Typically, this CRC16 implementation is even faster, as it uses 32 bit SSE calculations and then trucates the output to 16 bits. Also comes with AVX-512 support. https://github.com/awesomized/crc-fast-rust
