Clearfog pro : how use the switch and eth1

OCTEON CN9130
22

Hi i thought i’d provide more details.

Eth1 which seems to be configured sgmii link mode remains down.

ethtool output:

root@8782b935e01b:~# ethtool eth1
Settings for eth1:
        Supported ports: [ TP    MII ]
        Supported link modes:   10baseT/Half 10baseT/Full
                                100baseT/Half 100baseT/Full
                                1000baseT/Full
        Supported pause frame use: Symmetric Receive-only
        Supports auto-negotiation: Yes
        Supported FEC modes: Not reported
        Advertised link modes:  10baseT/Half 10baseT/Full
                                100baseT/Half 100baseT/Full
                                1000baseT/Full
        Advertised pause frame use: Symmetric Receive-only
        Advertised auto-negotiation: Yes
        Advertised FEC modes: Not reported
        Speed: Unknown!
        Duplex: Unknown! (255)
        Auto-negotiation: on
        master-slave cfg: preferred slave
        master-slave status: unknown
        Port: MII
        PHYAD: 1
        Transceiver: external
        Link detected: no

I obviously notice the unknown against speed and negation. attemtp to change this via ethtool i get a NETLINK errors:

oot@8782b935e01b:~# ethtool -s eth1 autoneg off speed 1000
netlink error: link settings update failed
netlink error: Invalid argument

dmesg output for eth1:

root@8782b935e01b:~# dmesg | grep eth1
[    2.049956] mvpp2 f2000000.ethernet eth1: Using firmware node mac address 06:f9:68:35:a2:9c
[  959.126486] mvpp2 f2000000.ethernet eth1: PHY [f212a200.mdio-mii:01] driver [Generic PHY] (irq=POLL)
[  959.137183] mvpp2 f2000000.ethernet eth1: configuring for phy/sgmii link mode
[ 1071.527988] br0: port 1(eth1) entered blocking state
[ 1071.533071] br0: port 1(eth1) entered disabled state
[ 1071.538173] mvpp2 f2000000.ethernet eth1: entered allmulticast mode
[ 1071.544926] mvpp2 f2000000.ethernet eth1: entered promiscuous mode


root@8782b935e01b:~# ip link
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN mode DEFAULT group default qlen 1000
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
2: eth0: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc mq state DOWN mode DEFAULT group default qlen 2048
    link/ether 8e:74:99:e6:ac:57 brd ff:ff:ff:ff:ff:ff
3: eth1: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc mq master br0 state DOWN mode DEFAULT group default qlen 2048
    link/ether 06:f9:68:35:a2:9c brd ff:ff:ff:ff:ff:ff
4: eth2: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc mq master br0 state DOWN mode DEFAULT group default qlen 2048
    link/ether aa:a4:4a:7a:30:eb brd ff:ff:ff:ff:ff:ff
5: br0: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc noqueue state DOWN mode DEFAULT group default qlen 1000
    link/ether ae:2f:d8:fd:6a:69 brd ff:ff:ff:ff:ff:ff

my pastbin of dmesg.
https://pastebin.com/uX8NEftB

I’ve setup interface br0. eth1 and eth2 both masters of br0.

eth1 is PHY in DTS.

Any help if appreciated - I getting really tempted to jump back to my old build which has worked since 2022.

this current build - Debian bookworm clean build using build scripts and patched. kernel-6.10. (same eth1 issue occurs on ubuntu 24.04, kernel 6.13).

Is it something to do with SGMII, do i have to configure anything, are they are hardware changes required on clearfog base as illuded to above. I’ve been trawling the internet and finding very little that can help me diagnose this. Even looking at how i can disable SGMII, as i say this doesn’t occur with my old build.

regards
Adrian

23

done it - after comparing it with my old build, I noticed the new build was using generic PHY linux driver. doh.

[  959.126486] mvpp2 f2000000.ethernet eth1: PHY [f212a200.mdio-mii:01] driver [Generic PHY]

[  351.012757] mvpp2 f2000000.ethernet eth1: PHY [f212a200.mdio-mii:01] driver [Marvell 88E1510] (irq=POLL)

eth1 is now up. Strange eth1 would stay down using the generic driver, but Eth2 would go up. Assume it must be something to do with eth1 being PHY interface.