| Age | Commit message (Collapse) | Author | Lines |
|---|
|
|
|
|
|
Rollup of 7 pull requests
Successful merges:
- #106038 (use implied bounds when checking opaque types)
- #111366 (Make `NonUseContext::AscribeUserTy` carry `ty::Variance`)
- #111375 (CFI: Fix SIGILL reached via trait objects)
- #111439 (Fix backtrace normalization in ice-bug-report-url.rs)
- #111444 (Only warn single-use lifetime when the binders match.)
- #111459 (Update browser-ui-test version to 0.16.0)
- #111460 (Improve suggestion for `self: Box<self>`)
Failed merges:
- #110454 (Require impl Trait in associated types to appear in method signatures)
r? `@ghost`
`@rustbot` modify labels: rollup
|
|
Sync portable-simd to 2023 May 10
Take 2.
r? `@ghost`
|
|
Improve suggestion for `self: Box<self>`
Fixes #110642
|
|
Update browser-ui-test version to 0.16.0
This new version brings one major improvement: it allows to use the original color format in checks (I plan to slowly continue converting colors back to their "original" format, ie the one used in CSS).
It also provides some improvements in some commands API.
r? `````@notriddle`````
|
|
Only warn single-use lifetime when the binders match.
Fixes https://github.com/rust-lang/rust/issues/111400
|
|
Fix backtrace normalization in ice-bug-report-url.rs
This test case currently fails on s390x, and probably other platforms where the last line of a backtrace does not contain and " at <source location>" specification.
The problem with the existing normalization lines
// normalize-stderr-test "\s*\d{1,}: .*\n" -> ""
// normalize-stderr-test "\s at .*\n" -> ""
is that \s matches all whitespace, including newlines, so the first (but not second) of these regexes may merge multiple lines. Thus the output differs depending on which of these matches on the last line of a backtrace.
As the whitespace used in backtraces is just normal space characters, change both regexes to just match at least one space character instead:
// normalize-stderr-test " +\d{1,}: .*\n" -> ""
// normalize-stderr-test " + at .*\n" -> ""
|
|
CFI: Fix SIGILL reached via trait objects
Fix #106547 by transforming the concrete self into a reference to a trait object before emitting type metadata identifiers for trait methods.
|
|
Make `NonUseContext::AscribeUserTy` carry `ty::Variance`
Close #108267
|
|
use implied bounds when checking opaque types
During opaque type inference, we check for the well-formedness of the hidden type in the opaque type's own environment, not the one of the defining site, which are different in the case of TAIT.
However in the case of associated-type-impl-trait, we don't use implied bounds from the impl header. This caused us to reject the following:
```rust
trait Service<Req> {
type Output;
fn call(req: Req) -> Self::Output;
}
impl<'a, Req> Service<&'a Req> for u8 {
type Output= impl Sized; // we can't prove WF of hidden type `WF(&'a Req)` although it's implied by the impl
//~^ ERROR type parameter Req doesn't live long enough
fn call(req: &'a Req) -> Self::Output {
req
}
}
```
although adding an explicit bound would make it pass:
```diff
- impl<'a, Req> Service<&'a Req> for u8 {
+ impl<'a, Req> Service<&'a Req> for u8 where Req: 'a, {
```
I believe it should pass as we already allow the concrete type to be used:
```diff
impl<'a, Req> Service<&'a Req> for u8 {
- type Output= impl Sized;
+ type Output= &'a Req;
```
Fixes #95922
Builds on #105982
cc ``@lcnr`` (because implied bounds)
r? ``@oli-obk``
|
|
|
|
Simplify the implementation of iterators over slices of ZSTs
Currently, slice iterators over ZSTs store `end = start.wrapping_byte_add(len)`.
That's slightly convenient for `is_empty`, but kinda annoying for pretty much everything else -- see bugs like #42789, for example.
This PR instead changes it to just `end = ptr::invalid(len)` instead.
That's easier to think about (IMHO, at least) as well as easier to represent.
`next` is still to big to get inlined into the mir-opt/pre-codegen/ tests, but if I bump the inline threshold to force it to show the whole thing, this implementation is also less MIR:
```
> git diff --numstat
241 370 tests/mir-opt/pre-codegen/slice_iter.forward_loop.PreCodegen.after.mir
255 329 tests/mir-opt/pre-codegen/slice_iter.reverse_loop.PreCodegen.after.mir
184 216 tests/mir-opt/pre-codegen/slice_iter.slice_iter_mut_next_back.PreCodegen.after.mir
182 254 tests/mir-opt/pre-codegen/slice_iter.slice_iter_next.PreCodegen.after.mir
```
(That's ≈70 lines less for `Iter::next`, for example.)
r? `@ghost`
~~Built atop #111282, so draft until that lands.~~
|
|
It isn't clear to me why these error patterns do not trigger,
but I am not going to waste time analyzing bugs in compiletest.
|
|
API changes resulted in subtle MIR and impl differences
|
|
Sync up to rust-lang/portable-simd@852762563aa890286eda2f668b8af30f8aa84216
|
|
|
|
update Miri
r? `@ghost`
|
|
This test case currently fails on s390x, and probably other
platforms where the last line of a backtrace does not contain
and " at <source location>" specification.
The problem with the existing normalization lines
// normalize-stderr-test "\s*\d{1,}: .*\n" -> ""
// normalize-stderr-test "\s at .*\n" -> ""
is that \s matches all whitespace, including newlines, so the
first (but not second) of these regexes may merge multiple
lines. Thus the output differs depending on which of these
matches on the last line of a backtrace.
As the whitespace used in backtraces is just normal space
characters, change both regexes to just match at least one
space character instead:
// normalize-stderr-test " +\d{1,}: .*\n" -> ""
// normalize-stderr-test " + at .*\n" -> ""
|
|
|
|
|
|
|
|
Shrink `SelectionError` a lot
`SelectionError` used to be 80 bytes (on 64 bit). That's quite big. Especially because the selection cache contained `Result<_, SelectionError>. The Ok type is only 32 bytes, so the 80 bytes significantly inflate the size of the cache.
Most variants of the `SelectionError` seem to be hard errors, only `Unimplemented` shows up in practice (for cranelift-codegen, it occupies 23.4% of all cache entries). We can just box away the biggest variant, `OutputTypeParameterMismatch`, to get the size down to 16 bytes, well within the size of the Ok type inside the cache.
|
|
|
|
|
|
Rollup of 7 pull requests
Successful merges:
- #108705 (Prevent ICE with broken borrow in closure)
- #111292 (Fix mishandled `--check-cfg` arguments order)
- #111382 (Isolate coverage FFI type layouts from their underlying LLVM C++ types)
- #111385 (vec-shrink-panik: update expectations to work on LLVM 17)
- #111389 (Add esp-idf platform-support page)
- #111432 (Use visit_assign to detect SSA locals.)
- #111448 (Use proper impl self type for alias impl in rustdoc)
Failed merges:
r? `@ghost`
`@rustbot` modify labels: rollup
|
|
Use proper impl self type for alias impl in rustdoc
We don't want to use `type_of(type_alias)`, we want to use `type_of(impl)` -- this will give us the self type of the impl *properly substituted* in the case that it's an alias.
Fixes #111420
|
|
Use visit_assign to detect SSA locals.
I screwed up the logic in 3c43b61b870add2daddbd8e480477e5a8aa409c2.
Fixes https://github.com/rust-lang/rust/issues/111426
|
|
Add esp-idf platform-support page
As mentioned in https://github.com/rust-lang/rust/pull/111369#issuecomment-1539881848, the initial PR for esp-idf didn't include a platform support document which is required for the merge of https://github.com/rust-lang/rust/pull/111369.
|
|
vec-shrink-panik: update expectations to work on LLVM 17
For some reason, the called function is `cleanup` on LLVM 17 instead of `filter`.
r? `@Amanieu`
|
|
Isolate coverage FFI type layouts from their underlying LLVM C++ types
I noticed that several of the types used to send coverage information through FFI are not properly isolated from the layout of their corresponding C++ types in the LLVM API.
This PR adds more explicitly-defined FFI struct/enum types in `CoverageMappingWrapper.cpp`, so that Rust source files in `rustc_codegen_ssa` and `rustc_codegen_llvm` aren't directly exposed to LLVM C++ types.
|
|
Fix mishandled `--check-cfg` arguments order
This PR fixes a bug in `--check-cfg` where the order of `--check-cfg=names(a)` and `--check-cfg=values(a,…)` would trip the compiler.
Fixes https://github.com/rust-lang/rust/issues/111291
cc `@taiki-e` `@petrochenkov`
|
|
Prevent ICE with broken borrow in closure
r? `@Nilstrieb`
Fixes #108683
This solution isn't ideal, I'm hoping to find a way to continue compilation without ICEing.
|
|
|
|
Optimize dataflow-const-prop place-tracking infra
Optimization opportunities found while investigating https://github.com/rust-lang/rust/pull/110719
Computing places breadth-first ensures that we create short projections before deep projections, since the former are more likely to be propagated.
The most relevant is the pre-computation of flooded places. Callgrind showed `flood_*` methods and especially `preorder_preinvoke` were especially hot. This PR attempts to pre-compute the set of `ValueIndex` that `preorder_invoke` would visit.
Using this information, we make some `PlaceIndex` inaccessible when they contain no `ValueIndex`, allowing to skip computations for those places.
cc `@jachris` as original author
|
|
Currently, slice iterators over ZSTs store `end = start.wrapping_byte_add(len)`.
That's slightly convenient for `is_empty`, but kinda annoying for pretty much everything else -- see bugs like 42789, for example.
This PR instead changes it to just `end = ptr::invalid(len)` instead.
That's easier to think about (IMHO, at least) as well as easier to represent.
|
|
|
|
Include the current Crate name in the measureme output name
See https://rust-lang.zulipchat.com/#narrow/stream/182449-t-compiler.2Fhelp/topic/measureme.20flamegraph.20panics/near/356367013
cc `@andjo403`
Currently, attempting to use `MIRIFLAGS=-Zmiri-measureme=miri cargo miri test` on a crate with multiple test targets (which is very common) will produce a corrupted measureme output file, because the various interpreter processes will stomp each other's output.
This change does not entirely prevent this, but the various test targets seem to always have different crate names, so if nothing else this will make the broken measureme files much harder to encounter by accident, while also making it clear what they are all for.
|
|
|
|
|
|
- disable the AVX512 variant for now (flaky)
- tell Clippy to knock it off
|
|
|
|
- Fix LANES over-replace
- Bring in traits
- Use less inference-heavy types
|
|
For some reason, the called function is `cleanup` on LLVM 17 instead of
`filter`.
r? @Amanieu
|
|
Don't force include Windows goop when documenting
Why do we need to include all the windows bits on non-windows platforms? Let's try not doing that.
Possible alternative to #111394, if it works.
|
|
Rustup
|
|
|
|
|
|
Rollup of 6 pull requests
Successful merges:
- #110673 (Make alias bounds sound in the new solver (take 2))
- #110747 (Encode types in SMIR)
- #111095 (Correctly handle associated items of a trait inside a `#[doc(hidden)]` item)
- #111381 (Keep encoding attributes for closures)
- #111408 (Fix incorrect implication of transmuting slices)
- #111410 (Switch to `EarlyBinder` for `thir_abstract_const` query)
Failed merges:
r? `@ghost`
`@rustbot` modify labels: rollup
|
|
|
