Audit Symbols by camsteffen · Pull Request #152624 · rust-lang/rust

camsteffen · 2026-02-14T17:22:49Z

Remove unused symbols and hoist some symbols to Clippy.

rustbot · 2026-02-14T17:22:52Z

Some changes occurred in src/tools/clippy

cc @rust-lang/clippy

rustbot · 2026-02-14T17:22:55Z

rustbot has assigned @adwinwhite.
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

Why was this reviewer chosen?

The reviewer was selected based on:

Owners of files modified in this PR: compiler
compiler expanded to 68 candidates
Random selection from 15 candidates

camsteffen · 2026-02-14T17:29:36Z

compiler/rustc_span/src/symbol.rs

-        new_v1,
-        new_v1_formatted,
        next,
-        niko,


@nikomatsakis sorry

samueltardieu

With the removal of the slice duplicate, you can r=me for the Clippy part.

View changes since this review

src/tools/clippy/clippy_utils/src/sym.rs

bjorn3 · 2026-02-14T22:53:58Z

Some of these symbols were added as a perf optimization to avoid having to allocate for identifiers that are commonly used. And pre-interned symbols are represented with less bytes in the crate metadata.

bjorn3 · 2026-02-14T23:02:11Z

compiler/rustc_span/src/symbol.rs

-        BTreeEntry,
        BTreeMap,
-        BTreeSet,
-        BinaryHeap,


These were added in camsteffen@eada4d1 They are still diagnostics items.

Since then, #138682 gave Clippy the ability to define their own Symbols. Many diagnostic items exist only for Clippy, so I figured only Clippy benefits from the pre-defined Symbols.

bjorn3 · 2026-02-14T23:03:46Z

compiler/rustc_span/src/symbol.rs

        SymbolIntern,
        Sync,
        SyncUnsafeCell,
-        T,


This is a very common identifier.

I'd rather try to pre-intern all ASCII letters a-zA-Z, similarly to sym::integer, and see if it makes any difference. Many of them are common.
(Not as a part of this PR.)

Makes sense. I wonder if there should be a special place for pre-interned symbols that are not sym::_ consts, just to ensure that silly people like me won't spend time auditing them.

Mark-Simulacrum · 2026-02-14T23:42:16Z

@bors try @rust-timer queue

Audit Symbols

camsteffen · 2026-02-14T23:46:35Z

Some of these symbols were added as a perf optimization to avoid having to allocate for identifiers that are commonly used. And pre-interned symbols are represented with less bytes in the crate metadata.

Hmm, I didn't realize that. I can look for names that are common and add those back. So IIUC we may want to keep as_str since that is a common method name, but not necessarily keep String since most crates don't define a String.

I'll wait for the perf run for now.

bjorn3 · 2026-02-15T00:19:01Z

but not necessarily keep String since most crates don't define a String

But they do reference it. Each crate has an independent symbol interner. There is no cross-crate importing of symbols.

adwinwhite · 2026-02-15T01:20:50Z

@bjorn3 I'm not familiar with this part and won't have much time looking into it either. Would you like to take over the review since you're already reviewing it?

rust-bors · 2026-02-15T01:53:26Z

☀️ Try build successful (CI)
Build commit: 6f9923e (6f9923e5e0a6b621aa74d57d64d3ff834b60a8cc, parent: a33907a7a5381473eec8bcfa0c56e05a856a911c)

rust-timer · 2026-02-15T02:33:44Z

Finished benchmarking commit (6f9923e): comparison URL.

Overall result: ❌✅ regressions and improvements - no action needed

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.

@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
Regressions ❌ (secondary)	0.3%	[0.3%, 0.3%]	2
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-0.3%	[-0.5%, -0.2%]	3
All ❌✅ (primary)	-	-	0

Max RSS (memory usage)

Results (primary 3.0%, 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)	3.0%	[3.0%, 3.0%]	1
Regressions ❌ (secondary)	2.3%	[1.3%, 2.8%]	3
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-6.8%	[-6.8%, -6.8%]	1
All ❌✅ (primary)	3.0%	[3.0%, 3.0%]	1

Cycles

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

Binary size

Results (primary -0.0%, 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)	-0.0%	[-0.0%, -0.0%]	4
Improvements ✅ (secondary)	-0.0%	[-0.0%, -0.0%]	14
All ❌✅ (primary)	-0.0%	[-0.0%, -0.0%]	4

Bootstrap: 483.617s -> 479.802s (-0.79%)
Artifact size: 397.94 MiB -> 397.91 MiB (-0.01%)

Audit Symbols *[View all comments](https://triagebot.infra.rust-lang.org/gh-comments/rust-lang/rust/pull/152624)* Remove unused symbols and hoist some symbols to Clippy.

jhpratt · 2026-02-21T02:51:45Z

CI appears stuck (x86_64-gnu-aux specifically)

@bors yield

rust-bors · 2026-02-21T02:51:49Z

Auto build cancelled. Cancelled workflows:

https://github.com/rust-lang/rust/actions/runs/22244490978

The next pull request likely to be tested is #152624.

Audit Symbols *[View all comments](https://triagebot.infra.rust-lang.org/gh-comments/rust-lang/rust/pull/152624)* Remove unused symbols and hoist some symbols to Clippy.

jhpratt · 2026-02-21T02:52:35Z

@bors yield

Let's give a rollup a chance first

rust-bors · 2026-02-21T02:52:38Z

Auto build cancelled. Cancelled workflows:

https://github.com/rust-lang/rust/actions/runs/22248970210

The next pull request likely to be tested is #152924.

JonathanBrouwer · 2026-02-21T10:10:46Z

The last job is aarch64-apple which tends to finish if you just give it time, so this time I'm not worried about it being stuck :)

rust-bors · 2026-02-21T10:15:30Z

☀️ Test successful - CI
Approved by: petrochenkov
Duration: 3h 31m 2s
Pushing 3fc3732 to main...

github-actions · 2026-02-21T10:19:06Z

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 1c00e98 (parent) -> 3fc3732 (this PR)

Test differences

Show 1 test diff

1 doctest diff were found. These are ignored, as they are noisy.

Test dashboard

Run

cargo run --manifest-path src/ci/citool/Cargo.toml -- \
    test-dashboard 3fc37321c49cb8aed32da426de6145e6d57105da --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: 2h 32m -> 1h 54m (-25.0%)
dist-x86_64-apple: 2h 36m -> 2h 8m (-17.8%)
aarch64-msvc-2: 1h 40m -> 1h 52m (+12.6%)
dist-apple-various: 1h 51m -> 1h 38m (-11.8%)
x86_64-msvc-ext1: 2h 3m -> 2h 14m (+9.2%)
aarch64-apple: 3h 44m -> 3h 23m (-9.2%)
x86_64-gnu-llvm-21-1: 1h 13m -> 1h 7m (-8.9%)
dist-aarch64-msvc: 1h 38m -> 1h 46m (+8.4%)
dist-ohos-aarch64: 1h 16m -> 1h 22m (+7.9%)
dist-various-1: 1h 6m -> 1h 11m (+7.1%)

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 · 2026-02-21T10:55:35Z

Finished benchmarking commit (3fc3732): 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)	0.2%	[0.2%, 0.2%]	1
Regressions ❌ (secondary)	0.2%	[0.1%, 0.2%]	2
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-0.5%	[-0.7%, -0.3%]	4
All ❌✅ (primary)	0.2%	[0.2%, 0.2%]	1

Max RSS (memory usage)

Results (secondary 1.9%)

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)	1.9%	[0.9%, 3.5%]	8
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	-	-	0

Cycles

Results (secondary -4.1%)

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)	-4.1%	[-4.1%, -4.1%]	1
All ❌✅ (primary)	-	-	0

Binary size

Results (primary -0.0%, 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
Improvements ✅ (primary)	-0.0%	[-0.0%, -0.0%]	1
Improvements ✅ (secondary)	-0.0%	[-0.0%, -0.0%]	14
All ❌✅ (primary)	-0.0%	[-0.0%, -0.0%]	1

Bootstrap: 480.967s -> 482.367s (0.29%)
Artifact size: 397.91 MiB -> 397.92 MiB (0.00%)

nnethercote · 2026-02-21T19:44:21Z

@camsteffen: I'm curious, how did you go about this? Did you use automation, tedious trial and error, something else? Thanks.

samueltardieu · 2026-02-21T20:00:39Z

In Clippy, we have a all_symbols_used test to ensure that we don't define symbols that aren't later referenced in the source. This is a bit quick-n-dirty but it does the job so far.

camsteffen · 2026-02-21T20:02:25Z

A mix of automation and trial and error. I wrote some scripts using grep to generate a list of usages that I could diff against a list of definitions. A lot of usages don't include sym:: so it was far from perfect. Especially large sets like feature names. I'm not a total script wizard so that part was AI assisted (don't @ me). Unfortunately not a method I can happily advocate for repeating. Maybe a better approach would be a custom HIR pass generate a list of usages.

@samueltardieu Ah I didn't know about that. Looks roughly similar to what I did. But unfortunately rustc has a lot more hard to grep cases.

RalfJung · 2026-02-22T10:53:51Z

compiler/rustc_span/src/symbol.rs

-        sys_mutex_lock,
-        sys_mutex_try_lock,
-        sys_mutex_unlock,


FWIW these were used in Miri. I guess we can re-intern them all the time there instead or figure out how to use Clippy's approach, but a heads-up would have been nice.

Hm, do we need to add a feature to x.py's check for miri with genmc perhaps? I think that's where the usage is in miri from a quick grep locally, which is probably why this got missed in PR CI etc.?

Yeah that's why it got missed in CI (though a quick grep would have found it).

GenMC mode needs a bit more work before we want to enable it in the nightly distribution. Is the idea that ./x check miri could enable genmc even if ./x build miri does not?

Mark-Simulacrum · 2026-02-24T03:20:06Z

Perf regressions look close to noise thresholds (though maybe not entirely noise). Regardless I don't think it's worth trying to iterate on the specific set of symbols too much, it's effectively neutral (mean including non-relevant is +0.01% on instructions, +0.00% on cycles).

rustbot assigned adwinwhite Feb 14, 2026

camsteffen commented Feb 14, 2026

View reviewed changes

samueltardieu reviewed Feb 14, 2026

View reviewed changes

src/tools/clippy/clippy_utils/src/sym.rs Outdated Show resolved Hide resolved

This comment has been minimized.

Sign in to view

camsteffen force-pushed the symbols-audit branch from 14ad4f5 to f35741a Compare February 14, 2026 18:08

This comment has been minimized.

Sign in to view

camsteffen force-pushed the symbols-audit branch from f35741a to dc753d0 Compare February 14, 2026 19:08

bjorn3 reviewed Feb 14, 2026

View reviewed changes

This comment has been minimized.

Sign in to view

rust-bors bot pushed a commit that referenced this pull request Feb 14, 2026

Auto merge of #152624 - camsteffen:symbols-audit, r=<try>

6f9923e

Audit Symbols

rustbot added the S-waiting-on-perf Status: Waiting on a perf run to be completed. label Feb 14, 2026

This comment has been minimized.

Sign in to view

rustbot removed the S-waiting-on-perf Status: Waiting on a perf run to be completed. label Feb 15, 2026

This comment has been minimized.

Sign in to view

rust-bors bot added merged-by-bors This PR was explicitly merged by bors. and removed S-waiting-on-bors Status: Waiting on bors to run and complete tests. Bors will change the label on completion. labels Feb 21, 2026

rust-bors bot merged commit 3fc3732 into rust-lang:main Feb 21, 2026
12 checks passed

rustbot added this to the 1.95.0 milestone Feb 21, 2026

rustbot added the perf-regression Performance regression. label Feb 21, 2026

GrigorenkoPV mentioned this pull request Feb 21, 2026

pre-intern single-letter sym::[a-zA-Z] #152948

Open

camsteffen deleted the symbols-audit branch February 21, 2026 21:29

RalfJung reviewed Feb 22, 2026

View reviewed changes

Mark-Simulacrum added the perf-regression-triaged The performance regression has been triaged. label Feb 24, 2026

Uh oh!

Comments

Conversation

camsteffen commented Feb 14, 2026 • edited by rustbot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

rustbot commented Feb 14, 2026

Uh oh!

rustbot commented Feb 14, 2026

Uh oh!

Choose a reason for hiding this comment

Uh oh!

samueltardieu 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!

This comment has been minimized.

This comment has been minimized.

bjorn3 commented Feb 14, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

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!

Mark-Simulacrum commented Feb 14, 2026

Uh oh!

This comment has been minimized.

This comment has been minimized.

camsteffen commented Feb 14, 2026

Uh oh!

bjorn3 commented Feb 15, 2026

Uh oh!

adwinwhite commented Feb 15, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

rust-bors bot commented Feb 15, 2026

Uh oh!

This comment has been minimized.

rust-timer commented Feb 15, 2026

Overall result: ❌✅ regressions and improvements - no action needed

Uh oh!

This comment has been minimized.

This comment has been minimized.

jhpratt commented Feb 21, 2026

Uh oh!

rust-bors bot commented Feb 21, 2026

Uh oh!

This comment has been minimized.

jhpratt commented Feb 21, 2026

Uh oh!

rust-bors bot commented Feb 21, 2026

Uh oh!

This comment has been minimized.

JonathanBrouwer commented Feb 21, 2026

Uh oh!

rust-bors bot commented Feb 21, 2026

Uh oh!

Uh oh!

github-actions bot commented Feb 21, 2026

Test differences

Job duration changes

Uh oh!

rust-timer commented Feb 21, 2026

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

Uh oh!

nnethercote commented Feb 21, 2026

Uh oh!

samueltardieu commented Feb 21, 2026

Uh oh!

camsteffen commented Feb 21, 2026

Uh oh!

RalfJung Feb 22, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

camsteffen commented Feb 14, 2026 •

edited by rustbot

Loading

samueltardieu left a comment •

edited by rustbot

Loading

bjorn3 commented Feb 14, 2026 •

edited

Loading

adwinwhite commented Feb 15, 2026 •

edited

Loading

RalfJung Feb 22, 2026 •

edited

Loading