ggml : AVX2 optimization for vec_dot_q4_3_q8_0 and refactoring (#1099) · andresalarcong/llama.cpp@c5aa5e5 · GitHub

Commit

ggml : AVX2 optimization for vec_dot_q4_3_q8_0 and refactoring (ggerg…

Browse files

…anov#1099)

* AVX2 optimization for vec_dot_q4_3_q8_0 and refactoring

* finish AVX vectorization of quantize_row_q8_0

* Rename hsum_int_8 to hsum_i32_8

Loading branch information

sw authored Apr 22, 2023

1 parent e9a9cb0 commit c5aa5e5