The example[0] is misleading with the given yaml[1]. We need to change the example yaml and the given example.
The performance operator uses CPUSets lists (perf doc and cpuset doc) meaning that instead of isolation cpu 0 and 4 it isolates 0 to 4 (same story for reserved cpu config). Additionally as per the same performance-addon-operators doc the union of reserved CPUs and isolated CPUs should include all online CPUs which this section of doc does not mention.
[0] - given example
Apply the isolated and reserved CPUs in the PerformanceProfile YAML. For example, you could set logical cores CPU0 and CPU4 as isolated, and logical cores CPU1 and CPU5 as reserved. When you configure reserved and isolated CPUs, the infra containers in pods use the reserved CPUs and the application containers use the isolated CPUs.
Which section(s) is the issue in?
https://docs.openshift.com/container-platform/4.8/scalability_and_performance/cnf-performance-addon-operator-for-low-latency-nodes.html#configuring_hyperthreading_for_a_cluster_cnf-master
Procedure step number 2
What needs fixing?
The example[0] is misleading with the given yaml[1]. We need to change the example yaml and the given example. The performance operator uses CPUSets lists (perf doc and cpuset doc) meaning that instead of isolation cpu 0 and 4 it isolates 0 to 4 (same story for
reserved
cpu config). Additionally as per the same performance-addon-operators doc the union ofreserved
CPUs andisolated
CPUs should include all online CPUs which this section of doc does not mention.[0] - given example
[1] - yaml