Allow passing a specific date to `bump-stage0`
This allows regenerating `src/stage0.json` on changes to the tool, without needing to hard-code the date in the source. It was useful for https://github.com/rust-lang/rust/pull/106394, which added clippy to the list of required components.
r? `@pietroalbini`
Suggest `impl Fn*` and `impl Future` in `-> _` return suggestions
Follow-up to #106172, only the last commit is relevant. Can rebase once that PR is landed for easier review.
Suggests `impl Future` and `impl Fn{,Mut,Once}` in `-> _` return suggestions.
r? `@estebank`
rustdoc: remove legacy font-feature-settings CSS
According to [caniuse], these vendor prefixes aren't needed in any supported web browsers as defined in [RFC 1985].
* The last version of Chrome that required a vendor prefix was version 47. The current version is 108.
* Firefox 33 is the last version that required a vendor prefix. The [current Firefox ESR] is version 102.
* The last version of Safari that required a vendor prefix was version 9.1. The current version is 16.
* The last version of Safari/iOS that required a vendor prefix was version 9.3. The current version is 16.
* Edge never required vendor prefixes.
* UCAndroid never required vendor prefixes.
[caniuse]: https://caniuse.com/?search=font-feature-settings
[RFC 1985]: https://rust-lang.github.io/rfcs/1985-tiered-browser-support.html
[current Firefox ESR]: https://wiki.mozilla.org/Releases
Jsondoclint: Add `--verbose` and `--json-output` options
There quite helpful for manually using jsondoclint as a debugging tool, especially on large files (these were written to look into core.json).
r? rustdoc
default OOM handler: use non-unwinding panic, to match std handler
The OOM handler in std will by default abort. This adjusts the default in liballoc to do the same, using the `can_unwind` flag on the panic info to indicate a non-unwinding panic.
In practice this probably makes little difference since the liballoc default will only come into play in no-std situations where people write a custom panic handler, which most likely will not implement unwinding. But still, this seems more consistent.
Cc `@rust-lang/wg-allocators,` https://github.com/rust-lang/rust/issues/66741
Simplify some iterator combinators
Saw a `flat_map().next()` instead of a `find_map()` in some method probe code, so did a quick search for some similar usages.
Ensure `lld` is supported with `download-ci-llvm`
This PR:
- ensures LLD's step in bootstrap's dist, but it's not strictly necessary since dist will already package it when it's present.
- makes bootstrap's `native::LLD` step support using the packaged `ci-llvm/bin/lld`, instead of building it from source (which would most likely not be available today, nor in the future where `download-ci-llvm = if-available` is the default).
If I understand correctly, `--enable-full-tools` will also enable `rust.lld`, and this is why LLD is already packaged today in the `rust-dev` component on the main targets (and why `-Zgcc-ld=lld` does work there).
That means it's likely that this PR will not be able to land before I've reworked and landed #101792: if LLD is available in `download-ci-llvm`, the `needs-rust-lld` tests should start being executed on the x64 macOS test builders, and CI would fail today.
I've tested locally that building with `download-ci-llvm = true` and `lld = true` with the LLVM submodule unregistered was successful, and that `rust-lld` and the various `lld-wrapper`s are present and `-Zgcc-ld=lld` works as well, on a few different platforms:
- `x86_64-unknown-linux-gnu`
- `aarch64-apple-darwin`
- `x86_64-pc-windows-msvc` (with `-Clinker=rust-lld` rather than `-Zgcc-ld=lld`)
- `x86_64-apple-darwin`, with the `MACOSX_DEPLOYMENT_TARGET` workaround for #101653
I don't think we really need to bump the `download-ci-llvm-stamp` in this case, since `./build/$triple/ci-llvm/bin/lld` is present on all the above targets already, but have added it mechanically, and it should probably be removed to avoid unnecessary downloads/churn.
Fixes#98340
Supersedes #100010
Refactoring report_method_error
While working on #105732, I found it's hard to follow this long function,
so I tried to make it shorter.
It's not easy for code reviewing, since so many lines of code changes,
but only the positions are changed.
Generally, I extract two sub-methods from `report_method_error`:
397b66e77b/compiler/rustc_hir_typeck/src/method/suggest.rs (L117)
to `note_candidates_on_method_error`
And this long block:
397b66e77b/compiler/rustc_hir_typeck/src/method/suggest.rs (L263)
to `report_no_match_method_error`.
r? `@compiler-errors`
According to [caniuse], these vendor prefixes aren't needed in any supported
web browsers as defined in [RFC 1985].
* The last version of Chrome that required a vendor prefix was version 47.
The current version is 108.
* Firefox 33 is the last version that required a vendor prefix. The
[current Firefox ESR] is version 102.
* The last version of Safari that required a vendor prefix was version 9.1.
The current version is 16.
* The last version of Safari/iOS that required a vendor prefix was version
9.3. The current version is 16.
* Edge never required vendor prefixes.
* UCAndroid never required vendor prefixes.
[caniuse]: https://caniuse.com/?search=font-feature-settings
[RFC 1985]: https://rust-lang.github.io/rfcs/1985-tiered-browser-support.html
[current Firefox ESR]: https://wiki.mozilla.org/Releases
Rollup of 8 pull requests
Successful merges:
- #104552 (warn newer available version of the x tool)
- #105681 (some fixes/improvements to mir::visit module)
- #106005 (Test the borrowck behavior of if-let guards)
- #106356 (clean: Remove `ctor_kind` from `VariantStruct`.)
- #106365 (Grammar : Missing "is" in format specifier diagnostic)
- #106388 (rustdoc: remove legacy box-sizing CSS)
- #106392 (`has_overflow` only if value is *not* within limit)
- #106402 (Fix dupe word typos)
Failed merges:
r? `@ghost`
`@rustbot` modify labels: rollup
rustdoc: remove legacy box-sizing CSS
According to [caniuse], these vendor prefixes aren't needed in any supported web browsers as defined in [RFC 1985]
* The last version of Chrome that required a vendor prefix was version 9. The current version is 108.
* Firefox 28 is the last version that required a vendor prefix. The [current Firefox ESR] is version 102.
* The last version of Safari that required a vendor prefix was version 5. The current version is 16.
* The last version of Safari/iOS that required a vendor prefix was version 4. The current version is 16.
* Edge never required vendor prefixes.
* UCAndroid never required vendor prefixes.
[caniuse]: https://caniuse.com/?search=box-sizing
[RFC 1985]: https://rust-lang.github.io/rfcs/1985-tiered-browser-support.html
[current Firefox ESR]: https://wiki.mozilla.org/Releases
Test the borrowck behavior of if-let guards
Add some tests to make sure that if-let guards behave the same as if guards with respect to borrow-checking. Most of them are a naive adaptation, replacing an `if` guard with `if let Some(())`.
This includes regression tests for notable issues that arose for if guards (#24535, #27282, #29723, #31287) as suggested in https://github.com/rust-lang/rust/issues/51114#issuecomment-900470419.
cc `@pnkfelix` are there any other tests that you would want to see?
cc tracking issue #51114
warn newer available version of the x tool
This PR adds a check to `tidy` to assert that the installed version of `x` is equal to the version in `src/tools/x/Cargo.toml`. It adds a `--wrapper-version` argument to `x` to determine the version at runtime, .
It does not warn if `x` has not yet been installed, on the assumption that the user isn't interested in using it.
This is needed for json output, but even without that, it increases
performance massivly. On my machine, in reduces the time to check
core.json from 40.190s to 11.333s.
Without verbose:
0:61941:36627 not in index or paths, but refered to at '$.index["0:62007"].inner.for.inner.id' and 12 more
With verbose:
0:10808:27206 not in index or paths, but refered to at '$.index["0:10813"].inner.for.inner.id', '$.index["0:52495"].inner.for.inner.id', '$.index["a:0:2666:215-0:10808:27206"].inner.for.inner.id', '$.index["a:0:2680:223-0:10808:27206"].inner.for.inner.id', '$.index["a:0:2730:7845-0:10808:27206"].inner.for.inner.id', '$.index["a:0:7731:21706-0:10808:27206"].inner.for.inner.id', '$.index["a:0:7732:21705-0:10808:27206"].inner.for.inner.id'
Only specify `--target` by default for `-Zgcc-ld=lld` on wasm
On macOS, it's not yet clear which cases of clang/OS/target/SDK version impact how to find ld/lld/rust-lld. The `--target` argument is not needed on our current targets with a vanilla config to do so, but may be in some cases ?
That is, things look to be different and more subtle than suggested in https://github.com/rust-lang/rust/issues/97402#issuecomment-1147404520.
Specifying this argument unconditionally currently breaks `-Zgcc-ld=lld` on the 10.7+ targets on x64 macOS. Vanilla configs on x64 and aarch64 don't seem to need it to be able to find `rust-lld`.
This fixes#101653 on macOS (I've tried on x64 and aarch64: vanilla installs with the most recent command line tools available for each of the OS versions I have access to, `run-make/issue-71519` passes).
I didn't expect the previous PRs to fail because of the existing tests, but CI does not actually run those tests yet, which explains the regressions. I was hoping to fix those in this PR but it's more involved (building lld is required for the tests to run, llvm/lld is not built on the test builders but on the dist builders, the dist builders don't run tests). This PR is just to unblock current users on macOS who have reported issues, but a regression could happen in the future by mistake until then.
r? `@petrochenkov`
`Split*::as_str` refactor
I've made this patch almost a year ago, so the rename and the behavior change are in one commit, sorry 😅
This fixes#84974, as it's required to make other changes work.
This PR
- Renames `as_str` method of string `Split*` iterators to `remainder` (it seems like the `as_str` name was confusing to users)
- Makes `remainder` return `Option<&str>`, to distinguish between "the iterator is exhausted" and "the tail is empty", this was [required on the tracking issue](https://github.com/rust-lang/rust/issues/77998#issuecomment-832696619)
r? `@m-ou-se`
Only include metadata for non-dynamic libraries in rustc-dev
The actual object code should be linked from librustc_driver.so, which is still included in rustc-dev. This saves on download time and disk usage.
Fixes https://github.com/rust-lang/rust/issues/103538
do not add noalias in return position
`noalias` as a return attribute in LLVM indicates that the returned pointer does not alias anything else that is reachable from the caller, *including things reachable before this function call*. This is clearly not the case with a function like `fn id(Box<T>) -> Box<T>`, so we cannot use this attribute.
Fixes https://github.com/rust-lang/unsafe-code-guidelines/issues/385 (including an actual miscompilation that `@comex` managed to produce).
