Re: [PATCH v2 1/1] arm64: dts: qcom: monaco-evk: Add Interface Plus Mezzanine

[Krzysztof Kozlowski <krzk@kernel.org>]

On 27/02/2026 10:50, Umang Chheda wrote:
> Hello Krzysztof,
> 
> On 2/24/2026 3:39 AM, Dmitry Baryshkov wrote:
>> On Mon, Feb 23, 2026 at 09:37:53PM +0100, Krzysztof Kozlowski wrote:
>>> On 23/02/2026 20:02, Dmitry Baryshkov wrote:
>>>>>> So I presume what you're saying is that we should at most declare one
>>>>>> level of non-controlled fixed regulators?
>>>>> In general, non-controller fixed regulators should not be there at all,
>>>>> except when they serve certain purpose, like fulfill the binding
>>>>> requirement. It's their only point.
>>>>>
>>>>> And a chain of:
>>>>>
>>>>> A -> B -> C -> device
>>>>>
>>>>> is completely redundant if all A+B+C are non-controlled.
>>>> I think that came from me. I don't consider that to be completely
>>>> redundant. It helps in reviews and in some understanding of the board
>>>> logic. I'm not asking to implement all the intermediate regulators, but
>>>> to implement the meaningful relationship between end-user regulators.
>>> These are not end-user regulators. These are fixed things which no one
>>> touches and no one needs. There is no single purpose for user-space to
>>> see them.
>>>
>>> Why do you not insist on defining all of such external oscilators, rest
>>> of regulators, all possible little ICs?
>> So, where is the boundary from you point of view? Do we define fixed
>> regulators powering DRM bridges / USB hubs and other similar devices?
>> Or do we do it only if the bindings require us to do it?
> 
> Can you help share your point of view on the above query from Dmitry ? In this case to adhere to  bindings requirements

I already replied in this thread. No point to repeat in multiple places.

Best regards,
Krzysztof