How to compile dSFMT

This document explains how to compile dSFMT for users who are using UNIX like systems (for example Linux, Free BSD, cygwin, osx, etc) on terminal. I can't help those who use IDE (Integrated Development Environment,) please see your IDE's help to use SIMD feature of your CPU.

1. First Step: Compile test programs using Makefile.

1-1. Compile standard C test program.

Check if dSFMT.c and Makefile are in your current directory. If not, cd to the directory where they exist. Then, type

make std

If it causes an error, try to type

cc -DDSFMT_MEXP=19937 -o test-std-M19937 dSFMT.c test.c

or try to type

gcc -DDSFMT_MEXP=19937 -o test-std-M19937 dSFMT.c test.c

If success, then check the test program. Type

./test-std-M19937 -v

You will see many random numbers displayed on your screen. If you want to check these random numbers are correct output, redirect output to a file and diff it with dSFMT.19937.out.txt, like this:

./test-std-M19937 -v > foo.txt
diff -w foo.txt dSFMT.19937.out.txt

Silence means they are the same because diff reports the difference of two files.

If you want to know the generation speed of dSFMT, type

./test-std-M19937 -s

It is very slow. To make it fast, compile it with -O3 option. If your compiler is gcc, you should specify -fno-strict-aliasing option with -O3. type

gcc -O3 -fno-strict-aliasing -DDSFMT_MEXP=19937 -o test-std-M19937 dSFMT.c test.c
./test-std-M19937 -s

If you are using gcc 4.0, you will get more performance of dSFMT by giving additional options --param max-inline-insns-single=1800, --param inline-unit-growth=500 and --param large-function-growth=900.

1-2. Compile SSE2 test program.

If your CPU supports SSE2 and you can use gcc version 3.4 or later, you can make test-sse2-M19937. To do this, type

make sse2

or type

gcc -O3 -msse2 -fno-strict-aliasing -DHAVE_SSE2=1 -DDSFMT_MEXP=19937 -o test-sse2-M19937 dSFMT.c test.c

If everything works well,

./test-sse2-M19937 -s

shows much shorter time than test-std-M19937 -s.

1-3. Compile AltiVec test program.

If you are using Macintosh computer with PowerPC G4 or G5, and your gcc version is later 3.3, you can make test-alti-M19937. To do this, type

make osx-alti

or type

gcc -O3 -faltivec -fno-strict-aliasing -DHAVE_ALTIVEC=1 -DDSFMT_MEXP=19937 -o test-alti-M19937 dSFMT.c test.c

If everything works well,

./test-alti-M19937 -s

shows much shorter time than test-std-M19937 -s.

1-4. Compile and check output automatically.

To make test program and check output automatically for all supported SFMT_MEXPs of dSFMT, type

make std-check

To check test program optimized for SSE2, type

make sse2-check

To check test program optimized for OSX PowerPC AltiVec, type

make osx-alti-check

These commands may take some time.

2. Second Step: Use dSFMT pseudorandom number generator with your C program.

2-1. Use sequential call and static link.

Here is a very simple program sample1.c which calculates PI using Monte-Carlo method.

#include <stdio.h>
#include <stdlib.h>
#include "dSFMT.h"

int main(int argc, char* argv[]) {
    int i, cnt, seed;
    double x, y, pi;
    const int NUM = 10000;
    dsfmt_t dsfmt;

    if (argc >= 2) {
	seed = strtol(argv[1], NULL, 10);
    } else {
	seed = 12345;
    }
    cnt = 0;
    dsfmt_init_gen_rand(&dsfmt, seed);
    for (i = 0; i < NUM; i++) {
	x = dsfmt_genrand_close_open(&dsfmt);
	y = dsfmt_genrand_close_open(&dsfmt);
	if (x * x + y * y < 1.0) {
	    cnt++;
	}
    }
    pi = (double)cnt / NUM * 4;
    printf("%f\n", pi);
    return 0;
}
      

To compile sample1.c with dSFMT.c with the period of 2607, type

gcc -DDSFMT_MEXP=521 -o sample1 dSFMT.c sample1.c

If your CPU supports SSE2 and you want to use optimized dSFMT for SSE2, type

gcc -msse2 -DDSFMT_MEXP=521 -DHAVE_SSE2 -o sample1 dSFMT.c sample1.c

If your Computer is Apple PowerPC G4 or G5 and you want to use optimized dSFMT for AltiVec, type

gcc -faltivec -DDSFMT_MEXP=521 -DHAVE_ALTIVEC -o sample1 dSFMT.c sample1.c

2-2. Use block call and static link.

Here is sample2.c which modifies sample1.c. The block call dsfmt_fill_array_close_open is much faster than sequential call, but it needs an aligned memory. The standard function to get an aligned memory is posix_memalign, but it isn't usable in every OS.

#include <stdio.h>
#define _XOPEN_SOURCE 600
#include <stdlib.h>
#include "dSFMT.h"

int main(int argc, char* argv[]) {
    int i, j, cnt, seed;
    double x, y, pi;
    const int NUM = 10000;
    const int R_SIZE = 2 * NUM;
    int size;
    double *array;
    dsfmt_t dsfmt;

    if (argc >= 2) {
	seed = strtol(argv[1], NULL, 10);
    } else {
	seed = 12345;
    }
    size = dsfmt_get_min_array_size();
    if (size < R_SIZE) {
	size = R_SIZE;
    }
#if defined(__APPLE__) || \
    (defined(__FreeBSD__) && __FreeBSD__ >= 3 && __FreeBSD__ <= 6)
    printf("malloc used\n");
    array = malloc(sizeof(double) * size);
    if (array == NULL) {
	printf("can't allocate memory.\n");
	return 1;
    }
#elif defined(_POSIX_C_SOURCE)
    printf("posix_memalign used\n");
    if (posix_memalign((void **)&array, 16, sizeof(double) * size) != 0) {
	printf("can't allocate memory.\n");
	return 1;
    }
#elif defined(__GNUC__) && (__GNUC__ > 3 || (__GNUC__ == 3 && __GNUC_MINOR__ >= 3))
    printf("memalign used\n");
    array = memalign(16, sizeof(double) * size);
    if (array == NULL) {
	printf("can't allocate memory.\n");
	return 1;
    }
#else /* in this case, gcc doesn't suppport SSE2 */
    array = malloc(sizeof(double) * size);
    if (array == NULL) {
	printf("can't allocate memory.\n");
	return 1;
    }
#endif
    cnt = 0;
    j = 0;
    dsfmt_init_gen_rand(&dsfmt, seed);
    dsfmt_fill_array_close_open(&dsfmt, array, size);
    for (i = 0; i < NUM; i++) {
	x = array[j++];
	y = array[j++];
	if (x * x + y * y < 1.0) {
	    cnt++;
	}
    }
    free(array);
    pi = (double)cnt / NUM * 4;
    printf("%f\n", pi);
    return 0;
}
      

To compile sample2.c with dSFMT.c with the period of 22281, type

gcc -DDSFMT_MEXP=2203 -o sample2 dSFMT.c sample2.c

If your CPU supports SSE2 and you want to use optimized dSFMT for SSE2, type

gcc -msse2 -DDSFMT_MEXP=2203 -DHAVE_SSE2 -o sample2 dSFMT.c sample2.c

If your computer is Apple PowerPC G4 or G5 and you want to use optimized dSFMT for AltiVec, type

gcc -faltivec -DDSFMT_MEXP=2203 -DHAVE_ALTIVEC -o sample2 dSFMT.c sample2.c

2-3. Initialize dSFMT using dsfmt_init_by_array function.

Here is sample3.c which modifies sample1.c. The 32-bit integer seed can only make 232 kinds of initial state, to avoid this problem, dSFMT provides dsfmt_init_by_array function. This sample uses dsfmt_init_by_array function which initialize the internal state array with an array of 32-bit. The size of an array can be larger than the internal state array and all elements of the array are used for initialization, but too large array is wasteful.

#include <stdio.h>
#include <string.h>
#include "dSFMT.h"

int main(int argc, char* argv[]) {
    int i, cnt, seed_cnt;
    double x, y, pi;
    const int NUM = 10000;
    uint32_t seeds[100];
    dsfmt_t dsfmt;

    if (argc >= 2) {
	seed_cnt = 0;
	for (i = 0; (i < 100) && (i < strlen(argv[1])); i++) {
	    seeds[i] = argv[1][i];
	    seed_cnt++;
	}
    } else {
	seeds[0] = 12345;
	seed_cnt = 1;
    }
    cnt = 0;
    dsfmt_init_by_array(&dsfmt, seeds, seed_cnt);
    for (i = 0; i < NUM; i++) {
	x = dsfmt_genrand_close_open(&dsfmt);
	y = dsfmt_genrand_close_open(&dsfmt);
	if (x * x + y * y < 1.0) {
	    cnt++;
	}
    }
    pi = (double)cnt / NUM * 4;
    printf("%f\n", pi);
    return 0;
}
      

To compile sample3.c, type

gcc -DDSFMT_MEXP=1279 -o sample3 dSFMT.c sample3.c

Now, seed can be a string. Like this:

./sample3 your-full-name