[PATCH] rockchip: enabled LAN port on NanoPi R2S

[Chen-Yu Tsai]

On Mon, Apr 5, 2021 at 4:53 PM Tianling Shen <cnsztl at gmail.com> wrote:
>
> Hi Chen-Yu,
>
> On 2021-04-05 16:14, Chen-Yu Tsai <wens at kernel.org> wrote:
> >
> > Hi,
> >
> > On Mon, Apr 5, 2021 at 3:46 PM Tianling Shen <cnsztl at gmail.com> wrote:
> > >
> > > From: David Bauer <mail at david-bauer.net>
> > >
> > > Enable the USB3 port on the FriendlyARM NanoPi R2S.
> > > This is required for the USB3 attached LAN port to work.
> > >
> > > Signed-off-by: David Bauer <mail at david-bauer.net>
> > > Signed-off-by: Tianling Shen <cnsztl at gmail.com>
> > > ---
> > >  .../boot/dts/rockchip/rk3328-nanopi-r2s.dts   | 23 +++++++++++++++++++
> > >  1 file changed, 23 insertions(+)
> > >
> > > diff --git a/arch/arm64/boot/dts/rockchip/rk3328-nanopi-r2s.dts b/arch/arm64/boot/dts/rockchip/rk3328-nanopi-r2s.dts
> > > index faf496d789cf..6ba9799a95c5 100644
> > > --- a/arch/arm64/boot/dts/rockchip/rk3328-nanopi-r2s.dts
> > > +++ b/arch/arm64/boot/dts/rockchip/rk3328-nanopi-r2s.dts
> > > @@ -37,6 +37,18 @@
> > >                 };
> > >         };
> > >
> > > +       vcc_rtl8153: vcc-rtl8153-regulator {
> > > +               compatible = "regulator-fixed";
> > > +               gpio = <&gpio2 RK_PC6 GPIO_ACTIVE_HIGH>;
> > > +               pinctrl-names = "default";
> > > +               pinctrl-0 = <&rtl8153_en_drv>;
> > > +               regulator-always-on;
> > > +               regulator-name = "vcc_rtl8153";
> > > +               regulator-min-microvolt = <5000000>;
> > > +               regulator-max-microvolt = <5000000>;
> >
> > This is just a simple switch, not an actual regulator.
> > It would make more sense to drop the voltage range and
> > instead have the implementation pass-through voltage
> > constraints from its parent.
>
> Thanks, I'll remove them in v2.
>
> >
> > > +               enable-active-high;
> > > +       };
> > > +
> > >         leds {
> > >                 compatible = "gpio-leds";
> > >                 pinctrl-0 = <&lan_led_pin>,  <&sys_led_pin>, <&wan_led_pin>;
> > > @@ -265,6 +277,12 @@
> > >                         };
> > >                 };
> > >         };
> > > +
> > > +       usb {
> > > +               rtl8153_en_drv: rtl8153-en-drv {
> > > +                       rockchip,pins = <2 RK_PC6 RK_FUNC_GPIO &pcfg_pull_none>;
> > > +               };
> > > +       };
> > >  };
> > >
> > >  &io_domains {
> > > @@ -364,6 +382,11 @@
> > >         dr_mode = "host";
> > >  };
> > >
> > > +&usbdrd3 {
> > > +       dr_mode = "host";
> > > +       status = "okay";
> >
> > Please also add a device node for the actual Ethernet controller, and
> > set up an aliases node for it, so that the bootloader has some place
> > to fill in a MAC address.
>
> But there's no valid (unique) MAC address for both this or on-board ethernet...
> They're non-existent in design.

Correct. And thanks for confirming that it's not just me and Robin that
got boards without the MAC address EEPROM...

If the user sets some MAC address in the bootloader environment by hand,
the bootloader could still potentially pass that MAC address to Linux
through the device tree. Whether the board has a valid address or not
is beside the point.


ChenYu


> Thanks,
> Tianling.
>
> >
> >
> > ChenYu
> >
> > > +};
> > > +
> > >  &usb_host0_ehci {
> > >         status = "okay";
> > >  };
> > > --
> > > 2.17.1
> > >
> > >
> > > _______________________________________________
> > > linux-arm-kernel mailing list
> > > linux-arm-kernel at lists.infradead.org
> > > http://lists.infradead.org/mailman/listinfo/linux-arm-kernel