[dpdk-dev] System hangs when using dpdk in a guest machine

Nikos Anastopoulos n.anastop at gmail.com
Thu May 15 16:24:55 CEST 2014


Hi,

I am experiencing an irrecoverable system hang when trying to run the
'helloworld' dpdk example from within a CentOS 6.5 guest (4 VCPUS, 7GB
RAM), running on VirtualBox (Windows host).

Specifically, I have successfully built dpdk-1.6.0r2 for the
"x86_64-ivshmem-linuxapp-gcc" target.
I perform the following initialization actions without problem (all as root):
1. mkdir -p /dev/hugepages
2. mount -t hugetlbfs nodev /dev/hugepages
3. echo 0 > /proc/sys/kernel/randomize_va_space
4. modprobe uio
5. insmod $RTE_SDK/$RTE_TARGET/kmod/igb_uio.ko

Afterwards I launch the app as root as follows:
$ ./helloworld -c f -n 4

EAL: Detected lcore 0 as core 0 on socket 0
EAL: Detected lcore 1 as core 1 on socket 0
EAL: Detected lcore 2 as core 2 on socket 0
EAL: Detected lcore 3 as core 3 on socket 0
EAL: Skip lcore 4 (not detected)
EAL: Skip lcore 5 (not detected)
EAL: Skip lcore 6 (not detected)
EAL: Skip lcore 7 (not detected)
EAL: Skip lcore 8 (not detected)
...
EAL: Skip lcore 62 (not detected)
EAL: Skip lcore 63 (not detected)
EAL: Searching for IVSHMEM devices...
EAL: No IVSHMEM configuration found!
EAL: Setting up memory...

The app freezes at this point, and after a while the guest OS freezes
as well, forcing me to power it off from the host (Virtualbox). After
a guest reboot I cannot find anything in the kernel logs related with
the hang.

Note also that the above steps have been validated, and the app has
successfully run, on a native CentOS 6.5 machine.

Any idea of what could be the cause for this behaviour?

Thanks in advance,
Nikos


More information about the dev mailing list