@brettcannon, this is what I was talking about. 😄

⚠️⚠️⚠️ Buildbot failure ⚠️⚠️⚠️

Hi! The buildbot s390x RHEL8 LTO 3.x has failed when building commit b2fc549.

What do you need to do:

1. Don't panic.
2. Check the buildbot page in the devguide if you don't know what the buildbots are or how they work.
3. Go to the page of the buildbot that failed (https://buildbot.python.org/all/#builders/567/builds/3626) and take a look at the build logs.
4. Check if the failure is related to this commit (b2fc549) or if it is a false positive.
5. If the failure is related to this commit, please, reflect that on the issue and make a new Pull Request with a fix.

You can take a look at the buildbot page here:

https://buildbot.python.org/all/#builders/567/builds/3626

Failed tests:

• test_asyncio

Failed subtests:

• test_subprocess_consistent_callbacks - test.test_asyncio.test_subprocess.SubprocessThreadedWatcherTests.test_subprocess_consistent_callbacks

Summary of the results of the build (if available):

== Tests result: FAILURE then FAILURE ==

415 tests OK.

10 slowest tests:

• test_tools: 3 min 16 sec
• test_gdb: 2 min 48 sec
• test_concurrent_futures: 2 min 31 sec
• test_multiprocessing_spawn: 1 min 38 sec
• test_multiprocessing_fork: 1 min 27 sec
• test_signal: 1 min 14 sec
• test_multiprocessing_forkserver: 1 min 13 sec
• test_asyncio: 1 min 4 sec
• test_logging: 44.6 sec
• test_math: 41.1 sec

1 test failed:
test_asyncio

18 tests skipped:
test_check_c_globals test_devpoll test_ioctl test_kqueue
test_launcher test_msilib test_nis test_peg_generator
test_perf_profiler test_startfile test_tix test_tkinter test_ttk
test_winconsoleio test_winreg test_winsound test_wmi
test_zipfile64

1 re-run test:
test_asyncio

Total duration: 5 min 7 sec

Traceback (most recent call last):
  File "/home/dje/cpython-buildarea/3.x.edelsohn-rhel8-z.lto/build/Lib/test/test_asyncio/test_subprocess.py", line 771, in test_subprocess_consistent_callbacks
    self.loop.run_until_complete(main())
  File "/home/dje/cpython-buildarea/3.x.edelsohn-rhel8-z.lto/build/Lib/asyncio/base_events.py", line 664, in run_until_complete
    return future.result()
           ^^^^^^^^^^^^^^^
  File "/home/dje/cpython-buildarea/3.x.edelsohn-rhel8-z.lto/build/Lib/test/test_asyncio/test_subprocess.py", line 763, in main
    self.assertEqual(events, [
AssertionError: Lists differ: [('pi[69 chars]), 'process_exited', 'pipe_connection_lost', '[17 chars]ost'] != [('pi[69 chars]), 'pipe_connection_lost', 'pipe_connection_lo[17 chars]ted']


Traceback (most recent call last):
  File "/home/dje/cpython-buildarea/3.x.edelsohn-rhel8-z.lto/build/Lib/test/test_asyncio/test_subprocess.py", line 771, in test_subprocess_consistent_callbacks
    self.loop.run_until_complete(main())
  File "/home/dje/cpython-buildarea/3.x.edelsohn-rhel8-z.lto/build/Lib/asyncio/base_events.py", line 664, in run_until_complete
    return future.result()
           ^^^^^^^^^^^^^^^
  File "/home/dje/cpython-buildarea/3.x.edelsohn-rhel8-z.lto/build/Lib/test/test_asyncio/test_subprocess.py", line 763, in main
    self.assertEqual(events, [
AssertionError: Lists differ: ['process_exited', ('pipe_data_received', 1, b'stdout')] != [('pipe_data_received', 1, b'stdout'), ('p[95 chars]ted']

I'm fairly sure that buildbot failure is unrelated to this PR. If the buildbot fails the next commit then I'll look into this failure further.

OOI, did you benchmark this? Startup times are sensitive to the changes in the import system.