yeah I pulled a similar looking list of tests for the PROFILE_TASK to suggest to the Docker Python image maintainers - see https://github.com/docker-library/python/pull/404 and https://github.com/docker-library/python/issues/160.

I haven't run comparisons of that vs the default overly long profile task, but a comparison against a non PGO enabled cpython 3.7 build shows the types of speedups i'd expect from a decent PGO build.

doing proper comparisons against a full build would take a lot more time than I have right now.

It'd be reasonable to change the default task to something that is faster to build so long as we don't appear to be making major sacrifices in speed somewhere.

(your benchmark table's pickle result changes are inconsistent and odd, but that benchmark may not be very meaningful as written, i haven't looked at it)