Tell LLVM about read-only captures #145259

nikic · 2025-08-11T16:03:10Z

&Freeze parameters are not only readonly within the function, but any captures of the pointer can also only be used for reads. This can now be encoded using the captures(address, read_provenance) attribute.

nikic · 2025-08-11T16:03:59Z

@bors try @rust-timer queue

Tell LLVM about read-only captures

rust-bors · 2025-08-11T18:17:39Z

☀️ Try build successful (CI)
Build commit: 6bc3793 (6bc37934c6eaa84bfd300c54189bec62e15d42ed, parent: 577166503aee7290e09374da21f4045c455acfd5)

rust-timer · 2025-08-11T19:27:15Z

Finished benchmarking commit (6bc3793): comparison URL.

Overall result: ❌✅ regressions and improvements - please read the text below

Benchmarking this pull request means it may be perf-sensitive – we'll automatically label it not fit for rolling up. You can override this, but we strongly advise not to, due to possible changes in compiler perf.

Next Steps: If you can justify the regressions found in this try perf run, please do so in sufficient writing along with @rustbot label: +perf-regression-triaged. If not, please fix the regressions and do another perf run. If its results are neutral or positive, the label will be automatically removed.

@bors rollup=never
@rustbot label: -S-waiting-on-perf +perf-regression

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%	[0.3%, 0.4%]	5
Regressions ❌ (secondary)	1.0%	[0.9%, 1.6%]	7
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-0.8%	[-0.8%, -0.7%]	5
All ❌✅ (primary)	0.3%	[0.3%, 0.4%]	5

Max RSS (memory usage)

Results (secondary 3.9%)

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

Cycles

Results (secondary -2.1%)

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
Improvements ✅ (secondary)	-2.1%	[-2.1%, -2.1%]	1
All ❌✅ (primary)	-	-	0

Binary size

Results (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
Regressions ❌ (secondary)	0.0%	[0.0%, 0.0%]	23
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	-	-	0

Bootstrap: 465.465s -> 464.036s (-0.31%)
Artifact size: 377.33 MiB -> 377.30 MiB (-0.01%)

nikic · 2025-08-19T10:40:10Z

tests/ui/codegen/equal-pointers-unequal/as-cast/segfault.rs

@@ -55,6 +55,7 @@ fn main() {
    // The `Box` has been deallocated by now, so this is a dangling reference!
    let r: &u8 = &*r;
    println!("{:p}", r);
+    println!("{}", i);


These adjustments are necessary so these tests still get miscompiled.

nikic · 2025-08-19T10:43:50Z

r? compiler

wesleywiser

Thanks @nikic!

r=me with or without suggestion

tests/codegen-llvm/read-only-capture-opt.rs

`&Freeze` parameters are not only `readonly` within the function, but any captures of the pointer can also only be used for reads. This can now be encoded using the `captures(address, read_provenance)` attribute.

The capture of i in assert_ne!() is now known read-only, which enables early SROA. Block this by passing i to println, where we currently cannot recognize this.

nikic · 2025-08-20T19:12:27Z

@bors r=wesleywiser

bors · 2025-08-20T19:12:29Z

📌 Commit dd151be has been approved by wesleywiser

It is now in the queue for this repository.

bors · 2025-08-20T23:41:44Z

⌛ Testing commit dd151be with merge 125ff8a...

bors · 2025-08-21T02:46:42Z

☀️ Test successful - checks-actions
Approved by: wesleywiser
Pushing 125ff8a to master...

github-actions · 2025-08-21T02:49:42Z

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 040a98a (parent) -> 125ff8a (this PR)

Test differences

Show 7 test diffs

Stage 1

[codegen] tests/codegen-llvm/read-only-capture-opt.rs: [missing] -> ignore (ignored when the LLVM version 19.1.1 is older than 21.0.0) (J0)
[codegen] tests/codegen-llvm/read-only-capture-opt.rs: [missing] -> ignore (ignored when the LLVM version 20.1.2 is older than 21.0.0) (J2)

Stage 2

[codegen] tests/codegen-llvm/read-only-capture-opt.rs: [missing] -> ignore (ignored when the LLVM version 19.1.1 is older than 21.0.0) (J1)
[codegen] tests/codegen-llvm/read-only-capture-opt.rs: [missing] -> pass (J3)
[codegen] tests/codegen-llvm/read-only-capture-opt.rs: [missing] -> ignore (ignored when the LLVM version 20.1.2 is older than 21.0.0) (J4)

Additionally, 2 doctest diffs were found. These are ignored, as they are noisy.

Job group index

J0: x86_64-gnu-llvm-19-3
J1: aarch64-gnu-llvm-19-1, x86_64-gnu-llvm-19-2
J2: x86_64-gnu-llvm-20-3
J3: aarch64-apple, aarch64-gnu, aarch64-msvc-1, arm-android, armhf-gnu, dist-i586-gnu-i586-i686-musl, i686-gnu-1, i686-gnu-nopt-1, i686-msvc-1, test-various, x86_64-gnu, x86_64-gnu-nopt, x86_64-gnu-stable, x86_64-mingw-1, x86_64-msvc-1
J4: x86_64-gnu-llvm-20-2

Test dashboard

Run

cargo run --manifest-path src/ci/citool/Cargo.toml -- \
    test-dashboard 125ff8a788c5d6a66917f499abdc00051afe6886 --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-aarch64-linux: 8945.1s -> 5954.9s (-33.4%)
dist-apple-various: 4248.7s -> 5367.6s (26.3%)
dist-x86_64-apple: 6496.1s -> 8134.3s (25.2%)
dist-aarch64-msvc: 5256.9s -> 6130.5s (16.6%)
i686-msvc-2: 9200.7s -> 7840.8s (-14.8%)
pr-check-1: 1583.7s -> 1371.4s (-13.4%)
dist-various-1: 3853.5s -> 4265.6s (10.7%)
aarch64-gnu-llvm-19-2: 2413.3s -> 2230.4s (-7.6%)
dist-x86_64-msvc-alt: 9908.0s -> 9159.0s (-7.6%)
x86_64-msvc-ext2: 6232.4s -> 5763.1s (-7.5%)

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 · 2025-08-21T03:59:20Z

Finished benchmarking commit (125ff8a): comparison URL.

Overall result: ❌✅ regressions and improvements - 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)	0.8%	[0.8%, 0.8%]	1
Regressions ❌ (secondary)	0.3%	[0.2%, 0.4%]	7
Improvements ✅ (primary)	-0.2%	[-0.2%, -0.2%]	1
Improvements ✅ (secondary)	-1.3%	[-1.7%, -0.2%]	8
All ❌✅ (primary)	0.3%	[-0.2%, 0.8%]	2

Max RSS (memory usage)

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

Cycles

Results (secondary -1.7%)

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)	3.7%	[3.7%, 3.7%]	1
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-3.0%	[-3.3%, -2.6%]	4
All ❌✅ (primary)	-	-	0

Binary size

Results (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
Regressions ❌ (secondary)	0.0%	[0.0%, 0.0%]	1
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	-	-	0

Bootstrap: 471.047s -> 471.807s (0.16%)
Artifact size: 378.31 MiB -> 378.19 MiB (-0.03%)

panstromek · 2025-08-25T16:47:10Z

perf triage:

include-blob regressions are noise. serde_derive looks real but it's tiny (adding a println statement). Seems like the biggest query differences are in LLVM, that seems to match the PR area.

@rustbot label: +perf-regression-triaged

This comment has been minimized.

Sign in to view

rust-bors bot added a commit that referenced this pull request Aug 11, 2025

Auto merge of #145259 - nikic:read-only-capture, r=<try>

6bc3793

Tell LLVM about read-only captures

rustbot added the S-waiting-on-perf Status: Waiting on a perf run to be completed. label Aug 11, 2025

This comment has been minimized.

Sign in to view

rustbot added perf-regression Performance regression. and removed S-waiting-on-perf Status: Waiting on a perf run to be completed. labels Aug 11, 2025

nikic force-pushed the read-only-capture branch from 2f266b1 to a68e8be Compare August 19, 2025 10:37

nikic marked this pull request as ready for review August 19, 2025 10:38

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 19, 2025

nikic commented Aug 19, 2025

View reviewed changes

rustbot assigned wesleywiser Aug 19, 2025

This comment has been minimized.

Sign in to view

nikic force-pushed the read-only-capture branch from a68e8be to 66fd5d8 Compare August 19, 2025 11:29

wesleywiser approved these changes Aug 20, 2025

View reviewed changes

tests/codegen-llvm/read-only-capture-opt.rs Outdated Show resolved Hide resolved

nikic added 2 commits August 20, 2025 19:08

Tell LLVM about read-only captures

d71ed8d

`&Freeze` parameters are not only `readonly` within the function, but any captures of the pointer can also only be used for reads. This can now be encoded using the `captures(address, read_provenance)` attribute.

Adjust test to still show miscompile

dd151be

The capture of i in assert_ne!() is now known read-only, which enables early SROA. Block this by passing i to println, where we currently cannot recognize this.

nikic force-pushed the read-only-capture branch from 66fd5d8 to dd151be Compare August 20, 2025 17:08

bors removed the S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. label Aug 20, 2025

bors added the S-waiting-on-bors Status: Waiting on bors to run and complete tests. Bors will change the label on completion. label Aug 20, 2025

bors added the merged-by-bors This PR was explicitly merged by bors. label Aug 21, 2025

bors merged commit 125ff8a into rust-lang:master Aug 21, 2025
11 checks passed

rustbot added this to the 1.91.0 milestone Aug 21, 2025

rustbot added the perf-regression-triaged The performance regression has been triaged. label Aug 25, 2025

Tell LLVM about read-only captures #145259

Tell LLVM about read-only captures #145259

Uh oh!

Conversation

nikic commented Aug 11, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

nikic commented Aug 11, 2025

Uh oh!

This comment has been minimized.

This comment has been minimized.

This comment has been minimized.

rust-bors bot commented Aug 11, 2025

Uh oh!

This comment has been minimized.

rust-timer commented Aug 11, 2025

Overall result: ❌✅ regressions and improvements - please read the text below

Uh oh!

Choose a reason for hiding this comment

Uh oh!

nikic commented Aug 19, 2025

Uh oh!

This comment has been minimized.

wesleywiser left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

nikic commented Aug 20, 2025

Uh oh!

bors commented Aug 20, 2025

Uh oh!

bors commented Aug 20, 2025

Uh oh!

bors commented Aug 21, 2025

Uh oh!

Uh oh!

github-actions bot commented Aug 21, 2025

Test differences

Stage 1

Stage 2

Job duration changes

Uh oh!

rust-timer commented Aug 21, 2025

Overall result: ❌✅ regressions and improvements - please read the text below

Uh oh!

panstromek commented Aug 25, 2025

Uh oh!

Uh oh!

nikic commented Aug 11, 2025 •

edited

Loading