| Age | Commit message (Collapse) | Author | Lines |
|---|
|
|
|
|
|
|
|
|
|
|
|
r=GuillaumeGomez,ChrisDenton,wesleywiser
Add support for Arm64EC to the Standard Library
Adds the final pieces so that the standard library can be built for arm64ec-pc-windows-msvc (initially added in #119199)
* Bumps `windows-sys` to 0.56.0, which adds support for Arm64EC.
* Correctly set the `isEC` parameter for LLVM's `writeArchive` function.
* Add `#![feature(asm_experimental_arch)]` to library crates where Arm64EC inline assembly is used, as it is currently unstable.
|
|
Simplify `static_assert_size`s.
We want to run them on all 64-bit platforms.
r? `@ghost`
|
|
llvm: update riscv target feature to match LLVM 19
In llvm/llvm-project@9067070d91e9d8cdd8509ffa56a076f08a3d7281 they ended up largely reverting
llvm/llvm-project@e81796671890b59c110f8e41adc7ca26f8484d20. This means the change we did in
rust-lang/rust@b378059e6b2573c5356423fa31d184a89a3b6029 is now only corrct for LLVM 18...so we have to adjust again.
``@rustbot`` label: +llvm-main
|
|
Move confusing comment about otherwise blocks in `lower_match_tree`
This comment was historically inside a block guarded by `if let Some(otherwise_block) = otherwise`.
When #120978 made the “otherwise block” non-optional, it also flattened that region of code. Doing so left this comment awkwardly stranded above an unrelated line of code, without its original context.
We can restore that context by moving it above the declaration of `otherwise`.
r? ``@Nadrieril``
|
|
Match ergonomics 2024: miscellaneous code cleanups
- Store `ByRef` instead of `BindingAnnotation` in `PatInfo`
- Rename `BindingAnnotation` to `BindingMode`
r? ``@Nadrieril``
cc #123076
``@rustbot`` label A-patterns
|
|
We want to run them on all 64-bit platforms.
|
|
r=RalfJung
Delay interning errors to after validation
fixes https://github.com/rust-lang/rust/issues/122398
fixes #122548
This improves diagnostics since validation errors are usually more helpful compared with interning errors that just make broad statements about the entire constant
r? `@RalfJung`
|
|
In llvm/llvm-project@9067070d91e9d8cdd8509ffa56a076f08a3d7281 they ended
up largely reverting
llvm/llvm-project@e81796671890b59c110f8e41adc7ca26f8484d20. This means
the change we did in
rust-lang/rust@b378059e6b2573c5356423fa31d184a89a3b6029 is now only
corrct for LLVM 18...so we have to adjust again.
@rustbot label: +llvm-main
|
|
Codegen ZSTs without an allocation
This makes sure that &[] is equivalent to unsafe code (from_raw_parts(dangling, 0)). No new stable guarantee is intended about whether or not we do this, this is just an optimization.
This regressed in #67000 (no comments I can see about that regression in the PR, though it did change the test modified here). We had previously performed this optimization since #63635.
|
|
Rollup of 9 pull requests
Successful merges:
- #116957 (meta: notify #t-rustdoc Zulip stream on backport nominations)
- #122201 (Document overrides of `clone_from()` in core/std)
- #122723 (Use same file permissions for ar_archive_writer as the LLVM archive writer)
- #124030 (interpret: pass MemoryKind to adjust_alloc_base_pointer)
- #124037 (Don't ascend into parent bodies when collecting stmts for possible return suggestion)
- #124049 (Stabilize `const_io_structs`)
- #124062 (Add another expression to weird-exprs.rs)
- #124066 (Don't error on subtyping of equal types)
- #124073 (Remove libc from rust_get_test_int uses)
r? `@ghost`
`@rustbot` modify labels: rollup
|
|
Don't error on subtyping of equal types
fixes https://github.com/rust-lang/rust/issues/124054 🤦
fixes #124075
fixes https://github.com/rust-lang/rust/issues/124079
r? `@compiler-errors`
|
|
Don't ascend into parent bodies when collecting stmts for possible return suggestion
Fixes #124022
|
|
interpret: pass MemoryKind to adjust_alloc_base_pointer
Another puzzle piece for https://github.com/rust-lang/miri/pull/3475.
The 2nd commit renames base_pointer -> root_pointer; that's how Tree Borrows already calls them and I think the term is more clear than "base pointer". In particular, this distinguishes it from "base address", since a root pointer can point anywhere into an allocation, not just its base address.
https://github.com/rust-lang/rust/pull/124018 has been rolled up already so I couldn't add it there any more.
r? ```@oli-obk```
|
|
Use same file permissions for ar_archive_writer as the LLVM archive writer
This is required to switch to ar_archive_writer in the future without regressions. In addition to this PR support for reading thin archives needs to be added (https://github.com/rust-lang/rust/issues/107407) to fix all known regressions.
Fixes https://github.com/rust-lang/rust/issues/107495
|
|
Silence some follow-up errors on trait impls in case the trait has conflicting or otherwise incoherent impls
fixes #123292
Also removes a bunch of extra diagnostics that were introduced in https://github.com/rust-lang/rust/pull/121154 and https://github.com/rust-lang/rust/pull/120558
|
|
|
|
|
|
|
|
static libraries with ar_archive_writer
Fixes #107495
|
|
|
|
This comment was historically inside a block guarded by
`if let Some(otherwise_block) = otherwise`.
When #120978 made the otherwise block non-optional, it also flattened that
region of code. Doing so left this comment awkwardly stranded above an
unrelated line of code, without its original context.
We can restore that context by moving it above the declaration of `otherwise`.
|
|
|
|
validation produces much higher quality errors and already handles most of the cases
|
|
|
|
|
|
|
|
|
|
also in Miri, "base tag" -> "root tag"
|
|
|
|
Make the comments for `ReturnDest` variants doc comments
This PR converts the documentation for `ReturnDest` from comments to doc comments.
|
|
Fix empty-set symbol in comments
The symbol in the original code is U+00D8 "LATIN CAPITAL LETTER O WITH STROKE" (https://en.wikipedia.org/wiki/%C3%98) which is an uppercase letter in Danish, Norwegian, Faroese, and Southern Sámi alphabets.
The symbol that was intended is U+2205 "EMPTY SET" (https://en.wikipedia.org/wiki/Empty_set#Notation).
| Before | After |
|---|---|
|  |  |
|
|
Delay span bug when `Self` kw resolves to `DefKind::{Mod,Trait}`
Catch the case where `kw::Self` is recovered in the parser and causes us to subsequently resolve `&self`'s implicit type to something that's not a type.
This check could be made more accurate, though I'm not sure how hard we have to try here.
Fixes #123988
|
|
Qualifier tweaking
Adding and removing qualifiers in some cases that make things nicer. Details in individual commits.
r? `@compiler-errors`
|
|
|
|
|
|
This makes sure that &[] is just as efficient as indirecting through
unsafe code (from_raw_parts). No new stable guarantee is intended about
whether or not we do this, this is just an optimization.
Co-authored-by: Ralf Jung <post@ralfj.de>
|
|
Rollup of 7 pull requests
Successful merges:
- #123673 (Don't ICE for kind mismatches during error rendering)
- #123675 (Taint const qualifs if a static is referenced that didn't pass wfcheck)
- #123975 (Port the 2 `rust-lld` run-make tests to `rmake`)
- #124000 (Use `/* value */` as a placeholder)
- #124013 (Box::into_raw: make Miri understand that this is a box-to-raw cast)
- #124027 (Prefer identity equality over equating types during coercion.)
- #124036 (Remove `default_hidden_visibility: false` from wasm targets)
r? `@ghost`
`@rustbot` modify labels: rollup
|
|
Remove `default_hidden_visibility: false` from wasm targets
To the best of my ability I believe that this is no longer necessary. I don't fully recall why this was first added but I believe it had to do with symbols all being exported by default and this was required to undo that. Regardless nowadays the default output of rustc seems suitable so it seems best to keep wasm in line with other targets.
|
|
Prefer identity equality over equating types during coercion.
These types are always generic only over their own generic parameters with no inference variables involved.
r? `@compiler-errors`
I love touching code that [hasn't changed meaningfully since 2016](https://github.com/rust-lang/rust/pull/41937)
|
|
Use `/* value */` as a placeholder
The expression `value` isn't a valid suggestion; let's use `/* value */` as a placeholder (which is also invalid) since it more clearly signals to the user that they need to fill it in with something meaningful. This parallels the suggestions we have in a couple other places, like arguments.
We could also print the type name instead of `/* value */`, especially if it's suggestable, but I don't care strongly about that.
|
|
Taint const qualifs if a static is referenced that didn't pass wfcheck
It is correct to only check the signature here, as the ICE is caused by `USE_WITH_ERROR` trying to allocate memory to store the result of `WITH_ERROR` before evaluating it.
fixes #123153
|
|
Don't ICE for kind mismatches during error rendering
fixes #123457
also some test suite cleanups to make backtraces easier to read
|
|
Simplify shallow resolver to just fold ty/consts
Probably faster than using a whole folder?
|
|
suggestion
|
|
To the best of my ability I believe that this is no longer necessary. I
don't fully recall why this was first added but I believe it had to do
with symbols all being exported by default and this was required to undo
that. Regardless nowadays the default output of rustc seems suitable so
it seems best to keep wasm in line with other targets.
|
