[dpdk-users] TestPMD fails with DPDK 16.07 + VMware Direct I/o Passthrough of X520-SR2 NICs
Hi, I have 2 10G X520-SR2 Ethernet adapters that i want to use in a VM with DPDK. So, I configured using VMware vSphere these two PCI devices to be in Passthrough (Direct I/o not SR-IOV) mode and attached these 2 I/fs to the VM instance. On the VM instance, the 2 PCI devices do show up ok (lspci and ethtool showing driver info below). The X520-SR2 adapters came up with ixgbe driver on the guest kernel. I am able to attach the I/f to DPDK using dpdk-devbind and to use igb_uio. However testpmd application fails to recognize the interface. Any ideas why? I saw a couple of links showing that there could be timing issues with ixgbe in earlier releases of dpdk, that could lead to such an issue, Is this the case even now? http://dpdk.org/ml/archives/dev/2014-January/001205.html http://www.intel.com/content/dam/www/public/us/en/documents/release-notes/dpdk-release-notes.pdf - Section 7.11 Any pointers would be really helpful. Is DPDK expected to work on VM without SR_IOV as well? Some info below: # dpdk-devbind -b igb_uio :1b:00.1 # dpdk-devbind --status Network devices using DPDK-compatible driver :1b:00.1 'Ethernet Server Bypass Adapter X520-SR2' drv=igb_uio unused=ixgbe Network devices using kernel driver === :02:01.0 '82545EM Gigabit Ethernet Controller (Copper)' if=eth1 drv=e1000 unused=igb_uio :0b:00.0 'VMXNET3 Ethernet Controller' if=eth0 drv=vmxnet3 unused=igb_uio *Active* :13:00.0 'VMXNET3 Ethernet Controller' if=eth2 drv=vmxnet3 unused=igb_uio :1b:00.0 'Ethernet Server Bypass Adapter X520-SR2' if=eth3 drv=ixgbe unused=igb_uio Other network devices = # testpmd -w :1b:00.1 EAL: Detected 14 lcore(s) PMD: bnxt_rte_pmd_init() called for (null) EAL: No probed ethernet devices # lspci | grep -i ethernet 02:01.0 Ethernet controller: Intel Corporation 82545EM Gigabit Ethernet Controller (Copper) (rev 01) 0b:00.0 Ethernet controller: VMware VMXNET3 Ethernet Controller (rev 01) 13:00.0 Ethernet controller: VMware VMXNET3 Ethernet Controller (rev 01) 1b:00.0 Ethernet controller: Intel Corporation Ethernet Server Bypass Adapter X520-SR2 (rev 01) 1b:00.1 Ethernet controller: Intel Corporation Ethernet Server Bypass Adapter X520-SR2 (rev 01) [root@ # ethtool -i eth4 driver: ixgbe version: 3.13.0.14-bypass-BYPASS-PFNIC firmware-version: 0x853e bus-info: :1b:00.1 supports-statistics: yes supports-test: yes supports-eeprom-access: yes supports-register-dump: yes supports-priv-flags: no
[dpdk-users] Dell PowerEdge CPU IERR caused by DPDK 16.07
On Mon, Nov 21, 2016 at 3:45 PM, Norbert Nyak? wrote: > As soon as I ran the testpipeline sample application on one of the > servers it stops responding and on the LCD info screen and in the > logs CPU IERR error messages appear > ?difficult to *conclusively* conclude ?that this might be _caused_ by dpdk. what is the error that you see (not just the fact that you saw a CPU-IERR) ? for some more information have a look here: http://www.dell.com/support/article/in/en/indhs1/SLN292634. maybe reseating memory modules (or some such) on the machine where you see this might fix the issue ? one laborious but possible approach to conclude that this is caused (rather than just _triggered_) by dpdk would be to invert machine configuration and then retest ? ?-- thanks anupam ? In the beginning was the lambda, and the lambda was with Emacs, and Emacs was the lambda.
[dpdk-users] Error "EAL: PCI device 0000:04:00.0 on NUMA socket -1 EAL: probe driver: 8086:1521 rte_igb_pmd" with uio_pci_generic driver in DPDK in box sponsered by intel
Hi, We are using DPDK in box (Minnowboard Turbot) sponsered by intel. We installed dpdk as per the installation guide and did modprobe uio_pci_generic. We have used uio_pci_generic and successfully binded the NICs as below. dpdk at dpdk-Minnowboard-Turbot-D0-PLATFORM:/usr/src/dpdk-16.07/tools$ ./dpdk-devbind.py -s Network devices using DPDK-compatible driver :04:00.0 'I350 Gigabit Network Connection' drv=uio_pci_generic unused=igb :04:00.1 'I350 Gigabit Network Connection' drv=uio_pci_generic unused=igb Network devices using kernel driver === :03:00.0 'RTL8111/8168/8411 PCI Express Gigabit Ethernet Controller' if=enp3s0 drv=r8169 unused=uio_pci_generic *Active* *Huge Page Configuration:* -bash-4.3$ cat /proc/meminfo | grep Huge AnonHugePages:239616 kB HugePages_Total: 64 HugePages_Free:0 HugePages_Rsvd:0 HugePages_Surp:0 Hugepagesize: 2048 kB But when we run ?./testpmd? to do the basic testing, We are facing the segmentation fault. dpdk at dpdk-Minnowboard-Turbot-D0-PLATFORM:/usr/src/dpdk-16.07/x86_64-native-linuxapp-gcc/app/test Launching app EAL: Detected 2 lcore(s) EAL: Probing VFIO support... PMD: bnxt_rte_pmd_init() called for (null) EAL: PCI device :04:00.0 on NUMA socket -1 EAL: probe driver: 8086:1521 rte_igb_pmd Segmentation fault (core dumped) *GDB Debugging:* Thread 1 "test" received signal SIGSEGV, Segmentation fault. 0x005fa9b8 in eth_igb_dev_init () (gdb) bt #0 0x005fa9b8 in eth_igb_dev_init () #1 0x005844f2 in rte_eth_dev_init () #2 0x0059b958 in pci_probe_all_drivers () #3 0x0059be24 in rte_eal_pci_probe () #4 0x00590317 in rte_eal_init () #5 0x0042dd35 in main () (gdb) list 1 ../sysdeps/x86_64/dl-procinfo.c: No such file or directory. (gdb) Please help us to resolve this issue.
[dpdk-users] Dell PowerEdge CPU IERR caused by DPDK 16.07
Hi, Thank you for your help! There was also a ?bus fatal error" in the iDRAC / OMSA log and I also looked at /var/log/* but there was no any suspicious log entry, I think it because the machine rebooted as soon as the error triggered. In my opinion it is not hardware fault because we have two machines with the same configs and both behaves the same, but we tried with the same network card with a different machine and there was no error. Also we will try it with a different network card soon. Since my first mail we updated the BIOS of the machines but it did not help. Maybe it can be some firmware bug in the server which incorrectly identifies CPU fault. Best regards, Norbert 2016-11-22 7:48 GMT+01:00 Anupam Kapoor : > > On Mon, Nov 21, 2016 at 3:45 PM, Norbert Nyak? > wrote: >> >> As soon as I ran the testpipeline sample application on one of the >> servers it stops responding and on the LCD info screen and in the >> logs CPU IERR error messages appear > > > difficult to *conclusively* conclude that this might be _caused_ by dpdk. > > what is the error that you see (not just the fact that you saw a CPU-IERR) ? > for some more information have a look here: > http://www.dell.com/support/article/in/en/indhs1/SLN292634. maybe reseating > memory modules (or some such) on the machine where you see this might fix > the issue ? > > one laborious but possible approach to conclude that this is caused (rather > than just _triggered_) by dpdk would be to invert machine configuration and > then retest ? > > -- > thanks > anupam > > In the beginning was the lambda, and the lambda was with Emacs, and Emacs > was the lambda.