]> git.lizzy.rs Git - rust.git/log
rust.git
2 years agoExpose iter::ByRefSized as unstable feature and use it
Paolo Barbolini [Sat, 11 Jun 2022 22:55:09 +0000 (00:55 +0200)]
Expose iter::ByRefSized as unstable feature and use it

2 years agoAdd VecDeque::extend from TrustedLen specialization
Paolo Barbolini [Sat, 11 Jun 2022 22:54:20 +0000 (00:54 +0200)]
Add VecDeque::extend from TrustedLen specialization

2 years agoAdd VecDeque::extend TrustedLen benchmark
Paolo Barbolini [Sat, 11 Jun 2022 20:53:46 +0000 (22:53 +0200)]
Add VecDeque::extend TrustedLen benchmark

2 years agoAuto merge of #97892 - klensy:fix-spaces, r=oli-obk
bors [Fri, 17 Jun 2022 17:30:16 +0000 (17:30 +0000)]
Auto merge of #97892 - klensy:fix-spaces, r=oli-obk

diagnostics: remove trailing spaces

Remove few occurrences of trailing spaces and drive by fix of needless alloc of const string.

2 years agoAuto merge of #98097 - lqd:const-alloc-hash, r=oli-obk
bors [Fri, 17 Jun 2022 15:10:04 +0000 (15:10 +0000)]
Auto merge of #98097 - lqd:const-alloc-hash, r=oli-obk

ctfe: limit hashing of big const allocations when interning

Const allocations are only hashed for interning. However, they can be large, making the hashing expensive especially since it uses `FxHash`: it's better suited to short keys, not potentially big buffers like the actual bytes of allocation and the associated 1/8th sized `InitMask`.

We can partially hash these fields when they're large, hashing the length, and head and tail of these buffers, to
limit possible collisions while avoiding most of the hashing work.

r? `@ghost`

2 years agoAuto merge of #98193 - Dylan-DPC:rollup-5gihjij, r=Dylan-DPC
bors [Fri, 17 Jun 2022 12:41:25 +0000 (12:41 +0000)]
Auto merge of #98193 - Dylan-DPC:rollup-5gihjij, r=Dylan-DPC

Rollup of 5 pull requests

Successful merges:

 - #95392 (std: Stabilize feature try_reserve_2 )
 - #97798 (Hide irrelevant lines in suggestions to allow for suggestions that are far from each other to be shown)
 - #97844 (Windows: No panic if function not (yet) available)
 - #98013 (Subtype FRU fields first in `type_changing_struct_update`)
 - #98191 (Remove the rest of unnecessary `to_string`)

Failed merges:

r? `@ghost`
`@rustbot` modify labels: rollup

2 years agoRollup merge of #98191 - TaKO8Ki:remove-rest-of-unnecessary-to-string, r=Dylan-DPC
Dylan DPC [Fri, 17 Jun 2022 10:21:50 +0000 (12:21 +0200)]
Rollup merge of #98191 - TaKO8Ki:remove-rest-of-unnecessary-to-string, r=Dylan-DPC

Remove the rest of unnecessary `to_string`

I removed most of unnecessary `to_string` in #98043. This patch removes the rest of them I missed.

2 years agoRollup merge of #98013 - compiler-errors:guide-inference-2, r=lcnr
Dylan DPC [Fri, 17 Jun 2022 10:21:49 +0000 (12:21 +0200)]
Rollup merge of #98013 - compiler-errors:guide-inference-2, r=lcnr

Subtype FRU fields first in `type_changing_struct_update`

So this fixes a subtle bug that `type_changing_struct_update` introduced, where it'll no longer coerce the base expr correctly. I actually think this code is easier to understand now, too.

r? `@lcnr` since you reviewed the last one

2 years agoRollup merge of #97844 - ChrisDenton:dont-panic, r=JohnTitor
Dylan DPC [Fri, 17 Jun 2022 10:21:48 +0000 (12:21 +0200)]
Rollup merge of #97844 - ChrisDenton:dont-panic, r=JohnTitor

Windows: No panic if function not (yet) available

In some situations (e.g. #97814) it is possible for required functions to be called before they've had a chance to be loaded. Therefore, we make it possible to recover from this situation simply by looking at error codes.

`@rustbot` label +O-windows

2 years agoRollup merge of #97798 - WaffleLapkin:allow_for_suggestions_that_are_quite_far_away_f...
Dylan DPC [Fri, 17 Jun 2022 10:21:48 +0000 (12:21 +0200)]
Rollup merge of #97798 - WaffleLapkin:allow_for_suggestions_that_are_quite_far_away_from_each_other, r=estebank

Hide irrelevant lines in suggestions to allow for suggestions that are far from each other to be shown

This is an attempt to fix suggestions one part of which is 6 lines or more far from the first. I've noticed "the problem" (of not showing some parts of the suggestion) here: https://github.com/rust-lang/rust/pull/97759#discussion_r889689230.

I'm not sure about the implementation (this big closure is just bad and makes already complicated code even more so), but I want to at least discuss the result.

Here is an example of how this changes the output:

Before:
```text
help: consider enclosing expression in a block
  |
3 ~     'l: { match () { () => break 'l,
4 |
5 |
6 |
7 |
8 |
...
```

After:
```text
help: consider enclosing expression in a block
  |
3 ~     'l: { match () { () => break 'l,
4 |
...
31|
32~ } };
  |
```

r? `@estebank`
`@rustbot` label +A-diagnostics +A-suggestion-diagnostics

2 years agoRollup merge of #95392 - Xuanwo:stablize_try_reserve_2, r=dtolnay
Dylan DPC [Fri, 17 Jun 2022 10:21:46 +0000 (12:21 +0200)]
Rollup merge of #95392 - Xuanwo:stablize_try_reserve_2, r=dtolnay

std: Stabilize feature try_reserve_2

This PR intends to stabilize feature `try_reserve_2`, closes https://github.com/rust-lang/rust/issues/91789

This PR will also replace the previous PR: https://github.com/rust-lang/rust/pull/95139

2 years agoAuto merge of #98106 - cjgillot:split-definitions, r=michaelwoerister
bors [Fri, 17 Jun 2022 10:00:11 +0000 (10:00 +0000)]
Auto merge of #98106 - cjgillot:split-definitions, r=michaelwoerister

Split up `Definitions` and `ResolverAstLowering`.

Split off https://github.com/rust-lang/rust/pull/95573

r? `@michaelwoerister`

2 years agoremove the rest of unnecessary `to_string`
Takayuki Maeda [Fri, 17 Jun 2022 09:48:09 +0000 (18:48 +0900)]
remove the rest of unnecessary `to_string`

2 years agoAuto merge of #97863 - JakobDegen:bitset-choice, r=nnethercote
bors [Fri, 17 Jun 2022 07:35:22 +0000 (07:35 +0000)]
Auto merge of #97863 - JakobDegen:bitset-choice, r=nnethercote

`BitSet` related perf improvements

This commit makes two changes:
 1. Changes `MaybeLiveLocals` to use `ChunkedBitSet`
 2. Overrides the `fold` method for the iterator for `ChunkedBitSet`

I have local benchmarks verifying that each of these changes individually yield significant perf improvements to #96451 . I'm hoping this will be true outside of that context too. If that is not the case, I'll try to gate things on where they help as needed

r? `@nnethercote` who I believe was working on closely related things, cc `@tmiasko` because of the destprop pr

2 years agoAuto merge of #98160 - nnethercote:mv-finish-out-of-Encoder, r=bjorn3
bors [Fri, 17 Jun 2022 05:13:07 +0000 (05:13 +0000)]
Auto merge of #98160 - nnethercote:mv-finish-out-of-Encoder, r=bjorn3

Move `finish` out of the `Encoder` trait.

This simplifies things, but requires making `CacheEncoder` non-generic.

(This was previously merged as commit 4 in #94732 and then was reverted
in #97905 because it caused a perf regression.)

r? `@ghost`

2 years agoSubtype FRU fields first in type_changing_struct_update
Michael Goulet [Sat, 11 Jun 2022 20:21:38 +0000 (13:21 -0700)]
Subtype FRU fields first in type_changing_struct_update

2 years agoAuto merge of #98143 - cuviper:futex-rwlock-inline, r=thomcc
bors [Fri, 17 Jun 2022 02:32:14 +0000 (02:32 +0000)]
Auto merge of #98143 - cuviper:futex-rwlock-inline, r=thomcc

Add `#[inline]` to small fns of futex `RwLock`

The important methods like `read` and `write` were already inlined,
which can propagate all the way to inlining in user code, but these
small state functions were left behind as normal calls. They should
almost always be inlined as well, as they're just a few instructions.

2 years agoAuto merge of #98181 - JohnTitor:rollup-65ztwnz, r=JohnTitor
bors [Thu, 16 Jun 2022 23:50:20 +0000 (23:50 +0000)]
Auto merge of #98181 - JohnTitor:rollup-65ztwnz, r=JohnTitor

Rollup of 5 pull requests

Successful merges:

 - #97377 (Do not suggest adding semicolon/changing delimiters for macros in item position that originates in macros)
 - #97675 (Make `std::mem::needs_drop` accept `?Sized`)
 - #98118 (Test NLL fix of bad lifetime inference for reference captured in closure.)
 - #98166 (Add rustdoc-json regression test for #98009)
 - #98169 (Keyword docs: Link to wikipedia article for dynamic dispatch)

Failed merges:

r? `@ghost`
`@rustbot` modify labels: rollup

2 years agoRollup merge of #98169 - pierwill:dyn-disp, r=JohnTitor
Yuki Okushi [Thu, 16 Jun 2022 22:16:58 +0000 (07:16 +0900)]
Rollup merge of #98169 - pierwill:dyn-disp, r=JohnTitor

Keyword docs: Link to wikipedia article for dynamic dispatch

2 years agoRollup merge of #98166 - GuillaumeGomez:regression-test-98009, r=matthiaskrgr
Yuki Okushi [Thu, 16 Jun 2022 22:16:57 +0000 (07:16 +0900)]
Rollup merge of #98166 - GuillaumeGomez:regression-test-98009, r=matthiaskrgr

Add rustdoc-json regression test for #98009

Fixes #98009.

There was apparently nothing to be done...

cc `@matthiaskrgr`
r? `@notriddle`

2 years agoRollup merge of #98118 - steffahn:scoped-threads-nll-test, r=m-ou-se
Yuki Okushi [Thu, 16 Jun 2022 22:16:56 +0000 (07:16 +0900)]
Rollup merge of #98118 - steffahn:scoped-threads-nll-test, r=m-ou-se

Test NLL fix of bad lifetime inference for reference captured in closure.

This came up as a use-case for `thread::scope` API that only compiles successfully since `feature(nll)` got stabilized recently.

Closes #93203 which had been re-opened for tracking this very test case to be added.

2 years agoRollup merge of #97675 - nvzqz:unsized-needs-drop, r=dtolnay
Yuki Okushi [Thu, 16 Jun 2022 22:16:55 +0000 (07:16 +0900)]
Rollup merge of #97675 - nvzqz:unsized-needs-drop, r=dtolnay

Make `std::mem::needs_drop` accept `?Sized`

This change attempts to make `needs_drop` work with types like `[u8]` and `str`.

This enables code in types like `Arc<T>` that was not possible before, such as https://github.com/rust-lang/rust/pull/97676.

2 years agoRollup merge of #97377 - ChayimFriedman2:issue-91800, r=estebank
Yuki Okushi [Thu, 16 Jun 2022 22:16:54 +0000 (07:16 +0900)]
Rollup merge of #97377 - ChayimFriedman2:issue-91800, r=estebank

Do not suggest adding semicolon/changing delimiters for macros in item position that originates in macros

Fixes #91800.

2 years agoAuto merge of #97936 - nnethercote:compile-unicode_normalization-faster, r=oli-obk
bors [Thu, 16 Jun 2022 21:09:30 +0000 (21:09 +0000)]
Auto merge of #97936 - nnethercote:compile-unicode_normalization-faster, r=oli-obk

Compile `unicode-normalization` faster

Various optimizations and cleanups aimed at improving compilation of `unicode-normalization`, which is notable for having several very large `match`es with many char ranges.

Best reviewed one commit at a time.

r? `@oli-obk`

2 years agoadjust const alloc interning partial hash comments
Rémy Rakic [Thu, 16 Jun 2022 18:04:52 +0000 (20:04 +0200)]
adjust const alloc interning partial hash comments

2 years agoAuto merge of #98173 - flip1995:clippyup, r=Manishearth
bors [Thu, 16 Jun 2022 18:13:02 +0000 (18:13 +0000)]
Auto merge of #98173 - flip1995:clippyup, r=Manishearth

Update Clippy

r? `@Manishearth`

2 years agoctfe: limit hashing of big const allocations when interning
Rémy Rakic [Tue, 14 Jun 2022 14:03:26 +0000 (16:03 +0200)]
ctfe: limit hashing of big const allocations when interning

Big const allocations hash a large amount of data for interning:
the whole bytes buffer, and the 1/8th sized initmask, with FxHash.
This hash function is made for shorter keys.

This only hashes the length, and head and tail of these buffers, to
limit possible collisions while avoiding most of the hashing work.

2 years agoUpdate Cargo.lock
flip1995 [Thu, 16 Jun 2022 15:39:29 +0000 (17:39 +0200)]
Update Cargo.lock

2 years agoMerge commit 'd7b5cbf065b88830ca519adcb73fad4c0d24b1c7' into clippyup
flip1995 [Thu, 16 Jun 2022 15:39:06 +0000 (17:39 +0200)]
Merge commit 'd7b5cbf065b88830ca519adcb73fad4c0d24b1c7' into clippyup

2 years agoKeyword docs: Link to wikipedia article for dynamic dispatch
pierwill [Thu, 16 Jun 2022 14:58:57 +0000 (09:58 -0500)]
Keyword docs: Link to wikipedia article for dynamic dispatch

2 years agoAuto merge of #9007 - flip1995:rustup, r=flip1995
bors [Thu, 16 Jun 2022 14:07:23 +0000 (14:07 +0000)]
Auto merge of #9007 - flip1995:rustup, r=flip1995

Rustup

r? `@ghost`

changelog: none

2 years agoBump nightly version -> 2022-06-16
flip1995 [Thu, 16 Jun 2022 14:04:12 +0000 (16:04 +0200)]
Bump nightly version -> 2022-06-16

2 years agoMerge remote-tracking branch 'upstream/master' into rustup
flip1995 [Thu, 16 Jun 2022 13:33:15 +0000 (15:33 +0200)]
Merge remote-tracking branch 'upstream/master' into rustup

2 years agobless clippy ui tests
Maybe Waffle [Thu, 16 Jun 2022 14:00:32 +0000 (18:00 +0400)]
bless clippy ui tests

2 years agoAdd back MAX_SUGGESTION_HIGHLIGHT_LINES so clippy is happy & buildable
Maybe Waffle [Mon, 13 Jun 2022 14:28:31 +0000 (18:28 +0400)]
Add back MAX_SUGGESTION_HIGHLIGHT_LINES so clippy is happy & buildable

2 years agoTry to clean up code...
Maybe Waffle [Mon, 13 Jun 2022 11:22:26 +0000 (15:22 +0400)]
Try to clean up code...

I'm not sure if I succeeded

2 years ago--bless ui
Maybe Waffle [Mon, 6 Jun 2022 15:26:50 +0000 (19:26 +0400)]
--bless ui

2 years agoImprove suggestions when its parts are far from each other
Maybe Waffle [Mon, 6 Jun 2022 14:04:42 +0000 (18:04 +0400)]
Improve suggestions when its parts are far from each other

Previously we only show at most 6 lines of suggestions and, if the
suggestions are more than 6 lines apart, we've just showed ... at the
end. This is probably fine, but quite confusing in my opinion.

This commit is an attempt to show ... in places where there is nothing
to suggest instead, for example:

Before:
```text
help: consider enclosing expression in a block
  |
3 ~     'l: { match () { () => break 'l,
4 |
5 |
6 |
7 |
8 |
...
```

After:
```text
help: consider enclosing expression in a block
  |
3 ~     'l: { match () { () => break 'l,
4 |
...
31|
32~ } };
  |
```

2 years agoFix a typo
Maybe Waffle [Mon, 6 Jun 2022 13:42:53 +0000 (17:42 +0400)]
Fix a typo

!(a & b) = !a | !b

2 years agobless clippy tests
klensy [Wed, 15 Jun 2022 11:15:54 +0000 (14:15 +0300)]
bless clippy tests

2 years agoAuto merge of #9005 - flip1995:book_ci_2, r=xFrednet
bors [Thu, 16 Jun 2022 12:49:20 +0000 (12:49 +0000)]
Auto merge of #9005 - flip1995:book_ci_2, r=xFrednet

Build mdbook in remark workflow

r? `@xFrednet`

Supersedes #8959

changelog: none

2 years agoAdd rustdonc-json regression test for #98009
Guillaume Gomez [Thu, 16 Jun 2022 12:13:57 +0000 (14:13 +0200)]
Add rustdonc-json regression test for #98009

2 years agoBuild mdbook in remark workflow
flip1995 [Thu, 16 Jun 2022 12:10:35 +0000 (14:10 +0200)]
Build mdbook in remark workflow

This is just to ensure that the book builds all time to not get in trouble when syncing with rust-lang/rust

2 years agobless rustdoc-ui tests
klensy [Wed, 8 Jun 2022 19:33:47 +0000 (22:33 +0300)]
bless rustdoc-ui tests

2 years agodrive by fix needless allocation of const string number
klensy [Wed, 8 Jun 2022 18:23:39 +0000 (21:23 +0300)]
drive by fix needless allocation of const string number

2 years ago fix one more case of trailing space
klensy [Wed, 8 Jun 2022 18:07:59 +0000 (21:07 +0300)]
 fix one more case of trailing space

2 years agodiagnostics: fix trailing space
klensy [Wed, 8 Jun 2022 17:34:57 +0000 (20:34 +0300)]
diagnostics: fix trailing space

2 years agoAuto merge of #97842 - notriddle:notriddle/tuple-docs, r=jsha,GuillaumeGomez
bors [Thu, 16 Jun 2022 11:13:30 +0000 (11:13 +0000)]
Auto merge of #97842 - notriddle:notriddle/tuple-docs, r=jsha,GuillaumeGomez

Improve the tuple and unit trait docs

* Reduce duplicate impls; show only the `(T,)` and include a sentence saying that there exists ones up to twelve of them.
* Show `Copy` and `Clone`.
* Show auto traits like `Send` and `Sync`, and blanket impls like `Any`.

Here's the new version:

* <https://notriddle.com/notriddle-rustdoc-test/std/primitive.tuple.html>
* <https://notriddle.com/notriddle-rustdoc-test/std/primitive.unit.html>

2 years agoAuto merge of #98161 - matthiaskrgr:rollup-8csenk9, r=matthiaskrgr
bors [Thu, 16 Jun 2022 07:45:03 +0000 (07:45 +0000)]
Auto merge of #98161 - matthiaskrgr:rollup-8csenk9, r=matthiaskrgr

Rollup of 4 pull requests

Successful merges:

 - #97757 (Support lint expectations for `--force-warn` lints (RFC 2383))
 - #98125 (Entry and_modify doc)
 - #98137 (debuginfo: Fix NatVis for Rc and Arc with unsized pointees.)
 - #98147 (Make #[cfg(bootstrap)] not error in proc macros on later stages )

Failed merges:

r? `@ghost`
`@rustbot` modify labels: rollup

2 years agoRollup merge of #98147 - est31:bootstrap_cfg_flag, r=Mark-Simulacrum
Matthias Krüger [Thu, 16 Jun 2022 07:10:23 +0000 (09:10 +0200)]
Rollup merge of #98147 - est31:bootstrap_cfg_flag, r=Mark-Simulacrum

Make #[cfg(bootstrap)] not error in proc macros on later stages

As was discovered in https://github.com/rust-lang/rust/pull/93628#issuecomment-1154697627,
adding #[cfg(bootstrap)] to a rust-internal proc macro crate
would yield an unexpected cfg name error, at least on later
stages wher the bootstrap cfg arg wasn't set.

rustc already passes arguments to mark bootstrap as expected,
however the means of delivery through the RUSTFLAGS env var
is unable to reach proc macro crates, as described
in the issue linked in the code this commit touches.

This wouldn't be an issue for cfg args that get passed through
RUSTFLAGS, as they would never become *active* either, so
any usage of one of these flags in a proc macro's code would
legitimately yield a lint warning. But since dc30258,
rust takes extra measures to pass --cfg=bootstrap even in
proc macros, by passing it via the wrapper. Thus, we need
to send the flags to mark bootstrap as expected also from the
wrapper, so that #[cfg(bootstrap)] also works from proc macros.

I want to thank `Urgau` and `jplatte` for helping me find the cause of this. :heart:

2 years agoRollup merge of #98137 - michaelwoerister:fix-unsized-rc-arc-natvis, r=wesleywiser
Matthias Krüger [Thu, 16 Jun 2022 07:10:22 +0000 (09:10 +0200)]
Rollup merge of #98137 - michaelwoerister:fix-unsized-rc-arc-natvis, r=wesleywiser

debuginfo: Fix NatVis for Rc and Arc with unsized pointees.

Currently, the NatVis for `Rc<T>` and `Arc<T>` does not support unsized `T`. For both `Rc<T>` and `Rc<dyn SomeTrait>` the visualizers fail:

```txt
    [Reference count] : -> must be used on pointers and . on structures
    [Weak reference count] : -> must be used on pointers and . on structures
```

This PR fixes the visualizers. For slices we can even give show the elements, so one now gets something like:

```txt
slice_rc         : { len=3 }
    [Length]         : 3
    [Reference count] : 41
    [Weak reference count] : 2
    [0]              : 1
    [1]              : 2
    [2]              : 3
```

r? `@wesleywiser`

2 years agoRollup merge of #98125 - KarlWithK:entry_add_modify_doc, r=Dylan-DPC
Matthias Krüger [Thu, 16 Jun 2022 07:10:21 +0000 (09:10 +0200)]
Rollup merge of #98125 - KarlWithK:entry_add_modify_doc, r=Dylan-DPC

Entry and_modify doc

This PR modifies the documentation for [HashMap](https://doc.rust-lang.org/std/collections/struct.HashMap.html#) and [BTreeMap](https://doc.rust-lang.org/std/collections/struct.BTreeMap.html#) by introducing examples for `and_modify`. `and_modify` is a function that tends to give more idiomatic rust code when dealing with these data structures -- yet it lacked examples and was hidden away. This PR adds that and addresses #98122.

I've made some choices which I tried to explain in my commits. This is my first time contributing to rust, so hopefully, I made the right choices.

2 years agoRollup merge of #97757 - xFrednet:rfc-2383-expect-with-force-warn, r=wesleywiser...
Matthias Krüger [Thu, 16 Jun 2022 07:10:20 +0000 (09:10 +0200)]
Rollup merge of #97757 - xFrednet:rfc-2383-expect-with-force-warn, r=wesleywiser,flip1995

Support lint expectations for `--force-warn` lints (RFC 2383)

Rustc has a `--force-warn` flag, which overrides lint level attributes and forces the diagnostics to always be warn. This means, that for lint expectations, the diagnostic can't be suppressed as usual. This also means that the expectation would not be fulfilled, even if a lint had been triggered in the expected scope.

This PR now also tracks the expectation ID in the `ForceWarn` level. I've also made some minor adjustments, to possibly catch more bugs and make the whole implementation more robust.

This will probably conflict with https://github.com/rust-lang/rust/pull/97718. That PR should ideally be reviewed and merged first. The conflict itself will be trivial to fix.

---

r? `@wesleywiser`

cc: `@flip1995` since you've helped with the initial review and also discussed this topic with me. :upside_down_face:

Follow-up of: https://github.com/rust-lang/rust/pull/87835

Issue: https://github.com/rust-lang/rust/issues/85549

Yeah, and that's it.

2 years agoMove `finish` out of the `Encoder` trait.
Nicholas Nethercote [Thu, 16 Jun 2022 06:00:25 +0000 (16:00 +1000)]
Move `finish` out of the `Encoder` trait.

This simplifies things, but requires making `CacheEncoder` non-generic.

(This was previously merged as commit 4 in #94732 and then was reverted
in #97905 because it caused a perf regression.)

2 years agoSupport lint expectations for `--force-warn` lints (RFC 2383)
xFrednet [Sun, 5 Jun 2022 10:33:45 +0000 (12:33 +0200)]
Support lint expectations for `--force-warn` lints (RFC 2383)

2 years agoAuto merge of #98108 - SpriteOvO:doc_auto_cfg-feature-rmv-fix, r=notriddle,GuillaumeGomez
bors [Thu, 16 Jun 2022 05:02:10 +0000 (05:02 +0000)]
Auto merge of #98108 - SpriteOvO:doc_auto_cfg-feature-rmv-fix, r=notriddle,GuillaumeGomez

Rustdoc: Fix stab disappearing and exclude cfg "doc" and "doctest"

Fixes #98065 Context: https://github.com/rust-lang/rust/issues/43781#issuecomment-1154226733

r? `@GuillaumeGomez`

2 years agoDo not suggest adding semicolon/changing delimiters for macros in item position that...
Chayim Refael Friedman [Tue, 24 May 2022 23:41:34 +0000 (23:41 +0000)]
Do not suggest adding semicolon/changing delimiters for macros in item position that originates in macros

2 years agoAuto merge of #98103 - exrook:btreemap-alloc, r=Amanieu
bors [Thu, 16 Jun 2022 02:15:55 +0000 (02:15 +0000)]
Auto merge of #98103 - exrook:btreemap-alloc, r=Amanieu

BTreeMap: Support custom allocators (v1.5)

Related:
https://github.com/rust-lang/wg-allocators/issues/7
https://github.com/TimDiekmann/alloc-wg

Blocked on:
~~#77187~~
~~#78459~~
~~#95036~~

previous: #77438

2 years agostd: Stabilize feature try_reserve_2
Xuanwo [Mon, 28 Mar 2022 02:01:59 +0000 (10:01 +0800)]
std: Stabilize feature try_reserve_2

Signed-off-by: Xuanwo <github@xuanwo.io>
2 years agocompare_const_vals: add a special case for certain ranges.
Nicholas Nethercote [Thu, 9 Jun 2022 00:23:39 +0000 (10:23 +1000)]
compare_const_vals: add a special case for certain ranges.

This commit removes the `a == b` early return, which isn't useful in
practice, and replaces it with one that helps matches with many ranges,
including char ranges.

2 years agocompare_const_vals: Use infallible evaluation.
Nicholas Nethercote [Thu, 9 Jun 2022 01:15:03 +0000 (11:15 +1000)]
compare_const_vals: Use infallible evaluation.

Because these evaluations can never fail.

2 years agoRemove `ty` arg from `compare_const_vals`.
Nicholas Nethercote [Wed, 8 Jun 2022 23:46:23 +0000 (09:46 +1000)]
Remove `ty` arg from `compare_const_vals`.

It's now only used in no-longer-interesting assertion.

2 years agoRemove `from_bool` closure.
Nicholas Nethercote [Wed, 8 Jun 2022 23:38:30 +0000 (09:38 +1000)]
Remove `from_bool` closure.

The code is clearer and simpler without it. Note that the `a == b` early
return at the top of the function means the `a == b` test at the end of
the function could never succeed.

2 years agoInline and remove `fallback` closure.
Nicholas Nethercote [Wed, 8 Jun 2022 23:38:09 +0000 (09:38 +1000)]
Inline and remove `fallback` closure.

2 years agoRemove one use of `compare_const_vals`.
Nicholas Nethercote [Wed, 8 Jun 2022 23:27:12 +0000 (09:27 +1000)]
Remove one use of `compare_const_vals`.

A direct comparison has the same effect. This also avoids the need for a
type test within `compare_const_vals`.

2 years agoAssert type equality of `a` and `b` in `compare_const_vals`.
Nicholas Nethercote [Wed, 8 Jun 2022 23:24:08 +0000 (09:24 +1000)]
Assert type equality of `a` and `b` in `compare_const_vals`.

Because they're always equal.

2 years agoRemove dead code from `compare_const_vals`.
Nicholas Nethercote [Wed, 8 Jun 2022 23:08:04 +0000 (09:08 +1000)]
Remove dead code from `compare_const_vals`.

It's never executed when running the entire test suite. I think it's
because of the early return at the top of the function if `a.ty() != ty`
succeeds.

2 years agosimplify_match_pair: avoid the second comparison if possible.
Nicholas Nethercote [Thu, 9 Jun 2022 01:51:41 +0000 (11:51 +1000)]
simplify_match_pair: avoid the second comparison if possible.

Also, the `try_to_bits` always succeeds, so use `unwrap`.

2 years agosort_candidates: avoid the second comparison if possible.
Nicholas Nethercote [Wed, 8 Jun 2022 07:02:06 +0000 (17:02 +1000)]
sort_candidates: avoid the second comparison if possible.

This is a performance win for `unicode-normalization`.

The commit also removes the closure, which isn't necessary. And
reformulates the comparison into a form I find easier to read.

2 years agoconst_range_contains: avoid the second comparison if possible.
Nicholas Nethercote [Wed, 8 Jun 2022 04:49:32 +0000 (14:49 +1000)]
const_range_contains: avoid the second comparison if possible.

This is a performance win for `unicode-normalization`.

Also, I find the new formulation easier to read.

2 years agoAuto merge of #98152 - JohnTitor:rollup-osr17j6, r=JohnTitor
bors [Wed, 15 Jun 2022 23:49:06 +0000 (23:49 +0000)]
Auto merge of #98152 - JohnTitor:rollup-osr17j6, r=JohnTitor

Rollup of 7 pull requests

Successful merges:

 - #97202 (os str capacity documentation)
 - #97964 (Fix suggestions for `&a: T` parameters)
 - #98053 (Fix generic impl rustdoc json output)
 - #98059 (Inline `const_eval_select`)
 - #98092 (Fix sidebar items expand collapse)
 - #98119 (Refactor path segment parameter error)
 - #98135 (Add regression test for #93775)

Failed merges:

r? `@ghost`
`@rustbot` modify labels: rollup

2 years agoThe type of the slice length field is architecture dependent
Wesley Wiser [Wed, 15 Jun 2022 22:35:41 +0000 (18:35 -0400)]
The type of the slice length field is architecture dependent

2 years agoRollup merge of #98135 - JohnTitor:issue-93022-93775, r=compiler-errors
Yuki Okushi [Wed, 15 Jun 2022 22:24:44 +0000 (07:24 +0900)]
Rollup merge of #98135 - JohnTitor:issue-93022-93775, r=compiler-errors

Add regression test for #93775

Closes #93775, also closes #93022 as it should have the same root cause
r? ```@compiler-errors```

Signed-off-by: Yuki Okushi <jtitor@2k36.org>
2 years agoRollup merge of #98119 - EdwinRy:path-parenthesized-type-error, r=estebank
Yuki Okushi [Wed, 15 Jun 2022 22:24:43 +0000 (07:24 +0900)]
Rollup merge of #98119 - EdwinRy:path-parenthesized-type-error, r=estebank

Refactor path segment parameter error

This PR attempts to rewrite the error handling for an unexpected parenthesised type parameters to:
- Use provided data instead of re-parsing the whole span
- Add a multipart suggestion to reflect on the changes with an underline
- Remove the unnecessary "if" nesting

2 years agoRollup merge of #98092 - GuillaumeGomez:fix-sidebar-items-expand-collapse, r=notriddle
Yuki Okushi [Wed, 15 Jun 2022 22:24:42 +0000 (07:24 +0900)]
Rollup merge of #98092 - GuillaumeGomez:fix-sidebar-items-expand-collapse, r=notriddle

Fix sidebar items expand collapse

The collapse/expand event was not working for the items in the source code viewer sidebar (talking about these items:

![Screenshot from 2022-06-14 11-21-58](https://user-images.githubusercontent.com/3050060/173543346-af056928-e921-458f-b918-60f6fd0ecbde.png)

).

This PR fixes it and adds a GUI test to prevent another regression.

r? ```@notriddle```

2 years agoRollup merge of #98059 - tmiasko:inline-const-eval-select, r=Amanieu
Yuki Okushi [Wed, 15 Jun 2022 22:24:41 +0000 (07:24 +0900)]
Rollup merge of #98059 - tmiasko:inline-const-eval-select, r=Amanieu

Inline `const_eval_select`

To avoid circular link time dependency between core and compiler
builtins when building with `-Zshare-generics`.

r? ```@Amanieu```

2 years agoRollup merge of #98053 - GuillaumeGomez:fix-generic-impl-json-ice, r=notriddle
Yuki Okushi [Wed, 15 Jun 2022 22:24:40 +0000 (07:24 +0900)]
Rollup merge of #98053 - GuillaumeGomez:fix-generic-impl-json-ice, r=notriddle

Fix generic impl rustdoc json output

Fixes #97986.

The problem in case of generic trait impl is that the trait's items are the same for all the types afterward. But since they're the same, it's safe for rustdoc-json to just ignore them.

A little representation of what's going on:

```rust
trait T {
    fn f(); // <- defid 0
}

impl<Y> T for Y {
    fn f() {} // <- defid 1
}

struct S; // <- defid 1 (since it matches `impl<Y> T for Y`
```

cc ```@Urgau```

r? ```@CraftSpider```

2 years agoRollup merge of #97964 - WaffleLapkin:fix_borrow_par_suggestions, r=compiler-errors
Yuki Okushi [Wed, 15 Jun 2022 22:24:39 +0000 (07:24 +0900)]
Rollup merge of #97964 - WaffleLapkin:fix_borrow_par_suggestions, r=compiler-errors

Fix suggestions for `&a: T` parameters

I've accidentally discovered that we have broken suggestions for `&a: T` parameters:
```rust
fn f(&mut bar: u32) {}

fn main() {
    let _ = |&mut a| ();
}
```
```text
error[E0308]: mismatched types
 --> ./t.rs:1:6
  |
1 | fn f(&mut bar: u32) {}
  |      ^^^^^^^^-----
  |      |         |
  |      |         expected due to this
  |      expected `u32`, found `&mut _`
  |      help: did you mean `bar`: `&u32`
  |
  = note:           expected type `u32`
          found mutable reference `&mut _`

error[E0308]: mismatched types
 --> ./t.rs:4:23
  |
4 |     let _: fn(u32) = |&mut a| ();
  |                       ^^^^^--
  |                       |    |
  |                       |    expected due to this
  |                       expected `u32`, found `&mut _`
  |                       help: did you mean `a`: `&u32`
  |
  = note:           expected type `u32`
          found mutable reference `&mut _`
```

It's hard to see, but
1. The help span is overlapping with "expected" spans
2. It suggests `fn f( &u32) {}` (no `mut` and lost parameter name) and `|&u32 ()` (no closing `|` and lost parameter name)

I've tried to fix this.

r? ``@compiler-errors``

2 years agoRollup merge of #97202 - joshtriplett:os-str-capacity-documentation, r=dtolnay
Yuki Okushi [Wed, 15 Jun 2022 22:24:38 +0000 (07:24 +0900)]
Rollup merge of #97202 - joshtriplett:os-str-capacity-documentation, r=dtolnay

os str capacity documentation

This is based on https://github.com/rust-lang/rust/pull/95394 , with expansion and consolidation
to address comments from `@dtolnay` and other `@rust-lang/libs-api` team members.

2 years agoMake #[cfg(bootstrap)] not error in proc macros on later stages
est31 [Wed, 15 Jun 2022 19:31:28 +0000 (21:31 +0200)]
Make #[cfg(bootstrap)] not error in proc macros on later stages

As was discovered in https://github.com/rust-lang/rust/pull/93628#issuecomment-1154697627 ,
adding #[cfg(bootstrap)] to a rust-internal proc macro crate
would yield an unexpected cfg name error, at least on later
stages wher the bootstrap cfg arg wasn't set.

rustc already passes arguments to mark bootstrap as expected,
however the means of delivery through the RUSTFLAGS env var
is unable to reach proc macro crates, as described
in the issue linked in the code this commit touches.

This wouldn't be an issue for cfg args that get passed through
RUSTFLAGS, as they would never become *active* either, so
any usage of one of these flags in a proc macro's code would
legitimately yield a lint warning. But since dc302587e2cf5105a3a864319d7e7bcb434bba20,
rust takes extra measures to pass --cfg=bootstrap even in
proc macros, by passing it via the wrapper. Thus, we need
to send the flags to mark bootstrap as expected also from the
wrapper, so that #[cfg(bootstrap)] also works from proc macros.

I want to thank Urgau and jplatte for helping me find the cause of this. ❤️

2 years agoAuto merge of #97178 - sunfishcode:ownedfd-and-dup, r=joshtriplett
bors [Wed, 15 Jun 2022 21:08:08 +0000 (21:08 +0000)]
Auto merge of #97178 - sunfishcode:ownedfd-and-dup, r=joshtriplett

Add a `BorrowedFd::try_clone_to_owned` and accompanying documentation

Add a `BorrowedFd::try_clone_to_owned`, which returns a new `OwnedFd` sharing the underlying file description. And similar for `BorrowedHandle` and `BorrowedSocket` on WIndows.

This is similar to the existing `OwnedFd::try_clone`, but it's named differently to reflect that it doesn't return `Result<Self, ...>`. I'm open to suggestions for better names.

Also, extend the `unix::io` documentation to mention that `dup` is permitted on `BorrowedFd`.

This was originally requsted [here](https://github.com/rust-lang/rust/issues/88564#issuecomment-910786081). At the time I wasn't sure whether it was desirable, but it does have uses and it helps clarify the API. The documentation previously didn't rule out using `dup` on a `BorrowedFd`, but the API only offered convenient ways to do it from an `OwnedFd`. With this patch, the API allows one to do `try_clone` on any type where it's permitted.

2 years agoAuto merge of #9002 - andylizi:fix-never-loop, r=Manishearth
bors [Wed, 15 Jun 2022 18:36:32 +0000 (18:36 +0000)]
Auto merge of #9002 - andylizi:fix-never-loop, r=Manishearth

Fix false positive for `never_loop` struct expression fields

Fixes #9001.

changelog: [`never_loop`]: Now checks for `continue` in struct expression

2 years agoAdd `#[inline]` to small fns of futex `RwLock`
Josh Stone [Wed, 15 Jun 2022 17:48:52 +0000 (10:48 -0700)]
Add `#[inline]` to small fns of futex `RwLock`

The important methods like `read` and `write` were already inlined,
which can propagate all the way to inlining in user code, but these
small state functions were left behind as normal calls. They should
almost always be inlined as well, as they're just a few instructions.

2 years agoAuto merge of #97665 - c410-f3r:assert-compiler, r=oli-obk
bors [Wed, 15 Jun 2022 17:43:01 +0000 (17:43 +0000)]
Auto merge of #97665 - c410-f3r:assert-compiler, r=oli-obk

[RFC 2011] Minimal initial implementation

Tracking issue: #44838
Third step of #96496

Implementation has ~290 LOC with the bare minimum to be in a functional state. Currently only searches for binary operations to mimic what `assert_eq!` and `assert_ne!` already do.

r? `@oli-obk`

2 years agoConsume resolutions for lowering separately.
Camille GILLOT [Wed, 15 Jun 2022 17:42:43 +0000 (19:42 +0200)]
Consume resolutions for lowering separately.

2 years agoUpdate next_node_id.
Camille GILLOT [Wed, 15 Jun 2022 17:41:41 +0000 (19:41 +0200)]
Update next_node_id.

2 years agoRemove unused item_generics_num_lifetimes.
Camille GILLOT [Wed, 15 Jun 2022 17:19:53 +0000 (19:19 +0200)]
Remove unused item_generics_num_lifetimes.

2 years agoExclude cfg "doc" and "doctest" from feature `doc_auto_cfg`
Sprite [Tue, 14 Jun 2022 19:34:00 +0000 (03:34 +0800)]
Exclude cfg "doc" and "doctest" from feature `doc_auto_cfg`

2 years agoFix `cfg(not)` and `cfg(all)` causing Rustdoc stab to disappear
Sprite [Tue, 14 Jun 2022 19:06:28 +0000 (03:06 +0800)]
Fix `cfg(not)` and `cfg(all)` causing Rustdoc stab to disappear

2 years agoAdd the new stability attributes, for Windows.
Dan Gohman [Wed, 15 Jun 2022 16:46:56 +0000 (09:46 -0700)]
Add the new stability attributes, for Windows.

2 years agoRevise the documentation for `try_clone`.
Dan Gohman [Thu, 19 May 2022 13:59:12 +0000 (06:59 -0700)]
Revise the documentation for `try_clone`.

On Unix, describe these in terms of the underlying "file description". On
Windows, describe them in terms of the underlying "object".

2 years agoAdd `BorrowedFd::try_clone_to_owned`.
Dan Gohman [Thu, 19 May 2022 13:43:37 +0000 (06:43 -0700)]
Add `BorrowedFd::try_clone_to_owned`.

And `BorrowedHandle::try_clone_to_owned` and
`BorrowedSocket::try_clone_to_owned` on Windows.

2 years agoDocument that `BorrowedFd` may be used to do a `dup`.
Dan Gohman [Thu, 19 May 2022 13:41:35 +0000 (06:41 -0700)]
Document that `BorrowedFd` may be used to do a `dup`.

2 years agodebuginfo: Fix NatVis for Rc and Arc with unsized pointees.
Michael Woerister [Wed, 15 Jun 2022 14:49:52 +0000 (16:49 +0200)]
debuginfo: Fix NatVis for Rc and Arc with unsized pointees.

2 years agoAuto merge of #95897 - AzureMarker:feature/horizon-std, r=nagisa
bors [Wed, 15 Jun 2022 14:21:28 +0000 (14:21 +0000)]
Auto merge of #95897 - AzureMarker:feature/horizon-std, r=nagisa

STD support for the Nintendo 3DS

Rustc already supports compiling for the Nintendo 3DS using the `armv6k-nintendo-3ds` target (Tier 3). Until now though, only `core` and `alloc` were supported. This PR adds standard library support for the Nintendo 3DS. A notable exclusion is `std::thread` support, which will come in a follow-up PR as it requires more complicated changes.

This has been a joint effort by `@Meziu,` `@ian-h-chamberlain,` myself, and prior work by `@rust3ds` members.

### Background

The Nintendo 3DS (Horizon OS) is a mostly-UNIX looking system, with the caveat that it does not come with a full libc implementation out of the box. On the homebrew side (I'm not under NDA), the libc interface is partially implemented by the [devkitPro](https://devkitpro.org/wiki/devkitPro_pacman) toolchain and a user library like [`libctru`](https://github.com/devkitPro/libctru). This is important because there are [some possible legal barriers](https://github.com/rust-lang/rust/pull/88529#issuecomment-919938396) to linking directly to a library that uses the underlying platform APIs, since they might be considered a trade secret or under NDA.

To get around this, the standard library impl for the 3DS does not directly depend on any platform-level APIs. Instead, it expects standard libc functions to be linked in. The implementation of these libc functions is left to the user. Some functions are provided by the devkitPro toolchain, but in our testing, we used the following to fill in the other functions:
- [`libctru`] - provides more basic APIs, such as `nanosleep`. Linked in by way of [`ctru-sys`](https://github.com/Meziu/ctru-rs/tree/master/ctru-sys).
- [`pthread-3ds`](https://github.com/Meziu/pthread-3ds) - provides pthread APIs for `std::thread`. Implemented using [`libctru`].
- [`linker-fix-3ds`](https://github.com/Meziu/rust-linker-fix-3ds) - fulfills some other missing libc APIs. Implemented using [`libctru`].

For more details, see the `src/doc/rustc/src/platform-support/armv6k-nintendo-3ds.md` file added in this PR.

### Notes
We've already upstreamed changes to the [`libc`] crate to support this PR, as well as the upcoming threading PR. These changes have all been released as of 0.2.121, so we bump the crate version in this PR.
Edit: After some rebases, the version bump has already been merged so it doesn't appear in this PR.

A lot of the changes in this PR are straightforward, and follow in the footsteps of the ESP-IDF target: https://github.com/rust-lang/rust/pull/87666.

The 3DS does not support user space process spawning, so these APIs are unimplemented (similar to ESP-IDF).

[`libctru`]: https://github.com/devkitPro/libctru
[`libc`]: https://github.com/rust-lang/libc

2 years agoAdd regression test for #93775
Yuki Okushi [Wed, 15 Jun 2022 12:31:47 +0000 (21:31 +0900)]
Add regression test for #93775

Signed-off-by: Yuki Okushi <jtitor@2k36.org>
2 years agoAuto merge of #98131 - JohnTitor:rollup-c17vjdy, r=JohnTitor
bors [Wed, 15 Jun 2022 11:04:23 +0000 (11:04 +0000)]
Auto merge of #98131 - JohnTitor:rollup-c17vjdy, r=JohnTitor

Rollup of 3 pull requests

Successful merges:

 - #95118 (Implement stabilization of `#[feature(io_safety)]`.)
 - #98110 (Make `ExprKind::Closure` a struct variant.)
 - #98115 (Remove `rustc_deprecated` diagnostics)

Failed merges:

r? `@ghost`
`@rustbot` modify labels: rollup

2 years ago[RFC 2011] Minimal initial implementation
Caio [Wed, 15 Jun 2022 10:37:40 +0000 (07:37 -0300)]
[RFC 2011] Minimal initial implementation

2 years agoRollup merge of #98115 - jhpratt:remove-rustc_deprecated, r=compiler-errors
Yuki Okushi [Wed, 15 Jun 2022 10:37:15 +0000 (19:37 +0900)]
Rollup merge of #98115 - jhpratt:remove-rustc_deprecated, r=compiler-errors

Remove `rustc_deprecated` diagnostics

Follow-up on #95960. The diagnostics will remain until the next bootstrap, at which point people will have had six weeks to adjust.

``@rustbot`` label +A-diagnostics

r? ``@compiler-errors``

2 years agoRollup merge of #98110 - cjgillot:closure-brace, r=Aaron1011
Yuki Okushi [Wed, 15 Jun 2022 10:37:14 +0000 (19:37 +0900)]
Rollup merge of #98110 - cjgillot:closure-brace, r=Aaron1011

Make `ExprKind::Closure` a struct variant.

Simple refactor since we both need it to introduce additional fields in `ExprKind::Closure`.

r? ``@Aaron1011``