[dpdk-users] pdump can only run once

[Violet Wang <violet127001@gmail.com>]

Hi all,

I need your help to solve an issue when using dpdk-pdump.



I wrote a pktgen-like program and started it first, and it can run well.

# ./my_program -l 0-2



When I started dpdk-pdump to capture packets for the first time, I can
capture packets, but it didn't show the captured packet counter (PDUMP
DEBUG STATS) when I stopped it.

# ./dpdk-pdump -l 0-2 -- --pdump 'port=0,queue=*,tx-dev=/tmp/test.pcap'

EAL: Detected 8 lcore(s)

EAL: Probing VFIO support...

EAL: WARNING: Address Space Layout Randomization (ASLR) is enabled in the
kernel.

EAL:    This may cause issues with mapping memory into secondary processes

EAL: PCI device 0000:03:00.0 on NUMA socket -1

EAL:   Invalid NUMA socket, default to 0

EAL:   probe driver: 15ad:7b0 net_vmxnet3

EAL: PCI device 0000:0b:00.0 on NUMA socket -1

EAL:   Invalid NUMA socket, default to 0

EAL:   probe driver: 15ad:7b0 net_vmxnet3

EAL: PCI device 0000:13:00.0 on NUMA socket -1

EAL:   Invalid NUMA socket, default to 0

EAL:   probe driver: 15ad:7b0 net_vmxnet3

PMD: Initializing pmd_pcap for net_pcap_tx_0

PMD: Creating pcap-backed ethdev on numa socket 4294967295

Port 1 MAC: 00 00 00 01 02 03



However, when I tried to start dpdk-pdump again, it was terminated. Here's
what I got:

# ./dpdk-pdump -l 0-2 -- --pdump 'port=0,queue=*,tx-dev=/tmp/test.pcap'

EAL: Detected 8 lcore(s)

EAL: Probing VFIO support...

EAL: PCI device 0000:03:00.0 on NUMA socket -1

EAL:   Invalid NUMA socket, default to 0

EAL:   probe driver: 15ad:7b0 net_vmxnet3

EAL: PCI device 0000:0b:00.0 on NUMA socket -1

EAL:   Invalid NUMA socket, default to 0

EAL:   probe driver: 15ad:7b0 net_vmxnet3

EAL: PCI device 0000:13:00.0 on NUMA socket -1

EAL:   Invalid NUMA socket, default to 0

EAL:   probe driver: 15ad:7b0 net_vmxnet3

RING: Cannot reserve memory

EAL: Error - exiting with code: 1

  Cause: File exists

Done.



Also, I observed some phenomenon:

1. If I stopped my program, dpdk-pdump can work again for the first time,
but it still failed when I tried to use it again.

2. I've tried the same codes on ubuntu-18.04, and the issue was not
observed.



I've checked these:

1.Before compiling DPDK, set both "CONFIG_RTE_LIBRTE_PMD_PCAP" and
"CONFIG_RTE_LIBRTE_PDUMP" to y.

2.Disabled ASRL with "echo 0 | sudo tee /proc/sys/kernel/randomize_va_space"

3.Increased hugepage to 2048, but the issue was not gone.



*Here's my environment:

OS: CentOS Linux release 7.6.1810 (Core)

kernel version: 3.10.0-957.el7.x86_64

CPU: 8 core

Hard disk/ram disk: 64 GB/32 GB

Numa node: 2

DPDK version: 17.11

used config: defconfig_x86_64-native-linuxapp-gcc

hugepage: 1024



*Here's hugepage information:

# grep Huge /proc/meminfo

AnonHugePages:     10240 kB

HugePages_Total:    2048

HugePages_Free:     1536

HugePages_Rsvd:        0

HugePages_Surp:        0

Hugepagesize:       2048 kB



*Here's numa node information:

# dmesg | grep -i numa

[ 0.000000] NUMA: Node 0 [mem 0x00000000-0x0009ffff] + [mem
0x00100000-0xbfffffff] -> [mem 0x00000000-0xbfffffff]

[ 0.000000] NUMA: Node 0 [mem 0x00000000-0xbfffffff] + [mem
0x100000000-0x83fffffff] -> [mem 0x00000000-0x83fffffff]



I'd be very apricated for your help. Many thanks for your reading.

Best Regards,

Violet