Re: [PATCH v5 3/3] arm64: dts: qcom: sa8540p-ride: Add ethernet nodes

[Stephen Boyd]

Quoting Andrew Halaney (2023-04-13 14:01:27)
> On Thu, Apr 13, 2023 at 01:47:19PM -0700, Stephen Boyd wrote:
> > Quoting Andrew Halaney (2023-04-13 12:15:41)
> > >  arch/arm64/boot/dts/qcom/sa8540p-ride.dts | 179 ++++++++++++++++++++++
> > >  1 file changed, 179 insertions(+)
> > > 
> > > diff --git a/arch/arm64/boot/dts/qcom/sa8540p-ride.dts b/arch/arm64/boot/dts/qcom/sa8540p-ride.dts
> > > index 40db5aa0803c..650cd54f418e 100644
> > > --- a/arch/arm64/boot/dts/qcom/sa8540p-ride.dts
> > > +++ b/arch/arm64/boot/dts/qcom/sa8540p-ride.dts
> > > @@ -28,6 +28,65 @@ aliases {
> > >         chosen {
> > >                 stdout-path = "serial0:115200n8";
> > >         };
> > > +
> > > +       mtl_rx_setup: rx-queues-config {
> > 
> > Is there a reason why this isn't a child of an ethernet node?
> > 
> > 
> 
> I debated if it was more appropriate to:
> 
>     1. make a duplicate in each ethernet node (ethernet0/1)
>     2. Put it in one and reference from both
>     3. have it floating around independent like this, similar to what is
>        done in sa8155p-adp.dts[0]
> 
> I chose 3 as it seemed cleanest, but if there's a good argument for a
> different approach I'm all ears!

I wonder if it allows the binding checker to catch bad properties by
having it under the ethernet node? That's the only thing I can think of
that may be improved, but I'll let binding reviewers comment here.