[dpdk-users] Correct setup of sfc

[Filip Janiszewski]

Hi,

I'm trying to test a SF card (Flareon Ultra SFN7142Q Dual-Port 40GbE) in
our testing box, the details of the device are:

.
Solarstorm firmware update utility [v7.1.1]
Copyright Solarflare Communications 2006-2018, Level 5 Networks 2002-2005

enp101s0f0 - MAC: 00-0F-53-2C-3A-10
    Firmware version:   v7.1.1
    Controller type:    Solarflare SFC9100 family
    Controller version: v6.2.7.1000
    Boot ROM version:   v5.1.0.1005

The Boot ROM firmware is up to date
The controller firmware is up to date

enp101s0f1 - MAC: 00-0F-53-2C-3A-11
    Firmware version:   v7.1.1
    Controller type:    Solarflare SFC9100 family
    Controller version: v6.2.7.1000
    Boot ROM version:   v5.1.0.1005

The Boot ROM firmware is up to date
The controller firmware is up to date
.

The DPDK lib (version 18.05) has been build with
CONFIG_RTE_LIBRTE_SFC_EFX_PMD=y as per instructions, the sfc driver
seems to be loaded, the output of 'lsmod | grep sfc' is:

.
sfc                   470393  0
.

In the kernel logs i see that the card links are ready:

.
[  178.496825] sfc 0000:65:00.0 enp101s0f0: link up at 40000Mbps
full-duplex (MTU 1500)
[  178.496856] IPv6: ADDRCONF(NETDEV_CHANGE): enp101s0f0: link becomes ready
[  178.497789] sfc 0000:65:00.1 enp101s0f1: link up at 40000Mbps
full-duplex (MTU 1500)
[  178.498356] IPv6: ADDRCONF(NETDEV_CHANGE): enp101s0f1: link becomes ready
.

But DPDK seems to not recognize properly the card, it is visible in the
initial EAL logs:

.
EAL: Detected 28 lcore(s)
EAL: Detected 1 NUMA nodes
EAL: Multi-process socket /var/run/dpdk/rte/mp_socket
EAL: Probing VFIO support...
EAL: PCI device 0000:00:1f.6 on NUMA socket 0
EAL:   probe driver: 8086:15b8 net_e1000_em
EAL: PCI device 0000:65:00.0 on NUMA socket 0
EAL:   probe driver: 1924:923 net_sfc_efx
EAL: PCI device 0000:65:00.1 on NUMA socket 0
EAL:   probe driver: 1924:923 net_sfc_efx
EAL: PCI device 0000:b3:00.0 on NUMA socket 0
EAL:   probe driver: 15b3:1015 net_mlx5
net_mlx5: MPLS over GRE/UDP tunnel offloading disabled due to old
OFED/rdma-core version or firmware configuration
EAL: PCI device 0000:b3:00.1 on NUMA socket 0
EAL:   probe driver: 15b3:1015 net_mlx5
net_mlx5: MPLS over GRE/UDP tunnel offloading disabled due to old
OFED/rdma-core version or firmware configuration
.

but perhaps the driver probe net_sfc_efx seems to not work as
rte_eth_dev_count_avail returns only 2 instead of 4 (there's another MLX
card on board).

Any suggestion on what might be missing here?

Thanks

-- 
BR, Filip
+48 666 369 823