Files
Xenia-Canary/src
Wunkolo 8e1b286578 [x64] Add GFNI and AVX512VBMI optimizations for VECTOR_SHL(Int8)
Uses `{v}gf2p8mulb` as a general int8-multiplication instruction to do
variable Int8 bit-shifts.
Uses `{v}gf2p8affineqb` in the case that all of the shift values are the same.

Based on this original PR I had made to Xenia:
https://github.com/xenia-project/xenia/pull/2247
and this little write-up I did on this method:
https://wunkolo.github.io/post/2020/11/gf2p8affineqb-int8-shifting/
2026-04-23 22:07:39 +02:00
..