Re: [vpp-dev] [csit-dev] API Change Proposal: explicit FIB table create and delete

Hi All, In the absence of any objections I have done: https://gerrit.fd.io/r/#/c/7819/ I’ll have a crack at the necessary CSIT changes. Is this: https://wiki.fd.io/view/CSIT/Tutorials/Vagrant/Virtualbox/Ubuntu still the recommended way to test CSIT code changes? Thanks, neale From: Dave

Re: [vpp-dev] user space TCP stack

Hi Hamid, Yes, we do have a userspace TCP stack but it is still under development. You can find examples of external apps here [1] and internal apps here [2, 3]. All of these use the binary api to interact with the session layer code. We’ll soon publish a wrapper library that should make

[vpp-dev] API Change: Dedicated SW interface Event

Hi All, I would like to propose the addition of a dedicated SW interface event message type rather than overload the set flags request. The over-loading of types causes problems for the automatic API generation tools. https://gerrit.fd.io/r/#/c/7925/ regards, neale

Re: [vpp-dev] MEMIF Throuput Problem

You are passing each packet twice trough hostvpp so effectivelly your hostvpp performance is 8 Mpps per core. There are several factors which can impact performance (cpu speed, numa locality, memory channel utilisation) but still you cannot expect order of magnitude better numbers. Best

[vpp-dev] MEMIF Throuput Problem

Hi , I am insterested in connecting two instance of vpp by memif (one vpp is running in host and another vpp is running in a lxc container). I have achieved functionality goal by memif but I have problem in performance test. I have done the following steps respectively: 1. first of all I installed

[vpp-dev] user space TCP stack

​Hi, I trying to use mTCP in my application but it is very primitive, for example it permits only one epoll instance and does not have shutdown API equivalent. Does VPP support user space TCP stack similar to mTCP? if so where is a sample application using it? Thanks, hamid

Re: [vpp-dev] 回复: vpp cpu usage utility

Hi Dave, Thanks for your inputs.. I just experimenting based on your inputs.. when i push < 7Gbps i don't see any packet losses on dpdk side and the vector/call is still ~1.1. As you mentioned <2.0,means has lot of room left.. then i just increased the traffic to achieve bit rate of about 8gbps

Re: [vpp-dev] 17.07 Release

In general, there is no limit to the number of MPLS labels. If you can be more specific about what you are referring to (i.e. how many labels can be pushed per LSP, or how many MPLS lookups/pops per packet) then I can give you a more definitive answer. /neale -Original Message- From: