Failure in amdgpu_device_ip_init triggers NULL point dereference in amdgpu_pci_probe on kernel 5.8.14

Brief summary of the problem:

Booting with kernel 5.8.14 locks up a HP ProBook 455R G6 containing an AMD Ryzen 5 3500U with kernel logs showing

amdgpu 0000:05:00.0: [drm:amdgpu_ring_test_helper [amdgpu]] *ERROR* ring gfx test failed (-110)
[drm:amdgpu_device_ip_init [amdgpu]] *ERROR* hw_init of IP block <gfx_v9_0> failed -110
amdgpu 0000:05:00.0: amdgpu: amdgpu_device_ip_init failed
amdgpu 0000:05:00.0: amdgpu: Fatal error during GPU init

which later on triggers a NULL pointer dereference in amdgpu_pci_probe.

The system does boot using modprobe.blacklist=amdgpu and is serviceable using nomodeset.

Hardware description:

• CPU: AMD Ryzen 5 3500U
• GPU: Picasso IGD [1002:15d8]
• System Memory: 16 GB
• Display(s): Internal at 1920x1080
• Type of Diplay Connection: eDP

System infomration:

• Distro name and Version: OpenSUSE Tumbleweed
• Kernel version: 5.8.4
• AMD package version: No package

Attached files:

• Kernel log immediately after manually running modprobe amdgpu