| 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
| ||
|
Date: Tue, 12 Dec 2023 12:27:32 +0300
From: Serge Semin <fancer.lancer@...il.com>
To: Richard Tresidder <rtresidd@...ctromag.com.au>
Cc: Andrew Lunn <andrew@...n.ch>, Jakub Kicinski <kuba@...nel.org>,
vinschen@...hat.com, netdev@...r.kernel.org
Subject: Re: STMMAC Ethernet Driver support
On Tue, Dec 12, 2023 at 11:57:22AM +0800, Richard Tresidder wrote:
>
> <font face="monospace">Richard Tresidder</font>
>
>
> On 12/12/2023 12:16 am, Andrew Lunn wrote:
> > > We use the SOC's internal STMMAC interface to connect to a Marvel switch IC
> > > and expose each port individually using vlan, I'd forgot that part.
> > > It's an 88E6352-xx-TFJ2I000 device utilising the 'marvell,mv88e6085'
> > > compatible driver in drivers\net\dsa\mv88e6xxx
> > Its odd you need VLANs. Each port should already be exposed to the
> > host as netdev interfaces. That is what DSA does.
> >
> > Andrew
> Hi Andrew
> I'll read further on that one as this is the first time I've had to dig
> into this side of the system.
> It had always "just worked".
> The ports show up in an 'ip l' list in the same style as a vlan with an @
> symbol, naming isn't quite vlan style though.
> That in concert with the fact this 'vlan_feature' line broke things has
> possibly distorted my view of how they're propagated.
> It's a rather trimmed down busybox image, so I'm missing some tools I'd
> usually use to examine stuff.
>
> This is the config in the dts
> **************************************
> //------------------------------------------------------------------------------
> // connected to dsa network switch
> &gmac1 {
> clock-names = "stmmaceth", "clk_ptp_ref";
Just a side note irrelevant to the topic. You might be interested to
know STMMAC driver expects to have the PTP reference clock source
passed with the "ptp_ref" name, not "clk_ptp_ref".
-Serge(y)
> clocks = <&emac1_clk &hps_eosc1>;
> f2h_ptp_ref_clk;
> fixed-link {
> speed = <1000>;
> full-duplex;
> };
> };
>
> //------------------------------------------------------------------------------
> &mdio1 {
> #address-cells = <1>;
> #size-cells = <0>;
>
> switch0: switch0@0 {
> compatible = "marvell,mv88e6085";
> #address-cells = <1>;
> reg = <0>;
> //reset-gpios = <&pio_a0 2 GPIO_ACTIVE_LOW>;
>
> dsa,member = <0 0>;
>
> ports {
> #address-cells = <1>;
> #size-cells = <0>;
>
> port@2 {
> reg = <2>;
> label = "lan1";
> phy-handle = <&switch1phy2>;
> };
>
> port@3 {
> reg = <3>;
> label = "lan2";
> phy-handle = <&switch1phy3>;
> };
>
> port@4 {
> reg = <4>;
> label = "lan3";
> phy-handle = <&switch1phy4>;
> };
>
> port@5 {
> reg = <5>;
> label = "wifi";
> fixed-link {
> speed = <100>;
> full-duplex;
> };
> };
>
> port@6 {
> reg = <6>;
> label = "cpu";
> ethernet = <&gmac1>;
> fixed-link {
> speed = <1000>;
> full-duplex;
> };
> };
>
> };
>
> mdio {
> #address-cells = <1>;
> #size-cells = <0>;
> switch1phy2: switch1phy2@2 {
> reg = <2>;
> marvell,reg-init = <0 0x10 0 0x0200>; // Sense only on Rx Energy
> Detect, no FLPs sents
> };
> switch1phy3: switch1phy3@3 {
> reg = <3>;
> marvell,reg-init = <0 0x10 0 0x0200>; // Sense only on Rx Energy
> Detect, no FLPs sents
> };
> switch1phy4: switch1phy4@4 {
> reg = <4>;
> marvell,reg-init = <0 0x10 0 0x0200>; // Sense only on Rx Energy
> Detect, no FLPs sents
> };
> };
>
> };
> };
> *************************************************
>
> This is how they appear using 'ip l'
> The @ symbol got me as I've usually associated this with vlan's in my day to
> day networking.
> But the config files don't configure them as vlans.
> *************************************************
> 1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue qlen 1000
> link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
> 2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1508 qdisc mq qlen 1000
> link/ether 04:91:62:cf:4f:06 brd ff:ff:ff:ff:ff:ff
> 3: sit0@...E: <NOARP> mtu 1480 qdisc noop qlen 1000
> link/sit 0.0.0.0 brd 0.0.0.0
> 4: lan1@...0: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc noqueue
> qlen 1000
> link/ether 04:91:62:cf:4f:06 brd ff:ff:ff:ff:ff:ff
> 5: lan2@...0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue qlen
> 1000
> link/ether 04:91:62:cf:4f:06 brd ff:ff:ff:ff:ff:ff
> 6: lan3@...0: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc noqueue
> qlen 1000
> link/ether 04:91:62:cf:4f:06 brd ff:ff:ff:ff:ff:ff
> 7: wifi@...0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue qlen
> 1000
> link/ether 04:91:62:cf:4f:06 brd ff:ff:ff:ff:ff:ff
>
> *************************************************
>
> And these are the systemd config files
> No vlan mentioned in there..
> *************************************************
> /etc/systemd/network/eth0.network
> [Match]
> Name=eth0
>
> [Network]
>
> /etc/systemd/network/lan.network
> [Match]
> Name=lan*
>
> [Network]
> DHCP=yes
> BindCarrier=eth0
> LinkLocalAddressing=yes
>
> [DHCP]
> UseDomains=yes
> ClientIdentifier=mac
>
> *************************************************
>
> Cheers
> Richard
>
Powered by blists - more mailing lists