[dpdk-users] Using KNI with virtio-net-pci

[dpdk-users] Using KNI with virtio-net-pci

[Ruslan Osmanov]

Hi,

I'm going to develop a DPDK application on a laptop, but the laptop's
hardware is not supported by DPDK. Furtunately, DPDK supports
paravirtualized devices(http://www.dpdk.org/doc/nics) including
virtio-net.  

So I'm trying to configure a QEMU guest for running the Kernel NIC
Interface(KNI) on a virtio-net-pci device. The problem is that the KNI
sample application doesn't accept the virtio-pci driver.


QEMU command:

eth_device=virtio-net-pci
exec qemu-system-x86_64 -enable-kvm \
  -cpu host -smp 2 \
  -vga std \
  -mem-prealloc -mem-path /dev/hugepages \
  -drive file=GentooVM.img,if=virtio \
  -netdev user,id=vmnic,hostname=gentoo \
  -device $eth_device,netdev=vmnic \
  -m 1024M \
  -monitor stdio \
  -name "Gentoo VM"

Running the KNI sample application in the guest:

sudo ./examples/kni/build/app/kni -c 0x3 -n 4 -- \
-P -p 0x1 --config="(0,0,1)"

EAL: Detected lcore 0 as core 0 on socket 0
EAL: Detected lcore 1 as core 0 on socket 0
EAL: Support maximum 128 logical core(s) by configuration.
EAL: Detected 2 lcore(s)
EAL: Probing VFIO support...
EAL:   IOMMU type 1 (Type 1) is supported
EAL:   IOMMU type 8 (No-IOMMU) is not supported
EAL: VFIO support initialized
EAL: Setting up physically contiguous memory...
...
EAL: WARNING: cpu flags constant_tsc=yes nonstop_tsc=no -> using
unreliable clock cycles !
EAL: Master lcore 0 is ready (tid=657d58c0;cpuset=[0])
PMD: rte_igbvf_pmd_init():  >>
EAL: lcore 1 is ready (tid=305ff700;cpuset=[1])
EAL: PCI device 0000:00:03.0 on NUMA socket -1
EAL:   probe driver: 1af4:1000 rte_virtio_pmd
EAL:   Not managed by a supported kernel driver(0), skipped
PMD: virtio_read_caps(): failed to map pci device!
PMD: vtpci_init(): trying with legacy virtio pci.
Segmentation fault

$ lspci
...
00:03.0 Ethernet controller: Red Hat, Inc Virtio network device

I've noticed that pci_scan_one() function sets dev->kdrv =
RTE_KDRV_NONE in lib/librte_eal/linuxapp/eal/eal_pci.c, while the
driver is detected as virtio-pci
(from /sys/bus/pci/devices/0000:00:03.0/driver).

Is it even possible to run KNI with virtio-net-pci device?

If it's impossible, then are there other options?

-- 
Ruslan Osmanov

Re: [dpdk-users] Using KNI with virtio-net-pci

[Ruslan Osmanov]

nobody can answer this simple question? I'm just trying to understand
why the sample KNI application fails.

sudo ./examples/kni/build/app/kni -c 0x3  -- -P -p 0x1 \
  --config="(0,1,0)"

EAL: Master lcore 0 is ready (tid=5f4e48c0;cpuset=[0])
PMD: rte_igbvf_pmd_init():  >>
EAL: lcore 1 is ready (tid=2b9ff700;cpuset=[1])
EAL: PCI device 0000:00:03.0 on NUMA socket -1
EAL:   probe driver: 8086:100e rte_em_pmd
EAL:   Not managed by a supported kernel driver, skipped
EAL: PCI device 0000:00:04.0 on NUMA socket -1
EAL:   probe driver: 1af4:1000 rte_virtio_pmd
EAL:   Not managed by a supported kernel driver, skipped
PMD: virtio_read_caps(): failed to map pci device!
PMD: vtpci_init(): trying with legacy virtio pci.
Segmentation fault


python2 ./tools/cpu_layout.py 
============================================================
Core and Socket Information (as reported by /proc/cpuinfo)
============================================================

cores =  [0]
sockets =  [0, 1]

       Socket 0      Socket 1      
       --------      --------      
Core 0 [0]           [1]           


Sorry, but the output is not very informative. Please help me to figure
out what's wrong with it.

On Thu, 2 Jun 2016 10:43:23 +0700
Ruslan Osmanov <rrosmanov@gmail.com> wrote:

> Hi,
> 
> I'm going to develop a DPDK application on a laptop, but the laptop's
> hardware is not supported by DPDK. Furtunately, DPDK supports
> paravirtualized devices(http://www.dpdk.org/doc/nics) including
> virtio-net.  
> 
> So I'm trying to configure a QEMU guest for running the Kernel NIC
> Interface(KNI) on a virtio-net-pci device. The problem is that the KNI
> sample application doesn't accept the virtio-pci driver.
> 
> 
> QEMU command:
> 
> eth_device=virtio-net-pci
> exec qemu-system-x86_64 -enable-kvm \
>   -cpu host -smp 2 \
>   -vga std \
>   -mem-prealloc -mem-path /dev/hugepages \
>   -drive file=GentooVM.img,if=virtio \
>   -netdev user,id=vmnic,hostname=gentoo \
>   -device $eth_device,netdev=vmnic \
>   -m 1024M \
>   -monitor stdio \
>   -name "Gentoo VM"
> 
> Running the KNI sample application in the guest:
> 
> sudo ./examples/kni/build/app/kni -c 0x3 -n 4 -- \
> -P -p 0x1 --config="(0,0,1)"
> 
> EAL: Detected lcore 0 as core 0 on socket 0
> EAL: Detected lcore 1 as core 0 on socket 0
> EAL: Support maximum 128 logical core(s) by configuration.
> EAL: Detected 2 lcore(s)
> EAL: Probing VFIO support...
> EAL:   IOMMU type 1 (Type 1) is supported
> EAL:   IOMMU type 8 (No-IOMMU) is not supported
> EAL: VFIO support initialized
> EAL: Setting up physically contiguous memory...
> ...
> EAL: WARNING: cpu flags constant_tsc=yes nonstop_tsc=no -> using
> unreliable clock cycles !
> EAL: Master lcore 0 is ready (tid=657d58c0;cpuset=[0])
> PMD: rte_igbvf_pmd_init():  >>
> EAL: lcore 1 is ready (tid=305ff700;cpuset=[1])
> EAL: PCI device 0000:00:03.0 on NUMA socket -1
> EAL:   probe driver: 1af4:1000 rte_virtio_pmd
> EAL:   Not managed by a supported kernel driver(0), skipped
> PMD: virtio_read_caps(): failed to map pci device!
> PMD: vtpci_init(): trying with legacy virtio pci.
> Segmentation fault
> 
> $ lspci
> ...
> 00:03.0 Ethernet controller: Red Hat, Inc Virtio network device
> 
> I've noticed that pci_scan_one() function sets dev->kdrv =
> RTE_KDRV_NONE in lib/librte_eal/linuxapp/eal/eal_pci.c, while the
> driver is detected as virtio-pci
> (from /sys/bus/pci/devices/0000:00:03.0/driver).
> 
> Is it even possible to run KNI with virtio-net-pci device?
> 
> If it's impossible, then are there other options?
> 


-- 
Ruslan Osmanov

Re: [dpdk-users] Using KNI with virtio-net-pci

[Ferruh Yigit]

On 6/2/2016 4:43 AM, Ruslan Osmanov wrote:
> Hi,
> 
> I'm going to develop a DPDK application on a laptop, but the laptop's
> hardware is not supported by DPDK. Furtunately, DPDK supports
> paravirtualized devices(http://www.dpdk.org/doc/nics) including
> virtio-net.  
> 
> So I'm trying to configure a QEMU guest for running the Kernel NIC
> Interface(KNI) on a virtio-net-pci device. The problem is that the KNI
> sample application doesn't accept the virtio-pci driver.
> 
> 
> QEMU command:
> 
> eth_device=virtio-net-pci
> exec qemu-system-x86_64 -enable-kvm \
>   -cpu host -smp 2 \
>   -vga std \
>   -mem-prealloc -mem-path /dev/hugepages \
>   -drive file=GentooVM.img,if=virtio \
>   -netdev user,id=vmnic,hostname=gentoo \
>   -device $eth_device,netdev=vmnic \
>   -m 1024M \
>   -monitor stdio \
>   -name "Gentoo VM"
> 
> Running the KNI sample application in the guest:
> 
> sudo ./examples/kni/build/app/kni -c 0x3 -n 4 -- \
> -P -p 0x1 --config="(0,0,1)"
> 
> EAL: Detected lcore 0 as core 0 on socket 0
> EAL: Detected lcore 1 as core 0 on socket 0
> EAL: Support maximum 128 logical core(s) by configuration.
> EAL: Detected 2 lcore(s)
> EAL: Probing VFIO support...
> EAL:   IOMMU type 1 (Type 1) is supported
> EAL:   IOMMU type 8 (No-IOMMU) is not supported
> EAL: VFIO support initialized
> EAL: Setting up physically contiguous memory...
> ...
> EAL: WARNING: cpu flags constant_tsc=yes nonstop_tsc=no -> using
> unreliable clock cycles !
> EAL: Master lcore 0 is ready (tid=657d58c0;cpuset=[0])
> PMD: rte_igbvf_pmd_init():  >>
> EAL: lcore 1 is ready (tid=305ff700;cpuset=[1])
> EAL: PCI device 0000:00:03.0 on NUMA socket -1
> EAL:   probe driver: 1af4:1000 rte_virtio_pmd
> EAL:   Not managed by a supported kernel driver(0), skipped
> PMD: virtio_read_caps(): failed to map pci device!
> PMD: vtpci_init(): trying with legacy virtio pci.
> Segmentation fault
> 
> $ lspci
> ...
> 00:03.0 Ethernet controller: Red Hat, Inc Virtio network device
> 
> I've noticed that pci_scan_one() function sets dev->kdrv =
> RTE_KDRV_NONE in lib/librte_eal/linuxapp/eal/eal_pci.c, while the
> driver is detected as virtio-pci
> (from /sys/bus/pci/devices/0000:00:03.0/driver).
> 
> Is it even possible to run KNI with virtio-net-pci device?
> 
> If it's impossible, then are there other options?
> 

Hi Ruslan,

I tried with same QEMU command with you and able to run KNI sample
application. I am using dpdk on latest master branch.

A few questions:

> EAL: PCI device 0000:00:03.0 on NUMA socket -1
> EAL:   probe driver: 1af4:1000 rte_virtio_pmd
> EAL:   Not managed by a supported kernel driver(0), skipped
1)
According above log, device not bind to a supported driver, this looks
like your problem. Can you please send output of:
dpdk_nic_bind.py -s

2)
Can you run testpmd? Or problem is just KNI?

3)
Are you using dpdk16.04?

Regards,
ferruh

Re: [dpdk-users] Using KNI with virtio-net-pci

[Ferruh Yigit]

On 6/28/2016 4:24 AM, Ruslan Osmanov wrote:
> Hi,
> 
> yes, the problem with running the KNI application was that I skipped
> binding NIC. Now the application is running on two NICs based on host
> TAP devices using igb_uio driver.
Good to hear that it is working now.

> 
> Actually, I've figured this out a couple of weeks ago. However, I still
> have difficulties testing the KNI app. Particularly, I can't see
> traffic going through NICs via `tcpdump -i vEth0`. The latter command
> displays only BOOT/DHCP requests from 0.0.0.0. For example, it doesn't
> show ICMP packets, when I `ping 192.168.1.10`(vEth0).
> 

I use netns if all interfaces are in same box, it should work with
updating routing tables but it is confusing and I am usually doing it
wrong J, can you please try with netns?

Regards,
ferruh