diff options
| author | Ville Syrjälä <[email protected]> | 2022-10-31 13:57:03 +0000 |
|---|---|---|
| committer | Ville Syrjälä <[email protected]> | 2022-11-02 06:55:56 +0000 |
| commit | 1086af67b9ab5229a8166909df01f289bc9a17bd (patch) | |
| tree | 986891cd52657983d2fbc2c429ba41f483a998c6 /drivers/gpu/drm/amd/amdgpu/amdgpu_sync.c | |
| parent | drm/i915/selftests: Test RING_TIMESTAMP on gen4/5 (diff) | |
| download | kernel-1086af67b9ab5229a8166909df01f289bc9a17bd.tar.gz kernel-1086af67b9ab5229a8166909df01f289bc9a17bd.zip | |
drm/i915/selftests: Run the perf MI_BB tests on gen4/5
Now that we know the ring timestamp frequency on gen4/5 we
can run the perf tests that depend on sampling the timestamp.
On g4x/ilk we must read the udw of the 64bit timestamp
register. Details in {g4x,gen5)_read_clock_frequency().
When executing the read via the CS i965 doesn't seem to need
the double read trick that CPU mmio reads need.
Signed-off-by: Ville Syrjälä <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
Reviewed-by: Lionel Landwerlin <[email protected]>
Diffstat (limited to 'drivers/gpu/drm/amd/amdgpu/amdgpu_sync.c')
0 files changed, 0 insertions, 0 deletions