CS348Project/llama.cpp
5
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-10-27 08:21:30 +00:00
Code Issues Packages Projects Releases Wiki Activity

ggml : move AMX to the CPU backend (#10570)

Browse Source 
* ggml : move AMX to the CPU backend

---------

Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
This commit is contained in:
Diego Devesa
2024-11-29 21:54:58 +01:00
committed by GitHub
parent b782e5c7d4
commit 7cc2d2c889
64 changed files with 514 additions and 801 deletions
Download Patch File Download Diff File Expand all files Collapse all files

20
ggml/src/ggml-cpu/amx/amx.h Normal file
View File

@@ -0,0 +1,20 @@
#include "ggml-backend.h"
#include "ggml-cpu-impl.h"
#ifdef __cplusplus
extern "C" {
#endif
#if defined(__AMX_INT8__) && defined(__AVX512VNNI__)
ggml_backend_buffer_type_t ggml_backend_amx_buffer_type(void);
bool ggml_backend_amx_buft_is_amx(ggml_backend_buffer_type_t buft);
bool ggml_backend_amx_device_supports_op(const struct ggml_tensor * op);
void ggml_backend_amx_mul_mat(const struct ggml_compute_params * params, struct ggml_tensor * dst);
size_t ggml_backend_amx_desired_wsize(const struct ggml_tensor * dst);
#endif
#ifdef __cplusplus
}
#endif