Github user shinrich commented on the issue: https://github.com/apache/trafficserver/pull/1259 Some details in the jira description. We noticed this when we moved to a newer compiler and started building against openssl 1.0.2. Somewhere in that change, the communication patterns must have resulted in more, smaller reads which created longer io block chains. We saw a very noticeable increase in CPU utilization, and perf top pointed the finger at read_avail.
--- If your project is set up for it, you can reply to this email and have your reply appear on GitHub as well. If your project does not have this feature enabled and wishes so, or if the feature is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. ---