Add Drop::pin_drop for pinned drops

[frank-king]

View all comments

This PR is part of the pin_ergonomics experiment (the tracking issue is #130494). It allows implementing Drop with a pinned self receiver, which is required for safe pin-projection.

Implementations:

• At least and at most one of drop and pin_drop should be implemented.
• No direct call of drop or pin_drop. They should only be called by the drop glue.
• pin_drop must and must only be used with types that support pin-projection (i.e. types with #[pin_v2]).
• Allows writing fn drop(&pin mut self) and desugars to fn pin_drop(&pin mut self). (Will be in the next PRs)

[frank-king]

The CI failed because Drop::drop becomes a provided method. I'm afraid it might not be a good way to hack into librustdoc. Is there any other good way to fix it?

[rust-bors]

💔 Test for edafd5c failed: CI. Failed job:

• auto - x86_64-gnu-llvm-21-3 (web logs, enhanced plaintext logs)

[petrochenkov]

Broken links in the reference again.
@rustbot author

[ehuss]

I think for fixing the test failure, I think I would prefer to do something like this:

diff --git a/src/ci/docker/scripts/x86_64-gnu-llvm3.sh b/src/ci/docker/scripts/x86_64-gnu-llvm3.sh
index 17eb2cea59a..d2989b3aae6 100755
--- a/src/ci/docker/scripts/x86_64-gnu-llvm3.sh
+++ b/src/ci/docker/scripts/x86_64-gnu-llvm3.sh
@@ -4,7 +4,9 @@ set -ex

 ##### Test stage 1 #####

-../x.py --stage 1 test --skip src/tools/tidy
+# linkchecker is skipped because mixing old rustc/rustdoc with new standard
+# library causes problems with generating correct links.
+../x.py --stage 1 test --skip src/tools/tidy --skip src/tools/linkchecker

 # Run the `mir-opt` tests again but this time for a 32-bit target.
 # This enforces that tests using `// EMIT_MIR_FOR_EACH_BIT_WIDTH` have

Generally I think it is too problematic to try to get the linkchecks working with --stage 1.

cc @jieyouxu @Kobzol in case you want to weigh in on this kind of change.

This was previously discussed at https://rust-lang.zulipchat.com/#narrow/channel/326414-t-infra.2Fbootstrap/topic/Skipping.20stage.201.20tests/with/583642338 for a closely related issue. In this case, I don't think we can employ easy tricks to make this work in the general case.

[jieyouxu]

Generally I think it is too problematic to try to get the linkchecks working with --stage 1.

I would tend to agree. Should we extract the linkchecker change to a separate PR (I can review that bit)? (Though I'm fine with leaving the change in this PR too.)

[frank-king]

I skipped the link check according to @ehuss's suggestion, and let's try it again?

@rustbot ready

[rustbot]

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.

[petrochenkov]

@bors r+

[rust-bors]

📌 Commit 7c656d8 has been approved by petrochenkov

It is now in the queue for this repository.

[rust-log-analyzer]

The job x86_64-gnu-llvm-21-1 failed! Check out the build log: (web) (plain enhanced) (plain)

Click to see the possible cause of the failure (guessed by this bot)

test expr_store::scope::tests::test_resolve_local_name_shadow ... ok
test expr_store::tests::body::array_element_cfg ... ok
test expr_store::scope::tests::while_let_adds_binding ... ok
test expr_store::tests::body::block::inner_item_smoke ... ok
error: test failed, to rerun pass `-p hir-def --lib`

Caused by:
  process didn't exit successfully: `/checkout/obj/build/x86_64-unknown-linux-gnu/stage2-tools/x86_64-unknown-linux-gnu/release/deps/hir_def-c439368f2b64a049 '--skip=tests::smoke_test_real_sysroot_cargo' --skip=check_code_formatting -Z unstable-options --format json` (signal: 11, SIGSEGV: invalid memory reference)
Bootstrap failed while executing `--stage 2 test --skip tests --skip coverage-map --skip coverage-run --skip library --skip tidyselftest`
Build completed unsuccessfully in 1:11:59
  local time: Sat May  9 18:30:26 UTC 2026
  network time: Sat, 09 May 2026 18:30:26 GMT
##[error]Process completed with exit code 1.

[rust-bors]

💔 Test for f56dcfe failed: CI. Failed job:

• auto - x86_64-gnu-llvm-21-1 (web logs, enhanced plaintext logs)

[P8L1]

@petrochenkov
@frank-king

I investigated the x86_64-gnu-llvm-21-1 SIGSEGV from the auto run.

The failed job crashed while running the stage2 rust-analyser hir_def libtest binary:

hir_def ... '--skip=tests::smoke_test_real_sysroot_cargo' --skip=check_code_formatting -Z unstable-options --format json
signal: 11, SIGSEGV

I tried to reproduce this on a temporary branch from the exact failing auto head SHA, focused reproduction run:

https://github.com/P8L1/rust/actions/runs/25625570467/

The hir_def libtest process ran successfully:

running 480 tests
Build completed successfully
exit code: 0

No SIGSEGV reproduced, and I did not make any Rust code changes.

Given that, I don’t think there is enough evidence for a code fix here yet. Could we retry the failed auto/CI job?

[petrochenkov]

@bors retry

[bjorn3]

It is not possible to manually call drop, so maybe it would be possible to only codegen drop or pin_drop depending on which was implemented and then have the drop glue directly call the correct method?

View changes since the review

[bjorn3]

@frank-king ^

[frank-king]

Opened a PR: #156495

[rust-bors]

☀️ Test successful - CI
Approved by: petrochenkov
Duration: 3h 15m 9s
Pushing e8f92f5 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 99eed20 (parent) -> e8f92f5 (this PR)

Test differences

Show 30 test diffs

Stage 0

• basic_broken: pass -> [missing] (J0)
• broken_fragment_local: pass -> [missing] (J0)
• broken_fragment_remote: pass -> [missing] (J0)
• broken_intra_doc_link: pass -> [missing] (J0)
• broken_redir: pass -> [missing] (J0)
• directory_link: pass -> [missing] (J0)
• redirect_loop: pass -> [missing] (J0)
• valid: pass -> [missing] (J0)

Stage 1

• [ui] tests/ui/pin-ergonomics/pinned-drop-check.rs: [missing] -> pass (J0)
• [ui] tests/ui/pin-ergonomics/pinned-drop-unsafety-check.rs: [missing] -> pass (J0)
• [ui] tests/ui/pin-ergonomics/pinned-drop.rs: [missing] -> pass (J0)

Stage 2

• [ui] tests/ui/pin-ergonomics/pinned-drop-check.rs: [missing] -> pass (J1)
• [ui] tests/ui/pin-ergonomics/pinned-drop-unsafety-check.rs: [missing] -> pass (J1)
• [ui] tests/ui/pin-ergonomics/pinned-drop.rs: [missing] -> pass (J1)

Additionally, 16 doctest diffs were found. These are ignored, as they are noisy.

Job group index

• J0: x86_64-gnu-llvm-21-3, x86_64-gnu-llvm-22-3
• J1: aarch64-apple, aarch64-gnu, aarch64-gnu-llvm-21-1, aarch64-msvc-1, arm-android, armhf-gnu, dist-i586-gnu-i586-i686-musl, i686-gnu-1, i686-gnu-nopt-1, i686-msvc-1, optional-x86_64-gnu-parallel-frontend, 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-stable, x86_64-mingw-1, x86_64-msvc-1

Test dashboard

Run

cargo run --manifest-path src/ci/citool/Cargo.toml -- \
    test-dashboard e8f92f576974f7d1d5ef5dfafc56eae303b4c73e --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 40m -> 2h 18m (+38.4%)
2. dist-ohos-armv7: 59m 43s -> 1h 14m (+24.1%)
3. dist-x86_64-musl: 2h 15m -> 1h 43m (-23.5%)
4. dist-arm-linux-musl: 1h 42m -> 1h 20m (-21.7%)
5. dist-i686-msvc: 2h 14m -> 1h 47m (-20.1%)
6. dist-powerpc64le-linux-gnu: 1h 34m -> 1h 16m (-19.0%)
7. x86_64-mingw-1: 2h 41m -> 2h 12m (-18.3%)
8. dist-aarch64-apple: 1h 55m -> 2h 16m (+17.5%)
9. dist-android: 29m 10s -> 32m 18s (+10.7%)
10. x86_64-gnu-llvm-21: 1h 16m -> 1h 24m (+9.9%)

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 (e8f92f5): comparison URL.

Overall result: no relevant changes - no action needed

@rustbot label: -perf-regression

Instruction count

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

Max RSS (memory usage)

Results (primary 1.5%, 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)	1.5%	[1.5%, 1.5%]	1
Regressions ❌ (secondary)	2.3%	[1.0%, 3.6%]	2
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-3.0%	[-3.0%, -3.0%]	1
All ❌✅ (primary)	1.5%	[1.5%, 1.5%]	1

Cycles

Results (secondary 3.4%)

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)	3.4%	[3.4%, 3.4%]	1
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	-	-	0

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
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-0.1%	[-0.1%, -0.0%]	27
Improvements ✅ (secondary)	-0.1%	[-0.1%, -0.0%]	19
All ❌✅ (primary)	-0.1%	[-0.1%, -0.0%]	27

Bootstrap: 499.454s -> 501.011s (0.31%)
Artifact size: 397.23 MiB -> 397.14 MiB (-0.02%)