Anton Bachin
Anton Bachin
I tried reproducing this on my macOS system, with what may be success. The program was: ```ocaml (* ocamlfind opt -linkpkg -package lwt.unix thread_count.ml && ./a.out *) open Lwt.Infix let...
Okay, the attached commit at least keeps the worker thread count synchronized with reality (at least more so than before). Making Lwt "handle" the failure internally, act as if the...
Thanks for the response. Is it correct to assume that you want the whole `EAGAIN` fixed, not just have a correct thread count? :)
Agreed, I remember worrying about thread releasing as well. I think a workaround that may be practical for now is to call `Lwt_unix.set_pool_size` and set the size to something relatively...
Thanks for the info. I'm renaming the issue to reflect the work remaining. Thanks for reporting!
> I think it is around 2x / 3x time. Is this based on a measurement? I would not have expected the impact to be such, though I suppose it's...
We would probably need to restore the old "combine expressions" for this (arithmetic/diffs on coverage output). If there is indeed interest in this, I can add them to Bisect_ppx.
> it is based on CI timings, bisect job: 73 mins, regular job ~ 40 mins. Ok. Just want to note: that difference would not be only from the instrumentation...
I'll add a bit of explanation to base it on: > I didn't understand why "h" at the end of "length" is highlighted or why `Option.iter2` isn't highlighted. Bisect instruments...
If you mean a real example, I believe there is not.