Well that didn’t show much.
The 5900X is notorious for WHEA 18/19 errors, which are hard crashes caused by the CPU core power states dropping too low when idle or under light loads.
Edit: This is Windows based info but has some relevance.
You could try the following.
Boot into your BIOS and navigate to your AMD Overclocking settings. Change Power Supply Idle Control from “Auto” to Typical Current Idle.
I did the bios change. Now we wait. (unless any other suggestions pop-up - I’m happy to try anything!)
If it’s gpu related I have some more kernel parameters you can try. See first of all if that setting change makes any difference.
Edit: You can try these individually or all at once.
amdgpu.ppfeaturemask=0xffffffff
amdgpu.aspm=0
pcie_aspm=off
These are kernel parameters. Are you using systemd-boot or grub?
Edit: If using grub you add them to the command line in /etc/default/grub and save and then run the update grub command sudo grub-mkconfig -o /boot/grub/grub.cfg
If you are using systemd-boot you add them to the command line in /etc/kernel/cmdline
Then run sudo reinstall-kernels
The crashes usually occur when I use the browser or interact with UIs in games. I’ve never had it occur when I was just playing a game normally. I’ll observe how it unfolds. The sheer amount of crashing conditioned me not to use my PC normally, and it is hard to get over. I haven’t played games much at all due to the fear of progression loss. ( 0 _ 0 )
I have the default which I believe is systemd-boot.
I’ve read pcie_aspm=off and it prevents a connection drop. Is there a parameter to prevent ethernet connection drops? I get quite frequent notification pop-ups saying there is a network connectivity loss. Could those all be connected?
Should I follow https://discovery.endeavouros.com/system-rescue/systemd-boot/2022/12/?
When I sudo nano /etc/kernel/cmdline, I get nvme_load=YES nowatchdog rw root=UUID=bc8408f1-6c3c-4b8f-a6e8-f22d295a83d9
Is amdgpu.ppfeaturemask=0xffffffff necessary?
I seem to get those too.
It’s just something you can try.
Edit: You are just adding this as a kernel parameter to what is already existing leaving a space between. Then run the reinstall kernels command.
Edit: If you find the kernel parameters don’t work you can remove them and then run the reinstall kernels command again.
Edit: These are just things you can try because I don’t have a crystal ball to determine exactly the reason the crashes are happening so it’s a little bit of trial and error which may or may not give the results we are looking for.
@pallid
Hardware is tricky and a lot of issues that happen can be related to many different things such as poor implementation of the UEFI Bios or firmware, power or software.
Edit: Everything in your hardware output looks fine and your UEFI Bios (Firmware) is up to date and you have checked memory and hard drive so you must look for other things.
Edit: Also if I’m not mistaken you are up to date? Kde just had some more updates.
@pallid
I have the 5800X with RX 6750XT and have no issues! Not a gamer though.
Edit:
[ricklinux@rick-ms7c91 ~]$ inxi -CG
CPU:
Info: 8-core model: AMD Ryzen 7 5800X bits: 64 type: MT MCP cache: L2: 4 MiB
Speed (MHz): avg: 1754 min/max: 556/4854 cores: 1: 1754 2: 1754 3: 1754
4: 1754 5: 1754 6: 1754 7: 1754 8: 1754 9: 1754 10: 1754 11: 1754 12: 1754
13: 1754 14: 1754 15: 1754 16: 1754
Graphics:
Device-1: Advanced Micro Devices [AMD/ATI] Navi 22 [Radeon RX 6700/6700
XT/6750 XT / 6800M/6850M XT] driver: amdgpu v: kernel
Display: wayland server: X.org v: 1.21.1.23 with: Xwayland v: 24.1.12
compositor: kwin_wayland driver: X: loaded: amdgpu
unloaded: modesetting,radeon dri: radeonsi gpu: amdgpu
resolution: 1920x1080~60Hz
API: EGL v: 1.5 drivers: radeonsi,swrast
platforms: gbm,wayland,x11,surfaceless,device
API: OpenGL v: 4.6 vendor: amd mesa v: 26.1.3-arch1.2 renderer: AMD
Radeon RX 6750 XT (radeonsi navi22 ACO DRM 3.64 7.0.14-arch1-1)
API: Vulkan v: 1.4.350 drivers: radv surfaces: N/A
Info: Tools: api: clinfo, eglinfo, glxinfo, vulkaninfo
de: kscreen-console,kscreen-doctor wl: wayland-info
x11: xdpyinfo, xprop, xrandr
I’m late getting back to you, I’m sorry.
Smart tests usually guage the health of your SSD/HDD.
Others have addressed that in this thread.
Personally I use gnome-disks (not in /extra anymore!) and a standalone called HDSentinel (https://www.hdsentinel.com/hard_disk_sentinel_linux.php) but since you revealed it only crashes when you do gaming stuff it might not be SSD. But you never know.
extra/gnome-disk-utility 46.1-2 (gnome)
Disk Management Utility for GNOME
My -Ss search had an “s” at the end of it!!! No wonder!!! D’oh!
@pallid
Still having crashes?
I’ve just gotten back from the long shift at work. I’ll see. Hopefully not. I’d say I’m most likely to find out tomorrow - assuming they persist - since I’ll probably fall asleep in a second.
Thank you for checking in. ଘ(੭*ˊᵕˋ)੭* ̀ˋ
I haven’t had any further crashes - so far, anyway.
I can’t tell which step helped, sorry.
If you ever find a solution to the ethernet disconnecting, please, share it! @ricklinux
Thank you so much to everyone for your patience and help. ʕ·͡ᴥ·ʔ
I crashed again, but the error looked different. I didn’t see it mentioning hardware error. There was a lot of numbers separated by periods.
Journalctl -1 result: https://dpaste.com/4GF7SWGDT
Is it related to the new kernel commands?
Where did you see this?
Not sure what you are referring to?
The error I saw was during the reboot after crash.
After removing the kernel lines, I am back to getting the hardware errors upon crash.
I’m not sure what you mean here after removing the kernel lines? Also hardware errors upon crash. I’d have to see it.
Edit: have you tried disabling global c-states in the Bios for CPU
Edit: Have you tried this kernel parameter?
amdgpu.ppfeaturemask=0xffffffff amdgpu.aspm=0
When I said kernel lines, I meant parameters.
The crashes had gotten worse. I crashed into a black screen which wouldn’t go away, with no text. I had to unplug my computer to turn it off, it wouldn’t even react to the power button being held.
I already had amdgpu.aspm=0, I will add amdgpu.ppfeaturemask=0xffffffffnow. I am very distraught by the amount of crashing.
I thought journalctl shows the errors, no? I’m not quick enough to take a photo with my phone. 
I’ll check the BIOS c-states.
The error I last saw (excluding the total black screen) mentioned the CPU, I think…?
Is there any way to check the start-up error if journalctl doesn’t show it?