Avoid a collection and iteration on empty passes
Just some mini optimization I saw in the wild. This way, we avoid a `collect` and `map` on an empty `passes`. Honestly, I don't even think this is big enough of a change to make a benchmark, but I'd still like to see results.
Based on [this book](https://nnethercote.github.io/perf-book/iterators.html#collect-and-extend)
[rustdoc] Correctly generate path for non-local items in source code pages
While browsing some crates using the "jump to def" feature, I realized that a lot of items didn't have a link generated. The reason is because we only cache foreign items if they appear in the documented API. This means that for the others, we need to infer them.
r? ``@notriddle``
For a rigid projection, recursively look at the self type's item bounds to fix the `associated_type_bounds` feature
Given a deeply nested rigid projection like `<<<T as Trait1>::Assoc1 as Trait2>::Assoc2 as Trait3>::Assoc3`, this PR adjusts both trait solvers to look at the item bounds for all of `Assoc3`, `Assoc2`, and `Assoc1` in order to satisfy a goal. We do this because the item bounds for projections may contain relevant bounds for *other* nested projections when the `associated_type_bounds` (ATB) feature is enabled. For example:
```rust
#![feature(associated_type_bounds)]
trait Trait1 {
type Assoc1: Trait2<Assoc2: Foo>;
// Item bounds for `Assoc1` are:
// `<Self as Trait1>::Assoc1: Trait2`
// `<<Self as Trait1>::Assoc1 as Trait2>::Assoc2: Foo`
}
trait Trait2 {
type Assoc2;
}
trait Foo {}
fn hello<T: Trait1>(x: <<T as Trait1>::Assoc1 as Trait2>::Assoc2) {
fn is_foo(_: impl Foo) {}
is_foo(x);
// Currently fails with:
// ERROR the trait bound `<<Self as Trait1>::Assoc1 as Trait2>::Assoc2: Foo` is not satisfied
}
```
This has been a long-standing place of brokenness for ATBs, and is also part of the reason why ATBs currently desugar so differently in various positions (i.e. sometimes desugaring to param-env bounds, sometimes desugaring to RPITs, etc). For example, in RPIT and TAIT position, `impl Foo<Bar: Baz>` currently desugars to `impl Foo<Bar = impl Baz>` because we do not currently take advantage of these nested item bounds if we desugared them into a single set of item bounds on the opaque. This is obviously both strange and unnecessary if we just take advantage of these bounds as we should.
## Approach
This PR repeatedly peels off each projection of a given goal's self type and tries to match its item bounds against a goal, repeating with the self type of the projection. This is pretty straightforward to implement in the new solver, only requiring us to loop on the self type of a rigid projection to discover inner rigid projections, and we also need to introduce an extra probe so we can normalize them.
In the old solver, we can do essentially the same thing, however we rely on the fact that projections *should* be normalized already. This is obviously not always the case -- however, in the case that they are not fully normalized, such as a projection which has both infer vars and, we bail out with ambiguity if we hit an infer var for the self type.
## Caveats
⚠️ In the old solver, this has the side-effect of actually stalling some higher-ranked trait goals of the form `for<'a> <?0 as Tr<'a>>: Tr2`. Because we stall them, they no longer are eagerly treated as error -- this cause some existing `known-bug` tests to go from fail -> pass.
I'm pretty unconvinced that this is a problem since we make code that we expect to pass in the *new* solver also pass in the *old* solver, though this obviously doesn't solve the *full* problem.
## And then also...
We also adjust the desugaring of ATB to always desugar to a regular associated bound, rather than sometimes to an impl Trait **except** for when the ATB is present in a `dyn Trait`. We need to lower `dyn Trait<Assoc: Bar>` to `dyn Trait<Assoc = impl Bar>` because object types need all of their associated types specified.
I would also be in favor of splitting out the ATB feature and/or removing support for object types in order to stabilize just the set of positions for which the ATB feature is consistent (i.e. always elaborates to a bound).
Build DebugInfo for async closures
The test is pretty bare, because I don't really know how to write debuginfo tests. I'd like to land this first, and then flesh it out correctly one it's no longer ICEing on master (which breaks people's ability to test using async closures).
r? oli-obk cc `@rust-lang/wg-debugging` (if any of y'all want to help me write a more fleshed out async closures test)
Emit more specific diagnostics when enums fail to cast with `as`
Fixes#120756
Changes this diagnostic reported in the issue:
```
error[E0605]: non-primitive cast: `Bad` as `u32`
--> src/main.rs:18:10
|
18 | dbg!(bad as u32);
| ^^^^^^^^^^ an `as` expression can only be used to convert between primitive types or to coerce to a specific trait object
```
to this:
```
error[E0605]: non-primitive cast: `Bad` as `u32`
--> src/main.rs:18:10
|
18 | dbg!(bad as u32);
| ^^^^^^^^^^ an `as` expression can be used to convert enum types to numeric types only if the enum type is unit-only or field-less
|
= note: see https://doc.rust-lang.org/reference/items/enumerations.html#casting for more information
```
This change is only for enums. The diagnostic remains unchanged for all other cases.
Improve `Option::inspect` docs
* Refer to the function as "a function" instead of "the provided closure" since it is not necessarily a closure.
* State that the original Option/Result is returned.
* Adjust the example for `Option::inspect` to use chaining.
improve normalization of `Pointee::Metadata`
This PR makes it so that `<Wrapper<Tail> as Pointee>::Metadata` is normalized to `<Tail as Pointee>::Metadata` if we don't know `Wrapper<Tail>: Sized`. With that, the trait solver can prove projection predicates like `<Wrapper<Tail> as Pointee>::Metadata == <Tail as Pointee>::Metadata`, which makes it possible to use the metadata APIs to cast between the tail and the wrapper:
```rust
#![feature(ptr_metadata)]
use std::ptr::{self, Pointee};
fn cast_same_meta<T: ?Sized, U: ?Sized>(ptr: *const T) -> *const U
where
T: Pointee<Metadata = <U as Pointee>::Metadata>,
{
let (thin, meta) = ptr.to_raw_parts();
ptr::from_raw_parts(thin, meta)
}
struct Wrapper<T: ?Sized>(T);
fn cast_to_wrapper<T: ?Sized>(ptr: *const T) -> *const Wrapper<T> {
cast_same_meta(ptr)
}
```
Previously, this failed to compile:
```
error[E0271]: type mismatch resolving `<Wrapper<T> as Pointee>::Metadata == <T as Pointee>::Metadata`
--> src/lib.rs:16:5
|
15 | fn cast_to_wrapper<T: ?Sized>(ptr: *const T) -> *const Wrapper<T> {
| - found this type parameter
16 | cast_same_meta(ptr)
| ^^^^^^^^^^^^^^ expected `Wrapper<T>`, found type parameter `T`
|
= note: expected associated type `<Wrapper<T> as Pointee>::Metadata`
found associated type `<T as Pointee>::Metadata`
= note: an associated type was expected, but a different one was found
```
(Yes, you can already do this with `as` casts. But using functions is so much ✨ *safer* ✨, because you can't change the metadata on accident.)
---
This PR essentially changes the built-in impls of `Pointee` from this:
```rust
// before
impl Pointee for u8 {
type Metadata = ();
}
impl Pointee for [u8] {
type Metadata = usize;
}
// ...
impl Pointee for Wrapper<u8> {
type Metadata = ();
}
impl Pointee for Wrapper<[u8]> {
type Metadata = usize;
}
// ...
// This impl is only selected if `T` is a type parameter or unnormalizable projection or opaque type.
fallback impl<T: ?Sized> Pointee for Wrapper<T>
where
Wrapper<T>: Sized
{
type Metadata = ();
}
// This impl is only selected if `T` is a type parameter or unnormalizable projection or opaque type.
fallback impl<T /*: Sized */> Pointee for T {
type Metadata = ();
}
```
to this:
```rust
// after
impl Pointee for u8 {
type Metadata = ();
}
impl Pointee for [u8] {
type Metadata = usize;
}
// ...
impl<T: ?Sized> Pointee for Wrapper<T> {
// in the old solver this will instead project to the "deep" tail directly,
// e.g. `Wrapper<Wrapper<T>>::Metadata = T::Metadata`
type Metadata = <T as Pointee>::Metadata;
}
// ...
// This impl is only selected if `T` is a type parameter or unnormalizable projection or opaque type.
fallback impl<T /*: Sized */> Pointee for T {
type Metadata = ();
}
```
Startup objects disappearing from sysroot
When launching tests with --keep-stage option, startup objects such as rsbegin.o an rsend.o may disappear from the corresponding stageN compiler.
Fix issue #120784
Fix `ErrorGuaranteed` unsoundness with stash/steal.
When you stash an error, the error count is incremented. You can then use the non-zero error count to get an `ErrorGuaranteed`. You can then steal the error, which decrements the error count. You can then cancel the error.
Example code:
```
fn unsound(dcx: &DiagCtxt) -> ErrorGuaranteed {
let sp = rustc_span::DUMMY_SP;
let k = rustc_errors::StashKey::Cycle;
dcx.struct_err("bogus").stash(sp, k); // increment error count on stash
let guar = dcx.has_errors().unwrap(); // ErrorGuaranteed from error count > 0
let err = dcx.steal_diagnostic(sp, k).unwrap(); // decrement error count on steal
err.cancel(); // cancel error
guar // ErrorGuaranteed with no error emitted!
}
```
This commit fixes the problem in the simplest way: by not counting stashed errors in `DiagCtxt::{err_count,has_errors}`.
However, just doing this without any other changes leads to over 40 ui test failures. Mostly because of uninteresting extra errors (many saying "type annotations needed" when type inference fails), and in a few cases, due to delayed bugs causing ICEs when no normal errors are printed.
To fix these, this commit adds `DiagCtxt::stashed_err_count`, and uses it in three places alongside `DiagCtxt::{has_errors,err_count}`. It's dodgy to rely on it, because unlike `DiagCtxt::err_count` it can go up and down. But it's needed to preserve existing behaviour, and at least the three places that need it are now obvious.
r? oli-obk
Fix more `ty::Error` ICEs in MIR passes
Fixes#120791 - Add a check for `ty::Error` in the `ByMove` coroutine pass
Fixes#120816 - Add a check for `ty::Error` in the MIR validator
Also a drive-by fix for a FIXME I had asked oli to add
r? oli-obk
A drive-by rewrite of `give_region_a_name()`
This drive-by rewrite makes the cache-updating nature of the method clearer, using the Entry API into the hash table for region names to capture the update-insert nature of the method. May be marginally more efficient since it only runtime-borrows and indexes the map once, but in this context the performance impact is almost certainly completely negligible.
Note that this commit should preserve all externally visible behaviour. Notably, it preserves the debug logging:
1. printing even in the case of a `None` for the new computed name, and
2. only printing on new values, begin silent on reused values
Invert diagnostic lints.
That is, change `diagnostic_outside_of_impl` and `untranslatable_diagnostic` from `allow` to `deny`, because more than half of the compiler has been converted to use translated diagnostics.
This commit removes more `deny` attributes than it adds `allow` attributes, which proves that this change is warranted.
r? ````@davidtwco````
core/time: avoid divisions in Duration::new
In our (decently large) code base, we use `SystemTime::UNIX_EPOCH.elapsed()` in a lot of places & often in a loop or in the hot path. On [Unix](https://github.com/rust-lang/rust/blob/1.75.0/library/std/src/sys/unix/time.rs#L153-L162) at least, it seems we do calculations before hand to ensure that nanos is within the valid range, yet `Duration::new()` still checks it again, using 2 divisions. It seems like adding a branch can make this function 33% faster on ARM64 in the cases where nanos is already in the valid range & seems to have no effect in the other case.
Benchmarks:
M1 Pro (14-inch base model):
```
duration/current/checked
time: [1.5945 ns 1.6167 ns 1.6407 ns]
Found 5 outliers among 100 measurements (5.00%)
2 (2.00%) high mild
3 (3.00%) high severe
duration/current/unchecked
time: [1.5941 ns 1.6051 ns 1.6179 ns]
Found 2 outliers among 100 measurements (2.00%)
1 (1.00%) high mild
1 (1.00%) high severe
duration/branched/checked
time: [1.1997 ns 1.2048 ns 1.2104 ns]
Found 8 outliers among 100 measurements (8.00%)
4 (4.00%) high mild
4 (4.00%) high severe
duration/branched/unchecked
time: [1.5881 ns 1.5957 ns 1.6039 ns]
Found 6 outliers among 100 measurements (6.00%)
3 (3.00%) high mild
3 (3.00%) high severe
```
EC2 c7gd.16xlarge (Graviton 3):
```
duration/current/checked
time: [2.7996 ns 2.8000 ns 2.8003 ns]
Found 5 outliers among 100 measurements (5.00%)
2 (2.00%) low severe
3 (3.00%) low mild
duration/current/unchecked
time: [2.9922 ns 2.9925 ns 2.9928 ns]
Found 7 outliers among 100 measurements (7.00%)
4 (4.00%) low severe
1 (1.00%) low mild
2 (2.00%) high mild
duration/branched/checked
time: [2.0830 ns 2.0843 ns 2.0857 ns]
Found 3 outliers among 100 measurements (3.00%)
1 (1.00%) low severe
1 (1.00%) low mild
1 (1.00%) high mild
duration/branched/unchecked
time: [2.9879 ns 2.9886 ns 2.9893 ns]
Found 5 outliers among 100 measurements (5.00%)
3 (3.00%) low severe
2 (2.00%) low mild
```
EC2 r7iz.16xlarge (Intel Xeon Scalable-based (Sapphire Rapids)):
```
duration/current/checked
time: [980.60 ps 980.79 ps 980.99 ps]
Found 10 outliers among 100 measurements (10.00%)
4 (4.00%) low severe
2 (2.00%) low mild
3 (3.00%) high mild
1 (1.00%) high severe
duration/current/unchecked
time: [979.53 ps 979.74 ps 979.96 ps]
Found 6 outliers among 100 measurements (6.00%)
2 (2.00%) low severe
1 (1.00%) low mild
2 (2.00%) high mild
1 (1.00%) high severe
duration/branched/checked
time: [938.72 ps 938.96 ps 939.22 ps]
Found 4 outliers among 100 measurements (4.00%)
1 (1.00%) low mild
1 (1.00%) high mild
2 (2.00%) high severe
duration/branched/unchecked
time: [1.0103 ns 1.0110 ns 1.0118 ns]
Found 10 outliers among 100 measurements (10.00%)
2 (2.00%) low mild
7 (7.00%) high mild
1 (1.00%) high severe
```
Bench code (ran using stable 1.75.0 & criterion latest 0.5.1):
I couldn't find any benches for `Duration` in this repo, so I just copied the relevant types & recreated it.
```rust
use criterion::{black_box, criterion_group, criterion_main, Criterion};
pub fn duration_bench(c: &mut Criterion) {
const NANOS_PER_SEC: u32 = 1_000_000_000;
#[derive(Clone, Copy, PartialEq, Eq, PartialOrd, Ord, Hash)]
#[repr(transparent)]
struct Nanoseconds(u32);
impl Default for Nanoseconds {
#[inline]
fn default() -> Self {
// SAFETY: 0 is within the valid range
unsafe { Nanoseconds(0) }
}
}
#[derive(Clone, Copy, PartialEq, Eq, PartialOrd, Ord, Hash, Default)]
pub struct Duration {
secs: u64,
nanos: Nanoseconds, // Always 0 <= nanos < NANOS_PER_SEC
}
impl Duration {
#[inline]
pub const fn new_current(secs: u64, nanos: u32) -> Duration {
let secs = match secs.checked_add((nanos / NANOS_PER_SEC) as u64) {
Some(secs) => secs,
None => panic!("overflow in Duration::new"),
};
let nanos = nanos % NANOS_PER_SEC;
// SAFETY: nanos % NANOS_PER_SEC < NANOS_PER_SEC, therefore nanos is within the valid range
Duration { secs, nanos: unsafe { Nanoseconds(nanos) } }
}
#[inline]
pub const fn new_branched(secs: u64, nanos: u32) -> Duration {
if nanos < NANOS_PER_SEC {
// SAFETY: nanos < NANOS_PER_SEC, therefore nanos is within the valid range
Duration { secs, nanos: unsafe { Nanoseconds(nanos) } }
} else {
let secs = match secs.checked_add((nanos / NANOS_PER_SEC) as u64) {
Some(secs) => secs,
None => panic!("overflow in Duration::new"),
};
let nanos = nanos % NANOS_PER_SEC;
// SAFETY: nanos % NANOS_PER_SEC < NANOS_PER_SEC, therefore nanos is within the valid range
Duration { secs, nanos: unsafe { Nanoseconds(nanos) } }
}
}
}
let mut group = c.benchmark_group("duration/current");
group.bench_function("checked", |b| {
b.iter(|| black_box(Duration::new_current(black_box(1_000_000_000), black_box(1_000_000))));
});
group.bench_function("unchecked", |b| {
b.iter(|| {
black_box(Duration::new_current(black_box(1_000_000_000), black_box(2_000_000_000)))
});
});
drop(group);
let mut group = c.benchmark_group("duration/branched");
group.bench_function("checked", |b| {
b.iter(|| {
black_box(Duration::new_branched(black_box(1_000_000_000), black_box(1_000_000)))
});
});
group.bench_function("unchecked", |b| {
b.iter(|| {
black_box(Duration::new_branched(black_box(1_000_000_000), black_box(2_000_000_000)))
});
});
}
criterion_group!(duration_benches, duration_bench);
criterion_main!(duration_benches);
```
Make privacy visitor use types more (instead of HIR)
r? ``@petrochenkov``
This is a prerequisite to normalizing projections, as otherwise we have too many invalid bound vars (hir_ty_to_ty is creating types that have bound vars, but no binder).
The commits are still chaotic, I'm gonna clean them up, but I just wanted to let you know about the general direction and wondering if we could land this before adding normalization, as normalization is where behavioral changes happen, and I'd like to keep that part as minimal as possible.
[context can be found on zulip](https://rust-lang.zulipchat.com/#narrow/stream/315482-t-compiler.2Fetc.2Fopaque-types/topic/weak.20type.20aliases.20and.20privacy)
Toggle assert_unsafe_precondition in codegen instead of expansion
The goal of this PR is to make some of the unsafe precondition checks in the standard library available in debug builds. Some UI tests are included to verify that it does that.
The diff is large, but most of it is blessing mir-opt tests and I've also split up this PR so it can be reviewed commit-by-commit.
This PR:
1. Adds a new intrinsic, `debug_assertions` which is lowered to a new MIR NullOp, and only to a constant after monomorphization
2. Rewrites `assume_unsafe_precondition` to check the new intrinsic, and be monomorphic.
3. Skips codegen of the `assume` intrinsic in unoptimized builds, because that was silly before but with these checks it's *very* silly
4. The checks with the most overhead are `ptr::read`/`ptr::write` and `NonNull::new_unchecked`. I've simply added `#[cfg(debug_assertions)]` to the checks for `ptr::read`/`ptr::write` because I was unable to come up with any (good) ideas for decreasing their impact. But for `NonNull::new_unchecked` I found that the majority of callers can use a different function, often a safe one.
Yes, this PR slows down the compile time of some programs. But in our benchmark suite it's never more than 1% icount, and the average icount change in debug-full programs is 0.22%. I think that is acceptable for such an improvement in developer experience.
https://github.com/rust-lang/rust/issues/120539#issuecomment-1922687101
Always check the result of `pthread_mutex_lock`
Fixes#120147.
Instead of manually adding a list of "good" platforms, I've simply made the check unconditional. pthread's mutex is already quite slow on most platforms, so one single well-predictable branch shouldn't hurt performance too much.
When launching tests with --keep-stage option, startup objects
such as rsbegin.o an rsend.o may disappear from the corresponding
stageN compiler.
Fix issue #120784
When you stash an error, the error count is incremented. You can then
use the non-zero error count to get an `ErrorGuaranteed`. You can then
steal the error, which decrements the error count. You can then cancel
the error.
Example code:
```
fn unsound(dcx: &DiagCtxt) -> ErrorGuaranteed {
let sp = rustc_span::DUMMY_SP;
let k = rustc_errors::StashKey::Cycle;
dcx.struct_err("bogus").stash(sp, k); // increment error count on stash
let guar = dcx.has_errors().unwrap(); // ErrorGuaranteed from error count > 0
let err = dcx.steal_diagnostic(sp, k).unwrap(); // decrement error count on steal
err.cancel(); // cancel error
guar // ErrorGuaranteed with no error emitted!
}
```
This commit fixes the problem in the simplest way: by not counting
stashed errors in `DiagCtxt::{err_count,has_errors}`.
However, just doing this without any other changes leads to over 40 ui
test failures. Mostly because of uninteresting extra errors (many saying
"type annotations needed" when type inference fails), and in a few
cases, due to delayed bugs causing ICEs when no normal errors are
printed.
To fix these, this commit adds `DiagCtxt::stashed_err_count`, and uses
it in three places alongside `DiagCtxt::{has_errors,err_count}`. It's
dodgy to rely on it, because unlike `DiagCtxt::err_count` it can go up
and down. But it's needed to preserve existing behaviour, and at least
the three places that need it are now obvious.
