EngineX-Ascend/enginex-ascend-910-llama.cpp
10
0
Fork 0
Code Issues Pull Requests Actions 4 Projects Releases Wiki Activity

zdnn: refactor codebase + add docs (#16178)

Browse Source 
* zdnn: initial matmul refactor

Signed-off-by: Aaron Teo <aaron.teo1@ibm.com>

* ggml-zdnn: rm static from funcs

Signed-off-by: Aaron Teo <aaron.teo1@ibm.com>

* ggml-zdnn: update ggml-zdnn.h

Signed-off-by: Aaron Teo <aaron.teo1@ibm.com>

* ggml-zdnn: change header files to hpp

Signed-off-by: Aaron Teo <aaron.teo1@ibm.com>

* ggml-zdnn: switch to common.hpp

Signed-off-by: Aaron Teo <aaron.teo1@ibm.com>

* ggml-zdnn: move mulmat forward around

Signed-off-by: Aaron Teo <aaron.teo1@ibm.com>

* ggml-zdnn: rm inline from utils

Signed-off-by: Aaron Teo <aaron.teo1@ibm.com>

* ggml-zdnn: code cleanup

Signed-off-by: Aaron Teo <aaron.teo1@ibm.com>

* docs: add zDNN docs

Signed-off-by: Aaron Teo <aaron.teo1@ibm.com>

---------

Signed-off-by: Aaron Teo <aaron.teo1@ibm.com>
This commit is contained in:
Aaron Teo
2025-09-23 14:53:05 +08:00
committed by GitHub
parent 0bc7cc7154
commit 264f1b5187
11 changed files with 337 additions and 269 deletions
Download Patch File Download Diff File Expand all files Collapse all files

19
ggml/src/ggml-zdnn/utils.hpp Normal file
View File

@@ -0,0 +1,19 @@
#ifndef GGML_ZDNN_UTILITIES_HPP
#define GGML_ZDNN_UTILITIES_HPP
#include "common.hpp"
zdnn_data_types ggml_zdnn_type_mapping(ggml_type type);
void ggml_zdnn_create_tensor(zdnn_tensor_desc & pre_tfm_desc,
zdnn_tensor_desc & tfm_desc,
zdnn_ztensor & ztensor,
const ggml_tensor * src,
const int64_t * ne,
const zdnn_data_layouts layout);
void ggml_zdnn_load_tensor(zdnn_ztensor & ztensor, void * buffer);
void ggml_zdnn_init_tensor(ggml_backend_zdnn_buffer * buffer, const ggml_tensor * tensor);
#endif // GGML_ZDNN_UTILITIES_HPP