On Mon, Oct 12, 2026 at 04:41:04PM +0800, Yongxing Mou wrote:
> 
> 
> On 10/5/2026 9:02 AM, Dmitry Baryshkov wrote:
> > On Tue, Sep 15, 2026 at 05:27:48PM +0800, Yongxing Mou wrote:
> > > Kaanapali (SM8850) has an additional INT2 GDSC that powers a subset
> > > of the DPU source pipes (VIG2/VIG3/DMA5), on top of the CORE_GDSC
> > > that powers the rest of the MDSS hardware.
> > > 
> > > Describe INT2 GDSC as an additional power domain of the DPU device
> > > rather than of the whole MDSS, since it only supplies DPU-local
> > > source pipes. Restrict the schema change to the Kaanapali DPU
> > > compatible via an if/then, so other platforms sharing
> > > qcom,sm8650-dpu.yaml (which don't have an INT2 GDSC) keep the
> > > existing single power-domain constraint.
> > > 
> > > Signed-off-by: Yongxing Mou <[email protected]>
> > > ---
> > >   .../bindings/display/msm/dpu-common.yaml           | 19 
> > > ++++++++++++++++--
> > >   .../bindings/display/msm/qcom,sm8650-dpu.yaml      | 23 
> > > ++++++++++++++++++++++
> > >   2 files changed, 40 insertions(+), 2 deletions(-)
> > > 
> > > diff --git 
> > > a/Documentation/devicetree/bindings/display/msm/dpu-common.yaml 
> > > b/Documentation/devicetree/bindings/display/msm/dpu-common.yaml
> > > index 3f953aa5e694..19b791a857b5 100644
> > > --- a/Documentation/devicetree/bindings/display/msm/dpu-common.yaml
> > > +++ b/Documentation/devicetree/bindings/display/msm/dpu-common.yaml
> > > @@ -25,8 +25,7 @@ properties:
> > >     interrupts:
> > >       maxItems: 1
> > > -  power-domains:
> > > -    maxItems: 1
> > > +  power-domains: true
> > >     operating-points-v2: true
> > >     opp-table:
> > > @@ -53,4 +52,20 @@ required:
> > >     - operating-points-v2
> > >     - ports
> > > +allOf:
> > > +  - if:
> > > +      properties:
> > > +        compatible:
> > > +          contains:
> > > +            const: qcom,kaanapali-dpu
> > > +    then:
> > > +      properties:
> > > +        power-domains:
> > > +          minItems: 2
> > > +          maxItems: 2
> > > +    else:
> > > +      properties:
> > > +        power-domains:
> > > +          maxItems: 1
> > > +
> > >   additionalProperties: true
> > > diff --git 
> > > a/Documentation/devicetree/bindings/display/msm/qcom,sm8650-dpu.yaml 
> > > b/Documentation/devicetree/bindings/display/msm/qcom,sm8650-dpu.yaml
> > > index aa0cf0ec5b93..88aa71d2b912 100644
> > > --- a/Documentation/devicetree/bindings/display/msm/qcom,sm8650-dpu.yaml
> > > +++ b/Documentation/devicetree/bindings/display/msm/qcom,sm8650-dpu.yaml
> > > @@ -61,6 +61,29 @@ required:
> > >     - clocks
> > >     - clock-names
> > > +allOf:
> > > +  - if:
> > > +      properties:
> > > +        compatible:
> > > +          contains:
> > > +            const: qcom,kaanapali-dpu
> > > +    then:
> > > +      properties:
> > > +        power-domains:
> > > +          items:
> > > +            - description: MMCX power domain
> > > +            - description: INT2 GDSC power domain
> > > +
> > > +        power-domain-names:
> > > +          minItems: 2
> > > +          maxItems: 2
> > > +          items:
> > > +            - const: mmcx
> > > +            - const: int2
> > 
> > Nit: I'd suggest more generic names here. If the next platform switches
> > from "mmcx" to some other domain, while keeping the split architecture,
> > the name becomes unapplicable (in the past we have seen platforms using
> > CX instead of MMCX). What about using names like "core", "int2"?
> > 
> Agreed, using "mmcx" would be too provider-specific, as the DPU performance
> domain can be CX on other platforms.
> 
> However, "core" might be confusing here because DISP_CC_MDSS_CORE_GDSC is
> already attached to the parent MDSS device. The first domain on the DPU is
> used for OPP performance-state voting, so would "perf" and "int2" be clearer
> generic names?

Responded on the other patch. I think that here we should be listing two
dispcc domains and then "core" and "int2" would be proper names for
them.

> > > +
> > > +      required:
> > > +        - power-domain-names
> > > +
> > >   unevaluatedProperties: false
> > >   examples:
> > > 
> > > -- 
> > > 2.43.0
> > > 
> > 
> 

-- 
With best wishes
Dmitry

Reply via email to