Make crc32 100x faster on x86-64

leejet / stable-diffusion.cpp

Stable Diffusion and Flux in pure C/C++

MIT License

3.31k stars 279 forks source link

Make crc32 100x faster on x86-64 #278

Closed jart closed 4 months ago

jart commented 4 months ago

This change makes checkpoints load significantly faster by optimizing pkzip's cyclic redundancy check. This code was developed by Intel and Google and Mozilla. See Chromium's zlib codebase for further details.

leejet commented 4 months ago

Amazing work! Thank you for your contribution.