Consistent crash on Linux AMD GPU when loading the game on a Tuesday evening

[damien31]

I run a fairly standard Arch setup with an AMD GPU (RADV drivers). The game does not typically crash at all, however for the last two weeks, when attempting to load the visuals for getting a new rank which happens on a Tuesday evening, not only will the game crash, but it will crash my GPU. I use Vulkan rendering, but I can try changing it to DirectX and see if the issue occurs again next week. The following is from my system's logs at the time of the crash:

Nov 12 20:43:31 Damien-Laptop kernel: amdgpu 0000:07:00.0: amdgpu: [gfxhub] page fault (src_id:0 ring:24 vmid:6 pasid:32785, for process MainThrd pid 3554784 thread VKRenderThread pid 3554918)
Nov 12 20:43:31 Damien-Laptop kernel: amdgpu 0000:07:00.0: amdgpu:   in page starting at address 0x00005d0000000000 from client 0x1b (UTCL2)
Nov 12 20:43:31 Damien-Laptop kernel: amdgpu 0000:07:00.0: amdgpu: GCVM_L2_PROTECTION_FAULT_STATUS:0x00601431
Nov 12 20:43:31 Damien-Laptop kernel: amdgpu 0000:07:00.0: amdgpu:          Faulty UTCL2 client ID: SQC (data) (0xa)
Nov 12 20:43:31 Damien-Laptop kernel: amdgpu 0000:07:00.0: amdgpu:          MORE_FAULTS: 0x1
Nov 12 20:43:31 Damien-Laptop kernel: amdgpu 0000:07:00.0: amdgpu:          WALKER_ERROR: 0x0
Nov 12 20:43:31 Damien-Laptop kernel: amdgpu 0000:07:00.0: amdgpu:          PERMISSION_FAULTS: 0x3
Nov 12 20:43:31 Damien-Laptop kernel: amdgpu 0000:07:00.0: amdgpu:          MAPPING_ERROR: 0x0
Nov 12 20:43:31 Damien-Laptop kernel: amdgpu 0000:07:00.0: amdgpu:          RW: 0x0
Nov 12 20:43:31 Damien-Laptop kernel: amdgpu 0000:07:00.0: amdgpu: [gfxhub] page fault (src_id:0 ring:24 vmid:6 pasid:32785, for process MainThrd pid 3554784 thread VKRenderThread pid 3554918)
Nov 12 20:43:31 Damien-Laptop kernel: amdgpu 0000:07:00.0: amdgpu:   in page starting at address 0x00005d0000000000 from client 0x1b (UTCL2)
Nov 12 20:43:31 Damien-Laptop kernel: amdgpu 0000:07:00.0: amdgpu: GCVM_L2_PROTECTION_FAULT_STATUS:0x00000000
Nov 12 20:43:31 Damien-Laptop kernel: amdgpu 0000:07:00.0: amdgpu:          Faulty UTCL2 client ID: CB/DB (0x0)
Nov 12 20:43:31 Damien-Laptop kernel: amdgpu 0000:07:00.0: amdgpu:          MORE_FAULTS: 0x0
Nov 12 20:43:31 Damien-Laptop kernel: amdgpu 0000:07:00.0: amdgpu:          WALKER_ERROR: 0x0
Nov 12 20:43:31 Damien-Laptop kernel: amdgpu 0000:07:00.0: amdgpu:          PERMISSION_FAULTS: 0x0
Nov 12 20:43:31 Damien-Laptop kernel: amdgpu 0000:07:00.0: amdgpu:          MAPPING_ERROR: 0x0
Nov 12 20:43:31 Damien-Laptop kernel: amdgpu 0000:07:00.0: amdgpu:          RW: 0x0

Given that this has not happened with any other application, I think this is due to some interaction between Deadlock, and the AMD drivers.