is_multicore Releases.v5_0_0 is false!

[talex5]

ocaml-version claims that 5.0.0 isn't multicore:

utop # Ocaml_version.(Configure_options.is_multicore Releases.v5_0_0);;
- : bool = false

This is causing CI to fail (e.g. mirage/mirage-crypto#169 (comment)).

[dra27]

cf. #58 (comment). There's an open question as to what this function means for 5.x. In the past, it was referring to if there was a multicore variant of OCaml, but OCaml 5.x is multicore.

[talex5]

The doc comment says:

(** [is_multicore t] is [true] if this version is a multicore-capable release. *)

Releases.v5_0_0 is surely a multicore-capable release.

[dra27]

Yes, indeed - I'm not arguing, it's just that the function isn't actually used that way (even in ocaml-version - see later in the opam package selection) so it's just not an instantly obvious fix (I thought there was a workaround for this in ocaml-ci in progress?)

[tmcgilchrist]

There is a fix in progress for this. Antonin was looking into it.

…

On Mon, 6 Feb 2023 at 12:36 am, David Allsopp ***@***.***> wrote: Yes, indeed - I'm not arguing, it's just that the function isn't actually used that way (even in ocaml-version - see later in the opam package selection) so it's just not an instantly obvious fix (I thought there was a workaround for this in ocaml-ci in progress?) — Reply to this email directly, view it on GitHub <#59 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AABJXONKSLPW42CLZ7DYIZTWV6UENANCNFSM6AAAAAAUQPWVYQ> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>