Re: [RFC PATCH 01/18] dt-bindings: net: wireless: describe the ath12k AHB module

[Kalle Valo <kvalo@kernel.org>]

Krzysztof Kozlowski <krzk@kernel.org> writes:

> On 04/09/2024 13:28, Raj Kumar Bhagat wrote:
>
>> On 9/4/2024 1:08 PM, Krzysztof Kozlowski wrote:
>>> On 04/09/2024 09:23, Raj Kumar Bhagat wrote:
>>>>>> +  qcom,board_id:
>>>>>> +    $ref: /schemas/types.yaml#/definitions/uint32
>>>>>> +    description:
>>>>>> +      Board id value, it identifies the variant of ath12k WiFi device.
>>>>>
>>>>> Please do not upstream whatever you have in your vendor tree... You even
>>>>> copied its style! There is no such property in upstream.
>>>>>
>>>>> NAK, you are duplicating compatibles.
>>>>>
>>>>>
>>>>
>>>> qcom,board_id is a new parameter introduced to distinguish between different variants
>>>> of ath12k devices that share the same compatible value. This differentiation is crucial
>>>> as these variants have different qcom,board_id values (with the same compatible) based
>>>
>>> You must explain why compatible cannot be used for that.
>>>
>> 
>> The compatible property is used to define the variant of the Ath12k device. However, the
>> Ath12k device variant does not change with different capabilities or configurations
>> (e.g., operating band). The qcom,board_id parameter is introduced to define these
>> configurations, not the device variant itself.
>
> operating aspects are run-time, so not suitable here. It's all still
> vague for me.

Yeah, this is weird and I would hope there is a better way to do it.
Unfortunately I'm not able to study this in detail right now but, like
you said, most likely this is just directly copied from QSDK (their
downstream).

-- 
https://patchwork.kernel.org/project/linux-wireless/list/

https://wireless.wiki.kernel.org/en/developers/documentation/submittingpatches