Re: [PATCH v2 2/4] dt-bindings: usb: chipidea,usb2-tegra: Document nvidia,external-control property

[sashiko-bot@kernel.org]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Medium] Adding `nvidia,external-control` to prevent the host from accessing the controller is an architectural anti-pattern that duplicates the purpose of the standard `status` property and misattributes resource ownership.
--

commit ba773e6bd558f2861c226bd39627e8f4363d1a0d
Author: Svyatoslav Ryhel <clamor95@gmail.com>

dt-bindings: usb: chipidea,usb2-tegra: Document nvidia,external-control property

This commit documents the nvidia,external-control property for USB controllers.
It indicates that the controller is configured externally, typically for a
modem requiring precise control over the USB bus in HSIC mode.

> diff --git a/Documentation/devicetree/bindings/usb/chipidea,usb2-tegra.yaml b/Documentation/devicetree/bindings/usb/chipidea,usb2-tegra.yaml
> index 78046f8a63ad5..d99a18900b5a8 100644
> --- a/Documentation/devicetree/bindings/usb/chipidea,usb2-tegra.yaml
> +++ b/Documentation/devicetree/bindings/usb/chipidea,usb2-tegra.yaml
[ ... ]
> @@ -40,6 +40,13 @@ properties:
>      type: boolean
>      deprecated: true
>  
> +  nvidia,external-control:
> +    description:
> +      Indicates that the controller is configured externally and that the host
> +      should not attempt to access it. This is usually used by a modem that
> +      requires precise bus configuration.
> +    type: boolean
> +

[Severity: Medium]
Does this new property duplicate the purpose of the standard status property
and misattribute resource ownership?

Device tree is intended to describe hardware, not software workarounds. If
the hardware block is exclusively controlled by an external entity like a
modem and should not be accessed by the host OS, standard semantics generally
dictate using status = "disabled" or status = "reserved".

If the modem relies on the host to provide shared resources like clocks or
resets to bring the bus up before enumeration, could those resources be
modeled via a power-sequencer or assigned directly to the modem hardware
node? 

Modeling the resources this way might avoid using the USB controller node as a
silent resource enabler, which would help keep the hardware description
decoupled from software policy.

>  required:
>    - compatible
>    - reg

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260523083013.46372-1-clamor95@gmail.com?part=2