CS348Project/llama.cpp
5
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-11-09 10:17:06 +00:00
Code Issues Packages Projects Releases Wiki Activity

ggml-zdnn: attempt at fixing tensor views during matmul

Browse Source 
Signed-off-by: Aaron Teo <aaron.teo1@ibm.com>
This commit is contained in:
Aaron Teo
2025-07-31 16:55:53 +08:00
parent 732df731ba
commit 6b6ebb9bee
1 changed files with 2 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
ggml/src/ggml-zdnn/ggml-zdnn.cpp
View File

@@ -112,8 +112,8 @@ static void ggml_zdnn_mul_mat_op(ggml_backend_zdnn_context * ctx, const ggml_ten
const ggml_tensor * inputs = src1;
ggml_tensor * output = dst;
ggml_backend_zdnn_buffer * weights_extra = (ggml_backend_zdnn_buffer *)weights->extra;
ggml_backend_zdnn_buffer * inputs_extra = (ggml_backend_zdnn_buffer *)inputs->extra;
ggml_backend_zdnn_buffer * weights_extra = (ggml_backend_zdnn_buffer *)(weights->view_src ? weights->view_src->extra : weights->extra);
ggml_backend_zdnn_buffer * inputs_extra = (ggml_backend_zdnn_buffer *)(inputs->view_src ? inputs->view_src->extra : inputs->extra);
ggml_backend_zdnn_buffer * output_extra = (ggml_backend_zdnn_buffer *)output->extra;
zdnn_tensor_desc ptd_bias, td_bias;