NLL: Limit two-phase borrows to autoref-introduced borrows
This imposes a restriction on two-phase borrows so that it only applies to autoref-introduced borrows.
The goal is to ensure that our initial deployment of two-phase borrows is very conservative. We want it to still cover the `v.push(v.len());` example, but we do not want it to cover cases like `let imm = &v; let mu = &mut v; mu.push(imm.len());`
(Why do we want it to be conservative? Because when you are not conservative, then the results you get, at least with the current analysis, are tightly coupled to details of the MIR construction that we would rather remain invisible to the end user.)
Fix#46747
I decided, for this PR, to add a debug-flag `-Z two-phase-beyond-autoref`, to re-enable the more general approach. But my intention here is *not* that we would eventually turn on that debugflag by default; the main reason I added it was that I thought it was useful for writing tests to be able to write source that looks like desugared MIR.
This commit disallows acquiring a function pointer to functions tagged as
`#[rustc_args_required_const]`. This is intended to be used as future-proofing
for the stdsimd crate to avoid taking a function pointer to any intrinsic which
has a hard requirement that one of the arguments is a constant value.
The x86_64 SysV ABI should use exact sizes for small structs passed in
registers, i.e. a struct that occupies 3 bytes should use an i24,
instead of the i32 it currently uses.
Refs #45543
Added `-Z two-phase-beyond-autoref` to bring back old behavior (mainly
to allow demonstration of desugared examples).
Updated tests to use aforementioned flag when necessary. (But in each
case where I added the flag, I made sure to also include a revision
without the flag so that one can readily see what the actual behavior
we expect is for the initial deployment of NLL.)
Fixes the hash test to recognize that MirValidated can change when changing
around labels, and add a new test that makes sure we're lowering loop statements
correctly.
As opposed to using weirdness involving pretending the body block
is the loop block. This does not pass tests
This commit is [ci skip] because I know it doesn't pass tests yet.
Somehow this commit introduces nondeterminism into the handling of
loops.
Warn about more ignored bounds in type aliases
It seems that all bounds in type aliases are entirely ignored, not just type bounds. This extends the warning appropriately.
I assume this should be made a hard error with the next epoch? I can't see any reason to accept these programs. (And suddenly enforcing these type bounds would be a breaking change.)
rustc: Add `#[rustc_args_required_const]`
This commit adds a new unstable attribute to the compiler which requires that
arguments to a function are always provided as constants. The primary use case
for this is SIMD intrinsics where arguments are defined by vendors to be
constant and in LLVM they indeed must be constant as well.
For now this is mostly just a semantic guarantee in rustc that an argument is a
constant when invoked, phases like trans don't actually take advantage of it
yet. This means that we'll be able to use this in stdsimd but we won't be able
to remove the `constify_*` macros just yet. Hopefully soon though!
Implement RFC 2052 (Epochs)
This adds -Zepochs and uses it for tyvar_behind_raw_pointer (#46906)
When we move this to --epoch=XXX, we'll need to gate the 2018 epoch on nightly, but not the 2015 one. I can make these changes here itself though it's kinda pointless given that the entire flag is nightly-only.
r? @nikomatsakis @aturon
cc #44581 (epoch tracking)
cc #46906 (tyvar_behind_raw_pointer)
Implement TrustedLen for Take<Repeat> and Take<RangeFrom>
This will allow optimization of simple `repeat(x).take(n).collect()` iterators, which are currently not vectorized and have capacity checks.
This will only support a few aggregates on `Repeat` and `RangeFrom`, which might be enough for simple cases, but doesn't optimize more complex ones. Namely, Cycle, StepBy, Filter, FilterMap, Peekable, SkipWhile, Skip, FlatMap, Fuse and Inspect are not marked `TrustedLen` when the inner iterator is infinite.
Previous discussion can be found in #47082
r? @alexcrichton
Add filtering options to `rustc_on_unimplemented`
- Add filtering options to `rustc_on_unimplemented` for local traits, filtering on `Self` and type arguments.
- Add a way to provide custom notes.
- Tweak binops text.
- Add filter to detect wether `Self` is local or belongs to another crate.
- Add filter to `Iterator` diagnostic for `&str`.
Partly addresses #44755 with a different syntax, as a first approach. Fixes#46216, fixes#37522, CC #34297, #46806.
[NLL] Improve DefiningTy::Const
Fixes#47590 by fixing the way DefiningTy represents constants. Previously, constants were represented using just the type of the variable. However, this will fail to capture early-bound regions as NLL inference vars, resulting in an ICE when we try to compute region VIDs a little bit later in the universal
region resolution process. (ref #47590)
Commit 7ed00caacc improved our error reporting by including the target
function in our error messages when there is an argument count mismatch.
A simple example from the UI tests is:
```
error[E0593]: function is expected to take a single 2-tuple as argument, but it takes 0 arguments
--> $DIR/closure-arg-count.rs:32:53
|
32 | let _it = vec![1, 2, 3].into_iter().enumerate().map(foo);
| ^^^ expected function that takes a single 2-tuple as argument
...
44 | fn foo() {}
| -------- takes 0 arguments
```
However, this assumed the target span was always available. This does
not hold true if the target function is in `std` or another crate. A
simple example from #48046 is assigning `str::split` to a function type
with a different number of arguments.
Fix by removing all of the labels and suggestions related to the target
span when it's not found.
Fixes#48046
Fixes#47590 by fixing the way DefiningTy represents constants. Previously,
constants were represented using just the type of the variable. However, this
will fail to capture early-bound regions as NLL inference vars, resulting in an
ICE when we try to compute region VIDs a little bit later in the universal
region resolution process.
MIR-borrowck: augmented assignment causes duplicate errors
Fixes#45697. This PR resolves the error duplication. I attempted to replace the existing sets since there were quite a few but only managed to replace two of them.
r? @nikomatsakis
Warn when rustc output conflicts with existing directories
When the compiled executable would conflict with a directory, display a
rustc error instead of a verbose and potentially-confusing linker
error. This is a usability improvement, and doesn’t actually change
behaviour with regards to compilation success. This addresses the
concern in #35887. Fixes#13098.
Previously, when the type of a method receiver could not be determined,
the error message would, potentially confusingly, highlight the span of
the entire method call.
Resolves#36598, resolves#42234.
This commit adds a new unstable attribute to the compiler which requires that
arguments to a function are always provided as constants. The primary use case
for this is SIMD intrinsics where arguments are defined by vendors to be
constant and in LLVM they indeed must be constant as well.
For now this is mostly just a semantic guarantee in rustc that an argument is a
constant when invoked, phases like trans don't actually take advantage of it
yet. This means that we'll be able to use this in stdsimd but we won't be able
to remove the `constify_*` macros just yet. Hopefully soon though!
save-analysis: avoid implicit unwrap
When looking up a field defintion, since the name might be incorrect in the field init shorthand case.
cc https://github.com/rust-lang-nursery/rls/issues/699
r? @eddyb
add documentation from doc(include) to analysis data
cc #44732
Currently save-analysis only loads docs from plain doc comments and doc attributes. Since `#[doc(include="filename.md")]` doesn't create a plain doc attribute when it loads the file, we need to be sure to pick up this info for the analysis data.
Fix overflow when performing drop check calculations in NLL
Clearing out the infcx's region constraints after processing each type
ends up interacting badly with normalizing associated types. This commit
keeps all region constraints intact until the end of
TypeLivenessGenerator.add_drop_live_constraint, ensuring that normalized
types are able to re-use existing inference variables.
Fixes#47589
