Re: [PATCH 1/1] dt-bindings: mmc: sdhci-msm: Fix issues in yaml bindings

[Dmitry Baryshkov]

On 2 July 2022 18:22:19 GMT+03:00, Rob Herring <robh@xxxxxxxxxx> wrote:
>On Sat, Jul 2, 2022 at 3:14 AM Dmitry Baryshkov
><dmitry.baryshkov@xxxxxxxxxx> wrote:
>>
>>
>>
>> On 2 July 2022 01:06:48 GMT+03:00, Rob Herring <robh@xxxxxxxxxx> wrote:
>> >On Sat, May 14, 2022 at 4:01 PM Bhupesh Sharma
>> ><bhupesh.sharma@xxxxxxxxxx> wrote:
>> >>
>> >> Rob pointed some remaining issues in the sdhci-msm yaml
>> >> bindings (via [1]).
>> >>
>> >> Fix the same by first using the 'mmc-controller.yaml' as
>> >> 'ref' and thereafter also fix the issues reported by
>> >> 'make dtbs_check' check.
>> >>
>> >> [1]. https://lore.kernel.org/linux-arm-msm/YnLmNCwNfoqZln12@xxxxxxxxxxxxxxxxxx/
>> >>
>> >> Fixes: a45537723f4b ("dt-bindings: mmc: sdhci-msm: Convert bindings to yaml")
>> >> Cc: Bjorn Andersson <bjorn.andersson@xxxxxxxxxx>
>> >> Cc: Rob Herring <robh@xxxxxxxxxx>
>> >> Cc: Ulf Hansson <ulf.hansson@xxxxxxxxxx>
>> >> Signed-off-by: Bhupesh Sharma <bhupesh.sharma@xxxxxxxxxx>
>> >> ---
>> >> -> This patch uses the dts changes sent (here: https://lore.kernel.org/linux-arm-msm/20220514215424.1007718-1-bhupesh.sharma@xxxxxxxxxx/), for fixing the dtbs_check errors.
>> >> -> This patch is rebased on 'linux-next/master'
>> >>
>> >>  .../devicetree/bindings/mmc/sdhci-msm.yaml    | 52 ++++++++++++++++---
>> >>  1 file changed, 44 insertions(+), 8 deletions(-)
>> >
>> >There's another issue with this applied:
>> >
>> >Documentation/devicetree/bindings/mmc/sdhci-msm.example.dtb:
>> >mmc@8804000: Unevaluated properties are not allowed
>> >('operating-points-v2' was unexpected)
>> >
>> >Should just need a 'operating-points-v2: true' line.
>> >
>> >This won't show up until a fix for 'unevaluatedProperties' handling is
>> >applied. But first I need all the issues fixed.
>>
>> Could you please add a dt-validate (?) argument so that we can validate new schemas with unevaluatedProperties working as expected, while keeping default behaviour intact (while it gets sorted out)?
>>
>
>I think that wouldn't work well because the schemas have to be
>reprocessed when such an option changes. Though kbuild does look for
>command line changes...
>
>In any case, I'm going to commit this to the main branch in a few
>days. There aren't many warnings left.

Ack, thanks for the explanation.


-- 
With best wishes
Dmitry