On Fri, Dec 01, 2023 at 09:03:09AM +0100, Krzysztof Kozlowski wrote: > On 01/12/2023 05:37, Chen-Yu Tsai wrote: > > On Thu, Nov 30, 2023 at 4:52 PM Krzysztof Kozlowski > > <krzysztof.kozlowski@xxxxxxxxxx> wrote: > >> > >> On 30/11/2023 09:33, Chen-Yu Tsai wrote: > >>> "interrupts-extended" provides a more concise way of describing external > >>> GPIO interrupts. > >>> > >>> Allow using this instead of "interrupts" plus "interrupt-parent" for > >>> cros-ec. > >>> > >>> Signed-off-by: Chen-Yu Tsai <wenst@xxxxxxxxxxxx> > >>> --- > >>> .../devicetree/bindings/mfd/google,cros-ec.yaml | 9 ++++++++- > >>> 1 file changed, 8 insertions(+), 1 deletion(-) > >>> > >>> diff --git a/Documentation/devicetree/bindings/mfd/google,cros-ec.yaml b/Documentation/devicetree/bindings/mfd/google,cros-ec.yaml > >>> index e1ca4f297c6d..e514eac9f4fc 100644 > >>> --- a/Documentation/devicetree/bindings/mfd/google,cros-ec.yaml > >>> +++ b/Documentation/devicetree/bindings/mfd/google,cros-ec.yaml > >>> @@ -75,6 +75,9 @@ properties: > >>> interrupts: > >>> maxItems: 1 > >>> > >>> + interrupts-extended: > >>> + maxItems: 1 > >>> + > >> > >> I don't understand why you need it. You already have interrupts there, > >> so this is redundant. I suggest to drop the patch or provide real > >> rationale (which I doubt you can get :) ). > > > > I'm sorry, I should've included this in the commit message. > > > > The other half of the patch gets rid of a DT validation fail when > > interrupts-extended is used instead of interrupts: > > > > .../arch/arm64/boot/dts/mediatek/mt8195-cherry-tomato-r1.dtb: ec@0: > > 'interrupts' is a required property > > from schema $id: http://devicetree.org/schemas/mfd/google,cros-ec.yaml# > > > > That looks either like a bug in bindings or in dtschema or in your DTS. > This patch should not be needed to solve that. I'll be commiting a dtschema fix soon. There's quite a few places needing the fixup. Rob