optimize field ordering by grouping m*2^n-sized fields with equivalently aligned ones
```rust
use std::ptr::addr_of;
use std::mem;
struct Foo {
word: u32,
byte: u8,
ary: [u8; 4]
}
fn main() {
let foo: Foo = unsafe { mem::zeroed() };
println!("base: {:p}\nword: {:p}\nbyte: {:p}\nary: {:p}", &foo, addr_of!(foo.word), addr_of!(foo.byte), addr_of!(foo.ary));
}
```
prints
```
base: 0x7fffc1a8a668
word: 0x7fffc1a8a668
byte: 0x7fffc1a8a66c
ary: 0x7fffc1a8a66d
```
I.e. the `u8` in the middle causes the array to sit at an odd offset, which might prevent optimizations, especially on architectures where unaligned loads are costly.
Note that this will make field ordering niche-dependent, i.e. a `Bar<T>` with `T=char` and `T=u32` may result in different field order, this may break some code that makes invalid assumptions about `repr(Rust)` types.
Bump `fd-lock` in `bootstrap` again
Followup to https://github.com/rust-lang/rust/pull/103778
Sorry for the quick succession but this fixes one more building issue for Tier 3 `windows-gnullvm` that I have previously missed, and it would be nice to have it in the release.
Lower return type outside async block creation
This allows feeding a different output type to async blocks with a different `ImplTraitContext`. Spotted this while working on #104321
Refactor must_use lint into two parts
Before, the lint did the checking for `must_use` and pretty printing the types in a special format in one pass, causing quite complex and untranslatable code.
Now the collection and printing is split in two. That should also make it easier to translate or extract the type pretty printing in the future.
Also fixes an integer overflow in the array length pluralization
calculation.
fixes#104352
Followup to https://github.com/rust-lang/rust/pull/103778
Sorry for the quick succession but this fixes one more building issue for Tier 3 `windows-gnullvm` that I have previously missed, and it would be nice to have it in the release.
Use `tcx.require_lang_item` instead of unwrapping lang items
I clearly remember esteban telling me that there is `require_lang_item` but he was from a phone atm and I couldn't find it, so I didn't use it. Stumbled on it today, so here we are :)
rustdoc: remove no-op CSS `.popover::before / a.test-arrow { display: inline-block }`
Since this box is absolutely positioned, its display type is [blockified] anyway. We just need to make sure it isn't `display: none`.
[blockified]: https://www.w3.org/TR/css-display-3/#transformations
Add failing test for projections used as const generic
Based on the experiment done in https://github.com/rust-lang/rust/pull/104443, we realized it's currently not possible to support projections in const generics. More information about it in https://github.com/rust-lang/rust/pull/104443#discussion_r1029375633.
This PR adds the UI test in any case so we can gather data in order to work towards adding `TyAlias` into the ABI in the future.
r? ``@oli-obk``
Forbid inlining `thread_local!`'s `__getit` function on Windows
Sadly, this will make things slower to avoid UB in an edge case, but it seems hard to avoid... and really whenever I look at this code I can't help but think we're asking for trouble.
It's pretty dodgy for us to leave this as a normal function rather than `#[inline(never)]`, given that if it *does* get inlined into a dynamically linked component, it's extremely unsafe (you get some other thread local, or if you're lucky, crash). Given that it's pretty rare for people to use dylibs on Windows, the fact that we haven't gotten bug reports about it isn't really that convincing. Ideally we'd come up with some kind of compiler solution (that avoids paying for this cost when static linking, or *at least* for use within the same crate...), but it's not clear what that looks like.
Oh, and because all this is only needed when we're implementing `thread_local!` with `#[thread_local]`, this patch adjusts the `cfg_attr` to be `all(windows, target_thread_local)` as well.
r? ``@ChrisDenton``
See also #84933, which is about improving the situation.
Simd contains fix
Fixes#104726
The bug was introduced by an improvement late in the original PR (#103779) which added the backtracking when the last and first byte of the needle were the same. That changed the meaning of the variable for the last probe offset, which I should have split into the last byte offset and last probe offset. Not doing so lead to incorrect loop conditions.
Update Clippy
r? `@Manishearth`
Sorry for taking so long. There were so many blockers and so little time. This situation should be mitigated with #104007 in the future.
disable strict-provenance-violating doctests in Miri
Most of these are on deprecated unstable functions anyway. This lets us run the remaining doctests with `-Zmiri-strict-provenance`, which I think is a win.
r? `@thomcc`
Remove a lifetime resolution hack from `compare_predicate_entailment`
This is not needed anymore, probably due to #102334 equating the function signatures fully in `collect_trait_impl_trait_tys`. Also, the assertion in in #102903 makes sure that this is actually fixed, so I'm pretty confident this isn't needed.
Only declare bindings for if-let guards once per arm
Currently, each candidate for a match arm uses separate locals for the bindings in the if-let guard, causing problems (#88015) when those branches converge in the arm body.
Fixes#88015 (🤞)
Check generics parity before collecting return-position `impl Trait`s in trait
The only thing is that this duplicates the error message for number of generics mismatch, but we already deduplicate that error message in Cargo. I could add a flag to delay the error if the reviewer cares.
Fixes#104281
Also drive-by adds a few comments to the `collect_trait_impl_trait_tys` method, and removes an unused argument from `compare_number_of_generics`.
