Re: [PATCH v3 2/6] dt-bindings: media: camss: Add qcom,kaanapali-camss binding

[Vijay Kumar Tumati]

On 10/28/2025 9:56 AM, Krzysztof Kozlowski wrote:
> On 28/10/2025 16:22, Vijay Kumar Tumati wrote:
> > > > +  interconnects:
> > > > +    maxItems: 2
> > > > +
> > > > +  interconnect-names:
> > > > +    items:
> > > > +      - const: ahb
> > > > +      - const: hf_mnoc
> > > Why previously this was called hf_0 but now hf?
> > Hi Krzysztof, the interconnect driver exposes only one node 'camnoc_hf'
> Drivers don't matter. Interconnect driver does not matter here. You
> describe this hardware, not interconnect driver.
>
> Keep it consistent with previous devices, whichever these are.
There has been a mixed usage in the previous devices but this (hf) is 
consistent with the recently added ones, like x1e80100 and 2290.
> > to the camera driver, with it internally managing the voting on hf_0 and
> > hf_1 clients. The traffic from the Real Time blocks in camera go through
> > both HF_0 and HF_1, with the former being the primary. This change
> > correctly represents that the BW voting is for the whole of the HF
> > client. Please let me know if you have any further questions and we
> > would be happy to answer. Thank you.
> > > > +
> > > > +  iommus:
> > > > +    maxItems: 1
> > > > +
> > > > +  power-domains:
> > > > +    items:
> > > > +      - description:
> > > > +          TFE0 GDSC - Thin Front End, Global Distributed Switch Controller.
> > > > +      - description:
> > > > +          TFE1 GDSC - Thin Front End, Global Distributed Switch Controller.
> > > > +      - description:
> > > > +          TFE2 GDSC - Thin Front End, Global Distributed Switch Controller.
> > > > +      - description:
> > > > +          Titan GDSC - Titan ISP Block Global Distributed Switch Controller.
> > > > +
> > > > +  power-domain-names:
> > > > +    items:
> > > > +      - const: tfe0
> > > > +      - const: tfe1
> > > > +      - const: tfe2
> > > Why not using the same names as before? It really does not matter that
> > > it is thin or image, all of them are the same because only the
> > > difference against top matters.
> > Right, this is done to maintain the consistency with the clock driver on
> Sorry, this makes no sense. This device has nothing to do with clock
> driver. Don't ever use clock drivers as arguments for doing something in
> completely different place.
>
> Not mentioning that drivers don't matter much for the bindings, so I
> really do not get what you try to explain here.

Understood. I meant to say that it is consistent with the naming for the 
TFE device that is available on Kaanapali. If our intention is to keep 
the names in the bindings same as previous generations despite the 
changing HW architectures, we could change these to IFEs, to be 
consistent with previous  generations. Please advise. Appreciate your 
inputs here.

> Best regards,
> Krzysztof