drm/amdkfd: Handle deallocated VPGRs in gfx11+ trap handler

Browse source

A wavefront may deallocate its VGPRs at the end of a program while
waiting for memory transactions to complete. If it subsequently
receives a context save exception it will be unable to save,
since this requires VGPRs. In this case the trap handler should
terminate the wavefront.

Fixes intermittent VM faults under context switching load.

V2: Use S_ENDPGM instead of S_ENDPGM_SAVED for performance counters

Signed-off-by: Jay Cornwall <jay.cornwall@amd.com>
Reviewed-by: Lancelot Six <lancelot.six@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

...

This commit is contained in:

Jay Cornwall 2024-05-28 15:55:56 -05:00 • committed by Alex Deucher

parent 4002a6c55e

commit c5afb313e7

2 changed files with 366 additions and 346 deletions

Show all changes Ignore whitespace when comparing lines Ignore changes in amount of whitespace Ignore changes in whitespace at EOL

Download patch file Download diff file Expand all files Collapse all files

735

drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler.h

View file

17

drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler_gfx10.asm

View file