OSDN Git Service

drm/amd/display: Support reading hw state from debugfs file
authorNicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Wed, 15 Aug 2018 16:00:23 +0000 (12:00 -0400)
committerAlex Deucher <alexander.deucher@amd.com>
Mon, 27 Aug 2018 20:20:49 +0000 (15:20 -0500)
commit46659a83e4662ed92000ec13445b8c0ca96fd2cc
tree71f5e2fd083a41b56c86e29351d2a89ce55dcfad
parent1f6010a96273c3111ecdc12aa274c932da920493
drm/amd/display: Support reading hw state from debugfs file

[Why]

Logging hardware state can be done by triggering a write to the
debugfs file. It would also be useful to be able to read the hardware
state from the debugfs file to be able to generate a clean log without
timestamps.

[How]

Usage: cat /sys/kernel/debug/dri/0/amdgpu_dm_dtn_log

Threading is an obvious concern when dealing with multiple debugfs
operations and blocking on global state in dm or dc seems unfavorable.

Adding an extra parameter for the debugfs log context state is the
implementation done here. Existing code that made use of DTN_INFO
and its associated macros needed to be refactored to support this.

We don't know the size of the log in advance so it reallocates the
log string dynamically. Once the log has been generated it's copied
into the user supplied buffer for the debugfs. This allows for seeking
support but it's worth nothing that unlike triggering output via
dmesg the hardware state might change in-between reads if your buffer
size is too small.

Signed-off-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Reviewed-by: Jordan Lazare <Jordan.Lazare@amd.com>
Acked-by: Leo Li <sunpeng.li@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c
drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c
drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c
drivers/gpu/drm/amd/display/dc/dm_services.h
drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h
drivers/gpu/drm/amd/display/include/logger_interface.h
drivers/gpu/drm/amd/display/include/logger_types.h