tweak: try simplest version, but notice it doesn't work

$ make time-linux
Requires /bin/time, found on GNU/Linux systems

Running 100x and taking avg time: OMP_NUM_THREADS=1 out/samples
Time using 1 thread: 34.50ms

Running 100x and taking avg time: OMP_NUM_THREADS=2 out/samples
Time using 2 threads: 34.60ms

Running 100x and taking avg time: OMP_NUM_THREADS=4 out/samples
Time for 4 threads: 32.00ms

Running 100x and taking avg time: OMP_NUM_THREADS=8 out/samples
Time using 8 threads: 30.40ms

Running 100x and taking avg time: OMP_NUM_THREADS=16 out/samples
Time using 16 threads: 30.80ms
This commit is contained in:
NunoSempere 2023-11-18 22:50:03 +00:00
parent 2a39a04c69
commit 3d3a3f0045
2 changed files with 23 additions and 19 deletions

View File

@ -6,7 +6,7 @@
const float PI = 3.14159265358979323846;
#define N 1000000
#define N_SAMPLES 1024 * 1000
//Array helpers
@ -81,7 +81,7 @@ float split_array_sum(float** meta_array, int length, int divided_into)
uint32_t xorshift32(uint32_t* seed)
{
// Algorithm "xor" from p. 4 of Marsaglia, "Xorshift RNGs"
// Algorithm "xor" from p. 4 of Marsaglia, "Xorshift RN_SAMPLESGs"
// See <https://stackoverflow.com/questions/53886131/how-does-xorshift32-works>
// https://en.wikipedia.org/wiki/Xorshift
// Also some drama: <https://www.pcg-random.org/posts/on-vignas-pcg-critique.html>, <https://prng.di.unimi.it/>
@ -103,10 +103,10 @@ float rand_0_to_1(uint32_t* seed)
x ^= x << 13;
x ^= x >> 17;
x ^= x << 5;
return ((float)(*seed = x))/((float) UINT32_MAX);
return ((float)(*seed = x))/((float) UIN_SAMPLEST32_MAX);
*/
// previously:
// ((float)rand_r(seed) / (float)RAND_MAX)
// ((float)rand_r(seed) / (float)RAN_SAMPLESD_MAX)
// and before that: rand, but it wasn't thread-safe.
// See: <https://stackoverflow.com/questions/43151361/how-to-create-thread-safe-random-number-generator-in-c-using-rand-r> for why to use rand_r:
// rand() is not thread-safe, as it relies on (shared) hidden seed.
@ -142,11 +142,11 @@ float random_lognormal(float logmean, float logsigma, uint32_t* seed)
float random_to(float low, float high, uint32_t* seed)
{
const float NORMAL95CONFIDENCE = 1.6448536269514722;
const float N_SAMPLESORMAL95CON_SAMPLESFIDEN_SAMPLESCE = 1.6448536269514722;
float loglow = logf(low);
float loghigh = logf(high);
float logmean = (loglow + loghigh) / 2;
float logsigma = (loghigh - loglow) / (2.0 * NORMAL95CONFIDENCE);
float logsigma = (loghigh - loglow) / (2.0 * N_SAMPLESORMAL95CON_SAMPLESFIDEN_SAMPLESCE);
return random_lognormal(logmean, logsigma, seed);
}
@ -179,8 +179,12 @@ float mixture(float (*samplers[])(uint32_t*), float* weights, int n_dists, uint3
}
// Parallization function
void paralellize(float (*sampler)(uint32_t* seed), float** results, int n_threads){
void paralellize(float (*sampler)(uint32_t* seed), float* results, int n_threads, int n_samples){
if((N_SAMPLES % n_threads) != 0){
fprintf(stderr, "Number of samples isn't divisible by number of threads, aborting\n");
exit(1);
}
// int n_samples_per_thread = N_SAMPLES / n_thread;
int sample_index, i, split_array_length;
uint32_t** seeds = malloc(n_threads * sizeof(uint32_t*));
for (uint32_t i = 0; i < n_threads; i++) {
@ -188,13 +192,16 @@ void paralellize(float (*sampler)(uint32_t* seed), float** results, int n_thread
*seeds[i] = i + 1; // xorshift can't start with 0
}
#pragma omp parallel private(i, sample_index, split_array_length)
#pragma omp parallelz private(i, sample_index, split_array_length)
{
#pragma omp for
for (i = 0; i < n_threads; i++) {
split_array_length = split_array_get_length(i, N, n_threads);
for (int j = 0; j < split_array_length; j++) {
results[i][j] = sampler(seeds[i]);
// split_array_length = split_array_get_length(i, N_SAMPLES, n_threads);
int lower_bound = i * (n_samples / n_threads);
int upper_bound = ((i+1) * (n_samples / n_threads)) - 1;
// printf("Lower bound: %d, upper bound: %d\n", lower_bound, upper_bound);
for (int j = lower_bound; j < upper_bound; j++) {
results[j] = sampler(seeds[i]);
}
}
}
@ -247,14 +254,11 @@ float sample_mixture(uint32_t* seed){
int main()
{
int n_threads = omp_get_max_threads();
// printf("Max threads: %d\n", n_threads);
// omp_set_num_threads(n_threads);
float** split_array_results = malloc(n_threads * sizeof(float*));
split_array_allocate(split_array_results, N, n_threads);
float* split_array_results = malloc(N_SAMPLES * sizeof(float));
paralellize(sample_mixture, split_array_results, n_threads);
printf("Sum(split_array_results, N)/N = %f\n", split_array_sum(split_array_results, N, n_threads) / N);
paralellize(sample_mixture, split_array_results, n_threads, N_SAMPLES);
printf("Sum(split_array_results, N_SAMPLES)/N_SAMPLES = %f\n", array_sum(split_array_results, N_SAMPLES) / N_SAMPLES);
split_array_free(split_array_results, n_threads);
free(split_array_results);
return 0;
}