Two system crashes pop!os, amd or nvidia problems or maybe related

Operating Systems & Open Source Linux
,
1

Crash 1 screen froze, had to do a hard reset. two days ago
I’m using pop!os with kde plasma (x11), linux kernel 7.0.11-76070011-generic, radeon 740M and rtx 4050 max-q. Victus HP Laptop 15-fb3093dx. I had a gpu crash two days ago (10/12) I’ve been trying to figure out. This one didn’t restart the whole system, but the amd gpu apparently hung up or something, the screen froze completely and stayed frozen for at least 30 minutes. I left it frozen because i was busy at the time. i looked at the logs and after the laptop woke up, it started getting a ton of PME: Spurious native interrupt! messages. Like, almost 8000 of them. later that evening the amd gpu crashedwhile firefox was running. The logs have stuff like ring gfx_0.0.0 timeout, MES failed to respond to msg=RESET, Ring gfx_0.0.0 reset failed. The logs showed it did reset the gpu but there were some more errors after that. Anyway I had to force a hard reset. It started back up normally and everything seemed to work just fine. But I’m not really sure what’s going on with it. From what I can tell (which isn’t much) it might be something with the amd driver, gpu scheduling (idk, I googled it but I’m not sure), PCIe power management, or some combination? (So, like, pretty much anything, lol). firefox was apparently using the gpu when it happened, but i don’t know if firefox actually caused anything. The thing that seems weird to me is all those PCIe messages starting after it woke up from sleep, and they kept going even after the gpu crashed/recovered.

Crash 2 from today (actually now as i’m posting this it’s ‘yesterday’) , entire computer restarted
Same computer, and also nvidia driver 595.104.02. Was using it earlier today, everything was working perfectly normally, then out of nowhere the screen went completely black for like 4 or 5 sec, i was in the middle of typing on the terminal and the entire computer just restarted. No warning or anything, just crashed and rebooted, and I lost the stuff I had open that wasn’t saved. I looked at the logs again, (and everything seemed to be running perfectly again like, for user experience) and couldn’t find anything showing what happened. They just kind of stop, and then the next boot starts. i did see when it started back up, the nvidia driver was being weird about initializing the rtx 4050. There were errors like unexpected WPR2 already up, Cannot initialize GSP firmware RM, RmInitAdapter failed! (0x62:0x40:2168). It even said the gpu was probably in a “bad state” and might need to be reset.

idk. I’m just trying to figur out what the cause is. nvidia drivers? gsp firmware? Or like, do i have it backwards ? Did the computer crash cause the errors when it restarted. Idk. the gpu seems to be working fine, nvidia-smi shows it and stuff. but with the crash i had two days ago, the amd gpu could be related or unrelated, but I figured I’d mention it because I don’t know if there’s something going on with the kernel, the gpu drivers (x2), or whatever is causing the crashes.

I attaxhed the logs I got from the crashes, and the stuff from when it woke up, amd gpu errors, lastfew minutes before the restart, nvidia errors after it booted back up. Also the system/hardware info in case any of that helps.

system_info

crash1_suspend

crash1_amdgpu

crash2_before_reboot

crash2_after_reboot

2

The two crashes are likely separate problems. In the first, the 740M’s graphics ring hung while Firefox was rendering and the driver recovered with a reset.

The NVIDIA GSP errors after the second crash are most likely a result of the 4050 not powering off fully during the reset.

Update linux-firmware and the kernel to the newest versions Pop!_OS offers.

If Firefox triggers the hang again, set gfx.webrender.software to true in about:config and see whether the hangs stop.

For the reboot that left nothing in the log, check for a saved crash record after the next one:

sudo ls /var/lib/systemd/pstore

If it’s empty, update the BIOS on the Victus to the latest version.

Drafted with AI.

3

Does POP offer KDE out of the box? Or did you add that yourself? What version? Who is still using X11? No modern DE offers X11.*

If you added KDE yourself, maybe use a distro built for KDE. Pop isn’t a great distro to begin with with the old Ubuntu 24.04 base.

Adding a DE to a distro is easy. But to make it work well, takes more work. If it was easy, all distros would offer all DE. They don’t, and there is a reason.

If you want Ubuntu with an old KDE, use Kubuntu. They paid KDE to make KDE 6.6 an LTS for them. Otherwise use a Fedora or Arch distro that keeps updating to new KDE version. Most KDE (except 6.6) are not supported for long - they require updating to new versions (like to 6.7, 6 8…). Most LTS distros have you stuck on old unsupported KDE versions

Did it work in the past, or is this a new installation and never worked?

*KDE 6.7.7 will be the last update next week. So yes, technically there still is a current KDE with X11. But just for some weeks. 6.8. already is the current version and has no X11. Gnome already abandoned X11 this spring. Cosmic never had X11.