Script D: Switchdev Mode with LAG Configuration

The following commands are used to create and bring up a VF in switchdev mode, and to configure LAG in active-backup mode. Note:ice_​lag DDP package can be downloaded from Git Repository and it can be found in this path: linux-firmware/intel/ice/ddp-lag). =========================================================================================== #!/bin/bash set -x #set -e DEVLINK=devlink PF1=ens2f0 PF1_PCI="pci/0000:17:00.0" PF2=ens2f1 BR=br1 VF1_PCI=0000:17:01.0 VF1=ens2f0v0 VF1_MAC= ea:fe:27:fa:5d:a6 VF1_IP=172.31.123.11 MASK=24 VF1_PR=eth0 UL_PR=ens2f0 BOND=bond0 #1. Load bonding driver modprobe bonding mode=active-backup miimon=100 max_bonds=1 or ip link add $<BOND> type bond mode active-backup miimon 100 #2. Create a bridge ip link add $<BR> type bridge #3. Change the eswitch mode to switchdev and create a VF devlink dev eswitch set $<PF1_PCI> mode switchdev echo <num_of_VFs> > /sys/class/net/$<PF1>/device/sriov_numvfs #4. Add the PF interfaces to bond ip l s $<PF1> master $<BOND> ip l s $<PF2> master $<BOND> #5. Verify in the dmesg that the PF1 is the active one dmesg -wH #6. Add the bond interface to the bridge ip link set $<BOND>master $<BR> #7. Add the port representor to the bridge ip l s $<VF_PR> master $<BR> #8. Bring all the interfaces up ip l s $<PF1> up ip l s $<PF2> up ip l s $<VF_PR> up ip l s $<BOND>up ip l s $<BR> up #9. Configure IP address on the VF and send traffic to the VF ip addr add $VF1_IP/$MASK dev $VF1 #10. Check the bonding mode and current active slave cat /proc/net/bonding/$BOND #11. To offload tc filters to the hardware hw-tc-offload must be enabled on the VFs Port Representor (VF_PR) ethtool -K $PF1 hw-tc-offload on ethtool -K $VF1_PR hw-tc-offload on # Verify settings: ethtool -k $PF1 | grep "hw-tc" ethtool -k $VF1_PR | grep "hw-tc" #12. Enable ingress qdisc on PF1 (Uplink port) and VF Port Representors $TC qdisc add dev $PF1_ingress $TC qdisc add dev $VF1 ingress #13. Add filter with skip_sw to offload to hardware $TC filter add dev $VF1_PR ingress protocol ip prio 1 flower src_mac $VF1_MAC dst_mac $PEER_MAC skip_sw action mirred egress redirect dev $PF1 $TC filter add dev $PF1 ingress protocol ip prio 1 flower src_mac $PEER_MAC dst_mac $VF1_MAC skip_sw action mirred egress redirect dev $VF1_PR #14. Change active slave to second PF ifenslave $BOND $PF2 -c or echo $PF2 > /sys/class/net/$BOND/bonding/active_slave #15. Verify the active slave change in the dmesg and also in the bonding mode cat /proc/net/bonding/$BOND ===========================================================================================