vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

[SYCL] remove global variables (#7710)

Browse source 
* separate DPCT helpers outside

* replace global variables with context

* remove useless extra

* update mul_mat condition

* remove duplicate buft initialization

* remove duplicate extra and global work group size

* remove useless backend check

* remove duplicated extras

* use macro for group_size and remove cuda-related
This commit is contained in:
Meng, Hengyu 2024-06-15 14:05:10 +08:00 • committed by GitHub
parent f8ec8877b7
commit 7b2f4a7d19
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
9 changed files with 4142 additions and 4835 deletions
Download patch file Download diff file Expand all files Collapse all files

11
ggml-sycl.h
View file

@ -8,14 +8,12 @@
#include "ggml.h"
#include "ggml-backend.h"
#include "ggml-sycl/presets.hpp"
#ifdef __cplusplus
extern "C" {
#endif
#define GGML_SYCL_MAX_DEVICES 48
#define GGML_SYCL_NAME "SYCL"
// backend API
GGML_API ggml_backend_t ggml_backend_sycl_init(int device);
@ -33,13 +31,6 @@ GGML_API GGML_CALL void ggml_sycl_get_gpu_list(int *id_list, int max_len);
GGML_API GGML_CALL void ggml_sycl_get_device_description(int device, char *description, size_t description_size);
GGML_API GGML_CALL int ggml_backend_sycl_get_device_count();
GGML_API GGML_CALL void ggml_backend_sycl_get_device_memory(int device, size_t *free, size_t *total);
GGML_API GGML_CALL int ggml_backend_sycl_get_device_index(int device_id);
// TODO: these are temporary
// ref: https://github.com/ggerganov/llama.cpp/pull/6022#issuecomment-1992615670
GGML_API GGML_CALL int ggml_backend_sycl_get_device_id(int device_index);
GGML_API GGML_CALL void ggml_backend_sycl_set_single_device_mode(int main_gpu_id);
GGML_API GGML_CALL void ggml_backend_sycl_set_mul_device_mode();
// SYCL doesn't support registering host memory, keep here for reference
// GGML_API GGML_CALL bool ggml_backend_sycl_register_host_buffer(void * buffer, size_t size);