bors [Sun, 14 Aug 2022 01:29:58 +0000 (01:29 +0000)]
Auto merge of #100511 - compiler-errors:rollup-vrte4w5, r=compiler-errors
Rollup of 11 pull requests
Successful merges:
- #100355 (rustdoc: Rename ``@has` FILE PATTERN` to ``@hasraw` FILE PATTERN`)
- #100407 (avoid some int2ptr casts in thread_local_key tests)
- #100434 (Fix HIR pretty printing of let else)
- #100438 (Erase regions better in `promote_candidate`)
- #100445 (adapt test for msan message change)
- #100447 (Remove more Clean trait implementations)
- #100464 (Make `[rust] use-lld=true` work on windows)
- #100475 (Give a helpful diagnostic when the next struct field has an attribute)
- #100490 (wf: correctly `shallow_resolve` consts)
- #100501 (nicer Miri backtraces for from_exposed_addr)
- #100509 (merge two test directories that mean the same thing)
Michael Goulet [Sat, 13 Aug 2022 21:10:13 +0000 (14:10 -0700)]
Rollup merge of #100490 - lcnr:wf-consts, r=jackh726
wf: correctly `shallow_resolve` consts
`shallow_resolve` on `InferConst` is always a noop. this is mostly irrelevant as inference vars should already be resolved at most - if not all - call sites. Haven't actually looked too deeply into whether this was a problem.
Michael Goulet [Sat, 13 Aug 2022 21:10:11 +0000 (14:10 -0700)]
Rollup merge of #100464 - khyperia:lld-icf-on-windows, r=jyn514
Make `[rust] use-lld=true` work on windows
Before, it would fail with "error: ignoring unknown argument '-Wl,--icf=all'"
This option was introduced in https://github.com/rust-lang/rust/pull/99062 (well, technically https://github.com/rust-lang/rust/pull/99680)
See zulip thread: https://rust-lang.zulipchat.com/#narrow/stream/182449-t-compiler.2Fhelp/topic/rust-lld.3A.20error.3A.20ignoring.20unknown.20argument.20'-Wl.2C--icf.3Dall'
Michael Goulet [Sat, 13 Aug 2022 21:10:08 +0000 (14:10 -0700)]
Rollup merge of #100445 - krasimirgg:llvm-16-msan, r=tmiasko
adapt test for msan message change
LLVM commit https://github.com/llvm/llvm-project/commit/057cabd997aeaef136e1e14f2ee645bd5bb197dd removed the function from the msan error message. This adapts our test accordingly.
Found via our experimental rust + llvm @ HEAD bot:
https://buildkite.com/llvm-project/rust-llvm-integrate-prototype/builds/12634#018289fe-b0bc-4bab-89b3-fb1d4e38f6db
Noah Lev [Sat, 13 Aug 2022 04:56:16 +0000 (00:56 -0400)]
rustdoc: Fix incorrect usage of `@!has` and `@!matches`
`@!has` (and `@!matches`) with two arguments used to treat the second
argument as a literal string of HTML code. Now, that feature has been
renamed into `@!hasraw` (and `@!matchesraw`), and the arity-2 `@!has`
version is an error.
These uses thought the second argument was being treated as an XPath, as
with the arity-3 version, but in fact was being treated as literal HTML.
Because these were checking for the *absence* of the string, the tests
silently did nothing -- an XPath string won't ever be showing up in the
test's generated HTML!
Noah Lev [Wed, 10 Aug 2022 04:52:04 +0000 (21:52 -0700)]
Rename `@{has,matches}-literal` to `...text`
Reasons:
1. It's shorter.
2. `@matches-literal` seems to contradict itself: a regex is
intrinsically not a literal match, while it is still a textual match.
bors [Sat, 13 Aug 2022 02:41:27 +0000 (02:41 +0000)]
Auto merge of #100341 - andrewpollack:fuchsia-llvm-libunwind, r=tmandry
Use llvm-libunwind="in-tree" for Fuchsia targets
With updates to Fuchsia CI's Zircon libraries #99833, we can introduce `llvm-libunwind="in-tree"` for Fuchsia targets. This PR restores functionality removed from https://github.com/rust-lang/rust/pull/93604#issuecomment-1136515651.
bors [Fri, 12 Aug 2022 16:32:24 +0000 (16:32 +0000)]
Auto merge of #100456 - Dylan-DPC:rollup-fn17z9f, r=Dylan-DPC
Rollup of 9 pull requests
Successful merges:
- #100022 (Optimize thread ID generation)
- #100030 (cleanup code w/ pointers in std a little)
- #100229 (add -Zextra-const-ub-checks to enable more UB checking in const-eval)
- #100247 (Generalize trait object generic param check to aliases.)
- #100255 (Adding more verbose documentation for `std::fmt::Write`)
- #100366 (errors: don't fail on broken primary translations)
- #100396 (Suggest const and static for global variable)
- #100409 (rustdoc: don't generate DOM element for operator)
- #100443 (Add two let else regression tests)
Dylan DPC [Fri, 12 Aug 2022 15:09:18 +0000 (20:39 +0530)]
Rollup merge of #100443 - est31:let_else_regression_tests, r=Mark-Simulacrum
Add two let else regression tests
Adds a regression test for #94176, as it was fixed by #98574 but doesn't have a regression test. The PR also incorporates a commit from #94012 which added a test for an issue discovered in that PR.
Originally they have been part of #99291, but I've moved them out in the hopes of getting them merged more quickly, as that PR is already open since a month, and so that #99291 can focus on the drop order part of things.
Dylan DPC [Fri, 12 Aug 2022 15:09:17 +0000 (20:39 +0530)]
Rollup merge of #100409 - jsha:highlight-lighter, r=GuillaumeGomez
rustdoc: don't generate DOM element for operator
In our source page highlighting, we were generating `<span class="op">` tags for all "operators", including e.g. `<` `>` around generic parameters, `*`, `&`. This contributed significantly to DOM size, but we don't actually style `.op` except in the ayu theme.
Remove the styles for `.op` in ayu, and stop generating the `<span>`s.
This reduces DOM size of an example page[1] from 265,938 HTML elements to 242,165 elements, a 9% reduction.
Dylan DPC [Fri, 12 Aug 2022 15:09:14 +0000 (20:39 +0530)]
Rollup merge of #100366 - davidtwco:translation-never-fail, r=compiler-errors
errors: don't fail on broken primary translations
If a primary bundle doesn't contain a message then the fallback bundle is used. However, if the primary bundle's message is broken (e.g. it refers to a interpolated variable that the compiler isn't providing) then this would just result in a compiler panic. While there aren't any primary bundles right now, this is the type of issue that could come up once translation is further along.
r? ```@compiler-errors``` (since this comes out of a in-person discussion we had at RustConf)
Dylan DPC [Fri, 12 Aug 2022 15:09:12 +0000 (20:39 +0530)]
Rollup merge of #100247 - cjgillot:verify-dyn-trait-alias-defaults, r=lcnr
Generalize trait object generic param check to aliases.
The current algorithm only checks that `Self` does not appear in defaults for traits. This is not sufficient for trait aliases.
This PR moves the check to trait object elaboration, which sees through trait aliases.
bors [Fri, 12 Aug 2022 13:51:26 +0000 (13:51 +0000)]
Auto merge of #100328 - davidtwco:perf-implications, r=nnethercote
passes: load `defined_lib_features` query less
Hopefully addresses the perf regressions from #99212 (see #99905).
Re-structure the stability checks for library features to avoid calling `defined_lib_features` for any more crates than necessary for each of the implications or local feature attributes that need validation.
bors [Fri, 12 Aug 2022 09:49:55 +0000 (09:49 +0000)]
Auto merge of #99624 - vincenzopalazzo:macros/unix_error, r=Amanieu
promote debug_assert to assert when possible and useful
This PR fixed a very old issue https://github.com/rust-lang/rust/issues/94705 to clarify and improve the POSIX error checking, and some of the checks are skipped because can have no benefit, but I'm sure that this can open some interesting discussion.
LLVM commit https://github.com/llvm/llvm-project/commit/057cabd997aeaef136e1e14f2ee645bd5bb197dd removed the function from the msan error message. This adapts our test accordingly.
Found via our experimental rust + llvm @ HEAD bot:
https://buildkite.com/llvm-project/rust-llvm-integrate-prototype/builds/12634#018289fe-b0bc-4bab-89b3-fb1d4e38f6db
bors [Fri, 12 Aug 2022 06:45:35 +0000 (06:45 +0000)]
Auto merge of #100435 - ehuss:update-cargo, r=ehuss
Update cargo
8 commits in ce40690a5e4e315d3dab0aae1eae69d0252c52ac..efd4ca3dc0b89929dc8c5f5c023d25978d76cb61
2022-08-09 22:32:17 +0000 to 2022-08-12 01:28:28 +0000
- Use `std::thread::scope` to replace crossbeam (rust-lang/cargo#10977)
- [docs] Remove extra "in" from `cargo-test.md` (rust-lang/cargo#10978)
- Enable two windows tests (rust-lang/cargo#10930)
- Improve error msg for get target runner (rust-lang/cargo#10968)
- Ensure rustc-echo-wrapper works with an overridden build.target-dir (rust-lang/cargo#10962)
- Switch back to `available_parallelism` (rust-lang/cargo#10969)
- Only override published resolver when the workspace is different (rust-lang/cargo#10961)
- Add `CARGO_LOG` to "Environment variables Cargo reads" (rust-lang/cargo#10967)
bors [Fri, 12 Aug 2022 02:58:51 +0000 (02:58 +0000)]
Auto merge of #99464 - nikic:llvm-15, r=cuviper
Update to LLVM 15
For preliminary testing. Some LLVM 15 compatibility fixes were applied separately in #99512.
Release timeline:
* LLVM 15 branched on Jul 26.
* The final LLVM 15.0.0 release is scheduled for Sep 6.
* Current nightly (1.65.0) is scheduled for Nov 3.
Changes in this PR (apart from the LLVM update):
* Pass `--set llvm.allow-old-toolchain` for many Docker images. LLVM 16 will require GCC >= 7.1, while LLVM 15 still allows older compilers with an option. Specify the option for builders still using GCC 5.4. #95026 updated some of the used toolchains, but not all.
* Use the `+atomics-32` target feature for thumbv6m.
* Explicitly link libatomic when cross-compiling LLVM to 32-bit target.
* Explicitly disable zstd support, to avoid libzstd.so dependency.
In our source page highlighting, we were generating `<span class="op">`
tags for all "operators", including e.g. `<` `>` around generic
parameters, `*`, `&`. This contributed significantly to DOM size, but
we don't actually style `.op` except in the ayu theme.
Remove the styles for `.op` in ayu, and stop generating the `<span>`s.
This reduces DOM size of an example page[1] from 265,938 HTML elements
to 242,165 elements, a 9% reduction.
Matthias Krüger [Thu, 11 Aug 2022 20:53:06 +0000 (22:53 +0200)]
Rollup merge of #100350 - jhpratt:stringify-vis, r=cjgillot
Stringify non-shorthand visibility correctly
This makes `stringify!(pub(in crate))` evaluate to `pub(in crate)` rather than `pub(crate)`, matching the behavior before the `crate` shorthand was removed. Further, this changes `stringify!(pub(in super))` to evaluate to `pub(in super)` rather than the current `pub(super)`. If the latter is not desired (it is _technically_ breaking), it can be undone.
Matthias Krüger [Thu, 11 Aug 2022 20:53:02 +0000 (22:53 +0200)]
Rollup merge of #99992 - jyn514:shell-scripts, r=Mark-Simulacrum
Add `x.sh` and `x.ps1` shell scripts
This is a more ambitious version of https://github.com/rust-lang/rust/pull/98716.
It still changes the x.py shebang back to python3, for compatibility with non-Unix systems,
but also adds alternative entrypoints for systems without `python3` installed.
These scripts will be necessary for the rust entrypoint (#94829), so I see
little downside in adding them early.
I'll update the dev-guide to suggest using these instead of x.py once this is merged.
Matthias Krüger [Thu, 11 Aug 2022 20:53:01 +0000 (22:53 +0200)]
Rollup merge of #99511 - RalfJung:raw_eq, r=wesleywiser
make raw_eq precondition more restrictive
Specifically, don't allow comparing pointers that way. Comparing pointers is subtle because you have to talk about what happens to the provenance.
This matches what [Miri already implements](https://play.rust-lang.org/?version=nightly&mode=debug&edition=2021&gist=9eb1dfb8a61b5a2d4a7cee43df2717af), and all existing users are fine with this.
If raw_eq on pointers is ever desired, we can adjust the intrinsic spec and Miri implementation as needed, but for now that seems just unnecessary. Also, this is a const intrinsic, and in const, comparing pointers this way is *not possible* -- so if we allow the intrinsic to compare pointers in general, we need to impose an extra restrictions saying that in const-context, pointers are *not* okay.
Matthias Krüger [Thu, 11 Aug 2022 20:52:57 +0000 (22:52 +0200)]
Rollup merge of #99337 - jsha:simplify-highlight, r=GuillaumeGomez
rustdoc: simplify highlight.rs
Split render_with_highlighting, which took many optional parameters, into three functions for specific purposes, which each take a smaller number of mostly required parameters.
Remove some plumbing to pass through an "edition" parameter, which was used solely to avoid highlighting some 2021 Edition keywords in non-2021 code.
I've tested a build of std docs before and after, and this does not change the generated HTML at all.
Followup from https://github.com/rust-lang/rust/pull/91264#discussion_r901151101
bors [Thu, 11 Aug 2022 18:05:55 +0000 (18:05 +0000)]
Auto merge of #100416 - Dylan-DPC:rollup-m344lh1, r=Dylan-DPC
Rollup of 11 pull requests
Successful merges:
- #92744 (Check if enum from foreign crate has any non exhaustive variants when attempting a cast)
- #99110 (Determine match_has_guard from candidates instead of looking up thir table again)
- #100184 (Stabilize ptr_const_cast)
- #100192 ( Remove duplicated temporaries creating during box derefs elaboration)
- #100232 (Do not consider method call receiver as an argument in AST.)
- #100287 (linux: Use `pthread_setname_np` instead of `prctl`)
- #100351 (Use `&mut Diagnostic` instead of `&mut DiagnosticBuilder` unless needed)
- #100370 (Remove more Clean trait implementations)
- #100391 (Improve size assertions)
- #100398 (Improve `-Zhir-stats`)
- #100403 (Improve error messages when running rustdoc GUI tests)
Dylan DPC [Thu, 11 Aug 2022 17:17:07 +0000 (22:47 +0530)]
Rollup merge of #100403 - GuillaumeGomez:improve-messages, r=jsha
Improve error messages when running rustdoc GUI tests
There was already a message on how to install `browser-ui-test`, so nothing to be done there. However, we didn't show how to update its version, so the first commit adds it.
Another pain point was how to fix the unexpected crash in `browser-ui-test` (because of a missing `--no-sandbox`, still no idea why it became mandatory a few months ago on some linux distributions...). It now looks like this:
```console
Running 1 rustdoc-gui (8 concurrently) ...
ERROR: puppeteer failed when trying to create a new page. Please try again with `--no-sandbox`
`browser-ui-test` crashed unexpectedly. Please try again with adding `--test-args --no-sandbox` at the end. For example: `x.py test src/test/rustdoc-gui --test-args --no-sandbox`
Build completed unsuccessfully in 0:00:03
```
Thanks to `@jsha` for suggesting these improvements!
Dylan DPC [Thu, 11 Aug 2022 17:17:02 +0000 (22:47 +0530)]
Rollup merge of #100287 - cuviper:no-linux-prctl, r=Mark-Simulacrum
linux: Use `pthread_setname_np` instead of `prctl`
This function is available on Linux since glibc 2.12, musl 1.1.16, and
uClibc 1.0.20. The main advantage over `prctl` is that it properly
represents the pointer argument, rather than a multi-purpose `long`,
so we're better representing strict provenance (#95496).
Dylan DPC [Thu, 11 Aug 2022 17:16:59 +0000 (22:46 +0530)]
Rollup merge of #100192 - tmiasko:rm-duplicated-locals, r=nagisa
Remove duplicated temporaries creating during box derefs elaboration
Temporaries created with `MirPatch::new_temp` will be declared after
patch application. Remove manually created duplicate declarations.
Removing duplicates exposes another issue. Visitor elaborates
terminator twice and attempts to access new, but not yet available,
local declarations. Remove duplicated call to `visit_terminator`.
bors [Thu, 11 Aug 2022 10:47:48 +0000 (10:47 +0000)]
Auto merge of #100315 - compiler-errors:norm-ct-in-proj, r=lcnr
Keep going if normalized projection has unevaluated consts in `QueryNormalizer`
#100312 was the wrong approach, I think this is the right one.
When normalizing a type, if we see that it's a projection, we currently defer to `tcx.normalize_projection_ty`, which normalizes the projections away but doesn't touch the unevaluated constants. So now we just continue to fold the type if it has unevaluated constants so we make sure to evaluate those too, if we can.
Currently it's reported as either `TraitItem` or `ImplItem`. This commit
changes it to `AssocItem`, because having the report match the type name
is (a) consistent with other types, and (b) the trait/impl split isn't
that important here.
This commit:
- Adds a comment explaining which `visit_*` methods should be
implemented.
- Adds and removes some `visit_*` methods accordingly, improving
coverage, and avoiding some double counting.