Re: [PATCH v4 4/7] dt-bindings: sram: Merge Renesas SRAM bindings into generic

[Geert Uytterhoeven]

Hi Krzysztof,

On Fri, Nov 1, 2019 at 11:54 AM Krzysztof Kozlowski <krzk@xxxxxxxxxx> wrote:
> On Fri, 1 Nov 2019 at 11:08, Geert Uytterhoeven <geert@xxxxxxxxxxxxxx> wrote:
> > On Mon, Oct 21, 2019 at 6:15 PM Krzysztof Kozlowski <krzk@xxxxxxxxxx> wrote:
> > > The Renesas SRAM bindings list only compatible so integrate them into
> > > generic SRAM bindings schema.
> > >
> > > Signed-off-by: Krzysztof Kozlowski <krzk@xxxxxxxxxx>
> >
> > Thanks for your patch, whcih is now commit 0759b09eadd0d9a1 ("dt-bindings:
> > sram: Merge Renesas SRAM bindings into generic") in Rob's for-next branch.
> >
> > > --- a/Documentation/devicetree/bindings/sram/renesas,smp-sram.txt
> > > +++ /dev/null
> > > @@ -1,27 +0,0 @@
> > > -* Renesas SMP SRAM
> > > -
> > > -Renesas R-Car Gen2 and RZ/G1 SoCs need a small piece of SRAM for the jump stub
> > > -for secondary CPU bringup and CPU hotplug.
> > > -This memory is reserved by adding a child node to a "mmio-sram" node, cfr.
> > > -Documentation/devicetree/bindings/sram/sram.txt.
> > > -
> > > -Required child node properties:
> > > -  - compatible: Must be "renesas,smp-sram",
> > > -  - reg: Address and length of the reserved SRAM.
> > > -    The full physical (bus) address must be aligned to a 256 KiB boundary.
> > > -
> > > -
> > > -Example:
> > > -
> > > -       icram1: sram@e63c0000 {
> > > -               compatible = "mmio-sram";
> > > -               reg = <0 0xe63c0000 0 0x1000>;
> > > -               #address-cells = <1>;
> > > -               #size-cells = <1>;
> > > -               ranges = <0 0 0xe63c0000 0x1000>;
> > > -
> > > -               smp-sram@0 {
> > > -                       compatible = "renesas,smp-sram";
> > > -                       reg = <0 0x10>;
> > > -               };
> >
> > > --- a/Documentation/devicetree/bindings/sram/sram.yaml
> > > +++ b/Documentation/devicetree/bindings/sram/sram.yaml
> >
> > > @@ -186,3 +187,17 @@ examples:
> > >              reg = <0x1ff80 0x8>;
> > >          };
> > >      };
> > > +
> > > +  - |
> > > +    sram@e63c0000 {
> > > +        compatible = "mmio-sram";
> > > +        reg = <0xe63c0000 0x1000>;
> >
> > Is there any specific reason you converted the example from 64-bit to
> > 32-bit addressing?
> > All Renesas SoCs using this have #address-cells and #size-cells = <2>.
>
> I should mention it in commit msg. The reason is because examples are
> compiled inside a {} with address/size cells of 1. Instead of

Thanks, that's what I was already afraid of...

> conversion maybe it would be reasonable to put it inside additional
> node adjusting the address/size cells.

I think it's fine to leave it as-as, though.  If we ever get to DT-ize
secondary CPU startup on EMMA Mobile EV2 or SH-Mobile AG5, we'll have
users without LPAE ;-)

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