Add `MaybeDangling` to `core` #149775

WaffleLapkin · 2025-12-08T15:49:00Z

This is the library part of adding MaybeDangling. Note that it doesn't actually do anything described in its docs (yet), I'll make a separate PR for that.

Tracking issue: #118166.

r? libs
cc @RalfJung

rustbot · 2025-12-10T12:07:01Z

The Miri subtree was changed

cc @rust-lang/miri

RalfJung

Thanks for working on this :-)

View changes since this review

library/core/src/mem/maybe_dangling.rs

rustbot · 2025-12-13T13:16:20Z

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.

Mark-Simulacrum

r=me in general with commits squashed, possibly comments addressed (happy to leave specifics up to you)

View changes since this review

Mark-Simulacrum · 2025-12-13T22:33:37Z

library/core/src/mem/maybe_dangling.rs

+///
+/// Even though the `Box`e's destructor is not run (and thus we don't have a double free bug), this
+/// code is still UB. This is because when moving `boxed` into `forget`, its validity invariants
+/// are asserted, causing UB since the `Box` is dangling.


Maybe worth clarifying, e.g., "The safety comment is as such wrong, as moving the boxed variable as part of the call is a use"?

Mark-Simulacrum · 2025-12-13T22:34:50Z

library/core/src/mem/maybe_dangling.rs

+/// **not** dangling -- functions like [`as_ref`] and [`into_inner`] are safe. It is not sound to
+/// return a dangling reference in a `MaybeDangling` to safe code. However, it *is* sound
+/// to hold such values internally inside your code -- and there's no way to do that without
+/// this type.


It sounds like there are alternatives to this type, though, right? Is it worth saying something about ManuallyDrop or some other construct that's already stable?

Hm... yeah fair, some other types will use this type and thus get the same effect; in particular, ManuallyDrop.

Mark-Simulacrum · 2025-12-13T22:37:17Z

library/core/src/mem/maybe_dangling.rs

+/// (and [boxes]) still must be aligned and non-null.
+///
+/// Additionally note that safe code can still assume that the inner value in a `MaybeDangling` is
+/// **not** dangling -- functions like [`as_ref`] and [`into_inner`] are safe. It is not sound to


How do we deal with the implicit Drop for the inner type? Should callers who place e.g. a dangling Box inside just be careful to call forget rather than letting it Drop?

Mark-Simulacrum · 2025-12-13T22:38:25Z

library/core/src/mem/maybe_dangling.rs

+
+use crate::{mem, ptr};
+
+/// Allows wrapped [references] and [boxes] to dangle.


Can we add a "Not yet actually implemented" or perhaps make maybe_dangling an incomplete lang feature? I worry a bit that as of this PR, from what I can tell, this is UB to use as documented, right? Since the compiler bits aren't there yet.

Co-authored-by: Ralf Jung <post@ralfj.de>

WaffleLapkin · 2025-12-17T21:29:37Z

I think I've addressed all the review comments.
@bors r=Mark-Simulacrum

bors · 2025-12-17T21:29:40Z

📌 Commit 25476d8 has been approved by Mark-Simulacrum

It is now in the queue for this repository.

…Simulacrum Add `MaybeDangling` to `core` This is the library part of adding `MaybeDangling`. Note that it doesn't actually do anything described in its docs (yet), I'll make a separate PR for that. Tracking issue: #118166. r? libs cc `@RalfJung`

bors · 2025-12-17T21:51:53Z

⌛ Testing commit 25476d8 with merge 3d0287a...

bors · 2025-12-17T23:30:25Z

💔 Test failed - checks-actions

- fixup `BTreeMap` gdb provider - fixup `ManuallyDrop` natvis thingy Now that `MaybeUninit` contains `ManuallyDrop` which contains `MaybeDangling` (tbc this is the addition), we need to unwrap one more layer.

WaffleLapkin · 2025-12-26T21:05:07Z

@bors try jobs=aarch64-msvc-1

Add `MaybeDangling` to `core` try-job: aarch64-msvc-1

rust-bors · 2025-12-26T23:19:00Z

☀️ Try build successful (CI)
Build commit: 553d580 (553d5804c0571418a8c80ae1c8f3449274bc3f7d, parent: 82dd3cb008233bfe50ba6b8d6618e6bbd6054eb1)

WaffleLapkin · 2025-12-27T09:44:08Z

@bors r=Mark-Simulacrum

bors · 2025-12-27T09:44:11Z

📌 Commit 6b4f4f5 has been approved by Mark-Simulacrum

It is now in the queue for this repository.

bors · 2025-12-27T09:45:24Z

⌛ Testing commit 6b4f4f5 with merge 0850949...

bors · 2025-12-27T12:57:52Z

☀️ Test successful - checks-actions
Approved by: Mark-Simulacrum
Pushing 0850949 to main...

github-actions · 2025-12-27T13:00:39Z

What is this?

This is an experimental post-merge analysis report that shows differences in test outcomes between the merged PR and its parent PR.

Comparing c7aa99f (parent) -> 0850949 (this PR)

Test differences

Show 184 test diffs

184 doctest diffs were found. These are ignored, as they are noisy.

Test dashboard

Run

cargo run --manifest-path src/ci/citool/Cargo.toml -- \
    test-dashboard 08509492139968a96a005ba811a995e2f1d6a2ac --output-dir test-dashboard

And then open test-dashboard/index.html in your browser to see an overview of all executed tests.

Job duration changes

x86_64-gnu-debug: 7263.3s -> 6500.0s (-10.5%)
dist-apple-various: 3957.2s -> 3571.3s (-9.8%)
x86_64-rust-for-linux: 2913.3s -> 2651.7s (-9.0%)
x86_64-msvc-1: 8297.4s -> 9020.0s (+8.7%)
x86_64-gnu: 8208.5s -> 7543.4s (-8.1%)
x86_64-gnu-gcc: 3165.6s -> 3414.0s (+7.8%)
aarch64-apple: 10290.9s -> 9513.2s (-7.6%)
dist-ohos-aarch64: 4527.8s -> 4213.2s (-6.9%)
aarch64-msvc-2: 6400.5s -> 5977.0s (-6.6%)
x86_64-gnu-distcheck: 7868.9s -> 7361.0s (-6.5%)

How to interpret the job duration changes?

Job durations can vary a lot, based on the actual runner instance
that executed the job, system noise, invalidated caches, etc. The table above is provided
mostly for t-infra members, for simpler debugging of potential CI slow-downs.

rust-timer · 2025-12-27T13:39:32Z

Finished benchmarking commit (0850949): comparison URL.

Overall result: ❌ regressions - 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

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

	mean	range	count
Regressions ❌ (primary)	0.3%	[0.1%, 0.6%]	9
Regressions ❌ (secondary)	0.2%	[0.1%, 0.5%]	7
Improvements ✅ (primary)	-0.3%	[-0.3%, -0.3%]	1
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	0.2%	[-0.3%, 0.6%]	10

Max RSS (memory usage)

Results (primary 0.9%, secondary -0.5%)

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

	mean	range	count
Regressions ❌ (primary)	4.8%	[2.8%, 6.8%]	2
Regressions ❌ (secondary)	3.0%	[2.2%, 3.7%]	2
Improvements ✅ (primary)	-1.7%	[-2.2%, -0.9%]	3
Improvements ✅ (secondary)	-2.3%	[-4.1%, -1.4%]	4
All ❌✅ (primary)	0.9%	[-2.2%, 6.8%]	5

Cycles

This benchmark run did not return any relevant results for this metric.

Binary size

Results (primary 0.6%, 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)	0.8%	[0.0%, 1.2%]	8
Regressions ❌ (secondary)	0.0%	[0.0%, 0.0%]	1
Improvements ✅ (primary)	-1.1%	[-1.1%, -1.1%]	1
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	0.6%	[-1.1%, 1.2%]	9

Bootstrap: 480.868s -> 482.408s (0.32%)
Artifact size: 390.80 MiB -> 390.78 MiB (-0.00%)

RalfJung · 2025-12-28T12:09:21Z

This seems to have caused Miri regressions: rust-lang/miri#4793

panstromek · 2025-12-29T14:06:37Z

perf triage:

Are regressions expected here? Some of those could be noise, it seems like the previous PR had more positive noise (post merge results look better than pre-merge), so this could be just return to normal. The incremental ones look real, though, they have some new deltas in detailed results, for example here for cargo debug.

rustbot assigned Mark-Simulacrum Dec 8, 2025

WaffleLapkin marked this pull request as ready for review December 8, 2025 15:49

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 Dec 8, 2025

This comment has been minimized.

Sign in to view

RalfJung reviewed Dec 13, 2025

View reviewed changes

library/core/src/mem/maybe_dangling.rs Outdated Show resolved Hide resolved

library/core/src/mem/maybe_dangling.rs Outdated Show resolved Hide resolved

This comment has been minimized.

Sign in to view

This comment was marked as resolved.

Sign in to view

rustbot added has-merge-commits PR has merge commits, merge with caution. S-waiting-on-author Status: This is awaiting some action (such as code changes or more information) from the author. labels Dec 13, 2025

WaffleLapkin force-pushed the core-mem-maybe-dangling branch from c7db108 to 95eee60 Compare December 13, 2025 13:16

rustbot removed S-waiting-on-author Status: This is awaiting some action (such as code changes or more information) from the author. has-merge-commits PR has merge commits, merge with caution. labels Dec 13, 2025

Mark-Simulacrum reviewed Dec 13, 2025

View reviewed changes

WaffleLapkin and others added 2 commits December 17, 2025 22:26

add MaybeDangling to core

4ea5cac

Co-authored-by: Ralf Jung <post@ralfj.de>

use MaybeDangling in ManuallyDrop

722e310

WaffleLapkin force-pushed the core-mem-maybe-dangling branch from 95eee60 to 25476d8 Compare December 17, 2025 21:28

bors 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 Dec 17, 2025

This comment has been minimized.

Sign in to view

bors added S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. and removed S-waiting-on-bors Status: Waiting on bors to run and complete tests. Bors will change the label on completion. labels Dec 17, 2025

WaffleLapkin added 2 commits December 26, 2025 22:02

fixup debugger files wrt to MaybeDangling

d42f7ea

- fixup `BTreeMap` gdb provider - fixup `ManuallyDrop` natvis thingy Now that `MaybeUninit` contains `ManuallyDrop` which contains `MaybeDangling` (tbc this is the addition), we need to unwrap one more layer.

bless a mir-opt and a miri tests

6b4f4f5

WaffleLapkin force-pushed the core-mem-maybe-dangling branch from 25476d8 to 6b4f4f5 Compare December 26, 2025 21:03

This comment has been minimized.

Sign in to view

rust-bors bot added a commit that referenced this pull request Dec 26, 2025

Auto merge of #149775 - WaffleLapkin:core-mem-maybe-dangling, r=<try>

553d580

Add `MaybeDangling` to `core` try-job: aarch64-msvc-1

bors 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 Dec 27, 2025

bors added the merged-by-bors This PR was explicitly merged by bors. label Dec 27, 2025

bors merged commit 0850949 into rust-lang:main Dec 27, 2025
13 checks passed

rustbot added this to the 1.94.0 milestone Dec 27, 2025

rustbot added the perf-regression Performance regression. label Dec 27, 2025

This was referenced Dec 27, 2025

[TOY] Run MIR opts before coroutine state transform #145330

Draft

Ignore #[doc(hidden)] items when computing trimmed paths for printing #148623

Open

ptr_aligment_type: add more APIs #148764

Draft

Stable inline const using unstable intrinsics #149817

Open

WaffleLapkin deleted the core-mem-maybe-dangling branch December 27, 2025 20:45

dtolnay mentioned this pull request Dec 28, 2025

MaybeDangling, catch_unwind rust-lang/miri#4793

Closed


		use crate::{mem, ptr};

		/// Allows wrapped [references] and [boxes] to dangle.

Uh oh!

Add MaybeDangling to core #149775

Add MaybeDangling to core #149775

Uh oh!

Conversation

WaffleLapkin commented Dec 8, 2025

Uh oh!

This comment has been minimized.

rustbot commented Dec 10, 2025

Uh oh!

RalfJung left a comment • edited by rustbot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

This comment has been minimized.

This comment was marked as resolved.

rustbot commented Dec 13, 2025

Uh oh!

Mark-Simulacrum left a comment • edited by rustbot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

WaffleLapkin commented Dec 17, 2025

Uh oh!

bors commented Dec 17, 2025

Uh oh!

bors commented Dec 17, 2025

Uh oh!

This comment has been minimized.

bors commented Dec 17, 2025

Uh oh!

WaffleLapkin commented Dec 26, 2025

Uh oh!

This comment has been minimized.

rust-bors bot commented Dec 26, 2025

Uh oh!

WaffleLapkin commented Dec 27, 2025

Uh oh!

bors commented Dec 27, 2025

Uh oh!

bors commented Dec 27, 2025

Uh oh!

bors commented Dec 27, 2025

Uh oh!

Uh oh!

github-actions bot commented Dec 27, 2025

Test differences

Job duration changes

Uh oh!

rust-timer commented Dec 27, 2025

Overall result: ❌ regressions - please read the text below

Uh oh!

RalfJung commented Dec 28, 2025

Uh oh!

panstromek commented Dec 29, 2025

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

8 participants

Add `MaybeDangling` to `core` #149775

Add `MaybeDangling` to `core` #149775

RalfJung left a comment •

edited by rustbot

Loading

Mark-Simulacrum left a comment •

edited by rustbot

Loading