Re: [PATCH 05/18] MIPS: DTS: jz4780: fix pinctrl as reported by dtbscheck

From: H. Nikolaus Schaller
Date: Sat Apr 09 2022 - 09:23:07 EST




> Am 09.04.2022 um 15:13 schrieb Krzysztof Kozlowski <krzysztof.kozlowski@xxxxxxxxxx>:
>
> On 09/04/2022 15:04, H. Nikolaus Schaller wrote:
>>
>>
>>> Am 09.04.2022 um 13:13 schrieb Krzysztof Kozlowski <krzysztof.kozlowski@xxxxxxxxxx>:
>>>
>>> On 08/04/2022 20:37, H. Nikolaus Schaller wrote:
>>>> arch/mips/boot/dts/ingenic/ci20.dtb: pin-controller@10010000: $nodename:0: 'pin-controller@10010000' does not match '^(pinctrl|pinmux)(@[0-9a-f]+)?$'
>>>> From schema: Documentation/devicetree/bindings/pinctrl/ingenic,pinctrl.yaml
>>>>
>>>> Signed-off-by: H. Nikolaus Schaller <hns@xxxxxxxxxxxxx>
>>>> ---
>>>> arch/mips/boot/dts/ingenic/jz4780.dtsi | 2 +-
>>>> 1 file changed, 1 insertion(+), 1 deletion(-)
>>>>
>>>> diff --git a/arch/mips/boot/dts/ingenic/jz4780.dtsi b/arch/mips/boot/dts/ingenic/jz4780.dtsi
>>>> index 5f44cf004d473..b5299eaffb84a 100644
>>>> --- a/arch/mips/boot/dts/ingenic/jz4780.dtsi
>>>> +++ b/arch/mips/boot/dts/ingenic/jz4780.dtsi
>>>> @@ -155,7 +155,7 @@ rtc_dev: rtc@10003000 {
>>>> clock-names = "rtc";
>>>> };
>>>>
>>>> - pinctrl: pin-controller@10010000 {
>>>> + pinctrl: pinctrl@10010000 {
>>>
>>> Do it once for all DTSes, not one file at a time. There are four more
>>> places with this.
>>
>> Well, automation has no notion of "similarity" in this case to
>> merge several patches.
>
> What does that mean? One cannot create multiple patches and apply them?

This patch set was created by some automatic scripts. And they produce one patch
per group of warnings.

But here you ask me to merge 4 unrelated topics into a single one.

Or do you mean something else?

>
>> And they are not related. Every one is based on a different .yaml
>> schema file.
>
> Which does not matter, because the name of the node does not matter. We
> enforce it in schema to makes things organized and easier in testing.
> This does not fix any real problem, just the problem we created by
> ourselves with schema.
>
>>
>> That in all cases the result looks similar comes from similar
>> requirements by the schemata and has no inherent connection.
>
> All schemas will require it, won't they? The same for arm...

We may be talking about different things here.

My understanding:
you ask me to merge 5/18, 8/18, 9/18, 12/18 because they contain "controller" in the node-name.

Right? If not then we must clarify that first.

For pinctrl it is not allowed to have a -controller suffix while for the other it is mandatory
by the schema pattern.

BR and thanks,
Nikolaus