Re: [PATCH RESEND v5 1/2] dt-binding: pinctrl: Add NPCM8XX pinctrl and GPIO documentation

[Krzysztof Kozlowski <krzysztof.kozlowski@xxxxxxxxxx>]

On 27/08/2023 17:30, Tomer Maimon wrote:
> Hi Krzysztof,
> 
> Thanks for your comment
> 
> On Sun, 27 Aug 2023 at 14:13, Krzysztof Kozlowski
> <krzysztof.kozlowski@xxxxxxxxxx> wrote:
>>
>> On 27/08/2023 11:44, Tomer Maimon wrote:
>>>>> +      pinctrl: pinctrl@f0800260 {
>>>>> +        compatible = "nuvoton,npcm845-pinctrl";
>>>>> +        ranges = <0x0 0x0 0xf0010000 0x8000>;
>>>>> +        #address-cells = <1>;
>>>>> +        #size-cells = <1>;
>>>>> +        nuvoton,sysgcr = <&gcr>;
>>>>> +
>>>>> +        gpio0: gpio@f0010000 {
>>>>
>>>> unit-address should be 0.
>>>>
>>>> Otherwise,
>>> The unit-address is correct f0010000
>>
>> Then how does it pass W=1 builds? How unit address can be f0010000 but
>> reg is 0? Really...
> Maybe because the ranges are ranges = <0x0 0x0 0xf0010000 0x8000>?

And how does this mapping should cause the unit address to not match the
reg? What type of rule is it?

Except also incorrect address in pinctrl node.. but your DTS
nuvoton-common-npcm8xx.dtsi has so many other bugs (duplicated nodes,
not matching, unit addresses), that I don't wonder that you do not see
other errors. But that's not a reason to add more. Rob gave you quite
specific advice, so I really do not understand why do you keep arguing
with it?


> I didn't get any warning regarding the unit-address
> bash-4.2$ make ARCH=arm64 dt_binding_check W=1

DTS, not binding.

BTW, your patches have errors. Please fix them:
patch:226: new blank line at EOF.
warning: 1 line adds whitespace errors.


Best regards,
Krzysztof