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

initial support

Browse source
This commit is contained in:
Ariadne 2024-03-01 17:04:31 +08:00
parent 8150be09d9
commit 705f1237bf
1 changed files with 1 additions and 1 deletions
Download patch file Download diff file Expand all files Collapse all files

2
ggml-sycl.cpp
View file

@ -8989,7 +8989,7 @@ static void dequantize_mul_mat_vec_q4_0_sycl(const void *vx, const dfloat *y,
float *dst, const int ncols, float *dst, const int ncols,
const int nrows, const int nrows,
dpct::queue_ptr stream) { dpct::queue_ptr stream) {
GGML_ASSERT(ncols % GGML_SYCL_DMMV_X == 0); // GGML_ASSERT(ncols % GGML_SYCL_DMMV_X == 0);
// const int block_num_y = (nrows + GGML_SYCL_MMV_Y - 1) / GGML_SYCL_MMV_Y; // const int block_num_y = (nrows + GGML_SYCL_MMV_Y - 1) / GGML_SYCL_MMV_Y;
// // the number of rows may exceed maximum grid size in the y or z dimensions, use the x dimension instead // // the number of rows may exceed maximum grid size in the y or z dimensions, use the x dimension instead
// const sycl::range<3> block_nums(1, 1, block_num_y); // const sycl::range<3> block_nums(1, 1, block_num_y);