multithread malloc

test2.cpp

//------------------------------------------------------------------------
#include <iostream>
#include <vector>
#include <omp.h>
#include <chrono>
//------------------------------------------------------------------------
const int N_THREADS = 16;
//------------------------------------------------------------------------
int
func1(int n, int i){
  int sum = 0;
  for(int k=0;k<2;k++){
    std::vector <int> v;
    for(int j=0;j<n;j++){
      v.push_back(j*i);
    }
    sum = sum + v[n/2];
  }
  return sum;
}
//------------------------------------------------------------------------
int
main(void){
  auto t1 = std::chrono::system_clock::now();
  #pragma omp parallel for schedule(static)
  for(int i=0;i<N_THREADS;i++){
    func1(10000000,i);
  }
  #pragma omp barrier
  auto t2 = std::chrono::system_clock::now();
  auto elapsed = std::chrono::duration_cast<std::chrono::milliseconds>(t2-t1).count();
  std::cout << "Parallel " << elapsed << std::endl;
  t1 = std::chrono::system_clock::now();
  func1(10000000,0);
  t2 = std::chrono::system_clock::now();
  elapsed = std::chrono::duration_cast<std::chrono::milliseconds>(t2-t1).count();
  std::cout << "Serial " << elapsed << std::endl;
}
//------------------------------------------------------------------------

all: a.out b.out

a.out: test2.cpp
        icpc -O3 -qopenmp -std=c++11 $< -o $@

b.out: test2.cpp
        g++ -O3 -fopenmp -std=c++11 $< -o $@

run:
        export OMP_NUM_THREADS=1;./a.out
        export OMP_NUM_THREADS=2;./a.out
        export OMP_NUM_THREADS=8;./a.out
        export OMP_NUM_THREADS=16;./a.out
        export OMP_NUM_THREADS=24;./a.out
        export OMP_NUM_THREADS=1;./b.out
        export OMP_NUM_THREADS=2;./b.out
        export OMP_NUM_THREADS=8;./b.out
        export OMP_NUM_THREADS=16;./b.out
        export OMP_NUM_THREADS=24;./b.out

clean:
        rm -f a.out b.out