[PATCH 0/3] rtc: rzn1: support XTAL clk and SCMP method

[PATCH 0/3] rtc: rzn1: support XTAL clk and SCMP method

[Wolfram Sang]

So far, the code and the binding for the RZ/N1D RTC assumed an input
clock of 32768Hz, so it was not explicitly described. It makes sense to
do this, though. For one reason, clocks with other frequencies might be
used. This RTC supports that via the SCMP counting method. The other
reason is, the upcoming R-Car Gen5 has only the SCMP method described,
so we need to use it there later.

This series lets the driver handle the optional "xtal" clock and switch
to the SCMP method if suitable. It has been tested on a Renesas RZ/N1D
board with hacked devicetree values.

A branch with updated DTs can be found here:

git://git.kernel.org/pub/scm/linux/kernel/git/wsa/linux.git renesas/n1d/rtc

Looking forward to comments!


Changes since RFC:
* add binding tag from Krzysztof (Thanks!)
* new patch to disable controller before initially modifying CTL0
* cosmetic changes

Wolfram Sang (3):
  dt-bindings: rtc: rzn1: add optional second clock
  rtc: rzn1: Disable controller before initialization
  rtc: rzn1: support input frequencies other than 32768Hz

 .../bindings/rtc/renesas,rzn1-rtc.yaml        |  8 ++-
 drivers/rtc/rtc-rzn1.c                        | 59 +++++++++++++++----
 2 files changed, 53 insertions(+), 14 deletions(-)

-- 
2.47.2

[PATCH 1/3] dt-bindings: rtc: rzn1: add optional second clock

[Wolfram Sang]

The external crystal can be a second clock input. It is needed for the
SCMP counting method which allows using crystals different than 32768Hz.
It is also needed for an upcoming SoC which only supports the SCMP
method.

Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
---
 .../devicetree/bindings/rtc/renesas,rzn1-rtc.yaml         | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/Documentation/devicetree/bindings/rtc/renesas,rzn1-rtc.yaml b/Documentation/devicetree/bindings/rtc/renesas,rzn1-rtc.yaml
index f6e0c613af67..f6fdcc7090b6 100644
--- a/Documentation/devicetree/bindings/rtc/renesas,rzn1-rtc.yaml
+++ b/Documentation/devicetree/bindings/rtc/renesas,rzn1-rtc.yaml
@@ -33,10 +33,14 @@ properties:
       - const: pps
 
   clocks:
-    maxItems: 1
+    minItems: 1
+    maxItems: 2
 
   clock-names:
-    const: hclk
+    minItems: 1
+    items:
+      - const: hclk
+      - const: xtal
 
   power-domains:
     maxItems: 1
-- 
2.47.2

Re: [PATCH 1/3] dt-bindings: rtc: rzn1: add optional second clock

[Geert Uytterhoeven]

Hi Wolfram,

On Wed, 19 Mar 2025 at 12:03, Wolfram Sang
<wsa+renesas@sang-engineering.com> wrote:
> The external crystal can be a second clock input. It is needed for the
> SCMP counting method which allows using crystals different than 32768Hz.
> It is also needed for an upcoming SoC which only supports the SCMP
> method.
>
> Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
> Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>

Thanks for your patch!

Reviewed-by: Geert Uytterhoeven <geert+renesas@glider.be>

> --- a/Documentation/devicetree/bindings/rtc/renesas,rzn1-rtc.yaml
> +++ b/Documentation/devicetree/bindings/rtc/renesas,rzn1-rtc.yaml
> @@ -33,10 +33,14 @@ properties:
>        - const: pps
>
>    clocks:
> -    maxItems: 1
> +    minItems: 1
> +    maxItems: 2
>
>    clock-names:
> -    const: hclk
> +    minItems: 1
> +    items:
> +      - const: hclk
> +      - const: xtal

Shouldn't the second clock become required? Or do you plan to make
that change after all upstream DTS files have been updated?

This is orthogonal to the driver considering it optional to maintain
compatibility with old DTBs.

>
>    power-domains:
>      maxItems: 1

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

Re: [PATCH 1/3] dt-bindings: rtc: rzn1: add optional second clock

[Geert Uytterhoeven]

Hi Wolfram,

On Thu, 10 Apr 2025 at 17:08, Geert Uytterhoeven <geert@linux-m68k.org> wrote:
> On Wed, 19 Mar 2025 at 12:03, Wolfram Sang
> <wsa+renesas@sang-engineering.com> wrote:
> > The external crystal can be a second clock input. It is needed for the
> > SCMP counting method which allows using crystals different than 32768Hz.
> > It is also needed for an upcoming SoC which only supports the SCMP
> > method.
> >
> > Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
> > Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
>
> Thanks for your patch!
>
> Reviewed-by: Geert Uytterhoeven <geert+renesas@glider.be>
>
> > --- a/Documentation/devicetree/bindings/rtc/renesas,rzn1-rtc.yaml
> > +++ b/Documentation/devicetree/bindings/rtc/renesas,rzn1-rtc.yaml
> > @@ -33,10 +33,14 @@ properties:
> >        - const: pps
> >
> >    clocks:
> > -    maxItems: 1
> > +    minItems: 1
> > +    maxItems: 2
> >
> >    clock-names:
> > -    const: hclk
> > +    minItems: 1
> > +    items:
> > +      - const: hclk
> > +      - const: xtal
>
> Shouldn't the second clock become required? Or do you plan to make
> that change after all upstream DTS files have been updated?

Upon second thought: this xtal clock is documented to be the "rtc"
input to the RZ/N1 system controller[1], so it looks like the original
idea was to obtain it through the system controller.  Unfortunately
the clock driver[2] does not use the rtc input clock, nor provides it
to consumers.

So either we fix that, or we go with your solution...

[1] https://elixir.bootlin.com/linux/latest/source/Documentation/devicetree/bindings/clock/renesas,r9a06g032-sysctrl.yaml#L32
[2] https://elixir.bootlin.com/linux/latest/source/drivers/clk/renesas/r9a06g032-clocks.c

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

Re: [PATCH 1/3] dt-bindings: rtc: rzn1: add optional second clock

[Wolfram Sang]

[-- Attachment #1: Type: text/plain, Size: 922 bytes --]


> > >    clock-names:
> > > -    const: hclk
> > > +    minItems: 1
> > > +    items:
> > > +      - const: hclk
> > > +      - const: xtal
> >
> > Shouldn't the second clock become required? Or do you plan to make
> > that change after all upstream DTS files have been updated?

True, we should make the second clock a requirement from now on.

> Upon second thought: this xtal clock is documented to be the "rtc"
> input to the RZ/N1 system controller[1], so it looks like the original
> idea was to obtain it through the system controller.  Unfortunately
> the clock driver[2] does not use the rtc input clock, nor provides it
> to consumers.

So, it would basically be a pass-through? I don't see any register in
SYSCTRL handling the external RTC clock.

> So either we fix that, or we go with your solution...

If it is a pass-through, I wonder what it would gain us, but I can do
that if there are reasons for it.


[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]

Re: [PATCH 1/3] dt-bindings: rtc: rzn1: add optional second clock

[Geert Uytterhoeven]

Hi Wolfram,

On Thu, 10 Apr 2025 at 22:23, Wolfram Sang
<wsa+renesas@sang-engineering.com> wrote:
> > > >    clock-names:
> > > > -    const: hclk
> > > > +    minItems: 1
> > > > +    items:
> > > > +      - const: hclk
> > > > +      - const: xtal
> > >
> > > Shouldn't the second clock become required? Or do you plan to make
> > > that change after all upstream DTS files have been updated?
>
> True, we should make the second clock a requirement from now on.
>
> > Upon second thought: this xtal clock is documented to be the "rtc"
> > input to the RZ/N1 system controller[1], so it looks like the original
> > idea was to obtain it through the system controller.  Unfortunately
> > the clock driver[2] does not use the rtc input clock, nor provides it
> > to consumers.
>
> So, it would basically be a pass-through? I don't see any register in
> SYSCTRL handling the external RTC clock.

I assume you are right, I didn't study RZ/N1D in detail.

> > So either we fix that, or we go with your solution...
>
> If it is a pass-through, I wonder what it would gain us, but I can do
> that if there are reasons for it.

Let's go for your solution.

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

Re: [PATCH 1/3] dt-bindings: rtc: rzn1: add optional second clock

[Wolfram Sang]

[-- Attachment #1: Type: text/plain, Size: 164 bytes --]


> > If it is a pass-through, I wonder what it would gain us, but I can do
> > that if there are reasons for it.
> 
> Let's go for your solution.

Thanks!


[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]