Ubuntu Version:
Ubuntu 24.04.3 LTS
Desktop Environment (if applicable):
Gnome
Problem Description:
Although I have seen chatter around AMD’s iGPU on L1T, my unit has been working for 3+ month already without any issue. So I still hope this is relevant to some of the updates, applied over the last 7-10 days. Unless the funny word “degradation” is in order.
This is related to MS Edge(dev) running MS Teams as an app. The system just hangs (cursor, keyboard… but one time I was in a call, and could hear other people. A few times (I am getting the impression that if I don’t move the mousepointer) I managed to get with a few seconds of frozen screen. But most of the times - 5-15 seconds of frozen screen → black screen → session killed. One time (today) it managed to stay in the black screen (responsive, can type, but not tty).
From the perspective of Teams (yes, every time it happened, Teams was involved), it doesn’t need to be a call (video, or even audio). Simply moving it from minimized (yes, I forgot the word), or even switching to a different chat can cause this.
I tried looking through journalctl, but for odd reason, didn’t find anything for that time period (started accusing a faulty m.2 drive). But I decided to keep journalctl -t in background, and try to capture the problem. Which I did. And which I’m sharing.
I know that maybe “go buy a dGPU, pleb” would be an option (I am already considering it), but the system is strictly for work, and the iGPU was more than enough for my line of work.
P.S. I do realize that this may better be posted as a bug report, but I didn’t find where to do so (hopefully some moder will address this question 
)
Relevant System Information:
Kernel: Linux 6.14.0-29-generic
Mobo: MSI x870 Tomahawk (not newest bios, but lets leave this aside)
CPU: Ryzen 9900x
GPU: no discrete GPU, using iGPU
Screenshots or Error Messages:
Sep 04 16:44:08 amdpc kernel: amdgpu 0000:6e:00.0: amdgpu: [gfxhub] page fault (src_id:0 ring:24 vmid:7 pasid:32771)
Sep 04 16:44:08 amdpc kernel: amdgpu 0000:6e:00.0: amdgpu: in process msedge pid 7090 thread msedge:cs0 pid 7119
Sep 04 16:44:08 amdpc kernel: amdgpu 0000:6e:00.0: amdgpu: in page starting at address 0x0000000000000000 from client 0x1b (UTCL2)
Sep 04 16:44:08 amdpc kernel: amdgpu 0000:6e:00.0: amdgpu: GCVM_L2_PROTECTION_FAULT_STATUS:0x00701430
Sep 04 16:44:08 amdpc kernel: amdgpu 0000:6e:00.0: amdgpu: Faulty UTCL2 client ID: SQC (data) (0xa)
Sep 04 16:44:08 amdpc kernel: amdgpu 0000:6e:00.0: amdgpu: MORE_FAULTS: 0x0
Sep 04 16:44:08 amdpc kernel: amdgpu 0000:6e:00.0: amdgpu: WALKER_ERROR: 0x0
Sep 04 16:44:08 amdpc kernel: amdgpu 0000:6e:00.0: amdgpu: PERMISSION_FAULTS: 0x3
Sep 04 16:44:08 amdpc kernel: amdgpu 0000:6e:00.0: amdgpu: MAPPING_ERROR: 0x0
Sep 04 16:44:08 amdpc kernel: amdgpu 0000:6e:00.0: amdgpu: RW: 0x0
Sep 04 16:44:19 amdpc kernel: amdgpu 0000:6e:00.0: amdgpu: Dumping IP State
Sep 04 16:44:19 amdpc kernel: amdgpu 0000:6e:00.0: amdgpu: Dumping IP State Completed
Sep 04 16:44:19 amdpc kernel: amdgpu 0000:6e:00.0: amdgpu: ring gfx_0.0.0 timeout, but soft recovered
Sep 04 16:44:19 amdpc kernel: amdgpu 0000:6e:00.0: amdgpu: Dumping IP State
Sep 04 16:44:19 amdpc kernel: amdgpu 0000:6e:00.0: amdgpu: Dumping IP State Completed
Sep 04 16:44:19 amdpc kernel: amdgpu 0000:6e:00.0: amdgpu: ring gfx_0.1.0 timeout, but soft recovered