Add a test for #37333
The test checks that we reuse the CGU of a crate when the implementation
details of an `extern crate` have changed.
Signed-off-by: Michal Terepeta <michal.terepeta@gmail.com>
no longer promote non-pattern const functions
This is trying to pack-pedal a bit on promotion feature creep, as proposed by @eddyb [here](https://github.com/rust-lang/const-eval/issues/19#issuecomment-472799062): possibly, a sane subset of `const fn` that we could promote are those that are just constructors -- the same subset that we might want to allow in pattern position at some point.
So, this removes the `rustc_promotable` attribute from the three functions they identified that do not fit this pattern. The first step is to run crater to see if there is code in the wild that relies on this being promotable.
r? @oli-obk
Suggest calling method when first argument is `self`
Closes: #66782
I've explored different approaches for this MR but I think the most straightforward is the best one.
I've tried to find out if the methods for given type exist (to maybe have a better suggestion), but we don't collect them anywhere and collecting them is quite problematic. Moreover, collecting all the methods would require rewriting big part of the code and also could potentially include performance degradation, which I don't think is necessary for this simple case.
- Use the span of the predicate
- Use the def's description instead of "struct/enum" (notably incorrect
for unions)
- Align formatting with other error messages
Attempt to fix intermittent failures of pgo-branch-weights test.
This PR tries to fix the intermittent failures of the pgo-branch-weights test (https://github.com/rust-lang/rust/issues/67746). The failing instances show no `!prof` annotations in LLVM IR. One possible cause is that the instrumented binary did not record anything. This is something I've occasionally seen happen for similarly small programs when using GNU ld as linker. The linker would not properly append the instruction counter sections, leading to most counters being dropped. This PR makes the test use the Gold linker instead.
It also makes each command exit immediately on failure so we can pinpoint the failure source better, should there still be a problem.
r? @Mark-Simulacrum
Rollup of 10 pull requests
Successful merges:
- #67450 (Allow for setting a ThinLTO import limit during bootstrap)
- #67595 (Suggest adding a lifetime constraint for opaque type)
- #67636 (allow rustfmt key in [build] section)
- #67736 (Less-than is asymmetric, not antisymmetric)
- #67762 (Add missing links for insecure_time)
- #67783 (Warn for bindings named same as variants when matching against a borrow)
- #67796 (Ensure that we process projections during MIR inlining)
- #67807 (Use drop instead of the toilet closure `|_| ()`)
- #67816 (Clean up err codes)
- #67825 (Minor: change take() docs grammar to match other docs)
Failed merges:
r? @ghost
Ensure that we process projections during MIR inlining
Fixes#67710
Previously, we were not calling `super_place`, which resulted in us
failing to update any local references that occur in
ProjectionElem::Index. This caused the post-inlining MIR to contain a
reference to a local ID from the inlined callee, leading to an ICE
due to a type mismatch.
allow rustfmt key in [build] section
Permit using `rustfmt` in `config.toml`. It will allow to not download `rustfmt` binary, which is not possible for at least some tiers-3 platforms.
Fixes: #67624
r? @Mark-Simulacrum
Suggest adding a lifetime constraint for opaque type
Fixes#67577, where code like this:
```
struct List {
data: Vec<String>,
}
impl List {
fn started_with<'a>(&'a self, prefix: &'a str) -> impl Iterator<Item=&'a str> {
self.data.iter().filter(|s| s.starts_with(prefix)).map(|s| s.as_ref())
}
}
```
will show this error:
```
Compiling playground v0.0.1 (/playground)
error[E0597]: `prefix` does not live long enough
--> src/lib.rs:6:47
|
5 | fn started_with<'a>(&'a self, prefix: &'a str) -> impl Iterator<Item=&'a str> {
| -- lifetime `'a` defined here --------------------------- opaque type requires that `prefix` is borrowed for `'a`
...
```
but without suggesting the lovely `help: you can add a constraint..`.
r? @estebank
Allow for setting a ThinLTO import limit during bootstrap
The benchmarks in https://github.com/rust-lang/rust/pull/66625 have shown that a lower ThinLTO import limit can be a net win for bootstrap times. This PR:
- exposes the setting to `config.toml`,
- defaults to a lower limit if `incremental = true` in `config.toml`, and
- sets a lower limit for `x86_64-gnu-llvm-7` CI image in order to make the jobs complete more quickly (which remains to be tested).
This setting will affect how the compiler and it's tools are compiled. It will not affect the settings the compiler uses when compiling user code.
r? @pietroalbini
cc @rust-lang/infra
A bit speculative at this stage but profiling shows that type folding
takes up a substantial amount of time during normalization which may
indicate that many types may be folded despite there being nothing to
normalize
Add Iterator::try_find
I found a need for this fn, and created this PR.
Tracking issue: #63178
I did a fair amount of thinking about the function name, and settled on the current one.
I don't see other anything else that's non-trivial here, but I'm open for debate. I just want this functionality to be there.
It couples with the `collect` trick for collecting `Result<Vec<T>, E>` from `Iterator<Item = Result<T, E>>`.
UPD:
I've already looked at `fallible_iterator` crate, but I don't think it supports my use case.
The main problem is that I can't construct a failable iterator. I have a regular iterator, and I just need to apply a predicate that can fail via `find` method.
UPD: `fallible_iterator` would work, but it's not elegant cause I'd have to make a failable iterator by mapping iterator with `Result::Ok` first.