[PATCH v9 3/5] dt-bindings: clock: meson: add A1 PLL clock controller bindings

[Dmitry Rokosov]

Hello Krzysztof,

Thank you for quick review!

On Fri, Mar 03, 2023 at 09:28:22AM +0100, Krzysztof Kozlowski wrote:
> On 01/03/2023 19:37, Dmitry Rokosov wrote:
> > Add the documentation for Amlogic A1 PLL clock driver, and A1 PLL
> > clock controller bindings.
> > Also include new A1 clock controller dt bindings to MAINTAINERS.
> > 
> > Signed-off-by: Jian Hu <jian.hu at amlogic.com>
> > Signed-off-by: Dmitry Rokosov <ddrokosov at sberdevices.ru>
> > ---
> >  .../bindings/clock/amlogic,a1-pll-clkc.yaml   | 59 +++++++++++++++++++
> >  MAINTAINERS                                   |  1 +
> >  include/dt-bindings/clock/a1-pll-clkc.h       | 20 +++++++
> >  3 files changed, 80 insertions(+)
> >  create mode 100644 Documentation/devicetree/bindings/clock/amlogic,a1-pll-clkc.yaml
> >  create mode 100644 include/dt-bindings/clock/a1-pll-clkc.h
> > 

[...]

> > +
> > +examples:
> > +  - |
> > +    #include <dt-bindings/clock/a1-clkc.h>
> 
> Does not look like you tested the bindings. Please run `make
> dt_binding_check` (see
> Documentation/devicetree/bindings/writing-schema.rst for instructions).
> 

I always run dt binding tests before sending the patch series.
It's 'must have' prerequisite along with checkpatch.

The first one (dt_binding_check):
    $ make ARCH=arm64 INSTALL_MOD_PATH=${INSTALL_MOD_PATH} \
           CROSS_COMPILE="${CROSS_COMPILE}" DEPMOD=${DEPMOD}
           INSTALL_MOD_STRIP=1 -C ${KERNEL_PATH} \
           dt_binding_check DT_SCHEMA_FILES=${DT_SCHEMA_PATH}

The second one (dtbs_check):
    $ make ARCH=arm64 INSTALL_MOD_PATH=${INSTALL_MOD_PATH} \
           CROSS_COMPILE="${CROSS_COMPILE}" DEPMOD=${DEPMOD} \
           INSTALL_MOD_STRIP=1 -C ${KERNEL_PATH} \
           dtbs_check DT_SCHEMA_FILES=${DT_SCHEMA_PATH}

But, as you mentioned in the another patchset, I didn't take into account
bisectability. In other words, I didn't execute above sanity check on the
each patchset.
Thank you for good catch, I'll fix it in the today v10 patch series.

> > +    apb {
> > +        #address-cells = <2>;
> > +        #size-cells = <2>;
> > +
> > +        clock-controller at 7c80 {
> > +            compatible = "amlogic,a1-pll-clkc";
> > +            reg = <0 0x7c80 0 0x18c>;
> > +            #clock-cells = <1>;
> > +            clocks = <&clkc_periphs CLKID_FIXPLL_IN>,
> > +                     <&clkc_periphs CLKID_HIFIPLL_IN>;
> > +            clock-names = "fixpll_in", "hifipll_in";
> > +        };
> > +    };
> > diff --git a/MAINTAINERS b/MAINTAINERS
> > index 39ff1a717625..8438bc9bd636 100644
> > --- a/MAINTAINERS
> > +++ b/MAINTAINERS
> > @@ -1895,6 +1895,7 @@ L:	linux-amlogic at lists.infradead.org
> >  S:	Maintained
> >  F:	Documentation/devicetree/bindings/clock/amlogic*
> >  F:	drivers/clk/meson/
> > +F:	include/dt-bindings/clock/a1*
> >  F:	include/dt-bindings/clock/gxbb*
> >  F:	include/dt-bindings/clock/meson*
> >  
> > diff --git a/include/dt-bindings/clock/a1-pll-clkc.h b/include/dt-bindings/clock/a1-pll-clkc.h
> 
> Filename matching bindings, so amlogic,a1-pll-clkc.h

You are totally right. But looks like other amlogic clock bindings don't
follow this rule.
So I'll change my patch series and send another patch series with fixup other
amlogic clock bindings.
> 
> > new file mode 100644
> > index 000000000000..3a559518c6e6
> > --- /dev/null
> > +++ b/include/dt-bindings/clock/a1-pll-clkc.h
> > @@ -0,0 +1,20 @@
> > +/* SPDX-License-Identifier: (GPL-2.0+ OR MIT) */
> 
> Any particular reason for using license other than in binding? Was it
> intentional (e.g. because it is derivative work)?
> 

No reason, actually. I've just used license which was introduced in the
previous patch series versions by Jian Hu. I suppose, standard license
usage should be okay.

[...]

-- 
Thank you,
Dmitry