Does/Could promesa's channel implementation suffer from this same memory leak issue as the new core.async alpha? https://mail.openjdk.org/pipermail/loom-dev/2026-January/008172.html https://www.reddit.com/r/Clojure/comments/1qatx3x/comment/nz600g0/
I'm pretty sure yes. In any case, from my point of view, leaving a thread (go-block) incomplete, just leaving channel unclose and waiting it magically GC is a bit strange. A good practica for avoid strange situations is just have good teardown of all processes.
I never suffered this issue (the memory leak) with vthreads because I always tend to have code that has clear "tear-down". Always closing all channel and stopping all possible v/threads and channels