[PATCH 2/3] arm64: dts: freescale: add initial device tree for TQMa93xx/MBa93xxLA-MINI

Andrew Lunn andrew at lunn.ch
Thu Mar 19 06:30:31 PDT 2026


On Thu, Mar 19, 2026 at 12:25:11PM +0100, Alexander Stein wrote:
> Hi Andrew,
> 
> Am Mittwoch, 18. März 2026, 15:49:38 CET schrieb Andrew Lunn:
> > On Tue, Mar 17, 2026 at 12:17:01PM +0100, Alexander Stein wrote:
> > > +	mdio {
> > > +		compatible = "snps,dwmac-mdio";
> > > +		#address-cells = <1>;
> > > +		#size-cells = <0>;
> > > +
> > > +		ethphy_eqos: ethernet-phy at 0 {
> > > +			compatible = "ethernet-phy-ieee802.3-c22";
> > > +			reg = <0>;
> > > +			pinctrl-names = "default";
> > > +			pinctrl-0 = <&pinctrl_eqos_phy>;
> > > +			interrupt-parent = <&gpio3>;
> > > +			interrupts = <26 IRQ_TYPE_EDGE_FALLING>;
> > 
> > Hi Alexander
> > 
> > Haven't we been here before? EDGE interrupts are generally wrong for
> > PHYs.
> 
> Whoops. Sorry, this slipped in from the downstream kernel. Will fix that.

You might want to fix your downstream as well, since this is not a
Mainline thing.

> > > +&fec {
> > > +	pinctrl-names = "default";
> > > +	pinctrl-0 = <&pinctrl_fec>;
> > > +	phy-mode = "rgmii-id";
> > > +	phy-handle = <&ethphy_fec>;
> > > +	fsl,magic-packet;
> > 
> > Has WoL been tested?
> 
> Yes, it works for both interfaces. &eqos doesn't need that property though.

Great. For the next year or two, i might be asking this all the time,
so maybe call it out in the commit message.

   Andrew



More information about the linux-arm-kernel mailing list