common : more accurate sampling timing (#17382)

* common : more accurate sampling timing * eval-callback : minor fixes * cont : add time_meas impl * cont : fix log msg [no ci] * cont : fix multiple definitions of time_meas * llama-cli : exclude chat template init from time measurement * cont : print percentage of unaccounted time * cont : do not reset timings
2025-11-20 13:40:10 +02:00
parent 5088b435d4
commit 196f5083ef
7 changed files with 102 additions and 33 deletions
@@ -26,7 +26,6 @@
 #include <sstream>
 #include <string>
 #include <thread>
-#include <unordered_map>
 #include <unordered_set>
 #include <vector>

@@ -60,6 +59,14 @@
 #pragma warning(disable: 4244 4267) // possible loss of data
 #endif

+common_time_meas::common_time_meas(int64_t & t_acc, bool disable) : t_start_us(disable ? -1 : ggml_time_us()), t_acc(t_acc) {}
+
+common_time_meas::~common_time_meas() {
+    if (t_start_us >= 0) {
+        t_acc += ggml_time_us() - t_start_us;
+    }
+}
+
 //
 // CPU utils
 //