RE: [PATCH v2 09/14] dt-bindings: pinctrl: realtek: Rename 'realtek,duty-cycle' to 'realtek,pulse-width-adjust'

[Yu-Chun Lin [林祐君]]

> On Fri, Mar 06, 2026 at 03:52:39PM +0800, Yu-Chun Lin wrote:
> > From: Tzuyi Chang <tychang@xxxxxxxxxxx>
> >
> > Rename 'realtek,duty-cycle' to 'realtek,pulse-width-adjust'.
> >
> > The previous name was misleading because this hardware block is not a
> > PWM generator. It does not generate a signal with a specific frequency
> > and duty ratio.
> >
> > Instead, it provides a fixed nanosecond-level adjustment to the
> > rising/ falling edges of an existing signal.
> >
> > Signed-off-by: Tzuyi Chang <tychang@xxxxxxxxxxx>
> > Co-developed-by: Yu-Chun Lin <eleanor.lin@xxxxxxxxxxx>
> > Signed-off-by: Yu-Chun Lin <eleanor.lin@xxxxxxxxxxx>
> > ---
> >  .../bindings/pinctrl/realtek,rtd1315e-pinctrl.yaml       | 9 ++++++---
> >  .../bindings/pinctrl/realtek,rtd1319d-pinctrl.yaml       | 9 ++++++---
> >  .../bindings/pinctrl/realtek,rtd1619b-pinctrl.yaml       | 9 ++++++---
> >  3 files changed, 18 insertions(+), 9 deletions(-)
> >
> > diff --git
> > a/Documentation/devicetree/bindings/pinctrl/realtek,rtd1315e-pinctrl.y
> > aml
> > b/Documentation/devicetree/bindings/pinctrl/realtek,rtd1315e-pinctrl.y
> > aml index 90bd49d87d2e..7b3888b2cea8 100644
> > ---
> > a/Documentation/devicetree/bindings/pinctrl/realtek,rtd1315e-pinctrl.y
> > aml
> > +++ b/Documentation/devicetree/bindings/pinctrl/realtek,rtd1315e-pinct
> > +++ rl.yaml
> > @@ -133,10 +133,13 @@ patternProperties:
> >          minimum: 0
> >          maximum: 7
> >
> > -      realtek,duty-cycle:
> > +      realtek,pulse-width-adjust:
> 
> No, that's ABI break without explanation. "misleading" is not the sufficient
> argument for breaking ABI.
> You are stuck with the ABI you added back in 2023.
> 
> Best regards,
> Krzysztof

The reason I renamed 'realtek,duty-cycle' to 'realtek,pulse-width-adjust' is that
during the v1 review, it was pointed out that the name caused confusion with
standard PWM bindings [1]. Since I need to modify the driver logic to address
this, I intended to update the binding simultaneously.

Although the binding was added in 2023, there are currently no DTS files in the
mainline kernel that use the 1315e, 1319d, or 1619b pinctrl device nodes.
Therefore, changing this property name will not break any existing device support
in the mainline tree.

As for the patch structure, I separated the binding and driver changes solely
because checkpatch.pl warned:
"WARNING: DT binding docs and includes should be a separate patch."

Given that there are no actual users of this property in the current mainline DTS,
would it be acceptable to proceed with this rename?

[1] https://lore.kernel.org/all/d5be357c14b84155adfa8a9f00a64d83@xxxxxxxxxxx/

Best regards,
Yu-Chun