My server will no longer reboot with kernels from about 5.18.5 or so on. The reboot stops after everything is shut down and reboot is called, but the next line on the error output on the screen says something like:
kvm: exiting virtual machine
At this point the box is hardlocked and I have to use the hardware reset button to finish the reboot (not good on remote admin'ed machines) I have two boxes that have developed this issue and both have older AMD video cards, so I wonder if this failure to reboot is related to:
Could the amdgpu null-pointer bug on boot be what is causing the reboot to hang? How would you test/verify?