Unable to use paravirtualized port created from vhost-user-client after dpdk-devbind

Question

Some context first:

I installed and configured OVS-DPDK on VM0 [ubuntu + qemu/kvm + ovs-dpdk]. As a guest running on top of VM0 I have VM1 [ubuntu + dpdk].

After a bit of googling I was able to create vhost-user-client port in OVS and consume it with VM1.

At this point I can see the product of vhost-user-client port by running:

$ sudo dpdk-devbind.py --status
Network devices using kernel driver
===================================
0000:01:00.0 'Virtio network device 1041' if=enp1s0 drv=virtio-pci unused=vfio-pci *Active*
0000:07:00.0 'Virtio network device 1041' if=enp7s0 drv=virtio-pci unused=vfio-pci *Active* <--- this is it

However after binding it using command:

sudo dpdk-devbind.py --bind=virtio-pci 0000:07:00.0

When running dpdk-testpmd, I get nb forwarding ports=0

$ sudo dpdk-testpmd
EAL: Detected 3 lcore(s)
EAL: Detected 1 NUMA nodes
EAL: Detected static linkage of DPDK
EAL: Multi-process socket /var/run/dpdk/rte/mp_socket
EAL: Selected IOVA mode 'PA'
EAL: No available hugepages reported in hugepages-1048576kB
EAL: Probing VFIO support...
EAL: VFIO support initialized
EAL: Probe PCI driver: net_virtio (1af4:1041) device: 0000:01:00.0 (socket 0)
testpmd: No probed ethernet devices
testpmd: create a new mbuf pool <mb_pool_0>: n=163456, size=2176, socket=0
testpmd: preferred mempool ops selected: ring_mp_mc
Done
No commandline core given, start packet forwarding
io packet forwarding - ports=0 - cores=0 - streams=0 - NUMA support enabled, MP allocation mode: native

  io packet forwarding packets/burst=32
  nb forwarding cores=1 - nb forwarding ports=0
Press enter to exit

I dont see any clues for this in dmesg and not sure where else to look.

How can I debug this issue

Answer 1

As hinted in the comments, DPDK ports will work with uio_pci_generic | vfio-pci | igb_uio. Hence for a virtio network device to be used with DPDK it needs binding with vfio-pci.

Please follow the steps as

1. ifconfig enp7s0 down
2. sudo modprobe uio
3. sudo modprobe vfio-pci (for newer kernel this can be skipped)
4. sudo echo 1 > /sys/module/vfio/parameters/enable_unsafe_noiommu_mode
5. sudo dpdk-devbind.py --bind=vfio-pci 0000:07:00.0

One done successfully dpdk application with virtio net can be used.

Note: as mentioned in the comments, it is not clear why virtio net device 0000:07:00.0 is bind with virtio-pci