port lint attributes by mejrs · Pull Request #162811 · rust-lang/rust

mejrs · 2026-09-15T13:17:22Z

Ports the lint attributes. I've kept this as small as possible, but there is a decent amount of clean up/refactoring that can be done afterwards,

r? @JonathanBrouwer @jdonszelmann
cc @Bryntet

rustbot · 2026-09-15T13:17:25Z

Some changes occurred in compiler/rustc_passes/src/check_attr.rs

cc @jdonszelmann, @JonathanBrouwer

These commits modify the Cargo.lock file. Unintentional changes to Cargo.lock can be introduced when switching branches and rebasing PRs.

If this was unintentional then you should revert the changes before this PR is merged.
Otherwise, you can ignore this comment.

Some changes occurred in compiler/rustc_attr_ir

cc @jdonszelmann, @JonathanBrouwer

Some changes occurred in compiler/rustc_attr_parsing

cc @jdonszelmann, @JonathanBrouwer

clippy is developed in its own repository. If possible, consider making this change to rust-lang/rust-clippy instead.

cc @rust-lang/clippy

mejrs · 2026-09-15T13:21:15Z

+LL |     let y: u32 = (x?).try_into().unwrap();
+   |                  +  +++++++++++++++++++++


Uh yeah I don't know what's happening here

View changes since the review

I think this is minor enough that I'll put it on my todo list as something that we should fix after merging this PR

JonathanBrouwer · 2026-09-15T13:23:07Z

I'll take a look at this on Friday! Thanks for doing this!

Bryntet · 2026-09-15T13:27:28Z

Thanks for picking this up ❤️

JonathanBrouwer · 2026-09-15T13:34:09Z

cc @cjgillot @petrochenkov in case you'd like to take a look (as reviewers of the original pr #155691)

mejrs · 2026-09-15T13:34:44Z

I'll take a look at this on Friday! Thanks for doing this!

I'll go ahead and cherrypick some changes from this PR then, some things can be split off.

JonathanBrouwer · 2026-09-15T14:10:12Z

Blocked on #162813

mejrs · 2026-09-15T15:58:16Z

@bors try @rust-timer queue

port lint attributes

rust-bors · 2026-09-15T16:59:49Z

☀️ Try build successful (CI)
Build commit: 837b948 (837b9481206c510936be02f4b200f8db8aae1286)
Base parent: 1936d44 (1936d44457352da43b2d740672b940ebc715ac38)

rust-timer · 2026-09-15T17:47:26Z

Finished benchmarking commit (837b948): comparison URL.

Overall result: ❌✅ regressions and improvements - BENCHMARK(S) FAILED

Benchmarking means the PR may be perf-sensitive. It's automatically marked not fit for rolling up. Overriding is possible but disadvised: it risks changing compiler perf.

Next, please: If you can, justify the regressions found in this try perf run in writing along with @rustbot label: +perf-regression-triaged. If not, fix the regressions and do another perf run. Neutral or positive results will clear the label automatically.

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

❗ ❗ ❗ ❗ ❗
Warning ⚠️: The following benchmark(s) failed to build:

image-0.25.6
Job failure

❗ ❗ ❗ ❗ ❗

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)	0.6%	[0.2%, 3.5%]	197
Regressions ❌ (secondary)	0.9%	[0.1%, 5.8%]	170
Improvements ✅ (primary)	-36.2%	[-77.2%, -10.4%]	4
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	-0.1%	[-77.2%, 3.5%]	201

Max RSS (memory usage)

Results (primary -7.1%, 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)	1.6%	[0.6%, 3.1%]	3
Regressions ❌ (secondary)	3.5%	[2.4%, 4.1%]	3
Improvements ✅ (primary)	-20.2%	[-22.6%, -17.8%]	2
Improvements ✅ (secondary)	-7.1%	[-9.4%, -3.6%]	3
All ❌✅ (primary)	-7.1%	[-22.6%, 3.1%]	5

Cycles

Results (primary -7.6%, secondary 2.7%)

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

	mean	range	count
Regressions ❌ (primary)	3.0%	[1.3%, 4.3%]	11
Regressions ❌ (secondary)	3.1%	[2.0%, 5.2%]	16
Improvements ✅ (primary)	-36.8%	[-77.7%, -10.0%]	4
Improvements ✅ (secondary)	-4.5%	[-4.5%, -4.5%]	1
All ❌✅ (primary)	-7.6%	[-77.7%, 4.3%]	15

Binary size

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

Bootstrap: 495.925s -> 496.275s (0.07%)
Artifact size: 406.90 MiB -> 407.07 MiB (0.04%)

port lint attributes

rust-bors · 2026-10-04T15:02:41Z

☀️ Try build successful (CI)
Build commit: be9d55c (be9d55ce5f1537cfe68101e2ffd17e70d4fd3b0e)
Base parent: 4ddbc06 (4ddbc06ea09abcda34f80baf31cc9bc2686b0ae6)

rust-timer · 2026-10-04T16:13:01Z

Finished benchmarking commit (be9d55c): comparison URL.

Overall result: ❌✅ regressions and improvements - please read:

Benchmarking means the PR may be perf-sensitive. It's automatically marked not fit for rolling up. Overriding is possible but disadvised: it risks changing compiler perf.

Next, please: If you can, justify the regressions found in this try perf run in writing along with @rustbot label: +perf-regression-triaged. If not, fix the regressions and do another perf run. Neutral or positive results will clear the label automatically.

@bors rollup=never rustc-perf
@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)	0.2%	[0.2%, 0.2%]	5
Regressions ❌ (secondary)	0.2%	[0.1%, 0.3%]	17
Improvements ✅ (primary)	-0.2%	[-0.8%, -0.1%]	25
Improvements ✅ (secondary)	-0.2%	[-0.5%, -0.1%]	17
All ❌✅ (primary)	-0.2%	[-0.8%, 0.2%]	30

Max RSS (memory usage)

Results (primary -2.6%, secondary -3.2%)

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)	-2.6%	[-6.0%, -1.0%]	3
Improvements ✅ (secondary)	-3.2%	[-3.7%, -2.5%]	3
All ❌✅ (primary)	-2.6%	[-6.0%, -1.0%]	3

Cycles

Results (primary 4.1%, secondary 4.9%)

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

	mean	range	count
Regressions ❌ (primary)	4.1%	[4.1%, 4.1%]	1
Regressions ❌ (secondary)	4.9%	[2.1%, 12.9%]	10
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	4.1%	[4.1%, 4.1%]	1

Binary size

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

Bootstrap: 491.424s -> 491.568s (0.03%)
Artifact size: 408.69 MiB -> 408.55 MiB (-0.03%)

JonathanBrouwer · 2026-10-04T17:24:27Z

Perf seems slightly worse than the previous run but it's still positive, and the behavior is more correct now, so let's keep things this way

JonathanBrouwer

r=me
Thanks so much for all the effort! This has been a lot of work, both from you and @Bryntet

@petrochenkov do you still want to take a look?
@bors rollup=never

View changes since this review

JonathanBrouwer · 2026-10-09T07:24:21Z

I went through the PR and collected a list of items for follow up PRs. Not all are things we definitely want to do, but at least we can consider them:

parenthesis https://github.com/rust-lang/rust/pull/162811/changes#r4016038072
duplicate https://github.com/mejrs/rust/blob/3d3790ab08c1ae5de09a9e304e403bb9ace3430e/tests/ui/lint/rfc-2383-lint-reason/lint-attribute-only-with-reason.rs
see if we can remove rest: Option<Box[Symbol]>> for perf
lint resolution in parsing port lint attributes #162811 (comment)
unhelpful error port lint attributes #162811 (comment)
unused attribute code duplication port lint attributes #162811 (comment)
otherwise don't emit error when parsing pre expansion lints
remove unsafe? port lint attributes #162811 (comment)
can we avoid calling meta_item_list? port lint attributes #162811 (comment)
Does making tool a Sym rather than a Ident improve perf?
Can we unify LintCheckKind and Level?

petrochenkov · 2026-10-12T13:35:27Z

-#![attr = Feature([fn_delegation#0])]
-extern crate std;
-#[attr = PreludeImport]
-use std::prelude::rust_2021::*;


Why did the pretty-printing change, even for things unrelated to lint attributes?
The previous version preserved the relative order of comments and attributes, and the new version does not, so this seems like a regression.

View changes since the review

The relative order is that comments come before unparsed attributes, but there are no longer any unparsed attributes in this file.

See

rust/compiler/rustc_hir_pretty/src/lib.rs

Lines 290 to 302 in 0093837

// Print all attributes, regardless of actual style, as inner attributes

// since this is the crate root with nothing above it to print outer

// attributes.

for attr in s.attrs(hir::CRATE_HIR_ID) {

s.print_attribute_as_style(attr, ast::AttrStyle::Inner);

}

// When printing the AST, we sometimes need to inject `#[no_std]` here.

// Since you can't compile the HIR, it's not necessary.

s.print_mod(krate);

s.print_remaining_comments();

s.s.eof()

rust/compiler/rustc_hir_pretty/src/lib.rs

Lines 102 to 130 in 0093837

fn print_attribute_as_style(&mut self, attr: &Attribute, style: ast::AttrStyle) {

match &attr {

Attribute::Unparsed(unparsed) => {

self.maybe_print_comment(unparsed.span.lo());

match style {

ast::AttrStyle::Inner => self.word("#!["),

ast::AttrStyle::Outer => self.word("#["),

}

self.print_attr_item(&unparsed, unparsed.span);

self.word("]");

self.hardbreak()

}

Attribute::Parsed(AttributeKind::DocComment { kind, comment, .. }) => {

self.word(rustc_ast_pretty::pprust::state::doc_comment_to_string(

*kind, style, *comment,

));

self.hardbreak()

}

Attribute::Parsed(pa) => {

match style {

ast::AttrStyle::Inner => self.word("#![attr = "),

ast::AttrStyle::Outer => self.word("#[attr = "),

}

pa.print_attribute(self);

self.word("]");

self.hardbreak()

}

}

}

(all the other pretty tests already have this regression, this is the only one that had a lint check attribute)

The relative order is that comments come before unparsed attributes

I don't see any reasons to print parsed and unparsed attributes in different locations, and the unparsed location looks more correct.
But it's not an issue for this PR, if it's pre-existing.

It has existed since parsed attributes, example:

// comment #![feature()]

hir pretty:

#![attr = Feature([])] extern crate std; #[attr = PreludeImport] use std::prelude::rust_2024::*; // comment

It seems that correctly printing comments requires a span, but parsed attributes generally don't have one, so it's not done for the Parsed variant

It seems that correctly printing comments requires a span, but parsed attributes generally don't have one, so it's not done for the Parsed variant

I see, perhaps it makes sense to do it for parsed attributes that do have a span for other reasons.

That would make it inconsistent whether comments or crate attributes are printed first 🤔

Having a span for all parsed attributes would also be annoying (aside from the ~decent amount of work), it'd probably increase the size of Attribute/AttributeKind and I'm not sure how to best deal with combined attributes (like multiple repr attributes).

petrochenkov · 2026-10-12T14:41:03Z

+                            &format!("{name}::{full_lint_name}")
+                        } else {
+                            full_lint_name
+                        };


Style nit: if let with short arms look better as matches (here and maybe in some other places).

View changes since the review

This is a bit too nitty for me, and while it's a wash to me which style this uses it's not going to last because I'm going to be rewriting this part soon. Next I want to do improvements to reduce the amount of strings that get needlessly allocated around here (this pr already got rid of some).

Do you strongly care about this? I don't want to do it, personally, but I also don't want to have a debate on it so I'll do it if you want me to.

petrochenkov · 2026-10-12T14:43:22Z

LGTM, most things I had issues with are already listed in #162811 (comment).

rustbot · 2026-10-12T19:14:05Z

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.

mejrs · 2026-10-12T21:00:26Z

I went through the PR and collected a list of items for follow up PRs. Not all are things we definitely want to do, but at least we can consider them:

Some thoughts on some of them:

* see if we can remove `rest: Option<Box[Symbol]>>` for perf

with register_tool comes the ability to define custom lint names (which can have as many segments as the user would like), so this will be a lang matter

* lint resolution in parsing [port lint attributes #162811 (comment)](https://github.com/rust-lang/rust/pull/162811#discussion_r4046648708)

If macro lints ever happen (in the form of #135432 or smth else) we likely need a tcx to resolve these lints, which cannot be done in attribute parsing

* remove unsafe? [port lint attributes #162811 (comment)](https://github.com/rust-lang/rust/pull/162811#discussion_r4173185326)

a "downcast_slice" method could also exist in std or e.g. rustc_data_structures.

* can we avoid calling `meta_item_list`? [port lint attributes #162811 (comment)](https://github.com/rust-lang/rust/pull/162811#discussion_r4173196063)

I think a "just try to parse this without actually erroring" might work (we kinda do this for diagnostic attributes), but it's a bit brittle for my liking, I'm also not sure I like ShouldEmit's state at this time.

Another option is to force pre-expansion lints to be crate level only (like #163714) but this would also be a lang (new keyword lint) and clippy matter.

* Can we unify `LintCheckKind` and `Level`?

Level also has a ForceWarn variant, which has no lint check attribute counterpart.

rustbot assigned JonathanBrouwer Sep 15, 2026

mejrs commented Sep 15, 2026

View reviewed changes

mejrs force-pushed the port_lint_check branch from 5f087e2 to c90513f Compare September 15, 2026 13:33

JonathanBrouwer mentioned this pull request Sep 15, 2026

Re-port lint attributes to attribute parsers #155691

Closed

petrochenkov self-assigned this Sep 15, 2026

mejrs commented Sep 15, 2026

View reviewed changes

Comment thread tests/ui/unpretty/exhaustive.hir.stdout

mejrs mentioned this pull request Sep 15, 2026

Pre lint port cleanups #162813

Merged

JonathanBrouwer added the S-blocked Status: Blocked on something else such as an RFC or other implementation work. label Sep 15, 2026

This comment has been minimized.

Sign in to view

mejrs force-pushed the port_lint_check branch from c90513f to 5c86bdf Compare September 15, 2026 14:32

This comment has been minimized.

Sign in to view

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

This comment has been minimized.

Sign in to view

rust-bors Bot pushed a commit that referenced this pull request Sep 15, 2026

Auto merge of #162811 - mejrs:port_lint_check, r=<try>

837b948

port lint attributes

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 Sep 15, 2026

This comment has been minimized.

Sign in to view

rust-bors Bot pushed a commit that referenced this pull request Oct 4, 2026

Auto merge of #162811 - mejrs:port_lint_check, r=<try>

be9d55c

port lint attributes

This comment has been minimized.

Sign in to view

mejrs force-pushed the port_lint_check branch from e325410 to 89b2c2c Compare October 4, 2026 15:20

rustbot removed the S-waiting-on-perf Status: Waiting on a perf run to be completed. label Oct 4, 2026

JonathanBrouwer approved these changes Oct 4, 2026

View reviewed changes

mejrs 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 Oct 4, 2026

This comment has been minimized.

Sign in to view

mejrs force-pushed the port_lint_check branch from 89b2c2c to 77dff77 Compare October 8, 2026 14:29

This comment has been minimized.

Sign in to view

petrochenkov reviewed Oct 12, 2026

View reviewed changes

petrochenkov 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 Oct 12, 2026

mejrs added 6 commits October 12, 2026 21:07

port lint attributes

9fee241

ditch attr/lint index

054a7a8

misc perf stuff

7ed0383

allocate less

d3f8f3f

comment on attr_id field

310c4f3

implement feedback

65e41fb

This comment has been minimized.

Sign in to view

fix attribute test

a0a416a

		LL \| let y: u32 = (x?).try_into().unwrap();
		\| + +++++++++++++++++++++

	// Print all attributes, regardless of actual style, as inner attributes
	// since this is the crate root with nothing above it to print outer
	// attributes.
	for attr in s.attrs(hir::CRATE_HIR_ID) {
	s.print_attribute_as_style(attr, ast::AttrStyle::Inner);
	}

	// When printing the AST, we sometimes need to inject `#[no_std]` here.
	// Since you can't compile the HIR, it's not necessary.

	s.print_mod(krate);
	s.print_remaining_comments();
	s.s.eof()

	fn print_attribute_as_style(&mut self, attr: &Attribute, style: ast::AttrStyle) {
	match &attr {
	Attribute::Unparsed(unparsed) => {
	self.maybe_print_comment(unparsed.span.lo());
	match style {
	ast::AttrStyle::Inner => self.word("#!["),
	ast::AttrStyle::Outer => self.word("#["),
	}
	self.print_attr_item(&unparsed, unparsed.span);
	self.word("]");
	self.hardbreak()
	}
	Attribute::Parsed(AttributeKind::DocComment { kind, comment, .. }) => {
	self.word(rustc_ast_pretty::pprust::state::doc_comment_to_string(
	kind, style, comment,
	));
	self.hardbreak()
	}
	Attribute::Parsed(pa) => {
	match style {
	ast::AttrStyle::Inner => self.word("#![attr = "),
	ast::AttrStyle::Outer => self.word("#[attr = "),
	}
	pa.print_attribute(self);
	self.word("]");
	self.hardbreak()
	}
	}
	}

Conversation

mejrs commented Sep 15, 2026 • edited by rustbot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

rustbot commented Sep 15, 2026

Uh oh!

mejrs Sep 15, 2026 • 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!

Choose a reason for hiding this comment

Uh oh!

JonathanBrouwer commented Sep 15, 2026

Uh oh!

Bryntet commented Sep 15, 2026

Uh oh!

JonathanBrouwer commented Sep 15, 2026

Uh oh!

mejrs commented Sep 15, 2026

Uh oh!

Uh oh!

JonathanBrouwer commented Sep 15, 2026

Uh oh!

This comment has been minimized.

mejrs commented Sep 15, 2026

Uh oh!

This comment has been minimized.

This comment has been minimized.

rust-bors Bot commented Sep 15, 2026

Uh oh!

This comment has been minimized.

rust-timer commented Sep 15, 2026

Overall result: ❌✅ regressions and improvements - BENCHMARK(S) FAILED

Uh oh!

This comment has been minimized.

This comment has been minimized.

rust-bors Bot commented Oct 4, 2026

Uh oh!

This comment has been minimized.

rust-timer commented Oct 4, 2026

Overall result: ❌✅ regressions and improvements - please read:

Uh oh!

JonathanBrouwer commented Oct 4, 2026

Uh oh!

JonathanBrouwer left a comment • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

This comment has been minimized.

This comment has been minimized.

JonathanBrouwer commented Oct 9, 2026

Uh oh!

petrochenkov Oct 12, 2026 • 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!

Choose a reason for hiding this comment

Uh oh!

mejrs Oct 12, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

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!

petrochenkov Oct 12, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

mejrs commented Sep 15, 2026 •

edited by rustbot

Loading

mejrs Sep 15, 2026 •

edited by rustbot

Loading

JonathanBrouwer left a comment •

edited

Loading

petrochenkov Oct 12, 2026 •

edited by rustbot

Loading

mejrs Oct 12, 2026 •

edited

Loading

petrochenkov Oct 12, 2026 •

edited

Loading

petrochenkov Oct 12, 2026 •

edited by rustbot

Loading