upstream-archive/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2026-08-05 17:40:46 -05:00
Code Issues Packages Projects Releases Wiki Activity

feat: cool profiler thingy

Browse Source
This commit is contained in:
Piotr Wilkin
2026-03-29 01:14:09 +01:00
committed by Piotr Wilkin
parent 0cea36222f
commit 7705d453a5
21 changed files with 2020 additions and 45 deletions
Download Patch File Download Diff File Expand all files Collapse all files

17
tools/server/server.cpp
View File

@@ -501,6 +501,23 @@ int llama_server(common_params & params, int argc, char ** argv) {
auto * ll_ctx = ctx_server.get_llama_context();
if (ll_ctx != nullptr) {
// export profiling data if profiling was enabled
if (params.profiling) {
ggml_backend_sched_t sched = llama_context_get_sched(ll_ctx);
if (sched != nullptr) {
if (params.profiling_output.empty()) {
ggml_backend_sched_print_profiling(sched);
} else {
int ret = ggml_backend_sched_export_profiling_json(sched, params.profiling_output.c_str());
if (ret == 0) {
SRV_INF("profiling data exported to: %s\n", params.profiling_output.c_str());
} else {
SRV_ERR("failed to export profiling data to: %s\n", params.profiling_output.c_str());
}
}
}
}
common_memory_breakdown_print(ll_ctx);
}
}