Mara Bos [Tue, 4 Jan 2022 15:10:14 +0000 (16:10 +0100)]
Simplify panicking mechanism of thread::scope.
It now panic!()s on its own, rather than resume_unwind'ing the panic
payload from the thread. Using resume_unwind skips the panic_handler,
meaning that the main thread would never have a panic handler run, which
can get confusing.
Matthias Krüger [Thu, 30 Dec 2021 12:12:33 +0000 (13:12 +0100)]
Rollup merge of #91519 - petrochenkov:cratexp2, r=Aaron1011
ast: Avoid aborts on fatal errors thrown from mutable AST visitor
Set the node to some dummy value and rethrow the error instead.
When using the old aborting `visit_clobber` in `InvocationCollector::visit_crate` the next tests abort due to fatal errors:
```
ui\modules\path-invalid-form.rs
ui\modules\path-macro.rs
ui\modules\path-no-file-name.rs
ui\parser\issues\issue-5806.rs
ui\parser\mod_file_with_path_attr.rs
```
Follow up to https://github.com/rust-lang/rust/pull/91313.
bors [Thu, 30 Dec 2021 11:34:24 +0000 (11:34 +0000)]
Auto merge of #89336 - Aaron1011:variance-struct-diag, r=cjgillot
Refactor variance diagnostics to work with more types
Instead of special-casing mutable pointers/references, we
now support general generic types (currently, we handle
`ty::Ref`, `ty::RawPtr`, and `ty::Adt`)
When a `ty::Adt` is involved, we show an additional note
explaining which of the type's generic parameters is
invariant (e.g. the `T` in `Cell<T>`). Currently, we don't
explain *why* a particular generic parameter ends up becoming
invariant. In the general case, this could require printing
a long 'backtrace' of types, so doing this would be
more suitable for a follow-up PR.
We still only handle the case where our variance switches
to `ty::Invariant`.
bors [Thu, 30 Dec 2021 02:53:09 +0000 (02:53 +0000)]
Auto merge of #91125 - eskarn:llvm-passes-plugin-support, r=nagisa
Allow loading LLVM plugins with both legacy and new pass manager
Opening a draft PR to get feedback and start discussion on this feature. There is already a codegen option `passes` which allow giving a list of LLVM pass names, however we currently can't use a LLVM pass plugin (as described here : https://llvm.org/docs/WritingAnLLVMPass.html), the only available passes are the LLVM built-in ones.
The proposed modification would be to add another codegen option `pass-plugins`, which can be set with a list of paths to shared library files. These libraries are loaded using the LLVM function `PassPlugin::Load`, which calls the expected symbol `lvmGetPassPluginInfo`, and register the pipeline parsing and optimization callbacks.
An example usage with a single plugin and 3 passes would look like this in the `.cargo/config`:
```toml
rustflags = [
"-C", "pass-plugins=/tmp/libLLVMPassPlugin",
"-C", "passes=pass1 pass2 pass3",
]
```
This would give the same functionality as the opt LLVM tool directly integrated in rust build system.
Additionally, we can also not specify the `passes` option, and use a plugin which inserts passes in the optimization pipeline, as one could do using clang.
Aaron Hill [Tue, 28 Sep 2021 01:22:44 +0000 (20:22 -0500)]
Refactor variance diagnostics to work with more types
Instead of special-casing mutable pointers/references, we
now support general generic types (currently, we handle
`ty::Ref`, `ty::RawPtr`, and `ty::Adt`)
When a `ty::Adt` is involved, we show an additional note
explaining which of the type's generic parameters is
invariant (e.g. the `T` in `Cell<T>`). Currently, we don't
explain *why* a particular generic parameter ends up becoming
invariant. In the general case, this could require printing
a long 'backtrace' of types, so doing this would be
more suitable for a follow-up PR.
We still only handle the case where our variance switches
to `ty::Invariant`.
bors [Wed, 29 Dec 2021 19:22:33 +0000 (19:22 +0000)]
Auto merge of #92244 - petrochenkov:alltraits, r=cjgillot
rustc_metadata: Encode list of all crate's traits into metadata
While working on https://github.com/rust-lang/rust/pull/88679 I noticed that rustdoc is casually doing something quite expensive, something that is used only for error reporting in rustc - collecting all traits from all crates in the dependency tree.
This PR trades some minor extra time spent by metadata encoder in rustc for major gains for rustdoc (and for rustc runs with errors, which execute the `all_traits` query for better diagnostics).
bors [Wed, 29 Dec 2021 12:07:33 +0000 (12:07 +0000)]
Auto merge of #92397 - matthiaskrgr:rollup-xnfou17, r=matthiaskrgr
Rollup of 7 pull requests
Successful merges:
- #92075 (rustdoc: Only special case struct fields for intra-doc links, not enum variants)
- #92118 (Parse and suggest moving where clauses after equals for type aliases)
- #92237 (Visit expressions in-order when resolving pattern bindings)
- #92340 (rustdoc: Start cleaning up search index generation)
- #92351 (Add long error explanation for E0227)
- #92371 (Remove pretty printer space inside block with only outer attrs)
- #92372 (Print space after formal generic params in fn type)
The pretty printer's `print_formal_generic_params` already prints formal generic params correctly with a space, we just need to call it when printing BareFn types instead of reimplementing the printing incorrectly without a space.
Matthias Krüger [Wed, 29 Dec 2021 09:17:10 +0000 (10:17 +0100)]
Rollup merge of #92237 - compiler-errors:issue-92100, r=cjgillot
Visit expressions in-order when resolving pattern bindings
[edited:] Visit the pattern's sub-expressions before defining any bindings.
Otherwise, we might get into a case where a Lit/Range expression in a pattern has a qpath pointing to a Ident pattern that is defined after it, causing an ICE when lowering to HIR. I have a more detailed explanation in the issue linked.
Matthias Krüger [Wed, 29 Dec 2021 09:17:08 +0000 (10:17 +0100)]
Rollup merge of #92118 - jackh726:type-alias-position-error, r=petrochenkov
Parse and suggest moving where clauses after equals for type aliases
~Mostly the same as #90076, but doesn't make any syntax changes.~ Whether or not we want to land the syntax changes, we should parse the invalid where clause position and suggest moving.
r? `@nikomatsakis`
cc `@petrochenkov` you might have thoughts on implementation
Matthias Krüger [Wed, 29 Dec 2021 09:17:08 +0000 (10:17 +0100)]
Rollup merge of #92075 - jyn514:resolve-cleanup, r=camelid
rustdoc: Only special case struct fields for intra-doc links, not enum variants
Variants are already handled by `resolve_str_path_error`, rustdoc doesn't need to consider them separately. Thanks `@camelid` for catching this!
Eventually I'd like to fix the "combine this with `variant_field`" comment but that needs `resolve_field` to take a `ty_res` parameter to avoid it being super hacky (cc https://github.com/rust-lang/rust/issues/83761#issuecomment-813026026).
bors [Wed, 29 Dec 2021 02:49:34 +0000 (02:49 +0000)]
Auto merge of #92283 - vacuus:print-generic-bounds, r=camelid,GuillaumeGomez
rustdoc: Remove `String` allocation in iteration in `print_generic_bounds`
(I realized only after making the commit that maybe I shouldn't refer to iteration as looping, but it's close enough)
The string representation of a `clean::GenericBound` instance (evaluated [here](https://github.com/rust-lang/rust/blob/master/src/librustdoc/html/format.rs#L397)) is deterministic for a given `self` (the instance), `cx` and `f`, and since `cx` and `f` are constant (as far as I can tell) for a given invocation of `print_generic_bounds`, `self` is the determining factor. Therefore, using the data in `self` shouldn't differ in effect from using its string representation.
Given the totality of the function calls needed to evaluate the string representation as well as the actual allocation, at the very least, this shouldn't negatively affect performance.
Noah Lev [Tue, 28 Dec 2021 21:15:02 +0000 (13:15 -0800)]
Add regression test for #59502
This issue was fixed using a hacky recursion "fuel" argument, but the
issue was never minimized nor was a regression test added. The
underlying bug is still unfixed, so this test should help with fixing it
and removing the `recurse` hack.
bors [Tue, 28 Dec 2021 16:59:56 +0000 (16:59 +0000)]
Auto merge of #92352 - matthiaskrgr:rollup-19fbq7u, r=matthiaskrgr
Rollup of 7 pull requests
Successful merges:
- #92076 (Ignore other `PredicateKind`s in rustdoc auto trait finder)
- #92219 (Remove VCVARS_BAT)
- #92238 (Add a test suite for stringify macro)
- #92330 (Add myself to .mailmap)
- #92333 (Tighten span when suggesting lifetime on path)
- #92335 (Document units for `std::column`)
- #92344 (:arrow_up: rust-analyzer)
Matthias Krüger [Tue, 28 Dec 2021 12:59:25 +0000 (13:59 +0100)]
Rollup merge of #92333 - compiler-errors:elided-lifetime-spans, r=cjgillot
Tighten span when suggesting lifetime on path
This is kind of a hack.
Really the issue here is that we want to suggest the segment's span if the path resolves to something defined outside of the macro, and the macro's span if it resolves to something defined within.. I'll look into seeing if we can do something like that.
Matthias Krüger [Tue, 28 Dec 2021 12:59:24 +0000 (13:59 +0100)]
Rollup merge of #92238 - dtolnay:stringifytest, r=Mark-Simulacrum
Add a test suite for stringify macro
This attempts to cover the behavior of `stringify!` on various interpolated syntax tree nodes.
The pretty printer has a history of unsightly whitespace (double spaces, missing spaces, spaces where there shouldn't be spaces) — #91437, #91562, #91568. There are several such issues left; the test cases that I consider to be currently behaving incorrectly are marked with `// FIXME` in the PR.
Matthias Krüger [Tue, 28 Dec 2021 12:59:23 +0000 (13:59 +0100)]
Rollup merge of #92219 - ehuss:remove-vcvars, r=Mark-Simulacrum
Remove VCVARS_BAT
This environment variable is no longer used. It was used in the original Azure Pipelines configuration (#60777). When GitHub Actions were added (#70190), it was no longer used, and I suspect it was just an oversight while transitioning the configuration.
Matthias Krüger [Tue, 28 Dec 2021 12:59:22 +0000 (13:59 +0100)]
Rollup merge of #92076 - Aaron1011:rustdoc-auto-trait-ignore, r=cjgillot
Ignore other `PredicateKind`s in rustdoc auto trait finder
Fixes #92073
There's not really anything we can do with them, and they're
causing ICEs. I'm not using a wildcard match, as we should check
that any new `PredicateKind`s are handled properly by rustdoc.
bors [Tue, 28 Dec 2021 04:19:01 +0000 (04:19 +0000)]
Auto merge of #92088 - camelid:intra-doc-cleanup, r=Manishearth
intra-doc: Use an enum to represent URL fragments
This is a step in the direction of computing the links more lazily,
which I think will simplify the implementation of intra-doc links.
This will also make it easier to eventually use the actual `Res` for
associated items, enum variants, and fields, rather than their HTML
page's `Res`.
Noah Lev [Tue, 28 Dec 2021 02:59:27 +0000 (18:59 -0800)]
Make `search_index` functions private where possible
Now the only two crate-public items are `build_index` and
`get_index_search_type` (because for some reason the latter is also used
in `formats::cache`).
Noah Lev [Tue, 28 Dec 2021 02:57:07 +0000 (18:57 -0800)]
Move `ExternalLocation` to `clean::types`
It was previously defined in `render::search_index` but wasn't used at
all there. `clean::types` seems like a better fit since that's where
`ExternalCrate` is defined.
Noah Lev [Tue, 28 Dec 2021 02:53:00 +0000 (18:53 -0800)]
rustdoc: Remove some unnecessary `cache` parameters
Based on
https://github.com/rust-lang/rust/pull/80883#issuecomment-774437832.
The `tcx` parameters do seem to be used though, so I only removed the
`cache` parameters.
Noah Lev [Sat, 18 Dec 2021 21:33:01 +0000 (13:33 -0800)]
intra-doc: Use an enum to represent URL fragments
This is a step in the direction of computing the links more lazily,
which I think will simplify the implementation of intra-doc links.
This will also make it easier to eventually use the actual `Res` for
associated items, enum variants, and fields, rather than their HTML
page's `Res`.
bors [Mon, 27 Dec 2021 21:55:52 +0000 (21:55 +0000)]
Auto merge of #92329 - matthiaskrgr:rollup-l3b4fl1, r=matthiaskrgr
Rollup of 7 pull requests
Successful merges:
- #90586 (Relax priv-in-pub lint on generic bounds and where clauses of trait impls.)
- #92112 (Fix the error of checking `base_expr` twice in type_changing_struct_update)
- #92147 (rustc_builtin_macros: make asm mod public for rustfmt)
- #92161 (resolve: Minor miscellaneous cleanups from #89059)
- #92264 (Remove `maybe_uninit_extra` feature from Vec docs)
- #92303 (Add test cases for issue #26186)
- #92307 (Fix minor typos)
Matthias Krüger [Mon, 27 Dec 2021 20:42:29 +0000 (21:42 +0100)]
Rollup merge of #92303 - Patrick-Poitras:issue-26186, r=jackh726
Add test cases for issue #26186
Closes #26186
It seems that issue #26186 has been solved at some point since the issue has been last updated. I've merged together the examples that were supplied by `@Osspial,` `@Mark-Simulacrum,` `@Diggsey,` `@eefriedman` and `@shepmaster,` so that we can add this to the testing suite and prevent these issues from re-occurring.
Matthias Krüger [Mon, 27 Dec 2021 20:42:25 +0000 (21:42 +0100)]
Rollup merge of #90586 - jswrenn:relax-privacy-lints, r=petrochenkov
Relax priv-in-pub lint on generic bounds and where clauses of trait impls.
The priv-in-pub lint is a legacy mechanism of the compiler, supplanted by a reachability-based [type privacy](https://github.com/rust-lang/rfcs/blob/master/text/2145-type-privacy.md) analysis. This PR does **not** relax type privacy; it only relaxes the lint (as proposed by the type privacy RFC) in the case of trait impls.
## Current Behavior
On public trait impls, it's currently an **error** to have a `where` bound constraining a private type with a trait:
```rust
pub trait Trait {}
pub struct Type {}
struct Priv {}
impl Trait for Priv {}
impl Trait for Type
where
Priv: Trait // ERROR
{}
```
...and it's a **warning** to have have a public type constrained by a private trait:
```rust
pub trait Trait {}
pub struct Type {}
impl Trait for Type
where
Pub: Priv // WARNING
{}
```
This lint applies to `where` clauses in other contexts, too; e.g. on free functions:
```rust
struct Priv<T>(T);
pub trait Pub {}
impl<T: Pub> Pub for Priv<T> {}
pub fn function<T>()
where
Priv<T>: Pub // WARNING
{}
```
**These constraints could be relaxed without issue.**
## New Behavior
This lint is relaxed for `where` clauses on trait impls, such that it's okay to have a `where` bound constraining a private type with a trait:
```rust
pub trait Trait {}
pub struct Type {}
struct Priv {}
impl Trait for Priv {}
impl Trait for Type
where
Priv: Trait // OK
{}
```
...and it's okay to have a public type constrained by a private trait:
```rust
pub trait Trait {}
pub struct Type {}
## Rationale
While the priv-in-pub lint is not essential for soundness, it *can* help programmers avoid pitfalls that would make their libraries difficult to use by others. For instance, such a lint *is* useful for free functions; e.g. if a downstream crate tries to call the `function` in the previous snippet in a generic context:
```rust
fn callsite<T>()
where
Priv<T>: Pub // ERROR: omitting this bound is a compile error, but including it is too
{
function::<T>()
}
```
...it cannot do so without repeating `function`'s `where` bound, which we cannot do because `Priv` is out-of-scope. A lint for this case is arguably helpful.
However, this same reasoning **doesn't** hold for trait impls. To call an unconstrained method on a public trait impl with private bounds, you don't need to forward those private bounds, you can forward the public trait:
```rust
mod upstream {
pub trait Trait {
fn method(&self) {}
}
pub struct Type<T>(T);
impl<T> Trait for Type<T>
where
Pub<T>: Priv // WARNING
{}
}
mod downstream {
use super::upstream::*;
fn function<T>(value: Type<T>)
where
Type<T>: Trait // <- no private deets!
{
value.method();
}
}
```
**This PR only eliminates the lint on trait impls.** It leaves it intact for all other contexts, including trait definitions, inherent impls, and function definitions. It doesn't need to exist in those cases either, but I figured I'd first target a case where it's mostly pointless.
## Other Notes
- See discussion [on zulip](https://rust-lang.zulipchat.com/#narrow/stream/213817-t-lang/topic/relax.20priv-in-pub.20lint.20for.20trait.20impl.20.60where.60.20bounds/near/222458397).
- This PR effectively reverts #79291.
bors [Sat, 25 Dec 2021 14:47:12 +0000 (14:47 +0000)]
Auto merge of #92227 - Kobzol:rustdoc-doc-hidden, r=jyn514
Rustdoc: use `is_doc_hidden` method on more places
While profiling `rustdoc`, I noticed that finding out if some item is marked with `#[doc(hidden)]` is relatively hot, so I tried to optimize it.
I noticed that there is already a method called `is_doc_hidden` on `TyCtxt`, but it wasn't used much, so I replaced the manual calls to `attrs(...).has_word(...)` with this method. Just by doing that, perf. was improved locally, although I'm not sure if the semantics of the previous calls and this method are the same?
As another step, I tried to querify `is_doc_hidden`, but I didn't include that here until we see the perf. results from the first commit and until I find whether this change is OK at all :)
bors [Fri, 24 Dec 2021 23:12:14 +0000 (23:12 +0000)]
Auto merge of #92229 - fee1-dead:fix-rustdoc-const-drop, r=dtolnay
Do not display `~const Drop` in rustdoc
Although `T: ~const Drop` is still at an experimental stage, we have already begun inserting these bounds in libstd. This change hides them from rustdoc because 1. `~const` should not be documented in general as it is not yet official syntax; 2. users are not expected to know what `~const Drop` means yet.