tqcq/llama.cpp
0
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-08-19 22:36:13 -04:00
Code Issues Packages Projects Releases Wiki Activity

ggml: move op parameters from tensors to ggml_tensor::op_params (#2333)

Browse Source 
* ggml: move op parameters from tensors to ggml_tensor::op_params

* alibi: use memcpy for float params

* remove `src[1] = NULL` in ops
This commit is contained in:
slaren
2023-07-23 14:36:02 +02:00
committed by GitHub
parent e76d630df1
commit 95a6c595e7
4 changed files with 226 additions and 486 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
ggml.h
View File

@@ -199,6 +199,7 @@
#define GGML_MAX_CONTEXTS 64
#define GGML_MAX_SRC 6
#define GGML_MAX_NAME 48
#define GGML_MAX_OP_PARAMS 32
#define GGML_DEFAULT_N_THREADS 4
@@ -418,6 +419,9 @@ extern "C" {
// compute data
enum ggml_op op;
// op params - allocated as int32_t for alignment
int32_t op_params[GGML_MAX_OP_PARAMS / sizeof(uint32_t)];
bool is_param;
struct ggml_tensor * grad;