kern/57477: radeon can hang after significant uptime, where the screen goes blank and no longer refreshes

[david%gutteridge.ca@localhost]

>Number:         57477
>Category:       kern
>Synopsis:       radeon can hang after significant uptime, where the screen goes blank and no longer refreshes
>Confidential:   no
>Severity:       serious
>Priority:       medium
>Responsible:    kern-bug-people
>State:          open
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Tue Jun 20 22:35:00 +0000 2023
>Originator:     David H. Gutteridge
>Release:        9.3_STABLE
>Organization:
TNF
>Environment:
NetBSD arcusxvii.nonus-porta.net 9.3_STABLE NetBSD 9.3_STABLE (GENERIC) #0: Sun May 28 10:23:12 UTC 2023  mkrepro%mkrepro.NetBSD.org@localhost:/usr/src/sys/arch/amd64/compile/GENERIC amd64
>Description:
I've found a laptop using radeon DRM will occasionally get into a state
where the screen goes blank and no longer refreshes. This happens after
an uptime of at least several days. When this occurs, the kernel
reports:

Jun  3 20:47:20 arcusxvii /netbsd: [ 890296.4391173] kern error: [drm:(/usr/src/sys/external/bsd/drm2/dist/drm/radeon/radeon_atom.c:748)atom_op_jump] *ERROR* atombios stuck in loop for more than 5secs aborting
Jun  3 20:47:20 arcusxvii /netbsd: [ 890296.4391173] kern error: [drm:(/usr/src/sys/external/bsd/drm2/dist/drm/radeon/radeon_atom.c:1211)atom_execute_table_locked] *ERROR* atombios stuck executing BE82 (len 114, WS 0, PS 4) @ 0xBEEB
Jun  3 20:47:20 arcusxvii /netbsd: [ 890296.4391173] kern error: [drm:(/usr/src/sys/external/bsd/drm2/dist/drm/radeon/radeon_atom.c:1211)atom_execute_table_locked] *ERROR* atombios stuck executing E802 (len 2592, WS 0, PS 8) @ 0xEDC6
Jun  3 20:47:26 arcusxvii /netbsd: [ 890302.0715951] kern error: [drm:(/usr/src/sys/external/bsd/drm2/dist/drm/radeon/radeon_atom.c:748)atom_op_jump] *ERROR* atombios stuck in loop for more than 5secs aborting
Jun  3 20:47:26 arcusxvii /netbsd: [ 890302.0715951] kern error: [drm:(/usr/src/sys/external/bsd/drm2/dist/drm/radeon/radeon_atom.c:1211)atom_execute_table_locked] *ERROR* atombios stuck executing BE82 (len 114, WS 0, PS 4) @ 0xBEEB
Jun  3 20:47:26 arcusxvii /netbsd: [ 890302.0715951] kern error: [drm:(/usr/src/sys/external/bsd/drm2/dist/drm/radeon/radeon_atom.c:1211)atom_execute_table_locked] *ERROR* atombios stuck executing E802 (len 2592, WS 0, PS 8) @ 0xE877
Jun  3 20:47:26 arcusxvii /netbsd: [ 890302.0715951] kern error: [drm:(/usr/src/sys/external/bsd/drm2/dist/drm/radeon/radeon_atombios_dp.c:712)radeon_dp_link_train_cr] *ERROR* clock recovery reached max voltage
Jun  3 20:47:26 arcusxvii /netbsd: [ 890302.0715951] kern error: [drm:(/usr/src/sys/external/bsd/drm2/dist/drm/radeon/radeon_atombios_dp.c:733)radeon_dp_link_train_cr] *ERROR* clock recovery failed
Jun  3 20:47:31 arcusxvii /netbsd: [ 890307.0938042] kern error: [drm:(/usr/src/sys/external/bsd/drm2/dist/drm/radeon/radeon_atom.c:748)atom_op_jump] *ERROR* atombios stuck in loop for more than 5secs aborting
Jun  3 20:47:31 arcusxvii /netbsd: [ 890307.0938042] kern error: [drm:(/usr/src/sys/external/bsd/drm2/dist/drm/radeon/radeon_atom.c:1211)atom_execute_table_locked] *ERROR* atombios stuck executing BE82 (len 114, WS 0, PS 4) @ 0xBEEB
Jun  3 20:47:31 arcusxvii /netbsd: [ 890307.0938042] kern error: [drm:(/usr/src/sys/external/bsd/drm2/dist/drm/radeon/radeon_atom.c:1211)atom_execute_table_locked] *ERROR* atombios stuck executing E802 (len 2592, WS 0, PS 8) @ 0xE877
Jun  3 20:47:31 arcusxvii /netbsd: [ 890307.0938042] kern error: [drm:(/usr/src/sys/external/bsd/drm2/dist/drm/radeon/radeon_atombios_dp.c:712)radeon_dp_link_train_cr] *ERROR* clock recovery reached max voltage
Jun  3 20:47:31 arcusxvii /netbsd: [ 890307.0938042] kern error: [drm:(/usr/src/sys/external/bsd/drm2/dist/drm/radeon/radeon_atombios_dp.c:733)radeon_dp_link_train_cr] *ERROR* clock recovery failed

I realize this isn't the most helpful PR, as it's against 9.3_STABLE.
I'm filing it for informational purposes for now, since I haven't found
any similar PR already filed.

>How-To-Repeat:
Run this hardware: KAVERI 0x1002:0x1309 0x103C:0x221C
>Fix:
Unknown.