electron-rare/prima-cpp
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

fix error in llama_context_n_gpu_layers

Browse Source
This commit is contained in:
Lizonghang
2025-01-15 10:08:41 +04:00
parent 5d9aadf3d5
commit 9279a2e3ff
3 changed files with 0 additions and 7 deletions
Download Patch File Download Diff File Expand all files Collapse all files
common/common.cpp
-1
View File
@@ -1440,7 +1440,6 @@ struct llama_init_result llama_init_from_gpt_params(gpt_params & params) {
params.n_gpu_layers = n_gpu_layers[my_rank];
cparams.n_gpu_layers = n_gpu_layers[my_rank];
mparams.n_gpu_layers = n_gpu_layers[my_rank];
llama_context_n_gpu_layers(lctx)[my_rank] = n_gpu_layers[my_rank];
#ifdef LLAMA_DEBUG
device_print_props(dev_info_set, n_world, model, cparams);
include/llama.h
-2
View File
@@ -465,8 +465,6 @@ extern "C" {
LLAMA_API uint32_t * llama_context_n_layer_window(struct llama_context * ctx);
LLAMA_API uint32_t * llama_context_n_gpu_layers(struct llama_context * ctx);
// Frees all allocated memory
LLAMA_API void llama_free(struct llama_context * ctx);
src/llama.cpp
-4
View File
@@ -20519,10 +20519,6 @@ uint32_t * llama_context_n_layer_window(struct llama_context * ctx) {
return ctx->cparams.n_layer_window;
}
uint32_t * llama_context_n_gpu_layers(struct llama_context * ctx) {
return ctx->cparams.n_gpu_layers;
}
void llama_free(struct llama_context * ctx) {
delete ctx;
}