Re: [PATCH 13/18] dt-bindings: fix jz4780-nemc issue as reported by dtbscheck

From: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
To: "H. Nikolaus Schaller" <hns@goldelico.com>
Cc: Rob Herring <robh+dt@kernel.org>,
	Paul Cercueil <paul@crapouillou.net>,
	Thomas Bogendoerfer <tsbogend@alpha.franken.de>,
	linux-kernel@vger.kernel.org, devicetree@vger.kernel.org,
	linux-mips@vger.kernel.org, letux-kernel@openphoenux.org
Subject: Re: [PATCH 13/18] dt-bindings: fix jz4780-nemc issue as reported by dtbscheck
Date: Sat, 9 Apr 2022 15:18:23 +0200	[thread overview]
Message-ID: <99a97b0f-19a5-136d-9160-c9fde6f3548c@linaro.org> (raw)
In-Reply-To: <B9FD64FE-82B0-4DC2-B4C3-BE266DAB28A0@goldelico.com>

On 09/04/2022 15:09, H. Nikolaus Schaller wrote:
> 
> 
>> Am 09.04.2022 um 13:26 schrieb Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>:
>>
>> On 08/04/2022 20:37, H. Nikolaus Schaller wrote:
>>> jz4780-nemc needs to be compatible to simple-mfd as well or we get
>>>
>>> arch/mips/boot/dts/ingenic/ci20.dtb: memory-controller@13410000: compatible: 'oneOf' conditional failed, one must be fixed:
>>> 	['ingenic,jz4780-nemc', 'simple-mfd'] is too long
>>> 	'ingenic,jz4725b-nemc' was expected
>>> 	'ingenic,jz4740-nemc' was expected
>>> 	From schema: Documentation/devicetree/bindings/memory-controllers/ingenic,nemc.yaml
>>>
>>> Signed-off-by: H. Nikolaus Schaller <hns@goldelico.com>
>>> ---
>>> .../devicetree/bindings/memory-controllers/ingenic,nemc.yaml    | 2 +-
>>> 1 file changed, 1 insertion(+), 1 deletion(-)
>>>
>>> diff --git a/Documentation/devicetree/bindings/memory-controllers/ingenic,nemc.yaml b/Documentation/devicetree/bindings/memory-controllers/ingenic,nemc.yaml
>>> index 24f9e19820282..3b1116588de3d 100644
>>> --- a/Documentation/devicetree/bindings/memory-controllers/ingenic,nemc.yaml
>>> +++ b/Documentation/devicetree/bindings/memory-controllers/ingenic,nemc.yaml
>>> @@ -17,7 +17,7 @@ properties:
>>>     oneOf:
>>>       - enum:
>>>           - 
>>> -          - ingenic,jz4780-nemc
>>> +          - [ , simple-mfd ]
>>
>> This is not correct representation. If you really need simple-mfd, then
>> this should be a separate item below oneOf.
> 
> Well, it is valid YAML syntax and seems to be accepted by dtbscheck.

It's not how we code it. Please do not introduce inconsistent - even if
valid - blocks.

> 
>> The true question is whether you need simple-mfd. Isn't the binding (and
>> the driver) expected to instantiate its children?
> 
> I had expected that but current ingenic,jz4780-nemc code doesn't.

Paul provided good reason for the simple-mfd. Use this one instead of dt
check warning. DT check warning means nothing, does not bring the actual
answer to "why", because it is artificial tool. The answer to "why" is
in what Paul wrote.

Best regards,
Krzysztof