Skip to main content ↑ ↓
Topic: Multifaceted Problem: Random Crashes and Empty Logs (Read 5875 times) previous topic - next topic
0 Members and 2 Guests are viewing this topic.

Re: Multifaceted Problem: Random Crashes and Empty Logs

Reply #15 –
I know how frustrating this is, I had multiple computers crash on me for weeks because of mesa. Perhaps you should try the latest mainline kernel in hopes of better cooperation? Or try some kernel parameters from the related Arch wiki entry.

Re: Multifaceted Problem: Random Crashes and Empty Logs

Reply #16 –
Quote from: Antilles – on
Alright, here are some new logs after the latest crash (in my absence, this time). Using Metalog, I've only one from it and two from the list that're generally required (as their output's the same as last time).

dmesg

kernel/current

Xorg.0.log

I do not know if there's considerable overlap between dmesg's output and that of Metalog's kernel log, but it seemed to have more helpful information. Seems the problems started at 11:23:58.


I don't see a crash FWIW

Re: Multifaceted Problem: Random Crashes and Empty Logs

Reply #17 –
Quote from: nous – on
I know how frustrating this is, I had multiple computers crash on me for weeks because of mesa. Perhaps you should try the latest mainline kernel in hopes of better cooperation? Or try some kernel parameters from the related Arch wiki entry.

Well, at least it isn't happening 3+ times back-to-back anymore. lol Unfortunately, it just got me while I was listening to a stream and playing Ultimate Doom.

That said, I am a bit surprised that you'd suggest something other than the LTS that I've been using--I thought that that was the most stable variant of the kernel. (Speaking of kernels, the way you spoke of what I have made it seem as if it wasn't the latest--I thought that a full system upgrade would update whichever kernel I have to the latest version.)

The kernel parameters look pretty relevant, actually--I'll have to give all of these a shot when I have more time.

Finally, I am curious about one thing: can a kernel module that modifies a Gamecube controller adapter's polling rate cause such a crash? It's also on the AUR. The crash does come even when the USB hub the adapter is connected to isn't powering the adapter's ports (or any other, for that matter).

EDIT: Another question: does changing kernel parameters require a reboot for the kernel to recognize the change?

Quote from: mrbrklyn – on
I don't see a crash FWIW

lol In all honesty, I'm right there with you. I only guessed at what might have been something, because there has to be some cause tied to this effect.

Re: Multifaceted Problem: Random Crashes and Empty Logs

Reply #18 –
Kernel - 061225

Another crash. The timestamp should be Jun 12 10:33:56. I have, after the last post, switched my PC's kernel to the latest mainline version (6.14.6.artix1-1). I noticed an error stating "flip_done timed out", but I remember trying the AMDGPU article's solution to no avail.

I'm instead going to try this kernel parameter: amdgpu.ppfeaturemask=0xfffd3fff. I found it in this Reddit topic.

Re: Multifaceted Problem: Random Crashes and Empty Logs

Reply #19 –
It's been two weeks since I implemented the aforementioned solution; so far, I haven't experienced a single crash--and it usually happens by this time. I'll report back after another two weeks of observation time.
 

Re: Multifaceted Problem: Random Crashes and Empty Logs

Reply #20 –
A month had passed without any crashes whatsoever...and then a little while ago, one occurred. Longest stretch of time, at least; I thought it might have been because my knee clipped my PC's case when I went AFK, but the errors appear to be exactly the same as they were in the past. My knee or shin has contacted the case multiple times over the month without anything happening, in addition.

Here's a kernel log.