[PATCH v3 1/6] dt-bindings: arm: qcom: Document Microsoft Surface Pro 11

[Konrad Dybcio]

On 12/18/25 5:35 PM, Jérôme de Bretagne wrote:
> Le jeu. 18 déc. 2025 à 15:15, Konrad Dybcio
> <konrad.dybcio at oss.qualcomm.com> a écrit :
>>
>> On 12/18/25 12:56 AM, Jérôme de Bretagne via B4 Relay wrote:
>>> From: Jérôme de Bretagne <jerome.debretagne at gmail.com>
>>>
>>> Add the compatibles for the Qualcomm-based Microsoft Surface Pro 11,
>>> using its Denali codename.
>>>
>>> The LCD models are using the Qualcomm Snapdragon X1 Plus (X1P64100),
>>> the OLED ones are using the Qualcomm Snapdragon X1 Elite (X1E80100).
>>>
>>> Due to the difference in how the built-in panel is being handled
>>> between the OLED variant and LCD one, it is required to have two
>>> separate DTBs, so document the compatible string for both variants.
>>>
>>> Signed-off-by: Jérôme de Bretagne <jerome.debretagne at gmail.com>
>>> ---
>>>  Documentation/devicetree/bindings/arm/qcom.yaml | 8 ++++++++
>>>  1 file changed, 8 insertions(+)
>>>
>>> diff --git a/Documentation/devicetree/bindings/arm/qcom.yaml b/Documentation/devicetree/bindings/arm/qcom.yaml
>>> index d84bd3bca2010508a8225b9549d8c634efa06531..7c99bc0d3aae3dc6e9c08fef0a535e114a3297a8 100644
>>> --- a/Documentation/devicetree/bindings/arm/qcom.yaml
>>> +++ b/Documentation/devicetree/bindings/arm/qcom.yaml
>>> @@ -1067,6 +1067,14 @@ properties:
>>>            - const: qcom,x1e78100
>>>            - const: qcom,x1e80100
>>>
>>> +      - items:
>>> +          - enum:
>>> +              - microsoft,denali-lcd
>>> +              - microsoft,denali-oled
>>> +          - const: microsoft,denali
>>> +          - const: qcom,x1p64100
>>> +          - const: qcom,x1e80100
>>
>> As the bot pointed out, this is valid, but not what you want:
>>
>> This expects
>>
>> "microsoft,denali-(lcd/oled)", "microsoft,denali", "qcom,x1p64100",
>> "qcom,x1e80100"
>>
>> whereas you're looking for 2 entries:
>>
>> - items:
>>         const: microsoft,denali-lcd
>>         const: microsoft,denali
>>         const: qcom,x1p64100
>>         const: qcom,x1e80100
>>
>> - items:
>>         const: microsoft,denali-oled
>>         const: microsoft,denali
>>         const: qcom,x1e80100
> 
> I was trying to group the 2 variants together, as they are very much
> related. Your description captures the 2 variants way more precisely,
> I will switch to 2 distinct entries in v4.

Unfortunately, in math terms, this is a:

ab + ac + bc

situation, as in you see there's a lot of repetition, but there's no way
to find a common neat factor for every entry

Konrad