[PATCH 8/9] dt-bindings: rtc: stmp3xxx-rtc: convert to dtschema
Javier Carrasco
javier.carrasco.cruz at gmail.com
Tue Apr 9 02:22:02 PDT 2024
On 4/9/24 09:40, Krzysztof Kozlowski wrote:
> On 08/04/2024 17:53, Javier Carrasco wrote:
>> Convert existing binding to dtschema to support validation.
>>
>> The 'fsl,imx28-rtc' compatible is currently not supported, and it is
>> only referenced in this binding and in nxp/mxs/imx28.dtsi. Therefore,
>> that compatible has been dropped, which triggers a warning when testing
>> the DT against the new binding.
>
> Instead document missing compatibles and mention this in commit msg.
>
There is no driver that will match 'fsl,imx28-rtc', only
'fsl,stmp3xxx-rtc', so I am not sure how to document the missing
compatible in a sensible way. My first suggestion to account for
undocumented strings would be:
compatible:
oneOf:
- items:
- enum:
- fsl,imx23-rtc
- fsl,imx28-rtc
- const: fsl,stmp3xxx-rtc
- const: fsl,stmp3xxx-rtc
Any suggestions or improvements?
>>
>> There is another reference to fsl,stmp3xxx-rtc in nxp/mxs/imx23.dtsi,
>> where another unsupported compatible 'fsl,imx23-rtc' is used, and the
>> same problem would arise when testing the file against the new binding.
>
> Please write concise messages... you have to paragraphs about the same?
> What is the difference here?
>
The difference is that 'fsl,imx23-rtc' was not even mentioned in any
binding, and it can only be found in imx23.dtsi. 'fsl,imx28-rtc' was
indeed mentioned in the txt binding.
My understanding after your comment is that we should gather
undocumented compatibles and add them to the bindings they would belong
to,no matter if they are used anywhere or not. I added this one to the
suggestion above as well.
>>
>> Signed-off-by: Javier Carrasco <javier.carrasco.cruz at gmail.com>
>> ---
>> .../devicetree/bindings/rtc/fsl,stmp3xxx-rtc.yaml | 45 ++++++++++++++++++++++
>> .../devicetree/bindings/rtc/stmp3xxx-rtc.txt | 21 ----------
>> 2 files changed, 45 insertions(+), 21 deletions(-)
>>
>
>
> Best regards,
> Krzysztof
>
Best regards,
Javier Carrasco
More information about the linux-arm-kernel
mailing list