This post provides a configuration example for Mellanox devices installed with MLNX_OFED running RoCE over a lossless network, in PCP-based QoS mode.
- Mellanox adapters and switches support DSCP based QoS and flow control, which is easier and simpler to configure and doesn't require VLANS, QoS is maintained across routers.
- QoS parameters are set on QP creation, When working with RDMA-CM it is possible to set QoS parameters for RDMA-CM created QPs
- Some of the configuration steps below can either be done permanently or temporarily (can be kept for the next boot).
For permanent configuration after running mlxconfig, a device reset (mlxfwreset) or host reboot is required.
Step 1 - Set QoS parameters
Map sk-prio 2 to SL 3 (Note: This command is nonpersistent)
# vconfig set_egress_map <vlan-interface> 2 3
[Optional] Set ToS to 106 (DSCP 26) for ALL RoCE traffic (Note: This command is nonpersistent)
# echo 106 > /sys/class/infiniband/<mlx-device>/tc/1/traffic_class
[Optional] Set the RDMA-CM ToS to 106 (DSCP 26) (Note: This command is nonpersistent)
# cma_roce_tos -d <mlx_dev> -t 106
[Optional] Enable ECN for TCP traffic (Note: This command is nonpersistent)
# sysctl -w net.ipv4.tcp_ecn=1
Step 2 - Enable PFC on RoCE prioritry
Activate PFC on priority 3
Method 1 - Using mlnx_qos tool (Note: This command is nonpersistent):
# mlnx_qos -i <interface> --pfc 0,0,0,1,0,0,0,0
Method 2 - Using LLDP DCBX, and configuration in the switch (Note: This requires LLDP service to be enabled in the switch)
# mlxconfig -d /dev/mst/mt4115_pciconf0 -y s LLDP_NB_DCBX_P1=TRUE LLDP_NB_TX_MODE_P1=2 LLDP_NB_RX_MODE_P1=2 LLDP_NB_DCBX_P2=TRUE LLDP_NB_TX_MODE_P2=2 LLDP_NB_RX_MODE_P2=2
<interface> refers to parent interface (for example ens2f0)
<vlan-interface> refers to vlan interface (for example ens2f0.100)
<mst-device> refers to MST device. (for example /dev/mst/mt4115_pciconf0)
<mlx-device> refers to mlx device (for example mlx5_0)