cpp-taskflow/benchmark/parallel_loop/taskflow.cpp at master · cg4115/cpp-taskflow · GitHub

History

52 lines (40 loc) · 1.29 KB

Raw

#include "matrix_multiplication.hpp"

#include <taskflow/taskflow.hpp>

// matrix_multiplication_taskflow

void matrix_multiplication_taskflow(unsigned num_threads) {

tf::Executor executor(num_threads);

tf::Taskflow taskflow;

auto pa = taskflow.parallel_for(0, N, 1, [&] (int i) {

for(int j=0; j<N; ++j) {

a[i][j] = i + j;

}

}, num_threads);

auto pb = taskflow.parallel_for(0, N, 1, [&] (int i) {

for(int j=0; j<N; ++j) {

b[i][j] = i * j;

}

}, num_threads);

auto pc = taskflow.parallel_for(0, N, 1, [&] (int i) {

for(int j=0; j<N; ++j) {

c[i][j] = 0;;

}

}, num_threads);

auto pr = taskflow.parallel_for(0, N, 1, [&] (int i) {

for(int j=0; j<N; ++j) {

for(int k=0; k<N; k++) {

c[i][j] += a[i][k] * b[k][j];

}

}, num_threads);

pa.second.precede(pr.first);

pb.second.precede(pr.first);

pc.second.precede(pr.first);

executor.run(taskflow).get();

//std::cout << reduce_sum() << std::endl;

}

std::chrono::microseconds measure_time_taskflow(unsigned num_threads) {

auto beg = std::chrono::high_resolution_clock::now();

matrix_multiplication_taskflow(num_threads);

auto end = std::chrono::high_resolution_clock::now();

return std::chrono::duration_cast<std::chrono::microseconds>(end - beg);

}

cpp-taskflow/benchmark/parallel_loop/taskflow.cpp at master · cg4115/cpp-taskflow · GitHub

Expand file tree

Breadcrumbs

taskflow.cpp

Latest commit

History

Breadcrumbs

taskflow.cpp

File metadata and controls