DPDK Segmentation Fault

Question

DPDK Segmentation Fault

Carisbrooke opened this issue 9 years ago · comments

Hello,

Have you ever seem this before, running on FC21:

NUSECONF=nuse.conf ./nuse ping www.google.com
<5>Linux version 4.0.0+ (root@roman) (gcc version 4.9.2 20150212 (Red Hat 4.9.2-6) (GCC) ) #0 Sat May 9 12:04:41 BST 2015
<6>NET: Registered protocol family 16
<6>NET: Registered protocol family 2
<6>TCP established hash table entries: 512 (order: 0, 4096 bytes)
<6>TCP bind hash table entries: 512 (order: 0, 4096 bytes)
<6>TCP: Hash tables configured (established 512 bind 512)
<6>UDP hash table entries: 128 (order: 0, 4096 bytes)
<6>UDP-Lite hash table entries: 128 (order: 0, 4096 bytes)
<6>NET: Registered protocol family 1
<6>Netfilter messages via NETLINK v0.30.
<6>nfnl_acct: registering with nfnetlink.
<6>nf_conntrack version 0.5.0 (32 buckets, 128 max)
<6>nf_tables: (c) 2007-2009 Patrick McHardy kaber@trash.net
<6>ip_set: protocol 6
<6>ipip: IPv4 over IPv4 tunneling driver
<6>nsc: GRE over IPv4 demultiplexor driver
<6>nsc: GRE over IPv4 tunneling driver
<6>nsc: (C) 2000-2006 Netfilter Core Team
<6>Initializing XFRM netlink socket
<6>NET: Registered protocol family 10
<6>nsc: Mobile IPv6
<6>nsc: IPv6 over IPv4 tunneling driver
<6>NET: Registered protocol family 17
<6>NET: Registered protocol family 15
<6>DCCP: Activated CCID 2 (TCP-like)
<6>DCCP: Activated CCID 3 (TCP-Friendly Rate Control)
<6>nsc: Hash tables configured (established 512 bind 512)
create vif dpdk0
address = 192.168.0.10
netmask = 255.255.255.0
macaddr = 00:01:01:01:01:01
type = 2
failed to get interface status
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: Detected lcore 4 as core 4 on socket 0
EAL: Detected lcore 5 as core 5 on socket 0
EAL: Detected lcore 6 as core 0 on socket 0
EAL: Detected lcore 7 as core 1 on socket 0
EAL: Detected lcore 8 as core 2 on socket 0
EAL: Detected lcore 9 as core 3 on socket 0
EAL: Detected lcore 10 as core 4 on socket 0
EAL: Detected lcore 11 as core 5 on socket 0
EAL: Support maximum 128 logical core(s) by configuration.
EAL: Detected 12 lcore(s)
./nuse: line 9: 1574 Segmentation fault (core dumped) LD_LIBRARY_PATH=.:../../../ LD_PRELOAD=liblinux.so:libnuse-linux.so $*

./nuse: line 9: 1574 Segmentation fault (core dumped) LD_LIBRARY_PATH=.:../../../ LD_PRELOAD=liblinux.so:libnuse-linux.so $*
[root@roman tools]# gdb ping core.1574
GNU gdb (GDB) Fedora 7.8.2-38.fc21
Copyright (C) 2014 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later http://gnu.org/licenses/gpl.html
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law. Type "show copying"
and "show warranty" for details.
This GDB was configured as "x86_64-redhat-linux-gnu".
Type "show configuration" for configuration details.
For bug reporting instructions, please see:
http://www.gnu.org/software/gdb/bugs/.
Find the GDB manual and other documentation resources online at:
http://www.gnu.org/software/gdb/documentation/.
For help, type "help".
Type "apropos word" to search for commands related to "word"...
Reading symbols from ping...Reading symbols from /root/net-next-nuse/arch/lib/tools/ping...(no debugging symbols found)...done.
(no debugging symbols found)...done.

warning: core file may not match specified executable file.
[New LWP 1574]
[New LWP 1575]
[New LWP 1576]
[New LWP 1577]
[New LWP 1578]
[New LWP 1579]
[New LWP 1580]
[Thread debugging using libthread_db enabled]
Using host libthread_db library "/lib64/libthread_db.so.1".
Core was generated by `ping www.google.com'.
Program terminated with signal SIGSEGV, Segmentation fault.
#0 0x00007f68902052a5 in nuse_bind (fd=7, name=0x7ffcf4f69fc0, namelen=110) at nuse-glue.c:198

198 struct SimSocket *kernel_socket = nuse_fd_table[fd].nuse_sock->kern_sock;
Missing separate debuginfos, use: debuginfo-install iputils-20140519-4.fc21.x86_64
(gdb) back
#0 0x00007f68902052a5 in nuse_bind (fd=7, name=0x7ffcf4f69fc0, namelen=110) at nuse-glue.c:198
#1 0x00007f68901f9026 in vfio_mp_sync_socket_setup () at /root/net-next-nuse/arch/lib/tools/dpdk/lib/librte_eal/linuxapp/eal/eal_pci_vfio_mp_sync.c:351
#2 pci_vfio_mp_sync_setup () at /root/net-next-nuse/arch/lib/tools/dpdk/lib/librte_eal/linuxapp/eal/eal_pci_vfio_mp_sync.c:379
#3 0x00007f68901f66c3 in rte_eal_pci_init () at /root/net-next-nuse/arch/lib/tools/dpdk/lib/librte_eal/linuxapp/eal/eal_pci.c:624
#4 0x00007f68901f1665 in rte_eal_init (argc=3, argv=) at /root/net-next-nuse/arch/lib/tools/dpdk/lib/librte_eal/linuxapp/eal/eal.c:755
#5 0x00007f689018a510 in dpdk_if_init (dpdk=0x7f6892a0c160) at nuse-vif-dpdk.c:180
#6 0x00007f689018a924 in nuse_vif_dpdk_create (ifname=0x7f6892a0df40 "dpdk0") at nuse-vif-dpdk.c:261
#7 0x00007f6890203650 in nuse_vif_create (type=NUSE_VIF_DPDK, ifname=0x7f6892a0df40 "dpdk0") at nuse-vif.c:69
#8 0x00007f6890208946 in nuse_netdev_create (vifcf=0x7f6892a0df40) at nuse.c:367
#9 0x00007f68902090cf in nuse_init () at nuse.c:524
#10 0x00007f6890cd4f2a in call_init.part () from /lib64/ld-linux-x86-64.so.2
#11 0x00007f6890cd503b in _dl_init_internal () from /lib64/ld-linux-x86-64.so.2
#12 0x00007f6890cc5d2a in _dl_start_user () from /lib64/ld-linux-x86-64.so.2
#13 0x0000000000000002 in ?? ()
#14 0x00007ffcf4f6c68b in ?? ()
#15 0x00007ffcf4f6c690 in ?? ()
#16 0x0000000000000000 in ?? ()

(gdb)

[root@roman tools]# dpdk/tools/dpdk_nic_bind.py --status

Network devices using DPDK-compatible driver

0000:03:00.1 'Ethernet Controller X710 for 10GbE SFP+' drv=igb_uio unused=i40e,vfio-pci

Network devices using kernel driver

0000:03:00.0 'Ethernet Controller X710 for 10GbE SFP+' if=enp3s0f0 drv=i40e unused=igb_uio,vfio-pci
0000:04:00.0 'I350 Gigabit Network Connection' if=enp4s0f0 drv=igb unused=igb_uio,vfio-pci
0000:04:00.1 'I350 Gigabit Network Connection' if=enp4s0f1 drv=igb unused=igb_uio,vfio-pci
0000:06:00.0 'SFC9020 [Solarstorm]' if=enp6s0f0 drv=sfc unused=igb_uio,vfio-pci
0000:06:00.1 'SFC9020 [Solarstorm]' if=enp6s0f1 drv=sfc unused=igb_uio,vfio-pci
0000:08:00.0 'I350 Gigabit Network Connection' if=eno1 drv=igb unused=igb_uio,vfio-pci Active
0000:08:00.1 'I350 Gigabit Network Connection' if=eno2 drv=igb unused=igb_uio,vfio-pci

Other network devices

[root@roman tools]#

[root@roman tools]# more nuse.conf

interface dpdk0
address 192.168.0.10
netmask 255.255.255.0
macaddr 00:01:01:01:01:01

if macaddr is not specified, random mac addr is used.

    viftype DPDK

route
network 0.0.0.0
netmask 0.0.0.0
gateway 192.168.0.1
[root@roman tools]#

I did have to patch the DPDK to work with 3.19 kernel.

Carisbrooke · Answer 1 · Sun May 10 2015 20:31:09 GMT+0800 (China Standard Time)

Have tried on CentOS 7.1 still failing in the same place, upgraded DPDK to the latest version. DPDK appears to be working but still coring

EAL: PCI memory mapped at 0x7f9258bec000
EAL: PCI memory mapped at 0x7f92e220d000
PMD: eth_i40e_dev_init(): FW 4.33 API 1.2 NVM 04.04.01 eetrack 80001867
PMD: eth_i40e_dev_init(): Failed to stop lldp
PMD: i40e_pf_parameter_init(): Max supported VSIs:66
PMD: i40e_pf_parameter_init(): PF queue pairs:64
PMD: i40e_pf_parameter_init(): Max VMDQ VSI num:63
PMD: i40e_pf_parameter_init(): VMDQ queue pairs:4
EAL: PCI device 0000:04:00.0 on NUMA socket 0
EAL: probe driver: 8086:1521 rte_igb_pmd
EAL: Not managed by a supported kernel driver, skipped
EAL: PCI device 0000:04:00.1 on NUMA socket 0
EAL: probe driver: 8086:1521 rte_igb_pmd
EAL: Not managed by a supported kernel driver, skipped
EAL: PCI device 0000:08:00.0 on NUMA socket 0
EAL: probe driver: 8086:1521 rte_igb_pmd
EAL: Not managed by a supported kernel driver, skipped
EAL: PCI device 0000:08:00.1 on NUMA socket 0
EAL: probe driver: 8086:1521 rte_igb_pmd
EAL: Not managed by a supported kernel driver, skipped
Interactive-mode selected
Configuring Port 0 (socket 0)
PMD: i40e_dev_tx_queue_setup(): Using simple tx path
PMD: i40e_dev_rx_queue_setup(): Rx Burst Bulk Alloc Preconditions are satisfied. Rx Burst Bulk Alloc function will be used on port=0, queue=0.
Port 0: 3C:FD:FE:00:48:80
Configuring Port 1 (socket 0)
PMD: i40e_dev_tx_queue_setup(): Using simple tx path
PMD: i40e_dev_rx_queue_setup(): Rx Burst Bulk Alloc Preconditions are satisfied. Rx Burst Bulk Alloc function will be used on port=1, queue=0.
Port 1: 3C:FD:FE:00:48:82
Checking link statuses...
Port 0 Link Down
Port 1 Link Up - speed 10000 Mbps - full-duplex
Done
testpmd>

Had to make the following chances to Makefile.dpdk

add -lrte_pmd_i40e
removed -lrte_pmd_virtio_uio

Hajime Tazaki · Answer 2 · Mon May 11 2015 03:15:36 GMT+0800 (China Standard Time)

thank you for your report.

I confirmed the issue. since linux-libos-tools was forked from net-next-nuse, DPDK (and netmap) are not well tested. I will work on this issue but it would be also great if you come up with the patches (and pull request in the end).

the issue in nuse_bind() would be that NUSE should handle the socket opened by dpdk library as a host fd, but tried to look up the internal fd table which is used by NUSE application (i.e., ping).

Carisbrooke · Answer 3 · Tue May 12 2015 21:50:05 GMT+0800 (China Standard Time)

Thanks, I'll try and have a look. Will not have time for a couple of weeks however.

Vincent Li · Answer 4 · Wed Jun 24 2015 03:13:51 GMT+0800 (China Standard Time)

Hi Hajime
I am getting similar coredump while testing NUSE and DPDK, what is the proper fix in regarding to "the issue in nuse_bind() would be that NUSE should handle the socket opened by dpdk library as a host fd, but tried to look up the internal fd table which is used by NUSE application (i.e., ping)."

int nuse_bind(int fd, const struct sockaddr *name, socklen_t namelen)
{
struct SimSocket *kernel_socket = nuse_fd_table[fd].nuse_sock->kern_sock;
int ret;

    ret = g_exported->sock_bind(kernel_socket, name, namelen);
    return ret;

}

the fd in nuse_bind is passed by vfio_mp_sync_socket_setup, nuse_bind should not look fd in nuse_fd_table[]? if not, what do you mean handle fd as host fd?

Hajime Tazaki · Answer 5 · Wed Jun 24 2015 07:21:36 GMT+0800 (China Standard Time)

I meant, if nuse_bind() can't find a fd in nuse_fd_table, then it means it wasn't opened by nuse_socket. in that case, host system call should be used.

write(2) system call (nuse_write()) is already implemented like that.

https://github.com/libos-nuse/linux-libos-tools/blob/master/nuse-glue.c#L279

though the issue of DPDK is not only the part of nuse_bind(), the above should be a proper way to fix it.

Vincent Li · Answer 6 · Thu Jun 25 2015 06:00:02 GMT+0800 (China Standard Time)

you mean like below?

int nuse_bind(int fd, const struct sockaddr *name, socklen_t namelen)
{
if (!nuse_fd_table[fd])
return host_bind(fd, name, namelen);

    struct SimSocket *kernel_socket = nuse_fd_table[fd].nuse_sock->kern_sock;
    int ret;

    ret = g_exported->sock_bind(kernel_socket, name, namelen);
    return ret;

}

Hajime Tazaki · Answer 7 · Thu Jun 25 2015 12:05:42 GMT+0800 (China Standard Time)

@vincentmli yes, something like that.

Hajime Tazaki · Answer 8 · Thu Jun 25 2015 12:18:05 GMT+0800 (China Standard Time)

to @vincentmli @Carisbrooke,

I have struggled a couple of combination between dpdk verison, host kernel version, on VM or bare-metal. and decided tentatively to revert dpdk version to the older one, 1.7.1, which we have been tested in the past with the older host kernel version (not LibOS kernel, around 3.13).

My Fedora21 on vmware has newer kernel (4.0.4), thus not worked and tested on VM yet.

the following outputs is the result with the commit 86eb208, which should work with the specific environment.

uname -a

Linux XXX 3.13.0-32-generic #57-Ubuntu SMP Tue Jul 15 03:51:08 UTC 2014 x86_64 x86_64 x86_64 GNU/Linux

and NIC.

$ ./dpdk/tools/dpdk_nic_bind.py --status

Network devices using DPDK-compatible driver
============================================
0000:01:00.0 'Ethernet 10G 2P X520 Adapter' drv=igb_uio unused=

Network devices using kernel driver
===================================
0000:00:19.0 '82578DM Gigabit Network Connection' if=eth0 drv=e1000e unused=igb_uio *Active*
0000:06:00.0 '82574L Gigabit Network Connection' if=p7p1 drv=e1000e unused=igb_uio

to build dpdk-enable NUSE

make library ARCH=lib
./dpdk-sdk-build.sh
make library ARCH=lib DPDK=yes

with a nuse.conf

interface dpdk0
        address 172.16.0.33
        netmask 255.255.255.0
        viftype DPDK

then execute it.

sudo NUSECONF=/tmp/nuse.conf ./nuse ping 172.16.0.2
<5>Linux version 4.1.0-rc7+ (tazaki@okzw3) (gcc version 4.8.2 (Ubuntu 4.8.2-19ubuntu1) ) #0 Fri Jun 19 17:54:12 JST 2015
<6>NET: Registered protocol family 16
<6>NET: Registered protocol family 2
<6>TCP established hash table entries: 512 (order: 0, 4096 bytes)
<6>TCP bind hash table entries: 512 (order: 0, 4096 bytes)
<6>TCP: Hash tables configured (established 512 bind 512)
<6>UDP hash table entries: 128 (order: 0, 4096 bytes)
<6>UDP-Lite hash table entries: 128 (order: 0, 4096 bytes)
<6>NET: Registered protocol family 1
<6>Netfilter messages via NETLINK v0.30.
<6>nfnl_acct: registering with nfnetlink.
<6>nf_conntrack version 0.5.0 (32 buckets, 128 max)
<6>nf_tables: (c) 2007-2009 Patrick McHardy <kaber@trash.net>
<6>ip_set: protocol 6
<6>ipip: IPv4 over IPv4 tunneling driver
<6>nsc: GRE over IPv4 demultiplexor driver
<6>nsc: GRE over IPv4 tunneling driver
<6>nsc: (C) 2000-2006 Netfilter Core Team
<6>Initializing XFRM netlink socket
<6>NET: Registered protocol family 10
<6>nsc: Mobile IPv6
<6>nsc: IPv6 over IPv4 tunneling driver
<6>NET: Registered protocol family 17
<6>NET: Registered protocol family 15
<6>nsc: L2TP core driver, V2.0
<6>nsc: L2TP IP encapsulation support (L2TPv3)
<6>nsc: L2TP netlink interface
<6>nsc: L2TP ethernet pseudowire support (L2TPv3)
<6>nsc: L2TP IP encapsulation support for IPv6 (L2TPv3)
<6>DCCP: Activated CCID 2 (TCP-like)
<6>DCCP: Activated CCID 3 (TCP-Friendly Rate Control)
<6>nsc: Hash tables configured (established 512 bind 512)
create vif dpdk0
  address = 172.16.0.33
  netmask = 255.255.255.0
  macaddr = 00:00:00:00:00:00
  type    = 2
failed to get interface status
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: Detected lcore 4 as core 0 on socket 0
EAL: Detected lcore 5 as core 1 on socket 0
EAL: Detected lcore 6 as core 2 on socket 0
EAL: Detected lcore 7 as core 3 on socket 0
EAL: Support maximum 64 logical core(s) by configuration.
EAL: Detected 8 lcore(s)
open: No such file or directory
EAL:   cannot open VFIO container, error 2 (No such file or directory)
EAL: VFIO support could not be initialized
EAL: Setting up memory...
EAL: Ask a virtual area of 0x400000 bytes
EAL: Virtual area found at 0x7fcb42c00000 (size = 0x400000)
EAL: Ask a virtual area of 0x2800000 bytes
EAL: Virtual area found at 0x7fcb40200000 (size = 0x2800000)
EAL: Ask a virtual area of 0x1800000 bytes
EAL: Virtual area found at 0x7fcb3a600000 (size = 0x1800000)
EAL: Ask a virtual area of 0x2000000 bytes
EAL: Virtual area found at 0x7fcb38400000 (size = 0x2000000)
EAL: Ask a virtual area of 0x400000 bytes
EAL: Virtual area found at 0x7fcb37e00000 (size = 0x400000)
EAL: Ask a virtual area of 0xc00000 bytes
(snip)
EAL: Virtual area found at 0x7fcb0de00000 (size = 0x200000)
EAL: Requesting 326 pages of size 2MB from socket 0
EAL: TSC frequency is ~1861995 KHz
EAL: Master core 0 is ready (tid=47ae9840)
EAL: PCI device 0000:01:00.0 on NUMA socket -1
EAL:   probe driver: 8086:154d rte_ixgbe_pmd
EAL:   PCI memory mapped at 0x7fcb47a31000
EAL:   PCI memory mapped at 0x7fcb47a2d000
epoll woke up for read with 7
EAL: PCI device 0000:01:00.1 on NUMA socket -1
EAL:   probe driver: 8086:154d rte_ixgbe_pmd
EAL:   0000:01:00.1 not managed by UIO driver, skipping
EAL: PCI device 0000:06:00.0 on NUMA socket -1
EAL:   probe driver: 8086:10d3 rte_em_pmd
EAL:   0000:06:00.0 not managed by UIO driver, skipping
EAL: PCI device 0000:01:00.1 on NUMA socket -1
EAL:   probe driver: 8086:154d rte_ixgbe_pmd
EAL:   0000:01:00.1 not managed by UIO driver, skipping
EAL: PCI device 0000:06:00.0 on NUMA socket -1
EAL:   probe driver: 8086:10d3 rte_em_pmd
EAL:   0000:06:00.0 not managed by UIO driver, skipping
  mac address for dpdk0 is randomized 02:00:d0:7b:b8:07
err ip_rt_ioctl to add route to 0.0.0.0 via 192.168.0.1 -101
nuse syscall proxy start at unix:///tmp/rump-server-nuse.27483
PING 172.16.0.2 (172.16.0.2) 56(84) bytes of data.
Warning: time of day goes back (-1424597300661860us), taking countermeasures.
64 bytes from 172.16.0.2: icmp_seq=1 ttl=64 time=1011 ms
64 bytes from 172.16.0.2: icmp_seq=2 ttl=64 time=0.174 ms
^C
--- 172.16.0.2 ping statistics ---
2 packets transmitted, 2 received, 0% packet loss, time 1007ms
rtt min/avg/max/mdev = 0.174/0.303/0.432/0.129 ms
finishing NUSE
rump_server finishing.

voila :)

I think I will close this issue if some of you guys successfully follow this instruction.
for more improvement such as newer DPDK support, finalized hugepage during exit, etc,
I would have another github issue.

Let me know how you experienced with the information.

Vincent Li · Answer 9 · Fri Jun 26 2015 02:14:20 GMT+0800 (China Standard Time)

thank you for your effort, I am sorry to bother you. I couldn't use bare metal machine like yours but VM environment on KVM.

for clarification, with DPDK 1.7.1, do I still need to modify nuse_bind to handle the fd as host fd ? I tried 1.7.1 and still got the same coredump backtrace, then modified nuse_bind to handle fd as host fd, got coredump again with nuse_listen, modified nuse_listen to handle fd as host fd, got coredump with nuse_accept, modified nuse_accept. no more coredump, but got:

<5>Linux version 4.1.0-rc7+ (root@ubuntu14) (gcc version 4.8.2 (Ubuntu 4.8.2-19ubuntu1) ) #0 Thu Jun 25 07:09:40 PDT 2015
<6>NET: Registered protocol family 16
<6>NET: Registered protocol family 2
<6>TCP established hash table entries: 512 (order: 0, 4096 bytes)
<6>TCP bind hash table entries: 512 (order: 0, 4096 bytes)
<6>TCP: Hash tables configured (established 512 bind 512)
<6>UDP hash table entries: 128 (order: 0, 4096 bytes)
<6>UDP-Lite hash table entries: 128 (order: 0, 4096 bytes)
<6>NET: Registered protocol family 1
<6>Netfilter messages via NETLINK v0.30.
<6>nfnl_acct: registering with nfnetlink.
<6>nf_conntrack version 0.5.0 (32 buckets, 128 max)
<6>nf_tables: (c) 2007-2009 Patrick McHardy kaber@trash.net
<6>ip_set: protocol 6
<6>ipip: IPv4 over IPv4 tunneling driver
<6>nsc: GRE over IPv4 demultiplexor driver
<6>nsc: GRE over IPv4 tunneling driver
<6>nsc: (C) 2000-2006 Netfilter Core Team
<6>Initializing XFRM netlink socket
<6>NET: Registered protocol family 10
<6>nsc: Mobile IPv6
<6>nsc: IPv6 over IPv4 tunneling driver
<6>NET: Registered protocol family 17
<6>NET: Registered protocol family 15
<6>nsc: L2TP core driver, V2.0
<6>nsc: L2TP IP encapsulation support (L2TPv3)
<6>nsc: L2TP netlink interface
<6>nsc: L2TP ethernet pseudowire support (L2TPv3)
<6>nsc: L2TP IP encapsulation support for IPv6 (L2TPv3)
<6>DCCP: Activated CCID 2 (TCP-like)
<6>DCCP: Activated CCID 3 (TCP-Friendly Rate Control)
<6>nsc: Hash tables configured (established 512 bind 512)
create vif dpdk0
address = 10.9.1.8
netmask = 255.255.255.0
macaddr = 00:01:01:01:01:01
type = 2
failed to get interface status

EAL: Detected 4 lcore(s)
EAL: Setting up memory...
EAL: Ask a virtual area of 0x200000 bytes
EAL: Virtual area found at 0x7f4077200000 (size = 0x200000)
EAL: Ask a virtual area of 0x200000 bytes
EAL: Virtual area found at 0x7f4076e00000 (size = 0x200000)
EAL: Ask a virtual area of 0x6c00000 bytes
EAL: Virtual area found at 0x7f4070000000 (size = 0x6c00000)
EAL: Ask a virtual area of 0x200000 bytes
EAL: Virtual area found at 0x7f406fc00000 (size = 0x200000)
EAL: Ask a virtual area of 0x200000 bytes
EAL: Virtual area found at 0x7f406f800000 (size = 0x200000)
EAL: Ask a virtual area of 0x200000 bytes
EAL: Virtual area found at 0x7f406f400000 (size = 0x200000)
EAL: Ask a virtual area of 0x200000 bytes
EAL: Virtual area found at 0x7f406f000000 (size = 0x200000)
EAL: Ask a virtual area of 0x800000 bytes
EAL: Virtual area found at 0x7f406e600000 (size = 0x800000)
EAL: Requesting 64 pages of size 2MB from socket 0
EAL: TSC frequency is ~3392295 KHz
EAL: WARNING: cpu flags constant_tsc=yes nonstop_tsc=no -> using unreliable clock cycles !
EAL: Master core 0 is ready (tid=7d083840)
EAL: PCI device 0000:00:03.0 on NUMA socket -1
EAL: probe driver: 1af4:1000 rte_virtio_pmd
EAL: 0000:00:03.0 not managed by VFIO driver, skipping
EAL: 0000:00:03.0 not managed by UIO driver, skipping
EAL: PCI device 0000:00:07.0 on NUMA socket -1
EAL: probe driver: 1af4:1000 rte_virtio_pmd
EAL: 0000:00:07.0 not managed by VFIO driver, skipping
EAL: PCI memory mapped at 0x7f407d088000
EAL: PCI device 0000:00:03.0 on NUMA socket -1
EAL: probe driver: 1af4:1000 rte_virtio_pmd
EAL: 0000:00:03.0 not managed by VFIO driver, skipping
EAL: 0000:00:03.0 not managed by UIO driver, skipping
dpdk_if_init:229: failed to setup tx queue <============
nuse_vif_dpdk_create:263: failed to init dpdk interface <=============
nuse_vif_dpdk_create:263: failed to init dpdk interface

here is my nic status:
root@ubuntu14:/home/vincent/net-next-nuse/arch/lib/tools# ./dpdk/tools/dpdk_nic_bind.py --status

Network devices using DPDK-compatible driver

0000:00:07.0 'Virtio network device' drv=igb_uio unused=

Network devices using kernel driver

0000:00:03.0 'Virtio network device' if= drv=virtio-pci unused=igb_uio

Other network devices

root@ubuntu14:/home/vincent/net-next-nuse/arch/lib/tools# lspci
00:00.0 Host bridge: Intel Corporation 440FX - 82441FX PMC [Natoma](rev 02)
00:01.0 ISA bridge: Intel Corporation 82371SB PIIX3 ISA [Natoma/Triton II]
00:01.1 IDE interface: Intel Corporation 82371SB PIIX3 IDE [Natoma/Triton II]
00:01.2 USB controller: Intel Corporation 82371SB PIIX3 USB [Natoma/Triton II](rev 01)
00:01.3 Bridge: Intel Corporation 82371AB/EB/MB PIIX4 ACPI (rev 03)
00:02.0 VGA compatible controller: Cirrus Logic GD 5446
00:03.0 Ethernet controller: Red Hat, Inc Virtio network device
00:04.0 Audio device: Intel Corporation 82801FB/FBM/FR/FW/FRW (ICH6 Family) High Definition Audio Controller (rev 01)
00:05.0 SCSI storage controller: Red Hat, Inc Virtio block device
00:06.0 RAM memory: Red Hat, Inc Virtio memory balloon
00:07.0 Ethernet controller: Red Hat, Inc Virtio network device

Hajime Tazaki · Answer 10 · Fri Jun 26 2015 22:50:11 GMT+0800 (China Standard Time)

@vincentmli in my bare-metal, I don't need to modify nuse_bind() etc and it's working.

Vincent Li · Answer 11 · Wed Jul 01 2015 04:26:01 GMT+0800 (China Standard Time)

@thehajime I can confirm it works on my bare-meal dell poweredge R220 with Intel 82571EB running on ubutnu 14.04.1 with kernel

Linux 3.13.0-32-generic #57-Ubuntu SMP Tue Jul 15 03:51:08 UTC 2014 x86_64 x86_64 x86_64 GNU/Linux

ubuntu 14.04.2 with kernel Linux ubuntu14 3.16.0-30-generic #40~14.04.1-Ubuntu SMP Thu Jan 15 17:43:14 UTC 2015 x86_64 x86_64 x86_64 GNU/Linux does not work

it is odd there would be such difference between ubuntu kernel 3.13.0-32 and 3.16.0-30, is this kernel or rump or dpdk issue?