Re: [PATCH v1 1/2] dt-bindings: phy: mtk-tphy: add property for software role switch

[Krzysztof Kozlowski <krzk@kernel.org>]

On 24/08/2025 08:25, Arseniy Velikanov wrote:
> On 15.08.2025 10:28, Krzysztof Kozlowski wrote:
>> On Fri, Aug 15, 2025 at 03:48:24AM +0400, Arseniy Velikanov wrote:
>>> Add a boolean property to enable software control of the PHY mode
>>> switching.
>>>
>>> Signed-off-by: Arseniy Velikanov <me@adomerle.pw>
>>> ---
>>>   Documentation/devicetree/bindings/phy/mediatek,tphy.yaml | 6 ++++++
>>>   1 file changed, 6 insertions(+)
>>>
>>> diff --git a/Documentation/devicetree/bindings/phy/mediatek,tphy.yaml b/Documentation/devicetree/bindings/phy/mediatek,tphy.yaml
>>> index b2218c151939..a90890d4a86f 100644
>>> --- a/Documentation/devicetree/bindings/phy/mediatek,tphy.yaml
>>> +++ b/Documentation/devicetree/bindings/phy/mediatek,tphy.yaml
>>> @@ -255,6 +255,12 @@ patternProperties:
>>>             use the property "mediatek,syscon-type" for newer SoCs that support it.
>>>           type: boolean
>>>   
>>> +      mediatek,software-role-switch:
>>
>> No, bindings do not describe software.
> Sorry, I didn't quite understand you.
> 
> With this property the driver changes phy mode manually using debug regs 


Here is the problem:     ^^^^^^^^, driver.

> instead of configuring hardware mode switching (because it's broken on 
> some platforms). So every time the role in usb-role-switch changes, it 
> has to switch phy mode. That's what I mean by "software".
> 
> I can rename this property to `mediatek,force-mode-switch` if that's a 
> issue.


Describe the hardware problem, characteristic, wiring or whatever else,
but not the drivers and not software.

force-media-switch is the same uninformative and same underlying
instruction for the driver.

Best regards,
Krzysztof

-- 
linux-phy mailing list
linux-phy@lists.infradead.org
https://lists.infradead.org/mailman/listinfo/linux-phy