Skip to contentSkip to conversation

dt-bindings: arm: mediatek: mmsys: change compatible for MT8195

PATCH In review Bo-Chen Chen 4 participants 9 messages Aug 25, 2022 – Sep 14, 2022

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

PATCH v2 1/1 dt-bindings: arm: mediatek: mmsys: change compatible for MT8195

From: "Jason-JH.Lin" <[email protected]>

For previous MediaTek SoCs, such as MT8173, there are 2 display HW pipelines binding to 1 mmsys with the same power domain, the same clock driver and the same mediatek-drm driver.

For MT8195, VDOSYS0 and VDOSYS1 are 2 display HW pipelines binding to 2 different power domains, different clock drivers and different mediatek-drm drivers.

Moreover, Hardware pipeline of VDOSYS0 has these components: COLOR, CCORR, AAL, GAMMA, DITHER. They are related to the PQ (Picture Quality) and they makes VDOSYS0 supports PQ function while they are not including in VDOSYS1.

Hardware pipeline of VDOSYS1 has the component ETHDR (HDR related component). It makes VDOSYS1 supports the HDR function while it's not including in VDOSYS0.

To summarize0:
Only VDOSYS0 can support PQ adjustment.
Only VDOSYS1 can support HDR adjustment.

Therefore, we need to separate these two different mmsys hardwares to 2 different compatibles for MT8195.

Fixes 81c5a41d10b9 ("dt-bindings: arm: mediatek: mmsys: add mt8195 SoC binding")

---
Changes for v2:
1. Add hardware difference for VDOSYS0 and VDOSYS1 in commit message.

---
 .../devicetree/bindings/arm/mediatek/mediatek,mmsys.yaml       | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
Documentation/devicetree/bindings/arm/mediatek/mediatek,mmsys.yaml +2 -1
index 6ad023eec193..bfbdd30d2092 100644
@@ -31,7 +31,8 @@ properties:
               - mediatek,mt8183-mmsys
               - mediatek,mt8186-mmsys
               - mediatek,mt8192-mmsys
-              - mediatek,mt8195-mmsys
+              - mediatek,mt8195-vdosys0
+              - mediatek,mt8195-vdosys1
               - mediatek,mt8365-mmsys
           - const: syscon
       - items:
Signature · 1 line
2.18.0
Acked-by on v2 1/1
#1 lore raw
Quoted 23 lines · Bo-Chen Chen

From: "Jason-JH.Lin" <[email protected]>

For previous MediaTek SoCs, such as MT8173, there are 2 display HW pipelines binding to 1 mmsys with the same power domain, the same clock driver and the same mediatek-drm driver.

For MT8195, VDOSYS0 and VDOSYS1 are 2 display HW pipelines binding to 2 different power domains, different clock drivers and different mediatek-drm drivers.

Moreover, Hardware pipeline of VDOSYS0 has these components: COLOR, CCORR, AAL, GAMMA, DITHER. They are related to the PQ (Picture Quality) and they makes VDOSYS0 supports PQ function while they are not including in VDOSYS1.

Hardware pipeline of VDOSYS1 has the component ETHDR (HDR related component). It makes VDOSYS1 supports the HDR function while it's not including in VDOSYS0.

To summarize0:
Only VDOSYS0 can support PQ adjustment.
Only VDOSYS1 can support HDR adjustment.

Quoted 7 lines

Therefore, we need to separate these two different mmsys hardwares to 2 different compatibles for MT8195.

Fixes 81c5a41d10b9 ("dt-bindings: arm: mediatek: mmsys: add mt8195 SoC binding")

---

Best regards,
Krzysztof

maintainer
#2 lore raw
In reply to Bo-Chen Chen
Quoted 41 lines · Bo-Chen Chen

From: "Jason-JH.Lin" <[email protected]>

For previous MediaTek SoCs, such as MT8173, there are 2 display HW pipelines binding to 1 mmsys with the same power domain, the same clock driver and the same mediatek-drm driver.

For MT8195, VDOSYS0 and VDOSYS1 are 2 display HW pipelines binding to 2 different power domains, different clock drivers and different mediatek-drm drivers.

Moreover, Hardware pipeline of VDOSYS0 has these components: COLOR, CCORR, AAL, GAMMA, DITHER. They are related to the PQ (Picture Quality) and they makes VDOSYS0 supports PQ function while they are not including in VDOSYS1.

Hardware pipeline of VDOSYS1 has the component ETHDR (HDR related component). It makes VDOSYS1 supports the HDR function while it's not including in VDOSYS0.

To summarize0:
Only VDOSYS0 can support PQ adjustment.
Only VDOSYS1 can support HDR adjustment.

Therefore, we need to separate these two different mmsys hardwares to 2 different compatibles for MT8195.

Fixes 81c5a41d10b9 ("dt-bindings: arm: mediatek: mmsys: add mt8195 SoC binding")

---
Changes for v2:
1. Add hardware difference for VDOSYS0 and VDOSYS1 in commit message.

---
  .../devicetree/bindings/arm/mediatek/mediatek,mmsys.yaml       | 3 ++-
  1 file changed, 2 insertions(+), 1 deletion(-)
Documentation/devicetree/bindings/arm/mediatek/mediatek,mmsys.yaml +0 -0
index 6ad023eec193..bfbdd30d2092 100644

Documentation/devicetree/bindings/arm/mediatek/mediatek,mmsys.yaml properties: · line 34

Quoted 5 lines
Documentation/devicetree/bindings/arm/mediatek/mediatek,mmsys.yaml +1 -1
@@ -31,7 +31,8 @@ properties:
                - mediatek,mt8183-mmsys
                - mediatek,mt8186-mmsys
                - mediatek,mt8192-mmsys
-              - mediatek,mt8195-mmsys
+              - mediatek,mt8195-vdosys0

Thanks for you patch. As I mentioned on v1, I propose to set mediatek,mt8195-mmsys as fallback for mediatek,mt8195-vdosys0 to not break backwards compatibility.

Apart from that, the binding change will need some changes to support the new binding. Please provide these together with this patch.

Regards,
Matthias

Quoted 3 lines
Documentation/devicetree/bindings/arm/mediatek/mediatek,mmsys.yaml +1 -0
+              - mediatek,mt8195-vdosys >                 - mediatek,mt8365-mmsys
            - const: syscon
        - items:
#3 lore raw
Quoted 69 lines · Matthias Brugger
Quoted 56 lines · Bo-Chen Chen

From: "Jason-JH.Lin" <[email protected]>

For previous MediaTek SoCs, such as MT8173, there are 2 display HW pipelines binding to 1 mmsys with the same power domain, the same clock driver and the same mediatek-drm driver.

For MT8195, VDOSYS0 and VDOSYS1 are 2 display HW pipelines binding to
2 different power domains, different clock drivers and different mediatek-drm drivers.

Moreover, Hardware pipeline of VDOSYS0 has these components: COLOR, CCORR, AAL, GAMMA, DITHER. They are related to the PQ (Picture Quality)
and they makes VDOSYS0 supports PQ function while they are not including in VDOSYS1.

Hardware pipeline of VDOSYS1 has the component ETHDR (HDR related component). It makes VDOSYS1 supports the HDR function while it's not
including in VDOSYS0.

To summarize0:
Only VDOSYS0 can support PQ adjustment.
Only VDOSYS1 can support HDR adjustment.

Therefore, we need to separate these two different mmsys hardwares to
2 different compatibles for MT8195.

Fixes 81c5a41d10b9 ("dt-bindings: arm: mediatek: mmsys: add mt8195

SoC binding")

---
Changes for v2:
1. Add hardware difference for VDOSYS0 and VDOSYS1 in commit message.
---
.../devicetree/bindings/arm/mediatek/mediatek,mmsys.yaml |
3 ++-

  1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/Documentation/devicetree/bindings/arm/mediatek/mediatek,mmsys.yam l b/Documentation/devicetree/bindings/arm/mediatek/mediatek,mmsys.yam l
index 6ad023eec193..bfbdd30d2092 100644
---
a/Documentation/devicetree/bindings/arm/mediatek/mediatek,mmsys.yam l
+++
b/Documentation/devicetree/bindings/arm/mediatek/mediatek,mmsys.yam l

Documentation/devicetree/bindings/arm/mediatek/mediatek,mmsys.yaml properties: · line 34

Quoted 5 lines
Documentation/devicetree/bindings/arm/mediatek/mediatek,mmsys.yaml +1 -1
@@ -31,7 +31,8 @@ properties:
                - mediatek,mt8183-mmsys
                - mediatek,mt8186-mmsys
                - mediatek,mt8192-mmsys
-              - mediatek,mt8195-mmsys
+              - mediatek,mt8195-vdosys0

Thanks for you patch. As I mentioned on v1, I propose to set mediatek,mt8195-mmsys as fallback for mediatek,mt8195-vdosys0 to not break
backwards compatibility.

Quoted 7 lines

Apart from that, the binding change will need some changes to support the new
binding. Please provide these together with this patch.

Regards,
Matthias

Hello Matthias,

Thanks for your comments.
The purpose of this patch is to confirm we can separate mt8195 mmsys into two compatibles. I think this modification is accepted.

After this, I think Jason-JH will push another series with this binding patch. In Jason-JH's series, we will modify mmsys driver based on this. And I think we don't need to keep "mediatek,mt8195-mmsys" if we also modify mmsys drivers in the same series.

Is it ok that postpones to pick this patch until we finish review follow-up series?

BRs,
Bo-Chen

Quoted 4 lines
Quoted 4 lines
(quoted diff) +1 -0
+              - mediatek,mt8195-vdosys >                 -

mediatek,mt8365-mmsys
- const: syscon
- items:

#4 lore raw
Quoted 78 lines · Bo-Chen Chen
Quoted 69 lines · Matthias Brugger
Quoted 56 lines · Bo-Chen Chen

From: "Jason-JH.Lin" <[email protected]>

For previous MediaTek SoCs, such as MT8173, there are 2 display HW pipelines binding to 1 mmsys with the same power domain, the same clock driver and the same mediatek-drm driver.

For MT8195, VDOSYS0 and VDOSYS1 are 2 display HW pipelines binding to
2 different power domains, different clock drivers and different mediatek-drm drivers.

Moreover, Hardware pipeline of VDOSYS0 has these components: COLOR, CCORR, AAL, GAMMA, DITHER. They are related to the PQ (Picture Quality)
and they makes VDOSYS0 supports PQ function while they are not including in VDOSYS1.

Hardware pipeline of VDOSYS1 has the component ETHDR (HDR related component). It makes VDOSYS1 supports the HDR function while it's not
including in VDOSYS0.

To summarize0:
Only VDOSYS0 can support PQ adjustment.
Only VDOSYS1 can support HDR adjustment.

Therefore, we need to separate these two different mmsys hardwares to
2 different compatibles for MT8195.

Fixes 81c5a41d10b9 ("dt-bindings: arm: mediatek: mmsys: add mt8195

SoC binding")

---
Changes for v2:
1. Add hardware difference for VDOSYS0 and VDOSYS1 in commit message.
---
.../devicetree/bindings/arm/mediatek/mediatek,mmsys.yaml |
3 ++-

  1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/Documentation/devicetree/bindings/arm/mediatek/mediatek,mmsys.yam l b/Documentation/devicetree/bindings/arm/mediatek/mediatek,mmsys.yam l
index 6ad023eec193..bfbdd30d2092 100644
---
a/Documentation/devicetree/bindings/arm/mediatek/mediatek,mmsys.yam l
+++
b/Documentation/devicetree/bindings/arm/mediatek/mediatek,mmsys.yam l

Documentation/devicetree/bindings/arm/mediatek/mediatek,mmsys.yaml properties: · line 34

Quoted 5 lines
Documentation/devicetree/bindings/arm/mediatek/mediatek,mmsys.yaml +1 -1
@@ -31,7 +31,8 @@ properties:
                - mediatek,mt8183-mmsys
                - mediatek,mt8186-mmsys
                - mediatek,mt8192-mmsys
-              - mediatek,mt8195-mmsys
+              - mediatek,mt8195-vdosys0

Thanks for you patch. As I mentioned on v1, I propose to set mediatek,mt8195-mmsys as fallback for mediatek,mt8195-vdosys0 to not break
backwards compatibility.

Quoted 7 lines

Apart from that, the binding change will need some changes to support the new
binding. Please provide these together with this patch.

Regards,
Matthias

Quoted 5 lines

Hello Matthias,

Thanks for your comments.
The purpose of this patch is to confirm we can separate mt8195 mmsys into two compatibles. I think this modification is accepted.

No, it is not accepted following Matthias comments. You received my ack based on assumption that ABI break is perfectly ok for platform maintainer, as he has decisive voice. If anyone is not happy with a ABI break, then his concerns must be addressed.

So let it be specific:
NAK.

Quoted 3 lines

After this, I think Jason-JH will push another series with this binding patch.

I don't know what do you mean here - another series on top of wrong patch?

Quoted 3 lines

In Jason-JH's series, we will modify mmsys driver based on this. And I think we don't need to keep "mediatek,mt8195-mmsys" if we also modify mmsys drivers in the same series.

This does not fux ABI break and broken bisectability.

Quoted 4 lines

Is it ok that postpones to pick this patch until we finish review follow-up series?

No. You got a clear review to fix.

Best regards,
Krzysztof

#5 lore raw
Quoted 114 lines · Krzysztof Kozlowski
Quoted 95 lines · Bo-Chen Chen
Quoted 85 lines · Matthias Brugger
Quoted 71 lines · Bo-Chen Chen

From: "Jason-JH.Lin" <[email protected]>

For previous MediaTek SoCs, such as MT8173, there are 2 display HW
pipelines binding to 1 mmsys with the same power domain, the same
clock driver and the same mediatek-drm driver.

For MT8195, VDOSYS0 and VDOSYS1 are 2 display HW pipelines binding
to
2 different power domains, different clock drivers and different
mediatek-drm drivers.

Moreover, Hardware pipeline of VDOSYS0 has these components:
COLOR,
CCORR, AAL, GAMMA, DITHER. They are related to the PQ (Picture Quality)
and they makes VDOSYS0 supports PQ function while they are not including in VDOSYS1.

Hardware pipeline of VDOSYS1 has the component ETHDR (HDR related
component). It makes VDOSYS1 supports the HDR function while it's
not
including in VDOSYS0.

To summarize0:
Only VDOSYS0 can support PQ adjustment.
Only VDOSYS1 can support HDR adjustment.

Therefore, we need to separate these two different mmsys hardwares
to
2 different compatibles for MT8195.

Fixes 81c5a41d10b9 ("dt-bindings: arm: mediatek: mmsys: add

mt8195
SoC binding")

---
Changes for v2:
1. Add hardware difference for VDOSYS0 and VDOSYS1 in commit message.
---

.../devicetree/bindings/arm/mediatek/mediatek,mmsys.yaml |
3 ++-

  1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/Documentation/devicetree/bindings/arm/mediatek/mediatek,mmsys .yam
l b/Documentation/devicetree/bindings/arm/mediatek/mediatek,mmsys .yam
l
index 6ad023eec193..bfbdd30d2092 100644
---
a/Documentation/devicetree/bindings/arm/mediatek/mediatek,mmsys .yam
l
+++
b/Documentation/devicetree/bindings/arm/mediatek/mediatek,mmsys .yam
l

Documentation/devicetree/bindings/arm/mediatek/mediatek,mmsys.yaml properties: · line 34

Quoted 5 lines
Documentation/devicetree/bindings/arm/mediatek/mediatek,mmsys.yaml +1 -1
@@ -31,7 +31,8 @@ properties:
                - mediatek,mt8183-mmsys
                - mediatek,mt8186-mmsys
                - mediatek,mt8192-mmsys
-              - mediatek,mt8195-mmsys
+              - mediatek,mt8195-vdosys0

Thanks for you patch. As I mentioned on v1, I propose to set mediatek,mt8195-mmsys as fallback for mediatek,mt8195-vdosys0 to not
break
backwards compatibility.

Quoted 8 lines

Apart from that, the binding change will need some changes to support
the new
binding. Please provide these together with this patch.

Regards,
Matthias

Quoted 6 lines

Hello Matthias,

Thanks for your comments.
The purpose of this patch is to confirm we can separate mt8195 mmsys
into two compatibles. I think this modification is accepted.

No, it is not accepted following Matthias comments. You received my ack
based on assumption that ABI break is perfectly ok for platform maintainer, as he has decisive voice. If anyone is not happy with a ABI
break, then his concerns must be addressed.

So let it be specific:
NAK.

Quoted 1 line
Quoted 7 lines
Quoted 3 lines

After this, I think Jason-JH will push another series with this binding
patch.

I don't know what do you mean here - another series on top of wrong patch?

Hello Krzysztof,

For this mt8195 mmsys binding separation, we still need to modify driver for this. The reason I send this patch is to confirm we can do this binding modification and I also think we can not pick this patch here.

We will push another series and it contains modification of binding and drivers. (The series will push by Jason-JH Lin)

Maybe I should use "RFC" for this series, and I think it's more correct.

BRs,
Bo-Chen

Quoted 16 lines
Quoted 4 lines

In Jason-JH's series, we will modify mmsys driver based on this. And I think we don't need to keep "mediatek,mt8195-mmsys" if we also
modify mmsys drivers in the same series.

This does not fux ABI break and broken bisectability.

Quoted 4 lines

Is it ok that postpones to pick this patch until we finish review follow-up series?

No. You got a clear review to fix.

Best regards,
Krzysztof

maintainer
#6 lore raw
Quoted 125 lines · Bo-Chen Chen
Quoted 114 lines · Krzysztof Kozlowski
Quoted 95 lines · Bo-Chen Chen
Quoted 85 lines · Matthias Brugger
Quoted 71 lines · Bo-Chen Chen

From: "Jason-JH.Lin" <[email protected]>

For previous MediaTek SoCs, such as MT8173, there are 2 display HW
pipelines binding to 1 mmsys with the same power domain, the same
clock driver and the same mediatek-drm driver.

For MT8195, VDOSYS0 and VDOSYS1 are 2 display HW pipelines binding
to
2 different power domains, different clock drivers and different
mediatek-drm drivers.

Moreover, Hardware pipeline of VDOSYS0 has these components:
COLOR,
CCORR, AAL, GAMMA, DITHER. They are related to the PQ (Picture Quality)
and they makes VDOSYS0 supports PQ function while they are not including in VDOSYS1.

Hardware pipeline of VDOSYS1 has the component ETHDR (HDR related
component). It makes VDOSYS1 supports the HDR function while it's
not
including in VDOSYS0.

To summarize0:
Only VDOSYS0 can support PQ adjustment.
Only VDOSYS1 can support HDR adjustment.

Therefore, we need to separate these two different mmsys hardwares
to
2 different compatibles for MT8195.

Fixes 81c5a41d10b9 ("dt-bindings: arm: mediatek: mmsys: add

mt8195
SoC binding")

---
Changes for v2:
1. Add hardware difference for VDOSYS0 and VDOSYS1 in commit message.
---

.../devicetree/bindings/arm/mediatek/mediatek,mmsys.yaml |
3 ++-

   1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/Documentation/devicetree/bindings/arm/mediatek/mediatek,mmsys .yam
l b/Documentation/devicetree/bindings/arm/mediatek/mediatek,mmsys .yam
l
index 6ad023eec193..bfbdd30d2092 100644
---
a/Documentation/devicetree/bindings/arm/mediatek/mediatek,mmsys .yam
l
+++
b/Documentation/devicetree/bindings/arm/mediatek/mediatek,mmsys .yam
l

Documentation/devicetree/bindings/arm/mediatek/mediatek,mmsys.yaml properties: · line 34

Quoted 5 lines
Documentation/devicetree/bindings/arm/mediatek/mediatek,mmsys.yaml +1 -1
@@ -31,7 +31,8 @@ properties:
                 - mediatek,mt8183-mmsys
                 - mediatek,mt8186-mmsys
                 - mediatek,mt8192-mmsys
-              - mediatek,mt8195-mmsys
+              - mediatek,mt8195-vdosys0

Thanks for you patch. As I mentioned on v1, I propose to set mediatek,mt8195-mmsys as fallback for mediatek,mt8195-vdosys0 to not
break
backwards compatibility.

Quoted 8 lines

Apart from that, the binding change will need some changes to support
the new
binding. Please provide these together with this patch.

Regards,
Matthias

Quoted 6 lines

Hello Matthias,

Thanks for your comments.
The purpose of this patch is to confirm we can separate mt8195 mmsys
into two compatibles. I think this modification is accepted.

No, it is not accepted following Matthias comments. You received my ack
based on assumption that ABI break is perfectly ok for platform maintainer, as he has decisive voice. If anyone is not happy with a ABI
break, then his concerns must be addressed.

So let it be specific:
NAK.

Quoted 1 line
Quoted 7 lines
Quoted 3 lines

After this, I think Jason-JH will push another series with this binding
patch.

I don't know what do you mean here - another series on top of wrong patch?

Hello Krzysztof,

Quoted 8 lines

For this mt8195 mmsys binding separation, we still need to modify driver for this. The reason I send this patch is to confirm we can do this binding modification and I also think we can not pick this patch here.

We will push another series and it contains modification of binding and drivers. (The series will push by Jason-JH Lin)

Sounds good. So lets wait for Jason-JH Lin to send this series and we can go on with the review.

Thanks!
Matthias

Quoted 23 lines

Maybe I should use "RFC" for this series, and I think it's more correct.

BRs,
Bo-Chen

Quoted 16 lines
Quoted 4 lines

In Jason-JH's series, we will modify mmsys driver based on this. And I think we don't need to keep "mediatek,mt8195-mmsys" if we also
modify mmsys drivers in the same series.

This does not fux ABI break and broken bisectability.

Quoted 4 lines

Is it ok that postpones to pick this patch until we finish review follow-up series?

No. You got a clear review to fix.

Best regards,
Krzysztof

maintainer
#7 lore raw
Quoted 132 lines · Matthias Brugger
Quoted 129 lines · Bo-Chen Chen
Quoted 113 lines · Krzysztof Kozlowski
Quoted 94 lines · Bo-Chen Chen
Quoted 84 lines · Matthias Brugger
Quoted 70 lines · Bo-Chen Chen

From: "Jason-JH.Lin" <[email protected]>

For previous MediaTek SoCs, such as MT8173, there are 2 display HW
pipelines binding to 1 mmsys with the same power domain, the same
clock driver and the same mediatek-drm driver.

For MT8195, VDOSYS0 and VDOSYS1 are 2 display HW pipelines binding
to
2 different power domains, different clock drivers and different
mediatek-drm drivers.

Moreover, Hardware pipeline of VDOSYS0 has these components:
COLOR,
CCORR, AAL, GAMMA, DITHER. They are related to the PQ (Picture Quality)
and they makes VDOSYS0 supports PQ function while they are not including in VDOSYS1.

Hardware pipeline of VDOSYS1 has the component ETHDR (HDR related
component). It makes VDOSYS1 supports the HDR function while it's
not
including in VDOSYS0.

To summarize0:
Only VDOSYS0 can support PQ adjustment.
Only VDOSYS1 can support HDR adjustment.

Therefore, we need to separate these two different mmsys hardwares
to
2 different compatibles for MT8195.

Fixes 81c5a41d10b9 ("dt-bindings: arm: mediatek: mmsys: add

mt8195
SoC binding")

---
Changes for v2:
1. Add hardware difference for VDOSYS0 and VDOSYS1 in commit message.
---
.../devicetree/bindings/arm/mediatek/mediatek,mmsys.yaml |
3 ++-

   1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/Documentation/devicetree/bindings/arm/mediatek/mediatek,mmsys .yam
l b/Documentation/devicetree/bindings/arm/mediatek/mediatek,mmsys .yam
l
index 6ad023eec193..bfbdd30d2092 100644
---
a/Documentation/devicetree/bindings/arm/mediatek/mediatek,mmsys .yam
l
+++
b/Documentation/devicetree/bindings/arm/mediatek/mediatek,mmsys .yam
l

(quoted diff) +0 -0
@@ -31,7 +31,8 @@ properties:
Quoted 5 lines

                 - mediatek,mt8183-mmsys
                 - mediatek,mt8186-mmsys
                 - mediatek,mt8192-mmsys
-              - mediatek,mt8195-mmsys
+              - mediatek,mt8195-vdosys0

Thanks for you patch. As I mentioned on v1, I propose to set mediatek,mt8195-mmsys as fallback for mediatek,mt8195-vdosys0 to not
break
backwards compatibility.

Quoted 8 lines

Apart from that, the binding change will need some changes to support
the new
binding. Please provide these together with this patch.

Regards,
Matthias

Quoted 6 lines

Hello Matthias,

Thanks for your comments.
The purpose of this patch is to confirm we can separate mt8195 mmsys
into two compatibles. I think this modification is accepted.

No, it is not accepted following Matthias comments. You received my ack
based on assumption that ABI break is perfectly ok for platform maintainer, as he has decisive voice. If anyone is not happy with a ABI
break, then his concerns must be addressed.

So let it be specific:
NAK.

Quoted 1 line
Quoted 7 lines
Quoted 3 lines

After this, I think Jason-JH will push another series with this binding
patch.

I don't know what do you mean here - another series on top of wrong patch?

Hello Krzysztof,

For this mt8195 mmsys binding separation, we still need to modify driver for this. The reason I send this patch is to confirm we can do this binding modification and I also think we can not pick this patch here.

Quoted 6 lines
Quoted 3 lines

We will push another series and it contains modification of binding and drivers. (The series will push by Jason-JH Lin)

Sounds good. So lets wait for Jason-JH Lin to send this series and we can go on with the review.

Just to be sure, this has not happened yet, correct?

Regards,
Matthias

Quoted 26 lines

Thanks!
Matthias

Quoted 23 lines

Maybe I should use "RFC" for this series, and I think it's more correct.

BRs,
Bo-Chen

Quoted 16 lines
Quoted 4 lines

In Jason-JH's series, we will modify mmsys driver based on this. And I think we don't need to keep "mediatek,mt8195-mmsys" if we also
modify mmsys drivers in the same series.

This does not fux ABI break and broken bisectability.

Quoted 4 lines

Is it ok that postpones to pick this patch until we finish review follow-up series?

No. You got a clear review to fix.

Best regards,
Krzysztof

#8 lore raw
Quoted 168 lines · Matthias Brugger
Quoted 156 lines · Matthias Brugger
Quoted 153 lines · Bo-Chen Chen
Quoted 133 lines · Krzysztof Kozlowski
Quoted 111 lines · Bo-Chen Chen
Quoted 100 lines · Matthias Brugger
Quoted 84 lines · Bo-Chen Chen

From: "Jason-JH.Lin" <[email protected]>

For previous MediaTek SoCs, such as MT8173, there are 2 display
HW
pipelines binding to 1 mmsys with the same power domain, the
same
clock driver and the same mediatek-drm driver.

For MT8195, VDOSYS0 and VDOSYS1 are 2 display HW pipelines
binding
to
2 different power domains, different clock drivers and different
mediatek-drm drivers.

Moreover, Hardware pipeline of VDOSYS0 has these components:
COLOR,
CCORR, AAL, GAMMA, DITHER. They are related to the PQ (Picture
Quality)
and they makes VDOSYS0 supports PQ function while they are not
including in VDOSYS1.

Hardware pipeline of VDOSYS1 has the component ETHDR (HDR related
component). It makes VDOSYS1 supports the HDR function while
it's
not
including in VDOSYS0.

To summarize0:
Only VDOSYS0 can support PQ adjustment.
Only VDOSYS1 can support HDR adjustment.

Therefore, we need to separate these two different mmsys hardwares
to
2 different compatibles for MT8195.

Fixes 81c5a41d10b9 ("dt-bindings: arm: mediatek: mmsys:

add
mt8195
SoC binding")

---
Changes for v2:
1. Add hardware difference for VDOSYS0 and VDOSYS1 in commit
message.
---
.../devicetree/bindings/arm/mediatek/mediatek,mmsys.yaml

Quoted 1 line

3 ++-

   1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/Documentation/devicetree/bindings/arm/mediatek/mediatek ,mmsys
.yam
l b/Documentation/devicetree/bindings/arm/mediatek/mediatek ,mmsys
.yam
l
index 6ad023eec193..bfbdd30d2092 100644
---
a/Documentation/devicetree/bindings/arm/mediatek/mediatek ,mmsys
.yam
l
+++
b/Documentation/devicetree/bindings/arm/mediatek/mediatek ,mmsys
.yam
l

Review properties: · line 34

Quoted 5 lines
(quoted diff) +1 -1
@@ -31,7 +31,8 @@ properties:
                 - mediatek,mt8183-mmsys
                 - mediatek,mt8186-mmsys
                 - mediatek,mt8192-mmsys
-              - mediatek,mt8195-mmsys
+              - mediatek,mt8195-vdosys0

Thanks for you patch. As I mentioned on v1, I propose to set
mediatek,mt8195-mmsys as fallback for mediatek,mt8195- vdosys0 to
not
break
backwards compatibility.

Quoted 9 lines

Apart from that, the binding change will need some changes to
support
the new
binding. Please provide these together with this patch.

Regards,
Matthias

Quoted 7 lines

Hello Matthias,

Thanks for your comments.
The purpose of this patch is to confirm we can separate mt8195
mmsys
into two compatibles. I think this modification is accepted.

No, it is not accepted following Matthias comments. You received my
ack
based on assumption that ABI break is perfectly ok for platform maintainer, as he has decisive voice. If anyone is not happy with a
ABI
break, then his concerns must be addressed.

So let it be specific:
NAK.

Quoted 1 line
Quoted 9 lines
Quoted 4 lines

After this, I think Jason-JH will push another series with this
binding
patch.

I don't know what do you mean here - another series on top of wrong
patch?

Hello Krzysztof,

For this mt8195 mmsys binding separation, we still need to modify driver for this. The reason I send this patch is to confirm we can do
this binding modification and I also think we can not pick this patch
here.

Quoted 9 lines
Quoted 4 lines

We will push another series and it contains modification of binding and
drivers. (The series will push by Jason-JH Lin)

Sounds good. So lets wait for Jason-JH Lin to send this series and we can go on
with the review.

Quoted 4 lines

Just to be sure, this has not happened yet, correct?

Regards,
Matthias

Yes, sorry for the late reply to you.
I'll start to send this series at 9/19.

Regards,
Jason-JH.Lin

Quoted 29 lines
Quoted 29 lines

Thanks!
Matthias

Quoted 25 lines

Maybe I should use "RFC" for this series, and I think it's more correct.

BRs,
Bo-Chen

Quoted 19 lines
Quoted 6 lines

In Jason-JH's series, we will modify mmsys driver based on this.
And I think we don't need to keep "mediatek,mt8195-mmsys" if we
also
modify mmsys drivers in the same series.

This does not fux ABI break and broken bisectability.

Quoted 5 lines

Is it ok that postpones to pick this patch until we finish review
follow-up series?

No. You got a clear review to fix.

Best regards,
Krzysztof

Signature · 1 line
Jason-JH Lin <[email protected]>