[dpdk-dev] DPDK 2.2 issue with dpdk kni-virtio

chintu hetam rometoroam at gmail.com
Fri Apr 8 17:48:08 CEST 2016


Hello All,

We are trying to test virtio-vhost-kni performance,

I able to start kni application but after that when i looked for
/sys/class/net/vEth0, there's no sock_fd and sock_en parameters, hence i am
stuck from attaching fd to virtio-qemu.

Host: Fedora 23
HugetblsFS: 1G pages

Following is my config.

[root at localhost dpdk-2.2.0]# ./tools/dpdk_nic_bind.py --status

Network devices using DPDK-compatible driver
============================================
0000:02:00.2 'I350 Gigabit Network Connection' drv=igb_uio unused=igb
0000:02:00.3 'I350 Gigabit Network Connection' drv=igb_uio unused=igb
0000:86:00.0 'Ethernet Controller 10-Gigabit X540-AT2' drv=igb_uio
unused=ixgbe
0000:86:00.1 'Ethernet Controller 10-Gigabit X540-AT2' drv=igb_uio
unused=ixgbe

Network devices using kernel driver
===================================
0000:02:00.0 'I350 Gigabit Network Connection' if=ens2f0 drv=igb
unused=igb_uio *Active*
0000:02:00.1 'I350 Gigabit Network Connection' if=ens2f1 drv=igb
unused=igb_uio
0000:89:00.0 'Ethernet Controller 10-Gigabit X540-AT2' if=enp137s0f0
drv=ixgbe unused=igb_uio
0000:89:00.1 'Ethernet Controller 10-Gigabit X540-AT2' if=enp137s0f1
drv=ixgbe unused=igb_uio

Other network devices
=====================
<none>


 examples/kni/build/app/kni -c 0xf -n 4 -- -p 0xc -P
--config="(2,0,1),(3,2,3)"
:
:
:
PP: Port ID: 3
APP: Rx lcore ID: 2, Tx lcore ID: 3
APP: Initialising port 2 ...
PMD: ixgbe_dev_rx_queue_setup(): sw_ring=0x7f1d3ffef6c0
sw_sc_ring=0x7f1d3ffef180 hw_ring=0x7f1d3ffefc00 dma_addr=0x533ffefc00
PMD: ixgbe_dev_tx_queue_setup(): sw_ring=0x7f1d3ffdcfc0
hw_ring=0x7f1d3ffdf000 dma_addr=0x533ffdf000
PMD: ixgbe_set_tx_function(): Using simple tx code path
PMD: ixgbe_set_tx_function(): Vector tx enabled.
PMD: ixgbe_set_rx_function(): Vector rx enabled, please make sure RX burst
size no less than 4 (port=2).
KNI: pci: 86:00:00       8086:1528
APP: Initialising port 3 ...
PMD: ixgbe_dev_rx_queue_setup(): sw_ring=0x7f1d3ffcc6c0
sw_sc_ring=0x7f1d3ffcc180 hw_ring=0x7f1d3ffccc00 dma_addr=0x533ffccc00
PMD: ixgbe_dev_tx_queue_setup(): sw_ring=0x7f1d3ffb9fc0
hw_ring=0x7f1d3ffbc000 dma_addr=0x533ffbc000
PMD: ixgbe_set_tx_function(): Using simple tx code path
PMD: ixgbe_set_tx_function(): Vector tx enabled.
PMD: ixgbe_set_rx_function(): Vector rx enabled, please make sure RX burst
size no less than 4 (port=3).
KNI: pci: 86:00:01       8086:1528

KNI: pci: 86:00:01       8086:1528

Checking link status
............................done
Port 2 Link Up - speed 100 Mbps - full-duplex
Port 3 Link Up - speed 100 Mbps - full-duplex
APP: Lcore 1 is writing to port 2
APP: Lcore 2 is reading from port 3
APP: Lcore 3 is writing to port 3
APP: Lcore 0 is reading from port 2

 ls /sys/class/net/vEth2/
addr_assign_type  broadcast        dev_id    duplex             ifalias
 link_mode         netdev_group  phys_port_name  proto_down  statistics
 type
address           carrier          dev_port  flags              ifindex
 mtu               operstate     phys_switch_id  queues      subsystem
uevent
addr_len          carrier_changes  dormant   gro_flush_timeout  iflink
name_assign_type  phys_port_id  power           speed       tx_queue_len

As this is sysfs, it has to be in kmod, but i don't see any mention of
these two parameters, what's going on?

Any help on what causes this issue?


-thanks-


More information about the dev mailing list