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

rebase

Browse source
This commit is contained in:
Neo Zhang 2024-07-02 14:04:40 +08:00
parent 776275cf9c
commit 8b6cb2c7f2
2 changed files with 3 additions and 4 deletions
Download patch file Download diff file Expand all files Collapse all files

6
ggml/src/ggml-sycl/common.hpp
View file

@ -217,11 +217,11 @@ struct ggml_sycl_device_info {
bool sycl_visible_devices_existed = false;
struct sycl_device_info {
int cc; // compute capability
int cc; // compute capability
// int nsm; // number of streaming multiprocessors
// size_t smpb; // max. shared memory per block
bool vmm; // virtual memory support
size_t total_vram;
bool vmm; // virtual memory support
size_t total_vram;
};
sycl_device_info devices[GGML_SYCL_MAX_DEVICES] = {};

1
src/llama.cpp
View file

@ -17620,7 +17620,6 @@ struct llama_context * llama_new_context_with_model(
#elif defined(GGML_USE_SYCL)
// with split_mode LLAMA_SPLIT_MODE_NONE or LLAMA_SPLIT_MODE_ROW, only the main GPU backend is used
if (model->split_mode == LLAMA_SPLIT_MODE_NONE || model->split_mode == LLAMA_SPLIT_MODE_ROW) {
ggml_sycl_set_single_device(model->main_gpu);
ggml_backend_t backend = ggml_backend_sycl_init(model->main_gpu);
if (backend == nullptr) {
LLAMA_LOG_ERROR("%s: failed to initialize SYCL%d backend\n", __func__, model->main_gpu);