| Message ID | 20240717-alpha-blending-v4-0-4b1c806c0749@mediatek.com (mailing list archive) |
|---|---|
| Headers | show |
| Series | Support alpha blending in MTK display driver | expand |
Hi, Shawn: Hsiao Chien Sung via B4 Relay <devnull+shawn.sung.mediatek.com@kernel.org> 於 2024年7月17日 週三 下午1:24寫道： > > Support "Pre-multiplied" and "None" blend mode on MediaTek's chips by > adding correct blend mode property when the planes init. > Before this patch, only the "Coverage" mode (default) is supported. For the whole series, applied to mediatek-drm-next [1], thanks. [1] https://git.kernel.org/pub/scm/linux/kernel/git/chunkuang.hu/linux.git/log/?h=mediatek-drm-next Regards, Chun-Kuang. > > Signed-off-by: Hsiao Chien Sung <shawn.sung@mediatek.corp-partner.google.com> > --- > Changes in v4: > - Add more information to the commit message > - Link to v3: https://lore.kernel.org/r/20240710-alpha-blending-v3-0-289c187f9c6f@mediatek.com > > Changes in v3: > - Remove the Change-Id > - Link to v2: https://lore.kernel.org/r/20240710-alpha-blending-v2-0-d4b505e6980a@mediatek.com > > Changes in v2: > - Remove unnecessary codes > - Add more information to the commit message > - Link to v1: https://lore.kernel.org/r/20240620-blend-v1-0-72670072ca20@mediatek.com > > --- > Hsiao Chien Sung (5): > drm/mediatek: Support "None" blending in OVL > drm/mediatek: Support "None" blending in Mixer > drm/mediatek: Support "Pre-multiplied" blending in OVL > drm/mediatek: Support "Pre-multiplied" blending in Mixer > drm/mediatek: Support alpha blending in display driver > > drivers/gpu/drm/mediatek/mtk_disp_ovl.c | 36 +++++++++++++++++++++++++-------- > drivers/gpu/drm/mediatek/mtk_ethdr.c | 13 +++++++++--- > drivers/gpu/drm/mediatek/mtk_plane.c | 11 ++++++++++ > 3 files changed, 49 insertions(+), 11 deletions(-) > --- > base-commit: 8ad49a92cff4bab13eb2f2725243f5f31eff3f3b > change-id: 20240710-alpha-blending-067295570863 > > Best regards, > -- > Hsiao Chien Sung <shawn.sung@mediatek.com> > >
Hi CK, On Wed, 2024-07-31 at 21:34 +0800, Chun-Kuang Hu wrote: > > External email : Please do not click links or open attachments until > you have verified the sender or the content. > Hi, Shawn: > > Hsiao Chien Sung via B4 Relay > <devnull+shawn.sung.mediatek.com@kernel.org> 於 2024年7月17日 週三 > 下午1:24寫道： > > > > Support "Pre-multiplied" and "None" blend mode on MediaTek's chips > by > > adding correct blend mode property when the planes init. > > Before this patch, only the "Coverage" mode (default) is supported. > > For the whole series, applied to mediatek-drm-next [1], thanks. > Thank you for the information. Regards, Shawn > [1] > https://git.kernel.org/pub/scm/linux/kernel/git/chunkuang.hu/linux.git/log/?h=mediatek-drm-next > > Regards, > Chun-Kuang. > > > > > Signed-off-by: Hsiao Chien Sung < > shawn.sung@mediatek.corp-partner.google.com> > > --- > > Changes in v4: > > - Add more information to the commit message > > - Link to v3: > https://lore.kernel.org/r/20240710-alpha-blending-v3-0-289c187f9c6f@mediatek.com > > > > Changes in v3: > > - Remove the Change-Id > > - Link to v2: > https://lore.kernel.org/r/20240710-alpha-blending-v2-0-d4b505e6980a@mediatek.com > > > > Changes in v2: > > - Remove unnecessary codes > > - Add more information to the commit message > > - Link to v1: > https://lore.kernel.org/r/20240620-blend-v1-0-72670072ca20@mediatek.com > > > > --- > > Hsiao Chien Sung (5): > > drm/mediatek: Support "None" blending in OVL > > drm/mediatek: Support "None" blending in Mixer > > drm/mediatek: Support "Pre-multiplied" blending in OVL > > drm/mediatek: Support "Pre-multiplied" blending in Mixer > > drm/mediatek: Support alpha blending in display driver > > > > drivers/gpu/drm/mediatek/mtk_disp_ovl.c | 36 > +++++++++++++++++++++++++-------- > > drivers/gpu/drm/mediatek/mtk_ethdr.c | 13 +++++++++--- > > drivers/gpu/drm/mediatek/mtk_plane.c | 11 ++++++++++ > > 3 files changed, 49 insertions(+), 11 deletions(-) > > --- > > base-commit: 8ad49a92cff4bab13eb2f2725243f5f31eff3f3b > > change-id: 20240710-alpha-blending-067295570863 > > > > Best regards, > > -- > > Hsiao Chien Sung <shawn.sung@mediatek.com> > > > >
Hi, On 2024-07-31 16:34 +03:00, Chun-Kuang Hu wrote: > Hi, Shawn: > > Hsiao Chien Sung via B4 Relay > <devnull+shawn.sung.mediatek.com@kernel.org> 於 2024年7月17日 週三 下午1:24寫道： >> >> Support "Pre-multiplied" and "None" blend mode on MediaTek's chips by >> adding correct blend mode property when the planes init. >> Before this patch, only the "Coverage" mode (default) is supported. > > For the whole series, applied to mediatek-drm-next [1], thanks. > > [1] https://git.kernel.org/pub/scm/linux/kernel/git/chunkuang.hu/linux.git/log/?h=mediatek-drm-next I am seeing broken colors on an MT8173 Chromebook with next-20240912, which goes away if I git-revert this series (commits 1f66fe62cc09 eb17c5909481 a3f7f7ef4bfe 59e9d9de25f0 4225d5d5e779). To illustrate, I took a picture [1] of some color mixing diagrams from Wikipedia [2]. Do you have an idea of what goes wrong? (I'm busy with too many things so I don't want to debug it now...) [1] https://i.imgur.com/tNFvovB.jpeg [2] https://en.wikipedia.org/wiki/Color_space#Generic
Hi, Jason: Please help to fix this issue. Otherwise, I would revert this series. Regards, CK On Thu, 2024-09-12 at 23:29 +0300, Alper Nebi Yasak wrote: > > External email : Please do not click links or open attachments until you have verified the sender or the content. > Hi, > > On 2024-07-31 16:34 +03:00, Chun-Kuang Hu wrote: > > Hi, Shawn: > > > > Hsiao Chien Sung via B4 Relay > > <devnull+shawn.sung.mediatek.com@kernel.org> 於 2024年7月17日 週三 下午1:24寫道： > >> > >> Support "Pre-multiplied" and "None" blend mode on MediaTek's chips by > >> adding correct blend mode property when the planes init. > >> Before this patch, only the "Coverage" mode (default) is supported. > > > > For the whole series, applied to mediatek-drm-next [1], thanks. > > > > [1] https://git.kernel.org/pub/scm/linux/kernel/git/chunkuang.hu/linux.git/log/?h=mediatek-drm-next > > I am seeing broken colors on an MT8173 Chromebook with next-20240912, > which goes away if I git-revert this series (commits 1f66fe62cc09 > eb17c5909481 a3f7f7ef4bfe 59e9d9de25f0 4225d5d5e779). > > To illustrate, I took a picture [1] of some color mixing diagrams from > Wikipedia [2]. Do you have an idea of what goes wrong? > > (I'm busy with too many things so I don't want to debug it now...) > > [1] https://i.imgur.com/tNFvovB.jpeg > [2] https://en.wikipedia.org/wiki/Color_space#Generic >
Hi Alper, On Thu, 2024-09-12 at 23:29 +0300, Alper Nebi Yasak wrote: > Hi, > > On 2024-07-31 16:34 +03:00, Chun-Kuang Hu wrote: > > Hi, Shawn: > > > > Hsiao Chien Sung via B4 Relay > > <devnull+shawn.sung.mediatek.com@kernel.org> 於 2024年7月17日 週三 > > 下午1:24寫道： > > > > > > Support "Pre-multiplied" and "None" blend mode on MediaTek's > > > chips by > > > adding correct blend mode property when the planes init. > > > Before this patch, only the "Coverage" mode (default) is > > > supported. > > > > For the whole series, applied to mediatek-drm-next [1], thanks. > > > > [1] > > https://git.kernel.org/pub/scm/linux/kernel/git/chunkuang.hu/linux.git/log/?h=mediatek-drm-next > > I am seeing broken colors on an MT8173 Chromebook with next-20240912, > which goes away if I git-revert this series (commits 1f66fe62cc09 > eb17c5909481 a3f7f7ef4bfe 59e9d9de25f0 4225d5d5e779). Thank you for your feedback. I apologize for the delay in my response; I was just notified of your email by my colleague. > > To illustrate, I took a picture [1] of some color mixing diagrams > from > Wikipedia [2]. Do you have an idea of what goes wrong? Based on the provided pictures, it appears that this issue is related to the way the MT8173 chip handles alpha channel (ARGB -> RGBA), which differes from the behavior of later chips. > > (I'm busy with too many things so I don't want to debug it now...) > > [1] https://i.imgur.com/tNFvovB.jpeg > [2] https://en.wikipedia.org/wiki/Color_space#Generic We will either submit a new fix ASAP or revert this series as a first step. Best regards, Shawn
Hi Alper, I think that's a platform issue because it works find in my MT8188 platform, but I don't have any MT8173 platform. Can you help me test the fix patch in your MT8173 platform? I'll provide a fix patch in 2 weeks via Google Chat. Regards, Jason-JH.Lin On Fri, 2024-09-13 at 01:25 +0000, CK Hu (胡俊光) wrote: > Hi, Jason: > > Please help to fix this issue. > Otherwise, I would revert this series. > > Regards, > CK > > On Thu, 2024-09-12 at 23:29 +0300, Alper Nebi Yasak wrote: > > > > External email : Please do not click links or open attachments > > until you have verified the sender or the content. > > Hi, > > > > On 2024-07-31 16:34 +03:00, Chun-Kuang Hu wrote: > > > Hi, Shawn: > > > > > > Hsiao Chien Sung via B4 Relay > > > <devnull+shawn.sung.mediatek.com@kernel.org> 於 2024年7月17日 週三 > > > 下午1:24寫道： > > > > > > > > Support "Pre-multiplied" and "None" blend mode on MediaTek's > > > > chips by > > > > adding correct blend mode property when the planes init. > > > > Before this patch, only the "Coverage" mode (default) is > > > > supported. > > > > > > For the whole series, applied to mediatek-drm-next [1], thanks. > > > > > > [1] > > > https://git.kernel.org/pub/scm/linux/kernel/git/chunkuang.hu/linux.git/log/?h=mediatek-drm-next > > > > I am seeing broken colors on an MT8173 Chromebook with next- > > 20240912, > > which goes away if I git-revert this series (commits 1f66fe62cc09 > > eb17c5909481 a3f7f7ef4bfe 59e9d9de25f0 4225d5d5e779). > > > > To illustrate, I took a picture [1] of some color mixing diagrams > > from > > Wikipedia [2]. Do you have an idea of what goes wrong? > > > > (I'm busy with too many things so I don't want to debug it now...) > > > > [1] https://i.imgur.com/tNFvovB.jpeg > > [2] https://en.wikipedia.org/wiki/Color_space#Generic > >
Hi, On 2024-09-13 05:56 +03:00, Jason-JH Lin (林睿祥) wrote: > Hi Alper, > > I think that's a platform issue because it works find in my MT8188 > platform, but I don't have any MT8173 platform. > > Can you help me test the fix patch in your MT8173 platform? > I'll provide a fix patch in 2 weeks via Google Chat. > > Regards, > Jason-JH.Lin I replied over Google Chat, but to make it clear for others on the mailing list: Yes, I'll try to test patches when you send them. > On Fri, 2024-09-13 at 01:25 +0000, CK Hu (胡俊光) wrote: >> Hi, Jason: >> >> Please help to fix this issue. >> Otherwise, I would revert this series. >> >> Regards, >> CK >> >> On Thu, 2024-09-12 at 23:29 +0300, Alper Nebi Yasak wrote: >> > >> > External email : Please do not click links or open attachments >> > until you have verified the sender or the content. >> > Hi, >> > >> > On 2024-07-31 16:34 +03:00, Chun-Kuang Hu wrote: >> > > Hi, Shawn: >> > > >> > > Hsiao Chien Sung via B4 Relay >> > > <devnull+shawn.sung.mediatek.com@kernel.org> 於 2024年7月17日 週三 >> > > 下午1:24寫道： >> > > > >> > > > Support "Pre-multiplied" and "None" blend mode on MediaTek's >> > > > chips by >> > > > adding correct blend mode property when the planes init. >> > > > Before this patch, only the "Coverage" mode (default) is >> > > > supported. >> > > >> > > For the whole series, applied to mediatek-drm-next [1], thanks. >> > > >> > > [1] >> > > https://git.kernel.org/pub/scm/linux/kernel/git/chunkuang.hu/linux.git/log/?h=mediatek-drm-next >> > >> > I am seeing broken colors on an MT8173 Chromebook with next- >> > 20240912, >> > which goes away if I git-revert this series (commits 1f66fe62cc09 >> > eb17c5909481 a3f7f7ef4bfe 59e9d9de25f0 4225d5d5e779). >> > >> > To illustrate, I took a picture [1] of some color mixing diagrams >> > from >> > Wikipedia [2]. Do you have an idea of what goes wrong? >> > >> > (I'm busy with too many things so I don't want to debug it now...) >> > >> > [1] https://i.imgur.com/tNFvovB.jpeg >> > [2] https://en.wikipedia.org/wiki/Color_space#Generic
Il 13/09/24 07:40, Alper Nebi Yasak ha scritto: > Hi, > > On 2024-09-13 05:56 +03:00, Jason-JH Lin (林睿祥) wrote: >> Hi Alper, >> >> I think that's a platform issue because it works find in my MT8188 >> platform, but I don't have any MT8173 platform. >> >> Can you help me test the fix patch in your MT8173 platform? >> I'll provide a fix patch in 2 weeks via Google Chat. Jason, Two weeks is a bit too late as CK already picked your series: reverting and reapplying creates noise in the commit log, and when this can be avoided it should be avoided. Since upstream is broken, please understand that this fix should be developed with *highest* priority (so, as immediate as possible). Thanks, Angelo >> >> Regards, >> Jason-JH.Lin > > I replied over Google Chat, but to make it clear for others on the > mailing list: Yes, I'll try to test patches when you send them. > >> On Fri, 2024-09-13 at 01:25 +0000, CK Hu (胡俊光) wrote: >>> Hi, Jason: >>> >>> Please help to fix this issue. >>> Otherwise, I would revert this series. >>> >>> Regards, >>> CK >>> >>> On Thu, 2024-09-12 at 23:29 +0300, Alper Nebi Yasak wrote: >>>> >>>> External email : Please do not click links or open attachments >>>> until you have verified the sender or the content. >>>> Hi, >>>> >>>> On 2024-07-31 16:34 +03:00, Chun-Kuang Hu wrote: >>>>> Hi, Shawn: >>>>> >>>>> Hsiao Chien Sung via B4 Relay >>>>> <devnull+shawn.sung.mediatek.com@kernel.org> 於 2024年7月17日 週三 >>>>> 下午1:24寫道： >>>>>> >>>>>> Support "Pre-multiplied" and "None" blend mode on MediaTek's >>>>>> chips by >>>>>> adding correct blend mode property when the planes init. >>>>>> Before this patch, only the "Coverage" mode (default) is >>>>>> supported. >>>>> >>>>> For the whole series, applied to mediatek-drm-next [1], thanks. >>>>> >>>>> [1] >>>>> https://git.kernel.org/pub/scm/linux/kernel/git/chunkuang.hu/linux.git/log/?h=mediatek-drm-next >>>> >>>> I am seeing broken colors on an MT8173 Chromebook with next- >>>> 20240912, >>>> which goes away if I git-revert this series (commits 1f66fe62cc09 >>>> eb17c5909481 a3f7f7ef4bfe 59e9d9de25f0 4225d5d5e779). >>>> >>>> To illustrate, I took a picture [1] of some color mixing diagrams >>>> from >>>> Wikipedia [2]. Do you have an idea of what goes wrong? >>>> >>>> (I'm busy with too many things so I don't want to debug it now...) >>>> >>>> [1] https://i.imgur.com/tNFvovB.jpeg >>>> [2] https://en.wikipedia.org/wiki/Color_space#Generic
On Fri, 2024-09-13 at 09:01 +0000, AngeloGioacchino Del Regno wrote: > Il 13/09/24 07:40, Alper Nebi Yasak ha scritto: > > Hi, > > > > On 2024-09-13 05:56 +03:00, Jason-JH Lin (林睿祥) wrote: > > > Hi Alper, > > > > > > I think that's a platform issue because it works find in my > > > MT8188 > > > platform, but I don't have any MT8173 platform. > > > > > > Can you help me test the fix patch in your MT8173 platform? > > > I'll provide a fix patch in 2 weeks via Google Chat. > > Jason, Two weeks is a bit too late as CK already picked your series: > reverting > and reapplying creates noise in the commit log, and when this can be > avoided it > should be avoided. > > Since upstream is broken, please understand that this fix should be > developed > with *highest* priority (so, as immediate as possible). > OK, I'll send the fix path right away! Regards, Jason-JH.Lin > Thanks, > Angelo > > > > > > > Regards, > > > Jason-JH.Lin > > > > I replied over Google Chat, but to make it clear for others on the > > mailing list: Yes, I'll try to test patches when you send them. > > > > > On Fri, 2024-09-13 at 01:25 +0000, CK Hu (胡俊光) wrote: > > > > Hi, Jason: > > > > > > > > Please help to fix this issue. > > > > Otherwise, I would revert this series. > > > > > > > > Regards, > > > > CK > > > > > > > > On Thu, 2024-09-12 at 23:29 +0300, Alper Nebi Yasak wrote: > > > > > > > > > > External email : Please do not click links or open > > > > > attachments > > > > > until you have verified the sender or the content. > > > > > Hi, > > > > > > > > > > On 2024-07-31 16:34 +03:00, Chun-Kuang Hu wrote: > > > > > > Hi, Shawn: > > > > > > > > > > > > Hsiao Chien Sung via B4 Relay > > > > > > <devnull+shawn.sung.mediatek.com@kernel.org> 於 2024年7月17日 > > > > > > 週三 > > > > > > 下午1:24寫道： > > > > > > > > > > > > > > Support "Pre-multiplied" and "None" blend mode on > > > > > > > MediaTek's > > > > > > > chips by > > > > > > > adding correct blend mode property when the planes init. > > > > > > > Before this patch, only the "Coverage" mode (default) is > > > > > > > supported. > > > > > > > > > > > > For the whole series, applied to mediatek-drm-next [1], > > > > > > thanks. > > > > > > > > > > > > [1] > > > > > > https://urldefense.com/v3/__https://git.kernel.org/pub/scm/linux/kernel/git/chunkuang.hu/linux.git/log/?h=mediatek-drm-next__;!!CTRNKA9wMg0ARbw!jrw0cZrT-6A9bviBj9yX0zzeM2NAPin74zALZAajn9k92ZFrRVd3iZVOe0xpjsr21_Xw5v8KDeeF7nDtvT5rhRAMDV7KZ-qQ0Q$ > > > > > > > > > > > > > > > > I am seeing broken colors on an MT8173 Chromebook with next- > > > > > 20240912, > > > > > which goes away if I git-revert this series (commits > > > > > 1f66fe62cc09 > > > > > eb17c5909481 a3f7f7ef4bfe 59e9d9de25f0 4225d5d5e779). > > > > > > > > > > To illustrate, I took a picture [1] of some color mixing > > > > > diagrams > > > > > from > > > > > Wikipedia [2]. Do you have an idea of what goes wrong? > > > > > > > > > > (I'm busy with too many things so I don't want to debug it > > > > > now...) > > > > > > > > > > [1] > > > > > https://urldefense.com/v3/__https://i.imgur.com/tNFvovB.jpeg__;!!CTRNKA9wMg0ARbw!jrw0cZrT-6A9bviBj9yX0zzeM2NAPin74zALZAajn9k92ZFrRVd3iZVOe0xpjsr21_Xw5v8KDeeF7nDtvT5rhRAMDV42KYfvHA$ > > > > > > > > > > [2] > > > > > https://urldefense.com/v3/__https://en.wikipedia.org/wiki/Color_space*Generic__;Iw!!CTRNKA9wMg0ARbw!jrw0cZrT-6A9bviBj9yX0zzeM2NAPin74zALZAajn9k92ZFrRVd3iZVOe0xpjsr21_Xw5v8KDeeF7nDtvT5rhRAMDV67WeSFhQ$ > > > > > > >
Hi Jason, On Sat, 2024-09-14 at 20:22 +0000, Jason-JH Lin (林睿祥) wrote: > On Fri, 2024-09-13 at 09:01 +0000, AngeloGioacchino Del Regno wrote: > > Il 13/09/24 07:40, Alper Nebi Yasak ha scritto: > > > Hi, > > > > > > On 2024-09-13 05:56 +03:00, Jason-JH Lin (林睿祥) wrote: > > > > Hi Alper, > > > > > > > > I think that's a platform issue because it works find in my > > > > MT8188 > > > > platform, but I don't have any MT8173 platform. > > > > > > > > Can you help me test the fix patch in your MT8173 platform? > > > > I'll provide a fix patch in 2 weeks via Google Chat. > > > > Jason, Two weeks is a bit too late as CK already picked your > > series: > > reverting > > and reapplying creates noise in the commit log, and when this can > > be > > avoided it > > should be avoided. > > > > Since upstream is broken, please understand that this fix should be > > developed > > with *highest* priority (so, as immediate as possible). > > > > OK, I'll send the fix path right away! I notice that I'm unable to see replies from external sources. Not sure if this is related to B4 relay, but I wanted to bring this to your attention as feedback while you may want to consider sending the patches with B4 relay as well. Regards, Shawn
Support "Pre-multiplied" and "None" blend mode on MediaTek's chips by adding correct blend mode property when the planes init. Before this patch, only the "Coverage" mode (default) is supported. Signed-off-by: Hsiao Chien Sung <shawn.sung@mediatek.corp-partner.google.com> --- Changes in v4: - Add more information to the commit message - Link to v3: https://lore.kernel.org/r/20240710-alpha-blending-v3-0-289c187f9c6f@mediatek.com Changes in v3: - Remove the Change-Id - Link to v2: https://lore.kernel.org/r/20240710-alpha-blending-v2-0-d4b505e6980a@mediatek.com Changes in v2: - Remove unnecessary codes - Add more information to the commit message - Link to v1: https://lore.kernel.org/r/20240620-blend-v1-0-72670072ca20@mediatek.com --- Hsiao Chien Sung (5): drm/mediatek: Support "None" blending in OVL drm/mediatek: Support "None" blending in Mixer drm/mediatek: Support "Pre-multiplied" blending in OVL drm/mediatek: Support "Pre-multiplied" blending in Mixer drm/mediatek: Support alpha blending in display driver drivers/gpu/drm/mediatek/mtk_disp_ovl.c | 36 +++++++++++++++++++++++++-------- drivers/gpu/drm/mediatek/mtk_ethdr.c | 13 +++++++++--- drivers/gpu/drm/mediatek/mtk_plane.c | 11 ++++++++++ 3 files changed, 49 insertions(+), 11 deletions(-) --- base-commit: 8ad49a92cff4bab13eb2f2725243f5f31eff3f3b change-id: 20240710-alpha-blending-067295570863 Best regards,