Hi Sudeep,
On Wed, Feb 16, 2022 at 3:40 PM Sudeep Holla <sudeep.holla@xxxxxxx> wrote:
> On Wed, Feb 16, 2022 at 02:21:43PM +0100, Geert Uytterhoeven wrote:
> > "make dt_binding_check":
> >
> > Documentation/devicetree/bindings/firmware/arm,scpi.example.dt.yaml: scpi: shmem: [[2], [1]] is too long
> >
> > Signed-off-by: Geert Uytterhoeven <geert+renesas@xxxxxxxxx>
> > ---
> > Exposed by commit 39bd2b6a3783b899 ("dt-bindings: Improve phandle-array
> > schemas").
>
> Interesting !
>
> > ---
> > Documentation/devicetree/bindings/firmware/arm,scpi.yaml | 1 +
> > 1 file changed, 1 insertion(+)
> >
> > diff --git a/Documentation/devicetree/bindings/firmware/arm,scpi.yaml b/Documentation/devicetree/bindings/firmware/arm,scpi.yaml
> > index 800417a76bda0bd9..3735f221fdf24e0a 100644
> > --- a/Documentation/devicetree/bindings/firmware/arm,scpi.yaml
> > +++ b/Documentation/devicetree/bindings/firmware/arm,scpi.yaml
> > @@ -51,6 +51,7 @@ properties:
> > be any memory reserved for the purpose of this communication between the
> > processors.
> > minItems: 1
> > + maxItems: 2
> >
>
> There is no max limit strictly speaking. The driver can use all the specified
> mboxes and associated shmem in round robin fashion. That is the reason I
> didn't add maxItems unlike the newer SCMI protocol which clearly restricts
> to one Tx and one Rx(much saner I must admit).
In the absence of maxItems, the validator assumes it is equal to minItems,
so we do need a sensible maxItems value here.
Any suggestions? 16? 64?
Thanks!
Gr{oetje,eeting}s,
Geert
--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@xxxxxxxxxxxxxx
In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds
