5.8. Combined parallel masked and taskloop Constructs

Just as the for and do constructs were combined with the parallel construct for convenience, so too, the combined parallel masked taskloop and parallel masked taskloop simd constructs have been created for convenience when using the taskloop construct.

In the following example the first taskloop construct is enclosed by the usual parallel and masked constructs to form a team of threads, and a single task generator (primary thread) for the taskloop construct.

The same OpenMP operations for the first taskloop are accomplished by the second taskloop with the parallel masked taskloop combined construct. The third taskloop uses the combined parallel masked taskloop simd construct to accomplish the same behavior as closely nested parallel masked, and taskloop simd constructs.

As with any combined construct the clauses of the components may be used with appropriate restrictions. The combination of the parallel masked construct with the taskloop or taskloop simd construct produces no additional restrictions.

//%compiler: clang
//%cflags: -fopenmp

/*
* name: parallel_masked_taskloop.1
* type: C
* version: omp_5.1
*/
#include <stdio.h>
#define N 100

int main()
{
   int i, a[N],b[N],c[N];

   for(int i=0; i<N; i++){ b[i]=i; c[i]=i; }  //init

   #pragma omp parallel
   #pragma omp masked
   #pragma omp taskloop                      // taskloop 1
   for(i=0;i<N;i++){ a[i] = b[i] + c[i]; }

   #pragma omp parallel masked taskloop      // taskloop 2
   for(i=0;i<N;i++){ b[i] = a[i] + c[i]; }

   #pragma omp parallel masked taskloop simd // taskloop 3
   for(i=0;i<N;i++){ c[i] = a[i] + b[i]; }

   printf(" %d %d\n",c[0],c[N-1]);  // 0 and 495
}

!!%compiler: gfortran
!!%cflags: -fopenmp

! name:       parallel_masked_taskloop.1
! type:       F-free
! version:    omp_5.1
program main

   integer, parameter :: N=100
   integer :: i, a(N),b(N),c(N)

   do i=1,N                            !! initialize
      b(i) = i
      c(i) = i
   enddo

   !$omp parallel
   !$omp masked
   !$omp taskloop                      !! taskloop 1
   do i=1,N
      a(i) = b(i) + c(i)
   enddo
   !$omp end taskloop
   !$omp end masked
   !$omp end parallel

   !$omp parallel masked taskloop      !! taskloop 2
   do i=1,N
      b(i) = a(i) + c(i)
   enddo
   !$omp end parallel masked taskloop

   !$omp parallel masked taskloop simd !! taskloop 3
   do i=1,N
      c(i) = a(i) + b(i)
   enddo
   !$omp end parallel masked taskloop simd

   print*,c(1),c(N)  !! 5 and 500

end program