On 27/06/2025 10:54, André Draszik wrote:
> Hi Krzysztof,
>
> On Thu, 2025-06-26 at 21:49 +0200, Krzysztof Kozlowski wrote:
>> On 24/05/2025 07:21, André Draszik wrote:
>>> +
>>> + gpio {
>>> + compatible = "maxim,max77759-gpio";
>>> +
>>> + gpio-controller;
>>> + #gpio-cells = <2>;
>>> + /*
>>> + * "Human-readable name [SIGNAL_LABEL]" where the
>>> + * latter comes from the schematic
>>> + */
>>> + gpio-line-names = "OTG boost [OTG_BOOST_EN]",
>>> + "max20339 IRQ [MW_OVP_INT_L]";
>>> +
>>> + interrupt-controller;
>>> + #interrupt-cells = <2>;
>>> + };
>>> +
>>> + nvmem-0 {
>>
>> Why is this called nvmem-0, not nvmem? Is there nvmem-1? I see binding
>> does it, but why?
>
> 'nvmem' is used/declared by nvmem-consumer.yaml as a phandle array
> already so using just 'nvmem' fails validation:
>
> Documentation/devicetree/bindings/mfd/maxim,max77759.example.dtb: pmic@66: nvmem: {'compatible': ['maxim,max77759-nvmem'], 'nvmem-
> layout': {'compatible': ['fixed-layout'], '#address-cells': 1, '#size-cells': 1, 'reboot-mode@0': {'reg': [[0, 4]]}, 'boot-reason@4':
> {'reg': [[4, 4]]}, 'shutdown-user-flag@8': {'reg': [[8, 1]]}, 'rsoc@10': {'reg': [[10, 2]]}}} is not of type 'array'
> from schema $id: http://devicetree.org/schemas/nvmem/nvmem-consumer.yaml#
>
> https://lore.kernel.org/all/20250226-max77759-mfd-v2-3-a65ebe2bc0a9@xxxxxxxxxx/
Heh, this should have been just folded into the parent as Rob suggested
during v2. Well, let's just merge it.
Best regards,
Krzysztof
