Mark simplify_aggregate_to_copy mir-opt as unsound

[jieyouxu]

Mark the simplify_aggregate_to_copy mir-opt added in #128299 as unsound as it seems to miscompile the MCVE reported in #132353. The mir-opt can be re-enabled once this case is fixed.

fn pop_min(mut score2head: Vec<Option<usize>>) -> Option<usize> {
    loop {
        if let Some(col) = score2head[0] {
            score2head[0] = None;
            return Some(col);
        }
    }
}

fn main() {
    let min = pop_min(vec![Some(1)]);
    println!("min: {:?}", min);
    // panic happens here on beta in release mode
    // but not in debug mode
    min.unwrap();
}

This MCVE is included as a run-pass ui regression test in the first commit. I built the ui test with a nightly manually, and can reproduce the behavioral difference with -C opt-level=0 and -C opt-level=1. Locally, this ui test will fail unless it was run on a compiler built with the second commit marking the mir-opt as unsound thus disabling it by default.

This PR partially reverts commit e7386b3, reversing changes made to 02b1be1. The mir-opt implementation is just marked as unsound but not reverted to make reland reviews easier. Test changes are reverted if they were not pure additions. Tests added by the original PR received -Z unsound-mir-opts compile-flags.

cc @DianQK @cjgillot (PR author and reviewer of #128299)

[rustbot]

r? @Nadrieril

rustbot has assigned @Nadrieril.
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

[rustbot]

Some changes occurred to MIR optimizations

cc @rust-lang/wg-mir-opt

[jieyouxu]

Nominating for beta-backport as this is a mir-opt miscompile.
@rustbot label +beta-nominated

[jieyouxu]

Changes since last review:

• Added -Zunsound-mir-opts only to tests that are pure additions in Simplify the canonical clone method and the copy-like forms to copy #128299.
• Retained changes to tests/codegen/enum/unreachable_enum_default_branch.rs.

[jieyouxu]

Probably r? mir-opt

[cjgillot]

Indeed, we aren't allowed to introduce new dereferences without checking for aliasing.
r=me

[cjgillot]

Could you copy this into mir-opt test which shows pop_min.GVN.diff?

[jieyouxu]

To clarify, do you mean retain this ui test (for now), but also create a new mir-opt test which shows the GVN diff?

[DianQK]

You can use this for a mir-opt test if you like:

#![allow(internal_features)]
#![feature(rustc_attrs, core_intrinsics)]

#[no_mangle]
fn foo(v: &mut Option<i32>) -> Option<i32> {
    if let &Some(col) = get(&v) {
        *v = None;
        return Some(col);
    } else {
        unsafe { std::intrinsics::unreachable() }
    }
}

#[no_mangle]
#[inline(never)]
#[rustc_nounwind]
fn get(v: &Option<i32>) -> &Option<i32> {
    v
}

[jieyouxu]

Thanks, I tried to add a GVN.diff test w/ this snippet, lmw if the test needs adjustments.

[jieyouxu]

Added a GVN diff mir-opt test as requested.
@rustbot ready

[cjgillot]

@bors r+

[bors]

📌 Commit 29db55e has been approved by cjgillot

It is now in the queue for this repository.

[cjgillot]

Thanks!

[klensy]

Maybe don't rollup pr where optimization was disabled?

[DianQK]

Maybe don't rollup pr where optimization was disabled?

cc @matthiaskrgr

[matthiaskrgr]

ah, didn't see that this affected default mir opt levels
@bors rollup=never
shoud have been set then instead of iffy :)

[jieyouxu]

Oh oops should have rollup=never yeah.

[apiraino]

Beta backport accepted as per compiler team on Zulip, once it gets merged. A backport PR will be authored by the release team at the end of the current development cycle.

@rustbot label +beta-accepted

[bors]

⌛ Testing commit cfb4c05 with merge a0d98ff...

[bors]

☀️ Test successful - checks-actions
Approved by: cjgillot,DianQK
Pushing a0d98ff to master...

[rust-timer]

Finished benchmarking commit (a0d98ff): comparison URL.

Overall result: ❌✅ regressions and improvements - please read the text below

Our benchmarks found a performance regression caused by this PR.
This might be an actual regression, but it can also be just noise.

Next Steps:

• If the regression was expected or you think it can be justified,
  please write a comment with sufficient written justification, and add
  @rustbot label: +perf-regression-triaged to it, to mark the regression as triaged.
• If you think that you know of a way to resolve the regression, try to create
  a new PR with a fix for the regression.
• If you do not understand the regression or you think that it is just noise,
  you can ask the @rust-lang/wg-compiler-performance working group for help (members of this group
  were already notified of this PR).

@rustbot label: +perf-regression
cc @rust-lang/wg-compiler-performance

Instruction count

This is the most reliable metric that we have; it was used to determine the overall result at the top of this comment. However, even this metric can sometimes exhibit noise.

	mean	range	count
Regressions ❌ (primary)	0.7%	[0.3%, 1.9%]	14
Regressions ❌ (secondary)	0.7%	[0.2%, 2.3%]	16
Improvements ✅ (primary)	-0.3%	[-0.3%, -0.3%]	1
Improvements ✅ (secondary)	-0.3%	[-0.4%, -0.3%]	5
All ❌✅ (primary)	0.7%	[-0.3%, 1.9%]	15

Max RSS (memory usage)

Results (primary -2.2%, secondary 0.6%)

This is a less reliable metric that may be of interest but was not used to determine the overall result at the top of this comment.

	mean	range	count
Regressions ❌ (primary)	3.9%	[3.5%, 4.3%]	2
Regressions ❌ (secondary)	2.7%	[1.9%, 4.5%]	4
Improvements ✅ (primary)	-6.2%	[-9.1%, -3.5%]	3
Improvements ✅ (secondary)	-2.2%	[-2.6%, -2.0%]	3
All ❌✅ (primary)	-2.2%	[-9.1%, 4.3%]	5

Cycles

Results (primary 0.9%)

This is a less reliable metric that may be of interest but was not used to determine the overall result at the top of this comment.

	mean	range	count
Regressions ❌ (primary)	0.9%	[0.9%, 0.9%]	2
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	0.9%	[0.9%, 0.9%]	2

Binary size

Results (primary 0.1%, secondary 0.3%)

This is a less reliable metric that may be of interest but was not used to determine the overall result at the top of this comment.

	mean	range	count
Regressions ❌ (primary)	0.2%	[0.0%, 1.3%]	25
Regressions ❌ (secondary)	0.6%	[0.2%, 1.2%]	12
Improvements ✅ (primary)	-0.1%	[-0.3%, -0.0%]	11
Improvements ✅ (secondary)	-0.0%	[-0.0%, -0.0%]	10
All ❌✅ (primary)	0.1%	[-0.3%, 1.3%]	36

Bootstrap: 784.059s -> 782.659s (-0.18%)
Artifact size: 333.59 MiB -> 333.43 MiB (-0.05%)

[lqd]

This deactivated an unsound MIR optimization until it is fixed.

@rustbot label: +perf-regression-triaged