[SOLVED] Random system freezes/reboots

Gricenchos · 2025-04-27 05:32:29

I've found out that it's possible to use the curve optimiser with the ryzenadj package. For now I've removed all kernel parameters and tried to add a bit more voltage to the CPU through that. Will report on the results.

GMP wrote:

Gricenchos wrote:
Edit 2: processor.max_cstate=1 doesn't seem to help much. Still getting freezes - it seems like kwin freezing? Unsure.
Thanks for the update.
The behavior you're describing now — graphical freezes (kwin locking up) instead of full kernel panic — is a different failure mode.
We need journal logs to be sure whether this is a GPU driver hang or something else.
Meanwhile:
Since
processor.max_cstate=1
alone is insufficient, the next consistent step is to also add:
idle=nomwait
This prevents the kernel from using the MWAIT instruction during idle, which can trigger deep sleep instability even when deeper C-states are disabled.
This is not random trial-and-error — it's a documented mitigation path for Ryzen C-state related issues.
No new kernel parameters beyond that at this stage. Clean isolation.
Edit:
Post journal from previous boot:
sudo journalctl -b -1 -p 3
or full log:
sudo journalctl -b -1 | curl -F 'file=@-' 0x0.st

Here's the full log, since it was a couple reboots ago:

http://0x0.st/8WXZ.txt

I will return the mentioned parameters when/if I get another freeze/panic and continue from there.

Last edited by Gricenchos (2025-04-27 05:35:19)

Gricenchos · 2025-04-29 11:56:28

It seems like curve optimiser have worked. Ryzenadj allowed me to put more voltage in the CPU via

sudo ryzenadj --set-coall=15

It allowed me a stable uptime of 24h. No kernel parameters. It, sadly, has to be run at every startup, is there a way to automatically execute it as a superuser?

I'll monitor the system behaviour for a few more days just in case and post logs if something happens again.

For some reason VLC, of all things, sometimes makes the system freeze? But no kernel panics on exiting suspend.

Last edited by Gricenchos (2025-04-29 15:25:41)

seth · 2025-04-29 15:16:36

You mean like a systemd service?
https://wiki.archlinux.org/title/System … unit_files

In case it is and please always remember to mark resolved threads by editing your initial posts subject - so others will know that there's no task left, but maybe a solution to find.
Thanks.

Gricenchos · 2025-04-29 15:28:10

seth wrote:

You mean like a systemd service?
https://wiki.archlinux.org/title/System … unit_files
In case it is and please always remember to mark resolved threads by editing your initial posts subject - so others will know that there's no task left, but maybe a solution to find.
Thanks.

Yes, thank you! I'll look into it.

I'll monitor the system behaviour for a couple more days just in case, and if it behaves, will mark the thread as solved, if that's fine.

Gricenchos · 2025-05-01 07:19:19

Well, after a couple days of monitoring, I can say that the kernel panics have stopped, but I still occasionally get system freeze on waking form suspend.

seth · 2025-05-01 13:15:25

That's probably different, how bad are those freezes? Can you still reboot w/ teh https://wiki.archlinux.org/title/Keyboa … el_(SysRq) (reisub - has to be explicitly enabled *before* the freeze)
Anything suspicious flaring up even during successful suspends?

Gricenchos · 2025-05-01 14:39:44

seth wrote:

That's probably different, how bad are those freezes? Can you still reboot w/ teh https://wiki.archlinux.org/title/Keyboa … el_(SysRq) (reisub - has to be explicitly enabled *before* the freeze)
Anything suspicious flaring up even during successful suspends?

No, SysRq reboot doesn't work during those. Nothing suspicious flares up during normal suspends either.

seth · 2025-05-01 15:10:22

Post a journal covering a successful suspend cycle - we'll just walk down the usual suspects from there…

Gricenchos · 2025-05-01 19:35:01

seth wrote:

Post a journal covering a successful suspend cycle - we'll just walk down the usual suspects from there…

Here's the journal after going to suspend and successfully waking from it:

http://0x0.st/84ql.txt

seth · 2025-05-01 20:03:18

Nope, journal covers 80 seconds, no sleep nor attempt.

Gricenchos · 2025-05-01 20:22:20

Sorry, my mistake. This should be the correct one:

http://0x0.st/84bM.txt

Went into suspend, woke up, ran journalctl --since "5 minutes ago"

seth · 2025-05-01 21:35:31

Nedxt to https://wiki.archlinux.org/title/Solid_ … leshooting the cs35l41-hda device seems rather exotic
Add "nvme_core.default_ps_max_latency_us=0 iommu=soft pcie_aspm=off pcie_port_pm=off" to the kernel parameters and see whether you still into wakeup freezes.

Gricenchos · 2025-05-02 16:07:03

seth wrote:

Nedxt to https://wiki.archlinux.org/title/Solid_ … leshooting the cs35l41-hda device seems rather exotic
Add "nvme_core.default_ps_max_latency_us=0 iommu=soft pcie_aspm=off pcie_port_pm=off" to the kernel parameters and see whether you still into wakeup freezes.

Sadly, those didn't do the trick, still got a freeze on a wakeup.

seth · 2025-05-02 22:06:08

Can you "modprobe -r cs35l41-hda" before the suspend or test to blacklist it for a while to see whether the freeze still happens without?

Gricenchos · 2025-05-03 05:31:28

Got another kernel panic today, on boot, interestingly. Here's the report:

https://tinyurl.com/yfa679xv

I suspect that the issue is the same, it's just that since I'm applying the curve optimiser by a system command, not through BIOS, it didn't have time to activate, and that's the reason for the panic.

seth wrote:

Can you "modprobe -r cs35l41-hda" before the suspend or test to blacklist it for a while to see whether the freeze still happens without?

Tried that command.

modprobe: FATAL: Module cs35l41-hda not found.

seth · 2025-05-03 06:51:11

It's actually named "snd-hda-scodec-cs35l41", sorry.

Gricenchos · 2025-05-03 07:03:38

seth wrote:

It's actually named "snd-hda-scodec-cs35l41", sorry.

it's alright. However, now modprobe -r doesn't work, as it reports the module to be in use.

seth · 2025-05-03 14:54:45

Probably by pipewire - did you try to just blacklist it?

Gricenchos · 2025-05-03 17:21:25

seth wrote:

Probably by pipewire - did you try to just blacklist it?

Sorry, didn't do that yet. Will try that now: is there no risk of breaking something, if it's being used currently?

seth · 2025-05-03 18:54:42

At worst your sound won't work while blacklisting that.

Gricenchos · 2025-05-03 20:11:33

seth wrote:

At worst your sound won't work while blacklisting that.

Blacklisted. Will report on the results.

Gricenchos · 2025-05-04 05:53:07

seth wrote:

It's actually named "snd-hda-scodec-cs35l41", sorry.

Okay, for some reason I can't blacklist that module? Whether I blacklist it though kernel parameters or /etc/modprope.d/blacklist.conf, it remains in use.

seth · 2025-05-04 05:58:06

module_blacklist=snd-hda-scodec-cs35l41

shows up in "cat /proc/cmdline" and what does

lsmod | grep cs35l41

then print?

Gricenchos · 2025-05-04 07:41:47

cat /proc/cmdline:

BOOT_IMAGE=/vmlinuz-linux-lts root=UUID=017f6ef1-19eb-4578-b831-fa02807eedc7 rw zswap.enabled=0 rootfstype=ext4 module_blacklist=snd-hda-scodec-cs35l41 loglevel=3 quiet

lsmod | grep cs35141

snd_hda_scodec_cs35l41_spi    12288  0
snd_hda_scodec_cs35l41_i2c    12288  0
snd_hda_scodec_cs35l41    77824  2 snd_hda_scodec_cs35l41_i2c,snd_hda_scodec_cs35l41_spi
snd_hda_cs_dsp_ctls    20480  1 snd_hda_scodec_cs35l41
snd_soc_cs_amp_lib     16384  1 snd_hda_scodec_cs35l41
snd_soc_cs35l41_lib    45056  3 snd_hda_scodec_cs35l41_i2c,snd_hda_scodec_cs35l41,snd_hda_scodec_cs35l41_spi
cs_dsp                 90112  3 snd_hda_scodec_cs35l41,snd_hda_cs_dsp_ctls,snd_soc_cs_amp_lib
snd                   155648  64 snd_ctl_led,snd_hda_codec_generic,snd_seq,snd_seq_device,snd_hda_codec_hdmi,snd_hwdep,snd_hda_intel,snd_hda_codec,snd_hda_codec_realtek,snd_sof,snd_hda_scodec_cs35l41,snd_timer,snd_compress,snd_soc_core,snd_pcm,snd_hda_cs_dsp_ctls

seth · 2025-05-04 12:49:31

"module_blacklist=snd_hda_scodec_cs35l41"?

Arch Linux

#51 2025-04-27 05:32:29

Re: [SOLVED] Random system freezes/reboots

#52 2025-04-29 11:56:28

Re: [SOLVED] Random system freezes/reboots

#53 2025-04-29 15:16:36

Re: [SOLVED] Random system freezes/reboots

#54 2025-04-29 15:28:10

Re: [SOLVED] Random system freezes/reboots

#55 2025-05-01 07:19:19

Re: [SOLVED] Random system freezes/reboots

#56 2025-05-01 13:15:25

Re: [SOLVED] Random system freezes/reboots

#57 2025-05-01 14:39:44

Re: [SOLVED] Random system freezes/reboots

#58 2025-05-01 15:10:22

Re: [SOLVED] Random system freezes/reboots

#59 2025-05-01 19:35:01

Re: [SOLVED] Random system freezes/reboots

#60 2025-05-01 20:03:18

Re: [SOLVED] Random system freezes/reboots

#61 2025-05-01 20:22:20

Re: [SOLVED] Random system freezes/reboots

#62 2025-05-01 21:35:31

Re: [SOLVED] Random system freezes/reboots

#63 2025-05-02 16:07:03

Re: [SOLVED] Random system freezes/reboots

#64 2025-05-02 22:06:08

Re: [SOLVED] Random system freezes/reboots

#65 2025-05-03 05:31:28

Re: [SOLVED] Random system freezes/reboots

#66 2025-05-03 06:51:11

Re: [SOLVED] Random system freezes/reboots

#67 2025-05-03 07:03:38

Re: [SOLVED] Random system freezes/reboots

#68 2025-05-03 14:54:45

Re: [SOLVED] Random system freezes/reboots

#69 2025-05-03 17:21:25

Re: [SOLVED] Random system freezes/reboots

#70 2025-05-03 18:54:42

Re: [SOLVED] Random system freezes/reboots

#71 2025-05-03 20:11:33

Re: [SOLVED] Random system freezes/reboots

#72 2025-05-04 05:53:07

Re: [SOLVED] Random system freezes/reboots

#73 2025-05-04 05:58:06

Re: [SOLVED] Random system freezes/reboots

#74 2025-05-04 07:41:47

Re: [SOLVED] Random system freezes/reboots

#75 2025-05-04 12:49:31

Re: [SOLVED] Random system freezes/reboots

Board footer