[PATCH v3 12/14] dt-bindings: mtd: gpmi-nand: Fix matching of clocks on different SoCs

[Krzysztof Kozlowski]

On Tue, 8 Sep 2020 at 18:51, Rob Herring <robh+dt at kernel.org> wrote:
>
> On Mon, Sep 7, 2020 at 12:10 AM Krzysztof Kozlowski <krzk at kernel.org> wrote:
> >
> > On Fri, Sep 04, 2020 at 04:36:39PM -0600, Rob Herring wrote:
> > > On Fri, Sep 4, 2020 at 9:25 AM Krzysztof Kozlowski <krzk at kernel.org> wrote:
> > > >
> > > > Driver requires different amount of clocks for different SoCs.  Describe
> > > > these requirements properly to fix dtbs_check warnings like:
> > > >
> > > >     arch/arm64/boot/dts/freescale/imx8mm-beacon-kit.dt.yaml: nand-controller at 33002000: clock-names:1: 'gpmi_apb' was expected
> > > >
> > > > Signed-off-by: Krzysztof Kozlowski <krzk at kernel.org>
> > > >
> > > > ---
> > > >
> > > > Changes since v1:
> > > > 1. Do not require order of clocks (use pattern).
> > >
> > > To the extent that you can, you should fix the order in dts files
> > > first. If we just adjust the schemas to match the dts files, then
> > > what's the point?
> >
> > The DTSes do not have mixed order of clocks between each other, as fair
> > as I remember. It was fix after Sasha Hauer comment that order is not
> > necessarily good.
> >
> > We have the clock-names property, why enforcing the order?
>
> Because DT/OpenFirmware has always had a defined order for property
> values. '*-names' is just extra information.

Thanks for the explanation. There are few nonobvious requirements
about writing schema which seems many (including me) miss. It might be
a good topic for some conference. Too bad ELCE CFP ended some time
ago. :)

Best regards,
Krzysztof