[PATCH v2 19/22] thunderbolt: Add Apple Silicon support

[Mika Westerberg]

Hi

On Wed, Sep 09, 2026 at 05:25:32PM +0200, Sven Peter wrote:
> On 09.09.26 17:20, Sven Peter wrote:
> > Hi,
> > 
> > On 09.09.26 08:06, Mika Westerberg wrote:
> > 
> > [...]
> > 
> > > 
> > > If it is possible, I suggest to use the same way we do in ACPI side here:
> > > 
> > > https://learn.microsoft.com/en-us/windows-hardware/drivers/pci/dsd-
> > > for-pcie-root-ports#map-native-protocols-pcie-displayport-tunneled-
> > > through-usb4-to-usb4-host-routers
> > > 
> > > It was specifically designed so that it should support DT as well (that's
> > > why it uses the device properties _DSD).
> > 
> > Oh, nice, I didn't know those existed. Konrad already has a PR open to
> > add these to dt-schema [1] and a PR that already makes use of these for
> > dwc3 [2].
> > I can just write generic code similar to the ACPI one then to add the
> > device links and split it off from this series, though I think that dt-
> > schema PR must be merged before we're allowed to use that
> > representation.
> > I'll adjust the dt-bindings here for v3 to just drop the representation
> > as OF ports/endpoints as well.
> > For the PCIe/DP tunnel notifications I should be able to use the same
> > representation, possibly with an additional usb4-adapter-index property
> > or similar to differentiate between the two DP IN adapters, once I get
> > to that then.
> 
> Though now that I read it again: do you know what PortInstance# in
> 
> Package () { "usb4-port-number", PortInstance#}, // PortInstance# is unused
> and can be 0
>       }
> 
> refers to? Is that meant to e.g. by 0 for the first DP IN adapter and 1 for
> the second? Or is that meant to be used when the NHI handles multiple Type-C
> ports? The description isn't clear to me...
> 

That's the USB4 port through the native PCIe/USB3 Gen X is tunneled through
but like I said in the previous email you don't need this for DP, only for
PCIe and USB3 (Gen X/T).