Description of problem:

the CRD API documentation on PerformanceProfile and Hugepages, says:

It is important to notice that setting hugepages default size to 1G will remove all 2M related folders from the node and it will be impossible to configure 2M hugepages under the node.

on: 

https://docs.openshift.com/container-platform/4.16/rest_api/node_apis/performanceprofile-performance-openshift-io-v2.html#spec-hugepages

on 4.16 doc, but even on 4.11.

I think this is wrong, at least I have tested and it is possible. Also, we have other documents were we give example of exactly what we are saying is not possible (configuring multiple sizes of hugepages with 1G default):

https://docs.openshift.com/container-platform/4.14/scalability_and_performance/low_latency_tuning/cnf-tuning-low-latency-nodes-with-perf-profile.html#cnf-allocating-multiple-huge-page-sizes_cnf-low-latency-perf-profile

Maybe this is an ancient limitation that now is not correct?

Note from msivak@redhat.com: This was introduced in 4.6 time by this PR: https://github.com/openshift-kni/performance-addon-operators/pull/321 - we need to validate two flows to make sure the comment is obsolete. NUMA specific hugepages and NUMA agnostic (evenly split) hugepages.

Version-Release number of selected component (if applicable):

    4.11 to 4.16

How reproducible:

Steps to Reproduce:

    1.
    2.
    3.
    

Actual results:

Expected results:

Additional info: