Re: [PATCH 1/6] dt-bindings: phy: qcom,hdmi-phy-other: use pxo clock

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On 09/09/2022 22:29, Dmitry Baryshkov wrote:
> 
> 
> On 9 September 2022 19:44:11 GMT+03:00, Krzysztof Kozlowski <krzysztof.kozlowski@xxxxxxxxxx> wrote:
>> On 09/09/2022 17:03, Dmitry Baryshkov wrote:
>>>>> +    then:
>>>>> +      properties:
>>>>> +        clocks:
>>>>> +          minItems: 1
>>>>> +          maxItems: 2
>>>>> +        clock-names:
>>>>> +          minItems: 1
>>>>> +          items:
>>>>> +            - const: slave_iface
>>>>> +            - const: pxo
>>>>
>>>> Why pxo is optional? Commit msg does not say much here.
>>>
>>> It's optional as it is not present in current DT files. The driver will 
>>> fallback to 'pxo_board' if the clock is not present.
>>>
>>>> It seems you also miss the DTS change adding the clock.
>>>
>>> Oh, I'll add it to v2.
>>
>> How about adding it to DTS and making it required in the bindings? I did
>> not check the driver, but isn't the driver fail if clock is missing thus
>> the clock is really required?
> 
> I had the impression that we cannot make a clock mandatory of it wasn't present before. Please correct me if I'm wrong.

We cannot break the ABI which means implementation must accept old DTS.
As you wrote below, the implementation will handle this case. Whether we
can require new DTS properties (if implementation respects ABI) is a
different problem and I believe that we can. Bindings can grow, even
with necessary changes, because no ABI is actually broken here.

> 
> No, the driver will not fail. It will fallback to the lookup of the `pxo_board' clock from the system clock list.

Ah, good!


Best regards,
Krzysztof



[Index of Archives]     [Device Tree Compilter]     [Device Tree Spec]     [Linux Driver Backports]     [Video for Linux]     [Linux USB Devel]     [Linux PCI Devel]     [Linux Audio Users]     [Linux Kernel]     [Linux SCSI]     [XFree86]     [Yosemite Backpacking]


  Powered by Linux