[Jules-Bertholet]

The checks affect inlining…

@rustbot label A-compiletest A-compiletest-directives

[rustbot]

compiletest directives have been modified. Please add or update docs for the
new or modified directive in src/doc/rustc-dev-guide/.

The rustc-dev-guide subtree was changed. If your future PRs only touch the subtree, consider submitting them directly to rust-lang/rustc-dev-guide, which is where the document is primarily maintained (and has faster CI).

cc @BoxyUwU, @tshepang

Some changes occurred in src/tools/compiletest

cc @jieyouxu

The run-make-support library was changed

cc @jieyouxu

[rustbot]

r? @wesleywiser

rustbot has assigned @wesleywiser.
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: @jieyouxu, @oli-obk, @wesleywiser, bootstrap
• @jieyouxu, @oli-obk, @wesleywiser, bootstrap expanded to 8 candidates
• Random selection from Mark-Simulacrum, clubby789, jieyouxu, oli-obk, wesleywiser

[saethlin]

This is a lot of copy+paste to make a single test behave better for your uncommon configuration.

Is it reasonable to extend the meaning of ignore-std-debug-assertions to mean debug assertions or overflow checks? The use case is the same.

[Jules-Bertholet]

Is it reasonable to extend the meaning of ignore-std-debug-assertions to mean debug assertions or overflow checks? The use case is the same.

This turns out to be a little complicated, because if exactly one of the two is enabled, you would want ignore-std-debug-assertions and needs-std-debug-assertions tests to both be ignored, but the compiletest code isn't set up for that.

[saethlin]

needs-std-debug-assertions was never used. It was added 2 years ago by #131913 so there's a small enough amount of git history I can reasonably grep for it. That PR alludes to a predecessor only-debug which I cannot find.

[Jules-Bertholet]

The point remains that the code is not set up for an ignores- without a corresponding needs-. I would rather have a little duplication than make big changes to address that