MS backed implementation (https://www.nuget.org/packages/System.IO.Hashing/) is recommended over this library as it is faster and better supported.
Nuget packets:
https://www.nuget.org/packages/XXHash.Native/ (Windows-x64, Linux-x64 SSE2)
https://www.nuget.org/packages/XXHash.Managed/
Yan Collet reference C implementation: https://github.com/Cyan4973/xxHash
Alexander Melnik port: https://github.com/uranium62/xxHash
Milosz Krajewski port: https://github.com/MiloszKrajewski/K4os.Hash.xxHash
Microsoft port: https://github.com/dotnet/runtime/tree/main/src/libraries/System.IO.Hashing/src/System/IO/Hashing
There are some others implementations available on nuget, but they are much slower or produce incorrect results.
BloomFilter avx2 optimized implementation is taken from Rocksdb. It can be hard to use as you should have good knowledge of your data in advance to tune number of parameters. Expect 1-2 bytes per key as good size estimation.
In case your input is larger than MaxArraySize in dotnet (~ 2 GiB), or it doesn't represented as continuous block of memory at once you should use streaming mode.
var state = new XXH3State(seed: 10);
using var fileStream = File.OpenRead("largefile");
await state.AppendAsync(fileStream);
var hash64 = state.GetXXH3_64();Calculating current hash value is non destructive and another portion of data can be added afterwise. Just keep in mind feeding very small blocks (like 1 byte) and repeatedly calculating hash value maybe slow. Note: This isn't replacement for HashCode.Combne as far as XXH3State is rather big (>500 bytes).