Any way to get AMD GPU to work on Fedora?

I recently installed Fedora 43 on my A16 with:

c3:00.0 VGA compatible controller: NVIDIA Corporation AD106M [GeForce RTX 4070 Max-Q / Mobile] (rev a1)
	Subsystem: Micro-Star International Co., Ltd. [MSI] Device 1462
	Kernel driver in use: nvidia
	Kernel modules: nouveau, nvidia_drm, nvidia
--
c5:00.0 Display controller: Advanced Micro Devices, Inc. [AMD/ATI] Strix [Radeon 880M / 890M] (rev c4)
	Subsystem: Micro-Star International Co., Ltd. [MSI] Device 1462
	Kernel driver in use: amdgpu
	Kernel modules: amdgpu

$ uname -a
Linux fedora 6.17.6-300.fc43.x86_64 #1 SMP PREEMPT_DYNAMIC Wed Oct 29 20:10:51 UTC 2025 x86_64 GNU/Linux

$ sudo cat /sys/kernel/debug/dri/1/amdgpu_firmware_info | grep -A 2 MES
MES_KIQ feature version: 6, firmware version: 0x0000006f
MES feature version: 1, firmware version: 0x00000080
VPE feature version: 60, firmware version: 0x00000039
VBIOS version: 113-STRIXEMU-001

Have been getting AMD GPU crashes (see journalctl logs below), especially more frequently after enabling the NVIDIA GPU.

I saw here: https://github.com/ROCm/ROCm/issues/5151

a recommendation to upgrade to the MES_KIQ firmware version 0x80 using the amdgpu-dkms-firmware package. But, that doesn’t seem to be supported forfedora, so I wondered if anyone has any advice what to do.

Various symptoms include:

  1. sporadic full or half gray screens
  2. delays of keypress echoing to the screen
  3. spontaneous reboots

These symptoms don’t produce any telltale logs to speak of, but every so often the system freezes and somehow recovers, and I do see this in the logs:

Oct 25 13:33:40 fedora kernel: amdgpu 0000:c5:00.0: amdgpu: [gfxhub] page fault (src_id:0 ring:24 vmid:7 pasid:32773)
Oct 25 13:33:40 fedora kernel: amdgpu 0000:c5:00.0: amdgpu:  in process chrome pid 3842 thread chrome:cs0 pid 4100)
Oct 25 13:33:40 fedora kernel: amdgpu 0000:c5:00.0: amdgpu:   in page starting at address 0x00001a943f7a0000 from client 10
Oct 25 13:33:40 fedora kernel: amdgpu 0000:c5:00.0: amdgpu: GCVM_L2_PROTECTION_FAULT_STATUS:0x00701430
Oct 25 13:33:40 fedora kernel: amdgpu 0000:c5:00.0: amdgpu:          Faulty UTCL2 client ID: SQC (data) (0xa)
Oct 25 13:33:40 fedora kernel: amdgpu 0000:c5:00.0: amdgpu:          MORE_FAULTS: 0x0
Oct 25 13:33:40 fedora kernel: amdgpu 0000:c5:00.0: amdgpu:          WALKER_ERROR: 0x0
Oct 25 13:33:40 fedora kernel: amdgpu 0000:c5:00.0: amdgpu:          PERMISSION_FAULTS: 0x3
Oct 25 13:33:40 fedora kernel: amdgpu 0000:c5:00.0: amdgpu:          MAPPING_ERROR: 0x0
Oct 25 13:33:40 fedora kernel: amdgpu 0000:c5:00.0: amdgpu:          RW: 0x0
Oct 25 13:33:50 fedora kernel: amdgpu 0000:c5:00.0: amdgpu: Dumping IP State
Oct 25 13:33:50 fedora kernel: amdgpu 0000:c5:00.0: amdgpu: Dumping IP State Completed
Oct 25 13:33:50 fedora kernel: amdgpu 0000:c5:00.0: amdgpu: [drm] AMDGPU device coredump file has been created
Oct 25 13:33:50 fedora kernel: amdgpu 0000:c5:00.0: amdgpu: [drm] Check your /sys/class/drm/card1/device/devcoredump/data
Oct 25 13:33:50 fedora kernel: amdgpu 0000:c5:00.0: amdgpu: ring gfx_0.0.0 timeout, but soft recovered

Any advice would be greatly appreciated!