diff mbox series

[v4,1/8] dt-bindings: arm: mediatek: mmsys: Add support for MT8195 VPPSYS

Message ID 20221207094921.15450-2-moudy.ho@mediatek.com (mailing list archive)
State New, archived
Headers show
Series add support MDP3 on MT8195 platform | expand

Commit Message

Moudy Ho Dec. 7, 2022, 9:49 a.m. UTC
For MT8195, VPPSYS0 and VPPSYS1 are 2 display pipes with
hardware differences in power domains, clocks and subsystem counts,
which should be determined by compatible names.

Signed-off-by: Moudy Ho <moudy.ho@mediatek.com>
Acked-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
---
 .../devicetree/bindings/arm/mediatek/mediatek,mmsys.yaml     | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

Comments

AngeloGioacchino Del Regno Dec. 7, 2022, 10:15 a.m. UTC | #1
Il 07/12/22 10:49, Moudy Ho ha scritto:
> For MT8195, VPPSYS0 and VPPSYS1 are 2 display pipes with
> hardware differences in power domains, clocks and subsystem counts,
> which should be determined by compatible names.
> 
> Signed-off-by: Moudy Ho <moudy.ho@mediatek.com>
> Acked-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>

Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Rob Herring Dec. 19, 2022, 1:44 p.m. UTC | #2
On Wed, Dec 7, 2022 at 3:49 AM Moudy Ho <moudy.ho@mediatek.com> wrote:
>
> For MT8195, VPPSYS0 and VPPSYS1 are 2 display pipes with
> hardware differences in power domains, clocks and subsystem counts,
> which should be determined by compatible names.
>
> Signed-off-by: Moudy Ho <moudy.ho@mediatek.com>
> Acked-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
> ---
>  .../devicetree/bindings/arm/mediatek/mediatek,mmsys.yaml     | 5 ++++-
>  1 file changed, 4 insertions(+), 1 deletion(-)

This landed in today's linux-next causing the following binding
warnings. Is this a fix? Only fixes for 6.2 should be added during the
merge window.

/builds/robherring/linux-dt/Documentation/devicetree/bindings/arm/mediatek/mediatek,mt8195-clock.example.dtb:
clock-controller@14000000: $nodename:0: 'clock-controller@14000000'
does not match '^syscon@[0-9a-f]+$'
        From schema:
/builds/robherring/linux-dt/Documentation/devicetree/bindings/arm/mediatek/mediatek,mmsys.yaml
/builds/robherring/linux-dt/Documentation/devicetree/bindings/arm/mediatek/mediatek,mt8195-clock.example.dtb:
clock-controller@14000000: compatible: 'oneOf' conditional failed, one
must be fixed:
        ['mediatek,mt8195-vppsys0'] is too short
        'mediatek,mt8195-vppsys0' is not one of
['mediatek,mt2701-mmsys', 'mediatek,mt2712-mmsys',
'mediatek,mt6765-mmsys', 'mediatek,mt6779-mmsys',
'mediatek,mt6795-mmsys', 'mediatek,mt6797-mmsys',
'mediatek,mt8167-mmsys', 'mediatek,mt8173-mmsys',
'mediatek,mt8183-mmsys', 'mediatek,mt8186-mmsys',
'mediatek,mt8188-vdosys0', 'mediatek,mt8192-mmsys',
'mediatek,mt8365-mmsys']
        'mediatek,mt8195-mmsys' was expected
        'mediatek,mt7623-mmsys' was expected
        From schema:
/builds/robherring/linux-dt/Documentation/devicetree/bindings/arm/mediatek/mediatek,mmsys.yaml
Matthias Brugger Dec. 19, 2022, 3:17 p.m. UTC | #3
On 19/12/2022 14:44, Rob Herring wrote:
> On Wed, Dec 7, 2022 at 3:49 AM Moudy Ho <moudy.ho@mediatek.com> wrote:
>>
>> For MT8195, VPPSYS0 and VPPSYS1 are 2 display pipes with
>> hardware differences in power domains, clocks and subsystem counts,
>> which should be determined by compatible names.
>>
>> Signed-off-by: Moudy Ho <moudy.ho@mediatek.com>
>> Acked-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
>> ---
>>   .../devicetree/bindings/arm/mediatek/mediatek,mmsys.yaml     | 5 ++++-
>>   1 file changed, 4 insertions(+), 1 deletion(-)
> 
> This landed in today's linux-next causing the following binding
> warnings. Is this a fix? Only fixes for 6.2 should be added during the
> merge window.
> 

No it's not a fix, it's queued for the next -rc phase. I didn't know of the 
limitation of linux-next only allowing fixes in the merge window.

Actually the patch is wrong. It shouldn't include "mediatek,mt8195-mmsys" as 
fallback for mediatek,mt8195-vppsys[0,1].

I'll revert the whole series from v6.2-tmp/dts64 and v6.2-tmp/soc

Regards,
Matthias

> /builds/robherring/linux-dt/Documentation/devicetree/bindings/arm/mediatek/mediatek,mt8195-clock.example.dtb:
> clock-controller@14000000: $nodename:0: 'clock-controller@14000000'
> does not match '^syscon@[0-9a-f]+$'
>          From schema:
> /builds/robherring/linux-dt/Documentation/devicetree/bindings/arm/mediatek/mediatek,mmsys.yaml
> /builds/robherring/linux-dt/Documentation/devicetree/bindings/arm/mediatek/mediatek,mt8195-clock.example.dtb:
> clock-controller@14000000: compatible: 'oneOf' conditional failed, one
> must be fixed:
>          ['mediatek,mt8195-vppsys0'] is too short
>          'mediatek,mt8195-vppsys0' is not one of
> ['mediatek,mt2701-mmsys', 'mediatek,mt2712-mmsys',
> 'mediatek,mt6765-mmsys', 'mediatek,mt6779-mmsys',
> 'mediatek,mt6795-mmsys', 'mediatek,mt6797-mmsys',
> 'mediatek,mt8167-mmsys', 'mediatek,mt8173-mmsys',
> 'mediatek,mt8183-mmsys', 'mediatek,mt8186-mmsys',
> 'mediatek,mt8188-vdosys0', 'mediatek,mt8192-mmsys',
> 'mediatek,mt8365-mmsys']
>          'mediatek,mt8195-mmsys' was expected
>          'mediatek,mt7623-mmsys' was expected
>          From schema:
> /builds/robherring/linux-dt/Documentation/devicetree/bindings/arm/mediatek/mediatek,mmsys.yaml
Krzysztof Kozlowski Dec. 19, 2022, 3:31 p.m. UTC | #4
On 19/12/2022 16:17, Matthias Brugger wrote:
> 
> 
> On 19/12/2022 14:44, Rob Herring wrote:
>> On Wed, Dec 7, 2022 at 3:49 AM Moudy Ho <moudy.ho@mediatek.com> wrote:
>>>
>>> For MT8195, VPPSYS0 and VPPSYS1 are 2 display pipes with
>>> hardware differences in power domains, clocks and subsystem counts,
>>> which should be determined by compatible names.
>>>
>>> Signed-off-by: Moudy Ho <moudy.ho@mediatek.com>
>>> Acked-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
>>> ---
>>>   .../devicetree/bindings/arm/mediatek/mediatek,mmsys.yaml     | 5 ++++-
>>>   1 file changed, 4 insertions(+), 1 deletion(-)
>>
>> This landed in today's linux-next causing the following binding
>> warnings. Is this a fix? Only fixes for 6.2 should be added during the
>> merge window.
>>
> 
> No it's not a fix, it's queued for the next -rc phase. I didn't know of the 
> limitation of linux-next only allowing fixes in the merge window.

Stephen asks for it every time. Now it is in separate email:

https://lore.kernel.org/linux-next/20221213120359.47fc3cad@canb.auug.org.au/

Previously it was in each linux-next announcement:
https://lore.kernel.org/linux-next/20220527164940.6ef5b25e@canb.auug.org.au/

> 
> Actually the patch is wrong. It shouldn't include "mediatek,mt8195-mmsys" as 
> fallback for mediatek,mt8195-vppsys[0,1].
> 
> I'll revert the whole series from v6.2-tmp/dts64 and v6.2-tmp/soc

Thanks. Please, drop all material for next cycle from linux-next.

Best regards,
Krzysztof
diff mbox series

Patch

diff --git a/Documentation/devicetree/bindings/arm/mediatek/mediatek,mmsys.yaml b/Documentation/devicetree/bindings/arm/mediatek/mediatek,mmsys.yaml
index 0711f1834fbd..493aa9e8d484 100644
--- a/Documentation/devicetree/bindings/arm/mediatek/mediatek,mmsys.yaml
+++ b/Documentation/devicetree/bindings/arm/mediatek/mediatek,mmsys.yaml
@@ -48,7 +48,10 @@  properties:
           - const: syscon
 
       - items:
-          - const: mediatek,mt8195-vdosys0
+          - enum:
+              - mediatek,mt8195-vdosys0
+              - mediatek,mt8195-vppsys0
+              - mediatek,mt8195-vppsys1
           - const: mediatek,mt8195-mmsys
           - const: syscon