Nicolás Ojeda Bär

This doesn't sound like a good idea to me. Many people open `Printf`, or alias it `module P = Printf` or do `let printf = Printf.printf`, or ... treating `Printf.printf`...

> is that because you couldn't find that manual or because you think that manual is so brief and so high-level that it doesn't actually serve the purpose of a...

> or `dune subst` shouldn't be executed? I don't remember why `dune subst` is in the script. Perhaps it is enough to remove it?

> I removed the `dune subst` action and re-uploaded the binaries for 0.21.0 and 0.20.1 (0.20.0 and 0.19.0 don't build), is the problem fixed? I don't have a working windows...

> Is the version correct for these ones? No:  I tried to reproduce locally by running the `build-mingw64.sh` script but it fails with  I will try to look...

(I can look into implementing this if there is a consensus.)

Note that the "hack" that you mention is already in `dune`'s codebase, it is used in `configurator`, see https://github.com/ocaml/dune/blob/bdfe5291bf2e080d64d8543c41ba55dd281bee22/otherlibs/configurator/src/v1.ml#L531-L537 and the code around it.

I am guessing this is due to the need for a specific implementation of the `caml_highres_clock` primitive in https://github.com/LexiFi/landmarks/blob/master/src/clock.c which uses the `rdtsc` instruction on x86. I'm not very familiar...

We don't have any `arm64` hardware to test this on, but you can give it a try by adding this to `clock.c`: ```c #elif defined(__aarch64__) uint64_t v; asm volatile ("mrs...

cc @mlasson