cufKernel.cuf

program incrementTest  
     implicit none
     integer, parameter :: n = 1024*1024
     integer :: a(n), i, b
     integer, device :: a_d(n)
     integer, parameter :: tPB = 256

     a = 1
     b = 3

     a_d = a


     ! A thread block of tPB threads is used when launching the kernel. The * specified for the first execution configuration
     ! parameter leaves the compiler free to calculate the number of thread blocks to launch in order to carry out the 
     ! operation in the loop. The execution configuration could have been specified as <<<*,*>>>, in which case the 
     ! compiler would choose the thread block size as well as the number of thread blocks to launch.
     !------------------------------------------------------------------------------------------------------------------------

     !$cuf kernel do <<<*,tPB>>>
     do i = 1, n
          a_d(i) = a_d(i) + b
     enddo

     a = a_d

     if (any(a /= 4)) then
          write(*,*) '**** Program Failed ****'
     else
          write(*,*) 'Program Passed'
     endif
end program incrementTest