[PATCH 3/5] arm64: dts: allwinner: a523: Add EMAC0 ethernet MAC

[Chen-Yu Tsai]

On Thu, Apr 24, 2025 at 11:28 AM Yixun Lan <dlan at gentoo.org> wrote:
>
> Hi Andre,
>
> On 01:43 Thu 24 Apr     , Andre Przywara wrote:
> > On Wed, 23 Apr 2025 22:03:24 +0800
> > Yixun Lan <dlan at gentoo.org> wrote:
> >
> > Hi Yixun,
> >
> > thanks for sending those patches!
> >
> > > Add EMAC0 ethernet MAC support which found on A523 variant SoCs,
> > > including the A527/T527 chips.
> >
> > maybe add here that MAC0 is compatible to the A64, and requires an
> > external PHY. And that we only add the RGMII pins for now.
> >
> ok
>
> > > Signed-off-by: Yixun Lan <dlan at gentoo.org>
> > > ---
> > >  arch/arm64/boot/dts/allwinner/sun55i-a523.dtsi | 42 ++++++++++++++++++++++++++
> > >  1 file changed, 42 insertions(+)
> > >
> > > diff --git a/arch/arm64/boot/dts/allwinner/sun55i-a523.dtsi b/arch/arm64/boot/dts/allwinner/sun55i-a523.dtsi
> > > index ee485899ba0af69f32727a53de20051a2e31be1d..c3ba2146c4b45f72c2a5633ec434740d681a21fb 100644
> > > --- a/arch/arm64/boot/dts/allwinner/sun55i-a523.dtsi
> > > +++ b/arch/arm64/boot/dts/allwinner/sun55i-a523.dtsi
> > > @@ -126,6 +126,17 @@ pio: pinctrl at 2000000 {
> > >                     interrupt-controller;
> > >                     #interrupt-cells = <3>;
> > >
> > > +                   emac0_pins: emac0-pins {
> >
> > Both the alias and the node name should contain rgmii instead of emac0,
> > as the other SoCs do, I think:
> >                       rgmii0_pins: rgmii0-pins {
> >
> ok
> > > +                           pins = "PH0", "PH1", "PH2", "PH3",
> > > +                                   "PH4", "PH5", "PH6", "PH7",
> > > +                                   "PH9", "PH10","PH13","PH14",
> > > +                                   "PH15","PH16","PH17","PH18";
> >
> > I think there should be a space behind each comma, and the
> > first quotation marks in each line should align.
> >
> will do
>
> > PH13 is EPHY-25M, that's the (optional) 25 MHz output clock pin, for
> > PHYs without a crystal. That's not controlled by the MAC, so I would
> > leave it out of this list, as also both the Avaota and the Radxa don't
> > need it. If there will be a user, they can add this separately.
> >
> make sense
>
> > > +                           allwinner,pinmux = <5>;
> > > +                           function = "emac0";
> > > +                           drive-strength = <40>;
> > > +                           bias-pull-up;
> >
> > Shouldn't this be push-pull, so no pull-up?
> >
> will drop

It would be better to have an explicit "bias-disable" here.
That way you are not depending on the bootloader setting a
certain state, or depending on the reset default.

ChenYu

> > The rest looks correct, when compared to the A523 manual.
> >
> thanks for review
>
> --
> Yixun Lan (dlan)