CS348Project/llama.cpp
5
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-10-27 08:21:30 +00:00
Code Issues Packages Projects Releases Wiki Activity

llama: print memory breakdown on exit (#15860)

Browse Source 
* llama: print memory breakdown on exit
This commit is contained in:
Johannes Gäßler
2025-09-24 16:53:48 +02:00
committed by GitHub
parent f2a789e334
commit e789095502
18 changed files with 243 additions and 12 deletions
Download Patch File Download Diff File Expand all files Collapse all files

24
include/llama.h
View File

@@ -1329,24 +1329,25 @@ extern "C" {
//
// Performance utils
//
// NOTE: Used by llama.cpp examples, avoid using in third-party apps. Instead, do your own performance measurements.
// NOTE: Used by llama.cpp examples/tools, avoid using in third-party apps. Instead, do your own performance measurements.
//
struct llama_perf_context_data {
double t_start_ms;
double t_load_ms;
double t_p_eval_ms;
double t_eval_ms;
// ms == milliseconds
double t_start_ms; // absolute start time
double t_load_ms; // time needed for loading the model
double t_p_eval_ms; // time needed for processing the prompt
double t_eval_ms; // time needed for generating tokens
int32_t n_p_eval;
int32_t n_eval;
int32_t n_reused; // number of times a ggml compute graph had been reused
int32_t n_p_eval; // number of prompt tokens
int32_t n_eval; // number of generated tokens
int32_t n_reused; // number of times a ggml compute graph had been reused
};
struct llama_perf_sampler_data {
double t_sample_ms;
double t_sample_ms; // time needed for sampling in ms
int32_t n_sample;
int32_t n_sample; // number of sampled tokens
};
LLAMA_API struct llama_perf_context_data llama_perf_context (const struct llama_context * ctx);
@@ -1358,6 +1359,9 @@ extern "C" {
LLAMA_API void llama_perf_sampler_print(const struct llama_sampler * chain);
LLAMA_API void llama_perf_sampler_reset( struct llama_sampler * chain);
// print a breakdown of per-device memory use via LLAMA_LOG:
LLAMA_API void llama_memory_breakdown_print(const struct llama_context * ctx);
//
// training
//