Re: [PATCH v3 1/6] dt-bindings: crypto: qcom,inline-crypto-engine: Fix legacy/new SoC strictness split

[Kuldeep Singh]

On 08-07-2026 12:13, Krzysztof Kozlowski wrote:
> On Mon, Jul 06, 2026 at 05:01:29PM +0530, Kuldeep Singh wrote:
>> Couple of already merged SoCs(like sc7280, sm8750, kaanapali etc.)
>> describe ICE as single clock historically which are recently updated
>> with mandatory 2 clocks.
>>
>> Keep only the known legacy compatibles flexible, and make strict
>> validation default(of power-domains and 2 clocks) for all other Soc
>> compatibles.
>>
>> This ensures old DTs are valid while ensuring any new SoC (like hawi,
>> milos, eliza) must follow latest requirements by default.
> 
> To re-iterate: You change the ABI for Hawi, this must be expressed and
> explained why.

Ohh I see, hawi is different case(compared to milos/eliza) where
bindings define the compatible but somehow was relaxed(got missed
probably) from 2 clocks recommendation.
Just hawi compatible is defined but no DT entry is present actually.

> I do not see any change in commit msg (listing "new SoC"
> is not what I meant is not relevant here - it even suggests like
> everything is here done without impact).

I'll update hawi as separate case in commit message to reflect properly.

-- 
Regards
Kuldeep