netdev - DSA: external phy address and port number of switch conflicts

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [thread-next>] [day] [month] [year] [list]

Date:   Thu, 13 Dec 2018 19:12:05 -0500
From:   John Rama <john.rama01@...il.com>
To:     netdev@...r.kernel.org
Subject: DSA: external phy address and port number of switch conflicts

Dear DSA experts

My name is John.

I'm working for our custom board which has the Marvel 88e5050 ethernet switch
as shown below, and trying to make the system works using DSA subsystem.

I found one problem and have a question.

- System
    - i.MX6 and Marvel 88e5050 Switch(PHY addr is 0x10)
    - port1-5 has integrated PHY
     - port7 is connected to Marvel 88e1510 PHY(PHY addr is 0x1)

                           0x10
    +------------+         +-----------------+
    |            |         | Marvel 88e5050  |
    |            | RGMII   |           port1 |
    |   IMX6     +---------+ port6           |
    |            |         |           port2 |
    |            | MDIO    |                 |
    |            +-+-------+           port3 |
    |            | |       |                 |
    +------------+ |       |           port4 |
                   |       |                 |
                 +-+-+     |           port5 |
                 |   |SGMII|                 |
                 |PHY|-----| port7           |
                 |   |     |                 |
                 +---+     +-----------------+
              Marvel 88e1510
                   0x1                        

- Device Tree setting
    In this configuration, our dts is as followings.

    dsa@0 {
      compatible = "marvell,dsa";
        #address-cells = <2>;
        #size-cells = <0>;

        dsa,ethernet = <&fec>;
        dsa,mii-bus = <&mdio>;

        switch@0 {
            #address-cells = <1>;
            #size-cells = <0>;
            reg = <0x10 0>;    /* MDIO address 16, switch 0 in tree */

            port@1 {
                reg = <1>;
                label = "port1";
            };

            port@2 {
                reg = <2>;
                label = "port2";
            };

            port@3 {
                reg = <3>;
                label = "port3";
            };

            port@4 {
                reg = <4>;
                label = "port4";
            };

            port@5 {
                reg = <5>;
                label = "port5";
            };

            port@6 {
                reg = <6>;
                label = "cpu";
            };

            port@7 {
                reg = <7>;
                label = "port7";
                phy-handle = <&ethphy1>;
                phy-mode = "sgmii";
            };                                       
        };
     };

    mdio: mdio {
        #address-cells = <1>;
        #size-cells = <0>;

        ethphy1: 88e1510{
            reg = <1>;
            phy-mode = "sgmii";
        };

    };
                                
- Problem
    dsa_slave_phy_connect() for port 7 results in attaching Generic PHY,
    as shown in the following log, which is unexpected result.
    dsa dsa@0 port7 (uninitialized): attached PHY at address 1 [Generic PHY]

    This is because port7 PHY addr (reg=<1>) and the port1 port number (reg=<1)> is conflicting.
    If I comment out the port1 setting from DTS, I can get following log,
    everything working fine and

    dsa dsa@0 port7 (uninitialized): attached PHY at address 1 [Marvel 88E1510]

    I'm using a little bit old kernel,4.1.15.
    So I checked the latest code of linux-next, but it seems there is similar problem.

- Question
    Is this system configuration is not supported by DSA ?
    Or am I misunderstand something ?
    Any insights are really appreciated.

John