| 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
| ||
|
Message-ID: <854a0d9c-17a2-c502-458d-4d02a2cd90bb@eks-engel.de>
Date: Wed, 5 Jun 2019 06:52:17 +0200
From: Benjamin Beckmeyer <beb@...-engel.de>
To: Andrew Lunn <andrew@...n.ch>
CC: <netdev@...r.kernel.org>
Subject: Re: DSA with MV88E6321 and imx28
Hey Andrew,
thanks for you reply.
I got the devicetree from somebody that is why German is in it. But first I wanted to get it running before I tidy it up.
The switch is strapped to single mode (so I can read SMI addresses 0x10-0x16 and 0x1b-0x1e directly). Do I have to
tell this the devicetree? I thought the driver will recognized that and I only give it the start SMI address of the switch?
I can read the internal PHYs with the indirectly read over the SMI COMMAND and SMI DATA registers.
The switch is not held in reset anymore, I take it out of reset in barebox manually every start, not a good solution but
for the moment to get it running it is okay.
The LCD port is just a GPIO. Like I said before, the devicetree is not completely from me. I've already started to write
a new one but then the MDIO bus can't be accessed anymore.
I will put some printk's in that function and see what is happening. What I think is confusing me, the switch is working
with internal addresses from 0-6 for the ports and you can read the port status registers at 0x10-0x16. And I'm not sure
which addresses should be in the switch mdio part.
Cheers,
Benjamin
> On Tue, Jun 04, 2019 at 03:07:25PM +0200, Benjamin Beckmeyer wrote:
>> Hi all,
>>
>> I'm working on a custom board with a 88E6321 and an i.MX28. Port 5 is directly connected per RMII to the CPU.
>> The switch is running in CPU attached mode. On Port 2 and 6 we have 2 external Micrel KSZ9031 PHYs.
>> Here is the snip of my device tree:
>>
>> &mac0 {
>> pinctrl-0 = <&mac0_pins_a &mac0_freigabe &mac0_lcd_d04>;
>> phy-supply = <®_3p3v>;
>> status = "okay";
>>
>> fixed-link = <1 1 100 0 0>;
> Hi Benjamin
>
> That is the old format for a fixed-link. Please use the new one.
>
>> /* this is done to remove enet_out */
>> clocks = <&clks 57>, <&clks 57>;
>> clock-names = "ipg", "ahb";
>>
>> /delete-property/ phy-reset-gpios;
>> /delete-property/ phy-reset-duration;
>> freigabe-gpios = <&gpio0 26 GPIO_ACTIVE_HIGH>;
> German in device tree?
>
>> trigger-gpios = <&gpio1 3 GPIO_ACTIVE_HIGH>;
>>
>> mdio {
>> #address-cells = <1>;
>> #size-cells = <0>;
>>
>> switch0: switch0@10 {
>> compatible = "marvell,mv88e6085";
>> reg = <0x10>;
> So you have the switch strapped to use a single address?
>
>> pinctrl-0 = <&lcd_d06_pins>;
> lcd ?
>
>> reset-gpios = <&gpio1 6 GPIO_ACTIVE_LOW>;
>>
>> dsa,member = <0 0>;
>>
>> ports {
>> #address-cells = <1>;
>> #size-cells = <0>;
>>
>> port@0 {
>> reg = <0x0>;
>> label = "Serdes0";
>> phy-handle = <&switch0phy0>;
>> };
>>
>> port@1 {
>> reg = <0x1>;
>> label = "Serdes1";
>> phy-handle = <&switch0phy1>;
>> };
>>
>> port@2 {
>> reg = <0x2>;
>> label = "lan1";
>> phy-handle = <&switch0phy2>;
>> };
>>
>> port@3 {
>> reg = <0x3>;
>> label = "lan2";
>> phy-handle = <&switch0phy3>;
>> };
>>
>> port@4 {
>> reg = <0x4>;
>> label = "lan3";
>> phy-handle = <&switch0phy4>;
>> };
>>
>> port5 {
>> reg = <0x5>;
>> label = "cpu";
>> ethernet = <&mac0>;
>> phy-mode = "rmii";
>> fixed-link {
>> speed = <100>;
>> full-duplex;
>> };
>> };
>>
>> port@6 {
>> reg = <0x6>;
>> label = "lan4";
>> phy-handle = <&switch0phy6>;
>> };
>> };
>>
>> mdio {
>> #address-cells = <1>;
>> #size-cells = <0>;
>> switch0phy0: switch0phy0@0 {
>> reg = <0xc>;
>> };
>> switch0phy1: switch0phy1@1 {
>> reg = <0xd>;
>> };
>> switch0phy2: switch0phy2@2 {
>> reg = <0x2>;
>> };
>> switch0phy3: switch0phy3@3 {
>> reg = <0x3>;
>> };
>> switch0phy4: switch0phy4@4 {
>> reg = <0x4>;
>> };
>> switch0phy6: switch0phy6@6 {
>> reg = <0x6>;
>> };
>> };
>> };
>> };
>> };
>>
>> I'm sure it must be wrong. Does the mdio part in between the switch part is for the internal
>> mdio bus?
> It is for the switch MDIO bus. For this generation of switch, it is
> both internal and external. Later generations have two MDIO busses,
> one internal and one external.
>
>> >From the outside I can read and write the SMI Register 0x10-0x16.
>>
>> Here is a snip from the bootup
>>
>> [ 1.377362] at24 0-0051: 256 byte 24c02 EEPROM, writable, 32 bytes/write
>> [ 1.391046] libphy: Fixed MDIO Bus: probed
>> [ 1.396763] libphy: mdio_driver_register: mv88e6085
>> [ 1.407168] fec 800f0000.ethernet (unnamed net_device) (uninitialized): Invalid MAC address: 00:00:00:00:00:00
>> [ 1.417279] fec 800f0000.ethernet (unnamed net_device) (uninitialized): Using random MAC address: 86:50:72:5d:79:ad
>> [ 1.429918] libphy: fec_enet_mii_bus: probed
>> [ 1.434374] mdio_bus 800f0000.ethernet-1:10: mdio_device_register
>> ---
>> [ 18.735835] Generic PHY fixed-0:00: attached PHY driver [Generic PHY] (mii_bus:phy_addr=fixed-0:00, irq=-1)
> You probably want to put some printk mv88e6xxx_detect(). Is it getting
> called? What value does id have? When these switches are held in
> reset, they don't respond on the bus at all, so you will get MDIO
> reads of 0xffff.
>
> Andrew
Powered by blists - more mailing lists