You can improve the performance of the CRC32 with slice-by-16; POC here. I've tested it extensively and all of the weird quirks (e.g. incrementing 16 times per loop instead of just adding and incrementing after each iteration) are the fastest possible method.
You can improve the performance of the CRC32 with slice-by-16; POC here. I've tested it extensively and all of the weird quirks (e.g. incrementing 16 times per loop instead of just adding and incrementing after each iteration) are the fastest possible method.