Re: [PATCH] schemas: Allow clocks: property in cache nodes

[Geert Uytterhoeven <geert@linux-m68k.org>]

Hi Conor,

On Sun, 3 May 2026 at 20:09, Conor Dooley <conor@kernel.org> wrote:
> On Sun, May 03, 2026 at 05:44:13PM +0200, Marek Vasut wrote:
> > Renesas R-Mobile APE6 currently describes clock which supply the cache
> > controller in their DT using "clocks" property. This is not the only
> > hardware that has cache controller clock controllable via some sort of
> > clock controller, for example Altera SoCFPGA Cyclone V and Arria V also
> > has controllable cache controller clock. Allow clocks: property in cache
> > controller node to allow users to fully describe such hardware.
>
> Hmm, shouldn't these cache controllers have dedicated bindings that
> enforce their clock requirements?

Just a few general comments:
  - All electronics needs power, so "power-domains" should always
    be permitted,
  - All synchronous logic needs a clock, so "clocks" should always
    be permitteded,
  - All reasonably complex circuits need a reset, so "resets" should
    always be permitted.

> > Signed-off-by: Marek Vasut <marek.vasut+renesas@mailbox.org>

As I have sent the same patch before[1]:
Acked-by: Geert Uytterhoeven <geert+renesas@glider.be>

> > --- a/dtschema/schemas/cache.yaml
> > +++ b/dtschema/schemas/cache.yaml
> > @@ -33,6 +33,9 @@ properties:
> >    compatible:
> >      const: cache
> >
> > +  clocks:
> > +    maxItems: 1
> > +
> >    power-domains:
> >      maxItems: 1

[1] https://lore.kernel.org/20260113075243.1192477-1-geert+renesas@glider.be/

Gr{oetje,eeting}s,

                        Geert

-- 
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org

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