Test interaction of unions with non-zero/niche-filling optimization
Notably this nails down part of the behavior that MaybeUninit assumes, e.g. that a Option<MaybeUninit<&u8>> does not take advantage of non-zero optimization, and thus is a safe construct.
It also verifies the status quo: that even unions that could theoretically take advantage of niches don't. (relevant: https://github.com/rust-lang/rust/issues/36394)
Mark core::alloc::Layout::from_size_align_unchecked const
Makes it possible (pending stabilization of #57563 (`const_fn`)) to rewrite code like
```rust
const BUFFER_SIZE: usize = 0x2000;
const BUFFER_ALIGN: usize = 0x1000;
fn foo() {
let layout = std::alloc::Layout::from_size_align(BUFFER_SIZE, BUFFER_ALIGN)
.unwrap();
let buffer = std::alloc::alloc(layout);
}
```
to
```rust
const BUFFER_LAYOUT: std::alloc::Layout = unsafe {
std::alloc::Layout::from_size_align_unchecked(0x2000, 0x1000)
};
fn foo() {
let buffer = std::alloc::alloc(BUFFER_LAYOUT);
}
```
which (although `unsafe` is used) looks somewhat cleaner and is easier to read.
Added ignore-sgx for appropriate tests in src/test
These are all the tests that make sense to ignore when targeting fortanix-unknonw-sgx, at least in test/runpass. Other suites not yet covered.
lint: convert incoherent_fundamental_impls into hard error
*Summary for affected authors:* If your crate depends on one of the following crates, please upgrade to a newer version:
- gtk-rs: upgrade to at least 0.4
- rusqlite: upgrade to at least 0.14
- nalgebra: upgrade to at least 0.15, or the last patch version of 0.14
- spade: upgrade or refresh the Cargo.lock file to use version 1.7
- imageproc: upgrade to at least 0.16 (newer versions no longer use nalgebra)
implement #46205
r? @nikomatsakis
Checking generic args after late bound region err.
Fixes#60622.
This PR fixes an ICE that occurs when a late bound region error is
emitted and that resulted in the rest of the generic arguments of a
function not being checked.
For example, you could specify a generic type parameter `T` in a function
call `foo<'_, T>()` to a function that doesn't have a generic type
parameter.
Since an error wasn't emitted from the function, compilation
continued to parts of typeck that didn't expect a generic type argument
in a call for a function that didn't have any generic type arguments.
This commit fixes an ICE that occurs when a late bound region error is
emitted and that resulted in the rest of the generic arguments of a
function not being checked.
For example, you could specify a generic type parameter `T` in a function
call `foo<'_, T>()` to a function that doesn't have a generic type
parameter.
Since an error wasn't emitted from the function, compilation
continued to parts of typeck that didn't expect a generic type argument
in a call for a function that didn't have any generic type arguments.
Include expression to wait for to the span of Await
Currently the span of `await!` only includes itself:
```rust
await!(3);
// ^^^^^
```
This PR changes it so that the span holds the whole `await!` expression:
```rust
await!(3);
// ^^^^^^^^^
forego caching for all participants in cycles, apart from root node
This is a targeted fix for #60010, which uncovered a pretty bad failure of our caching strategy in the face of coinductive cycles. The problem is explained in the comment in the PR on the new field, `in_cycle`, but I'll reproduce it here:
> Starts out as false -- if, during evaluation, we encounter a
> cycle, then we will set this flag to true for all participants
> in the cycle (apart from the "head" node). These participants
> will then forego caching their results. This is not the most
> efficient solution, but it addresses #60010. The problem we
> are trying to prevent:
>
> - If you have `A: AutoTrait` requires `B: AutoTrait` and `C: NonAutoTrait`
> - `B: AutoTrait` requires `A: AutoTrait` (coinductive cycle, ok)
> - `C: NonAutoTrait` requires `A: AutoTrait` (non-coinductive cycle, not ok)
>
> you don't want to cache that `B: AutoTrait` or `A: AutoTrait`
> is `EvaluatedToOk`; this is because they were only considered
> ok on the premise that if `A: AutoTrait` held, but we indeed
> encountered a problem (later on) with `A: AutoTrait. So we
> currently set a flag on the stack node for `B: AutoTrait` (as
> well as the second instance of `A: AutoTrait`) to supress
> caching.
>
> This is a simple, targeted fix. The correct fix requires
> deeper changes, but would permit more caching: we could
> basically defer caching until we have fully evaluated the
> tree, and then cache the entire tree at once.
I'm not sure what the impact of this fix will be in terms of existing crates or performance: we were accepting incorrect code before, so there will perhaps be some regressions, and we are now caching less.
As the comment above notes, we could do a lot better than this fix, but that would involve more invasive rewrites. I thought it best to start with something simple.
r? @pnkfelix -- but let's do crater/perf run
cc @arielb1
Add #[doc(hidden)] attribute on compiler generated module.
Resolves unavoidable `missing_docs` warning/error on proc-macro crates.
Resolves#42008.
Changes not yet tested locally, however I wanted to submit first since `rustc` takes forever to compile.
