Overhaul of the `AllocRef` trait to match allocator-wg's latest consens; Take 2
GitHub won't let me reopen#69889 so I make a new PR.
In addition to #69889 this fixes the unsoundness of `RawVec::into_box` when using allocators supporting overallocating. Also it uses `MemoryBlock` in `AllocRef` to unify `_in_place` methods by passing `&mut MemoryBlock`. Additionally, `RawVec` now checks for `size_of::<T>()` again and ignore every ZST. The internal capacity of `RawVec` isn't used by ZSTs anymore, as `into_box` now requires a length to be specified.
r? @Amanieu
fixesrust-lang/wg-allocators#38fixesrust-lang/wg-allocators#41fixesrust-lang/wg-allocators#44fixesrust-lang/wg-allocators#51
expand vec![] to Vec::new()
The current expansion of `vec![]` calls `into_vec` on a boxed slice, which results in longer IR, and even after optimization, some unwinding artifacts are still present in the IR. This PR uses `Vec::new()` for `vec![]`.
This also allows `vec![]` to be used in const expressions.
BTreeMap/BTreeSet: implement drain_filter
Provide an implementation of drain_filter for BTreeMap and BTreeSet. Should be optimal when the predicate picks only elements in leaf nodes with at least MIN_LEN remaining elements, which is a common case, at least when draining only a fraction of the map/set, and also when the predicate picks elements stored in internal nodes where the right subtree can easily let go of a replacement element.
The first commit adds benchmarks with an external, naive implementation. to compare how much this claimed optimality-in-some-cases is actually worth.
Optimize strip_prefix and strip_suffix with str patterns
As mentioned in https://github.com/rust-lang/rust/issues/67302#issuecomment-585639226.
I'm not sure whether adding these methods to `Pattern` is desirable—but they have default implementations so the change is backwards compatible. Plus it seems like they're slated for wholesale replacement soon anyway? #56345
----
Constructing a Searcher in strip_prefix and strip_suffix is
unnecessarily slow when the pattern is a fixed-length string. Add
strip_prefix and strip_suffix methods to the Pattern trait, and add
optimized implementations of these methods in the str implementation.
The old implementation is retained as the default for these methods.
Constructing a Searcher in strip_prefix and strip_suffix is
unnecessarily slow when the pattern is a fixed-length string. Add
strip_prefix and strip_suffix methods to the Pattern trait, and add
optimized implementations of these methods in the str implementation.
The old implementation is retained as the default for these methods.
BTreeMap testing: introduce symbolic constants and use height consistently
Doesn't change what or how much is tested, except for some exact integer types, just for convenience and because `node::CAPACITY` is a usize.
r? @RalfJung
They used to be covered by `optin_builtin_traits` but negative impls
are now applicable to all traits, not just auto traits.
This also adds docs in the unstable book for the current state of auto traits.
Currently, constructing a waker requires calling the unsafe
`Waker::from_raw` API. This API requires the user to manually construct
a vtable for the waker themself - which is both cumbersome and very
error prone. This API would provide an ergonomic, straightforward and
guaranteed memory-safe way of constructing a waker.
It has been our longstanding intention that the `Waker` type essentially
function as an `Arc<dyn Wake>`, with a `Wake` trait as defined here. Two
considerations prevented the original API from being shipped as simply
an `Arc<dyn Wake>`:
- We want to support futures on embedded systems, which may not have an
allocator, and in optimized executors for which this API may not be
best-suited. Therefore, we have always explicitly supported the
maximally-flexible (but also memory-unsafe) `RawWaker` API, and
`Waker` has always lived in libcore.
- Because `Waker` lives in libcore and `Arc` lives in liballoc, it has
not been feasible to provide a constructor for `Waker` from `Arc<dyn
Wake>`.
Therefore, the Wake trait was left out of the initial version of the
task waker API.
However, as Rust 1.41, it is possible under the more flexible orphan
rules to implement `From<Arc<W>> for Waker where W: Wake` in liballoc.
Therefore, we can now define this constructor even though `Waker` lives
in libcore.
This PR adds these APIs:
- A `Wake` trait, which contains two methods
- A required method `wake`, which is called by `Waker::wake`
- A provided method `wake_by_ref`, which is called by
`Waker::wake_by_ref` and which implementors can override if they
can optimize this use case.
- An implementation of `From<Arc<W>> for Waker where W: Wake + Send +
Sync + 'static`
- A similar implementation of `From<Arc<W>> for RawWaker`.
couple more clippy fixes (let_and_return, if_same_then_else)
* summarize if-else-code with identical blocks (clippy::if_same_then_else)
* don't create variable bindings just to return the bound value immediately (clippy::let_and_return)
Amend Rc/Arc::from_raw() docs regarding unsafety
[This](https://stackoverflow.com/questions/59671647/is-it-safe-to-clone-a-type-erased-arc-via-raw-pointer) question on SO boils down to "is it safe to `::from_raw()` a `Rc<T>`/`Arc<T>` using a dummy `T` even if `T` is never dereferenced via the new `Rc`/`Arc`?". It almost never is.
This PR amends the docs of `from_raw()` regarding this point.
BTreeMap: remove shared root
This replaces the shared root with `Option`s in the BTreeMap code, and then slightly cleans up the node manipulation code taking advantage of the removal of the shared root. I expect that further simplification is possible, but wanted to get this posted for initial review.
Note that `BTreeMap::new()` continues to not allocate.
Benchmarks seem within the margin of error/unaffected, as expected for an entirely predictable branch.
```
name alloc-bench-a ns/iter alloc-bench-b ns/iter diff ns/iter diff % speedup
btree::map::iter_mut_20 20 21 1 5.00% x 0.95
btree::set::clone_100 1,360 1,439 79 5.81% x 0.95
btree::set::clone_100_and_into_iter 1,319 1,434 115 8.72% x 0.92
btree::set::clone_10k 143,515 150,991 7,476 5.21% x 0.95
btree::set::clone_10k_and_clear 142,792 152,916 10,124 7.09% x 0.93
btree::set::clone_10k_and_into_iter 146,019 154,561 8,542 5.85% x 0.94
```
