Ever since the following upgrades:
python3-xen x86_64 2001:4.14.5-2.fc32 qubes-dom0-cached 60 k
xen x86_64 2001:4.14.5-2.fc32 qubes-dom0-cached 20 k
xen-hypervisor x86_64 2001:4.14.5-2.fc32 qubes-dom0-cached 7.8 M
xen-libs x86_64 2001:4.14.5-2.fc32 qubes-dom0-cached 625 k
xen-licenses x86_64 2001:4.14.5-2.fc32 qubes-dom0-cached 32 k
xen-runtime x86_64 2001:4.14.5-2.fc32 qubes-dom0-cached 19 M
My AMDGPU keeps crashing every 15 min with errors like:
dom0 kernel: radeon 0000:42:00.0: ring 4 stalled for more than 10113msec
dom0 kernel: radeon 0000:42:00.0: GPU lockup (current fence id 0x00000000001c4209 last fence id 0x00000000001c4219 on ring 4)
dom0 kernel: radeon 0000:42:00.0: Saved 1164 dwords of commands on ring 0.
dom0 kernel: radeon 0000:42:00.0: GPU softreset: 0x00000049
Swapped out the GPU with a spare in case it had gone bad, but didn't help. Currently running Linux dom0 5.10.112-1.fc32.qubes.x86_64. Tried booting with an earlier kernel, but no change. Any suggestions? Haven't tried downgrading Xen yet.