Re: [PATCH v13 5/5] dt-bindings: media: qcom,x1e80100-camss: Make vdd-csiphy supplies optional

On 28/07/2026 10:42, [email protected] wrote:
> [Severity: Medium]
> Since the CSIPHY nodes are being moved out to peer nodes, do other properties
> in this schema also need to be updated to prevent validation regressions?
> 
> For example, the reg-names, clock-names, and interrupt-names in this schema
> are currently defined as strict, fixed-order arrays that expect csiphy elements
> at specific indices. If those elements are removed from the CAMSS node to avoid
> resource conflicts, won't the remaining elements shift and cause
> dt_binding_check to fail?
> 
> Additionally, the ports schema mandates data-lanes on the CAMSS endpoints.
> Is this requirement still correct for internal links between the CAMSS core
> and a peer PHY, rather than physical lanes to the sensor?

A good guess but, no - illustrative of how LLMs "understand" stuff but 
actually don't understand stuff.

If the graph lookup and probe succeeds then the legacy regs/clocks are 
never mapped by camss - if that lookup _fails_ then you absolutely need 
the legacy stuff.

"Dr Chandra will I dream" ?

"I don't know"

---
bod