diff --git a/run.c b/run.c index fded591..2a9d65a 100644 --- a/run.c +++ b/run.c @@ -13,6 +13,7 @@ $ ./run #include #include #include +#include // ---------------------------------------------------------------------------- // Transformer and RunState structs, and related memory management @@ -212,6 +213,7 @@ void softmax(float* x, int size) { void matmul(float* xout, float* x, float* w, int n, int d) { // W (d,n) @ x (n,) -> xout (d,) + #pragma omp parallel for for (int i = 0; i < d; i++) { float val = 0.0f; for (int j = 0; j < n; j++) { @@ -372,6 +374,12 @@ int argmax(float* v, int n) { // ---------------------------------------------------------------------------- +long time_in_ms() { + struct timeval time; + gettimeofday(&time, NULL); + return time.tv_sec * 1000 + time.tv_usec / 1000; +} + int main(int argc, char *argv[]) { // poor man's C argparse @@ -438,7 +446,8 @@ int main(int argc, char *argv[]) { malloc_run_state(&state, &config); // the current position we are in - clock_t start = clock(); + long start = time_in_ms(); + int next; int token = 1; // 1 = BOS token in Llama-2 sentencepiece int pos = 0; @@ -469,9 +478,8 @@ int main(int argc, char *argv[]) { printf("\n"); // report our achieved tok/s - clock_t end = clock(); - double elapsed = (double)(end - start) / CLOCKS_PER_SEC; - printf("achieved tok/s: %f\n", config.seq_len / elapsed); + long end = time_in_ms(); + printf("achieved tok/s: %f\n", config.seq_len / (double)(end-start)*1000); // memory cleanup free_run_state(&state);