[PATCH v3 5/6] arm64: dts: rockchip: fix USB OTG power on Photonicat in battery mode

[Coia Prant]

Chukun Pan <amadeus at jmu.edu.cn> 于2026年11月4日周三 20:40写道：
>
> Hi,
>
> > regulator_5v_boost: regulator-5v-boost {
>
> I prefer to use vcc as a prefix for consistency.
>
> > +      * 5V_OUT is the OR-ed combination of the VBAT boost (via D5002)
> > +      * and VCCIN_5V (via D16).
> > +      * It has no enable pin of its own.
> > +      * Model it as a fixed regulator depending on the boost converter,
> > +      * so that VBAT is boosted when running on battery.
>
> What are the benefits of doing this? This is not a real regulator.
> It is the same path as 5V_BOOST. D16 controls which channel the VIN
> of VCC5V0_USB30_OTG0 comes from.

Hi,

Thanks for the feedback.

1. Naming: I'd prefer to keep the names as `regulator-5v-boost` and
   `regulator-5v-out`. These match the exact net names in the schematic
   (5V_BOOST and 5V_OUT). Adding a `vcc` prefix would break the 1:1
   mapping between the DTS and the schematic, making it harder for
   anyone tracing the board power.

2. Regarding the 5V_OUT node: I understand your point that D16/D5002
   are hardware ORing and there is no software control. However, 5V_OUT
   is a distinct net in the schematic. Keeping the node makes the DT
   power tree accurately reflect the hardware and clearly documents
   that both the USB VBUS and HDMI 5V domains share this OR-ed source.

   The `regulator-fixed` binding supports modeling always-on fixed
   rails, so this doesn't break the regulator semantics. If the kernel
   later supports multiple `vin-supply` inputs, we can update this node
   to list both the boost and VCCIN_5V explicitly. For now, chaining
   `vcc5v0_usb30_otg0` -> `5v_out` -> `5v_boost` is the most accurate
   representation.

Coia