cfg_select: Support unbraced expressions by joshtriplett · Pull Request #145233 · rust-lang/rust

joshtriplett · 2025-08-10T23:34:07Z

Tracking issue for cfg_select: #115585

When operating on expressions, cfg_select! can now handle expressions
without braces. (It still requires braces for other things, such as
items.)

Expand the test coverage and documentation accordingly.

I'm not sure whether deciding to extend cfg_select! in this way is T-lang or T-libs-api. I've labeled for both, with the request that both teams don't block on each other. :)

When operating on expressions, `cfg_select!` can now handle expressions without braces. (It still requires braces for other things, such as items.) Expand the test coverage and documentation accordingly.

rustbot · 2025-08-10T23:34:11Z

r? @jieyouxu

rustbot has assigned @jieyouxu.
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

compiler/rustc_parse/src/parser/cfg_select.rs

Amanieu · 2025-08-12T15:15:46Z

We discussed this in the @rust-lang/libs-api meeting and everyone was happy with this change.

traviscross · 2025-08-14T08:47:09Z

We discussed this in the lang meeting. We were all happy to see this land in nightly. Thanks to @joshtriplett for putting this forward.

jieyouxu · 2025-08-14T12:49:49Z

I updated the PR description to point to the tracking issue #115585. It seems like the tracking issue was very outdated, so I tried to update that as well (but I probably missed a few still). You may wish to update the tracking issue example (#115585 (comment)) to demonstrate this syntax allowance.

jieyouxu

Implementation wise, this looks good, thanks!

jieyouxu · 2025-08-14T12:51:35Z

@bors r+ rollup

bors · 2025-08-14T12:51:38Z

📌 Commit 38df158 has been approved by jieyouxu

It is now in the queue for this repository.

…eyouxu cfg_select: Support unbraced expressions Tracking issue for `cfg_select`: rust-lang#115585 When operating on expressions, `cfg_select!` can now handle expressions without braces. (It still requires braces for other things, such as items.) Expand the test coverage and documentation accordingly. --- I'm not sure whether deciding to extend `cfg_select!` in this way is T-lang or T-libs-api. I've labeled for both, with the request that both teams don't block on each other. :)

Rollup of 9 pull requests Successful merges: - #118087 (Add Ref/RefMut try_map method) - #140794 (Add information about group a lint belongs to) - #144947 (Fix description of unsigned `checked_exact_div`) - #145005 (strip prefix of temporary file names when it exceeds filesystem name length limit) - #145233 (cfg_select: Support unbraced expressions) - #145243 (take attr style into account in diagnostics) - #145353 (bootstrap: Fix jemalloc 64K page support for aarch64 tools) - #145379 (bootstrap: Support passing `--timings` to cargo) - #145389 ([rustdoc] Revert "rustdoc search: prefer stable items in search results") Failed merges: - #144983 (Rehome 37 `tests/ui/issues/` tests to other subdirectories under `tests/ui/`) - #145065 (resolve: Introduce `RibKind::Block`) r? `@ghost` `@rustbot` modify labels: rollup

…eyouxu cfg_select: Support unbraced expressions Tracking issue for `cfg_select`: rust-lang#115585 When operating on expressions, `cfg_select!` can now handle expressions without braces. (It still requires braces for other things, such as items.) Expand the test coverage and documentation accordingly. --- I'm not sure whether deciding to extend `cfg_select!` in this way is T-lang or T-libs-api. I've labeled for both, with the request that both teams don't block on each other. :)

joshtriplett · 2025-08-14T20:27:31Z

I updated the PR description to point to the tracking issue #115585. It seems like the tracking issue was very outdated, so I tried to update that as well (but I probably missed a few still). You may wish to update the tracking issue example (#115585 (comment)) to demonstrate this syntax allowance.

Done.

Rollup of 11 pull requests Successful merges: - #137872 (Include whitespace in "remove |" suggestion and make it hidden) - #144631 (Fix test intrinsic-raw_eq-const-bad for big-endian) - #145233 (cfg_select: Support unbraced expressions) - #145261 (Improve tracing in bootstrap) - #145324 (Rename and document `ONLY_HOSTS` in bootstrap) - #145353 (bootstrap: Fix jemalloc 64K page support for aarch64 tools) - #145379 (bootstrap: Support passing `--timings` to cargo) - #145397 (Rust documentation, use `rustc-dev-guide` :3) - #145398 (Use `default_field_values` in `Resolver`) - #145401 (cleanup: Remove useless `[T].iter().last()`) - #145403 (Adjust error message grammar to be less awkward) r? `@ghost` `@rustbot` modify labels: rollup

Rollup merge of #145233 - joshtriplett:cfg-select-expr, r=jieyouxu cfg_select: Support unbraced expressions Tracking issue for `cfg_select`: #115585 When operating on expressions, `cfg_select!` can now handle expressions without braces. (It still requires braces for other things, such as items.) Expand the test coverage and documentation accordingly. --- I'm not sure whether deciding to extend `cfg_select!` in this way is T-lang or T-libs-api. I've labeled for both, with the request that both teams don't block on each other. :)

Kobzol · 2025-08-15T06:48:35Z

@rust-timer build b7907e9

Testing for #145407.

rust-timer · 2025-08-15T09:00:19Z

Finished benchmarking commit (b7907e9): 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)	0.2%	[0.2%, 0.4%]	36
Regressions ❌ (secondary)	0.3%	[0.2%, 0.5%]	20
Improvements ✅ (primary)	-2.0%	[-3.0%, -0.9%]	2
Improvements ✅ (secondary)	-0.2%	[-0.2%, -0.1%]	9
All ❌✅ (primary)	0.1%	[-3.0%, 0.4%]	38

Max RSS (memory usage)

Results (primary -3.4%, secondary -2.0%)

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)	0.8%	[0.8%, 0.8%]	1
Improvements ✅ (primary)	-3.4%	[-3.4%, -3.4%]	1
Improvements ✅ (secondary)	-3.5%	[-4.0%, -2.9%]	2
All ❌✅ (primary)	-3.4%	[-3.4%, -3.4%]	1

Cycles

Results (primary -2.7%)

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)	-	-	0
Improvements ✅ (primary)	-2.7%	[-3.2%, -2.1%]	3
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	-2.7%	[-3.2%, -2.1%]	3

Binary size

Results (primary -0.2%, 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)	-0.2%	[-0.6%, -0.1%]	4
Improvements ✅ (secondary)	-0.1%	[-0.3%, -0.1%]	7
All ❌✅ (primary)	-0.2%	[-0.6%, -0.1%]	4

Bootstrap: 470.516s -> 470.62s (0.02%)
Artifact size: 377.36 MiB -> 377.36 MiB (-0.00%)

joshtriplett · 2025-08-15T09:21:24Z

The perf results there don't seem particularly likely from this PR; looking at some of the deltas they're in places like LLVM codegen, which this doesn't even touch. Are those spurious/noise?

Kobzol · 2025-08-15T09:22:43Z

No, it's just this PR. But I think that this might actually be noise, similar to your PR with multiple macro kinds. Maybe modifying the compiler around this area triggers some back-and-forth LLVM codegen differences.

Rollup of 11 pull requests Successful merges: - rust-lang/rust#137872 (Include whitespace in "remove |" suggestion and make it hidden) - rust-lang/rust#144631 (Fix test intrinsic-raw_eq-const-bad for big-endian) - rust-lang/rust#145233 (cfg_select: Support unbraced expressions) - rust-lang/rust#145261 (Improve tracing in bootstrap) - rust-lang/rust#145324 (Rename and document `ONLY_HOSTS` in bootstrap) - rust-lang/rust#145353 (bootstrap: Fix jemalloc 64K page support for aarch64 tools) - rust-lang/rust#145379 (bootstrap: Support passing `--timings` to cargo) - rust-lang/rust#145397 (Rust documentation, use `rustc-dev-guide` :3) - rust-lang/rust#145398 (Use `default_field_values` in `Resolver`) - rust-lang/rust#145401 (cleanup: Remove useless `[T].iter().last()`) - rust-lang/rust#145403 (Adjust error message grammar to be less awkward) r? `@ghost` `@rustbot` modify labels: rollup

…eyouxu cfg_select: Support unbraced expressions Tracking issue for `cfg_select`: rust-lang#115585 When operating on expressions, `cfg_select!` can now handle expressions without braces. (It still requires braces for other things, such as items.) Expand the test coverage and documentation accordingly. --- I'm not sure whether deciding to extend `cfg_select!` in this way is T-lang or T-libs-api. I've labeled for both, with the request that both teams don't block on each other. :)

Mark-Simulacrum · 2025-08-19T02:19:06Z

A few days later looking at the results of the parent rollup (https://perf.rust-lang.org/compare.html?start=898aff704d6f0d00343f21d31b8b9bfac8e43007&end=3507a749b365aae4eefa96ab700a9315d3280ee7&stat=instructions:u) in context of the subsequent perf, I think most of the affected benchmarks became much noisier around the time of this PR (not quite at the same time though, best as I can tell).

So I don't think we should look more into this particular PR at this time.

folkertdev · 2026-01-15T19:20:59Z

compiler/rustc_parse/src/parser/cfg_select.rs

+    if !classify::expr_is_complete(&expr) && p.token != token::CloseBrace && p.token != token::Eof {
+        p.expect(exp!(Comma))?;
+    } else {
+        let _ = p.eat(exp!(Comma));
    }


@joshtriplett what is happening here? @ehuss asked over at rust-lang/reference#2103 (comment) why the p.token != token::CloseBrace is needed here (given that the { /* ... */ } case is already handled above).

I suspect this is an attempt at good error messages? That isn't tested at all though.

@traviscross

…=JonathanBrouwer stabilize `cfg_select!` *[View all comments](https://triagebot.infra.rust-lang.org/gh-comments/rust-lang/rust/pull/149783)* tracking issue: rust-lang#115585 closes rust-lang#115585 reference PR: - rust-lang/reference#2103 # Request for Stabilization ## Summary The `cfg_select!` macro picks the expansion corresponding to the first `cfg` condition that evaluates to `true`. It simplifies complex conditional expressions. ```rust cfg_select! { unix => { fn foo() { /* unix specific functionality */ } } target_pointer_width = "32" => { fn foo() { /* non-unix, 32-bit functionality */ } } _ => { fn foo() { /* fallback implementation */ } } } let is_unix_str = cfg_select! { unix => "unix", _ => "not unix", }; println!("{is_unix_str}"); ``` ## Semantics The expansion of a `cfg_select!` call is the right-hand side of the first `cfg` rule that evaluates to true. This can be roughly expressed using this macro: ```rust macro_rules! cfg_select { ({ $($tt:tt)* }) => {{ $crate::cfg_select! { $($tt)* } }}; (_ => { $($output:tt)* }) => { $($output)* }; ( $cfg:meta => $output:tt $($( $rest:tt )+)? ) => { #[cfg($cfg)] $crate::cfg_select! { _ => $output } $( #[cfg(not($cfg))] $crate::cfg_select! { $($rest)+ } )? } } ``` The actual implementation uses a builtin macro so that `cfg_select!` can be used both in item and expression position. ## Documentation reference PR: - rust-lang/reference#2103 ## Tests The `cfg_select!` macro is already used extensively in the rust compiler codebase. It has several dedicated tests: - [`tests/ui/check-cfg/cfg-select.rs`](https://github.com/rust-lang/rust/blob/main/tests/ui/check-cfg/cfg-select.rs)tests that warnings are emitted when an unexpected `cfg` condition is used. - [`tests/ui/macros/cfg_select.rs`](https://github.com/rust-lang/rust/blob/main/tests/ui/macros/cfg_select.rs) tests that `cfg_select!` has the expected expansion, and tests that the expected syntax is accepted. ## History - rust-lang#115416 - rust-lang#117162 - rust-lang#133720 - rust-lang#135625 - rust-lang#137198 - rust-lang#138993 - rust-lang#138996 - rust-lang#143461 - rust-lang#143941 - rust-lang#145233 - rust-lang#148712 - rust-lang#149380 - rust-lang#149925 # Resolved questions # Unresolved questions The style team has decided on how to format `cfg_select!`, but this formatting has not yet been implemented. See rust-lang#144323. r? @traviscross   > [!NOTE] > # Concerns (0 active) > > - ~~[allowing-comma-after-closing-brace](rust-lang#149783 (comment) resolved in [this comment](rust-lang#149783 (comment)) > > *Managed by `@rustbot`—see [help](https://forge.rust-lang.org/triagebot/concern.html) for details.*

cfg_select: Support unbraced expressions

38df158

When operating on expressions, `cfg_select!` can now handle expressions without braces. (It still requires braces for other things, such as items.) Expand the test coverage and documentation accordingly.

rustbot assigned jieyouxu Aug 10, 2025

rustbot added the S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. label Aug 10, 2025

jieyouxu reviewed Aug 11, 2025

View reviewed changes

compiler/rustc_parse/src/parser/cfg_select.rs Show resolved Hide resolved

traviscross added the P-lang-drag-1 Lang team prioritization drag level 1. https://rust-lang.zulipchat.com/#narrow/channel/410516-t-lang label Aug 12, 2025

joshtriplett removed the I-libs-api-nominated Nominated for discussion during a libs-api team meeting. label Aug 13, 2025

This was referenced Aug 13, 2025

Formatting for cfg_select rust-lang/style-team#201

Open

add rustfmt support for cfg_select #144323

Open

jieyouxu mentioned this pull request Aug 14, 2025

Tracking issue for cfg_select (formerly cfg_match) #115585

Open

5 tasks

jieyouxu added the F-cfg_select `#![feature(cfg_select)]` label Aug 14, 2025

jieyouxu approved these changes Aug 14, 2025

View reviewed changes

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 Aug 14, 2025

GuillaumeGomez mentioned this pull request Aug 14, 2025

Rollup of 9 pull requests #145394

Closed

This was referenced Aug 14, 2025

Rollup of 12 pull requests #145406

Closed

Rollup of 11 pull requests #145407

Merged

bors merged commit 01bd889 into rust-lang:master Aug 15, 2025
10 checks passed

rustbot added this to the 1.91.0 milestone Aug 15, 2025

This comment has been minimized.

Sign in to view

rustbot added the perf-regression Performance regression. label Aug 15, 2025

joshtriplett deleted the cfg-select-expr branch August 15, 2025 09:17

traviscross added the perf-regression-triaged The performance regression has been triaged. label Aug 19, 2025

folkertdev mentioned this pull request Dec 8, 2025

stabilize cfg_select! #149783

Open

folkertdev reviewed Jan 15, 2026

View reviewed changes

Uh oh!

Comments

Conversation

joshtriplett commented Aug 10, 2025 • edited by rustbot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

rustbot commented Aug 10, 2025

Uh oh!

Uh oh!

Amanieu commented Aug 12, 2025

Uh oh!

traviscross commented Aug 14, 2025

Uh oh!

jieyouxu commented Aug 14, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

jieyouxu left a comment

Choose a reason for hiding this comment

Uh oh!

jieyouxu commented Aug 14, 2025

Uh oh!

bors commented Aug 14, 2025

Uh oh!

joshtriplett commented Aug 14, 2025

Uh oh!

Uh oh!

Kobzol commented Aug 15, 2025

Uh oh!

This comment has been minimized.

rust-timer commented Aug 15, 2025

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

Instruction count

Max RSS (memory usage)

Cycles

Binary size

Uh oh!

joshtriplett commented Aug 15, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Kobzol commented Aug 15, 2025

Uh oh!

Mark-Simulacrum commented Aug 19, 2025

Uh oh!

folkertdev Jan 15, 2026

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

10 participants

joshtriplett commented Aug 10, 2025 •

edited by rustbot

Loading

jieyouxu commented Aug 14, 2025 •

edited

Loading

joshtriplett commented Aug 15, 2025 •

edited

Loading