - <unknown>
   3:     0x7fa668d1b40f - <unknown>
   4:     0x7fa668d1afd0 - <unknown>
   5:     0x7fa668d1bfcb - <unknown>
   6:     0x7fa668d1bad2 - <unknown>
   7:     0x7fa668d1ba36 - <unknown>
   8:     0x7fa668d1ba21 - <unknown>
   9:     0x7fa668b0eb52 - <unknown>
  10:     0x7fa668b36c60 - <unknown>
  11:     0x56472ccc9b1a - <unknown>
  12:     0x56472cfb1a5b - <unknown>
  13:     0x7fa695af8f81 - <unknown>
  14:     0x7fa695b18c3a - <unknown>
  15:     0x7fa695b17b50 - <unknown>
  16:     0x7fa68c45bd75 - <unknown>
  17:     0x7fa695aee2e1 - <unknown>
  18:     0x7fa6934beeda - <unknown>
  19:     0x7fa693546ac4 - __clone
  20:                0x0 - <unknown>

And relevant journalctl -b output

Mar 08 17:39:51 Fedora-desktop flatpak[2923]: info: pulse-input: Started recording from 'easyeffects_source'
Mar 08 17:39:51 Fedora-desktop flatpak[2923]: info: [Loaded global audio device]: 'Mic/Aux'
Mar 08 17:39:51 Fedora-desktop flatpak[2923]: info: PipeWire initialized
Mar 08 17:39:51 Fedora-desktop flatpak[2923]: info: Switched to scene 'Scene'
Mar 08 17:39:51 Fedora-desktop flatpak[2923]: info: ------------------------------------------------
Mar 08 17:39:51 Fedora-desktop flatpak[2923]: info: Loaded scenes:
Mar 08 17:39:51 Fedora-desktop flatpak[2923]: info: - scene 'Scene':
Mar 08 17:39:51 Fedora-desktop flatpak[2923]: info:     - source: 'Screen Capture (PipeWire)' (pipewire-screen-capture-source)
Mar 08 17:39:51 Fedora-desktop flatpak[2923]: info: ------------------------------------------------
Mar 08 17:39:52 Fedora-desktop flatpak[2923]: info: [pipewire] Screencast session created
Mar 08 17:39:52 Fedora-desktop plasmashell[1491]: QFont::setPointSizeF: Point size <= 0 (0.000000), must be greater than 0
Mar 08 17:39:52 Fedora-desktop flatpak[2923]: info: [pipewire] Asking for monitor and window
Mar 08 17:39:52 Fedora-desktop pipewire[1336]: pw.node: (alsa_input.pci-0000_00_1b.0.analog-stereo-44) graph xrun not-triggered (38 suppressed)
Mar 08 17:39:52 Fedora-desktop pipewire[1336]: pw.node: (alsa_input.pci-0000_00_1b.0.analog-stereo-44) xrun state:0x7fc57dff3008 pending:5/6 s:444183529508 a:444184548360 f:444184562354 waiting:1018852 process:13994 status:triggered
Mar 08 17:39:52 Fedora-desktop pipewire[1336]: pw.node: (ee_sie_deepfilternet-133) xrun state:0x7fc57da47008 pending:0/1 s:444204890493 a:444204915500 f:444183614961 waiting:25007 process:18446744073688251077 status:awake
Mar 08 17:39:52 Fedora-desktop plasmashell[1491]: QFont::setPointSizeF: Point size <= 0 (0.000000), must be greater than 0
Mar 08 17:39:52 Fedora-desktop flatpak[2923]: info: [pipewire] source selected, setting up screencast
Mar 08 17:39:52 Fedora-desktop flatpak[2923]: info: [pipewire] Server version: 1.2.7
Mar 08 17:39:52 Fedora-desktop flatpak[2923]: info: [pipewire] Library version: 0.3.83
Mar 08 17:39:52 Fedora-desktop flatpak[2923]: info: [pipewire] Header version: 0.3.83
Mar 08 17:39:52 Fedora-desktop flatpak[2923]: info: [pipewire] Created stream 0x555f4b861980
Mar 08 17:39:52 Fedora-desktop flatpak[2923]: info: [pipewire] Stream 0x555f4b861980 state: "connecting" (error: none)
Mar 08 17:39:52 Fedora-desktop flatpak[2923]: info: [pipewire] Playing stream 0x555f4b861980
Mar 08 17:39:52 Fedora-desktop pipewire[1336]: mod.client-node: detected old client version 5
Mar 08 17:39:52 Fedora-desktop flatpak[2923]: info: [pipewire] Stream 0x555f4b861980 state: "paused" (error: none)
Mar 08 17:39:52 Fedora-desktop flatpak[2923]: info: [pipewire] Negotiated format:
Mar 08 17:39:52 Fedora-desktop flatpak[2923]: info: [pipewire]     Format: 12 (Spa:Enum:VideoFormat:BGRA)
Mar 08 17:39:52 Fedora-desktop flatpak[2923]: info: [pipewire]     Modifier: 0x0
Mar 08 17:39:52 Fedora-desktop flatpak[2923]: info: [pipewire]     Size: 1440x900
Mar 08 17:39:52 Fedora-desktop flatpak[2923]: info: [pipewire]     Framerate: 0/1
Mar 08 17:39:52 Fedora-desktop flatpak[2923]: info: [pipewire] Negotiated format:
Mar 08 17:39:52 Fedora-desktop flatpak[2923]: info: [pipewire]     Format: 12 (Spa:Enum:VideoFormat:BGRA)
Mar 08 17:39:52 Fedora-desktop flatpak[2923]: info: [pipewire]     Modifier: 0x100000000000001
Mar 08 17:39:52 Fedora-desktop flatpak[2923]: info: [pipewire]     Size: 1440x900
Mar 08 17:39:52 Fedora-desktop flatpak[2923]: info: [pipewire]     Framerate: 0/1
Mar 08 17:39:52 Fedora-desktop flatpak[2923]: info: [pipewire] Stream 0x555f4b861980 state: "streaming" (error: none)
Mar 08 17:39:52 Fedora-desktop plasmashell[1491]: QFont::setPointSizeF: Point size <= 0 (0.000000), must be greater than 0
Mar 08 17:39:52 Fedora-desktop flatpak[2923]: info: adding 21 milliseconds of audio buffering, total audio buffering is now 21 milliseconds (source: Desktop Audio)
Mar 08 17:39:53 Fedora-desktop flatpak[2923]: info: adding 21 milliseconds of audio buffering, total audio buffering is now 42 milliseconds (source: Mic/Aux)
Mar 08 17:39:55 Fedora-desktop pipewire[1336]: pw.node: (alsa_input.pci-0000_00_1b.0.analog-stereo-44) graph xrun not-triggered (90 suppressed)
Mar 08 17:39:55 Fedora-desktop pipewire[1336]: pw.node: (alsa_input.pci-0000_00_1b.0.analog-stereo-44) xrun state:0x7fc57dff3008 pending:5/6 s:447512078988 a:447522451016 f:447522453298 waiting:10372028 process:2282 status:triggered
Mar 08 17:39:55 Fedora-desktop pipewire[1336]: pw.node: (ee_sie_deepfilternet-133) xrun state:0x7fc57da47008 pending:0/1 s:447533432574 a:447533459028 f:447522283242 waiting:26454 process:18446744073698375830 status:awake
Mar 08 17:39:57 Fedora-desktop pipewire[1336]: pw.node: (alsa_input.pci-0000_00_1b.0.analog-stereo-44) graph xrun not-triggered (20 suppressed)
Mar 08 17:39:57 Fedora-desktop pipewire[1336]: pw.node: (alsa_input.pci-0000_00_1b.0.analog-stereo-44) xrun state:0x7fc57dff3008 pending:5/6 s:449581543954 a:449594653236 f:449594659880 waiting:13109282 process:6644 status:triggered
Mar 08 17:39:57 Fedora-desktop pipewire[1336]: pw.node: (ee_sie_deepfilternet-133) xrun state:0x7fc57da47008 pending:0/1 s:449602961067 a:449603052805 f:449594117967 waiting:91738 process:18446744073700616778 status:awake
Mar 08 17:40:01 Fedora-desktop pipewire[1336]: pw.node: (alsa_input.pci-0000_00_1b.0.analog-stereo-44) graph xrun not-triggered (17 suppressed)
Mar 08 17:40:01 Fedora-desktop pipewire[1336]: pw.node: (alsa_input.pci-0000_00_1b.0.analog-stereo-44) xrun state:0x7fc57dff3008 pending:5/6 s:453912171134 a:453914449678 f:453914451888 waiting:2278544 process:2210 status:triggered
Mar 08 17:40:01 Fedora-desktop pipewire[1336]: pw.node: (ee_sie_deepfilternet-133) xrun state:0x7fc57da47008 pending:0/1 s:453933581962 a:453933661218 f:453914295796 waiting:79256 process:18446744073690186194 status:awake
Mar 08 17:40:03 Fedora-desktop pipewire[1336]: pw.node: (alsa_input.pci-0000_00_1b.0.analog-stereo-44) graph xrun not-triggered (167 suppressed)
Mar 08 17:40:03 Fedora-desktop pipewire[1336]: pw.node: (alsa_input.pci-0000_00_1b.0.analog-stereo-44) xrun state:0x7fc57dff3008 pending:5/6 s:455917425738 a:455928491136 f:455928497723 waiting:11065398 process:6587 status:triggered
Mar 08 17:40:03 Fedora-desktop pipewire[1336]: pw.node: (ee_sie_deepfilternet-133) xrun state:0x7fc57da47008 pending:0/1 s:455938658526 a:455928455727 f:455928187536 waiting:18446744073699348817 process:18446744073709283425 status:triggered
Mar 08 17:40:03 Fedora-desktop pipewire[1336]: pw.node: (alsa_output.pci-0000_00_1b.0.analog-stereo-46) graph xrun not-triggered (118 suppressed)
Mar 08 17:40:03 Fedora-desktop pipewire[1336]: pw.node: (alsa_output.pci-0000_00_1b.0.analog-stereo-46) xrun state:0x7fc57e938008 pending:2/5 s:455918046633 a:455928381657 f:455928464172 waiting:10335024 process:82515 status:triggered
Mar 08 17:40:03 Fedora-desktop pipewire[1336]: pw.node: (ee_soe_spectrum-103) xrun state:0x7fc57da4a008 pending:0/2 s:455950072139 a:455928281197 f:455928370587 waiting:18446744073687760674 process:89390 status:triggered
Mar 08 17:40:03 Fedora-desktop flatpak[2923]: info: adding 64 milliseconds of audio buffering, total audio buffering is now 106 milliseconds (source: Mic/Aux)
Mar 08 17:40:05 Fedora-desktop pipewire[1336]: pw.node: (alsa_input.pci-0000_00_1b.0.analog-stereo-44) graph xrun not-triggered (257 suppressed)
Mar 08 17:40:05 Fedora-desktop pipewire[1336]: pw.node: (alsa_input.pci-0000_00_1b.0.analog-stereo-44) xrun state:0x7fc57dff3008 pending:1/6 s:457922790472 a:457944183508 f:457944196729 waiting:21393036 process:13221 status:triggered
Mar 08 17:40:05 Fedora-desktop pipewire[1336]: pw.node: (ee_sie_deepfilternet-133) xrun state:0x7fc57da47008 pending:-1/1 s:457944183508 a:457952827198 f:457952703834 waiting:8643690 process:18446744073709428252 status:awake
Mar 08 17:40:05 Fedora-desktop pipewire[1336]: pw.node: (ee_sie_spectrum-107) xrun state:0x7fc57da10008 pending:0/2 s:457952703834 a:457898736564 f:457898884247 waiting:18446744073655584346 process:147683 status:triggered
Mar 08 17:40:05 Fedora-desktop pipewire[1336]: pw.node: (ee_sie_output_level-116) xrun state:0x7fc57da36008 pending:0/2 s:457952809917 a:457952787376 f:457952809917 waiting:18446744073709529075 process:22541 status:triggered
Mar 08 17:40:05 Fedora-desktop pipewire[1336]: pw.node: (alsa_output.pci-0000_00_1b.0.analog-stereo-46) graph xrun not-triggered (166 suppressed)
Mar 08 17:40:05 Fedora-desktop pipewire[1336]: pw.node: (alsa_output.pci-0000_00_1b.0.analog-stereo-46) xrun state:0x7fc57e938008 pending:2/5 s:457948958741 a:457967573765 f:457967679787 waiting:18615024 process:106022 status:triggered
Mar 08 17:40:05 Fedora-desktop pipewire[1336]: pw.node: (ee_soe_spectrum-103) xrun state:0x7fc57da4a008 pending:0/2 s:457970054602 a:457967620836 f:457967770696 waiting:18446744073707117850 process:149860 status:triggered
Mar 08 17:40:06 Fedora-desktop audit[2884]: ANOM_ABEND auid=1000 uid=1000 gid=1000 ses=3 subj=unconfined_u:unconfined_r:unconfined_t:s0-s0:c0.c1023 pid=2884 comm="easyeffects" exe="/app/bin/easyeffects" sig=6 res=1
Mar 08 17:40:06 Fedora-desktop systemd-coredump[2999]: Process 2884 (easyeffects) of user 1000 terminated abnormally with signal 6/ABRT, processing...
Mar 08 17:40:06 Fedora-desktop audit: BPF prog-id=100 op=LOAD
Mar 08 17:40:06 Fedora-desktop audit: BPF prog-id=101 op=LOAD
Mar 08 17:40:06 Fedora-desktop audit: BPF prog-id=102 op=LOAD
Mar 08 17:40:06 Fedora-desktop systemd[1]: Started [email protected] - Process Core Dump (PID 2999/UID 0).
Mar 08 17:40:06 Fedora-desktop audit[1]: SERVICE_START pid=1 uid=0 auid=4294967295 ses=4294967295 subj=system_u:system_r:init_t:s0 msg='unit=systemd-coredump@2-2999-0 comm="systemd" exe="/usr/lib/systemd/systemd" hostname=? addr=? terminal=? res=success'
Mar 08 17:40:07 Fedora-desktop pipewire[1336]: pw.node: (alsa_input.pci-0000_00_1b.0.analog-stereo-44) graph xrun not-triggered (225 suppressed)
Mar 08 17:40:07 Fedora-desktop pipewire[1336]: pw.node: (alsa_input.pci-0000_00_1b.0.analog-stereo-44) xrun state:0x7fc57dff3008 pending:5/6 s:459928234898 a:459949694185 f:459949705384 waiting:21459287 process:11199 status:triggered
Mar 08 17:40:07 Fedora-desktop pipewire[1336]: pw.node: (ee_sie_deepfilternet-133) xrun state:0x7fc57da47008 pending:0/1 s:459949694185 a:458864017014 f:458863913081 waiting:18446744072623874445 process:18446744073709447683 status:triggered
Mar 08 17:40:07 Fedora-desktop pipewire[1336]: pw.node: (alsa_output.pci-0000_00_1b.0.analog-stereo-46) graph xrun not-triggered (179 suppressed)
Mar 08 17:40:07 Fedora-desktop pipewire[1336]: pw.node: (alsa_output.pci-0000_00_1b.0.analog-stereo-46) xrun state:0x7fc57e938008 pending:2/5 s:459954954995 a:459975945772 f:459976065446 waiting:20990777 process:119674 status:triggered
Mar 08 17:40:07 Fedora-desktop pipewire[1336]: pw.node: (ee_soe_spectrum-103) xrun state:0x7fc57da4a008 pending:0/2 s:459976155493 a:458804126865 f:458804231621 waiting:18446744072537522988 process:104756 status:triggered
Mar 08 17:40:08 Fedora-desktop systemd-coredump[3000]: Process 2884 (easyeffects) of user 1000 dumped core.

                                                       Stack trace of thread 9:
                                                       #0  0x00007fa6934c0eb4 n/a ([dso] + 0xf52eb4)
                                                       #1  0x00007fa69346edce n/a ([dso] + 0xf00dce)
                                                       #2  0x00007fa69345683f n/a ([dso] + 0xee883f)
                                                       #3  0x00007fa668c35057 n/a (n/a + 0x0)
                                                       #4  0x0000000000000002 n/a (n/a + 0x0)
                                                       ELF object binary architecture: AMD x86-64
Mar 08 17:40:08 Fedora-desktop systemd[1]: [email protected]: Deactivated successfully.
Mar 08 17:40:08 Fedora-desktop audit[1]: SERVICE_STOP pid=1 uid=0 auid=4294967295 ses=4294967295 subj=system_u:system_r:init_t:s0 msg='unit=systemd-coredump@2-2999-0 comm="systemd" exe="/usr/lib/systemd/systemd" hostname=? addr=? terminal=? res=success'
Mar 08 17:40:08 Fedora-desktop systemd[1]: [email protected]: Consumed 1.735s CPU time, 371.2M memory peak.
Mar 08 17:40:08 Fedora-desktop audit: BPF prog-id=102 op=UNLOAD
Mar 08 17:40:08 Fedora-desktop audit: BPF prog-id=101 op=UNLOAD
Mar 08 17:40:08 Fedora-desktop audit: BPF prog-id=100 op=UNLOAD
Mar 08 17:40:08 Fedora-desktop systemd[1259]: app-flatpak-com.github.wwmm.easyeffects-2871.scope: Consumed 14.280s CPU time, 198.9M memory peak.

@wwmm
Copy link
Owner

wwmm commented Mar 8, 2025

From my testing it seems easyeffects only crashes when the system and the app itself is under heavy load (I mean high CPU usage).

Under these circumstances it may make sense for the system to kill EasyEffects. By default the filters run in a realtime thread managed by PipeWire. If a heavy plugin runs on it when the system is under heavy load it may happen that the plugin takes too long to do what it has to do and the system decides to kill EasyEffects. It will depend on the CPU. And on how heavy is the load on it.

What is strange is the system killing EasyEffects when there is no heavy plugin in the pipeline and the CPU has almost no load. These cases are still a mystery.

@disposableone
Copy link
Author

disposableone commented Mar 8, 2025

Well, I do occasionally have strangely 27% CPU use when easyeffects is idle, which is really strange. (27% CPU meaning what top shows, meaning probably not a whole lot of CPU usage but more than should be). Maybe that's when my easyeffects gets killed. It's one of the Firefox tabs that starts doing that to easyeffects... That didn't use to happen some months ago...the only thing changed, I think, I started using firejail with firefox.

@wwmm
Copy link
Owner

wwmm commented Mar 8, 2025

Well, I do occasionally have strangely 27% CPU use when easyeffects is idle, which is really strange. (27% CPU meaning what top shows, meaning probably not a whole lot of CPU usage but more than should be). Maybe that's when my easyeffects gets killed. It's one of the Firefox tabs that starts doing that to easyeffects... That didn't use to happen some months ago...the only thing changed, I think, I started using firejail with firefox.

It should not be killed when the CPU load is only of 27%. When I think about heavy load I have in mind the cases where each core load is of at least 80% or 90 %. By default top shows values in a scale that goes from zero to (n cores) * 100 %. For example I have a Ryzen 9900X. As it has 24 virtual cores top scale goes to 2400%. Your load of 27 % is probably a light load if we take into account the maximum value of the scale top is using.

@disposableone
Copy link
Author

Well, I do occasionally have strangely 27% CPU use when easyeffects is idle, which is really strange. (27% CPU meaning what top shows, meaning probably not a whole lot of CPU usage but more than should be). Maybe that's when my easyeffects gets killed. It's one of the Firefox tabs that starts doing that to easyeffects... That didn't use to happen some months ago...the only thing changed, I think, I started using firejail with firefox.

It should not be killed when the CPU load is only of 27%. When I think about heavy load I have in mind the cases where each core load is of at least 80% or 90 %. By default top shows values in a scale that goes from zero to (n cores) * 100 %. For example I have a Ryzen 9900X. As it has 24 virtual cores top scale goes to 2400%. Your load of 27 % is probably a light load if we take into account the maximum value of the scale top is using.

But it's illegal to have 24 cores. Honestly, I think there's just some sort of a bug somewhere.

@wwmm
Copy link
Owner

wwmm commented Mar 9, 2025

But it's illegal to have 24 cores.

😄

Honestly, I think there's just some sort of a bug somewhere.

I agree. There is a non negligible number of users having EasyEffects killed by the system. What is frustrating is that as I have never seen EasyEffects being killed on my computer I do not know even how I could try to debug this. Specially considering that the system does not say why it is killing EasyEffects.

When the oom kills an app because it used too much RAM there is a clear message in the system logs about the oom killing it because of this. And I think that something similar also happens when an app spends too much time in the realtime thread. But in these "killed" reports that I have seen in the last months the system says absolutely nothing about why it is killing our process. I wonder if there is a way to make that message more verbose...

@disposableone
Copy link
Author

I agree. There is a non negligible number of users having EasyEffects killed by the system. What is frustrating is that as I have never seen EasyEffects being killed on my computer I do not know even how I could try to debug this. Specially considering that the system does not say why it is killing EasyEffects.

Try running Firefox in a firejail for a day or two while using youtube and soundcloud. There's also a constant Widevinecdm plugin crash on Soundcloud that may be somehow causing this (I filed a bug on that to Firefox).

When the oom kills an app because it used too much RAM there is a clear message in the system logs about the oom killing it because of this. And I think that something similar also happens when an app spends too much time in the realtime thread. But in these "killed" reports that I have seen in the last months the system says absolutely nothing about why it is killing our process. I wonder if there is a way to make that message more verbose...

I have to add, I only have seen easyeffects killed once when I was listening to sound. Most of the times, it's killed when no active sound processing is/should be done (but there seems to be non-idle CPU usage!). Maybe one way to investigate is to see how easyeffects can have CPU usage when it's completely idle (no sound is played). For example, when I open xAI's Grok in Firefox, it shows as input in easyeffects and then STAYS there even if there's no sound is played or even if I just browse x.com in that tab. That's weird, right? Could be related.

@wwmm
Copy link
Owner

wwmm commented Mar 9, 2025

Try running Firefox in a firejail

I did some quick tests and nothing unusual happened. Using it for long times will be annoying when even simple things like saving files are broken because of the apparpor sandbox. And considering what firejails aims to do testing with it may not make sense. Things may be broken just because apparmor is not allowing something that has to be allowed.

@disposableone
Copy link
Author

Maybe it's easyeffects itself. Now I have Clocks output stream stuck in Output tab of easyeffects, even though no alarms are going off or scheduled and the Clocks app (GNOME) is closed. That's not normal. I'm also getting lsp-catalog-* files in /dev/shm/ lately all the time.

@wwmm
Copy link
Owner

wwmm commented Mar 10, 2025

Maybe it's easyeffects itself. Now I have Clocks output stream stuck in Output tab of easyeffects, even though no alarms are going off or scheduled and the Clocks app (GNOME) is closed. That's not normal. I'm also getting lsp-catalog-* files in /dev/shm/ lately all the time.

When this happens are these streams also visible in pavucontrol? I can think of a few possibilities for this behavior:

  • For some reason EasyEffects is not removing them from its list. This by itself would not make EasyEffects be killed. And I would expect this to be happening often and to more people
  • PipeWire is not emitting the signal about these streams removal. What would also be strange.
  • The application is not destroying its streams for some reason. Firefox is among the ones that sometimes keep some streams alive for reasons that are not clear

In any case none of this should have EasyEffects being killed as a result. You may be facing more than one issue at the same time.

@ZephyrsI
Copy link

I did face this issue (Fedora 41) with the default Fedora kernel, dbus just keeps killing easyeffects after every Spotify music track change. I'm now using a different kernel and the issue never occurred again and I stopped thinking about it.

@disposableone
Copy link
Author

I did face this issue (Fedora 41) with the default Fedora kernel, dbus just keeps killing easyeffects after every Spotify music track change. I'm now using a different kernel and the issue never occurred again and I stopped thinking about it.

What's your kernel version now?

@ZephyrsI
Copy link

What's your kernel version now?

I switched to https://github.com/CachyOS/copr-linux-cachyos

@wwmm
Copy link
Owner

wwmm commented Mar 11, 2025

I did face this issue (Fedora 41) with the default Fedora kernel, dbus just keeps killing easyeffects after every Spotify music track change. I'm now using a different kernel and the issue never occurred again and I stopped thinking about it.

I wonder why. The kernel version by itself should not be the problem. Otherwise I would also have seen this on Arch Linux. Maybe Fedora is doing something when configuring its kernel that makes EasyEffects to be killed.

@ZephyrsI
Copy link

ZephyrsI commented Mar 11, 2025

I did face this issue (Fedora 41) with the default Fedora kernel, dbus just keeps killing easyeffects after every Spotify music track change. I'm now using a different kernel and the issue never occurred again and I stopped thinking about it.

I wonder why. The kernel version by itself should not be the problem. Otherwise I would also have seen this on Arch Linux. Maybe Fedora is doing something when configuring its kernel that makes EasyEffects to be killed.

Back then I thought the Crystalizer effect was the root cause of audio stutters and SIGKILLs, but the issue persisted no matter how I configured my effects.
But after I switched to a different kernel, everything works fine without me changing anything related to EE.

So, based on my case, I don't think these kind of issues are caused by easyeffects. Most Fedora users won't even think of switching to a different kernel.

@disposableone
Copy link
Author

I noticed this persistent weird stuff from the tab where I open xAI's Grok in Firefox. It starts outputting sound (as evidenced by an output source added to Easyeffects), but there's nothing play in that tab, it's just Grok, you know? There's no sound there. And then it starts ramping up CPU cycles on Easyeffects, even though nothing is playing. I close the tab, easyeffects CPu usage goes back to idle, close to zero or whatever. Very suspicious. And I think I noticed these kills started happening around when I started using Grok, as I think I already mentioned, but not in a way that I can attribute it to that (as mentioned earlier, no reason for the kills by the kernel).

@wwmm
Copy link
Owner

wwmm commented Mar 11, 2025

I noticed this persistent weird stuff from the tab where I open xAI's Grok in Firefox. It starts outputting sound (as evidenced by an output source added to Easyeffects), but there's nothing play in that tab, it's just Grok, you know? There's no sound there. And then it starts ramping up CPU cycles on Easyeffects, even though nothing is playing. I close the tab, easyeffects CPu usage goes back to idle, close to zero or whatever. Very suspicious. And I think I noticed these kills started happening around when I started using Grok, as I think I already mentioned, but not in a way that I can attribute it to that (as mentioned earlier, no reason for the kills by the kernel).

Hum... When you have time run pw-top in a terminal and take a look at the QUANTUM value for Firefox and your soundcard line when Grok is open. Maybe it is making Firefox to request very low latency values. This could explain the higher than normal CPU usage.

@ZephyrsI
Copy link

Mar 11 23:07:21 deimos wireplumber[2642]: pw.node: (bluez_output) graph xrun not-triggered (17 suppressed)
Mar 11 23:07:21 deimos wireplumber[2642]: pw.node: (-75) xrun state:0x7fa75d535008 pending:0/12 s:1386558639984 a:1386568669239 f:1386570555361 waiting:10029255 process:1886122 status:triggered
Mar 11 23:07:23 deimos wireplumber[2642]: pw.node: (bluez_output) graph xrun not-triggered (4 suppressed)
Mar 11 23:07:23 deimos wireplumber[2642]: pw.node: (-75) xrun state:0x7fa75d535008 pending:10/12 s:1388761715436 a:1388765110154 f:1388765152605 waiting:3394718 process:42451 status:triggered
Mar 11 23:07:23 deimos wireplumber[2642]: pw.node: (-191) xrun state:0x7fa74c438008 pending:0/1 s:1388796994324 a:1388797121516 f:1388761922725 waiting:127192 process:18446744073674352825 status:awake
Mar 11 23:07:25 deimos wireplumber[2642]: pw.node: (bluez_output) graph xrun not-triggered (4 suppressed)
Mar 11 23:07:25 deimos wireplumber[2642]: pw.node: (-75) xrun state:0x7fa75d535008 pending:8/12 s:1390820333664 a:1390826849827 f:1390828119327 waiting:6516163 process:1269500 status:triggered
Mar 11 23:07:25 deimos wireplumber[2642]: pw.node: (-110) xrun state:0x7fa74c43a008 pending:0/2 s:1390841668309 a:1390823685058 f:1390823759218 waiting:18446744073691568365 process:74160 status:triggered
Mar 11 23:07:28 deimos wireplumber[2642]: pw.node: (bluez_output) graph xrun not-triggered (1 suppressed)
Mar 11 23:07:28 deimos wireplumber[2642]: pw.node: (-75) xrun state:0x7fa75d535008 pending:3/12 s:1393807035502 a:1393810129818 f:1393811366861 waiting:3094316 process:1237043 status:triggered
Mar 11 23:07:28 deimos wireplumber[2642]: pw.node: (-129) xrun state:0x7fa74d001008 pending:0/2 s:1393833091401 a:1393833097361 f:1393809916892 waiting:5960 process:18446744073686371147 status:awake
Mar 11 23:07:33 deimos wireplumber[2642]: pw.node: (bluez_output) graph xrun not-triggered (43 suppressed)
Mar 11 23:07:33 deimos wireplumber[2642]: pw.node: (-75) xrun state:0x7fa75d535008 pending:1/12 s:1398798961129 a:1398804781740 f:1398808137877 waiting:5820611 process:3356137 status:triggered
Mar 11 23:07:36 deimos wireplumber[2642]: pw.node: (bluez_output) graph xrun not-triggered (8 suppressed)
Mar 11 23:07:36 deimos wireplumber[2642]: pw.node: (-75) xrun state:0x7fa75d535008 pending:0/12 s:1401550963426 a:1401558103906 f:1401558425532 waiting:7140480 process:321626 status:triggered
Mar 11 23:07:36 deimos systemd[2514]: dbus-:[email protected]: Main process exited, code=killed, status=9/KILL
Mar 11 23:07:36 deimos systemd[2514]: dbus-:[email protected]: Failed with result 'signal'.
Mar 11 23:07:36 deimos systemd[2514]: dbus-:[email protected]: Consumed 41.713s CPU time, 134.4M memory peak.

I booted back to Fedora's kernel Linux 6.12.13-200.fc41.x86_64 to check, the audio crackles a lot with wireplumber just spamming a lot of these, and then eventually dbus killed EE. This doesn't happen with cachyos kernel.

@wwmm
Copy link
Owner

wwmm commented Mar 11, 2025

I booted back to Fedora's kernel Linux 6.12.13-200.fc41.x86_64 to check, the audio crackles a lot with wireplumber just spamming a lot of these, and then eventually dbus killed EE. This doesn't happen with cachyos kernel.

That is a lot of xrun errors. Not good. It is a sign that the system is not being able to process audio buffers fast enough. I wonder what kind of configuration is being done in the standard Fedora kernel. Although I have kernel 6.13.6 on Arch Linux at this moment I also used 6.12.x when it was the one available on Arch's repositories. That kernel was totally fine. It has to be some kind of configuration choice that is doing this.

@disposableone
Copy link
Author

I booted back to Fedora's kernel `Linux 6.12.13-200.fc41.x86_64` to check, the audio crackles a lot with wireplumber just spamming a lot of these, and then eventually dbus killed EE. This doesn't happen with cachyos kernel.

Ha, I'm on 6.12.13. No crackling much, but lots of kills. I'll update soon, I guess, before anything.

@disposableone
Copy link
Author

Hum... When you have time run pw-top in a terminal and take a look at the QUANTUM value for Firefox and your soundcard line when Grok is open. Maybe it is making Firefox to request very low latency values. This could explain the higher than normal CPU usage.

Quantum for Firefox is 900, alsa output quantum is 512. Is that good?

@ZephyrsI
Copy link

I suspect the Crystalizer effect has some kind of part in my case, since this issue happens more frequently if I have it enabled, I don't know about other cases though.

@wwmm
Copy link
Owner

wwmm commented Mar 11, 2025

Quantum for Firefox is 900, alsa output quantum is 512. Is that good?

It depends on the sampling rate. But for the usual values of 48 kHz and 44.1 kHz it should not be a problem. Depending on the system it may cause crackling. But it should not use as much CPU as a quantum of 128 or 64. Something you may try is temporarily forcing a fixed and higher quantum like 1024 or 2048 https://gitlab.freedesktop.org/pipewire/pipewire/-/wikis/Guide-Latency and see if that changes anything.

I suspect the Crystalizer effect has some kind of part in my case, since this issue happens more frequently if I have it enabled, I don't know about other cases though.

This is one of our heaviest plugins and does several calls to the zita-convolver library. That in turns uses fftw3 a lot when doing Fast Fourier Transforms. Depending on what instructions the CPU has or not available this isn't a cheap operation at all. So it is not surprising this plugin makes it easier to cause xruns. And in turn the crash.

@wwmm
Copy link
Owner

wwmm commented Mar 11, 2025

Something you may try is temporarily forcing a fixed and higher quantum like 1024 or 2048

Always power of 2 values. They are more efficient. And if I am not mistaken the range goes from 32 up to 8192.

@disposableone
Copy link
Author

Ok, I removed Crystallizer effect from the rack, I don't wanna jinx it, but it seems to not crash anymore. We'll see in a few days for sure, of course.

Grok, though, keeps easyeffects not idle, with some CPU usage, nothing crazy, but what the hell is going on???

@wwmm
Copy link
Owner

wwmm commented Mar 12, 2025

Grok, though, keeps easyeffects not idle, with some CPU usage, nothing crazy, but what the hell is going on???

My guess is that it is the same or similar to the streams that Pavucontrol creates to measure volume levels. Is it recording audio from a device or playing to it?

@disposableone
Copy link
Author

Grok, though, keeps easyeffects not idle, with some CPU usage, nothing crazy, but what the hell is going on???

My guess is that it is the same or similar to the streams that Pavucontrol creates to measure volume levels. Is it recording audio from a device or playing to it?

No, it's Grok, it's not doing anything, it's just text, no audio input/output (supposedly). So it's possible it's measuring volume levels? That's bizarre.

@wwmm
Copy link
Owner

wwmm commented Mar 12, 2025

No, it's Grok, it's not doing anything, it's just text, no audio input/output (supposedly). So it's possible it's measuring volume levels? That's bizarre.

Hum... I am not familiar with it. Is it one of those AI chat bots for Firefox? If yes one possibility is that it creates a stream for "text to speech". That kind of functionality for people that can not type in a keyboard. For example blind people. Usually Firefox tries to use the package speex-dispatcher for this. Maybe Grok does something similar.

@disposableone
Copy link
Author

No, it's Grok, it's not doing anything, it's just text, no audio input/output (supposedly). So it's possible it's measuring volume levels? That's bizarre.

Hum... I am not familiar with it. Is it one of those AI chat bots for Firefox? If yes one possibility is that it creates a stream for "text to speech". That kind of functionality for people that can not type in a keyboard. For example blind people. Usually Firefox tries to use the package speex-dispatcher for this. Maybe Grok does something similar.

It's not a bot, it's x.com/i/grok webpage. I thought no website can use speech-dispatcher unless requested by user. I don't have speech-dispatcher installed.

@wwmm
Copy link
Owner

wwmm commented Mar 12, 2025

I thought no website can use speech-dispatcher unless requested by user. I don't have speech-dispatcher installed.

If I remember well speech-dispatcher creates its own stream independent from the browser and its name is in the stream name. If it were it you would recognize it immediately.

@wwmm
Copy link
Owner

wwmm commented Mar 12, 2025

Although I do not have a X account I've decided to type x.com/i/grok on Firefox. It did create a playback stream that plays silence. What will keep EasyEffects active like any other player stream. Weird... This playback stream seems useless... And the same thing happens on Chromium. So it is not something specific to Firefox.

What is curious is that on both browsers no stream is created if I just type x.com.

@disposableone
Copy link
Author

Ok, I am still getting easyeffects killed without Crystallizer on. I blame Grok tab doing something weird.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

4 participants