Initialize BTree nodes directly in the heap
We can avoid any stack-local nodes entirely by using `Box::new_uninit`, and since the nodes are mostly `MaybeUninit` fields, we only need a couple of actual writes before `assume_init`. This should help with the stack overflows in #81444, and may also improve performance in general.
r? `@Mark-Simulacrum`
cc `@ssomers`
This is a pure refactoring split out from #80689.
It represents the most invasive part of that PR, requiring changes in
every caller of `parse_outer_attributes`
In order to eagerly expand `#[cfg]` attributes while preserving the
original `TokenStream`, we need to know the range of tokens that
corresponds to every attribute target. This is accomplished by making
`parse_outer_attributes` return an opaque `AttrWrapper` struct. An
`AttrWrapper` must be converted to a plain `AttrVec` by passing it to
`collect_tokens_trailing_token`. This makes it difficult to accidentally
construct an AST node with attributes without calling `collect_tokens_trailing_token`,
since AST nodes store an `AttrVec`, not an `AttrWrapper`.
As a result, we now call `collect_tokens_trailing_token` for attribute
targets which only support inert attributes, such as generic arguments
and struct fields. Currently, the constructed `LazyTokenStream` is
simply discarded. Future PRs will record the token range corresponding
to the attribute target, allowing those tokens to be removed from an
enclosing `collect_tokens_trailing_token` call if necessary.
Rollup of 8 pull requests
Successful merges:
- #81811 (Fix doc test for Vec::retain(), now passes clippy::eval_order_dependence)
- #81900 (Organize trait test files)
- #81995 (Fix suggestion to introduce explicit lifetime)
- #82031 (Drop an unnecessary intermediate variable)
- #82033 (Refactor `get_word_attr` to return only `Option`)
- #82040 (Add test to prevent src link regression)
- #82041 (Add docs for shared_from_slice From impls)
- #82050 (Added tests to drain an empty vec)
Failed merges:
r? `@ghost`
`@rustbot` modify labels: rollup
Added tests to drain an empty vec
Discovered this kind of issue in an unrelated library.
The author copied the tests from here and AFAIK, there are no tests for this particular case.
https://github.com/LeonineKing1199/minivec/pull/19
Signed-off-by: Hanif Bin Ariffin <hanif.ariffin.4326@gmail.com>
Add docs for shared_from_slice From impls
The advantage of making these docs is mostly in pointing out that these
functions all make new allocations and copy/clone/move the source into them.
These docs are on the function, and not the `impl` block, to avoid showing
the "[+] show undocumented items" button.
CC #51430
Add test to prevent src link regression
Fixes#80502.
This PR is simply about adding a test to prevent a regression.
cc `@bugadani` `@CraftSpider`
r? `@camelid`
Refactor `get_word_attr` to return only `Option`
This commit removes `bool` from the return type of `NestedAttributesExt::get_word_attr` so it will return only `Option<ast::NestedMetaItem>` for less redundancy.
Closes#82016
r? `@jyn514`
Drop an unnecessary intermediate variable
Neither does it shorten the code nor does it provide a helpful name.
`@rustbot` modify labels +C-cleanup +T-compiler
r? `@varkor`
Fix suggestion to introduce explicit lifetime
Addresses #81650
Error message after fix:
```
error[E0311]: the parameter type `T` may not live long enough
--> src/main.rs:25:11
|
24 | fn play_with<T: Animal + Send>(scope: &Scope, animal: T) {
| -- help: consider adding an explicit lifetime bound...: `T: 'a +`
25 | scope.spawn(move |_| {
| ^^^^^
|
note: the parameter type `T` must be valid for the anonymous lifetime #2 defined on the function body at 24:1...
--> src/main.rs:24:1
|
24 | fn play_with<T: Animal + Send>(scope: &Scope, animal: T) {
| ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
note: ...so that the type `[closure@src/main.rs:25:17: 27:6]` will meet its required lifetime bounds
--> src/main.rs:25:11
|
25 | scope.spawn(move |_| {
| ^^^^^
```
Fix doc test for Vec::retain(), now passes clippy::eval_order_dependence
Doc test for Vec::retain() works correctly but is flagged by clippy::eval_order_dependence. Fix avoids the issue by using an iterator instead of an index.
Discovered this kind of issue in an unrelated library.
The author copied the tests from here and AFAIK, there are no tests for this particular case.
Signed-off-by: Hanif Bin Ariffin <hanif.ariffin.4326@gmail.com>
Rollup of 10 pull requests
Successful merges:
- #79775 (Fix injected errors when running doctests on a crate named after a keyword)
- #81012 (Stabilize the partition_point feature)
- #81479 (Allow casting mut array ref to mut ptr)
- #81506 (HWAddressSanitizer support)
- #81741 (Increment `self.index` before calling `Iterator::self.a.__iterator_ge…)
- #81850 (use RWlock when accessing os::env)
- #81911 (GAT/const_generics: Allow with_opt_const_param to return GAT param def_id)
- #82022 (Push a `char` instead of a `str` with len one into a String)
- #82023 (Remove unnecessary lint allow attrs on example)
- #82030 (Use `Iterator::all` instead of open-coding it)
Failed merges:
r? `@ghost`
`@rustbot` modify labels: rollup
The information logged here is of limited general interest, while at the
same times makes it impractical to simply enable logging and share the
resulting logs due to the amount of the output produced.
Reduce log level from info to debug for developer oriented information.
For example, when building cargo, this reduces the amount of logs
generated by `RUSTC_LOG=info cargo build` from 265 MB to 79 MB.
Continuation of changes from 81350.
GAT/const_generics: Allow with_opt_const_param to return GAT param def_id
Fixes#75415Fixes#79666
cc ```@lcnr```
I've absolutely no idea who to r? for this...
use RWlock when accessing os::env
Multiple threads modifying the current process environment is fairly uncommon. Optimize for the more common read case.
r? ````@m-ou-se````
Increment `self.index` before calling `Iterator::self.a.__iterator_ge…
…`t_unchecked` in `Zip` `TrustedRandomAccess` specialization
Otherwise if `Iterator::self.a.__iterator_get_unchecked` panics the
index would not have been incremented yet and another call to
`Iterator::next` would read from the same index again, which is not
allowed according to the API contract of `TrustedRandomAccess` for
`!Clone`.
Fixes https://github.com/rust-lang/rust/issues/81740
Allow casting mut array ref to mut ptr
Allow casting mut array ref to mut ptr
We now allow two new casts:
- mut array reference to mut ptr. Example:
let mut x: [usize; 2] = [0, 0];
let p = &mut x as *mut usize;
We allow casting const array references to const pointers so not
allowing mut references to mut pointers was inconsistent.
- mut array reference to const ptr. Example:
let mut x: [usize; 2] = [0, 0];
let p = &mut x as *const usize;
This was similarly inconsistent as we allow casting mut references to
const pointers.
Existing test 'vector-cast-weirdness' updated to test both cases.
Fixes#24151
The advantage of making these docs is mostly in pointing out that these
functions all make new allocations and copy/clone/move the source into them.
These docs are on the function, and not the `impl` block, to avoid showing
the "[+] show undocumented items" button.
CC #51430
Try fast_reject::simplify_type in coherence before doing full check
This is a reattempt at landing #69010 (by `@jonas-schievink).` The change adds a fast path for coherence checking to see if there's no way for types to unify since full coherence checking can be somewhat expensive.
This has big effects on code generated by the [`windows`](https://github.com/microsoft/windows-rs) which in some cases spends as much as 20% of compilation time in the `specialization_graph_of` query. In local benchmarks this took a compilation that previously took ~500 seconds down to ~380 seconds.
This is surely not going to make a difference on much smaller crates, so the question is whether it will have a negative impact. #69010 was closed because some of the perf suite crates did show small regressions.
Additional discussion of this issue is happening [here](https://rust-lang.zulipchat.com/#narrow/stream/247081-t-compiler.2Fperformance/topic/windows-rs.20perf).
