Rollup merge of #103353 - wesleywiser:fix_lld_thinlto_msvc, r=michaelwoerister
Fix Access Violation when using lld & ThinLTO on windows-msvc
Users report an AV at runtime of the compiled binary when using lld and ThinLTO on windows-msvc. The AV occurs when accessing a static value which is defined in one crate but used in another. Based on the disassembly of the cross-crate use, it appears that the use is not correctly linked with the definition and is instead assigned a garbage pointer value.
If we look at the symbol tables for each crates' obj file, we can see what is happening:
The use of the symbol has the "import" style symbol name but the declaration doesn't generate any symbol with the same name. As a result, linking the files generates a warning from lld:
> rust-lld: warning: bin.obj: locally defined symbol imported: reproducer::memrchr::FN::h612b61ca0e168901 (defined in lib.obj) [LNK4217]
and the symbol reference remains undefined at runtime leading to the AV.
To fix this, we just need to detect that we are performing ThinLTO (and thus, static linking) and omit the `dllimport` attribute on the extern item in LLVM IR.
Rollup merge of #101381 - Urgau:target-mixup-homogenous-floats, r=Amanieu
Test that target feature mix up with homogeneous floats is sound
This pull-request adds a test in `src/test/abi/` that test that target feature mix up with homogeneous floats is sound.
This is basically is ripoff of [src/test/ui/simd/target-feature-mixup.rs](https://github.com/rust-lang/rust/blob/47d1cdb0bcac8e417071ce1929d261efe2399ae2/src/test/ui/simd/target-feature-mixup.rs) but for floats and without `#[repr(simd)]`.
*Extracted from https://github.com/rust-lang/rust/pull/97559 since I don't yet know what to do with that PR.*
Rollup merge of #100508 - BoxyUwU:make_less_things_late_bound, r=nikomatsakis
avoid making substs of type aliases late bound when used as fn args
fixes #47511
fixes #85533
(although I did not know theses issues existed when i was working on this :upside_down_face:)
currently `Alias<...>` is treated the same as `Struct<...>` when deciding if generics should be late bound or early bound but this is not correct as `Alias` might normalize to a projection which does not constrain the generics.
I think this needs more tests before merging
more explanation of PR [here](https://hackmd.io/v44a-QVjTIqqhK9uretyQg?view)
Hackmd inline for future readers:
---
This assumes reader is familiar with the concept of early/late bound lifetimes. There's a section on rustc-dev-guide if not (although i think some details are a bit out of date)
## problem & background
Not all lifetimes on a fn can be late bound:
```rust
fn foo<'a>() -> &'a ();
impl<'a> Fn<()> for FooFnDef {
type Output = &'a (); // uh oh unconstrained lifetime
}
```
so we make make them early bound
```rust
fn foo<'a>() -> &'a ();
impl<'a> Fn<()> for FooFnDef<'a> {// wow look at all that lifetimey
type Output = &'a ();
}
```
(Closures have the same constraint however it is not enforced leading to soundness bugs, [#84385](https://github.com/rust-lang/rust/pull/84385) implements this "downgrading late bound to early bound" for closures)
lifetimes on fn items are only late bound when they are "constrained" by the fn args:
```rust
fn foo<'a>(_: &'a ()) -> &'a ();
// late bound, not present on `FooFnItem`
// vv
impl<'a> Trait<(&'a (),)> for FooFnItem {
type Output = &'a ();
}
// projections do not constrain inputs
fn bar<'a, T: Trait>(_: <T as Trait<'a>>::Assoc) -> &'a (); // early bound
// vv
impl<'a, T: Trait> Fn<(<T as Trait<'a>>::Assoc,)> for BarFnItem<'a, T> {
type Output = &'a ();
}
```
current logic for determining if inputs "constrain" a lifetime works off of HIR so does not normalize aliases. It also assumes that any path with no self type constrains all its substs (i.e. `Foo<'a, u32>` has no self type but `T::Assoc` does). This falls apart for top level type aliases (see linked issues):
```rust
type Alias<'a, T> = <T as Trait<'a>>::Assoc;
// wow look its a path with no self type uwu
// i bet that constrains `'a` so it should be latebound
// vvvvvvvvvvv
fn foo<'a, T: Trait>(_: Alias<'a, T>) -> &'a ();
// `Alias` normalized to make things clearer
// vvvvvvvvvvvvvvvvvvvvvvv
impl<'a, T: Trait> Fn<(<T as Trait<'a>>::Assoc,)> for FooFnDef<T> {
type Output = &'a ();
// oh no `'a` isnt constrained wah wah waaaah *trumbone noises*
// i think, idk what musical instrument that is
}
```
## solution
The PR solves this by having the hir visitor that checks for lifetimes in constraining uses check if the path is a `DefKind::Alias`. If it is we ""normalize"" it by calling `type_of` and walking the returned type. This is a bit hacky as it requires a mapping between the substs on the path in hir, and the generics of the `type Alias<...>` which is on the ty layer.
Alternative solutions may involve calculating the "late boundness" of lifetimes after/during astconv rather than relying on hir at all. We already have code to determine whether a lifetime SHOULD be late bound or not as this is currently how the error for `fn foo<'a, T: Trait>(_: Alias<'a, T>) -> &'a ();` gets emitted.
It is probably not possible to do this right now, late boundness is used by `generics_of` and `gather_explicit_predicates_of` as we currently do not put late bound lifetimes in `Generics`. Although this seems sus to me as the long term goal is to make all generics late bound which would result in `generics_of(function)` being empty? [#103448](https://github.com/rust-lang/rust/pull/103448) places all lifetimes in `Generics` regardless of late boundness so that may be a good step towards making this possible.
bors [Tue, 8 Nov 2022 22:50:12 +0000 (22:50 +0000)]
Auto merge of #104168 - GuillaumeGomez:rollup-tf4edqc, r=GuillaumeGomez
Rollup of 12 pull requests
Successful merges:
- #103928 (Add 'ty_error_with_guaranteed' and 'const_error_with_guaranteed')
- #104027 (Place config.toml in current working directory if config not found)
- #104093 (disable btree size tests on Miri)
- #104097 (run alloc benchmarks in Miri and fix UB)
- #104104 (Add split-debuginfo print option)
- #104109 (rustdoc: Add mutable to the description)
- #104113 (Fix `const_fn_trait_ref_impl`, add test for it)
- #104114 (Fix invalid background-image file name)
- #104132 (fix: lint against lint functions)
- #104139 (Clarify licensing situation of MPSC and SPSC queue)
- #104147 (Remove an address comparison from the parser)
- #104165 (Add llvm-main to triagebot.toml)
Failed merges:
- #104115 (Migrate crate-search element to CSS variables)
Guillaume Gomez [Tue, 8 Nov 2022 19:40:53 +0000 (20:40 +0100)]
Rollup merge of #104147 - WaffleLapkin:don't-compare-ptr-addresses-in-parser-, r=compiler-errors
Remove an address comparison from the parser
Originally this check was added in #68985, as suggested by https://github.com/rust-lang/rust/pull/68985/commits/940f65782cc5df7fecad27b38cc25b6d1eeaf2e8#r376850175. I don't think that this address check is a robust way of making parser more robust.
This code is also extensively tested by [`ui/parser/issues/issue-35813-postfix-after-cast.rs`](https://github.com/rust-lang/rust/blob/57d3c58ed6e0faf89a62411f96c000ffc9fd3937/src/test/ui/parser/issues/issue-35813-postfix-after-cast.rs).
Guillaume Gomez [Tue, 8 Nov 2022 19:40:53 +0000 (20:40 +0100)]
Rollup merge of #104139 - ferrocene:pa-channel-licensing, r=pnkfelix
Clarify licensing situation of MPSC and SPSC queue
Originally, these two files were licensed under the `BSD-2-Clause` license, as they were based off sample code on a blog licensing those snippets under that license:
In 2017 though, the author of that blog agreed to relicense their code under the standard `MIT OR Apache-2.0` license in https://github.com/rust-lang/rust/pull/42149. This PR clarifies the situation in the files by expanding the comment at the top of the file.
Guillaume Gomez [Tue, 8 Nov 2022 19:40:52 +0000 (20:40 +0100)]
Rollup merge of #104132 - Rejyr:rustc_lint-function-lints, r=davidtwco
fix: lint against lint functions
Add `#[rustc_lint_diagnostics]` to lint functions missing them. From [this discussion](https://github.com/rust-lang/rust/pull/101138#issuecomment-1306379999).
Guillaume Gomez [Tue, 8 Nov 2022 19:40:51 +0000 (20:40 +0100)]
Rollup merge of #104109 - yancyribbens:add-mutable-to-comment-for-align-to-mut, r=thomcc
rustdoc: Add mutable to the description
`mutable` is missing from the description. Currently the description for [align_to](https://github.com/rust-lang/rust/blob/master/library/core/src/slice/mod.rs#L3498) is the same as [align_to_mut](https://github.com/rust-lang/rust/blob/master/library/core/src/slice/mod.rs#L3559)
Guillaume Gomez [Tue, 8 Nov 2022 19:40:50 +0000 (20:40 +0100)]
Rollup merge of #104104 - kamirr:master, r=lcnr
Add split-debuginfo print option
This option prints all supported values for `-Csplit-debuginfo=..`, i.e. only stable ones on stable/beta and all of them on nightly/dev.
Motivated by 1.65.0 regression causing builds with the following entry in `Cargo.toml` to fail on Windows:
```toml
[profile.dev]
split-debuginfo = "unpacked"
```
See https://github.com/rust-lang/cargo/pull/11347 for details.
This will lead to closing https://github.com/rust-lang/rust/issues/103976.
Guillaume Gomez [Tue, 8 Nov 2022 19:40:50 +0000 (20:40 +0100)]
Rollup merge of #104097 - RalfJung:miri-alloc-benches, r=thomcc
run alloc benchmarks in Miri and fix UB
Miri since recently has a "fake monotonic clock" that works even with isolation. Its measurements are not very meaningful but it means we can run these benches and check them for UB.
And that's a good thing since there was UB here: fixes https://github.com/rust-lang/rust/issues/104096.
bors [Tue, 8 Nov 2022 19:35:08 +0000 (19:35 +0000)]
Auto merge of #103252 - lcnr:recompute_applicable_impls, r=jackh726
selection failure: recompute applicable impls
The way we currently skip errors for ambiguous trait obligations seems pretty fragile so we get some duplicate errors because of this.
Removing this info from selection errors changes this system to be closer to my image of our new trait solver and is also making it far easier to change overflow errors to be non-fatal :sparkles:
bors [Tue, 8 Nov 2022 14:37:40 +0000 (14:37 +0000)]
Auto merge of #103965 - petrochenkov:effvisperf3, r=oli-obk
resolve: More detailed effective visibility tracking for imports
Per-`DefId` tracking is not enough, due to glob imports in particular, which have a single `DefId` for the whole glob import item.
We need to track this stuff per every introduced name (`NameBinding`).
Also drop `extern` blocks from the effective visibility table, they are nominally private and it doesn't make sense to keep them there.
Later commits add some debug-only invariant checking and optimiaztions to mitigate regressions in https://github.com/rust-lang/rust/pull/103965#issuecomment-1304256445.
This is a bugfix and continuation of https://github.com/rust-lang/rust/pull/102026.
Dylan DPC [Tue, 8 Nov 2022 05:53:53 +0000 (11:23 +0530)]
Rollup merge of #104067 - jeremyd2019:patch-1, r=davidtwco
fix debuginfo for windows_gnullvm_base.rs
These lines (including the FIXME comment) were added to windows_gnu_base.rs in cf2c492ef8c87c049b4e3a62f43c841aafc88cba but windows_gnullvm_base.rs was not updated. This resulted in an error `LLVM ERROR: dwo only supported with ELF and Wasm` attempting to build on aarch64-pc-windows-gnullvm.
See also https://github.com/msys2/MINGW-packages/pull/13921#issuecomment-1304391707
Dylan DPC [Tue, 8 Nov 2022 05:53:52 +0000 (11:23 +0530)]
Rollup merge of #103987 - compiler-errors:no-in_tail_expr, r=eholk
Remove `in_tail_expr` from FnCtxt
Cleans up yet another unneeded member from `FnCtxt`. The `in_tail_expr` condition wasn't even correct -- it was set for true while typechecking the whole fn body.
Dylan DPC [Tue, 8 Nov 2022 05:53:51 +0000 (11:23 +0530)]
Rollup merge of #103651 - Alexendoo:parse-format-unicode-escapes, r=wesleywiser
Fix `rustc_parse_format` spans following escaped utf-8 multibyte chars
Currently too many skips are created for char escapes that are larger than 1 byte when encoded in UTF-8, [playground:](https://play.rust-lang.org/?version=stable&mode=debug&edition=2021&gist=c77a9dc669b69b167271b59ed2c8d88c)
```rust
fn main() {
format!("\u{df}{a}");
format!("\u{211d}{a}");
format!("\u{1f4a3}{a}");
}
```
```
error[[E0425]](https://doc.rust-lang.org/stable/error-index.html#E0425): cannot find value `a` in this scope
--> src/main.rs:2:22
|
2 | format!("\u{df}{a}");
| ^ not found in this scope
error[[E0425]](https://doc.rust-lang.org/stable/error-index.html#E0425): cannot find value `a` in this scope
--> src/main.rs:3:25
|
3 | format!("\u{211d}{a}");
| ^ not found in this scope
error[[E0425]](https://doc.rust-lang.org/stable/error-index.html#E0425): cannot find value `a` in this scope
--> src/main.rs:4:27
|
4 | format!("\u{1f4a3}{a}");
| ^ not found in this scope
```
This reduces the number of skips to account for that
Dylan DPC [Tue, 8 Nov 2022 05:53:50 +0000 (11:23 +0530)]
Rollup merge of #103446 - the8472:tra-array-chunks, r=Mark-Simulacrum
Specialize `iter::ArrayChunks::fold` for TrustedRandomAccess iterators
```
OLD:
test iter::bench_trusted_random_access_chunks ... bench: 368 ns/iter (+/- 4)
NEW:
test iter::bench_trusted_random_access_chunks ... bench: 30 ns/iter (+/- 0)
```
The resulting assembly is similar to #103166 but the specialization kicks in under different (partially overlapping) conditions compared to that PR. They're complementary.
In principle a TRA-based specialization could be applied to all `ArrayChunks` methods, including `next()` as we do for `Zip` but that would have all the same hazards as the Zip specialization. Only doing it for `fold` is far less hazardous. The downside is that it only helps with internal, exhaustive iteration. I.e. `for _ in` or `try_fold` will not benefit.
Note that the regular, `try_fold`-based and the specialized `fold()` impl have observably slightly different behavior. Namely the specialized variant does not fetch the remainder elements from the underlying iterator. We do have a few other places in the standard library where beyond-the-end-of-iteration side-effects are being elided under some circumstances but not others.
Inspired by https://old.reddit.com/r/rust/comments/yaft60/zerocost_iterator_abstractionsnot_so_zerocost/
Rejyr [Tue, 8 Nov 2022 00:23:29 +0000 (19:23 -0500)]
fix: lint against lint functions
fix: lint against the functions `LintContext::{lookup_with_diagnostics,lookup,struct_span_lint,lint}`, `TyCtxt::struct_lint_node`, `LintLevelsBuilder::struct_lint`.
The 8472 [Mon, 7 Nov 2022 23:13:26 +0000 (00:13 +0100)]
document and improve array Guard type
The type is unsafe and now exposed to the whole crate.
Document it properly and add an unsafe method so the
caller can make it visible that something unsafe is happening.
bors [Mon, 7 Nov 2022 10:36:48 +0000 (10:36 +0000)]
Auto merge of #101395 - saethlin:strict-provenance-codegen, r=nikic
Add a codegen test for rust-lang/rust#96152
This is a regression test for https://github.com/rust-lang/rust/issues/96152, it is intended to check that our codegen for a particular strict provenance pattern is always as good as the ptr2int2ptr/provenance-ignoring style.
bors [Mon, 7 Nov 2022 02:15:39 +0000 (02:15 +0000)]
Auto merge of #104083 - JohnTitor:rollup-lo3wbzs, r=JohnTitor
Rollup of 9 pull requests
Successful merges:
- #103885 (rustdoc: various cross-crate reexport fixes)
- #103914 (Make underscore_literal_suffix a hard error.)
- #104045 (Add type_array to BaseTypeMethods)
- #104056 (Vec: IntoIterator signature consistency)
- #104059 (Fix typo in `rustc_middle/lint.rs`)
- #104062 (rustdoc: remove unused CSS `#sidebar-filler`)
- #104065 (Migrate rust logo filter to CSS variables)
- #104066 (LLVM 16: Update RISCV data layout)
- #104074 (rustdoc: Add an example for round that is different from truncate)
jeremyd2019 [Sun, 6 Nov 2022 19:51:12 +0000 (11:51 -0800)]
fix debuginfo for windows_gnullvm_base.rs
These lines (including the FIXME comment) were added to windows_gnu_base.rs in cf2c492ef8c87c049b4e3a62f43c841aafc88cba but windows_gnullvm_base.rs was not updated. This resulted in an error `LLVM ERROR: dwo only supported with ELF and Wasm` attempting to build on aarch64-pc-windows-gnullvm.
Yuki Okushi [Mon, 7 Nov 2022 00:46:29 +0000 (09:46 +0900)]
Rollup merge of #104074 - yancyribbens:add-example-to-round, r=Mark-Simulacrum
rustdoc: Add an example for round that is different from truncate
The current examples for [round](https://github.com/rust-lang/rust/blob/master/library/std/src/f64.rs#L75) would have the same results as the example for [truncate](https://github.com/rust-lang/rust/blob/master/library/std/src/f64.rs#L95). This PR adds one more example to `round` that will have a different result from `truncate`.
Yuki Okushi [Mon, 7 Nov 2022 00:46:26 +0000 (09:46 +0900)]
Rollup merge of #104045 - Ayush1325:type_array, r=nikic
Add type_array to BaseTypeMethods
Moved `type_array` function to `rustc_codegen_ssa::BaseTypeMethods` trait. This allows using normal `alloca` function to create arrays as suggested in https://github.com/rust-lang/rust/pull/104022.
Yuki Okushi [Mon, 7 Nov 2022 00:46:25 +0000 (09:46 +0900)]
Rollup merge of #103885 - fmease:rustdoc-various-cross-crate-reexport-fixes, r=cjgillot,GuillaumeGomez
rustdoc: various cross-crate reexport fixes
Fixes for various smaller cross-crate reexport issues.
The PR is split into several commits for easier review. Will be squashed after approval.
Most notable changes:
* We finally render late-bound lifetimes in the generic parameter list of cross-crate functions & methods.
Previously, we would display the re-export of `pub fn f<'s>(x: &'s str) {}` as `pub fn f(x: &'s str)`
* We now render unnamed parameters of cross-crate functions and function pointers as underscores
since that's exactly what we do for local definitions, too. Mentioned as a bug in #44306.
* From now on, the rendering of cross-crate trait-object types is more correct:
* `for<>` parameter lists (for higher-ranked lifetimes) are now shown
* the return type of `Fn{,Mut,Once}` trait bounds is now displayed
Regarding the last list item, here is a diff for visualization (before vs. after):
The redundant `+ 'static` will be removed in a follow-up PR that will hide trait-object lifetime-bounds if they coincide with [their default](https://doc.rust-lang.org/reference/lifetime-elision.html#default-trait-object-lifetimes) (see [Zulip discussion](https://rust-lang.zulipchat.com/#narrow/stream/266220-rustdoc/topic/clean_middle_ty.3A.20I.20need.20to.20add.20a.20parameter/near/307143097)). `FIXME(fmease)`s were added.
bors [Sun, 6 Nov 2022 17:48:33 +0000 (17:48 +0000)]
Auto merge of #99943 - compiler-errors:tuple-trait, r=jackh726
Implement `std::marker::Tuple`, use it in `extern "rust-call"` and `Fn`-family traits
Implements rust-lang/compiler-team#537
I made a few opinionated decisions in this implementation, specifically:
1. Enforcing `extern "rust-call"` on fn items during wfcheck,
2. Enforcing this for all functions (not just ones that have bodies),
3. Gating this `Tuple` marker trait behind its own feature, instead of grouping it into (e.g.) `unboxed_closures`.
Still needing to be done:
1. Enforce that `extern "rust-call"` `fn`-ptrs are well-formed only if they have 1/2 args and the second one implements `Tuple`. (Doing this would fix ICE in #66696.)
2. Deny all explicit/user `impl`s of the `Tuple` trait, kinda like `Sized`.
3. Fixing `Tuple` trait built-in impl for chalk, so that chalkification tests are un-broken.
Open questions:
1. Does this need t-lang or t-libs signoff?