tqcq/llama.cpp
0
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-07-29 05:33:37 -04:00
Code Issues Packages Projects Releases Wiki Activity

CUDA: fix MMQ for non-contiguous src0, add tests (#10021)

Browse Source 
* CUDA: fix MMQ for non-contiguous src0, add tests

* revise test code
This commit is contained in:
Johannes Gäßler
2024-10-24 11:09:36 +02:00
committed by GitHub
parent 0a1c750c80
commit c39665f589
4 changed files with 73 additions and 29 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
ggml/src/ggml-cuda/mmq.cu
View File

@@ -8,8 +8,6 @@ void ggml_cuda_op_mul_mat_q(
const int64_t ne00 = src0->ne[0];
const int64_t nb01 = src0->nb[1];
const int64_t ne10 = src1->ne[0];
const int64_t ne11 = src1->ne[1];
GGML_ASSERT(ne10 % QK8_1 == 0);
@@ -17,7 +15,7 @@ void ggml_cuda_op_mul_mat_q(
const int64_t ne0 = dst->ne[0];
const int64_t row_diff = row_high - row_low;
const int64_t stride00 = nb01 / ggml_type_size(src0->type);
const int64_t stride00 = ne00 / ggml_blck_size(src0->type);
int id = ggml_cuda_get_device();
const int compute_capability = ggml_cuda_info().devices[id].cc;