Re: [PATCH 1/2] dt-bindings: clock: rockchip: add clk-out

From: Heiko Stübner

Date: Sat Nov 21 2026 - 08:27:36 EST


Am Samstag, 21. November 2026, 12:02:37 Mitteleuropäische Normalzeit schrieb Krzysztof Kozlowski:
> On 21/11/2026 11:58, ZhuXiangyu wrote:
> > Document the Rockchip clock output/input switch, a single GRF-
> > controlled gate used for signals such as the SAI MCLK output.
> >
> > Signed-off-by: ZhuXiangyu <1194781887@xxxxxx>
> > ---
> > .../bindings/clock/rockchip,clk-out.yaml | 81 +++++++++++++++++++
>
> Please follow carefully submitting patches document, including how to
> sign your work and how to send it correctly threaded. There are numerous
> guides already, so just follow them.
>
>
> > 1 file changed, 81 insertions(+)
> > create mode 100644 Documentation/devicetree/bindings/clock/rockchip,clk-out.yaml
> >
> > diff --git a/Documentation/devicetree/bindings/clock/rockchip,clk-out.yaml b/Documentation/devicetree/bindings/clock/rockchip,clk-out.yaml
> > new file mode 100644
> > index 00000000..bf2a9b8a
> > --- /dev/null
> > +++ b/Documentation/devicetree/bindings/clock/rockchip,clk-out.yaml
> > @@ -0,0 +1,81 @@
> > +# SPDX-License-Identifier: GPL-2.0-only OR BSD-2-Clause
> > +%YAML 1.2
> > +---
> > +$id: http://devicetree.org/schemas/clock/rockchip,clk-out.yaml#
> > +$schema: http://devicetree.org/meta-schemas/core.yaml#
> > +
> > +title: Rockchip Clock Output/Input Switch
> > +
> > +maintainers:
> > + - Heiko Stuebner <heiko@xxxxxxxxx>
> > +
> > +description: |
> > + Some Rockchip SoCs route a clock between an internal clock and a pin
> > + through a GRF bit, for example the SAI MCLK input/output switch. This
> > + driver exposes that single bit as a gate clock.
> > +
> > +properties:
> > + compatible:
> > + const: rockchip,clk-out
>
> Generic compatibles are not allowed. Do NOT send us downstream code,
> it's not acceptable.
>
> Not doing detailed review as it does not follow basic rules. Please go
> through for example DTS101 slides.

this whole implementation is bogous, but it seems the submitter found
the correct way to do this in [0] .

But of course the submitter still fails to follow basic styles there.
Broken threading and especially not stating that the new one superseded
this one.

And of course yet another patch bomb in the span of minutes accross so
many different subsystems - which I guess is not a human doing it.

@ZhuXiangyu: even if you're an LLM and new to the process, don't flood
the lists - especially not before you get the process right.


[0] https://lore.kernel.org/linux-rockchip/tencent_9138CD6157C88BC2D52FE6394AF83E560208@xxxxxx