The (internal) std::io::lazy::Lazy was used to lazily initialize the
stdout and stdin buffers (and mutexes). It uses atexit() to register a
destructor to flush the streams on exit, and mark the streams as
'closed'. Using the stream afterwards would result in a panic.
Stdout uses a LineWriter which contains a BufWriter that will flush the
buffer on drop. This one is important to be executed during shutdown,
to make sure no buffered output is lost. It also forbids access to
stdout afterwards, since the buffer is already flushed and gone.
Stdin uses a BufReader, which does not implement Drop. It simply forgets
any previously read data that was not read from the buffer yet. This
means that in the case of stdin, the atexit() function's only effect is
making stdin inaccessible to the program, such that later accesses
result in a panic. This is uncessary, as it'd have been safe to access
stdin during shutdown of the program.
---
This change removes the entire io::lazy module in favour of
SyncOnceCell. SyncOnceCell's fast path is much faster (a single atomic
operation) than locking a sys_common::Mutex on every access like Lazy
did.
However, SyncOnceCell does not use atexit() to drop the contained object
during shutdown.
As noted above, this is not a problem for stdin. It simply means stdin
is now usable during shutdown.
The atexit() call for stdout is moved to the stdio module. Unlike the
now-removed Lazy struct, SyncOnceCell does not have a 'gone and
unusable' state that panics. Instead of adding this again, this simply
replaces the buffer with one with zero capacity. This effectively
flushes the old buffer *and* makes any writes afterwards pass through
directly without touching a buffer, making print!() available during
shutdown without panicking.
- Suggest `x.py setup` if config.toml doesn't exist yet (twice, once
before and once after the build)
- Prompt for a profile if not given on the command line
- Print the configuration file that will be used
- Print helpful starting commands after setup
- Link to the dev-guide after finishing
- Note that distro maintainers will see the changelog warning
const_evaluatable_checked: extend predicate collection
We now walk the hir instead of using `ty` so that we get better spans here, While I am still not completely sure if that's
what we want in the end, it does seem a lot closer to the final goal than the previous version.
We also look into type aliases (and use a `TypeVisitor` here), about which I am not completely sure, but we will see how well this works.
We also look into fn decls, so the following should work now.
```rust
fn test<T>() -> [u8; std::mem::size_of::<T>()] {
[0; std::mem::size_of::<T>()]
}
```
Additionally, we visit the optional trait and self type of impls.
r? `@oli-obk`
Fix underflow when calculating the number of no-op jumps folded
When removing unwinds to no-op blocks and folding jumps to no-op blocks,
remove the unwind target first. Otherwise we cannot determine if target
has been already folded or not.
Previous implementation incorrectly assumed that all resume targets had
been folded already, occasionally resulting in an underflow:
```
remove_noop_landing_pads: removed 18446744073709551613 jumps and 3 landing pads
```
Stabilize intra-doc links
Fixes https://github.com/rust-lang/rust/issues/43466
Thanks to the great work of `@jyn514` in getting the [cross-crate reexport issue](https://github.com/rust-lang/rust/issues/65983) in intra-rustdoc links fixed, I think we're now in a position to stabilize this feature.
The tracking issue currently has two unresolved issues:
- <s>behavior around doc(hidden): This is fixed in https://github.com/rust-lang/rust/pull/73365, which is just waiting for CI and should land tomorrow. It's also a pretty niche bug so while I expect it to land soon I don't think we need to block stabilization on it anyway.</s>
- Non-identifier primitive types like slices: This was not a part of the original RFC anyway, and is a pretty niche use case
The feature itself, sans https://github.com/rust-lang/rust/issues/65983, has been shipped on nightly for three years now, with people using it on docs.rs. https://github.com/rust-lang/rust/issues/65983 itself is not an overwhelmingly central bit of functionality; the reason we elected to block stabilization on it was that back in 2017 it was not possible to fix the issue without some major refactorings of resolve, and we did not want to stabilize something that had such a potentially unfixable bug.
Given that we've fixed it, I see no reason to delay stabilization on this long awaited feature. It's possible that the latest patches have problems, however we _have_ done crater runs of some of the crucial parts. Furthermore, that's what the release trains are for, we will have a solid three months to let it ride the trains before it actually hits the stable compiler.
r? `@rust-lang/rustdoc`
Now that `PrimTy::name()` exists, there's no need to carry around the
name of the primitive that failed to resolve. This removes the variants
special-casing primitives in favor of `NotResolved`.
- Remove `NoPrimitiveImpl` and `NoPrimitiveAssocItem`
- Remove hacky `has_primitive` check in `resolution_failure()`
- Fixup a couple tests that I forgot to `--bless` before
- Add `PrimTy::name` and `PrimTy::name_str`
- Use those new functions to distinguish between the name in scope and
the canonical name
- Fix diagnostics for primitive types
- Add tests for primitives
Previously, these were spread throughout the codebase. This had two
drawbacks:
1. It caused the fast path to be slower: even if a link resolved,
rustdoc would still perform various lookups for the error diagnostic.
2. It was inconsistent and didn't always give all diagnostics (https://github.com/rust-lang/rust/issues/76925)
Now, diagnostics only perform expensive lookups in the error case.
Additionally, the error handling is much more consistent, both in
wording and behavior.
- Remove `CannotHaveAssociatedItems`, `NotInScope`, `NoAssocItem`, and `NotAVariant`
in favor of the more general `NotResolved`
`resolution_failure` will now look up which of the four above
categories is relevant, instead of requiring the rest of the code to
be consistent and accurate in which it picked.
- Remove unnecessary lookups throughout the intra-doc link pass. These
are now done by `resolution_failure`.
+ Remove unnecessary `extra_fragment` argument to `variant_field()`;
it was only used to do lookups on failure.
+ Remove various lookups related to associated items
+ Remove distinction between 'not in scope' and 'no associated item'
- Don't perform unnecessary copies
- Remove unused variables and code
- Update tests
- Note why looking at other namespaces is still necessary
- 'has no inner item' -> 'contains no item'
bless tests
revert const_type_id stabilization
This reverts #72488, which is currently on beta and scheduled to stabilize in `1.47.0`, based on https://github.com/rust-lang/rust/pull/75923#issuecomment-696676511
It turns out we might not be quite ready to stabilize `TypeId` in const contexts before having a chance to rework its internals. Since `TypeId` is a bit of an oddity we want to be careful about how those internals are currently being relied on while making changes. That will be easier to do without having to also consider compile-time contexts.
r? `@eddyb`
Rollup of 9 pull requests
Successful merges:
- #76898 (Record `tcx.def_span` instead of `item.span` in crate metadata)
- #76939 (emit errors during AbstractConst building)
- #76965 (Add cfg(target_has_atomic_equal_alignment) and use it for Atomic::from_mut.)
- #76993 (Changing the alloc() to accept &self instead of &mut self)
- #76994 (fix small typo in docs and comments)
- #77017 (Add missing examples on Vec iter types)
- #77042 (Improve documentation for ToSocketAddrs)
- #77047 (Miri: more informative deallocation error messages)
- #77055 (Add #[track_caller] to more panicking Cell functions)
Failed merges:
r? `@ghost`
This keeps only the `std` artifacts compiled by the given stage, not the
compiler. This is useful when working on the latter stages of the
compiler in tandem with the standard library, since you don't have to
rebuild the *entire* compiler when the standard library changes.
MIR pass to remove unneeded drops on types not needing drop
This is heavily dependent on MIR inlining running to actually see the drop statement.
Do we want to special case replacing a call to std::mem::drop with a goto aswell?
