linux-dev/drivers/gpu/drm/mediatek, branch linus/master

Merge tag 'mediatek-drm-next-5.19' of https://git.kernel.org/pub/scm/linux/kernel/git/chunkuang.hu/linux into drm-next

2022-05-06T07:26:01Z

Mediatek DRM Next for Linux 5.19 1. Add display support for MT8186 2. Miscellaneous refinement and fixup. Signed-off-by: Dave Airlie From: Chun-Kuang Hu Link: https://patchwork.freedesktop.org/patch/msgid/1651702965-23630-1-git-send-email-chunkuang.hu@kernel.org

drm/mediatek: mtk_dsi: Remove unnecessary print function dev_err()

2022-05-04T21:36:49Z

The print function dev_err() is redundant because platform_get_irq() already prints an error. Eliminate the follow coccicheck warning: ./drivers/gpu/drm/mediatek/mtk_dsi.c:1092:2-9: line 1092 is redundant because platform_get_irq() already prints an error Link: https://patchwork.kernel.org/project/linux-mediatek/patch/20220224012318.84935-1-yang.lee@linux.alibaba.com/ Reported-by: Abaci Robot Signed-off-by: Yang Li Signed-off-by: Chun-Kuang Hu

drm/mediatek: Add lut diff flag for new gamma hardware support

2022-04-28T22:14:49Z

mt8183 gamma module usage is different with before soc, gamma odd(index start from 0) lut value set to hardware register should be the difference of current lut value with last lut value. for example, chrome os user space set lut like this(only r chanel for example): 2 4 6 8 10 12. 1) mt8183 gamma driver should set the gamma lut to hardware register like this: 2 [2] 6 [2] 10 [2] the value with [] is the difference value 2)gamma hardware process display data with original lut Link: https://patchwork.kernel.org/project/linux-mediatek/patch/20220428085829.15855-2-yongqiang.niu@mediatek.com/ Signed-off-by: Yongqiang Niu Signed-off-by: Yongqiang Niu Reviewed-by: AngeloGioacchino Del Regno Signed-off-by: Chun-Kuang Hu

drm/mediatek: dpi: Use mt8183 output formats for mt8192

2022-04-21T22:20:56Z

The configuration for mt8192 was incorrectly using the output formats from mt8173. Since the output formats for mt8192 are instead the same ones as for mt8183, which require two bus samples per pixel, the pixelclock and DDR edge setting were misconfigured. This made external displays unable to show the image. Fix the issue by correcting the output format for mt8192 to be the same as for mt8183, fixing the usage of external displays for mt8192. Link: https://patchwork.kernel.org/project/linux-mediatek/patch/20220408013950.674477-1-nfraprado@collabora.com/ Fixes: be63f6e8601f ("drm/mediatek: dpi: Add output bus formats to driver data") Signed-off-by: Nícolas F. R. A. Prado Reviewed-by: AngeloGioacchino Del Regno Reviewed-by: Rex-BC Chen Signed-off-by: Chun-Kuang Hu

drm/mediatek: Add display support for MT8186

2022-04-21T22:13:44Z

Add mmsys driver data and compatible for MT8186 in mtk_drm_drv.c. Link: https://patchwork.kernel.org/project/linux-mediatek/patch/20220421051218.8652-2-rex-bc.chen@mediatek.com/ Signed-off-by: Yongqiang Niu Signed-off-by: Rex-BC Chen Reviewed-by: AngeloGioacchino Del Regno Signed-off-by: Chun-Kuang Hu

drm/mediatek: Fix mtk_cec_mask()

2022-04-06T14:34:28Z

In current implementation, mtk_cec_mask() writes val into target register and ignores the mask. After talking to our hdmi experts, mtk_cec_mask() should read a register, clean only mask bits, and update (val | mask) bits to the register. Link: https://patchwork.kernel.org/project/linux-mediatek/patch/20220315232301.2434-1-miles.chen@mediatek.com/ Fixes: 8f83f26891e1 ("drm/mediatek: Add HDMI support") Signed-off-by: Miles Chen Reviewed-by: AngeloGioacchino Del Regno Reviewed-by: Matthias Brugger Cc: Zhiqiang Lin Cc: CK Hu Cc: Matthias Brugger Cc: AngeloGioacchino Del Regno Signed-off-by: Chun-Kuang Hu

drm/mediatek: Fix DPI component detection for MT8192

2022-04-05T23:28:36Z

When support for MT8192 was added, the DPI device was not added to the list of components to look for. This causes the secondary display pipeline to not be able to fully bind, and the DRM driver subsequently defers probing. Add the DPI device compatible to list of DPI components to fix this. Link: https://patchwork.kernel.org/project/linux-mediatek/patch/20220225032754.140168-1-wenst@chromium.org/ Fixes: 01365f549c88 ("drm/mediatek: Add support for Mediatek SoC MT8192") Signed-off-by: Chen-Yu Tsai Signed-off-by: Chun-Kuang Hu

drm/mediatek: Add vblank register/unregister callback functions

2022-04-05T23:10:26Z

We encountered a kernel panic issue that callback data will be NULL when it's using in ovl irq handler. There is a timing issue between mtk_disp_ovl_irq_handler() and mtk_ovl_disable_vblank(). To resolve this issue, we use the flow to register/unregister vblank cb: - Register callback function and callback data when crtc creates. - Unregister callback function and callback data when crtc destroies. With this solution, we can assure callback data will not be NULL when vblank is disable. Link: https://patchwork.kernel.org/project/linux-mediatek/patch/20220321072320.15019-1-rex-bc.chen@mediatek.com/ Fixes: 9b0704988b15 ("drm/mediatek: Register vblank callback function") Signed-off-by: Rex-BC Chen Reviewed-by: jason-jh.lin Signed-off-by: Chun-Kuang Hu

drm/mediatek: mtk_drm_crtc: Use kmalloc in mtk_drm_crtc_duplicate_state

2022-04-05T22:27:23Z

Optimize mtk_drm_crtc_duplicate_state() by switching from kzalloc() to kmalloc(): the only variable of this struct that gets checked in other functions is `pending_config`, but if that's set to false, then all of the remaining variables will only ever be set, but not read - so, also set `pending_config` to false. This saves us some small overhead. Link: https://patchwork.kernel.org/project/linux-mediatek/patch/20220118133759.112458-2-angelogioacchino.delregno@collabora.com/ Signed-off-by: AngeloGioacchino Del Regno Signed-off-by: Chun-Kuang Hu

drm/mediatek: mtk_drm_plane: Use kmalloc in mtk_plane_duplicate_state

2022-04-05T22:27:23Z

There is no need to zero out the newly allocated memory because we are duplicating all members of struct mtk_plane_state: switch to kmalloc to save some overhead. Link: https://patchwork.kernel.org/project/linux-mediatek/patch/20220118133759.112458-1-angelogioacchino.delregno@collabora.com/ Signed-off-by: AngeloGioacchino Del Regno Signed-off-by: Chun-Kuang Hu