Re: the problem of rust :-(

[Greg Troxel <gdt%lexort.com@localhost>]

Tobias Nygren <tnn%NetBSD.org@localhost> writes:

> On Mon, 24 Jun 2024 09:11:09 +0000
> pin <voidpin%protonmail.com@localhost> wrote:
>
>> As we approach the Q2-freeze, the following can no longer be updated:
>>  - devel/cargo-c: pkgsrc version 0.9.32; current upstream release, 0.10.0 requires Rust-1.77.2
>>  - devel/cargo-modules: pkgsrc version 0.15.5; current upstream release, 0.16.0 requires Rust-1.78.0
>>  - editors/tp-note: pkgsrc version 1.24.2; current upstream release, 1.24.4 requires Rust 1.77.2
>> 
>> This is better than expected at this point.

That is a pretty short list of what's being held back, and it doesn't
seem like a major user problem, compared to entire computers breaking.

tp-note would seem to have some software engineering hygiene problems.
It is not ok to change prerequisites in a micro release.  Maybe I am
misreading (or maybe this is just evidence of rust culture :-( ).

> I can provide access to a 16-core aarch64 box if someone wants to
> assist with debugging the rust stack overflow problem (which I think
> is the major blocker we currently have?).
>
> I have on my TODO list to look at it but it is unlikely I will have time
> before Q2-freeze.

We have been in freeze since June 20, and in not-updating-rust prefreeze
since June 1.  I don't follow the "approach the freeze" language.

So it doesn't really matter if it is fixed before the branch is cut or
not.  We are well past the time where we can consider updating rust.

Except that if it is fixed in the next week, that is awesome and we can
see where we are for updating rust soon after the freeze is over.  While
I think aarch64 is perhaps the biggest deal, I am not sure we are ready
to throw earmv7hf-el overboard.