@Alex_H @Sh_Ra Have you both reported issue to upstream drm/amd ?
I’ve ran into the same page fault issue but not with a VM workload and I’m under BIOS 3.05.
EDIT : No graphical glitch on my side however, I haven’t used an external monitor with a higher res than 1920x1080
No, and I don’t really care who and which issue trackers/chip suppliers/whatever should be informed here. I’m just a basic end-user reporting Framework issues back to Framework 
My reproduction guide uses only the internal monitor (2256x1504). Just make sure to maximize the VM window and the browser inside it.
Who is responsible for this issue? It prevents me from running certain workloads on fullscreen.
I don’t expect it to get fixed today, I just want it to be seen.
Page faults like that are usually user mode driver (IE mesa) bugs. You should try to reproduce it using the latest mesa and if you can still reproduce it report it there.
I’m unable to reproduce the glitch on my end following the steps…
I ran the benchmark on host and inside the vm, still nothing.
BIOS 3.05, Pop!_OS, kernel 6.8.0, with 1 external monitor
I used these settings in virt-manager,
maximized the VM window and its containing browser window, yes
This issue is no longer reproducible, no matter what I try. Guys, 4k fullscreen Apps are here now. 
Truly the year of the Linux desktop.
Kernel: 6.9.11-200.fc40.x86_64
Glad to hear it @Alex_H ! My partner is also on a Framework 13 AMD, running Fedora 40, and says he occasionally finds his system suddenly slows to a crawl (1 frame rendered about every 3 seconds) before hanging completely after about 20 seconds. He does game development so I’m imagining it could be a GPU thing for him too (I watched it happen today while he had a game dev program - Godot - open)
I’ll tell him to make sure his kernel is up-to-date. Just to confirm, did you have to update the BIOS or anything like that as well? If so, does that happen automatically via the Software app or does he need to do anything to trigger it? He hasn’t done anything manual to update the BIOS since he bought the machine 6 months ago.
I don’t know which of the countless updates did it and when it was fixed. Just discovered it a few days ago. But yes, I’m on the latest everything and recall upgrading my BIOS a few months ago. It is just a simple fwupmgr command and a reboot, there are guides out there. That said, my problem could be different from the one you described. Won’t help you here, but check dmesg for logs.
I think a new post should be made for new issues, I’m still encountering issue on kernel 6.10.3 with the driver AMDGPU (BIOS 3.0.5) but they are very difficult to reproduce :
[61942.825121] amdgpu 0000:c1:00.0: amdgpu: [gfxhub] page fault (src_id:0 ring:24 vmid:7 pasid:32786)
[61942.825132] amdgpu 0000:c1:00.0: amdgpu: in process .kitty-wrapped pid 3017 thread kitty:cs0 pid 3018)
[61942.825137] amdgpu 0000:c1:00.0: amdgpu: in page starting at address 0x0000f5ae3ed33000 from client 10
[61942.825140] amdgpu 0000:c1:00.0: amdgpu: GCVM_L2_PROTECTION_FAULT_STATUS:0x00701430
[61942.825143] amdgpu 0000:c1:00.0: amdgpu: Faulty UTCL2 client ID: SQC (data) (0xa)
[61942.825146] amdgpu 0000:c1:00.0: amdgpu: MORE_FAULTS: 0x0
[61942.825148] amdgpu 0000:c1:00.0: amdgpu: WALKER_ERROR: 0x0
[61942.825150] amdgpu 0000:c1:00.0: amdgpu: PERMISSION_FAULTS: 0x3
[61942.825153] amdgpu 0000:c1:00.0: amdgpu: MAPPING_ERROR: 0x0
[61942.825155] amdgpu 0000:c1:00.0: amdgpu: RW: 0x0
[61952.994551] [drm:amdgpu_job_timedout [amdgpu]] *ERROR* ring gfx_0.0.0 timeout, but soft recovered
EDIT: However this does not do any graphical corruption, just crashes my GPU accelerated terminal, the desktop freezes for a bit and then unfreezes.
This is very likely a mesa bug, it’s not the same as previous.
Got the same issue here. I was playing 1080p video in Firefox and then the desktop froze and crashed
There is another thread talking about this happening with 6.10 on framework website but not with 6.9. does that match your behavior?
You should assist with bisecting if so.