Re: [PATCH v3 1/9] dt-bindings: memory: Document RZ/G3E support

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

Hi Biju,

On Mon, 31 Mar 2025 at 20:29, Biju Das <biju.das.jz@bp.renesas.com> wrote:
> > From: Geert Uytterhoeven <geert@linux-m68k.org>
> > On Mon, 31 Mar 2025 at 17:33, Biju Das <biju.das.jz@bp.renesas.com> wrote:
> > > > From: Geert Uytterhoeven <geert@linux-m68k.org> On Mon, 31 Mar 2025
> > > > at 16:34, Biju Das <biju.das.jz@bp.renesas.com> wrote:
> > > > > > From: Geert Uytterhoeven <geert@linux-m68k.org> On Mon, 31 Mar
> > > > > > 2025 at 15:54, Biju Das <biju.das.jz@bp.renesas.com> wrote:
> > > > > > > > From: Biju Das <biju.das.jz@bp.renesas.com> Document support
> > > > > > > > for the Expanded Serial Peripheral Interface (xSPI)
> > > > > > > > Controller in the Renesas RZ/G3E
> > > > > > > > (R9A09G047) SoC.
> > > > > > > >
> > > > > > > > Reviewed-by: Rob Herring (Arm) <robh@kernel.org>
> > > > > > > > Signed-off-by: Biju Das <biju.das.jz@bp.renesas.com>
> > > > > >
> > > > > > > > --- /dev/null
> > > > > > > > +++ b/Documentation/devicetree/bindings/memory-controllers/r
> > > > > > > > +++ enes
> > > > > > > > +++ as,r
> > > > > > > > +++ zg3e
> > > > > > > > +++ -xspi.yaml
> > > > > >
> > > > > > > > +    spi@11030000 {
> > > > > > > > +        compatible = "renesas,r9a09g047-xspi";
> > > > > > > > +        reg = <0x11030000 0x10000>, <0x20000000 0x10000000>;
> > > > > > > > +        reg-names = "regs", "dirmap";
> > > > > > > > +        interrupts = <GIC_SPI 228 IRQ_TYPE_EDGE_RISING>,
> > > > > > > > +                     <GIC_SPI 229 IRQ_TYPE_EDGE_RISING>;
> > > > > > > > +        interrupt-names = "pulse", "err_pulse";
> > > > > > > > +        clocks = <&cpg CPG_MOD 0x9f>, <&cpg CPG_MOD 0xa0>,
> > > > > > > > +                 <&cpg CPG_MOD 0xa1>, <&cpg CPG_MOD 0xa1>;
> > > > > > >
> > > > > > > On the next version I am going to update spix2 clk as <&cpg
> > > > > > > CPG_CORE R9A09G047_SPI_CLK_SPIX2>
> >
> > According to the RZ/G3E clock system diagram, (the parent of) clk_spi is derived from (the parent of)
> > clk_spix2, not the other way around?
> > So you can model clk_spi as a fixed divider clock with parent clk_spix2 and factor two.  I.e. provide
> > a new core clock R9A09G047_SPI_CLK_SPI instead of your proposed R9A09G047_SPI_CLK_SPIX2?
    ^^^^

> > > > > > What's spix2 clk? Ah, re-adding dropped line:
> > > > > >
> > > > > > > > +        clock-names = "ahb", "axi", "spi", "spix2";

> Can you please share your thoughts to handle this?

See above ^^^^ ;-)

> 1) Gate only spi clk

Gate only clk_spix2, which is the parent of clk_spi.
So enabling any of them will (propagate to) enable clk_spix2,
which uses the hardware gate.

> 2) For monitoring use both clock

Check only clk_spix2 for monitoring.

> 3) Clock specifier needs two distinct entries. So that consumer will get
>    proper rates for both clocks.

clk_spi would be a separate fixed-divider clock.

Does that make sense?
Thanks!

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