Ryzen 9 7900 GPU crashing on desktop

Ubuntu Version:

Description: Ubuntu 24.04.2 LTS
Release: 24.04
Codename: noble

Problem Description:
Hi,
I have a new PC and I’m struggling with the desktop. Few seconds (sometimes a full minute) after login, the screen freezes, gets blank and I find myself back at login screen. Sometimes the grey screen with the dead PC appears and asks me to log off.

I’ve tried different distributions. Always the same, the desktop doesn’t work. If i connect to console only, everything is fine.
if I deactivate the internal GPU and use a PCIe video card, it works as well. So it’s the GPU in the Ryzen CPU that is not working with Linux.
Is it too new?
I have dmesg info that I colelcted via SSH over my old Ubuntu PC. There you can see, that the AMD GPU is perofming some sort of reset, then follows the error.

Relevant System Information:
Board: ASUS TUF Gaming B650E-Wifi
CPU: Ryzen 9 7900
RAM: Kingston FURY Beast (2 x 32GB, 5600 MHz, DDR5-RAM, DIMM)
NVME SSD Crucial T700 w heatsink (2000 GB, M.2 2280)

uname -r
6.11.0-19-generic

lshw -c display

 *-display
       Beschreibung: VGA compatible controller
       Produkt: Raphael
       Hersteller: Advanced Micro Devices, Inc. [AMD/ATI]
       Physische ID: 0
       Bus-Informationen: pci@0000:0c:00.0
       Logischer Name: /dev/fb0
       Version: c4
       Breite: 64 bits
       Takt: 33MHz
       Fähigkeiten: pm pciexpress msi msix vga_controller bus_master cap_list fb
       Konfiguration: depth=32 driver=amdgpu latency=0 resolution=1920,1200
       Ressourcen: iomemory:fc0-fbf iomemory:fc0-fbf irq:84 memory:fce0000000-fcefffffff memory:fcf0000000-fcf01fffff ioport:e000(Größe=256) memory:f6a00000-f6a7ffff

Screenshots or Error Messages:

sudo dmesg -H | grep amdgpu


[  +0,001754] [drm] amdgpu kernel modesetting enabled.
[  +0,000024] amdgpu: vga_switcheroo: detected switching method \_SB_.PCI0.GP17.VGA_.ATPX handle
[  +0,000235] amdgpu: ATPX version 1, functions 0x00000000
[  +0,006839] amdgpu: Virtual CRAT table created for CPU
[  +0,000020] amdgpu: Topology: Add CPU node
[  +0,000123] amdgpu 0000:0c:00.0: enabling device (0006 -> 0007)
[  +0,000008] amdgpu 0000:0c:00.0: amdgpu: Fetched VBIOS from VFCT
[  +0,000003] amdgpu: ATOM BIOS: 102-RAPHAEL-008
[  +0,025247] amdgpu 0000:0c:00.0: vgaarb: deactivate vga console
[  +0,000004] amdgpu 0000:0c:00.0: amdgpu: Trusted Memory Zone (TMZ) feature disabled as experimental (default)
[  +0,000008] amdgpu 0000:0c:00.0: amdgpu: VRAM: 512M 0x000000F400000000 - 0x000000F41FFFFFFF (512M used)
[  +0,000002] amdgpu 0000:0c:00.0: amdgpu: GART: 1024M 0x0000000000000000 - 0x000000003FFFFFFF
[  +0,000089] [drm] amdgpu: 512M of VRAM memory ready
[  +0,000003] [drm] amdgpu: 31725M of GTT memory ready.
[  +0,001617] amdgpu 0000:0c:00.0: amdgpu: reserve 0xa00000 from 0xf41e000000 for PSP TMR
[  +0,041976] amdgpu 0000:0c:00.0: amdgpu: RAS: optional ras ta ucode is not available
[  +0,005911] amdgpu 0000:0c:00.0: amdgpu: RAP: optional rap ta ucode is not available
[  +0,000003] amdgpu 0000:0c:00.0: amdgpu: SECUREDISPLAY: securedisplay ta ucode is not available
[  +0,001765] amdgpu 0000:0c:00.0: amdgpu: SMU is initialized successfully!
[  +0,000481] snd_hda_intel 0000:0c:00.1: bound 0000:0c:00.0 (ops amdgpu_dm_audio_component_bind_ops [amdgpu])
[  +0,003241] kfd kfd: amdgpu: Allocated 3969056 bytes on gart
[  +0,000013] kfd kfd: amdgpu: Total number of KFD nodes to be created: 1
[  +0,000133] amdgpu: Virtual CRAT table created for GPU
[  +0,000624] amdgpu: Topology: Add dGPU node [0x164e:0x1002]
[  +0,000002] kfd kfd: amdgpu: added device 1002:164e
[  +0,000010] amdgpu 0000:0c:00.0: amdgpu: SE 1, SH per SE 1, CU per SH 2, active_cu_number 2
[  +0,000004] amdgpu 0000:0c:00.0: amdgpu: ring gfx_0.0.0 uses VM inv eng 0 on hub 0
[  +0,000001] amdgpu 0000:0c:00.0: amdgpu: ring gfx_0.1.0 uses VM inv eng 1 on hub 0
[  +0,000002] amdgpu 0000:0c:00.0: amdgpu: ring comp_1.0.0 uses VM inv eng 4 on hub 0
[  +0,000001] amdgpu 0000:0c:00.0: amdgpu: ring comp_1.1.0 uses VM inv eng 5 on hub 0
[  +0,000001] amdgpu 0000:0c:00.0: amdgpu: ring comp_1.2.0 uses VM inv eng 6 on hub 0
[  +0,000001] amdgpu 0000:0c:00.0: amdgpu: ring comp_1.3.0 uses VM inv eng 7 on hub 0
[  +0,000002] amdgpu 0000:0c:00.0: amdgpu: ring comp_1.0.1 uses VM inv eng 8 on hub 0
[  +0,000001] amdgpu 0000:0c:00.0: amdgpu: ring comp_1.1.1 uses VM inv eng 9 on hub 0
[  +0,000001] amdgpu 0000:0c:00.0: amdgpu: ring comp_1.2.1 uses VM inv eng 10 on hub 0
[  +0,000001] amdgpu 0000:0c:00.0: amdgpu: ring comp_1.3.1 uses VM inv eng 11 on hub 0
[  +0,000001] amdgpu 0000:0c:00.0: amdgpu: ring kiq_0.2.1.0 uses VM inv eng 12 on hub 0
[  +0,000001] amdgpu 0000:0c:00.0: amdgpu: ring sdma0 uses VM inv eng 13 on hub 0
[  +0,000002] amdgpu 0000:0c:00.0: amdgpu: ring vcn_dec_0 uses VM inv eng 0 on hub 8
[  +0,000001] amdgpu 0000:0c:00.0: amdgpu: ring vcn_enc_0.0 uses VM inv eng 1 on hub 8
[  +0,000001] amdgpu 0000:0c:00.0: amdgpu: ring vcn_enc_0.1 uses VM inv eng 4 on hub 8
[  +0,000001] amdgpu 0000:0c:00.0: amdgpu: ring jpeg_dec uses VM inv eng 5 on hub 8
[  +0,000291] amdgpu 0000:0c:00.0: amdgpu: Runtime PM not available
[  +0,000349] [drm] Initialized amdgpu 3.59.0 for 0000:0c:00.0 on minor 1
[  +0,005990] fbcon: amdgpudrmfb (fb0) is primary device
[  +0,000002] amdgpu 0000:0c:00.0: [drm] fb0: amdgpudrmfb frame buffer device
[  +0,218773] amdgpu 0000:0c:00.0: [drm] REG_WAIT timeout 1us * 100000 tries - optc31_disable_crtc line:145
[Mär21 18:32] amdgpu 0000:0c:00.0: amdgpu: ring gfx_0.0.0 timeout, signaled seq=1168, emitted seq=1170
[  +0,000008] amdgpu 0000:0c:00.0: amdgpu: Process information: process gnome-shell pid 3032 thread gnome-shel:cs0 pid 3060
[  +0,000003] amdgpu 0000:0c:00.0: amdgpu: GPU reset begin!
[  +0,071340] amdgpu 0000:0c:00.0: amdgpu: Dumping IP State
[  +0,000538] amdgpu 0000:0c:00.0: amdgpu: Dumping IP State Completed
[  +0,000005] amdgpu 0000:0c:00.0: amdgpu: MODE2 reset
[  +0,007644] amdgpu 0000:0c:00.0: amdgpu: GPU reset succeeded, trying to resume
[  +0,000066] amdgpu 0000:0c:00.0: amdgpu: PSP is resuming...
[  +0,021581] amdgpu 0000:0c:00.0: amdgpu: reserve 0xa00000 from 0xf41e000000 for PSP TMR
[  +0,200891] amdgpu 0000:0c:00.0: amdgpu: RAS: optional ras ta ucode is not available
[  +0,005611] amdgpu 0000:0c:00.0: amdgpu: RAP: optional rap ta ucode is not available
[  +0,000002] amdgpu 0000:0c:00.0: amdgpu: SECUREDISPLAY: securedisplay ta ucode is not available
[  +0,000002] amdgpu 0000:0c:00.0: amdgpu: SMU is resuming...
[  +0,000622] amdgpu 0000:0c:00.0: amdgpu: SMU is resumed successfully!
[  +0,002668] amdgpu 0000:0c:00.0: amdgpu: ring gfx_0.0.0 uses VM inv eng 0 on hub 0
[  +0,000003] amdgpu 0000:0c:00.0: amdgpu: ring gfx_0.1.0 uses VM inv eng 1 on hub 0
[  +0,000002] amdgpu 0000:0c:00.0: amdgpu: ring comp_1.0.0 uses VM inv eng 4 on hub 0
[  +0,000001] amdgpu 0000:0c:00.0: amdgpu: ring comp_1.1.0 uses VM inv eng 5 on hub 0
[  +0,000002] amdgpu 0000:0c:00.0: amdgpu: ring comp_1.2.0 uses VM inv eng 6 on hub 0
[  +0,000001] amdgpu 0000:0c:00.0: amdgpu: ring comp_1.3.0 uses VM inv eng 7 on hub 0
[  +0,000001] amdgpu 0000:0c:00.0: amdgpu: ring comp_1.0.1 uses VM inv eng 8 on hub 0
[  +0,000001] amdgpu 0000:0c:00.0: amdgpu: ring comp_1.1.1 uses VM inv eng 9 on hub 0
[  +0,000002] amdgpu 0000:0c:00.0: amdgpu: ring comp_1.2.1 uses VM inv eng 10 on hub 0
[  +0,000001] amdgpu 0000:0c:00.0: amdgpu: ring comp_1.3.1 uses VM inv eng 11 on hub 0
[  +0,000002] amdgpu 0000:0c:00.0: amdgpu: ring kiq_0.2.1.0 uses VM inv eng 12 on hub 0
[  +0,000001] amdgpu 0000:0c:00.0: amdgpu: ring sdma0 uses VM inv eng 13 on hub 0
[  +0,000002] amdgpu 0000:0c:00.0: amdgpu: ring vcn_dec_0 uses VM inv eng 0 on hub 8
[  +0,000001] amdgpu 0000:0c:00.0: amdgpu: ring vcn_enc_0.0 uses VM inv eng 1 on hub 8
[  +0,000001] amdgpu 0000:0c:00.0: amdgpu: ring vcn_enc_0.1 uses VM inv eng 4 on hub 8
[  +0,000002] amdgpu 0000:0c:00.0: amdgpu: ring jpeg_dec uses VM inv eng 5 on hub 8
[  +0,002526] amdgpu 0000:0c:00.0: amdgpu: recover vram bo from shadow start
[  +0,000002] amdgpu 0000:0c:00.0: amdgpu: recover vram bo from shadow done
[  +0,000015] amdgpu 0000:0c:00.0: amdgpu: GPU reset(2) succeeded!
[  +1,827686] [drm:amdgpu_cs_ioctl [amdgpu]] *ERROR* Failed to initialize parser -125!

What I’ve Tried:
Tried newer kernel, older kernel, other distributions (and desktops).Searched for a solution with linux-firmare-git but did not know where to get that or how to install.

Hope someone can give me a hint…

Greetings
Alex

Grab the daily Kubuntu 25.04 live image and give that a try.

Daily Kubuntu has same problems. Desktop goes black and returns, infobox appears saying desktop effects were restartet because of GPU reset

dmesg shows this:

amdgpu: ring gfx_0.1.0 timeout, signaled seq=7422, emitted seq= 7424
amdgpu: Process inrofmation: process kwin_wayland pid 5541 thread kwin_wayla:cs0 pid 5454
amdgpu: Starting gfx_0.1.0 ring reset
amdgpu: Ring gfx_0.1.0 reset failure
amdgpu: GPU reset begin!
amdgpu: MODE2 set
amdgpu: GPU reset succeeded, trying to resum

Tried ubuntu plucky puffin, that one goes directly to the “Oops” screen when desktop is loading.

The daily Kubuntu is already using the latest 6.14 kernel with Mesa 25 drivers, so if that didn’t work the card itself may be to fault. Check bios settings, I have heard that secure boot must be enabled for the newer cards to work properly.

I don’t have a physical card, the CPU has GPU function.
Already tried with and without secure boot, fast boot etc.