Re: [PATCH 1/3] dt-bindings: mtd: atmel-nand: convert txt to yaml

From: Balamanikandan.Gunasundar
Date: Mon Mar 25 2024 - 08:12:45 EST


On 22/03/24 1:00 pm, Conor Dooley wrote:
> On Fri, Mar 22, 2024 at 04:27:29AM +0000, Balamanikandan.Gunasundar@xxxxxxxxxxxxx wrote:
>> On 20/03/24 10:05 pm, Conor Dooley wrote:
>>> On Wed, Mar 20, 2024 at 11:22:07AM +0530, Balamanikandan Gunasundar wrote:
>
>>>> +allOf:
>>>> + - if:
>>>> + properties:
>>>> + compatible:
>>>> + contains:
>>>> + enum:
>>>> + - atmel,at91rm9200-nand-controller
>>>> + - atmel,at91sam9260-nand-controller
>>>> + - atmel,at91sam9261-nand-controller
>>>> + - atmel,at91sam9g45-nand-controller
>>>> + - atmel,sama5d3-nand-controller
>>>> + - microchip,sam9x60-nand-controller
>>>> + then:
>>>> + properties:
>>>> + "#address-cells":
>>>> + const: 2
>>>> +
>>>> + "#size-cells":
>>>> + const: 1
>>> Why is this in an if? Isn't this all of the devices in the binding?
>>>
>>
>> The default nand-controller.yaml defines this as const values.
>> (#address-cell : 1 and #size-cells : 1). I am trying to override this
>> const value.
>
> You're not overriding anything as you don't have a ref to
> nand-controller.yaml in this file, AFAICT. Why don't you?
>
>> May be I should think about better approach ?
>
> You should be able to apply this unconditionally for this file. I don't
> see why the if would be needed?
>
>
>>>> +patternProperties:
>>>> + "^nand@[a-f0-9]$":
>>>> + type: object
>>>> + $ref: nand-chip.yaml#
>>>> + description:
>>>> + NAND chip bindings. All generic properties described in
>>>> + Documentation/devicetree/bindings/mtd/{common,nand}.txt also apply to
>>>> + the NAND device node, and NAND partitions should be defined under the
>>>> + NAND node as described in
>>>> + Documentation/devicetree/bindings/mtd/partition.txt.
>>> These files do not exist.
>>>
>>
>> Apologies for copying the content from the text file. I will correct this.
>
> You don't need these comments at all I think. You have the ref to
> nand-chip.yaml, so at least the first text file reference can be
> removed.
>

Agree with this. I will remove it.

>> Yes. I should fix the alignment. I will send a v2 shortly
>
> I did make other comments, so I assume you agree with everything else I
> mentioned and will implement them in v2.

Yes. I agree with other comments as well. I will also address all the
comments with the other 2 patches and send a v2. Thanks for reviewing.

Regards,
Bala.

>
> Thanks,
> Conor.