Re: Dpdk-pdump not capturing packet for vmxnet3 device

[Stephen Hemminger <stephen@networkplumber.org>]

On Wed, 19 Mar 2025 07:21:30 +0000
"Rajesh Goel (rajegoel)" <rajegoel@cisco.com> wrote:

> Hi DPDK community,
> 
> I need some urgent help.
> 
> 1)      Iam running my dpdk application on a VMware based VM. Iam using 20.11.9 dpdk version
> Iam directly connected to the PCI device to receive pkts on my application.
> 
>       :
>       EAL: Detected 16 lcore(s)
>       EAL: Detected 1 NUMA nodes
>       EAL: Detected shared linkage of DPDK
>       EAL: Multi-process socket /var/run/dpdk/rte/mp_socket
>       EAL: Selected IOVA mode 'PA'
>       EAL: No available hugepages reported in hugepages-1048576kB
>       EAL: Probing VFIO support...
>       EAL: VFIO support initialized
>       EAL: Probe PCI driver: net_vmxnet3 (15ad:7b0) device: 0000:04:00.0 (socket 0)
>       EAL: No legacy callbacks, legacy socket not created
>       :
> 
> 2)      Iam trying to use dpdk-pdump application to capture packets.
> 
>       dpdk-pdump -l 3 -- --pdump 'device_id=0000:04:00.0,queue=*,rx-dev=/tmp/rx.pcap,total-num-mbufs=8192,ring-size=128'
> 
>       EAL: Detected 16 lcore(s)
>       EAL: Detected 1 NUMA nodes
>       EAL: Detected static linkage of DPDK
>       EAL: Multi-process socket /var/run/dpdk/rte/mp_socket_644376_50d5dc20612f8
>       EAL: Selected IOVA mode 'PA'
>       EAL: Probing VFIO support...
>       EAL: VFIO support initialized
>       EAL: Probe PCI driver: net_vmxnet3 (15ad:7b0) device: 0000:04:00.0 (socket 0)
>       Port 1 MAC: 02 70 63 61 70 00
>        core (3), capture for (1) tuples
>        - port 0 device (0000:04:00.0) queue 65535
>       ^C
> 
>       Signal 2 received, preparing to exit...
>       ##### PDUMP DEBUG STATS #####
>        -packets dequeued:                                       0
>        -packets transmitted to vdev:            0
>        -packets freed:                                  0
> 
> This is the mempool and ring which got created in app side
> 
>             ring <MP_pdump_pool_0>@0x10045db00
>               flags=0
>               size=16384
>               capacity=16383
>               ct=0
>               ch=0
>               pt=8192
>               ph=8192
>               used=8192
>               avail=8191
>             ring <rx_ring_0>@0x102802680
>               flags=0
>               size=128
>               capacity=127
>               ct=0
>               ch=0
>               pt=0
>               ph=0
>               used=0
>               avail=127
> 
> This is the port output, port 0 is application and port 1 is created by pdump and no pkt is seen to be captured.
> 
> Port 0 stats:
>  - link status: 1
>  - Pkts in:   50
>  - Pkts out:  0
>  - total ibytes: 4024
>  - total obytes: 0
>  - rx miss: 0
>  - In Errs:   0
>  - Out Errs:  0
>  - Mbuf Errs: 0
>  - Q Rx Packets:
> 
> Port 1 stats:
>  - link status: 1
>  - Pkts in:   0
>  - Pkts out:  0
>  - total ibytes: 0
>  - total obytes: 0
>  - rx miss: 0
>  - In Errs:   0
>  - Out Errs:  0
>  - Mbuf Errs: 0
>  - Q Rx Packets:
> 
> 3)      I see the same result while using dpdk-testpmd on the APP side.
> 
> 4)      As per this page, https://doc.dpdk.org/guides/tools/pdump.html.
>       Is this applicable to vmxnet3 device here ??
> 
> *       As of now the dpdk-pdump tool cannot capture the packets of virtual devices in the primary process due to a bug in the ethdev library. Due to this bug, in a multi process context, when the primary and secondary have different ports set, then the secondary process (here the dpdk-pdump tool) overwrites the rte_eth_devices[] entries of the primary process.
> 
> Thanks
> Rajesh
> 
> 

Try dpdk-dumpcap instead. Pdump is limited and will not get any new features.
I am going to propose deprecating pdump in 25.11.