vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Update ggml-cuda.cu

Browse source 
Co-authored-by: slaren <slarengh@gmail.com>
This commit is contained in:
Johannes Gäßler 2024-02-11 19:07:03 +01:00 • committed by GitHub
parent a3a46580f7
commit 763083e567
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
1 changed files with 1 additions and 1 deletions
Download patch file Download diff file Expand all files Collapse all files

2
ggml-cuda.cu
View file

@ -6898,7 +6898,7 @@ static void mul_mat_vec_q_cuda(
const dim3 block_nums(nblocks, 1, 1); const dim3 block_nums(nblocks, 1, 1);
const dim3 block_dims(WARP_SIZE, nwarps, 1); const dim3 block_dims(WARP_SIZE, nwarps, 1);
switch(ncols_y) { switch (ncols_y) {
case 1: case 1:
mul_mat_vec_q<1, qk, qi, block_q_t, vdr, vec_dot> mul_mat_vec_q<1, qk, qi, block_q_t, vdr, vec_dot>
<<<block_nums, block_dims, 0, stream>>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst); <<<block_nums, block_dims, 0, stream>>>(vx, vy, dst, ncols_x, nrows_x, nrows_y, nrows_dst);