AMD To Restore Retry Faults Behavior To Help With Raven Stability

Last week I reported on a possible workaround for helping AMD APUs with stability issues on recent Linux kernels. That behavior will now be the new temporary default in dealing with stability issues on Raven APUs.

AMDGPU DRM lead maintainer Alex Deucher sent out a patch reverting the earlier change in behavior. The retry faults behavior is now enabled as otherwise the disabled behavior "causes stability issues on some raven boards. Revert for now until a proper fix is completed."

The small patch will hopefully be mainlined soon and backported to Linux 5.4 but in the meantime affected users can also boot their system with amdgpu.noretry=0 to achieve the same experience.

Linux 5.4 has been particularly hard hitting for some Raven APU users with hangs. Though it's not all Raven users but appears to be partially dependent upon the motherboard/firmware at play.