/* -*- c++ -*- */ /* * Copyright 2002,2004,2013 Free Software Foundation, Inc. * * This file is part of GNU Radio * * GNU Radio is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License as published by * the Free Software Foundation; either version 3, or (at your option) * any later version. * * GNU Radio is distributed in the hope that it will be useful, * but WITHOUT ANY WARRANTY; without even the implied warranty of * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the * GNU General Public License for more details. * * You should have received a copy of the GNU General Public License * along with GNU Radio; see the file COPYING. If not, write to * the Free Software Foundation, Inc., 51 Franklin Street, * Boston, MA 02110-1301, USA. */ #ifdef HAVE_CONFIG_H #include "config.h" #endif #include <stdio.h> #include <stdlib.h> #include <sys/time.h> #ifdef HAVE_SYS_RESOURCE_H #include <sys/resource.h> #endif #include <unistd.h> #include <blocks/nco.h> #include <blocks/fxpt_nco.h> #include <string.h> #define ITERATIONS 20000000 #define BLOCK_SIZE (10 * 1000) // fits in cache #define FREQ 5003.123 static double timeval_to_double(const struct timeval *tv) { return (double)tv->tv_sec + (double)tv->tv_usec * 1e-6; } static void benchmark(void test (float *x, float *y), const char *implementation_name) { #ifdef HAVE_SYS_RESOURCE_H struct rusage rusage_start; struct rusage rusage_stop; #else double clock_start; double clock_end; #endif float output[2*BLOCK_SIZE]; float *x = &output[0], *y = &output[BLOCK_SIZE]; // touch memory memset(output, 0, 2*BLOCK_SIZE*sizeof(float)); // get starting CPU usage #ifdef HAVE_SYS_RESOURCE_H if(getrusage(RUSAGE_SELF, &rusage_start) < 0) { perror("getrusage"); exit(1); } #else clock_start = (double)clock() * (1000000. / CLOCKS_PER_SEC); #endif // do the actual work test(x, y); // get ending CPU usage #ifdef HAVE_SYS_RESOURCE_H if(getrusage(RUSAGE_SELF, &rusage_stop) < 0) { perror("getrusage"); exit(1); } // compute results double user = timeval_to_double(&rusage_stop.ru_utime) - timeval_to_double(&rusage_start.ru_utime); double sys = timeval_to_double(&rusage_stop.ru_stime) - timeval_to_double(&rusage_start.ru_stime); double total = user + sys; #else clock_end = (double)clock() * (1000000. / CLOCKS_PER_SEC); double total = clock_end - clock_start; #endif printf("%18s: cpu: %6.3f steps/sec: %10.3e\n", implementation_name, total, ITERATIONS / total); } // ---------------------------------------------------------------- // Don't compare the _vec with other functions since memory store's // are involved. void basic_sincos_vec(float *x, float *y) { gr::blocks::nco<float,float> nco; nco.set_freq(2 * M_PI / FREQ); for(int i = 0; i < ITERATIONS/BLOCK_SIZE; i++) { for(int j = 0; j < BLOCK_SIZE; j++) { nco.sincos(&x[2*j+1], &x[2*j]); nco.step(); } } } void native_sincos_vec(float *x, float *y) { gr::blocks::nco<float,float> nco; nco.set_freq(2 * M_PI / FREQ); for(int i = 0; i < ITERATIONS/BLOCK_SIZE; i++) { nco.sincos((gr_complex*)x, BLOCK_SIZE); } } void fxpt_sincos_vec(float *x, float *y) { gr::blocks::fxpt_nco nco; nco.set_freq (2 * M_PI / FREQ); for(int i = 0; i < ITERATIONS/BLOCK_SIZE; i++) { nco.sincos((gr_complex*)x, BLOCK_SIZE); } } // ---------------------------------------------------------------- void native_sincos(float *x, float *y) { gr::blocks::nco<float,float> nco; nco.set_freq(2 * M_PI / FREQ); for(int i = 0; i < ITERATIONS; i++) { nco.sincos(x, y); nco.step(); } } void fxpt_sincos(float *x, float *y) { gr::blocks::fxpt_nco nco; nco.set_freq(2 * M_PI / FREQ); for(int i = 0; i < ITERATIONS; i++) { nco.sincos(x, y); nco.step(); } } // ---------------------------------------------------------------- void native_sin(float *x, float *y) { gr::blocks::nco<float,float> nco; nco.set_freq(2 * M_PI / FREQ); for(int i = 0; i < ITERATIONS; i++) { *x = nco.sin(); nco.step(); } } void fxpt_sin(float *x, float *y) { gr::blocks::fxpt_nco nco; nco.set_freq(2 * M_PI / FREQ); for(int i = 0; i < ITERATIONS; i++) { *x = nco.sin(); nco.step(); } } // ---------------------------------------------------------------- void nop_fct(float *x, float *y) { } void nop_loop(float *x, float *y) { for(int i = 0; i < ITERATIONS; i++) { nop_fct(x, y); } } int main(int argc, char **argv) { benchmark(nop_loop, "nop loop"); benchmark(native_sin, "native sine"); benchmark(fxpt_sin, "fxpt sine"); benchmark(native_sincos, "native sin/cos"); benchmark(fxpt_sincos, "fxpt sin/cos"); benchmark(basic_sincos_vec, "basic sin/cos vec"); benchmark(native_sincos_vec, "native sin/cos vec"); benchmark(fxpt_sincos_vec, "fxpt sin/cos vec"); }