stabilize never type

[WaffleLapkin]

View all comments

This PR

• stabilizes the never type (aka !) (!!!)
• sets the never type fallback to ! on all editions (breaking change, see crater report analysis and refresher on never type fallback)
• makes Infallible an alias to !
• removes dependency_on_unit_never_type_fallback lint (there is no more never type fallback to () so this lint can't be triggered)

Cat

Tracking

• Tracking issue for promoting ! to a type (RFC 1216) #35121
• Tracking Issue for never type fallback change #148922

FCPs

• Edition 2024: Make ! fall back to ! #123508 (comment) (stabilization plan, T-lang)
• stabilize never type #155499 (comment) (never type stabilization modulo bugs, T-lang)
• Make Infallible = ! #155924 (comment) (make Infallible = !, T-libs)

Related changes

• Lint bump: Report never type lints in dependencies #141937
• Rust 2024 edition change:
  • Tracking Issue for Rust 2024: Make ! fall back to ! #123748
  • Edition 2024: Make ! fall back to ! #123508
• Various changes to lessen the effect of the breaking changes:
  • consider subtyping when checking if an infer var is sized #157820
  • Fix trait method resolution on an adjusted never type #156047
  • Remove From<!> for T *reservation* impl #160705
• Never type documentation changes: rewrite never type documentation #158370 (blocked on this pr)

Experiments

• Fallback change and never type stabilization:
  • Run: stabilize never type #155499 (comment)
  • Analysis: stabilize never type #155499 (comment)
  • Status of backports: stabilize never type #155499 (comment)
• Fallback change, never type stabilization and making Infallible = !: [experiment][1] mew #155500
• Fallback change, never type stabilization, making Infallible = !, and removing reservation impl: [experiment][2] meow mrrrumph #155501
• Never type stabilization and Infallible = ! change, without fallback change: [experiment][3] meowww #155657

---

Closes #35121
Closes #148922

[WaffleLapkin]

@bors try

[rust-bors]

☀️ Try build successful (CI)
Build commit: 1c4199d (1c4199da70f816f2239f41c55168ff75a3663718, parent: b2f1ccf524a3a4cf9c34545167cc23b659cf1cbd)

[WaffleLapkin]

@craterbot check p=1 crates=https://crater-reports.s3.amazonaws.com/pr-154887/retry-regressed-list.txt

[craterbot]

👌 Experiment pr-155499 created and queued.
🤖 Automatically detected try build 1c4199d
⚠️ Try build based on commit 2895d33, but latest commit is 3e8df8d. Did you forget to make a new try build?
🔍 You can check out the queue and this experiment's details.

ℹ️ Crater is a tool to run experiments across parts of the Rust ecosystem. Learn more

[craterbot]

🚧 Experiment pr-155499 is now running

ℹ️ Crater is a tool to run experiments across parts of the Rust ecosystem. Learn more

[craterbot]

🎉 Experiment pr-155499 is completed!
📊 3277 regressed and 0 fixed (9024 total)
📊 1149 spurious results on the retry-regressed-list.txt, consider a retry¹ if this is a significant amount.
📰 Open the summary report.

⚠️ If you notice any spurious failure please add them to the denylist!
ℹ️ Crater is a tool to run experiments across parts of the Rust ecosystem. Learn more

Footnotes

1. re-run the experiment with crates=https://crater-reports.s3.amazonaws.com/pr-155499/retry-regressed-list.txt ↩

[rust-bors]

📋 This PR cannot be approved because it currently has the following labels: needs-fcp, needs-reference-pr.

[JonathanBrouwer]

Way to ruin a dramatic moment bors /j
We should also merge the reference pr rust-lang/reference#2283
@bors r+

[rust-bors]

📌 Commit b74b72b has been approved by JonathanBrouwer

It is now in the queue for this repository.

[JonathanBrouwer]

@bors rollup=iffy causes breakage

[JonathanBrouwer]

💔 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: #161685 (comment)

[rust-bors]

This pull request was unapproved.

Auto build was cancelled due to unapproval. Cancelled workflows:

• https://github.com/rust-lang/rust/actions/runs/32761245655

View changes since this unapproval

[JonathanBrouwer]

Description:
A publicly-visible enum cannot be imported by its prior path. A pub use may have been removed, or the enum itself > may have been renamed or removed entirely.
ref: https://doc.rust-lang.org/cargo/reference/semver.html#item-remove
impl: https://github.com/obi1kenobi/cargo-semver-checks/tree/v0.50.0/src/lints/enum_missing.ron

Failed in:
enum core::convert::Infallible, previously in file library/core/src/convert/mod.rs:920

This is a false positive in cargo-semver-checks (cc @obi1kenobi)
The type is still there, it just turned from a pub enum into a pub type
@bors r+

[rust-bors]

📌 Commit 0a35c2a has been approved by JonathanBrouwer

It is now in the queue for this repository.

[asquared31415]

I believe pub enum -> not enum is a breaking change, but it's an intended breakage.

[scottmcm]

For an enum with variants I think it is, but in this weird case where it doesn't have any variants it might not be?

Regardless, it's something we've been working towards for years so this instance we'll let through for sure.

[rust-bors]

☀️ Test successful - CI
Approved by: JonathanBrouwer
Duration: 3h 7m 55s
Pushing e776960 to main...

[github-actions]

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 0a3fa2a (parent) -> e776960 (this PR)

Test differences

Show 1825 test diffs

Stage 1

• [ui] tests/ui/coercion/coerce-issue-49593-box-never.rs: [missing] -> pass (J1)
• [ui] tests/ui/coercion/coerce-issue-49593-box-never.rs#e2021: pass -> [missing] (J1)
• [ui] tests/ui/coercion/coerce-issue-49593-box-never.rs#e2024: pass -> [missing] (J1)
• [ui] tests/ui/feature-gates/feature-gate-never_type.rs: pass -> [missing] (J1)
• [ui] tests/ui/never_type/fallback_change/defaulted-never-note.rs: [missing] -> pass (J1)
• [ui] tests/ui/never_type/fallback_change/defaulted-never-note.rs#e2021: pass -> [missing] (J1)
• [ui] tests/ui/never_type/fallback_change/defaulted-never-note.rs#e2024: pass -> [missing] (J1)
• [ui] tests/ui/never_type/fallback_change/diverging-fallback-unconstrained-return.rs: [missing] -> pass (J1)
• [ui] tests/ui/never_type/fallback_change/diverging-fallback-unconstrained-return.rs#e2021: pass -> [missing] (J1)
• [ui] tests/ui/never_type/fallback_change/diverging-fallback-unconstrained-return.rs#e2024: pass -> [missing] (J1)
• [ui] tests/ui/never_type/fallback_change/dont-suggest-turbofish-from-expansion.rs: pass -> [missing] (J1)
• [ui] tests/ui/never_type/fallback_change/fallback-closure-ret.rs: [missing] -> pass (J1)
• [ui] tests/ui/never_type/fallback_change/fallback-closure-ret.rs#e2021: pass -> [missing] (J1)
• [ui] tests/ui/never_type/fallback_change/fallback-closure-ret.rs#e2024: pass -> [missing] (J1)
• [ui] tests/ui/never_type/fallback_change/fallback-closure-wrap.rs: [missing] -> pass (J1)
• [ui] tests/ui/never_type/fallback_change/fallback-closure-wrap.rs#e2021: pass -> [missing] (J1)
• [ui] tests/ui/never_type/fallback_change/fallback-closure-wrap.rs#e2024: pass -> [missing] (J1)
• [ui] tests/ui/never_type/fallback_change/lint-never-type-fallback-flowing-into-unsafe.rs: [missing] -> pass (J1)
• [ui] tests/ui/never_type/fallback_change/lint-never-type-fallback-flowing-into-unsafe.rs#e2015: pass -> [missing] (J1)
• [ui] tests/ui/never_type/fallback_change/lint-never-type-fallback-flowing-into-unsafe.rs#e2024: pass -> [missing] (J1)
• [ui] tests/ui/never_type/fallback_change/try-block-never-type-fallback.rs: [missing] -> pass (J1)
• [ui] tests/ui/never_type/fallback_change/try-block-never-type-fallback.rs#e2021: pass -> [missing] (J1)
• [ui] tests/ui/never_type/fallback_change/try-block-never-type-fallback.rs#e2024: pass -> [missing] (J1)
• [ui (polonius)] tests/ui/coercion/coerce-issue-49593-box-never.rs: [missing] -> pass (J2)
• [ui (polonius)] tests/ui/coercion/coerce-issue-49593-box-never.rs#e2021: pass -> [missing] (J2)
• [ui (polonius)] tests/ui/coercion/coerce-issue-49593-box-never.rs#e2024: pass -> [missing] (J2)
• [ui (polonius)] tests/ui/feature-gates/feature-gate-never_type.rs: pass -> [missing] (J2)
• [ui (polonius)] tests/ui/never_type/fallback_change/defaulted-never-note.rs: [missing] -> pass (J2)
• [ui (polonius)] tests/ui/never_type/fallback_change/defaulted-never-note.rs#e2021: pass -> [missing] (J2)
• [ui (polonius)] tests/ui/never_type/fallback_change/defaulted-never-note.rs#e2024: pass -> [missing] (J2)
• [ui (polonius)] tests/ui/never_type/fallback_change/diverging-fallback-unconstrained-return.rs: [missing] -> pass (J2)
• [ui (polonius)] tests/ui/never_type/fallback_change/diverging-fallback-unconstrained-return.rs#e2021: pass -> [missing] (J2)
• [ui (polonius)] tests/ui/never_type/fallback_change/diverging-fallback-unconstrained-return.rs#e2024: pass -> [missing] (J2)
• [ui (polonius)] tests/ui/never_type/fallback_change/dont-suggest-turbofish-from-expansion.rs: pass -> [missing] (J2)
• [ui (polonius)] tests/ui/never_type/fallback_change/fallback-closure-ret.rs: [missing] -> pass (J2)
• [ui (polonius)] tests/ui/never_type/fallback_change/fallback-closure-ret.rs#e2021: pass -> [missing] (J2)
• [ui (polonius)] tests/ui/never_type/fallback_change/fallback-closure-ret.rs#e2024: pass -> [missing] (J2)
• [ui (polonius)] tests/ui/never_type/fallback_change/fallback-closure-wrap.rs: [missing] -> pass (J2)
• [ui (polonius)] tests/ui/never_type/fallback_change/fallback-closure-wrap.rs#e2021: pass -> [missing] (J2)
• [ui (polonius)] tests/ui/never_type/fallback_change/fallback-closure-wrap.rs#e2024: pass -> [missing] (J2)
• [ui (polonius)] tests/ui/never_type/fallback_change/lint-never-type-fallback-flowing-into-unsafe.rs: [missing] -> pass (J2)
• [ui (polonius)] tests/ui/never_type/fallback_change/lint-never-type-fallback-flowing-into-unsafe.rs#e2015: pass -> [missing] (J2)
• [ui (polonius)] tests/ui/never_type/fallback_change/lint-never-type-fallback-flowing-into-unsafe.rs#e2024: pass -> [missing] (J2)
• [ui (polonius)] tests/ui/never_type/fallback_change/try-block-never-type-fallback.rs: [missing] -> pass (J2)
• [ui (polonius)] tests/ui/never_type/fallback_change/try-block-never-type-fallback.rs#e2021: pass -> [missing] (J2)
• [ui (polonius)] tests/ui/never_type/fallback_change/try-block-never-type-fallback.rs#e2024: pass -> [missing] (J2)

Stage 2

• [ui] tests/ui/coercion/coerce-issue-49593-box-never.rs: [missing] -> pass (J0)
• [ui] tests/ui/coercion/coerce-issue-49593-box-never.rs#e2021: pass -> [missing] (J0)
• [ui] tests/ui/coercion/coerce-issue-49593-box-never.rs#e2024: pass -> [missing] (J0)
• [ui] tests/ui/feature-gates/feature-gate-never_type.rs: pass -> [missing] (J0)
• [ui] tests/ui/never_type/fallback_change/defaulted-never-note.rs: [missing] -> pass (J0)
• [ui] tests/ui/never_type/fallback_change/defaulted-never-note.rs#e2021: pass -> [missing] (J0)
• [ui] tests/ui/never_type/fallback_change/defaulted-never-note.rs#e2024: pass -> [missing] (J0)
• [ui] tests/ui/never_type/fallback_change/diverging-fallback-unconstrained-return.rs: [missing] -> pass (J0)
• [ui] tests/ui/never_type/fallback_change/diverging-fallback-unconstrained-return.rs#e2021: pass -> [missing] (J0)
• [ui] tests/ui/never_type/fallback_change/diverging-fallback-unconstrained-return.rs#e2024: pass -> [missing] (J0)
• [ui] tests/ui/never_type/fallback_change/dont-suggest-turbofish-from-expansion.rs: pass -> [missing] (J0)
• [ui] tests/ui/never_type/fallback_change/fallback-closure-ret.rs: [missing] -> pass (J0)
• [ui] tests/ui/never_type/fallback_change/fallback-closure-ret.rs#e2021: pass -> [missing] (J0)
• [ui] tests/ui/never_type/fallback_change/fallback-closure-ret.rs#e2024: pass -> [missing] (J0)
• [ui] tests/ui/never_type/fallback_change/fallback-closure-wrap.rs: [missing] -> pass (J0)
• [ui] tests/ui/never_type/fallback_change/fallback-closure-wrap.rs#e2021: pass -> [missing] (J0)
• [ui] tests/ui/never_type/fallback_change/fallback-closure-wrap.rs#e2024: pass -> [missing] (J0)
• [ui] tests/ui/never_type/fallback_change/lint-never-type-fallback-flowing-into-unsafe.rs: [missing] -> pass (J0)
• [ui] tests/ui/never_type/fallback_change/lint-never-type-fallback-flowing-into-unsafe.rs#e2015: pass -> [missing] (J0)
• [ui] tests/ui/never_type/fallback_change/lint-never-type-fallback-flowing-into-unsafe.rs#e2024: pass -> [missing] (J0)
• [ui] tests/ui/never_type/fallback_change/try-block-never-type-fallback.rs: [missing] -> pass (J0)
• [ui] tests/ui/never_type/fallback_change/try-block-never-type-fallback.rs#e2021: pass -> [missing] (J0)
• [ui] tests/ui/never_type/fallback_change/try-block-never-type-fallback.rs#e2024: pass -> [missing] (J0)

Additionally, 1756 doctest diffs were found. These are ignored, as they are noisy.

Job group index

• J0: aarch64-apple-1, aarch64-apple-macos-26-1, aarch64-gnu, aarch64-gnu-llvm-21-1, aarch64-msvc-1, arm-android, armhf-gnu, i686-gnu-1, i686-gnu-nopt-1, i686-msvc-1, test-various, x86_64-gnu, x86_64-gnu-debug, x86_64-gnu-gcc, x86_64-gnu-llvm-21, x86_64-gnu-llvm-21-2, x86_64-gnu-llvm-22-2, x86_64-gnu-nopt, x86_64-gnu-parallel-frontend, x86_64-gnu-stable, x86_64-mingw-1, x86_64-msvc-1
• J1: x86_64-gnu-llvm-21-3, x86_64-gnu-llvm-22-3, x86_64-gnu-next-trait-solver-polonius
• J2: x86_64-gnu-next-trait-solver-polonius

Test dashboard

Run

cargo run --manifest-path src/ci/citool/Cargo.toml -- \
    test-dashboard e7769602aca3770e8d8ea55716becb22e839a579 --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

1. x86_64-msvc-ext1: 1h 27m -> 2h 9m (+47.7%)
2. x86_64-msvc-2: 1h 54m -> 2h 40m (+39.7%)
3. dist-i686-mingw: 2h 50m -> 1h 49m (-35.9%)
4. x86_64-gnu-llvm-21-2: 1h 31m -> 1h (-33.7%)
5. dist-x86_64-llvm-mingw: 1h 40m -> 2h 11m (+30.8%)
6. dist-ohos-armv7: 58m 7s -> 1h 15m (+30.6%)
7. dist-x86_64-illumos: 1h 25m -> 1h 51m (+29.9%)
8. dist-various-2: 39m 6s -> 49m 25s (+26.4%)
9. dist-x86_64-msvc: 3h 9m -> 2h 22m (-25.0%)
10. x86_64-gnu-next-trait-solver-polonius: 57m 10s -> 43m 2s (-24.7%)

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]

Finished benchmarking commit (e776960): comparison URL.

Overall result: ❌✅ regressions and improvements - 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
Regressions ❌ (secondary)	0.1%	[0.1%, 0.1%]	2
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-0.3%	[-0.4%, -0.1%]	10
All ❌✅ (primary)	-	-	0

Max RSS (memory usage)

Results (primary -3.9%, secondary -1.8%)

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)	-	-	0
Improvements ✅ (primary)	-3.9%	[-3.9%, -3.9%]	1
Improvements ✅ (secondary)	-1.8%	[-2.6%, -1.0%]	2
All ❌✅ (primary)	-3.9%	[-3.9%, -3.9%]	1

Cycles

Results (primary -1.2%, secondary -4.5%)

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

	mean	range	count
Regressions ❌ (primary)	3.7%	[3.7%, 3.7%]	1
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-2.4%	[-2.7%, -2.3%]	4
Improvements ✅ (secondary)	-4.5%	[-5.9%, -2.4%]	3
All ❌✅ (primary)	-1.2%	[-2.7%, 3.7%]	5

Binary size

Results (primary -0.1%, secondary -0.1%)

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

	mean	range	count
Regressions ❌ (primary)	0.1%	[0.1%, 0.1%]	3
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-0.1%	[-0.2%, -0.0%]	55
Improvements ✅ (secondary)	-0.1%	[-0.2%, -0.0%]	36
All ❌✅ (primary)	-0.1%	[-0.2%, 0.1%]	58

Bootstrap: 469.763s -> 471.419s (0.35%)
Artifact size: 400.31 MiB -> 400.34 MiB (0.01%)

[theemathas]

@JonathanBrouwer Not a false positive. I think that use std::convert::Infallible::{self}; will stop compiling after the change.

[obi1kenobi]

I think we're all on the same page about the (possible) breakage being fine for this PR.

Now purely for our collective curiosity ...

@JonathanBrouwer Not a false positive. I think that use std::convert::Infallible::{self}; will stop compiling after the change.

Even if that's the case, does this fall under the "useless code" exception? As in, if that and only that is the only breakage, is that truly SemVer-major given that that code by itself has no purpose?

For an enum with variants I think it is, but in this weird case where it doesn't have any variants it might not be?

This was also my first thought. But I admittedly haven't thought enough about the zero-variant case.

If we need to add a "has at least one variant" clause, I think that should be easy enough. Type aliases are also a bit of a nightmare for SemVer though, and I definitely haven't even worked out the full set of edge cases yet, let alone implemented all of them in c-s-c.

[obi1kenobi]

Also — 🎉🎉🎉 Congrats! So glad this is finally merged, and in such a nice round-numbered release too :)