Trying to shrink_to greater than capacity should be no-op
Per the discussion in https://github.com/rust-lang/rust/issues/56431, `shrink_to` shouldn't panic if you try to make a vector shrink to a capacity greater than its current capacity.
BTreeMap: test all borrowing interfaces and test more chaotic order behavior
Inspired by #81169, test what happens if you mess up order of the type with which you search (as opposed to the key type).
r? `@Mark-Simulacrum`
Add `unwrap_unchecked()` methods for `Option` and `Result`
In particular:
- `unwrap_unchecked()` for `Option`.
- `unwrap_unchecked()` and `unwrap_err_unchecked()` for `Result`.
These complement other `*_unchecked()` methods in `core` etc.
Currently there are a couple of places it may be used inside rustc (`LinkedList`, `BTree`). It is also easy to find other repositories with similar functionality.
Fixes#48278.
BTreeMap: bring back the key slice for immutable lookup
Pave the way for binary search, by reverting a bit of #73971, which banned `keys` for misbehaving while it was defined for every `BorrowType`. Adding some `debug_assert`s along the way.
r? `@Mark-Simulacrum`
libtest: Wait for test threads to exit after they report completion
Otherwise we can miss bugs where a test reports that it succeeded but then panics within a TLS destructor.
Example:
```rust
use std:🧵:sleep;
use std::time::Duration;
struct Foo;
impl Drop for Foo {
fn drop(&mut self) {
sleep(Duration::from_secs(1));
panic!()
}
}
thread_local!(static FOO: Foo = Foo);
#[test]
pub fn test() {
FOO.with(|_| {});
}
```
Before this fix, `cargo test` incorrectly reports success.
```console
$ cargo test
Finished test [unoptimized + debuginfo] target(s) in 0.01s
Running target/debug/deps/panicking_test-85130fa46b54f758
running 1 test
test test ... ok
test result: ok. 1 passed; 0 failed; 0 ignored; 0 measured; 0 filtered out; finished in 0.00s
$ echo $?
0
```
After this fix, the failure is visible. (The entire process is aborted due to #24479.)
```console
$ cargo test
Finished test [unoptimized + debuginfo] target(s) in 0.01s
Running target/debug/deps/panicking_test-76180625bc2ee3c9
running 1 test
thread 'test' panicked at 'explicit panic', src/main.rs:9:9
note: run with `RUST_BACKTRACE=1` environment variable to display a backtrace
fatal runtime error: failed to initiate panic, error 5
error: test failed, to rerun pass '--bin panicking-test'
Caused by:
process didn't exit successfully: `/tmp/panicking-test/target/debug/deps/panicking_test-76180625bc2ee3c9 --nocapture` (signal: 6, SIGABRT: process abort signal)
$ echo $?
101
```
mark raw_vec::ptr with inline
when a lot of vectors is used in a enum as in the example in #66617 if this function is not inlined and multiple cgus is used this results in huge compile times. with this fix the compile time is 6s from minutes for the example in #66617. I did not have the patience to wait for it to compile for more then 3 min.
This reduces the total complexity of checking timeouts from quadratic
to linear, and should also fix an unwrap of None on completion of an
already timed-out test.
Signed-off-by: Anders Kaseorg <andersk@mit.edu>
Prevent query cycles in the MIR inliner
r? `@eddyb` `@wesleywiser`
cc `@rust-lang/wg-mir-opt`
The general design is that we have a new query that is run on the `validated_mir` instead of on the `optimized_mir`. That query is forced before going into the optimization pipeline, so as to not try to read from a stolen MIR.
The query should not be cached cross crate, as you should never call it for items from other crates. By its very design calls into other crates can never cause query cycles.
This is a pessimistic approach to inlining, since we strictly have more calls in the `validated_mir` than we have in `optimized_mir`, but that's not a problem imo.
Otherwise we can miss bugs where a test reports that it succeeded but
then panics within a TLS destructor.
Signed-off-by: Anders Kaseorg <andersk@mit.edu>
Make std::future a re-export of core::future
After 1a764a7ef5, there are no `std::future`-specific items (except for `cfg(bootstrap)` items removed in 93eed402ad). So, instead of defining `std` own module, we can re-export the `core::future` directly.
Implement Error for &(impl Error)
Opening this up just to see what it breaks. It's unfortunate that `&(impl Error)` doesn't actually implement `Error`. If this direct approach doesn't work out then I'll try something different, like an `Error::by_ref` method.
**EDIT:** This is a super low-priority experiment so feel free to cancel it for more important crater runs! 🙂
-----
# Stabilization Report
## Why?
We've been working for the last few years to try "fix" the `Error` trait, which is probably one of the most fundamental in the whole standard library. One of its issues is that we commonly expect you to work with abstract errors through `dyn Trait`, but references and smart pointers over `dyn Trait` don't actually implement the `Error` trait. If you have a `&dyn Error` or a `Box<dyn Error>` you simply can't pass it to a method that wants a `impl Error`.
## What does this do?
This stabilizes the following trait impl:
```rust
impl<'a, T: Error + ?Sized + 'static> Error for &'a T;
```
This means that `&dyn Error` will now satisfy a `impl Error` bound.
It doesn't do anything with `Box<dyn Error>` directly. We discussed how we could do `Box<dyn Error>` in the thread here (and elsewhere in the past), but it seems like we need something like lattice-based specialization or a sprinkling of snowflake compiler magic to make that work. Having said that, with this new impl you _can_ now get a `impl Error` from a `Box<dyn Error>` by dereferencing it.
## What breaks?
A crater run revealed a few crates broke with something like the following:
```rust
// where e: &'short &'long dyn Error
err.source()
```
previously we'd auto-deref that `&'short &'long dyn Error` to return a `Option<&'long dyn Error>` from `source`, but now will call directly on `&'short impl Error`, so will return a `Option<&'short dyn Error>`. The fix is to manually deref:
```rust
// where e: &'short &'long dyn Error
(*err).source()
```
In the recent Libs meeting we considered this acceptable breakage.
Remove delay-binding for Win XP and Vista
The minimum supported Windows version is now Windows 7. Windows XP
and Windows Vista are no longer supported; both are already broken, and
require extra steps to use.
This commit removes the delayed-binding support for Windows API
functions that are present on all supported Windows targets. This has
several benefits: Removes needless complexity. Removes a load and
dynamic call on hot paths in mutex acquire / release. This may have
performance benefits.
* "Drop official support for Windows XP"
https://github.com/rust-lang/compiler-team/issues/378
* "Firefox has ended support for Windows XP and Vista"
https://support.mozilla.org/en-US/kb/end-support-windows-xp-and-vista
Inline methods of Path and OsString
These methods are not generic, and therefore aren't candidates for cross-crate inlining without an `#[inline]` attribute.
More clear documentation for NonNull<T>
Rephrase and hopefully clarify the discussion of covariance in `NonNull<T>` documentation.
I'm very much not an expert so someone should definitely double check the correctness of what I'm saying. At the same time, the new language makes more sense to me, so hopefully it also is more logical to others whose knowledge of covariance basically begins and ends with the [Rustonomicon chapter](https://doc.rust-lang.org/nomicon/subtyping.html).
Related to #48929.
Change branching in `iter.skip()`
Optimize branching in `Skip`, which was brought up in #80416.
This assumes that if `next` is called, it's likely that there will be more calls to `next`, and the branch for skip will only be hit once thus it's unlikely to take that path. Even w/o the `unlikely` intrinsic, it compiles more efficiently, I believe because the path where `next` is called is always taken.
It should be noted there are very few places in the compiler where `Skip` is used, so probably won't have a noticeable perf impact.
[New impl](https://godbolt.org/z/85rdj4)
[Old impl](https://godbolt.org/z/Wc74rh)
[Some additional asm examples](https://godbolt.org/z/feKzoz) although they really don't have a ton of difference between them.
Add doc aliases for memory allocations
This patch adds doc aliases for various C allocation functions, making it possible to search for the C-equivalent of a function and finding the (safe) Rust counterpart:
- `Vec::with_capacity` / `Box::new` / `vec!` -> alloc + malloc, allocates memory
- `Box::new_zeroed` -> calloc, allocates zeroed-out memory
- `Vec::{reserve,reserve_exact,try_reserve_exact,shrink_to_fit,shrink_to}` -> realloc, reallocates a previously allocated slice of memory
It's worth noting that `Vec::new` does not allocate, so we don't link to it. Instead people are probably looking for `Vec::with_capacity` or `vec!`. I hope this will allow people comfortable with the system allocation APIs to make it easier to find what they may be looking for.
Thanks!
Enforce statically that `MIN_NON_ZERO_CAP` is calculated at compile time
Previously, it would usually get computed by LLVM, but this enforces it. This removes the need for the comment saying "LLVM is smart enough".
I don't expect this to make a performance difference, but I do think it makes the performance properties easier to reason about.
Document why not use concat! in dbg! macro
Original title: Reduce code generated by `dbg!` macro
The expanded code before/after: <https://rust.godbolt.org/z/hE3j95>.
---
We cannot use `concat!` since `file!` could contains `{` or the expression is a block (`{ .. }`).
Using it will generated malformed format strings.
So let's document this reason why we don't use `concat!` macro at all.
Expand docs on Iterator::intersperse
Unstable feature in #79524. This expands on the docs to bring them more in line with how other methods of `Iterator` are demonstrated.
The minimum supported Windows version is now Windows 7. Windows XP
and Windows Vista are no longer supported; both are already broken, and
require extra steps to use.
This commit removes the delayed-binding support for Windows API
functions that are present on all supported Windows targets. This has
several benefits: Removes needless complexity. Removes a load and
dynamic call on hot paths in mutex acquire / release. This may have
performance benefits.
* "Drop official support for Windows XP"
https://github.com/rust-lang/compiler-team/issues/378
* "Firefox has ended support for Windows XP and Vista"
https://support.mozilla.org/en-US/kb/end-support-windows-xp-and-vista
