Re: [PATCH v4 4/4] usb: dwc3: Notify XHCI core of tunneled status

[Thinh Nguyen]

On Tue, Oct 20, 2026, Konrad Dybcio wrote:
> On 10/3/26 3:23 AM, Thinh Nguyen wrote:
> > On Tue, Sep 29, 2026, Konrad Dybcio wrote:
> >> From: Konrad Dybcio <konrad.dybcio@xxxxxxxxxxxxxxxx>
> >>
> >> The Thunderbolt framework relies on the USB core to create device links
> >> for tunneled ports, so that the USB3 controller is only kept
> >> runtime-resumed for the duration of the tunneling. This depends on
> >> first knowing whether a connection is tunneled or native.
> >>
> >> Add the logic to handle that for DWC3 controllers.
> >>
> >> Assisted-by: LLM
> >> Signed-off-by: Konrad Dybcio <konrad.dybcio@xxxxxxxxxxxxxxxx>
> >> ---
> 
> [...]
> 
> >> +	/* Not all DWC3 instances have CIO HW, trust the platform firmware */
> >> +	if (!device_property_present(dwc->dev, "usb4-host-interface"))
> > 
> > Can we pass this as a dwc3 property from your glue driver instead,
> > inferred from the compatible string? (ie. add dwc->has_cio)
> 
> We have both CIO and non-CIO hosts with the exact same compatible
> strings on the SoCs that support it :/

Ok.

> 
> Ack to other comments. If we need to wait for this dt property,
> would you be open to getting patches 1-3 merged in meantime?

I'm fine with 1-3 going in first, pending on Mathias's review.

> 
> Or maybe we can lean on the fact that there exist dwc3+ACPI devices
> that could consume this property too and sneak by ;)
> 

I'd prefer to wait for the DT property to be accepted first, since the
binding is still under discussion. I'd rather not use the ACPI case as
justification unless there's an actual user.

Thanks,
Thinh