Fix "since" field for `Once::is_complete`'s `#[stable]` attribute
It was accidentally merged with the wrong version in #68945. Thanks @jplatte for noticing.
This also needs to be beta backported.
Return feature gate as a `Symbol`
A minor refactoring that will be needed for #68940. That PR is blocked on me changing the error comments in a whole lot of UI tests.
r? @oli-obk
def_collector: Fully visit async functions
We forgot to visit attributes previously, it caused ICEs.
Special treatment of async functions is also moved from `visit_item` to `visit_fn` to reuse more of the default visitor.
Fixes https://github.com/rust-lang/rust/issues/67778.
Implement From<&mut str> for String
I ran into this missing impl when trying to do `String::from` on the result returned from this API in the `uuid` crate:
https://docs.rs/uuid/0.8.1/uuid/adapter/struct.Hyphenated.html#method.encode_lower
I wasn't sure what to put in the stability annotation. I'd appreciate some help with that :)
ast: `Mac`/`Macro` -> `MacCall`
It's now obvious that these refer to macro calls rather than to macro definitions.
It's also a single name instead of two different names in different places.
`rustc_expand` usually calls macro calls in a wide sense (including attributes and derives) "macro invocations", but structures and variants renamed in this PR are only relevant to fn-like macros, so it's simpler and clearer to just call them calls.
cc https://github.com/rust-lang/rust/pull/63586#discussion_r314232513
r? @eddyb
Add undo_leak to reset RefCell borrow state
This method is complementary for the feature cell_leak added in an
earlier PR. It allows *safely* reverting the effects of leaking a borrow guard by
statically proving that such a guard could not longer exist. This was
not added to the existing `get_mut` out of concern of impacting the
complexity of the otherwise pure pointer cast and because the name
`get_mut` poorly communicates the intent of resetting remaining borrows.
This is a follow-up to #68712 and uses the same tracking issue, #69099,
as these methods deal with the same mechanism and the idea came up
[in a review comment](https://github.com/rust-lang/rust/pull/68712#discussion_r384670041).
@dtolnay who reviewed the prior PR.
cc @RalfJung
Change "method" to "associated function"
r? @matthewjasper
cc @Centril @eddyb #67742
I'm opening this mostly as a test to see what the diagnostic changes would be. It seems that this makes them somewhat more verbose, and I'm not sure it's worth it...
The relevant changes are the last two commits (it is rebased on top of #67742)
Remove `sip::Hasher::short_write`.
`sip::Hasher::short_write` is currently unused. It is called by
`sip::Hasher::write_{u8,usize}`, but those methods are also unused,
because `DefaultHasher`, `SipHasher` and `SipHasher13` don't implement
any of the `write_xyz` methods, so all their write operations end up
calling `sip::Hasher::write`.
(I confirmed this by inserting a `panic!` in `sip::Hasher::short_write`
and running the tests -- they all passed.)
The alternative would be to add all the missing `write_xyz` methods.
This does give some significant speed-ups, but it hurts compile times a
little in some cases. See #69152 for details. This commit does the
conservative thing and doesn't change existing behaviour.
r? @rust-lang/libs
Emit 1-based column numbers in debuginfo
* Use byte offsets instead of char position offsets. Resolves#67360.
* Use 1-based offsets instead of 0-based ones. Resolves#65437.
* Consistently omit column information for msvc targets, matching clang behaviour (previously columns have been omitted from `DILocation`, but not from `DILexicalBlock`).
submodules: update clippy from 329923ed to fdce47ba
Changes:
````
add CR feedback
Improve documentation
Use `edition:2018` flag more widely
Update tests/ui/needless_doc_main.rs
Move links to the end of each chapter on adding_lints
Move links to the end of each chapter on CONTRIBUTING
Clean-up adding_lints.md
Clean-up CONTRIBUTING.md
needless_doc_main: only check rust code
Use `node_type_opt` over `node_type`
Fix doc
Fix ICE with trivial_bounds feature
clippy_lints: readme: don't mention crates.io since it is no longer used to publish clippy.
update rust-lang.github.io to rustc-dev-guide.rust-lang.org
Improve placeholder in map_unit_fn
Fix match single binding when in a let stmt
Improve error messages for {option,result}_map_unit_fn
Mention the setup instructions in CONTRIBUTING
redundant_pattern: take binding (ref, ref mut) into account in suggestion.
check_pat: delay creation of the "normal" vec until we reach the branch where is is actually needed
deps: bump itertools 0.8 -> 0.9
add lint on File::read_to_string and File::read_to_end
transition rustc-guide to rustc-dev-guide
Rename macro_use_import -> macro_use_imports
warn on macro_use attr
Fix deploy script for tag deploys
````
Fixes rust-lang#69957
Add `mir-opt` test for more precise drop elaboration
Depends on #69676. This test should ensure that the problem fixed in that PR does not reoccur.
This has been split out from #69676 since the test fails on certain targets where no cleanup blocks are emitted. I have to find the correct `ignore` directives.
r? @oli-obk
Don't store locals in generators that are immediately overwritten with the resume argument
This fixes https://github.com/rust-lang/rust/issues/69672 and makes https://github.com/rust-lang/rust/pull/69033 pass the async fn size tests again (in other words, there will be no size regression of async fn if both this and https://github.com/rust-lang/rust/pull/69033 land).
~~This is a small botch and I'd rather have a more precise analysis, but that seems much harder to pull off, so this special-cases `Yield` terminators that store the resume argument into a simple local (ie. without any field projections) and explicitly marks that local as "not live" in the suspend point of that yield. We know that this local does not need to be stored in the generator for this suspend point because the next resume would immediately overwrite it with the passed-in resume argument anyways. The local might still end up in the state if it is used across another yield.~~ (this now properly updates the dataflow framework to handle this case)
Optimize catch_unwind to match C++ try/catch
This refactors the implementation of catching unwinds to allow LLVM to inline the "try" closure directly into the happy path, avoiding indirection. This means that the catch_unwind implementation is (after this PR) zero-cost unless a panic is thrown.
https://rust.godbolt.org/z/cZcUSB is an example of the current codegen in a simple case. Notably, the codegen is *exactly the same* if `-Cpanic=abort` is passed, which is clearly not great.
This PR, on the other hand, generates the following assembly:
```asm
# -Cpanic=unwind:
push rbx
mov ebx,0x2a
call QWORD PTR [rip+0x1c53c] # <happy>
mov eax,ebx
pop rbx
ret
mov rdi,rax
call QWORD PTR [rip+0x1c537] # cleanup function call
call QWORD PTR [rip+0x1c539] # <unfortunate>
mov ebx,0xd
mov eax,ebx
pop rbx
ret
# -Cpanic=abort:
push rax
call QWORD PTR [rip+0x20a1] # <happy>
mov eax,0x2a
pop rcx
ret
```
Fixes#64224, and resolves#64222.
