Ensure that optimize attributes on closures are inherited by the shim.

[veluca93]

Tracking issue: #54882
Stabilization PR: #157273

[rustbot]

Some changes occurred to MIR optimizations

cc @rust-lang/wg-mir-opt

[rustbot]

r? @mejrs

rustbot has assigned @mejrs.
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: compiler
• compiler expanded to 73 candidates
• Random selection from 19 candidates

[wesleywiser]

r=me with those changes 🙂

View changes since this review

[mejrs]

r=me with those changes 🙂

View changes since this review

I still want to review this as well.

[veluca93]

@rustbot ready

(is there a way for me to re-trigger the CI for things that appear to be spurious failures like this one?)

[mejrs]

What is the behavior of this?

#![feature(optimize_attribute)]

fn main() {
    f(
        #[optimize(size)] || { },
    );
}

fn f(_: impl Fn()) {}

Note that the attribute is applied to a function argument, not the expression (hence it does not require #![feature(stmt_expr_attributes)]).

View changes since this review

[mejrs]

(I just noticed there's a stabilization PR in flight, I have raised it there as well)

[veluca93]

What is the behavior of this?

#![feature(optimize_attribute)]

fn main() {
    f(
        #[optimize(size)] || { },
    );
}

fn f(_: impl Fn()) {}

Note that the attribute is applied to a function argument, not the expression (hence it does not require #![feature(stmt_expr_attributes)]).

View changes since this review

In practice that's going to depend on inlining (i.e. if the closure gets MIR-inlined into f, the attribute has no effect, but if not, then the closure's body should be optimized for size).

[mejrs]

I see. Attributes in that position are somewhat ambiguous so a test would be nice, if it is feasible.

(is there a way for me to re-trigger the CI for things that appear to be spurious failures like this one?)

You can force push nothing.

[veluca93]

I see. Attributes in that position are somewhat ambiguous so a test would be nice, if it is feasible.

I think I did so.

(is there a way for me to re-trigger the CI for things that appear to be spurious failures like this one?)

You can force push nothing.

Good to know!

[mejrs]

@bors r=mejrs,wesleywiser rollup

[rust-bors]

📌 Commit 0959e4d has been approved by mejrs,wesleywiser

It is now in the queue for this repository.