Re: [PATCH v5 2/3] dt-bindings: cpufreq: qcom-cpufreq-nvmem: make cpr bindings optional

[Krzysztof Kozlowski]

On 08/02/2023 01:09, Christian Marangi wrote:
> On Wed, Feb 01, 2023 at 09:20:39AM +0100, Krzysztof Kozlowski wrote:
>> On 31/01/2023 16:18, Christian Marangi wrote:
>>> The qcom-cpufreq-nvmem driver supports 2 kind of devices:
>>> - pre-cpr that doesn't have power-domains and base everything on nvmem
>>>   cells and multiple named microvolt bindings.
>>>   Doesn't need required-opp binding in the opp nodes as they are only
>>>   used for genpd based devices.
>>> - cpr-based that require power-domain in the cpu nodes and use various
>>>   source to decide the correct voltage and freq
>>>   Require required-opp binding since they need to be linked to the
>>>   related opp-level.
>>>
>>> When the schema was introduced, it was wrongly set to always require these
>>> binding but this is not the case for pre-cpr devices.
>>>
>>> Make the power-domain and the required-opp optional and set them required
>>> only for qcs404 based devices.
>>>
>>> Fixes: ec24d1d55469 ("dt-bindings: opp: Convert qcom-nvmem-cpufreq to DT schema")
>>
>> Fixes go as first patches in the series.
>>
> 
> Hi,
> this is problematic. This documentation is a bit special.
> 
> v4 had this patch as first but this cause error with make
> dt_binding_check as the schema will be effectively empty (as it will
> have only if condition)
> 
> This is why I pushed v5 that swap this with the second patch and first
> add non conditional stuff to the schema and only with the second patch
> makes them conditional.
> 
> Any hint to handle this corner case? I'm having some diffiulties due to
> how special this is but we really need this fix since it's blocking the
> introduction of opp table for ipq806x and ipq807x (as the schema is
> currently flawed)

Let's then drop fixes tag, because it will only confuse any backporters.

Best regards,
Krzysztof