about summary refs log tree commit diff
AgeCommit message (Collapse)AuthorLines
2025-06-30Move lazy type alias checks to non-hir-wfckOli Scherer-91/+63
2025-06-30Use predicate spans instead of whole item spansOli Scherer-135/+113
2025-06-30Check variances in the non-hir wfcheckerOli Scherer-78/+105
2025-06-30Make variance wfcheck only use the HIR in the error pathOli Scherer-28/+26
2025-06-30Don't look at static items' HIR for wfcheckOli Scherer-78/+68
2025-06-30fix bitcast of single-element SIMD vectorsFolkert de Vries-18/+47
2025-06-30Unconditionally run `check_item_type` on all itemsOli Scherer-392/+593
2025-06-30Try increasing the repetition limitOli Scherer-1/+1
2025-06-30Require either wrapping nullary intrinsincs in const blocks or explicitly ↵Oli Scherer-1/+22
deciding not to
2025-06-30Remove the nullary intrinsic const eval logic and treat them like other ↵Oli Scherer-124/+78
intrinsics
2025-06-30Stop backends from needing to support nullary intrinsicsOli Scherer-4/+5
2025-06-30Stop backends from needing to support nullary intrinsicsOli Scherer-86/+67
2025-06-30Make check_param_wf only go through the HIR in the error pathOli Scherer-26/+31
2025-06-30Stop storing `Span` that is only used for checking for global bounds that do ↵Oli Scherer-21/+14
not hold
2025-06-30cleaned up some testsKivooeo-17/+45
2025-06-30Move `RUST_TEST_NOCAPTURE` warning to early config checkJieyou Xu-6/+7
2025-06-30Move compiletest `main.rs` to `src/bin/`Jieyou Xu-0/+4
To make it obvious `compiletest`-the-tool has two components: 1. The core compiletest library, and 2. The tool binary, which will be executed by bootstrap.
2025-06-30Remove `unused_trait_names`'s support inside macrosyanglsh-56/+6
2025-06-30fix: `unused_trait_names` FP when `as` name is from macro inputyanglsh-0/+39
2025-06-30Move some early config checks to the compiletest libJieyou Xu-14/+17
2025-06-30Merge pull request #2485 from dianne/update-cond-loop-desugaringnora-4/+2
update AST-to-HIR lowering examples for conditionals and loops
2025-06-30cleaned up some testsKivooeo-97/+152
2025-06-30Merge pull request #4429 from RalfJung/futex-val-too-bigRalf Jung-3/+9
linux futex: fix for val > i32::MAX
2025-06-30Handle build scripts better in `-Zmacro-stats` output.Nicholas Nethercote-1/+11
Currently all build scripts are listed as `build_script_build` in the stats header. This commit uses `CARGO_PKG_NAME` to improve that. I tried it on Bevy, it works well, giving output like this on the build script: ``` MACRO EXPANSION STATS: serde build script ``` and this on the crate itself: ``` MACRO EXPANSION STATS: serde ```
2025-06-30cleaned up some testsKivooeo-33/+57
2025-06-30linux futex: fix for val > i32::MAXRalf Jung-3/+9
2025-06-30cleaned up some testsKivooeo-27/+43
2025-06-30Merge pull request #20124 from zachs18/patch-1Lukas Wirth-5/+5
Remove last use of `rustc_pattern_analysis::Captures`
2025-06-29update AST-to-HIR lowering examples for conditionals and loopsdianne-4/+2
- `for` loops now use two `match`es for all of their bindings. I'm not sure this is the most helpful way of conveying that, but it's about as informative as before while staying brief. - `while let` and `if let` don't use `match`; they use `let` expressions in their conditions. Since `if let` no longer has significantly different desugaring and having a whole bullet point for `while` would feel redundant with `for`, I've removed those examples.
2025-06-30build-fail directive: make explanation more uniformTshepang Mbambo-3/+3
The last part of the paragraph did not fit
2025-06-29Remove last use of `rustc_pat_analysis::Captures`zachs18-5/+5
It's not necessary anymore due to Rust 2024 lifetime capture rules.
2025-06-30Improve macro-stats printing.Nicholas Nethercote-8/+17
By allowing long names to overlap with the "Uses" field when it has spare space. This avoids unnecessary line breaks in the output.
2025-06-30Augment the macro-stats test.Nicholas Nethercote-3/+10
With a long macro name that could fit on one line, but currently isn't formatted that way, because the name would overlap with the maximum width of the "Uses" column. (The next commit will fix this.)
2025-06-30Auto merge of #142941 - compiler-errors:shallow-bail, r=lcnrbors-19/+70
Shallowly bail from `coerce_unsized` more We do a *lot* of coercion in HIR typeck. Most of the time we're just coercing a type to itself, but we always try `coerce_unsized` even if it's not necessary. Let's avoid doing that by adding a fast path to `coerce_unsized`; see the comment in that function.
2025-06-29Merge pull request #723 from rust-lang/sync_from_rust_2025_06_28antoyo-450/+222
Sync from rust 2025/06/28
2025-06-29Show auto trait and blanket impls for `!`zachs18-0/+5
Add an empty `impl ! {}` so `rustdoc` generates auto trait impls and blanket impls on `!`'s documentation page.
2025-06-29Fix handling of __builtin_ia32_rdtscpAntoni Boucher-3/+2
2025-06-30Regenerated intrinsicsFractalFir-58/+66
2025-06-30Changed intrinsic generation to include full intrinsic names in panic messagesFractalFir-6/+6
2025-06-30Do not autofix comments containing bare CRSamuel Tardieu-0/+35
2025-06-29Fix `cast_possible_truncation` should not suggest inside const context (#15164)Samuel Tardieu-4/+28
Closes rust-lang/rust-clippy#15163 This lint should not give suggestions in const context for now, since a lot of the commonly used method of `Result` is not const. changelog: [`cast_possible_truncation`] fix improperly give suggestions inside const context
2025-06-29Merge pull request #4397 from tgross35/gettid-shimsRalf Jung-12/+312
Add shims for `gettid`-esque functions
2025-06-29Merge pull request #4422 from RalfJung/more-ciRalf Jung-31/+65
run CI on a bunch more architectures
2025-06-29Add shims for `gettid`-esque functionsTrevor Gross-12/+312
Various platforms provide a function to return the current OS thread ID, but they all use a slightly different name. Add shims for these functions for Apple, FreeBSD, and Windows, with tests to account for those and a few more platforms that are not yet supported by Miri. The syscall and extern symbol is included as well on Linux. These should be useful in general but will also help support printing the OS thread ID in panic messages [1]. [1]: https://github.com/rust-lang/rust/pull/115746 Squashed commit from Ralf: try_from_scalar: extend comment
2025-06-29add a riscv64 test jobRalf Jung-0/+5
2025-06-29remove duplicate clippy checkRalf Jung-4/+2
2025-06-29also test on s390x via qemuRalf Jung-0/+15
2025-06-29test on x86-32 and arm-32Ralf Jung-25/+41
2025-06-29fix type mismatches in native-lib/scalar_arguments testRalf Jung-2/+2
2025-06-29add float_nan test for powfRalf Jung-28/+54
plus various minor tweaks