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

rpc : make RPC servers come first in the device list (#9296)

Browse source 
* rpc : make RPC servers come first in the device list

* rpc : disable options for non-RPC builds

* rpc : rpc_count always zero for non-RPC builds
This commit is contained in:
Radoslav Gerganov 2024-09-04 11:08:32 +03:00 • committed by GitHub
parent 9379d3cc17
commit 82e3b03c11
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
3 changed files with 50 additions and 33 deletions
Download patch file Download diff file Expand all files Collapse all files

4
examples/llama-bench/llama-bench.cpp
View file

@ -299,7 +299,9 @@ static void print_usage(int /* argc */, char ** argv) {
printf(" --cpu-strict <0|1> (default: %s)\n", join(cmd_params_defaults.cpu_strict, ",").c_str());
printf(" --poll <0...100> (default: %s)\n", join(cmd_params_defaults.poll, ",").c_str());
printf(" -ngl, --n-gpu-layers <n> (default: %s)\n", join(cmd_params_defaults.n_gpu_layers, ",").c_str());
#ifdef GGML_USE_RPC
printf(" -rpc, --rpc <rpc_servers> (default: %s)\n", join(cmd_params_defaults.rpc_servers, ",").c_str());
#endif
printf(" -sm, --split-mode <none|layer|row> (default: %s)\n", join(transform_to_str(cmd_params_defaults.split_mode, split_mode_str), ",").c_str());
printf(" -mg, --main-gpu <i> (default: %s)\n", join(cmd_params_defaults.main_gpu, ",").c_str());
printf(" -nkvo, --no-kv-offload <0|1> (default: %s)\n", join(cmd_params_defaults.no_kv_offload, ",").c_str());
@ -482,12 +484,14 @@ static cmd_params parse_cmd_params(int argc, char ** argv) {
}
auto p = string_split<int>(argv[i], split_delim);
params.n_gpu_layers.insert(params.n_gpu_layers.end(), p.begin(), p.end());
#ifdef GGML_USE_RPC
} else if (arg == "-rpc" || arg == "--rpc") {
if (++i >= argc) {
invalid_param = true;
break;
}
params.rpc_servers.push_back(argv[i]);
#endif
} else if (arg == "-sm" || arg == "--split-mode") {
if (++i >= argc) {
invalid_param = true;