diff options
| author | Daniel Vetter <[email protected]> | 2017-05-09 14:03:25 +0000 |
|---|---|---|
| committer | Daniel Vetter <[email protected]> | 2017-05-10 08:21:08 +0000 |
| commit | d673c02c4bdbcbe3076a2680f9c954be26b525c8 (patch) | |
| tree | 6a3535fa5b309037993de557aaf8360caee0281e /drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c | |
| parent | drm/vgem: Convert to a struct drm_device subclass (diff) | |
| download | kernel-d673c02c4bdbcbe3076a2680f9c954be26b525c8.tar.gz kernel-d673c02c4bdbcbe3076a2680f9c954be26b525c8.zip | |
drm/vblank: Switch drm_driver->get_vblank_timestamp to return a bool
There's really no reason for anything more:
- Calling this while the crtc vblank stuff isn't set up is a driver
bug. Those places alrready DRM_ERROR.
- Calling this when the crtc is off is either a driver bug (calling
drm_crtc_handle_vblank at the wrong time) or a core bug (for
anything else). Again, we DRM_ERROR.
- EINVAL is checked at higher levels already, and if we'd use struct
drm_crtc * instead of (dev, pipe) it would be real obvious that
those are again core bugs.
The only valid failure mode is crap hardware that couldn't sample a
useful timestamp, to ask the core to just grab a not-so-accurate
timestamp. Bool is perfectly fine for that.
v2: Also fix up the one caller, I lost that in the shuffling (Jani).
v3: Fixup commit message (Neil).
Cc: Jani Nikula <[email protected]>
Cc: Mario Kleiner <[email protected]>
Cc: Eric Anholt <[email protected]>
Cc: Rob Clark <[email protected]>
Cc: [email protected]
Cc: [email protected]
Cc: Alex Deucher <[email protected]>
Cc: Christian König <[email protected]>
Cc: Ben Skeggs <[email protected]>
Reviewed-by: Neil Armstrong <[email protected]>
Acked-by: Ville Syrjälä <[email protected]>
Signed-off-by: Daniel Vetter <[email protected]>
Link: http://patchwork.freedesktop.org/patch/msgid/[email protected]
Diffstat (limited to 'drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c')
0 files changed, 0 insertions, 0 deletions
