Re: [linux-next-v2 4/5] arm64: dts: rockchip: Add support of interrupt to ethernet node on Rock 3A SBC

[Anand Moon]

Hi Michael,

Thanks for your review comments.

On Tue, 22 Nov 2022 at 21:40, Michael Riesch
<michael.riesch@xxxxxxxxxxxxxx> wrote:
>
> Hi Anand,
>
> On 11/16/22 21:01, Anand Moon wrote:
> > As per the shematic gmac1 support gpio interrupt controller
>
> Typo "shematic" -> "schematic"
Ok,
>
> > GMAC1_INT/PMEB_GPIO3_A7 add the support for this.
>
> Maybe split the commit message into two proper sentences.

Ok this Interrupt is used for Power Management Event (supports 3.3V pull up).
Set low if received a magic packet or wake up frame; active low.
so this interrupt is used for suspend / resume.

> >
> > Signed-off-by: Anand Moon <linux.amoon@xxxxxxxxx>
> > ---
> > v2: new patch added
> > ---
> >  arch/arm64/boot/dts/rockchip/rk3568-rock-3a.dts | 10 +++++++++-
> >  1 file changed, 9 insertions(+), 1 deletion(-)
> >
> > diff --git a/arch/arm64/boot/dts/rockchip/rk3568-rock-3a.dts b/arch/arm64/boot/dts/rockchip/rk3568-rock-3a.dts
> > index 5378254c57ca..9f84a23a8789 100644
> > --- a/arch/arm64/boot/dts/rockchip/rk3568-rock-3a.dts
> > +++ b/arch/arm64/boot/dts/rockchip/rk3568-rock-3a.dts
> > @@ -588,10 +588,14 @@ rgmii_phy1: ethernet-phy@0 {
> >               compatible = "ethernet-phy-ieee802.3-c22";
> >               reg = <0x0>;
> >               pinctrl-names = "default";
> > -             pinctrl-0 = <&eth_phy_rst>;
> > +             pinctrl-0 = <&eth_phy_rst>, <&eth_phy_int>;
> >               reset-assert-us = <20000>;
> >               reset-deassert-us = <100000>;
> >               reset-gpios = <&gpio3 RK_PB0 GPIO_ACTIVE_LOW>;
> > +             interrupt-parent = <&gpio3>;
> > +             /* GMAC1_INT/PMEB_GPIO3_A7 */
>
> This comment is pretty superfluous.

Ok will drop this.

>
> > +             interrupts = <RK_PA7 IRQ_TYPE_LEVEL_LOW>;
> > +             #interrupt-cells = <1>;
>
> I am not an expert here, but I believe #interrupt-cells = <1>; means
> that the phy provides an array of interrupts. Are you sure this is
> correct? I find it strange that the phy driver consumes one interrupt
> and provides N interrupts?!
>
Ok will drop this.

> >       };
> >  };
> >
> > @@ -630,6 +634,10 @@ vcc_mipi_en: vcc_mipi_en {
> >       };
> >
> >       ethernet {
> > +             eth_phy_int: eth-phy-int {
> > +                     rockchip,pins = <3 RK_PA7 RK_FUNC_GPIO &pcfg_pull_down>;
>
> Interrupt is active low and you pull down the line here? There is a pull
> up resistor on sheet 11 of the schematic, so this does not seem right at
> all.
>
Actually this GPIO3_A7_d hence pull_down, this pin level triggered
So I will set this as &pcfg_pull_none.

I don't have PoE set up to verify these changes as of now.

> Best regards,
> Michael

Thanks

-Anand