BTreeMap: disentangle Drop implementation from IntoIter
No longer require every `BTreeMap` to dig up its last leaf edge before dying. This speeds up the `clone_` benchmarks by 25% for normal keys and values (far less for huge values).
r? `@Mark-Simulacrum`
Reduce log level used by tracing instrumentation from info to debug
Restore log level to debug to avoid make info log level overly verbose (the uses of instrument attribute modified there, were for the most part a replacement for `debug!`; one use was novel).
Optimize Vec::retain
Use `copy_non_overlapping` instead of `swap` to reduce memory writes, like what we've done in #44355 and `String::retain`.
#48065 already tried to do this optimization but it is reverted in #67300 due to bad codegen of `DrainFilter::drop`.
This PR re-implement the drop-then-move approach. I did a [benchmark](https://gist.github.com/oxalica/3360eec9376f22533fcecff02798b698) on small-no-drop, small-need-drop, large-no-drop elements with different predicate functions. It turns out that the new implementation is >20% faster in average for almost all cases. Only 2/24 cases are slower by 3% and 5%. See the link above for more detail.
I think regression in may-panic cases is due to drop-guard preventing some optimization. If it's permitted to leak elements when predicate function of element's `drop` panic, the new implementation should be almost always faster than current one.
I'm not sure if we should leak on panic, since there is indeed an issue (#52267) complains about it before.
resolve: Cleanup visibility resolution for enum variants and trait items
by always delegating it to `fn resolve_visibility`.
Also remove some special treatment of visibility on enum variants and trait items remaining from pre-`pub(restricted)` times.
Special treatment like this was necessary before `pub(restricted)` had been implemented and only two visibilities existed - `pub` and non-`pub`.
Now it's no longer necessary and the desired behavior follows from `pub(restricted)`-style visibilities naturally assigned to enum variants and trait items.
update RLS and rustfmt
Fixes#81582 and fixes#81583
r? `@Xanewok`
I was originally surprised by the size of lockfile diff, though after looking at the RLS changes it makes a bit more sense to me now
Allow unstable features in some PGO benchmarks
Some of the benchmarks we're using for PGO require unstable features (such as compiling the standard library and some rustc-perf benchmarks), breaking CI on the beta and stable branches. For the past two releases we cherry-picked a commit directly onto the beta branch that unconditionally sets `RUSTC_BOOTSTRAP=1`, and this PR backports a similar change to the master branch.
The difference between this commit and the one we backported previously (483c1a83ca) is that this is more scoped in which benchmarks we allow unstable features, to prevent unintentionally enabling unstable features.
r? `@Mark-Simulacrum`
Update cargo
5 commits in 34170fcd6e0947808a1ac63ac85ffc0da7dace2f..ab64d1393b5b77c66b6534ef5023a1b89ee7bf64
2021-02-04 15:52:52 +0000 to 2021-02-10 00:19:10 +0000
- Allow `true` and `false` as options for `strip` option (rust-lang/cargo#9153)
- Change git dependencies to use `HEAD` by default (rust-lang/cargo#9133)
- appendix auth gcm link to new repo (which is xplat) (rust-lang/cargo#9152)
- Fix warnings of the new non_fmt_panic lint (rust-lang/cargo#9148)
- Fix panic with doc collision orphan. (rust-lang/cargo#9142)
Rename HIR UnOp variants
This renames the variants in HIR UnOp from
enum UnOp {
UnDeref,
UnNot,
UnNeg,
}
to
enum UnOp {
Deref,
Not,
Neg,
}
Motivations:
- This is more consistent with the rest of the code base where most enum
variants don't have a prefix.
- These variants are never used without the `UnOp` prefix so the extra
`Un` prefix doesn't help with readability. E.g. we don't have any
`UnDeref`s in the code, we only have `UnOp::UnDeref`.
- MIR `UnOp` type variants don't have a prefix so this is more
consistent with MIR types.
- "un" prefix reads like "inverse" or "reverse", so as a beginner in
rustc code base when I see "UnDeref" what comes to my mind is
something like `&*` instead of just `*`.
Use format string in bootstrap panic instead of a string directly
This fixes the following warning when compiling with nightly:
```
warning: panic message is not a string literal
--> src/bootstrap/builder.rs:1515:24
|
1515 | panic!(out);
| ^^^
|
= note: `#[warn(non_fmt_panic)]` on by default
= note: this is no longer accepted in Rust 2021
help: add a "{}" format string to Display the message
|
1515 | panic!("{}", out);
| ^^^^^
help: or use std::panic::panic_any instead
|
1515 | std::panic::panic_any(out);
| ^^^^^^^^^^^^^^^^^^^^^^
```
Found while working on https://github.com/rust-lang/rust/pull/79540. cc https://github.com/rust-lang/rust/pull/81645, which landed in 1.51.
Bump stabilization version for const int methods
These methods missed the beta cutoff. See #80962 for details.
`@rustbot` modify labels to +A-const-fn, +A-intrinsics
r? `@m-ou-se`
Make Vec::split_at_spare_mut public
This PR introduces a new method to the public API, under
`vec_split_at_spare` feature gate:
```rust
impl<T, A: Allocator> impl Vec<T, A> {
pub fn split_at_spare_mut(&mut self) -> (&mut [T], &mut [MaybeUninit<T>]);
}
```
The method returns 2 slices, one slice references the content of the vector,
and the other references the remaining spare capacity.
The method was previously implemented while adding `Vec::extend_from_within` in #79015,
and used to implement `Vec::spare_capacity_mut` (as the later is just a
subset of former one).
See also previous [discussion in `Vec::spare_capacity_mut` tracking issue](https://github.com/rust-lang/rust/issues/75017#issuecomment-770381335).
## Unresolved questions
- [ ] Should we consider changing the name? `split_at_spare_mut` doesn't seem like an intuitive name
- [ ] Should we deprecate `Vec::spare_capacity_mut`? Any usecase of `Vec::spare_capacity_mut` can be replaced with `Vec::split_at_spare_mut` (but not vise-versa)
r? `@KodrAus`
Add `Box::into_inner`.
This adds a `Box::into_inner` method to the `Box` type. <del>I actually suggest deprecating the compiler magic of `*b` if this gets stablized in the future.</del>
r? `@m-ou-se`
Remove usages of `expr_method_call` in derive(Ord,PartialOrd,RustcEncode,RustcDecode)
Preparing for deprecation of `expr_method_call` (#81295), by removing the remaining usages not covered by (#81294).
I am not sure about the changes to `derive(RustcEncode,RustcDecode)`
Borrowck: refactor visited map to a bitset
This PR refactors `Borrows` and the `precompute_borrows_out_of_scope` function so that this initial phase has a much reduced memory pressure. This is achieved by reducing what is stored on the heap, and also reusing heap memory as much as possible.
[experiment] remove `#[inline]` from rustc_query_system::plumbing
These functions have a ton of generic parameters and are instantiated
over and over again. Hopefully this will reduce binary bloat and speed
up bootstrapping times.
r? `@cjgillot`
Allow building rustdoc without first building rustc (MVP)
## Motivation
The compile times for rustc are extremely long and a major issue for
recruiting new contributors to rustdoc. People interested in joining
often give up after running into issues with submodules or python
versions. stage1 rustdoc fundamentally doesn't care about bootstrapping
or stages, it just needs `rustc_private` available.
## Summary of Changes
- Add an opt-in `[rust] download_rustc` option
- Determine the version of the compiler to download using `log --author=bors`
- Do no work for any component other than `Rustdoc` for any stage. Instead, copy the CI artifacts from the downloaded sysroot stage0/ to stage0-sysroot/ or stage1/ in `Sysroot`. This is done with an `ENABLE_DOWNLOAD_STAGE1` constant which is off by default.
- Don't download different versions of rustfmt or cargo - those should still use the beta version (rustfmt especially).
The vast majority of work is done in bootstrap.py, which downloads the artifacts and extracts them to stage0/ in place of the beta compiler. Rustbuild just takes care of copying the artifacts to stage1 if necessary.
## Future work
- I turned off verification for the commit tarballs because the .sha256 URLs gave a 404. This seems not ideal, it would be nice to start signing them.
- This will break if you rebase an old enough branch (I think commits are kept at most 160 days?). This doesn't need to be supported, but it would be nice to give a reasonable error. https://github.com/rust-lang/rust/pull/79540#issuecomment-751481291
- Right now, every time you rebase, stage0 tools (bootstrap, tidy, ...) will have to be recompiled. Additionally running `x.py setup tools` will compile rustbuild twice. Instead, this should download a separate beta compiler for stage0 and only use CI artifacts for stage1 onward. https://github.com/rust-lang/rust/pull/79540#issuecomment-757047321
- Add `x.py setup tools` to enable this conveniently (it doesn't make sense to use this for compiler developers). cb5d8c8522
- Compile a new version of tracing so that rustdoc still gets debug logging (since CI artifacts always disable `debug` and `trace` logging). https://github.com/rust-lang/rust/pull/79540#issuecomment-742756411, 6a5d512420
- Right now only rustdoc is ever rebuilt. This is not ideal and should probably at least compile compiler tools (rustfmt, clippy, miri). https://github.com/rust-lang/rust/pull/79540#issuecomment-775634693
- Using `git log --author=bors` sometimes breaks. This should use `git merge-base` instead. https://github.com/rust-lang/rust/pull/79540#discussion_r572572280
- It would be nice to support cross-compiling the standard library. Right now this gives an assertion failure I think.
Some of this work has already been done in (the history for) 673476c785.
Fix derived PartialOrd operators
The derived implementation of `partial_cmp` compares matching fields one
by one, stopping the computation when the result of a comparison is not
equal to `Some(Equal)`.
On the other hand the derived implementation for `lt`, `le`, `gt` and
`ge` continues the computation when the result of a field comparison is
`None`, consequently those operators are not transitive and inconsistent
with `partial_cmp`.
Fix the inconsistency by using the default implementation that fall-backs
to the `partial_cmp`. This also avoids creating very deeply nested
closures that were quite costly to compile.
Fixes#81373.
Helps with #81278, #80118.
This renames the variants in HIR UnOp from
enum UnOp {
UnDeref,
UnNot,
UnNeg,
}
to
enum UnOp {
Deref,
Not,
Neg,
}
Motivations:
- This is more consistent with the rest of the code base where most enum
variants don't have a prefix.
- These variants are never used without the `UnOp` prefix so the extra
`Un` prefix doesn't help with readability. E.g. we don't have any
`UnDeref`s in the code, we only have `UnOp::UnDeref`.
- MIR `UnOp` type variants don't have a prefix so this is more
consistent with MIR types.
- "un" prefix reads like "inverse" or "reverse", so as a beginner in
rustc code base when I see "UnDeref" what comes to my mind is
something like "&*" instead of just "*".
