David Allsopp

This is something which it would be nice to fix properly in opam, but we're a bit nervous that making a small tweak here runs a risk of making the...

@NathanReb - are you likely to be able to work on this in the next few months (just looking at milestones)?

Attempting to summarise the two main questions/points I have at this stage: 1. The native Windows implementation (especially on mingw-w64) is much better if it can use `_vscprintf_l`, which I...

Ah, it turns out the locales weren't being tested in various places! I've tweaked ocamltest to add `run_can_skip` as a setting. This defaults to false but if set to true...

Thanks, @shindere! To answer points in turn: - It's nice since #9797 that the only C code in ocamltest is the very important run_stub... so if we were to add...

Right, homework done properly this time and an extra commit pushed. It turns out that since Windows 10 1809, the QPC is locked at the Hyper-V value of 100ns so...

Performance improvement removed (it can be re-opened for 5.1, if necessary): this PR now just removes code. The change in `caml_time_counter` is to move from static initialisation in the function...

> Is it okay to stop generating `utils/config_boot.mli` and `utils/config_main.mli` as the PR does? > > These files look not so useful since the corresponding implementations won't actually be compiled...

> > These files are there so that [..] > > This reads like a design aspect that could be documented by a comment somewhere in the configuration / build...

Oops - it's not to do with the `IN_COREBOOT_CYCLE` stuff, it's simply that `utils/config.generated.ml` is erased during partialclean where it should now only be erased during `distclean`. PR incoming.