Optimize try_eval_bits to avoid layout queries
This specifically targets match checking, but is possibly more widely
useful as well. In code with large, single-value match statements, we
were previously spending a lot of time running layout_of for the
primitive types (integers, chars) -- which is essentially useless. This
optimizes the code to avoid those query calls by directly obtaining the
size for these types, when possible.
It may be worth considering adding a `size_of` query in the future which
might be far faster, especially if specialized for "const" cases --
match arms being the most obvious example. It's possibly such a function
would benefit from *not* being a query as well, since it's trivially
evaluatable from the sty for many cases whereas a query needs to hash
the input and such.
This specifically targets match checking, but is possibly more widely
useful as well. In code with large, single-value match statements, we
were previously spending a lot of time running layout_of for the
primitive types (integers, chars) -- which is essentially useless. This
optimizes the code to avoid those query calls by directly obtaining the
size for these types, when possible.
It may be worth considering adding a `size_of` query in the future which
might be far faster, especially if specialized for "const" cases --
match arms being the most obvious example. It's possibly such a function
would benefit from *not* being a query as well, since it's trivially
evaluatable from the sty for many cases whereas a query needs to hash
the input and such.
Point at enclosing match when expecting `()` in arm
When encountering code like the following:
```rust
fn main() {
match 3 {
4 => 1,
3 => {
println!("Yep it maches.");
2
}
_ => 2
}
println!("Bye!")
}
```
point at the enclosing `match` expression and suggest ignoring the
returned value:
```
error[E0308]: mismatched types
--> $DIR/match-needing-semi.rs:8:13
|
LL | / match 3 {
LL | | 4 => 1,
LL | | 3 => {
LL | | 2
| | ^ expected (), found integer
LL | | }
LL | | _ => 2
LL | | }
| | -- help: consider using a semicolon here
| |_____|
| expected this to be `()`
|
= note: expected type `()`
found type `{integer}
```
Fix#40799.
panic=abort support in libtest
Add experimental support for tests compiled with panic=abort. Enabled with `-Z panic_abort_tests`.
r? @alexcrichton
cc @cramertj
Rollup of 5 pull requests
Successful merges:
- #63492 (Remove redundancy from the implementation of C variadics.)
- #64589 (Differentiate AArch64 bare-metal targets between hf and non-hf.)
- #64799 (Fix double panic when printing query stack during an ICE)
- #64824 (No StableHasherResult everywhere)
- #64884 (Add pkg-config to dependency list if building for Linux on Linux)
Failed merges:
r? @ghost
No StableHasherResult everywhere
This removes the generic parameter on `StableHasher`, instead moving it to the call to `finish`. This has the side-effect of making all `HashStable` impls nicer, since we no longer need the verbose `<W: StableHasherResult>` that previously existed -- often forcing line wrapping.
This is done for two reasons:
* we should avoid false "generic" dependency on the result of StableHasher
* we don't need to codegen two/three copies of all the HashStable impls when they're transitively used to produce a fingerprint, u64, or u128. I haven't measured, but this might actually make our artifacts somewhat smaller too.
* Easier to understand/read/write code -- the result of the stable hasher is irrelevant when writing a hash impl.
Fix double panic when printing query stack during an ICE
On the latest nightly, any call to `bug` or `span_bug` will result in two panics - the first one as a normal result of calling `bug` / `span_bug`, and the second as a result of trying to print the query stack from the panic handler. This is caused by the query-printing code attempting to acquire a lock on `HandlerInnder`, which is still being held by `bug`.
This PR moves the actual panic out of `HandlerInner`, into `Handler`. This allows us to release the lock on `HandlerInner` before triggering the panic, ensuring that the panic handler will be able to acquire the lock if necessary.
Remove redundancy from the implementation of C variadics.
This cleanup was first described in https://github.com/rust-lang/rust/issues/44930#issuecomment-497163539:
* AST doesn't track `c_variadic: bool` anymore, relying solely on a trailing `CVarArgs` type in fn signatures
* HIR doesn't have a `CVarArgs` anymore, relying solely on `c_variadic: bool`
* same for `ty::FnSig` (see tests for diagnostics improvements from that)
* `{hir,mir}::Body` have one extra argument than the signature when `c_variadic == true`
* `rustc_typeck` and `rustc_mir::{build,borrowck}` need to give that argument the right type (which no longer uses a lifetime parameter, but a function-internal scope)
* `rustc_target::abi::call` doesn't need special hacks anymore (since it never sees the `VaListImpl` now, it's all inside the body)
r? @nagisa / @rkruppe cc @dlrobertson @oli-obk
added more context for duplicate lang item errors (fixes#60561)
Some more information about #60561 -- these errors are pretty common when one works in restrictive environments with `no_std` or customized `std`, but they don't provide much context for debugging, as any transitive dependency could have brought in `std` crate. With that, currently, one needs to use something like `cargo tree` and investigate transitive dependencies one by one.
It'll be more helpful to know at least the crate that uses `std` (which `cargo tree` doesn't show) to pin down this investigation when debugging.
I'm not sure what the best way to get this context is inside rustc internals (I'm new to them). I found that `all_crate_nums` query returns the crates in some dependency order, so printing out the name of the preceding crate seems to do the trick. But I welcome suggestions if this can be done in a better way.
Also adds an unstable flag to disable the ICE
(`-Zsuppress-const-validation-back-compat-ice`) so that nightly users do
not have to revert to a previous nightly if their code causes
disagreement between the validators.
check_match: improve diagnostics for `let A = 2;` with `const A: i32 = 3`
For example:
```
error[E0005]: refutable pattern in local binding: `std::i32::MIN..=1i32` and `3i32..=std::i32::MAX` not covered
--> $DIR/const-pat-non-exaustive-let-new-var.rs:2:9
|
LL | let A = 3;
| ^
| |
| interpreted as a constant pattern, not a new variable
| help: introduce a variable instead: `a_var`
...
LL | const A: i32 = 2;
| ----------------- constant defined here
```
r? @estebank
cc @matthiaskrgr @rpjohnst
hir: Disallow `target_feature` on constants
Fixes#64768.
This PR fixes an ICE when `#[target_feature]` is applied to constants by disallowing this with the same error as when `#[target_feature]` is applied to other places it shouldn't be.
I couldn't see anything in the [RFC](https://github.com/rust-lang/rfcs/blob/master/text/2045-target-feature.md) that suggested that `#[target_feature]` should be applicable to constants or any tests that suggested it should, though I might have missed something - if this is desirable in future, it remains possible to remove this error (but for the time being, I think this error is better than an ICE).
I also added some extra cases to the test for other places where `#[target_feature]` should not be permitted.
cc @gnzlbg
Account for tail expressions when pointing at return type
When there's a type mismatch we make an effort to check if it was
caused by a function's return type. This logic now makes sure to
only point at the return type if the error happens in a tail
expression.
Turn `walk_parent_nodes` method into an iterator.
CC #39968, CC #40799.
Remove unused DepTrackingMap
Deletes some related code (MemoizationMap trait, etc.)
I believe this became unused with red/green incremental introduction, but am uncertain.
There's no longer two distinct gcx and tcx lifetimes which made this
necessary (or, at least, the code compiles -- it's possible we got
better at normalizing, but that seems unlikely).
When there's a type mismatch we make an effort to check if it was
caused by a function's return type. This logic now makes sure to
only point at the return type if the error happens in a tail
expression.
