The fast async mutex which uses atomics with spinlock algorithm.
Spinlock algorithm integrated with the Rust futures concept without the overhead of any libs. Also when the MutexGuard is dropped,
a waker of the next locker will be executed.
It will be works with any async runtime in Rust, it may be a tokio, smol, async-std and etc..
```rust use fastasyncmutex::mutex::Mutex;
async fn main() { let mutex = Mutex::new(10); let guard = mutex.lock().await; assert_eq!(*guard, 10); } ```
There is result of benchmarks which runs on MacBook Pro (16-inch, 2019) 2,3 GHz 8-Core Intel Core i9 16GB RAM
Tests you can find in the benchmarks dir.
```
running 30 tests
test mutex::fastasyncmutex::tests::concurrencywithoutwaiting ... bench: 1,582,255 ns/iter (+/- 226,548)
test mutex::fastasyncmutex::tests::create ... bench: 0 ns/iter (+/- 0)
test mutex::fastasyncmutex::tests::stepbystepwithoutwaiting ... bench: 165,499 ns/iter (+/- 58,886)
test mutex::fastasyncmutexunordered::tests::concurrencywithoutwaiting ... bench: 1,559,522 ns/iter (+/- 213,015)
test mutex::fastasyncmutexunordered::tests::create ... bench: 0 ns/iter (+/- 0)
test mutex::fastasyncmutexunordered::tests::stepbystepwithoutwaiting ... bench: 92,905 ns/iter (+/- 13,518)
test mutex::futures::tests::concurrencywithoutwaiting ... bench: 1,636,030 ns/iter (+/- 109,052)
test mutex::futures::tests::create ... bench: 88 ns/iter (+/- 15)
test mutex::futures::tests::stepbystepwithoutwaiting ... bench: 210,283 ns/iter (+/- 15,931)
test mutex::smol::tests::concurrencywithoutwaiting ... bench: 1,870,199 ns/iter (+/- 142,146)
test mutex::smol::tests::create ... bench: 0 ns/iter (+/- 0)
test mutex::smol::tests::stepbystepwithoutwaiting ... bench: 343,573 ns/iter (+/- 32,607)
test mutex::tokio::tests::concurrencywithoutwaiting ... bench: 22,686,028 ns/iter (+/- 3,648,661)
test mutex::tokio::tests::create ... bench: 89 ns/iter (+/- 10)
test mutex::tokio::tests::stepbystepwithoutwaiting ... bench: 707,779 ns/iter (+/- 114,661)
test rwlock::fastasyncmutex::tests::concurrencyread ... bench: 1,698,161 ns/iter (+/- 153,215)
test rwlock::fastasyncmutex::tests::concurrencywrite ... bench: 1,584,764 ns/iter (+/- 115,173)
test rwlock::fastasyncmutex::tests::create ... bench: 0 ns/iter (+/- 0)
test rwlock::fastasyncmutex::tests::stepbystepread ... bench: 247,954 ns/iter (+/- 33,076)
test rwlock::fastasyncmutex::tests::stepbystepwriting ... bench: 153,154 ns/iter (+/- 27,294)
test rwlock::smol::tests::concurrencyread ... bench: 1,667,601 ns/iter (+/- 188,827)
test rwlock::smol::tests::concurrencywrite ... bench: 2,139,812 ns/iter (+/- 208,289)
test rwlock::smol::tests::create ... bench: 1 ns/iter (+/- 0)
test rwlock::smol::tests::stepbystepread ... bench: 196,376 ns/iter (+/- 26,671)
test rwlock::smol::tests::stepbystepwriting ... bench: 623,069 ns/iter (+/- 121,692)
test rwlock::tokio::tests::concurrencyread ... bench: 24,207,303 ns/iter (+/- 3,060,491)
test rwlock::tokio::tests::concurrencywrite ... bench: 23,943,930 ns/iter (+/- 1,776,313)
test rwlock::tokio::tests::create ... bench: 86 ns/iter (+/- 19)
test rwlock::tokio::tests::stepbystepread ... bench: 815,935 ns/iter (+/- 58,694)
test rwlock::tokio::tests::stepbystep_writing ... bench: 841,960 ns/iter (+/- 180,398)
test result: ok. 0 passed; 0 failed; 0 ignored; 30 measured; 0 filtered out ```
Licensed under either of
at your option.
Unless you explicitly state otherwise, any contribution intentionally submitted for inclusion in the work by you, as defined in the Apache-2.0 license, shall be dual licensed as above, without any additional terms or conditions.