coretest in miri: fix using unstable libtest features

[RalfJung]

Alternative (IMO preferable) to #153369. Also reverts that PR.

[rustbot]

r? @Mark-Simulacrum

rustbot has assigned @Mark-Simulacrum.
They will have a look at your PR within the next two weeks and either review your PR or reassign to another reviewer.

Use r? to explicitly pick a reviewer

Why was this reviewer chosen?

The reviewer was selected based on:

• Owners of files modified in this PR: bootstrap
• bootstrap expanded to 6 candidates
• Random selection from Mark-Simulacrum, clubby789, jieyouxu

[RalfJung]

@bors try jobs=x86_64-gnu-aux

[rustbot]

This PR was rebased onto a different main commit. Here's a range-diff highlighting what actually changed.

Rebasing is a normal part of keeping PRs up to date, so no action is needed—this note is just to help reviewers.

[rust-bors]

☀️ Try build successful (CI)
Build commit: dec0b01 (dec0b01a659033e24d58dc8a50312191a40fd290, parent: f8704be04fe1150527fc2cf21dd44327f0fe87fb)

[cuviper]

I think for the failure we saw on beta, you'll have to also set that here:
https://github.com/RalfJung/rust/blob/07a902a84445c32fa8106fb2b2ee44c38d43b4c3/src/ci/docker/host-x86_64/x86_64-gnu-miri/check-miri.sh

I was reproducing that failure on main by setting rust.channel = "beta" and running the command:

$ ./x miri --stage 2 library/core --test-args notest

... which is one of the last commands in that check script, with no MIRIFLAGS at all yet.
(It does work if I add the env-forward for that command though, as expected.)

[cuviper]

It might be easier to have bootstrap set/append that to MIRIFLAGS itself, rather than trying to find all the places we call it in CI.

[RalfJung]

Oh right, we're also invoking ./x miri there...

(It does work if I add the env-forward for that command though, as expected.)

Thanks for confirming!

[RalfJung]

It might be easier to have bootstrap set/append that to MIRIFLAGS itself, rather than trying to find all the places we call it in CI.

Fair point. I tried that in the latest revision that I just pushed.

[cuviper]

cc @Kobzol -- this is probably more like the bootstrap-oriented fix you had in mind.