Re: [PATCH v3 1/3] dt-bindings: media: camss: Add qcom,sm6350-camss

[Krzysztof Kozlowski <krzk@kernel.org>]

On 13/02/2026 14:15, Luca Weiss wrote:
> Add bindings for the Camera Subsystem on the SM6350 SoC.
> 
> Signed-off-by: Luca Weiss <luca.weiss@fairphone.com>
> ---
>  .../bindings/media/qcom,sm6350-camss.yaml          | 471 +++++++++++++++++++++
>  1 file changed, 471 insertions(+)

Just to be clear, because Bryan posted now x1e80100 camss patches
changing ABI. I was holding this patchset for camss/media maintainers to
come up with driver and any other necessary pieces (like guidance for
bindings) support for split CSI PHY. Finally after long time I gave up
and gave these bindings green light.

Additionally there is some claim that there is policy allowing to change
bindings. No. There is no change of bindings with few exceptions. Decide
now what bindings you want to have. One single node or split CSI PHY,
because you cannot take the first option but then six months later send
complete revamp of the ABI.

So please decide and answer, e.g. by taking this patch, that you
understand above and THIS will be your fixed ABI.

P.S. It is second time I am battling with fake policies invented around
camss. Previous was that supposedly we ask to order the
reg/interrupts/clocks by value. That was completely fake rule debunked
later when DT maintainers found out that you created such. Now another
rule that you can change a released ABI with users just calling them "no
upstream DTS users". I think this drivers/subsystem is somehow the most
fake-rule inventing and I never had to correct other folks so many
times. I don't know why.

Best regards,
Krzysztof