Esteban Küber [Tue, 22 Jun 2021 02:07:19 +0000 (19:07 -0700)]
Modify structured suggestion output
* On suggestions that include deletions, use a diff inspired output format
* When suggesting addition, use `+` as underline
* Color highlight modified span
bors [Wed, 11 Aug 2021 01:36:23 +0000 (01:36 +0000)]
Auto merge of #87254 - rusticstuff:rustc_codegen_llvm_dont_emit_zero_sized_padding, r=eddyb
LLVM codegen: Don't emit zero-sized padding for fields
Currently padding is emitted before fields of a struct and at the end of the struct regardless of the ABI. Even if no padding is required zero-sized padding fields are emitted. This is not useful and - more importantly - it make it impossible to generate the exact vector types that LLVM expects for certain ARM SIMD intrinsics. This change should unblock the implementation of many ARM intrinsics using the `unadjusted` ABI, see https://github.com/rust-lang/stdarch/issues/1143#issuecomment-827404092.
This is a proof of concept only because the field lookup now takes O(number of fields) time compared to O(1) before since it recalculates the mapping at every lookup. I would like to find out how big the performance impact actually is before implementing caching or restricting this behavior to the `unadjusted` ABI.
cc `@SparrowLii` `@bjorn3`
([Discussion on internals](https://internals.rust-lang.org/t/feature-request-add-a-way-in-rustc-for-generating-struct-type-llvm-ir-without-paddings/15007))
Yuki Okushi [Tue, 10 Aug 2021 19:18:47 +0000 (04:18 +0900)]
Rollup merge of #87895 - TheWastl:issue-87872, r=estebank
typeck: don't suggest inaccessible fields in struct literals and suggest ignoring inaccessible fields in struct patterns
Fixes #87872.
This PR adjusts the missing field diagnostic logic in typeck so that when any of the missing fields in a struct literal or pattern is inaccessible then the error is less confusing, even if some of the missing fields are accessible.
Yuki Okushi [Tue, 10 Aug 2021 19:18:41 +0000 (04:18 +0900)]
Rollup merge of #87865 - tsoutsman:master, r=GuillaumeGomez
Clarify terms in rustdoc book
Fixes #70898
I chose to completely remove the term directive over attribute because rustdoc has a lint called `invalid_codeblock_attributes` and the term attributes is used throughout the book. I slightly changed the introductory sentence to describe the relationship between annotations and attributes.
I also moved the text explaining the example from below the blocks to above the blocks which is more in line with the rest of the book. I also changed the description for the `should_panic` attribute as I found it a little confusing. Finally, some of the blocks were `text` and some were `rust` so I changed them all to `text` which is in line with the rest of the book.
Yuki Okushi [Tue, 10 Aug 2021 19:18:40 +0000 (04:18 +0900)]
Rollup merge of #87861 - tsoutsman:patch-1, r=GuillaumeGomez
Fix heading colours in Ayu theme
Closes #87828
The issue seems to stem from #87210 where code headings were changed from a heading containing a `code` element to a heading with the `code-header` class. `rustdoc.css` was updated, but `ayu.css` was missed.
Yuki Okushi [Tue, 10 Aug 2021 19:18:39 +0000 (04:18 +0900)]
Rollup merge of #87854 - BoxyUwU:var-None, r=oli-obk
correctly handle enum variants in `opt_const_param_of`
Fixes #87542
`opt_const_param_of` was returning `None` for args on an enum variant `Enum::Variant::<10>` because we called `generics_of` on the enum variant which has no generics.
Yuki Okushi [Tue, 10 Aug 2021 19:18:38 +0000 (04:18 +0900)]
Rollup merge of #87848 - godmar:@godmar/thread-join-documentation-fix, r=joshtriplett
removed references to parent/child from std::thread documentation
- also clarifies how thread.join and detaching of threads works
- the previous prose implied that there is a relationship between a
spawning thread and the thread being spawned, and that "child" threads
couldn't outlive their "parents" unless detached, which is incorrect.
Yuki Okushi [Tue, 10 Aug 2021 19:18:34 +0000 (04:18 +0900)]
Rollup merge of #87636 - Kixiron:unzip-option, r=scottmcm
Added the `Option::unzip()` method
* Adds the `Option::unzip()` method to turn an `Option<(T, U)>` into `(Option<T>, Option<U>)` under the `unzip_option` feature
* Adds tests for both `Option::unzip()` and `Option::zip()`, I noticed that `.zip()` didn't have any
* Adds `#[inline]` to a few of `Option`'s methods that were missing it
bors [Mon, 9 Aug 2021 08:19:19 +0000 (08:19 +0000)]
Auto merge of #87820 - elichai:patch-2, r=kennytm
Replace read_to_string with read_line in Stdin example
The current example results in infinitely reading from stdin, which can confuse newcomers trying to read from stdin.
(`@razmag` encountered this while learning the language from the docs)
unknown [Sun, 16 May 2021 03:28:16 +0000 (13:28 +1000)]
Add the aarch64-unknown-uefi target
* This commit adds the aarch64-unknown-uefi target and also adds it into
the supported targets list under the tier-3 target table.
* Uses the small code model by default
bors [Mon, 9 Aug 2021 01:08:52 +0000 (01:08 +0000)]
Auto merge of #87465 - audunhalland:refactor_typeck_primary_body_of, r=eddyb
Simplify typeck/primary_body_of, fix comment to match return signature
Hi, new contributor here! I'm carefully reading through the various modules just to learn. I noticed this function, `primary_body_of`, which has gone through a couple of refactors over time, adding new `Option`s to its returned tuple. Observations:
1. the `fn`'s documentation was not all up to date with the the current return signature.
2. `FnHeader` and `FnDecl` are always both `Some` or `None`. So I figured it might just return a reference to the full `hir::FnSig`, for simplicity and more precise typing. It's a pure refactor.
I'm learning better by working with code than just reading it, so here goes! If you want to avoid pure refactor PRs that don't really fix anything, I can revert the code change to only update the comment instead.
bors [Sun, 8 Aug 2021 13:11:09 +0000 (13:11 +0000)]
Auto merge of #87827 - eddyb:wrapperless-mem-replace, r=m-ou-se
Avoid using the `copy_nonoverlapping` wrapper through `mem::replace`.
This is a much simpler way to achieve the pre-#86003 behavior of `mem::replace` not needing dynamically-sized `memcpy`s (at least before inlining), than re-doing #81238 (which needs #86699 or something similar).
I didn't notice it until recently, but `ptr::write` already explicitly avoided using the wrapper, while `ptr::read` just called the wrapper (and was the reason for us observing any behavior change from #86003 in Rust-GPU).
<hr/>
The codegen test I've added fails without the change to `core::ptr::read` like this (ignore the `v0` mangling, I was using a worktree with it turned on by default, for this):
```llvm
13: ; core::intrinsics::copy_nonoverlapping::<u8>
14: ; Function Attrs: inlinehint nonlazybind uwtable
15: define internal void `@_RINvNtCscK5tvALCJol_4core10intrinsics19copy_nonoverlappinghECsaS4X3EinRE8_25mem_replace_direct_memcpy(i8*` %src, i8* %dst, i64 %count) unnamed_addr #0 {
16: start:
17: %0 = mul i64 %count, 1
18: call void `@llvm.memcpy.p0i8.p0i8.i64(i8*` align 1 %dst, i8* align 1 %src, i64 %0, i1 false)
not:17 !~~~~~~~~~~~~~~~~~~~~~ error: no match expected
19: ret void
20: }
```
With the `core::ptr::read` change, `core::intrinsics::copy_nonoverlapping` doesn't get instantiated and the test passes.
Klim Tsoutsman [Sun, 8 Aug 2021 12:35:51 +0000 (22:35 +1000)]
Fix heading colours in Ayu theme
Closes #87828
The issue seems to stem from #87210 where code headings were changed from a heading containing code to a heading with the `code-header` class. `rustdoc.css` was updated, but `ayu.css` was missed.
bors [Sun, 8 Aug 2021 10:23:17 +0000 (10:23 +0000)]
Auto merge of #87836 - tsoutsman:master, r=petrochenkov
Change proc_macro::Diagnostics docs
With Rust 1.54 attributes can invoke function-like macros. This allows functions generated by macros to have non-generic documentation. This pull request changes the documentation of `proc_macro::Diagnostics` to be more specific and include a link to the level.
## Example
Before:
```markdown
Adds a new child diagnostic message to `self` with the level
identified by this method’s name with the given `message`.
```
After:
```markdown
Adds a new child diagnostic message to self with the [`Level::Error`]
level, and the given `message`.
```
bors [Sun, 8 Aug 2021 07:15:48 +0000 (07:15 +0000)]
Auto merge of #87815 - BoxyUwU:cec-generics-of-ice, r=eddyb
encode `generics_of` for fields and ty params
Fixes #87674
Fixes #87603
ICE was caused by calling `generics_of` on a `DefId` without any `generics_of` results. This was happening when we call `generics_of` on parent `DefId`s of an unevaluated const when we evaluate it.
bors [Sun, 8 Aug 2021 01:52:32 +0000 (01:52 +0000)]
Auto merge of #86744 - ijackson:sink-default, r=dtolnay
impl Default, Copy, Clone for std::io::Sink and Empty
The omission of `Sink: Default` is causing me a slight inconvenience in a test harness. There seems little reason for this and `Empty` not to be `Clone` and `Copy` too.
I have made all three of these insta-stable, because:
AIUI `Copy` can only be derived, and I was not able to find any examples of how to unstably derive it. I think it is probably not possible.
I hunted through the git history for precedent and found
bors [Sat, 7 Aug 2021 23:24:15 +0000 (23:24 +0000)]
Auto merge of #87772 - npmccallum:naked_abi, r=Amanieu
Move naked function ABI check to its own lint
This check was previously categorized under the lint named
`UNSUPPORTED_NAKED_FUNCTIONS`. That lint is future incompatible and will
be turned into an error in a future release. However, as defined in the
Constrained Naked Functions RFC, this check should only be a warning.
This is because it is possible for a naked function to be implemented in
such a way that it does not break even the undefined ABI. For example, a
`jmp` to a `const`.
Therefore, this patch defines a new lint named
`UNDEFINED_NAKED_FUNCTION_ABI` which contains just this single check.
Unlike `UNSUPPORTED_NAKED_FUNCTIONS`, `UNDEFINED_NAKED_FUNCTION_ABI`
will not be converted to an error in the future.
bors [Sat, 7 Aug 2021 20:43:25 +0000 (20:43 +0000)]
Auto merge of #87743 - cuviper:opaque-calls, r=nikic
Prepare call/invoke for opaque pointers
Rather than relying on `getPointerElementType()` from LLVM function
pointers, we now pass the function type explicitly when building `call`
or `invoke` instructions.
Yuki Okushi [Sat, 7 Aug 2021 16:13:45 +0000 (01:13 +0900)]
Rollup merge of #87842 - steffahn:hidden_broken_intra_doc, r=dtolnay
Fix intra doc link in hidden doc of Iterator::__iterator_get_unchecked
Recently, I edited the import list of the `core::iter::traits::iterator` module (in #85874). This results in a broken intra doc link in a hidden documentation with the effect that `RUSTDOCFLAGS='--document-private-items --document-hidden-items' x doc library/std` fails. (This can be worked around by adding `-Arustdoc::broken-intra-doc-links`; still, it’s a broken link so let’s fix it.)
Yuki Okushi [Sat, 7 Aug 2021 16:13:44 +0000 (01:13 +0900)]
Rollup merge of #87838 - jetomit:add-readdir-note, r=dtolnay
Document that fs::read_dir skips . and ..
Hi,
I think this is worth noting in the docs since it differs from POSIX `readdir`. I didn’t put it under platform-specific notes because it seems to be consistent across platforms, and changing this behavior in the future could cause pretty nasty bugs.
Yuki Okushi [Sat, 7 Aug 2021 16:13:43 +0000 (01:13 +0900)]
Rollup merge of #87833 - m-rsha:patch-1, r=Aaron1011
Fix typo -- "The" -> "They"
I was reading through source code to try to learn more about the language and how to make the most of it when I ran into a small typo, so I figured I'd offer up a fix! I'm pretty new now, but hopefully I can offer up more substantial changes later. :D
Yuki Okushi [Sat, 7 Aug 2021 16:13:42 +0000 (01:13 +0900)]
Rollup merge of #87789 - tmiasko:panik, r=Mark-Simulacrum
Make vec-shrink-panic test compatible with v0 mangling
The v0 mangling includes an instantiating crate in a mangled name,
which crates a false positive match for a word `panic`.
Rename crate name / test case to avoid the issue.
Yuki Okushi [Sat, 7 Aug 2021 16:13:41 +0000 (01:13 +0900)]
Rollup merge of #87744 - Smittyvb:xpy-test-force-rerun, r=Mark-Simulacrum
Add x.py option to --force-rerun compiletest tests
This can be used like `./x.py test src/test/ui/abi/ --force-rerun`, and is useful when verifying that newly blessed tests don't change between test runs (such as due to being dependent on the current time or memory layout or RNG), without needing to change the test file or find the right file in `build` to remove.
Godmar Back [Sat, 7 Aug 2021 15:33:18 +0000 (11:33 -0400)]
removed references to parent/child from std::thread documentation
- also clarifies how thread.join and detaching of threads works
- the previous prose implied that there is a relationship between a
spawning thread and the thread being spawned, and that "child" threads
couldn't outlive their parents unless detached, which is incorrect.
bors [Sat, 7 Aug 2021 01:26:15 +0000 (01:26 +0000)]
Auto merge of #87408 - kornelski:try_reserve_error, r=yaahc
Hide allocator details from TryReserveError
I think there's [no need for TryReserveError to carry detailed information](https://github.com/rust-lang/rust/issues/48043#issuecomment-825139280), but I wouldn't want that issue to delay stabilization of the `try_reserve` feature.
So I'm proposing to stabilize `try_reserve` with a `TryReserveError` as an opaque structure, and if needed, expose error details later.
This PR moves the `enum` to an unstable inner `TryReserveErrorKind` that lives under a separate feature flag. `TryReserveErrorKind` could possibly be left as an implementation detail forever, and the `TryReserveError` get methods such as `allocation_size() -> Option<usize>` or `layout() -> Option<Layout>` instead, or the details could be dropped completely to make try-reserve errors just a unit struct, and thus smaller and cheaper.
await Marsha [Fri, 6 Aug 2021 23:40:36 +0000 (18:40 -0500)]
Fix typo -- "The" -> "They"
I was reading through source code to try to learn more about the language and how to make the most of it when I ran into a small typo, so I figured I'd offer up a fix! I'm pretty new now, but hopefully I can offer up more substantial changes later. :D
bors [Fri, 6 Aug 2021 18:23:14 +0000 (18:23 +0000)]
Auto merge of #87822 - JohnTitor:rollup-kxojii0, r=JohnTitor
Rollup of 7 pull requests
Successful merges:
- #85807 (bootstrap: Disable initial-exec TLS model on powerpc)
- #87761 (Fix overflow in rustc happening if the `err_count()` is reduced in a stage.)
- #87775 (Add hint for unresolved associated trait items if the trait has a single item)
- #87779 (Remove special case for statement `NodeId` assignment)
- #87787 (Use `C-unwind` ABI for `__rust_start_panic` in `panic_abort`)
- #87809 (Fix typo in the ptr documentation)
- #87816 (Sync rustc_codegen_cranelift)
Yuki Okushi [Fri, 6 Aug 2021 16:46:37 +0000 (01:46 +0900)]
Rollup merge of #87816 - bjorn3:sync_cg_clif-2021-08-06, r=bjorn3
Sync rustc_codegen_cranelift
05677b6bd6c938ed760835d9b1f6514992654ae3 removes two assertions that should have been removed in https://github.com/rust-lang/rust/pull/87515. They are no longer correct and trigger while compiling the sysroot.
Yuki Okushi [Fri, 6 Aug 2021 16:46:35 +0000 (01:46 +0900)]
Rollup merge of #87787 - hyd-dev:c-unwind, r=RalfJung
Use `C-unwind` ABI for `__rust_start_panic` in `panic_abort`
The function originally has `C` ABI but is called using `C-unwind` ABI in `std`:
https://github.com/rust-lang/rust/blob/d4ad1cfc63ba5824196bfb2370451ddb5af2e020/library/std/src/panicking.rs#L49-L54
Which might be [problematic](https://github.com/rust-lang/miri/pull/1745#discussion_r596096876) and triggers this [Miri error](https://github.com/rust-lang/rust/issues/87778#issuecomment-893306222):
```
error: Undefined Behavior: calling a function with ABI C using caller ABI C-unwind
--> /home/hyd-dev/.rustup/toolchains/miri/lib/rustlib/src/rust/library/std/src/panicking.rs:672:9
|
672 | __rust_start_panic(obj)
| ^^^^^^^^^^^^^^^^^^^^^^^ calling a function with ABI C using caller ABI C-unwind
|
= help: this indicates a bug in the program: it performed an invalid operation, and caused Undefined Behavior
= help: see https://doc.rust-lang.org/nightly/reference/behavior-considered-undefined.html for further information
```
Changing the ABI of the function to `C-unwind` fixes the error above.
Yuki Okushi [Fri, 6 Aug 2021 16:46:33 +0000 (01:46 +0900)]
Rollup merge of #87775 - Kobzol:single-associated-item-hint, r=oli-obk
Add hint for unresolved associated trait items if the trait has a single item
This PR introduces a special-cased hint for unresolved trait items paths. It is shown if:
- the path was not resolved to any existing trait item
- and no existing trait item's name was reasonably close with regard to edit distance
- and the trait only has a single item in the corresponding namespace
I didn't know where I should put tests, therefore so far I just managed to bless two existing tests. I would be glad for hints where should tests for a hint like this be created, how should they be named (with reference to the original issue?) and what tests should I create (is it enough to test it just for types? or create separate tests also for functions and constants?).
It could also be turned into a machine applicable suggestion I suppose.
This is my first `rustc` PR, so please go easy on me :)
Yuki Okushi [Fri, 6 Aug 2021 16:46:32 +0000 (01:46 +0900)]
Rollup merge of #87761 - rusticstuff:rustc_error_overflow, r=Mark-Simulacrum
Fix overflow in rustc happening if the `err_count()` is reduced in a stage.
This can happen if stashed diagnostics are removed or replaced with fewer errors. The semantics stay the same if built without overflow checks. Fixes #84219.
Background: I came across this independently by running `RUSTFLAGS="-C overflow-checks=on" ./x.py test`. Fixing this will allow us to move on and find further overflow errors with testing or fuzzing.
bors [Fri, 6 Aug 2021 15:35:01 +0000 (15:35 +0000)]
Auto merge of #87784 - rusticstuff:bootstrap_config_overflow_checks, r=Mark-Simulacrum
Add config.toml options for enabling overflow checks in rustc and std
The names are `overflow-checks` and `overflow-checks-std` and they work similar to `debug-assertions` and `debug-assertions-std`. Once added we can measure how big the performance impact actually is and maybe enable them for CI tests.
Enabling them already makes two ui tests fail:
```
failures:
[ui] ui/parser/item-free-const-no-body-semantic-fail.rs
[ui] ui/parser/item-free-static-no-body-semantic-fail.rs
```
(See #84219 and #87761.)