Remove some `unchecked_claim_error_was_emitted` calls
We want to drive the number of these calls down as much as possible. This PR gets rid of a bunch of them.
r? ``@oli-obk``
Don't use bashism in checktools.sh
`if [[` doesn't work because this is a `/bin/sh` script. We were never running the success side of this `if` at all.
Add parallel rustc ui tests
Updates #118698
Add some ui tests for parallel rustc front end
This is a relatively large feature so I think it's worth creating a new entity in tests/ui folder, so we need to modify the limit in tidy.
Use the same mir-opt bless targets on all platforms
This undoes some of the implementation in https://github.com/rust-lang/rust/pull/119035, but not the effect. Sorry for the churn, I've learned a lot about how all this works over the past few weeks.
The objective here is to make `x test mir-opt --bless` use the same set of targets on all platforms. It didn't do that from the start because bootstrap assumes that a target linker is available, so the availability of cross-linkers is how we ended up with `MIR_OPT_BLESS_TARGET_MAPPING` and poor support for blessing mir-opt tests from Aarch64 MacOS. This PR corrects that.
So I've adjusted the bless targets for mir-opt tests, as well as tweaked some of the logic in bootstrap about linker configuration so that we don't try to access the cache of cc/linker configuration when doing the mir-opt builds.
While working on that I realized that if I swapped from the `cargo rustc -p std` strategy to `cargo check` on the sysroot, I could use the existing code for check builds to bypass some linker logic. Sweet.
But just doing that doesn't work, because then mir-opt tests complain that they can't find an rlib for any of the standard library crates. That happens because nearly all the mir-opt tests are attempting to build `CrateType::Executable`. We already have all the MIR required for mir-opt tests from the rmeta files, but since rustc think we're trying to build an executable it demands we have access to all the upstream monomorphizations that only exist in rlibs, not the meta files in a MIR-only sysroot.
So to fix that, I've swapped all the mir-opt tests be passed `--crate-type=rlib`. That works, but leaves us with a few broken mir-opt tests which I've blessed or fixed up; we also lose MIR for some functions so I added `-Clink-dead-code` to paper over that. The inlining changes are because changing the crate-type perturbs the hashes that are compared here to sometimes let us do inlining even in a possibly-recursive call: 4cb17b4e78/compiler/rustc_mir_transform/src/inline.rs (L332-L341)
`main_args` calls `from_matches`, which does lots of initialization. If
anything goes wrong, `from_matches` emits an error message and returns
`Err(1)` (or `Err(3)`). `main_args` then turns the `Err(1)` into
`Err(ErrorGuaranteed)`, because that's what `catch_with_exit_code`
requires on error. But `catch_with_exit_code` doesn't do anything with
the `ErrorGuaranteed`, it just exits with `EXIT_FAILURE`.
We can avoid the creation of the `ErrorGuaranteed` (which requires
an undesirable `unchecked_claim_error_was_emitted` call), by changing
`from_matches` to instead eagerly abort if anything goes wrong. The
behaviour from the user's point of view is the same: an early abort with
an `EXIT_FAILURE` exit code.
And we can also simplify `from_matches` to return an `Option` instead of
a `Result`:
- Old `Err(0)` case --> `None`
- Old `Err(_)` case --> fatal error.
This requires similar changes to `ScrapeExamplesOptions::new` and
`load_call_locations`.
By making non-unicode arguments a fatal error instead of a warning, we
don't need to handle what comes after, which avoids the need for an
`unchecked_claim_error_was_emitted` call.
Rework support for async closures; allow them to return futures that borrow from the closure's captures
This PR implements a new lowering for async closures via `TyKind::CoroutineClosure` which handles the curious relationship between the closure and the coroutine that it returns.
I wrote up a bunch in [this hackmd](https://hackmd.io/`@compiler-errors/S1HvqQxca)` which will be copied to the dev guide after this PR lands, and hopefully left sufficient comments in the source code explaining why this change is as large as it is.
This also necessitates that they begin implementing the `AsyncFn`-family of traits, rather than the `Fn`-family of traits -- if you need `Fn` implementations, you should probably use the non-sugar `|| async {}` syntax instead.
Notably this PR does not yet implement `async Fn()` syntax sugar for bounds, but I expect to add those soon (**edit:** #120392). For now, users must use `AsyncFn()` traits directly, which necessitates adding the `async_fn_traits` feature gate as well. I will add this as a follow-up very soon.
r? oli-obk
This is based on top of #120322, but that PR is minimal.
All the other `emit`/`emit_diagnostic` methods were recently made
consuming (e.g. #119606), but this one wasn't. But it makes sense to.
Much of this is straightforward, and lots of `clone` calls are avoided.
There are a couple of tricky bits.
- `Emitter::primary_span_formatted` no longer takes a `Diagnostic` and
returns a pair. Instead it takes the two fields from `Diagnostic` that
it used (`span` and `suggestions`) as `&mut`, and modifies them. This
is necessary to avoid the cloning of `diag.children` in two emitters.
- `from_errors_diagnostic` is rearranged so various uses of `diag` occur
before the consuming `emit_diagnostic` call.
make matching on NaN a hard error, and remove the rest of illegal_floating_point_literal_pattern
These arms would never be hit anyway, so the pattern makes little sense. We have had a future-compat lint against float matches in general for a *long* time, so I hope we can get away with immediately making this a hard error.
This is part of implementing https://github.com/rust-lang/rfcs/pull/3535.
Closes https://github.com/rust-lang/rust/issues/41620 by removing the lint.
https://github.com/rust-lang/reference/pull/1456 updates the reference to match.
rustdoc: trait.impl, type.impl: sort impls to make it not depend on serialization order
Can be tested by running `cargo doc` with different rust versions on some crate and comparing `doc` folders: files in `trait.impl` and `type.impl` will sometimes have different order of impls.
Rollup of 8 pull requests
Successful merges:
- #120484 (Avoid ICE when is_val_statically_known is not of a supported type)
- #120516 (pattern_analysis: cleanup manual impls)
- #120517 (never patterns: It is correct to lower `!` to `_`.)
- #120523 (Improve `io::Read::read_buf_exact` error case)
- #120528 (Store SHOULD_CAPTURE as AtomicU8)
- #120529 (Update data layouts in custom target tests for LLVM 18)
- #120531 (Remove a bunch of `has_errors` checks that have no meaningful or the wrong effect)
- #120533 (Correct paths for hexagon-unknown-none-elf platform doc)
r? `@ghost`
`@rustbot` modify labels: rollup
CI: Use ninja on apple builders
This switches the apple builders to use ninja when building LLVM. My hope is that this should resolve the timeouts we have been experiencing since December. My theory is that something in the image update from [Dec 20](dec20a5272) is causing an issue with our build (or, perhaps more remotely, some [update to LLVM itself](https://github.com/rust-lang/rust/commits/master/src/llvm-project)).
The symptoms are that during the LLVM build it just hangs just before the install step. The last thing it prints is `[100%] Built target llvm-reduce` and then just hangs. Normally the next part should be `Install the project...` where it starts installing LLVM. I'm able to reproduce this without too much difficulty. I've been testing ninja, and it seems to be working better (however, my test isn't quite equivalent, since I'm getting sccache misses, and I can't update the S3 bucket).
Installing ninja takes about 7 to 10 seconds, so it shouldn't impact things. I can't determine if it will affect the overall build timing due to not being able to test with a warm S3 cache.
Correct paths for hexagon-unknown-none-elf platform doc
Update the library paths to correctly refer to libdemo{1,2}_hexagon and switch to the release build instead.
Update references to libstandalone to the specific G0/PIC archive instead.
Update the library paths to correctly refer to libdemo{1,2}_hexagon
and switch to the release build instead.
Update references to libstandalone to the specific G0/PIC archive instead.
rustdoc: Correctly handle attribute merge if this is a glob reexport
Fixes#120487.
The regression was introduced in https://github.com/rust-lang/rust/pull/113091. Only non-glob reexports should have been impacted.
cc `````@Nemo157`````
r? `````@notriddle`````
Don't hash lints differently to non-lints.
`Diagnostic::keys`, which is used for hashing and equating diagnostics, has a surprising behaviour: it ignores children, but only for lints. This was added in #88493 to fix some duplicated diagnostics, but it doesn't seem necessary any more.
This commit removes the special case and only four tests have changed output, with additional errors. And those additional errors aren't exact duplicates, they're just similar. For example, in src/tools/clippy/tests/ui/same_name_method.rs we currently have this error:
```
error: method's name is the same as an existing method in a trait
--> $DIR/same_name_method.rs:75:13
|
LL | fn foo() {}
| ^^^^^^^^^^^
|
note: existing `foo` defined here
--> $DIR/same_name_method.rs:79:9
|
LL | impl T1 for S {}
| ^^^^^^^^^^^^^^^^
```
and with this change we also get this error:
```
error: method's name is the same as an existing method in a trait
--> $DIR/same_name_method.rs:75:13
|
LL | fn foo() {}
| ^^^^^^^^^^^
|
note: existing `foo` defined here
--> $DIR/same_name_method.rs:81:9
|
LL | impl T2 for S {}
|
```
I think printing this second argument is reasonable, possibly even preferable to hiding it. And the other cases are similar.
r? `@estebank`
