root/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggerganov/llama.cpp.git synced 2024-12-30 13:24:35 +00:00
Code Issues Actions 2 Packages Projects Releases Wiki Activity

iq1_s: AVX2 finally works

Browse Source
This commit is contained in:
Iwan Kawrakow 2024-02-12 08:29:54 +02:00
parent 67e7c4238e
commit 2ffb05acc8
1 changed files with 2 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
ggml-quants.c
View File

@ -9333,8 +9333,8 @@ void ggml_vec_dot_iq1_s_q8_K (int n, float * GGML_RESTRICT s, size_t bs, const
const __m128i ql = _mm_loadu_si128((const __m128i*)qs); qs += 16;
memcpy(&aux64, sc, 8); sc += 8;
const __m128i qh = _mm_shuffle_epi8(_mm_set_epi64x(aux64 >> 4, aux64), shuffle_h);
const __m256i hbit = _mm256_cvtepi8_epi16(_mm_and_si128(qh, m8));
v_gindex = _mm256_or_si256(_mm256_cvtepi8_epi16(ql), _mm256_slli_epi16(hbit, 5));
const __m256i hbit = _mm256_cvtepu8_epi16(_mm_and_si128(qh, m8));
v_gindex = _mm256_or_si256(_mm256_cvtepu8_epi16(ql), _mm256_slli_epi16(hbit, 5));
const __m128i scales = _mm_or_si128(_mm_slli_epi16(_mm_and_si128(qh, m7), 1), m1);
for (int i32 = 0; i32 < 4; ++i32) {