Re: [RFC PATCH] dt-bindings: iio: adc: use spi-peripheral-props.yaml

[Krzysztof Kozlowski <krzysztof.kozlowski@xxxxxxxxxx>]

On 19/07/2022 00:00, Rob Herring wrote:
> On Sat, Jul 16, 2022 at 07:26:04PM +0100, Jonathan Cameron wrote:
>> On Fri, 15 Jul 2022 11:53:02 +0200
>> Krzysztof Kozlowski <krzysztof.kozlowski@xxxxxxxxxx> wrote:
>>
>>> Instead of listing directly properties typical for SPI peripherals,
>>> reference the spi-peripheral-props.yaml schema.  This allows using all
>>> properties typical for SPI-connected devices, even these which device
>>> bindings author did not tried yet.
>>>
>>> Remove the spi-* properties which now come via spi-peripheral-props.yaml
>>> schema, except for the cases when device schema adds some constraints
>>> like maximum frequency.
>>>
>>> Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@xxxxxxxxxx>
>>>
>>> ---
>>>
>>> This is an RFC with only some files changed, as I am still not sure of
>>> benefits for typical case - device node has just spi-max-frequency and
>>> nothing more.  I still find useful to reference the schema, but maybe I
>>> am missing something?
>>>
>>> Before doing wide-tree cleanup like this, I would be happy to receive
>>> some feedback whether this makes sense.
>>
>> Hi Krzysztof,
>>
>> This has the side effect of allowing spi-cpol / spi-cpha for devices
>> where they weren't previously allowed by the binding.  A typical device
>> only supports a subset of combinations of those.
>>
>> I'm not clear whether these should always be allowed (e.g. allow for inverters
>> etc in the path) or whether we should be enforcing the "correct"
>> settings for devices assuming they are directly connected.
>>
>> Currently we have a bunch of bindings that are documenting the allowed
>> flexibility - including cases where only particular combinations of these
>> settings are allowed.
>>
>> So we could either:
>> 1) Note that we've been doing it wrong and the binding should not enforce
>>    these constraints so remove them.
> 
> I'd lean towards this.
> 
>> 2) Add explicit spi-cpol: false statements etc the drivers where they
>>    are not allowed.
> 
> 3) Drop spi-cpol / spi-cpha from spi-peripheral-props.yaml. It's purpose 
> is to collect all possible SPI controller properties that are per child 
> node. Whereas we've said spi-cpol / spi-cpha are device specific 
> properties.

Thanks Rob and Jonathan. I can go with (3).

Best regards,
Krzysztof