This loop has less efficient memory access patterns than a nearby outer loop. To fix: Reorder the loops if possible.

Example (original code)

…
  do j=1,arrSize
    do i=1,arrSize
      do k=1,arrSize
        cMatrix(i,j) = cMatrix(i,j) + aMatrix(i,k) * bMatrix(k,j)
      end do
    end do
  end do
…


subroutine matrix_multiply(arrSize, aMatrix, bMatrix, cMatrix)
  implicit none
  real, intent(inout) :: cMatrix(:,:)
  real, intent(in)    :: aMatrix(:,:), bMatrix(:,:)
  integer, intent(in) :: arrSize
  integer :: i,j,k;

  do j=1,arrSize
    do i=1,arrSize
      do k=1,arrSize
        cMatrix(i,j) = cMatrix(i,j) + aMatrix(i,k) * bMatrix(k,j)
      end do
    end do
  end do

end subroutine matrix_multiply

Example (revised code)

…
  do j=1,arrSize
    do k=1,arrSize
      do i=1,arrSize
        cMatrix(i,j) = cMatrix(i,j) + aMatrix(i,k) * bMatrix(k,j)
      end do
    end do
  end do
…


subroutine matrix_multiply(arrSize, aMatrix, bMatrix, cMatrix)
  implicit none
  real, intent(inout) :: cMatrix(:,:)
  real, intent(in)    :: aMatrix(:,:), bMatrix(:,:)
  integer, intent(in) :: arrSize
  integer :: i,j,k;

  do j=1,arrSize
    do k=1,arrSize
      do i=1,arrSize
        cMatrix(i,j) = cMatrix(i,j) + aMatrix(i,k) * bMatrix(k,j)
      end do
    end do
  end do

end subroutine matrix_multiply

Interchanging is not always possible because of dependencies, which can lead to different results.

Reorder loops