Re: [PATCH RFC 1/2] regulator: dt-bindings: raa215300: add clock output

[Josua Mayer]

Am 03.05.26 um 02:57 schrieb Mark Brown:
> On Sat, May 02, 2026 at 06:07:04PM +0200, Josua Mayer wrote:
>
>> +  "#clock-cells":
>> +    const: 0
>> +
>> +  clock-output-names:
>> +    maxItems: 1
>> +
>>    clocks:
>>      description: |
>>        The clocks are optional. The RTC is disabled, if no clocks are
> Should there be a requirement for an input clock if a clock output is
> specified?

The input clock is same as for RTC, so I suppose yes.

In that case, which property should the dependency be based on?
#clock-cells ?
clock-output-names?

clock-output-names is optional, so I think #clock-cells is the only option.

Is it important to model this dependency?
Or is it sufficient that driver returns an error when clock is missing?