[dpdk-dev] segfault in vt_pci_init

[dpdk-dev] segfault in vt_pci_init

[Thomas F Herbert]

All,

I am trying to debug a problem:

I am running testpmd in a VM.

The stack trace is below.

Running uio_pci_generic. kvm-qemu 2.5.3, Centos 7
CentOS Linux release 7.2.1511 (Core)


Problem can be reproduced using either igb_uio or virtio_pci kernel drivers.

Currently running 3.18.35 kernel but problem can be reproduced with 3.10 
kernel shipped with Centos.

Problem originally was scene with 16.04. Now reproduced from top of 
master commit, 34d279906d588e349f3e69020a94d7f932bdf099

Currently running qemu/kvm 2.3: qemu-kvm-ev-2.3.0-31.el7_2.10.1.x86_64

Thanks for helping with more eyes on this problem. I am at OPNFV and 
this problem is currently holding up nfv deployment.

(gdb) run -c 0x1 -n 4 -- -i -- portmask 0x1 --ncores=1
Starting program: /home/heat-admin/dpdk/mybuild/app/testpmd -c 0x1 -n 4 
-- -i -- portmask 0x1 --ncores=1
[Thread debugging using libthread_db enabled]
Using host libthread_db library "/lib64/libthread_db.so.1".
EAL: Detected 4 lcore(s)
EAL: Probing VFIO support...
EAL: VFIO support initialized
[New Thread 0x7ffff73f0700 (LWP 30542)]
EAL: WARNING: cpu flags constant_tsc=yes nonstop_tsc=no -> using 
unreliable clock cycles !
[New Thread 0x7ffff6aec700 (LWP 30543)]
EAL: PCI device 0000:00:03.0 on NUMA socket -1
EAL:   probe driver: 1af4:1000 rte_virtio_pmd

Program received signal SIGSEGV, Segmentation fault.
0x0000000000616172 in vtpci_init (dev=0xbfb0f0, hw=0x7fffd57d0700,
     dev_flags=0x7fffffffdf9c)
     at /home/heat-admin/dpdk/drivers/net/virtio/virtio_pci.c:722
722                dev->devargs->type != RTE_DEVTYPE_WHITELISTED_PCI) {
Missing separate debuginfos, use: debuginfo-install 
glibc-2.17-106.el7_2.6.x86_64
(gdb) bt
#0  0x0000000000616172 in vtpci_init (dev=0xbfb0f0, hw=0x7fffd57d0700,
     dev_flags=0x7fffffffdf9c)
     at /home/heat-admin/dpdk/drivers/net/virtio/virtio_pci.c:722
#1  0x00000000006209fb in eth_virtio_dev_init (
     eth_dev=0xb659a0 <rte_eth_devices>)
     at /home/heat-admin/dpdk/drivers/net/virtio/virtio_ethdev.c:1108
#2  0x000000000048339c in rte_eth_dev_init (pci_drv=0x8d52a0 
<rte_virtio_pmd>,
     pci_dev=0xbfb0f0)
     at /home/heat-admin/dpdk/lib/librte_ether/rte_ethdev.c:288
#3  0x00000000004a0b7d in rte_eal_pci_probe_one_driver (
     dr=0x8d52a0 <rte_virtio_pmd>, dev=0xbfb0f0)
     at /home/heat-admin/dpdk/lib/librte_eal/common/eal_common_pci.c:214
#4  0x00000000004a0d3d in pci_probe_all_drivers (dev=0xbfb0f0)
     at /home/heat-admin/dpdk/lib/librte_eal/common/eal_common_pci.c:290
#5  0x00000000004a108a in rte_eal_pci_probe ()
     at /home/heat-admin/dpdk/lib/librte_eal/common/eal_common_pci.c:417
#6  0x0000000000492fd2 in rte_eal_init (argc=11, argv=0x7fffffffe368)
     at /home/heat-admin/dpdk/lib/librte_eal/linuxapp/eal/eal.c:874
#7  0x000000000043592d in main (argc=11, argv=0x7fffffffe368)
     at /home/heat-admin/dpdk/app/test-pmd/testpmd.c:2059
(gdb)

-- 
*Thomas F Herbert*
SDN Group
Office of Technology
*Red Hat*

Re: [dpdk-dev] segfault in vt_pci_init

[Thomas Monjalon]

2016-06-22 14:27, Thomas F Herbert:
> Program received signal SIGSEGV, Segmentation fault.
> 0x0000000000616172 in vtpci_init (dev=0xbfb0f0, hw=0x7fffd57d0700,
>      dev_flags=0x7fffffffdf9c)
>      at /home/heat-admin/dpdk/drivers/net/virtio/virtio_pci.c:722
> 722                dev->devargs->type != RTE_DEVTYPE_WHITELISTED_PCI) {

Please could you check with the current HEAD?
Maybe this commit can fix your issue:
	http://dpdk.org/commit/7e40200c56

Re: [dpdk-dev] segfault in vt_pci_init

[Thomas F Herbert]

Thomas,

Thanks. That fixes it of course.

--TFH

On 6/22/16 3:39 PM, Thomas Monjalon wrote:
> 2016-06-22 14:27, Thomas F Herbert:
>> Program received signal SIGSEGV, Segmentation fault.
>> 0x0000000000616172 in vtpci_init (dev=0xbfb0f0, hw=0x7fffd57d0700,
>>       dev_flags=0x7fffffffdf9c)
>>       at /home/heat-admin/dpdk/drivers/net/virtio/virtio_pci.c:722
>> 722                dev->devargs->type != RTE_DEVTYPE_WHITELISTED_PCI) {
> Please could you check with the current HEAD?
> Maybe this commit can fix your issue:
> 	http://dpdk.org/commit/7e40200c56

-- 
*Thomas F Herbert*
SDN Group
Office of Technology
*Red Hat*