Adam Caldwell b8fa7936bf Enable AVX-VNNI 256-bit path for Q8_K R8 matmul (#1460) ... Use the sign trick with dpbusd instead of maddubs+madd+add, replacing 3 AVX2 instructions with 1 fused VNNI instruction. Removes dead HAVE_FANCY_SIMD code left over from the R16 split. Co-authored-by: Adam Caldwell <accaldwell@users.noreply.github.com>		2026-03-19 06:56:11 +01:00
..
cmake	Merge mainline llama.cpp (#3)	2024-07-27 07:55:01 +02:00
include	Fix clang warnings on macOS (#1354)	2026-03-03 16:27:16 +01:00
src	Enable AVX-VNNI 256-bit path for Q8_K R8 matmul (#1460)	2026-03-19 06:56:11 +01:00
.gitignore	Merge mainline llama.cpp (#3)	2024-07-27 07:55:01 +02:00
CMakeLists.txt	Remove llamafile remnants (#1179)	2026-01-22 13:20:23 +02:00