# Changelog
4.3.1
* Add lock-free concurrent data structures: `Bag`, `LinkedList`, `Queue`, and `Stack`.
4.2.5
* Add `Guard::has_garbage`.
4.2.4
* Minor optimization.
4.2.2 - 4.2.3
* `Guard::accelerate` now only accelerates garbage collection of the current thread without affecting other threads.
4.2.1
* `u8` can be converted into `Epoch`.
4.2.0
* `Epoch` uses a range of `[0, 63]` `u8` values instead of rotating four values.
4.1.2
* More const functions.
4.1.1
* Let `miri` not execute Intel-specific code paths.
4.1.0
* The size of `Option<Guard>` is now that of `Guard`.
4.0.1
* Minor improvements to documentation.
4.0.0
* Bump MSRV to 1.85.0 / Edition 2024.
3.0.10
* Minor epoch update policy optimization.
* Minor `NonNull` optimization on `Owned` and `Shared`.
3.0.9
* Fix unsound `Sync` implementations of `AtomicShared` and `Shared`; previously, the `Sync` implementation allowed an arbitrary thread to own/drop the contained instance.
3.0.8
* Minor `const` optimization.
3.0.7
* Fix a use-after-free issue when thread-local storage is dropped.
3.0.5
* Fix minor linting errors.
3.0.4
* Adjust tests to be more Miri friendly.
3.0.3
* Fix a rare memory ordering issue when dropping thread-local storage.
3.0.2
* Make `SDD` much more friendly to Miri.
3.0.1
* Compatible with the [`Miri`](https://github.com/rust-lang/miri) memory leak checker.
* Make `Collectible` private since it is unsafe.
* Remove `Guard::defer` which depends on `Collectible`.
* Remove `prepare`.
2.1.0
* Minor performance optimization.
* Remove `Owned::release`.
2.0.0
* `{Owned, Shared}::release` no longer receives a `Guard`.
* `Link` is now public.
1.7.0
* Add `loom` support.
1.6.0
* Add `Guard::accelerate`.
1.5.0
* Fix `Guard::epoch` to return the correct epoch value.
1.4.0
* `Epoch` is now a 4-state type (3 -> 4).
1.3.0
* Add `Epoch`
* Add `Guard::epoch`.
1.2.0
* Remove `Collectible::drop_and_dealloc`.
1.1.0
* Add `prepare`.
1.0.1
* Relax trait bounds of `Guard::defer_execute`.
1.0.0
* Minor code cleanup.
0.2.0
* Make `Guard` `UnwindSafe`.
0.1.0
* Minor optimization.
0.0.1
* Initial commit: code copied from [`scalable-concurrent-containers`](https://github.com/wvwwvwwv/scalable-concurrent-containers).