Mirrors/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggerganov/llama.cpp.git synced 2024-12-27 06:39:25 +01:00
Code Issues Packages Projects Releases Wiki Activity

cuda : do not use dmmv if the tensor does not have enough cols

Browse Source
This commit is contained in:
slaren 2024-07-10 02:21:38 +02:00
parent 713665db2e
commit f15167a4c7
1 changed files with 2 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

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

@ -1875,7 +1875,8 @@ static void ggml_cuda_mul_mat(ggml_backend_cuda_context & ctx, const ggml_tensor
bool use_dequantize_mul_mat_vec = (ggml_is_quantized(src0->type) || src0->type == GGML_TYPE_F16)
&& src1->type == GGML_TYPE_F32 && dst->type == GGML_TYPE_F32
&& src0->ne[0] % GGML_CUDA_DMMV_X == 0 && src1->ne[1] == 1;
&& src0->ne[0] % GGML_CUDA_DMMV_X == 0 && src0->ne[0] >= GGML_CUDA_DMMV_X*2
&& src1->ne[1] == 1;
bool use_mul_mat_vec_q = ggml_is_quantized(src0->type)
&& src1->type == GGML_TYPE_F32 && dst->type == GGML_TYPE_F32
&& src1->ne[1] <= MMVQ_MAX_BATCH_SIZE;