ExternalVendorCode/whisper.cpp
1
0
Fork 0
mirror of https://github.com/ggerganov/whisper.cpp.git synced 2024-12-24 06:46:37 +00:00
Code Issues Actions 3 Packages Projects Releases Wiki Activity

Minor arithmetic improvement to mmvq wrapper kernel (llama/7172)

Browse Source
This commit is contained in:
Ouadie EL FAROUKI 2024-05-10 01:32:15 +01:00 committed by Georgi Gerganov
parent c114b75aee
commit fe454b8d9e
1 changed files with 11 additions and 9 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

8
ggml-sycl.cpp
View File

@ -8330,13 +8330,15 @@ static void mul_mat_vec_q(const void * __restrict__ vx, const void * __restrict_
const int blocks_per_row = ncols / qk;
const int blocks_per_warp = vdr * WARP_SIZE / qi;
const int qi_vdr = (qi / vdr); // N_threads processing 1 qk block
// partial sum for each thread
float tmp = 0.0f;
const block_q_t * x = (const block_q_t *) vx;
const block_q8_1 * y = (const block_q8_1 *) vy;
for (int i = item_ct1.get_local_id(2) / (qi / vdr); i < blocks_per_row;
for (int i = item_ct1.get_local_id(2) / qi_vdr; i < blocks_per_row;
i += blocks_per_warp) {
const int ibx = row * blocks_per_row + i; // x block index
@ -8344,8 +8346,8 @@ static void mul_mat_vec_q(const void * __restrict__ vx, const void * __restrict_
const int iqs =
vdr *
(item_ct1.get_local_id(2) %
(qi / vdr)); // x block quant index when casting the quants to int
(item_ct1.get_local_id(2) -
i * qi_vdr); // x block quant index when casting the quants to int
tmp += vec_dot_q_sycl(&x[ibx], &y[iby], iqs);
}