We are working with a Dual 10G Ethernet PHY Device (Vitesse vcs8488) and would like to represent it in the device tree for several of our boards. This device has some registers and pins that are common to both of the PHY units on the die. Other registers are standard IEEE803.3-C45.

Here is my first (hand-waving) take at how it should look:


ethernet-phy-nexus@0 {
        reg = <0x00>;
        /* The Vitesse VSC8488 is a dual-PHY where
         * some of the configuration is common across
         * both of the phy devices such as the reset
         * line and the base MDIO address.
         */
        compatible = "vitesse,vsc8488-nexus", "ethernet-phy-nexus";
        #address-cells = <1>;
        #size-cells = <0>;
        ranges;
        
        /* Hardware reset signal */
        reset = <&gpio 17 0>;

        /* Other common initialization (laser power, etc. )*/

        phy30: ethernet-phy@0 {
                /* Absolute address */
                reg = <0x0>;
                compatible = "vitesse,vsc8488", "ethernet-phy-ieee802.3-c45";
                interrupt-parent = <&gpio>;
                interrupts = <13 8>;

                /* TX Fault GPIO line */
                tx_fault = <1>;
                /* GPIO that enables output */
                txon = <4>;
                /* INT A GPIO output */
                inta = <5>;
        };

        phy31: ethernet-phy@1 {
                /* Absolute address */
                reg = <0x1>;
                compatible = "vitesse,vsc8488", "ethernet-phy-ieee802.3-c45";
                interrupt-parent = <&gpio>;
                interrupts = <13 8>;

                /* TX Fault GPIO line */
                tx_fault = <8>;
                /* GPIO that enables output */
                txon = <10>;
                /* INT A GPIO output */
                inta = <5>;
        };
};


We would then modify of_mdio.c so that when it encountered an "ethernet-phy-nexus", it would add all the children of the nexus.

Comments?  Objections?

Thanks, and let me know what you think,
David Daney

_______________________________________________
devicetree-discuss mailing list
devicetree-discuss@lists.ozlabs.org
https://lists.ozlabs.org/listinfo/devicetree-discuss

Reply via email to