On Tue, Jul 10, 2018 at 10:38 AM, Ben Pfaff <[email protected]> wrote: > On Tue, Jun 26, 2018 at 04:16:42PM -0700, Darrell Ball wrote: > > Signed-off-by: Darrell Ball <[email protected]> > > I wonder whether coverage counters would be a better alternative? They > use thread-local data so incrementing them is cheaper than atomic > increments, at least if they are accessed from multiple CPUs. >
Thanks for the pointer on the coverage counters; since the counters are for sanity failures, the performance does not matter much. That being said, I have recently made some changes to the Fragmentation series where some same sanity failures also need counting, since some now bypass conntrack. It is probably best to do both together, after the Fragmentation series. _______________________________________________ dev mailing list [email protected] https://mail.openvswitch.org/mailman/listinfo/ovs-dev
