Revert "Make `lto` and `linker-plugin-lto` work the same for `compiler_builtins` #146133

rcvalle · 2025-09-02T18:25:52Z

This reverts commit cf8753e (PR #145368) and fix the regressions reported at #145981, #146109, and #146145.

rustbot · 2025-09-02T18:25:56Z

rustbot has assigned @davidtwco.
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

rustbot · 2025-09-02T18:26:00Z

Some changes occurred in compiler/rustc_codegen_ssa

cc @WaffleLapkin

rcvalle · 2025-09-02T18:26:10Z

r? @bjorn3

bjorn3 · 2025-09-02T19:48:07Z

Can you please remove the issue references from the commit message. r=me with the commit message changed.

…r_builtins`" This reverts commit cf8753e and fixes the regressions reported.

rcvalle · 2025-09-03T06:21:04Z

r? @bjorn3

rustbot · 2025-09-03T06:21:07Z

Requested reviewer is already assigned to this pull request.

Please choose another assignee.

nikic · 2025-09-03T07:58:17Z

@bors r=bjorn3 rollup

bors · 2025-09-03T07:58:21Z

📌 Commit 916b55e has been approved by bjorn3

It is now in the queue for this repository.

nikic · 2025-09-03T07:58:55Z

@bors p=1 (multiple reported regressions)

Revert "Make `lto` and `linker-plugin-lto` work the same for `compiler_builtins` This reverts commit cf8753e and fix the regressions reported at rust-lang#145981 and rust-lang#146109.

Rollup of 15 pull requests Successful merges: - #143725 (core: add Peekable::next_if_map) - #145209 (Stabilize `path_add_extension`) - #145750 (raw_vec.rs: Remove superfluous fn alloc_guard) - #145962 (Ensure we emit an allocator shim when only some crate types need one) - #145963 (Add LSX accelerated implementation for source file analysis) - #146054 (add `#[must_use]` to `array::repeat`) - #146090 (Derive `PartialEq` for `InvisibleOrigin`) - #146120 (Correct typo in `rustc_errors` comment) - #146127 (Rename `ToolRustc` to `ToolRustcPrivate`) - #146131 (rustdoc-search: add test case for indexing every item type) - #146133 (Revert "Make `lto` and `linker-plugin-lto` work the same for `compiler_builtins`) - #146134 (llvm: nvptx: Layout update to match LLVM) - #146136 (docs(std): add missing closing code block fences in doc comments) - #146137 (Disallow frontmatter in `--cfg` and `--check-cfg` arguments) - #146140 (compiletest: cygwin follows windows in using PATH for dynamic libraries) r? `@ghost` `@rustbot` modify labels: rollup

RalfJung · 2025-09-03T12:11:38Z

Please also reference the PR when filing a revert, so there's a backlink in the original PR and one can see the revert there. (I did that now for this PR.)

Revert "Make `lto` and `linker-plugin-lto` work the same for `compiler_builtins` This reverts commit cf8753e (PR rust-lang#145368) and fix the regressions reported at rust-lang#145981 and rust-lang#146109.

Rollup of 16 pull requests Successful merges: - #143725 (core: add Peekable::next_if_map) - #145209 (Stabilize `path_add_extension`) - #145342 (fix drop scope for `super let` bindings within `if let`) - #145750 (raw_vec.rs: Remove superfluous fn alloc_guard) - #145962 (Ensure we emit an allocator shim when only some crate types need one) - #145963 (Add LSX accelerated implementation for source file analysis) - #146054 (add `#[must_use]` to `array::repeat`) - #146090 (Derive `PartialEq` for `InvisibleOrigin`) - #146120 (Correct typo in `rustc_errors` comment) - #146127 (Rename `ToolRustc` to `ToolRustcPrivate`) - #146133 (Revert "Make `lto` and `linker-plugin-lto` work the same for `compiler_builtins`) - #146134 (llvm: nvptx: Layout update to match LLVM) - #146136 (docs(std): add missing closing code block fences in doc comments) - #146137 (Disallow frontmatter in `--cfg` and `--check-cfg` arguments) - #146140 (compiletest: cygwin follows windows in using PATH for dynamic libraries) - #146156 (miri subtree update) r? `@ghost` `@rustbot` modify labels: rollup

bors · 2025-09-03T19:07:58Z

⌛ Testing commit 916b55e with merge a1208bf...

bors · 2025-09-03T22:15:51Z

☀️ Test successful - checks-actions
Approved by: bjorn3
Pushing a1208bf to master...

github-actions · 2025-09-03T22:19:07Z

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 fd75a9c (parent) -> a1208bf (this PR)

Test differences

Show 6 test diffs

Stage 1

[ui] tests/ui/sanitizer/cfi/no_builtins.rs: pass -> [missing] (J1)

Stage 2

[ui] tests/ui/sanitizer/cfi/no_builtins.rs: pass -> [missing] (J0)
[ui] tests/ui/sanitizer/cfi/no_builtins.rs: ignore (ignored on targets without CFI sanitizer) -> [missing] (J2)

Additionally, 3 doctest diffs were found. These are ignored, as they are noisy.

Job group index

Test dashboard

Run

cargo run --manifest-path src/ci/citool/Cargo.toml -- \
    test-dashboard a1208bf765ba783ee4ebdc4c29ab0a0c215806ef --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

dist-apple-various: 4386.1s -> 3149.7s (-28.2%)
dist-arm-linux-musl: 6060.9s -> 5568.7s (-8.1%)
dist-various-1: 4238.7s -> 3924.3s (-7.4%)
dist-x86_64-netbsd: 5095.5s -> 4723.3s (-7.3%)
dist-aarch64-msvc: 5654.9s -> 6032.1s (6.7%)
dist-various-2: 2321.0s -> 2171.9s (-6.4%)
dist-android: 1591.1s -> 1497.0s (-5.9%)
pr-check-2: 2179.8s -> 2306.7s (5.8%)
tidy: 191.7s -> 181.8s (-5.1%)
dist-aarch64-apple: 7391.3s -> 7762.7s (5.0%)

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 · 2025-09-03T23:26:20Z

Finished benchmarking commit (a1208bf): comparison URL.

Overall result: no relevant changes - no action needed

@rustbot label: -perf-regression

Instruction count

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

Max RSS (memory usage)

Results (primary 10.6%, secondary -1.4%)

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

	mean	range	count
Regressions ❌ (primary)	10.6%	[9.6%, 11.7%]	2
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-1.4%	[-1.4%, -1.4%]	1
All ❌✅ (primary)	10.6%	[9.6%, 11.7%]	2

Cycles

Results (primary 2.5%)

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

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

Binary size

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

Bootstrap: 466.494s -> 466.26s (-0.05%)
Artifact size: 388.33 MiB -> 388.41 MiB (0.02%)

Revert "Make `lto` and `linker-plugin-lto` work the same for `compiler_builtins` This reverts commit cf8753e (PR rust-lang#145368) and fix the regressions reported at rust-lang#145981, rust-lang#146109, and rust-lang#146145.

RalfJung · 2025-11-10T21:55:32Z

Is there any way to add a test to prevent this regression from occurring again?

dianqk · 2025-11-19T14:58:50Z

Is there any way to add a test to prevent this regression from occurring again?

I will add it.

dianqk · 2025-11-19T15:11:24Z

Is there any way to add a test to prevent this regression from occurring again?

For cargo: rust-lang/cargo#16277, I will also add a test to rustc later.

Add a test for rust-lang/rust#146133. `cargo +nightly-2025-08-29 test --test build-std -- lto` can reproduce the regression. This is not a bug from Cargo, but it requires `-Zbuild-std` in most use cases. The test case is from rust-lang/rust#146109. The point is that when rustc is invoked with -Clto=fat or -Clto=thin, it should perform LTO with ALL bitcodes. However, rust-lang/rust#145368 emits bitcodes for compiler_builtins but excludes it from LTO participation. As a result, the compiler_builtins bitcodes library is passed to the linker, but the linkers, such as the GNU ld or older versions of LLD, are unable to process bitcodes.

rustbot assigned davidtwco Sep 2, 2025

rustbot added PG-exploit-mitigations Project group: Exploit mitigations 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. labels Sep 2, 2025

This comment has been minimized.

Sign in to view

rustbot assigned bjorn3 and unassigned davidtwco Sep 2, 2025

rcvalle force-pushed the rust-cfi-fix-145981 branch from df97f0f to 0bc9d00 Compare September 2, 2025 18:27

This comment has been minimized.

Sign in to view

tgross35 changed the title ~~Revert "Make lto and linker-plugin-lto work the same for `compile…~~ Revert "Make lto and linker-plugin-lto work the same for compiler_builtins Sep 2, 2025

Revert "Make lto and linker-plugin-lto work the same for `compile…

916b55e

…r_builtins`" This reverts commit cf8753e and fixes the regressions reported.

rcvalle force-pushed the rust-cfi-fix-145981 branch from 0bc9d00 to 916b55e Compare September 2, 2025 20:11

rcvalle marked this pull request as draft September 2, 2025 22:27

rustbot 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-review Status: Awaiting review from the assignee but also interested parties. labels Sep 2, 2025

rcvalle marked this pull request as ready for review September 3, 2025 06:20

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 Sep 3, 2025

lqd mentioned this pull request Sep 3, 2025

staticlib with lto=fat now includes compiler_builtins as llvm bitcode after rustup update #146145

Closed

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 Sep 3, 2025

jhpratt mentioned this pull request Sep 3, 2025

Rollup of 15 pull requests #146153

Closed

RalfJung mentioned this pull request Sep 3, 2025

rustc-LLVM ERROR of ControlFlowIntegrit #142284

Open

Zalathar mentioned this pull request Sep 3, 2025

Rollup of 16 pull requests #146158

Closed

bors added the merged-by-bors This PR was explicitly merged by bors. label Sep 3, 2025

bors merged commit a1208bf into rust-lang:master Sep 3, 2025
11 checks passed

rustbot added this to the 1.91.0 milestone Sep 3, 2025

rcvalle deleted the rust-cfi-fix-145981 branch September 4, 2025 21:12

This was referenced Sep 4, 2025

regression with LTO + musl + build-std #146109

Closed

Cross linker lto error: conflict llvm version. #145981

Closed

Sanitizers target modificators #138736

Merged

ctz mentioned this pull request Sep 5, 2025

Introduce ApiMisuse error rustls/rustls#2620

Merged

dianqk mentioned this pull request Nov 19, 2025

test(build-std): Add test for LTO rust-lang/cargo#16277

Merged

Revert "Make lto and linker-plugin-lto work the same for compiler_builtins #146133

Revert "Make lto and linker-plugin-lto work the same for compiler_builtins #146133

Uh oh!

Conversation

rcvalle commented Sep 2, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

rustbot commented Sep 2, 2025

Uh oh!

rustbot commented Sep 2, 2025

Uh oh!

This comment has been minimized.

rcvalle commented Sep 2, 2025

Uh oh!

This comment has been minimized.

bjorn3 commented Sep 2, 2025

Uh oh!

rcvalle commented Sep 3, 2025

Uh oh!

rustbot commented Sep 3, 2025

Uh oh!

nikic commented Sep 3, 2025

Uh oh!

bors commented Sep 3, 2025

Uh oh!

nikic commented Sep 3, 2025

Uh oh!

RalfJung commented Sep 3, 2025

Uh oh!

bors commented Sep 3, 2025

Uh oh!

bors commented Sep 3, 2025

Uh oh!

Uh oh!

github-actions bot commented Sep 3, 2025

Test differences

Stage 1

Stage 2

Job duration changes

Uh oh!

rust-timer commented Sep 3, 2025

Overall result: no relevant changes - no action needed

Instruction count

Max RSS (memory usage)

Cycles

Binary size

Uh oh!

RalfJung commented Nov 10, 2025

Uh oh!

dianqk commented Nov 19, 2025

Uh oh!

dianqk commented Nov 19, 2025

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

9 participants

Revert "Make `lto` and `linker-plugin-lto` work the same for `compiler_builtins` #146133

Revert "Make `lto` and `linker-plugin-lto` work the same for `compiler_builtins` #146133

rcvalle commented Sep 2, 2025 •

edited

Loading