CS348Project/llama.cpp
5
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-11-21 12:16:57 +00:00
Code Issues Packages Projects Releases Wiki Activity

sync : ggml

Browse Source
This commit is contained in:
Georgi Gerganov
2024-08-27 22:01:45 +03:00
parent 3246fe84d7
commit 231cff5f6f
21 changed files with 1422 additions and 178 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
ggml/src/ggml-cuda/sumrows.cu
View File

@@ -16,7 +16,7 @@ static __global__ void k_sum_rows_f32(const float * x, float * dst, const int nc
}
}
static void sum_rows_f32_cuda(const float * x, float * dst, const int ncols, const int nrows, cudaStream_t stream) {
void sum_rows_f32_cuda(const float * x, float * dst, const int ncols, const int nrows, cudaStream_t stream) {
const dim3 block_dims(WARP_SIZE, 1, 1);
const dim3 block_nums(nrows, 1, 1);
k_sum_rows_f32<<<block_nums, block_dims, 0, stream>>>(x, dst, ncols);
@@ -32,7 +32,6 @@ void ggml_cuda_op_sum_rows(ggml_backend_cuda_context & ctx, ggml_tensor * dst) {
GGML_ASSERT( dst->type == GGML_TYPE_F32);
GGML_ASSERT(ggml_is_contiguous(src0));
const int64_t ncols = src0->ne[0];
const int64_t nrows = ggml_nrows(src0);