On 10/23/26 5:57 PM, Dumitru Ceara via dev wrote: > Hi Rosemarie, > > On 10/23/26 7:15 AM, Rosemarie O'Riorden wrote: >> This commit adds a test that developers can run to check if the changes >> they've made cause a regression in the performance of ovn-controller or >> ovn-northd. >> >> Run from inside the sandbox and compare two commits' runtime and peak >> memory usage for each process. >> > > Thanks for the new revision! I think this benchmark will be very useful > indeed. > >> This test creates a topology consisting of logical switches and routers, >> ACLs, port groups, NAT, DHCP, DNS, QoS, routing policies, and load >> balancers. The user can also provide their own database file for the >> benchmark. >> >> * To run the test with defaults (200 nodes, track northd and controller): >> ./ovn-benchmark.sh >> * To run with 50 nodes: ./ovn-benchmark.sh 50 >> * To run with 30 nodes and only track ovn-northd: >> ./ovn-benchmark.sh 30 northd >> * To run with Valgrind Massif tracking allocations: >> ./ovn-benchmark.sh --valgrind > > I'm still reviewing this patch but one thing I saw while testing it out > is that running ovn-benchmark.sh with valgrind when OVN has been built > with ubsan/asan doesn't work. We should probably document/check for that. FWIW, this is the same as building with asan/ubsan and then trying to run make check-valgrind. These two are just incompatible. Instrumentation is also incompatible with gdb. We may have a note on that, but it can also be treated as a common knowledge that isn't OVN-specific.
Best regards, Ilya Maximets. _______________________________________________ dev mailing list [email protected] https://mail.openvswitch.org/mailman/listinfo/ovs-dev
