tqcq/llama.cpp
0
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-06-29 12:35:16 +00:00
Code Issues Packages Projects Releases Wiki Activity

GGUF: C++ refactor, backend support, misc fixes (#11030)

Browse Source 
* GGUF: C++ refactor, backend support, misc fixes

remove ggml_tensor.backend

update CODEOWNERS [no ci]

remove gguf_get_data from API

revise GGUF API data types
This commit is contained in:
Johannes Gäßler
2025-01-07 18:01:58 +01:00
committed by GitHub
parent 017cc5f446
commit 53ff6b9b9f
21 changed files with 1795 additions and 1627 deletions
Download Patch File Download Diff File Expand all files Collapse all files

27
ggml/src/ggml-impl.h
View File

@ -3,6 +3,8 @@
// GGML internal header
#include "ggml.h"
#include "gguf.h"
#include <assert.h>
#include <math.h>
#include <stdlib.h> // load `stdlib.h` before other headers to work around MinGW bug: https://sourceforge.net/p/mingw-w64/bugs/192/
@ -551,22 +553,15 @@ static inline ggml_bf16_t ggml_compute_fp32_to_bf16(float s) {
#define GGML_FP32_TO_BF16(x) ggml_compute_fp32_to_bf16(x)
#define GGML_BF16_TO_FP32(x) ggml_compute_bf16_to_fp32(x)
// expose GGUF internals for test code
GGML_API size_t gguf_type_size(enum gguf_type type);
GGML_API struct gguf_context * gguf_init_from_file_impl(FILE * file, struct gguf_init_params params);
struct gguf_buf {
void * data;
size_t size;
size_t offset;
};
GGML_API struct gguf_buf gguf_buf_init(size_t size);
GGML_API void gguf_buf_free(struct gguf_buf buf);
GGML_API void gguf_write_to_buf(const struct gguf_context * ctx, struct gguf_buf * buf, bool only_meta);
#ifdef __cplusplus
}
#endif
#ifdef __cplusplus
#include <vector>
// expose GGUF internals for test code
GGML_API size_t gguf_type_size(enum gguf_type type);
GGML_API struct gguf_context * gguf_init_from_file_impl(FILE * file, struct gguf_init_params params);
GGML_API void gguf_write_to_buf(const struct gguf_context * ctx, std::vector<int8_t> & buf, bool only_meta);
#endif // __cplusplus