[Flang] Assignments for overlapping arrays of a derived type with FORALL/WHERE are slower than GFortran

[yus3710-fj]

The execution of the following codes is slower than GFortran.

• test.f90

module d
  type mytype1
    integer :: x,y,z
  end type
end module

subroutine ddaofn050(lhs,rhs)
  use d
  type(mytype1) :: rhs(:)
  type(mytype1) :: lhs(:)
  integer i
  forall(i=lbound(lhs,1):ubound(lhs,1)-3)
    lhs(i+3) = lhs(i)
  end forall
end subroutine

• main.f90

use d
implicit none

integer, parameter :: count = 1000
integer :: i,j,k
type(mytype1) :: t1(100),t2(100)
real :: start_t, end_t
interface
  subroutine ddaofn050(lhs,rhs)
    use d
    type(mytype1) :: rhs(:)
    type(mytype1) :: lhs(:)
  end subroutine
end interface

call cpu_time(start_t)
do i=1,count
  do j=1,count
    call ddaofn050(t1,t2)
  end do
end do
call cpu_time(end_t)

print *, end_t - start_t
end

• commands

$ flang -Ofast -mcpu=native test.f90 main.f90 && ./a.out # for A64FX/Grace
$ flang -Ofast -march=native test.f90 main.f90 && ./a.out # for Xeon

	Flang [s]	GFortran [s]
A64FX	60.676124	0.231041014
Grace	8.141944	0.0531440005
Xeon	11.817137	0.0444660001