Hi Neale, sorry for the late response cause of the Chinese National Day holiday. I noticed this patch has already merged to master branch, so I run the test on master branch directly and found the performance drop still exist compared to 18.07 which is almost 28%.
Regards, Yuwei From: [email protected] [mailto:[email protected]] On Behalf Of Neale Ranns via Lists.Fd.Io Sent: Thursday, October 4, 2018 9:43 PM To: Zhang, Yuwei1 <[email protected]>; [email protected] Cc: [email protected] Subject: Re: [vpp-dev] Master branch l2bd test perf dop Hi Yuwei, Can you test this, please: https://gerrit.fd.io/r/#/c/15100/ thanks, neale De : "Zhang, Yuwei1" <[email protected]<mailto:[email protected]>> Date : vendredi 28 septembre 2018 à 03:08 À : "Neale Ranns (nranns)" <[email protected]<mailto:[email protected]>>, "[email protected]<mailto:[email protected]>" <[email protected]<mailto:[email protected]>> Objet : RE: [vpp-dev] Master branch l2bd test perf dop Hi Neale, I assume the replications should related to the interfaces in the bridge, right? I just have 2 interfaces in the bridge which means one of the interface receive traffic and another send out. In the 64K size packet test case, the performance have almost 35% drop. I didn’t do other case yet. Regards, Yuwei From: Neale Ranns (nranns) [mailto:[email protected]] Sent: Thursday, September 27, 2018 8:31 PM To: Zhang, Yuwei1 <[email protected]<mailto:[email protected]>>; [email protected]<mailto:[email protected]> Subject: Re: [vpp-dev] Master branch l2bd test perf dop Hi Yuwei, There was a change to the l2flood node recently: https://gerrit.fd.io/r/#/c/13578/ where we use the buffer clone mechanism rather than free-recycle. I would expect the CPU cycles per invocation of the l2-flood node to increase, but the number of invocations of l2flood to decrease (w.r.t. the interface-tx node). How many replications does your test perform and is there a trend for perf change versus number of replications? Thanks, Neale De : <[email protected]<mailto:[email protected]>> au nom de Zhang Yuwei <[email protected]<mailto:[email protected]>> Date : jeudi 27 septembre 2018 à 05:02 À : "[email protected]<mailto:[email protected]>" <[email protected]<mailto:[email protected]>> Objet : [vpp-dev] Master branch l2bd test perf dop Hi All, In our recent test, I found a performance drop in master branch. I execute the l2bd test case in a 2.5GHZ CPU and found almost 35% drop compared to 18.07 release. My test is set two NIC ports to a same bridge domain and send traffic to test the l2 forward performance. I found in the master branch, l2flood function consume much more CPU cycles than 18.07 which means any test use the l2flood function will also have a performance drop. Can anybody kindly help to check this issue? Thanks a lot. Regards, Yuwei
-=-=-=-=-=-=-=-=-=-=-=- Links: You receive all messages sent to this group. View/Reply Online (#10795): https://lists.fd.io/g/vpp-dev/message/10795 Mute This Topic: https://lists.fd.io/mt/26289209/21656 Group Owner: [email protected] Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub [[email protected]] -=-=-=-=-=-=-=-=-=-=-=-
