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]> au nom de Zhang Yuwei <[email protected]> Date : jeudi 27 septembre 2018 à 05:02 À : "[email protected]" <[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 (#10691): https://lists.fd.io/g/vpp-dev/message/10691 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]] -=-=-=-=-=-=-=-=-=-=-=-
