[PATCH 1/9] dt-bindings: pwm: marvell,pxa: add optional property resets

[Alex Elder]

On 4/15/25 5:12 AM, Yixun Lan wrote:
> Hi Philipp,
> 
> On 17:54 Tue 15 Apr     , Guodong Xu wrote:
>> On Tue, Apr 15, 2025 at 4:53 PM Philipp Zabel <p.zabel at pengutronix.de> wrote:
>>>
>>> On Fr, 2025-04-11 at 17:44 +0100, Conor Dooley wrote:
>>>> On Fri, Apr 11, 2025 at 09:14:15PM +0800, Guodong Xu wrote:
>>>>> Add an optional resets property for the Marvell PWM PXA binding.
>>>>>
>>>>> Signed-off-by: Guodong Xu <guodong at riscstar.com>
>>>>> ---
>>>>>   Documentation/devicetree/bindings/pwm/marvell,pxa-pwm.yaml | 3 +++
>>>>>   1 file changed, 3 insertions(+)
>>>>>
>>>>> diff --git a/Documentation/devicetree/bindings/pwm/marvell,pxa-pwm.yaml b/Documentation/devicetree/bindings/pwm/marvell,pxa-pwm.yaml
>>>>> index 9ee1946dc2e1..9640d4b627c2 100644
>>>>> --- a/Documentation/devicetree/bindings/pwm/marvell,pxa-pwm.yaml
>>>>> +++ b/Documentation/devicetree/bindings/pwm/marvell,pxa-pwm.yaml
>>>>> @@ -31,6 +31,9 @@ properties:
>>>>>     clocks:
>>>>>       maxItems: 1
>>>>>
>>>>> +  resets:
>>>>> +    maxItems: 1
>>>>
>>>> Do any of the currently supported devices use a reset? If not, then add
>>>> this in tandem with the new compatible and only allow it there please.
>>>
>>> Also, if spacemit,k1-pwm can not work without the reset being
>>> deasserted, mark it as required.
> 
> If I inerpret correctly, only reset requires explicitly being de-asserted,
> need to mark as required? that's being said, if reset comes out as de-asserted
>   by default after power reset, then not necessary?
> (in other cases, some device block is in asserted state by default)

We can often benefit from the state that the boot loader has left
things in, but I think it's better not to assume it if possible.
I suppose it might not be required though.

Anyway, the reset line is available to use; why not require it?

					-Alex

> thanks
>>>
>>
>> Thank you Philipp. spacemit,k1-pwm can not work without the reset.
>> I will add that in the next version.
>>
>> -Guodong
>>
>>> The driver can still use reset_control_get_optional.
>>>
>>> regards
>>> Philipp
>