Please use this bug to backport https://github.com/openshift/cluster-network-operator/pull/530
Unless you have a reason not to do it for 4.4.0(high risk? not enough time?), you should do it for 4.4.0. Note that I didn't open the original bug, I just cloned it back to 4.4 and 4.3 because it was marked severy "urgent" in 4.5 and if a bug is marked severity "urgent" it stands to reason we need it fixed in older releases also. So you may want to ask Clayton for a second opinion.
The 4.3 bug 1820432 was seen on some customer Azure clusters, so as long as folks feel these changes are safe, I'd rather have them backported without waiting for 4.4.z, so we can get them VERIFIED in 4.4 and land the 4.3 backports sooner. Not my call, but weighing in in case a feather on the scale tips the decision ;).
Verified this bug on 4.4.0-0.nightly-2020-04-09-194724 oc get pod ovs-6rx8q -n openshift-sdn -o yaml | grep cpu cpu: 100m
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory, and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://access.redhat.com/errata/RHBA-2020:0581