[vpp-dev] Segmentation fault when dpdk number-rx-queues > 1 in startup.conf

2022-03-02 Thread Xu, Ting
Hi, all I meet one issue that when I set dpdk rx queue number larger than 1 (which also enables RSS) in startup.conf, for example: dev default { # Number of receive queues, enables RSS # Default is 1 num-rx-queues 2 } When start VPP, it will meet

Re: [vpp-dev] #vpp-hoststack - Issue with UDP receiver application with epoll_wait()

2022-03-02 Thread Kyle Upton (kyupton) via lists.fd.io
Hi Florin, Just converted the app to edge triggered. And it's working great. Thanks! -Kyle On 3/2/22, 6:23 PM, "Florin Coras" wrote: Hi Kyle, Is the app by chance trying to do epoll in level trigger mode? If yes, support for it was pushed in sometime around 21.06 so the code you

Re: [vpp-dev] #vpp-hoststack - Issue with UDP receiver application with epoll_wait()

2022-03-02 Thread Florin Coras
Hi Kyle, Is the app by chance trying to do epoll in level trigger mode? If yes, support for it was pushed in sometime around 21.06 so the code you are running might be missing some features. It’s also less tested when compared with edge trigger. Conversely, if the app's running epoll in edge

[vpp-dev] #vpp-hoststack - Issue with UDP receiver application with epoll_wait()

2022-03-02 Thread Kyle Upton (kyupton) via lists.fd.io
Hi, I have a simple UDP recvfrom() application (using LD_PRELOAD). When I just use recvfrom(), I can receive packets played from a small PCAP file. However, when I wrap the recvfrom() in an epoll_wait(), I only receive one packet. The second packet appears to be in the Rx fifo (tail is greater

Re: [vpp-dev] Strange - SRIOV VF VLAN ID is disappeared #dpdk #forwarding

2022-03-02 Thread Eric
On Wed, Mar 2, 2022 at 05:03 PM, Ray Kinsella wrote: > > I think your approach of configuring VLAN on the sub-interfaces is the > correct one. If I take the way, the VF1 on the host can't receive any Request Reply, VF1 in zone1 linux namespace,tcpdump -i can not catch expect packet, the

Re: [vpp-dev] Strange - SRIOV VF VLAN ID is disappeared #dpdk #forwarding

2022-03-02 Thread Ray Kinsella
Eric writes: > Thanks in advance > > the trace packet is: > > --- > Packet 17 > > 00:18:16:753781: dpdk-input > dp0 rx queue 0 > buffer 0x9679b: current data 0, length 60, buffer-pool 0, ref-count 1, > totlen-nifb 0, trace handle 0x10 > ext-hdr-valid > l4-cksum-computed l4-cksum-correct > PKT