Use `drop_guard` in some places in {core,alloc,std} by GrigorenkoPV · Pull Request #161702 · rust-lang/rust

GrigorenkoPV · 2026-08-24T21:37:43Z

Tracking issue: Tracking Issue for core::mem::DropGuard #144426
Will conflict with Stabilize core::mem::DropGuard #161520
Add no-value DropGuard::new(|| ...) #161550 would also be cool occasionally

Didn't touch the places where manual impl Drops had #[inline] on their fn drop or where the guard type had other impls beside Drop and/or was named a lot.

No LLMs used, only pure human slop.

rustbot · 2026-08-24T21:37:48Z

r? @clarfonthey

rustbot has assigned @clarfonthey.
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: libs
libs expanded to 12 candidates
Random selection from JohnTitor, Mark-Simulacrum, clarfonthey, nia-e

clarfonthey · 2026-08-25T02:15:45Z

The way the code was moved around makes this a bit more difficult to review, but otherwise, it looks good to me. Just a limitation of how diffs work.

I think this is a strict improvement regardless of stabilisation. I don't think this should require more than a rebase from Yosh, so, it should be fine that this conflicts with the stabilisation PR.

@bors r+ rollup

Thank you!

rust-bors · 2026-08-25T02:15:47Z

📌 Commit a5a9d5d has been approved by clarfonthey

It is now in the queue for this repository.

…they Use `drop_guard` in some places in {core,alloc,std} - Tracking issue: rust-lang#144426 - Will conflict with rust-lang#161520 - rust-lang#161550 would also be cool occasionally Didn't touch the places where manual `impl Drop`s had `#[inline]` on their `fn drop` or where the guard type had other `impl`s beside `Drop` and/or was named a lot. No LLMs used, only pure human slop.

Rollup of 9 pull requests Successful merges: - #160533 (dirfd dir operations (3/4)) - #161294 (add crashtests [6/N]) - #160392 (library: motor: bump moto-rt ABI ver to 17) - #161646 (interpret: fix after_local_read handling) - #160819 (Rework `div_ceil` for nonzero integers) - #161591 (run `classify-runtime-const` test for `f128`) - #161680 (renovate: add lockfile update warning) - #161702 (Use `drop_guard` in some places in {core,alloc,std}) - #161709 (Stop rustfmt deleting attributes in fn params)

JonathanBrouwer · 2026-08-25T05:04:55Z

💔 I suspect this PR failed tests as part of a rollup
@bors r-

After fixing the problem, consider running a try job for the failed job before re-approving.

Link to failure: #161715 (comment)

rust-bors · 2026-08-25T05:05:02Z

This pull request was unapproved.

This PR was contained in a rollup (#161715), which was unapproved.

View changes since this unapproval

GrigorenkoPV · 2026-08-25T09:40:07Z

@rustbot ready

yoshuawuyts · 2026-08-25T11:28:19Z

@GrigorenkoPV thank you for filing this PR; this is really cool to see and a very helpful reference for how DropGuard actually ends up working out in real codebases. I'm loving the 40% code reduction compared to manually creating guards for each instance. Imo it validates the usefulness of the API.

GrigorenkoPV · 2026-08-25T16:42:27Z

I'm loving the 40% code reduction compared to manually creating guards for each instance.

Not sure how accurate that metric is, but yeah, code reduction is definitely real.

It is especially nice to not have to write all the types out again for the struct fields. And the generics for both the struct and the Drop impl (since those cannot be reused from the outer scope; in one place the old approach even required a PhantomData just to capture one of the generic parameters, which, with this change, is not needed anymore).

Imo it validates the usefulness of the API.

Absolutely. I remember touching std (something to do with Drain or CloneUninit) at some point a few years ago and wondering why there was no "canonical" way to to scope guards.

clarfonthey · 2026-08-27T00:00:38Z

@bors try

clarfonthey · 2026-08-30T00:40:49Z

Apologies for forcing you to rebate again .-.

rustbot · 2026-08-30T14:08:37Z

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.

GrigorenkoPV · 2026-08-30T14:10:11Z

The grindset never stops

@rustbot ready

clarfonthey · 2026-08-30T14:37:00Z

@bors try jobs=dist-various-1,dist-various-2,test-various,x86_64-msvc-1,dist-x86_64-linux-quick

Use `drop_guard` in some places in {core,alloc,std} try-job: dist-various-1 try-job: dist-various-2 try-job: test-various try-job: x86_64-msvc-1 try-job: dist-x86_64-linux-quick

rust-bors · 2026-08-30T17:15:32Z

☀️ Try build successful (CI)
Build commit: c961452 (c9614527a3ca600da9c6321bd53eb843491e6cc4)
Base parent: 93635a5 (93635a5d547dd8a0c51553a15376012346f17261)

clarfonthey · 2026-08-30T17:17:02Z

@bors r+

rust-bors · 2026-08-30T17:17:05Z

📌 Commit 4f404c1 has been approved by clarfonthey

It is now in the queue for this repository.

…they Use `drop_guard` in some places in {core,alloc,std} - Tracking issue: rust-lang#144426 - Will conflict with rust-lang#161520 - rust-lang#161550 would also be cool occasionally Didn't touch the places where manual `impl Drop`s had `#[inline]` on their `fn drop` or where the guard type had other `impl`s beside `Drop` and/or was named a lot. No LLMs used, only pure human slop.

…uwer Rollup of 8 pull requests Successful merges: - #137720 (support `#[target_feature(enable = ...)]` on `#[naked]` functions) - #160923 (Handle multiple action records in EH personality function) - #161788 (Check to ensure we're running against the correct LLVM version) - #161644 (bootstrap: Flatten and rename `compute_src_directory_via_git`) - #161702 (Use `drop_guard` in some places in {core,alloc,std}) - #161931 (Revert "Add rustc_test_entrypoint_marker") - #162015 (touch up "get attribute" docs.) - #162019 (remove a couple of redundant clones, thanks clippy)

…they Use `drop_guard` in some places in {core,alloc,std} - Tracking issue: rust-lang#144426 - Will conflict with rust-lang#161520 - rust-lang#161550 would also be cool occasionally Didn't touch the places where manual `impl Drop`s had `#[inline]` on their `fn drop` or where the guard type had other `impl`s beside `Drop` and/or was named a lot. No LLMs used, only pure human slop.

…uwer Rollup of 11 pull requests Successful merges: - #160923 (Handle multiple action records in EH personality function) - #161788 (Check to ensure we're running against the correct LLVM version) - #158303 (Always escape grapheme extenders in `str::escape_debug`) - #161644 (bootstrap: Flatten and rename `compute_src_directory_via_git`) - #161702 (Use `drop_guard` in some places in {core,alloc,std}) - #161713 (Add config for gram editor to .gitignore) - #161792 (Fix and improve diagnostics for lint `rust_2021_prefixes_incompatible_syntax`) - #161931 (Revert "Add rustc_test_entrypoint_marker") - #162015 (touch up "get attribute" docs.) - #162019 (remove a couple of redundant clones, thanks clippy) - #162023 (Remove redundant braces from `NonZero` doctests)

rust-timer · 2026-08-31T08:27:17Z

Note

This PR was benchmarked as part of triage of its containing rollup: triage URL.

Finished benchmarking commit (a643d98): comparison URL.

Overall result: ❌ regressions - please read:

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

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

	mean	range	count
Regressions ❌ (primary)	0.5%	[0.1%, 1.9%]	57
Regressions ❌ (secondary)	0.6%	[0.2%, 1.9%]	29
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	0.5%	[0.1%, 1.9%]	57

Max RSS (memory usage)

Results (primary 1.1%, secondary 2.6%)

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

	mean	range	count
Regressions ❌ (primary)	1.1%	[0.7%, 2.1%]	4
Regressions ❌ (secondary)	2.6%	[2.6%, 2.6%]	1
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	1.1%	[0.7%, 2.1%]	4

Cycles

Results (primary 2.3%, secondary 2.5%)

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

	mean	range	count
Regressions ❌ (primary)	2.3%	[2.3%, 2.3%]	1
Regressions ❌ (secondary)	2.5%	[2.2%, 2.8%]	5
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	2.3%	[2.3%, 2.3%]	1

Binary size

Results (primary 1.2%, secondary 1.0%)

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

	mean	range	count
Regressions ❌ (primary)	1.2%	[0.2%, 2.6%]	90
Regressions ❌ (secondary)	1.0%	[0.4%, 2.9%]	71
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	1.2%	[0.2%, 2.6%]	90

Bootstrap: 475.19s -> 473.278s (-0.40%)
Artifact size: 402.86 MiB -> 403.43 MiB (0.14%)

panstromek · 2026-08-31T08:46:22Z

perf triage:

Hi! This looks like it caused a pretty broad perf regression in #162028. This is quite unexpected for a PR that is just a cleanup (I assume), even though it technically makes sense if you look at the detailed results.

Can we reduce the impact of this somehow? This is broad enough that I'd consider a revert, but if we have some idea to reduce the impact quickly in a followup, that'd probably be nicer (it looks like this is pretty bitrotty, looking at the history here). It looks like this is from more type instantiations and downstream codegen queries, but I wonder whether some changes here are more impactul than others.

yoshuawuyts

RE: perf regressions - right now DropGuard doesn't have #[inline] on the Drop::drop impl. This patch did replace a number of those in critical data structures (BTree, VecDeque, Slice). Maybe a first good candidate to try out would be to add #[inline] to the Drop impl and then check whether that affects the perf?

yoshuawuyts · 2026-08-31T10:22:19Z

-        // Run the destructor of the value even if the destructor of the key panics.
-        struct Dropper<'a, T>(&'a mut MaybeUninit<T>);
-        impl<T> Drop for Dropper<'_, T> {
-            #[inline]


For the perf regressions, this one stands out to me

yoshuawuyts · 2026-08-31T10:23:04Z

        // Dropping `guard` handles moving the remaining elements into place.
-        impl<'r, 'a, T, A: Allocator> Drop for DropGuard<'r, 'a, T, A> {
-            #[inline]
-            fn drop(&mut self) {


This one stands out to me as well.

yoshuawuyts · 2026-08-31T10:23:36Z

-                    num_init: usize,
-                }
-                impl<'a, T, A: Allocator> Drop for DropGuard<'a, T, A> {
-                    #[inline]


Third one here.

rustbot assigned clarfonthey Aug 24, 2026

rustbot added O-unix Operating system: Unix-like S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. T-libs Relevant to the library team, which will review and decide on the PR/issue. labels Aug 24, 2026

GrigorenkoPV force-pushed the drop-guard branch from b733c5d to a5a9d5d Compare August 24, 2026 21:55

rust-bors Bot added S-waiting-on-bors Status: Waiting on bors to run and complete tests. Bors will change the label on completion. and removed S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. labels Aug 25, 2026

Zalathar mentioned this pull request Aug 25, 2026

Rollup of 9 pull requests #161715

Closed

rust-bors Bot added S-waiting-on-author Status: This is awaiting some action (such as code changes or more information) from the author. and removed S-waiting-on-bors Status: Waiting on bors to run and complete tests. Bors will change the label on completion. labels Aug 25, 2026

GrigorenkoPV force-pushed the drop-guard branch from a5a9d5d to e990441 Compare August 25, 2026 09:37

rustbot added S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. and removed S-waiting-on-author Status: This is awaiting some action (such as code changes or more information) from the author. labels Aug 25, 2026

This was referenced Aug 25, 2026

Add no-value DropGuard::new(|| ...) #161550

Draft

Stabilize core::mem::DropGuard #161520

Open

This comment has been minimized.

Sign in to view

GrigorenkoPV force-pushed the drop-guard branch from e990441 to f6e2586 Compare August 26, 2026 16:11

This comment has been minimized.

Sign in to view

Use drop_guard in some places in {core,alloc,std}

4f404c1

GrigorenkoPV force-pushed the drop-guard branch from 3546f3a to 4f404c1 Compare August 30, 2026 14:08

rustbot added S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. and removed S-waiting-on-author Status: This is awaiting some action (such as code changes or more information) from the author. labels Aug 30, 2026

This comment has been minimized.

Sign in to view

rust-bors Bot added S-waiting-on-bors Status: Waiting on bors to run and complete tests. Bors will change the label on completion. and removed S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. labels Aug 30, 2026

JonathanBrouwer mentioned this pull request Aug 30, 2026

Rollup of 8 pull requests #162022

Closed

JonathanBrouwer mentioned this pull request Aug 30, 2026

Rollup of 11 pull requests #162024

Closed

This was referenced Aug 30, 2026

Rollup of 11 pull requests #162025

Closed

Rollup of 11 pull requests #162028

Merged

rust-bors Bot merged commit 27e2dc8 into rust-lang:main Aug 31, 2026
14 checks passed

yoshuawuyts reviewed Aug 31, 2026

View reviewed changes

Uh oh!

Uh oh!

Conversation

GrigorenkoPV commented Aug 24, 2026 • edited by rustbot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

rustbot commented Aug 24, 2026

Uh oh!

clarfonthey commented Aug 25, 2026

Uh oh!

rust-bors Bot commented Aug 25, 2026

Uh oh!

JonathanBrouwer commented Aug 25, 2026

Uh oh!

rust-bors Bot commented Aug 25, 2026

Uh oh!

GrigorenkoPV commented Aug 25, 2026

Uh oh!

yoshuawuyts commented Aug 25, 2026

Uh oh!

GrigorenkoPV commented Aug 25, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

This comment has been minimized.

This comment has been minimized.

clarfonthey commented Aug 27, 2026

Uh oh!

This comment has been minimized.

This comment has been minimized.

clarfonthey commented Aug 30, 2026

Uh oh!

rustbot commented Aug 30, 2026

Uh oh!

GrigorenkoPV commented Aug 30, 2026

Uh oh!

clarfonthey commented Aug 30, 2026

Uh oh!

This comment has been minimized.

rust-bors Bot commented Aug 30, 2026

Uh oh!

clarfonthey commented Aug 30, 2026

Uh oh!

rust-bors Bot commented Aug 30, 2026

Uh oh!

Uh oh!

rust-timer commented Aug 31, 2026

Overall result: ❌ regressions - please read:

Instruction count

Max RSS (memory usage)

Cycles

Binary size

Uh oh!

panstromek commented Aug 31, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

yoshuawuyts left a comment

Choose a reason for hiding this comment

Uh oh!

yoshuawuyts Aug 31, 2026

Choose a reason for hiding this comment

Uh oh!

yoshuawuyts Aug 31, 2026

Choose a reason for hiding this comment

Uh oh!

yoshuawuyts Aug 31, 2026

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

10 participants

GrigorenkoPV commented Aug 24, 2026 •

edited by rustbot

Loading

GrigorenkoPV commented Aug 25, 2026 •

edited

Loading

panstromek commented Aug 31, 2026 •

edited

Loading