Maybe we have a path forward for performance stackful coroutine? More pieces need to fall in place but this looks promising.

Several ways to deal with the task, which are the performance implications? Clearly coroutines aren't the best tool for the job here.

In which case you want one or the other? This is illustrated in the Rust case which has its own struggles, but the question applies more largely in my opinion.

Beware of the problems coroutines might bring. Some are "obvious" and related to concurrency in general, but some are running deep in their design in C++.