This is my implementation of the slice-by-8 algorithm for CRC32 generation.

I've added a flag CRC_ENABLE_SLICE_BY_8, I'd appreciate if someone can give
me a hint on how to set this up in the makefile config. I get the
impression that we want this to be on by default or for the major
architectures, but still allow people to disable it easily at build time.

Tests are in the previous patch, coverage there revolves around potential
alignment issues and using longer sample data to make sure we're hitting
the codepaths for length > 8 and length not evenly divisible by 8.

Attachment: gnulib_slice_by_8.patch
Description: Binary data

Reply via email to