Thanks for the explanation, understood. However, I don't think it would be terribly difficult to demonstrate that these changes do address actual pains(*) in the situations they're supposed to address them. And the pains in question might in real world occur quite rarely, unexpectedly, and take some time/cost to debug and reproduce, which is why it would be good to be able to preemptively address them.

(*) For some values of "pain" -- many of the changes here are just for outputting nice-to-know timing information.