SIMD-accelerated operations
| crate | version | docs |
|---|---|---|
| base64-simd |  |
 |
| hex-simd |  |
 |
| uuid-simd |  |
 |
The crates automatically select SIMD functions when available and provide fast fallback implementations. Benchmark results are available in simd-benches.
- Performance: To be the fastest
- Productivity: Efficient SIMD abstractions
- Ergonomics: Easy to use
This project relies heavily on unsafe code. We encourage everyone to review the code and report any issues.
Memory safety bugs and unsoundness issues are classified as critical bugs. They will be fixed as soon as possible.
This project accepts English or Chinese. All code, docs, PRs and issues should be written in English or Chinese.
This project contains multiple algorithms and implementations. Some of them are not original. We list the references here.
base64:
- http://0x80.pl/articles/index.html#base64-algorithm-new
- https://gist.github.com/aqrit/a2ccea48d7cac7e9d4d99f19d4759666
hex:
unicode:
If my open-source work has been helpful to you, please sponsor me.
Every little bit helps. Thank you!