root/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggerganov/llama.cpp.git synced 2024-12-26 03:14:35 +00:00
Code Issues Actions 7 Packages Projects Releases Wiki Activity

ggml : tiny ggml_vec_dot_q4_K_q8_K AVX2 improvement (#2819)

Browse Source
This commit is contained in:
Ronny Brendel 2023-08-28 14:51:08 +02:00 committed by GitHub
parent 35feac6560
commit 3af6b86301
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 2 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
k_quants.c
View File

@ -2694,13 +2694,13 @@ void ggml_vec_dot_q4_K_q8_K(const int n, float * restrict s, const void * restri
const __m256i q8l = _mm256_loadu_si256((const __m256i*)q8); q8 += 32; const __m256i q8l = _mm256_loadu_si256((const __m256i*)q8); q8 += 32;
__m256i p16l = _mm256_maddubs_epi16(q4l, q8l); __m256i p16l = _mm256_maddubs_epi16(q4l, q8l);
p16l = _mm256_madd_epi16(scale_l, p16l); p16l = _mm256_madd_epi16(scale_l, p16l);
sumi = _mm256_add_epi32(sumi, p16l);
const __m256i q8h = _mm256_loadu_si256((const __m256i*)q8); q8 += 32; const __m256i q8h = _mm256_loadu_si256((const __m256i*)q8); q8 += 32;
__m256i p16h = _mm256_maddubs_epi16(q4h, q8h); __m256i p16h = _mm256_maddubs_epi16(q4h, q8h);
p16h = _mm256_madd_epi16(scale_h, p16h); p16h = _mm256_madd_epi16(scale_h, p16h);
sumi = _mm256_add_epi32(sumi, p16h); const __m256i sumj = _mm256_add_epi32(p16l, p16h);
sumi = _mm256_add_epi32(sumi, sumj);
} }
__m256 vd = _mm256_set1_ps(d); __m256 vd = _mm256_set1_ps(d);