Mark more locals as moved to avoid building drops for them. by cjgillot · Pull Request #158281 · rust-lang/rust

cjgillot · 2026-06-23T02:51:13Z

MIR building skips generating drops for moved-from locals in the topmost scope. This was only used for call terminators, but can be generalized to many other moves. This PR generalizes this to aggregate construction and many other assignments.

This avoids generating drops that would then be removed by drop elaboration.

This PR changes borrowck behaviour: some programs that were rejected are now accepted, see the last commit. That particular case was wrongly rejected.

Fixes #156713

~~Based on #158279 to remove a lint false-positive~~

rustbot · 2026-06-23T02:51:17Z

Some changes occurred to MIR optimizations

cc @rust-lang/wg-mir-opt

rustbot · 2026-06-23T02:51:19Z

r? @JohnTitor

rustbot has assigned @JohnTitor.
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 21 candidates

theemathas · 2026-06-23T03:52:09Z

Does this have user-visible behavior changes? e.g., see #156713

cjgillot · 2026-06-23T22:56:19Z

Yes. With this change, both cases in #156713 pass. We also have a few user-visible consequences with the drop order lint.

JohnTitor · 2026-06-27T01:23:05Z

@rustbot reroll

cjgillot · 2026-07-09T00:03:04Z

@bors try @rust-timer queue

Mark more locals as moved to avoid building drops for them.

theemathas · 2026-07-09T00:20:18Z

Since this PR has insta-stable changes, I think this will need to go through an FCP. Could you describe what the user-facing changes are, and lang-nominate the PR?

rust-bors · 2026-07-09T02:10:56Z

☀️ Try build successful (CI)
Build commit: 9ef0dc4 (9ef0dc4705f7d34e7ff65cb13ffa27f4263fedbf)
Base parent: 14cae68 (14cae681329a63c622a6e1fbe1d30f9374bc51d8)

rust-timer · 2026-07-09T02:51:53Z

Finished benchmarking commit (9ef0dc4): comparison URL.

Overall result: ❌✅ regressions and improvements - please read:

Benchmarking means the PR may be perf-sensitive. It's automatically marked not fit for rolling up. Overriding is possible but disadvised: it risks changing compiler perf.

Next, please: If you can, justify the regressions found in this try perf run in writing along with @rustbot label: +perf-regression-triaged. If not, fix the regressions and do another perf run. Neutral or positive results will clear the label automatically.

@bors rollup=never
@rustbot label: -S-waiting-on-perf +perf-regression

Instruction count

Our most reliable metric. Used to determine the overall result above. However, even this metric can be noisy.

	mean	range	count
Regressions ❌ (primary)	1.0%	[0.2%, 16.0%]	150
Regressions ❌ (secondary)	33.2%	[0.1%, 740.0%]	177
Improvements ✅ (primary)	-0.4%	[-0.6%, -0.2%]	11
Improvements ✅ (secondary)	-0.4%	[-2.0%, -0.1%]	16
All ❌✅ (primary)	0.9%	[-0.6%, 16.0%]	161

Max RSS (memory usage)

Results (primary -1.2%, secondary 0.6%)

A less reliable metric. May be of interest, but not used to determine the overall result above.

	mean	range	count
Regressions ❌ (primary)	-	-	0
Regressions ❌ (secondary)	3.7%	[3.7%, 3.7%]	1
Improvements ✅ (primary)	-1.2%	[-2.7%, -0.4%]	4
Improvements ✅ (secondary)	-0.9%	[-1.0%, -0.8%]	2
All ❌✅ (primary)	-1.2%	[-2.7%, -0.4%]	4

Cycles

Results (primary 4.5%, secondary 75.6%)

A less reliable metric. May be of interest, but not used to determine the overall result above.

	mean	range	count
Regressions ❌ (primary)	4.5%	[2.0%, 7.2%]	9
Regressions ❌ (secondary)	90.1%	[0.8%, 234.3%]	22
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-4.4%	[-9.4%, -2.3%]	4
All ❌✅ (primary)	4.5%	[2.0%, 7.2%]	9

Binary size

Results (primary 0.0%, secondary 0.0%)

A less reliable metric. May be of interest, but not used to determine the overall result above.

	mean	range	count
Regressions ❌ (primary)	0.0%	[0.0%, 0.1%]	38
Regressions ❌ (secondary)	0.0%	[0.0%, 0.1%]	23
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	0.0%	[0.0%, 0.1%]	38

Bootstrap: 488.761s -> 487.76s (-0.20%)
Artifact size: 388.78 MiB -> 385.17 MiB (-0.93%)

cjgillot · 2026-07-11T18:48:13Z

@bors try @rust-timer queue

Mark more locals as moved to avoid building drops for them.

rust-bors · 2026-07-11T20:55:52Z

☀️ Try build successful (CI)
Build commit: 68a2829 (68a28293b11549460b78874559bdb450ef9105e7)
Base parent: acf6a00 (acf6a00aff5d8aa8892aadff0b48364b50f97cf4)

rust-timer · 2026-07-11T21:37:33Z

Finished benchmarking commit (68a2829): comparison URL.

Overall result: ❌✅ regressions and improvements - please read:

Benchmarking means the PR may be perf-sensitive. It's automatically marked not fit for rolling up. Overriding is possible but disadvised: it risks changing compiler perf.

Next, please: If you can, justify the regressions found in this try perf run in writing along with @rustbot label: +perf-regression-triaged. If not, fix the regressions and do another perf run. Neutral or positive results will clear the label automatically.

@bors rollup=never
@rustbot label: -S-waiting-on-perf +perf-regression

Instruction count

Our most reliable metric. Used to determine the overall result above. However, even this metric can be noisy.

	mean	range	count
Regressions ❌ (primary)	-	-	0
Regressions ❌ (secondary)	0.2%	[0.0%, 0.4%]	5
Improvements ✅ (primary)	-0.4%	[-0.9%, -0.2%]	20
Improvements ✅ (secondary)	-0.5%	[-0.8%, -0.0%]	16
All ❌✅ (primary)	-0.4%	[-0.9%, -0.2%]	20

Max RSS (memory usage)

Results (primary 1.7%, secondary 2.1%)

A less reliable metric. May be of interest, but not used to determine the overall result above.

	mean	range	count
Regressions ❌ (primary)	4.4%	[4.4%, 4.4%]	1
Regressions ❌ (secondary)	3.1%	[0.6%, 4.7%]	3
Improvements ✅ (primary)	-0.9%	[-0.9%, -0.9%]	1
Improvements ✅ (secondary)	-0.9%	[-0.9%, -0.9%]	1
All ❌✅ (primary)	1.7%	[-0.9%, 4.4%]	2

Cycles

Results (secondary 6.5%)

A less reliable metric. May be of interest, but not used to determine the overall result above.

	mean	range	count
Regressions ❌ (primary)	-	-	0
Regressions ❌ (secondary)	6.5%	[2.8%, 15.0%]	8
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	-	-	0

Binary size

This perf run didn't have relevant results for this metric.

Bootstrap: 489.809s -> 489.874s (0.01%)
Artifact size: 389.31 MiB -> 389.84 MiB (0.14%)

And some more cases along the way. We have a specific optimization to avoid generating useless drops, use it. In particular, aggregate construction are very similar to function calls for which this is designed.

rustbot · 2026-07-18T21:08:44Z

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.

dianne · 2026-07-26T10:34:55Z

        // look for moves of a local variable, like `MOVE(_X)`
-        let locals_moved = operands.iter().flat_map(|operand| match operand.node {
+        let local_moved = match operand {
            Operand::Copy(_) | Operand::Constant(_) | Operand::RuntimeChecks(_) => None,
            Operand::Move(place) => place.as_local(),
-        });
+        };

-        for local in locals_moved {
+        if let Some(local) = local_moved {
            // check if we have a Drop for this operand and -- if so
            // -- add it to the list of moved operands. Note that this
            // local might not have been an operand created for this
            // call, it could come from other places too.
-            if scope.drops.iter().any(|drop| drop.local == local && drop.kind == DropKind::Value) {
-                scope.moved_locals.push(local);
-            }
+            scope.moved_locals.insert(local);
        }


Nit: it looks like these matches could be combined.

Also, could you update the comments?

View changes since the review

dianne · 2026-07-26T11:15:24Z

+//! Regression test for issue #156713. In the `fails` case, borrowck was trying to check liveness
+//! of `bar` which had been moved to a match scrutinee.


I think it's slightly subtler than this: as I understand it, the problem with fails was the unwind edge from the drop of the operand of the tuple constructor. Since borrowck doesn't ignore dead unwind edges and the drop order on the unwind path from that was bad (it dropped foo before the match scrutinee temporary), we got an error.

View changes since the review

dianne · 2026-07-26T11:22:08Z

+// This compiles
+fn works() {
+    let foo = Foo;
+    let bar = Bar(&foo);
+    drop(match { (bar,) } {
+        args => args,
+    })
+}


Maybe a nit, but I'm not sure works fits as a test here. As I understand, the reason it compiled successfully before this PR is a longstanding MIR building bug: #47949 made it so the unwind path from the drop of the operand to the tuple constructor was missing the drop of the match scrutinee temporary, meaning there was no error from the match scrutinee being dropped after foo.

Having a test means we'll make sure it still works after #47949 is fixed, but then it feels like a more specific version of fails.

View changes since the review

rustbot added S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. T-compiler Relevant to the compiler team, which will review and decide on the PR/issue. T-libs Relevant to the library team, which will review and decide on the PR/issue. labels Jun 23, 2026

rustbot assigned JohnTitor Jun 23, 2026

theemathas added T-lang Relevant to the language team needs-fcp This change is insta-stable, or significant enough to need a team FCP to proceed. and removed T-libs Relevant to the library team, which will review and decide on the PR/issue. labels Jun 23, 2026

rustbot assigned jackh726 and unassigned JohnTitor Jun 27, 2026

cjgillot force-pushed the drop-moved-locals branch from 51861b3 to d128dba Compare July 1, 2026 23:37