Skip to content

ggml : fix IQ3_S AVX implementation #5834

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
Mar 2, 2024
Merged

ggml : fix IQ3_S AVX implementation #5834

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion ggml-quants.c
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -10231,7 +10231,7 @@ void ggml_vec_dot_iq3_s_q8_K (int n, float * GGML_RESTRICT s, size_t bs, const v
//const __m256i q2_2 = _mm256_i32gather_epi32((const int *)iq3s_grid, idx.vec[1], 4);
const __m256i q2_1 = _mm256_set_epi32(
iq3s_grid[idx.index[7]], iq3s_grid[idx.index[6]], iq3s_grid[idx.index[5]], iq3s_grid[idx.index[4]],
iq3s_grid[idx.index[3]], iq3s_grid[idx.index[1]], iq3s_grid[idx.index[1]], iq3s_grid[idx.index[0]]
iq3s_grid[idx.index[3]], iq3s_grid[idx.index[2]], iq3s_grid[idx.index[1]], iq3s_grid[idx.index[0]]
);
const __m256i q2_2 = _mm256_set_epi32(
iq3s_grid[idx.index[15]], iq3s_grid[idx.index[14]], iq3s_grid[idx.index[13]], iq3s_grid[idx.index[12]],
Expand Down