Changes in v9:
- As per previous conversation with CK Hu, added a commit that
de-commonizes the gamma setting function that was used in
both DISP_AAL and DISP_GAMMA, now each of them have their
own .gamma_set() callback (mtk_disp_gamma_set_common() has
been removed).
- Added a change to use bitfield macros in mtk_disp_aal.c
- Added a change to compress of_device_id entries in mtk_disp_aal.c
- Tested again on MT6795, MT8173, MT8186, MT8192, MT8195
Changes in v8:
- Changed lut_size to be a mtk_disp_gamma_set_common() function
parameter to pass lut size from AAL
Changes in v7:
- Added check for NULL dev for AAL-gamma case
- Added get_lut_size callback for AAL-gamma
- Added comment to clarify SoC 10/12 bits support and old vs new
register layout as suggested by Alexandre M.
Changes in v6:
- Fixed smatch warning in patch 11/11, ref.:
https://lore.kernel.org/all/202306101458.lRXHEE0Z-lkp@intel.com/
Changes in v5:
- Removed incorrect comment on default LUT size and bits
- Removed useless check for num_lut_banks
- Added comment about CMDQ implementation on patch 5
- Evaluated passing lut size/bits from AAL, idea discarded as
the implementation would be rather tricky while bringing no
benefits.
Changes in v4:
- Fixed assignment typo appeared in v3
Changes in v3:
- Fixed issues due to variables renaming during cleanup (oops)
- This is actually the right series, since v2 was taken from the
wrong kernel tree.... :-)
Changes in v2:
- Added explicit inclusion of linux/bitfield.h in patch [06/11]
This series adds support for GAMMA IP requiring and/or supporting
a 12-bits LUT using a slightly different register layout and programming
sequence for multiple LUT banks: this IP version is currently found
on a number of SoCs, not only including the Chromebook/IoT oriented
Kompanio 1200/1380 MT8195/MT8195T, but also Smartphone chips such as
the Dimensity 9200 (MT6985) and others.
This series was tested on MT8195, MT8192, MT8173, MT6795:
* MT6795, MT8192, MT8173: No regression, works fine.
* MT8195: Color correction is finally working!
AngeloGioacchino Del Regno (15):
drm/mediatek: gamma: Reduce indentation in mtk_gamma_set_common()
drm/mediatek: gamma: Support SoC specific LUT size
drm/mediatek: gamma: Improve and simplify HW LUT calculation
drm/mediatek: gamma: Enable the Gamma LUT table only after programming
drm/mediatek: gamma: Use bitfield macros
drm/mediatek: aal: Use bitfield macros
drm/mediatek: De-commonize disp_aal/disp_gamma gamma_set functions
drm/mediatek: gamma: Support specifying number of bits per LUT
component
drm/mediatek: gamma: Support multi-bank gamma LUT
drm/mediatek: gamma: Add support for 12-bit LUT and MT8195
drm/mediatek: gamma: Make sure relay mode is disabled
drm/mediatek: gamma: Program gamma LUT type for descending or rising
drm/mediatek: aal: Rewrite kerneldoc for struct mtk_disp_aal
drm/mediatek: gamma: Add kerneldoc for struct mtk_disp_gamma
drm/mediatek: aal: Compress of_device_id entries and add sentinel
Jason-JH.Lin (1):
drm/mediatek: gamma: Adjust mtk_drm_gamma_set_common parameters
drivers/gpu/drm/mediatek/mtk_disp_aal.c | 84 ++++++--
drivers/gpu/drm/mediatek/mtk_disp_drv.h | 3 +-
drivers/gpu/drm/mediatek/mtk_disp_gamma.c | 203 ++++++++++++++++----
drivers/gpu/drm/mediatek/mtk_drm_crtc.c | 8 +-
drivers/gpu/drm/mediatek/mtk_drm_crtc.h | 1 -
drivers/gpu/drm/mediatek/mtk_drm_ddp_comp.c | 2 +
drivers/gpu/drm/mediatek/mtk_drm_ddp_comp.h | 9 +
7 files changed, 256 insertions(+), 54 deletions(-)
--
2.41.0