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

Update llama.cpp

Browse source 
Co-authored-by: slaren <slarengh@gmail.com>
This commit is contained in:
Radoslav Gerganov 2024-06-03 14:59:49 +03:00 • committed by GitHub
parent 464c75c00e
commit 243a3e4bb2
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
1 changed files with 9 additions and 8 deletions
Download patch file Download diff file Expand all files Collapse all files

17
llama.cpp
View file

@ -16430,15 +16430,16 @@ struct llama_context * llama_new_context_with_model(
} }
#endif #endif
#if defined(GGML_USE_RPC) #if defined(GGML_USE_RPC)
for (int i = 0; i < (int)model->rpc_servers.size(); i++) { if (model->n_gpu_layers > 0) {
const char * endpoint = model->rpc_servers[i].c_str(); for (const auto & endpoint : model->rpc_servers) {
ggml_backend_t backend = ggml_backend_rpc_init(endpoint); ggml_backend_t backend = ggml_backend_rpc_init(endpoint.c_str());
if (backend == nullptr) { if (backend == nullptr) {
LLAMA_LOG_ERROR("%s: failed to initialize RPC to '%s'\n", __func__, endpoint); LLAMA_LOG_ERROR("%s: failed to initialize RPC to '%s'\n", __func__, endpoint.c_str());
llama_free(ctx); llama_free(ctx);
return nullptr; return nullptr;
}
ctx->backends.push_back(backend);
} }
ctx->backends.push_back(backend);
} }
#endif #endif
ctx->backend_cpu = ggml_backend_cpu_init(); ctx->backend_cpu = ggml_backend_cpu_init();