Re: [dpdk-users] Running the quick start

["De Lara Guarch, Pablo" <pablo.de.lara.guarch@intel.com>]

Hi Sandeep,

> -----Original Message-----
> From: users [mailto:users-bounces@dpdk.org] On Behalf Of Sandeep
> Rayapudi
> Sent: Tuesday, August 16, 2016 2:27 PM
> To: users@dpdk.org
> Subject: [dpdk-users] Running the quick start
> 
> Hi all,
> 
> I've installed DPDK on the machine that has Intel NICs. I followed the
> steps from the Quick Start (
> http://dpdk.org/doc/quick-starthttp://dpdk.org/doc/quick-start)
> After making the NICs DPDK-COMPATIBLE, they are listed correctly.
> 
> Network devices using DPDK-compatible driver
> ============================================
> 0000:04:00.0 '82599ES 10-Gigabit SFI/SFP+ Network Connection' drv=igb_uio
> unused=ixgbe
> 0000:04:00.1 '82599ES 10-Gigabit SFI/SFP+ Network Connection' drv=igb_uio
> unused=ixgbe
> 0000:05:00.0 '82599ES 10-Gigabit SFI/SFP+ Network Connection' drv=igb_uio
> unused=ixgbe
> 0000:05:00.1 '82599ES 10-Gigabit SFI/SFP+ Network Connection' drv=igb_uio
> unused=ixgbe
> 
> Network devices using kernel driver
> ===================================
> 0000:02:00.0 'NetXtreme BCM5719 Gigabit Ethernet PCIe' if=eno1 drv=tg3
> unused=igb_uio *Active*
> 0000:02:00.1 'NetXtreme BCM5719 Gigabit Ethernet PCIe' if=eno2 drv=tg3
> unused=igb_uio
> 0000:02:00.2 'NetXtreme BCM5719 Gigabit Ethernet PCIe' if=eno3 drv=tg3
> unused=igb_uio
> 0000:02:00.3 'NetXtreme BCM5719 Gigabit Ethernet PCIe' if=eno4 drv=tg3
> unused=igb_uio
> 
> Other network devices
> =====================
> <none>
> 
> For DPDK-compatible NICs, the names of these interfaces is not shown, so
> I'm unable to pass name of the NIC in iface parameter.
> 
> build/app/testpmd -c7 -n3 --vdev=eth_pcap0,iface=eth0
> --vdev=eth_pcap1,iface=eth1 --
>                   -i --nb-cores=2 --nb-ports=2 --total-num-mbufs=2048
> 
> 
> If I pass the original name of the NIC before it is made a DPDK compatible,
> it gives the error:
> ./build/app/testpmd -c7 -n3 --vdev=eth_pcap0,iface=eno49
> --vdev=eth_pcap1,iface=eno50 -- -i --nb-cores=2 --nb-ports=2
> --total-num-mbufs=2048
> EAL: Detected 28 lcore(s)
> EAL: Probing VFIO support...
> PMD: Initializing pmd_pcap for eth_pcap0
> PMD: Couldn't open eno49: eno49: SIOCETHTOOL(ETHTOOL_GET_TS_INFO)
> ioctl
> failed: No such device
> PMD: Couldn't open interface eno49
> EAL: failed to initialize eth_pcap0 device
> PANIC in rte_eal_init():
> Cannot init pmd devices
> 6: [./build/app/testpmd(_start+0x29) [0x436d49]]
> 5: [/lib/x86_64-linux-gnu/libc.so.6(__libc_start_main+0xf0)
> [0x7f39d56da830]]
> 4: [./build/app/testpmd(main+0x47) [0x436047]]
> 3: [./build/app/testpmd(rte_eal_init+0xed2) [0x48aff2]]
> 2: [./build/app/testpmd(__rte_panic+0xc9) [0x4324ca]]
> 1: [./build/app/testpmd(rte_dump_stack+0x2b) [0x49256b]]
> Aborted (core dumped)
> 
> Any inputs on how to run this?

Once you bind those interfaces to igb_uio driver, you don't have any interface names anymore,
as they are not handled by the kernel.
If you want to use those ports, just call testpmd, with no extra parameters, and it will use the four ports.
They will be sorted with the PCI address, so port 0 will be 0000:04:00.0, port 1, 0000:04:00.1, etc.
If you only want to use two ports, let's say 0000:04:00.0 and 0000:05:00.0, you can whitelist those ports,
so the app will not use the other ones:

build/app/testpmd -c7 -n3 -w 0000:04:00.0 -w 0000:05:00.0 -- -i

Hope this helps,
Pablo

> 
> Thanks,
> Sandeep