I disagree. The most busy thread you found with top -H was only 20% of 1 core. What is the overall CPU utilization on that host? If it isn't above 60-70% then there is no CPU bottleneck.
Seems unlikely, that would look like repeated calls rather than spinning and it's bounded.
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-2018:2819