------- Comment #11 from kargl at gcc dot gnu dot org 2009-11-15 18:38 -------
Jerry, I added
@@ -56,11 +55,11 @@ get_array_index (gfc_array_ref *ar, mpz_
for (i = 0; i < ar->dimen; i++)
{
e = gfc_copy_expr (ar->start[i]);
- re = gfc_simplify_expr (e, 1);
+ gfc_simplify_expr (e, 1);
if ((gfc_is_constant_expr (ar->as->lower[i]) == 0)
|| (gfc_is_constant_expr (ar->as->upper[i]) == 0)
- || (gfc_is_constant_expr (e) == 0))
+ /*|| (gfc_is_constant_expr (e) == 0)*/)
gfc_error ("non-constant array in DATA statement %L", &ar->where);
@@ -313,6 +314,10 @@ gfc_assign_data_value (gfc_expr *lvalue,
else
mpz_set (offset, index);
+gmp_printf ("offset: %Zd\n", offset);
+if (lvalue->ts.type == BT_DERIVED)
+ mpz_add_ui (offset, offset, 1);
+
in data.c to see the computed offset values. The first chunk comments out
a check for a constant ar->start[] value, so that I get back to where
get_data_index is called. For this code
implicit none
type :: a
real :: x(3)
end type a
integer, parameter :: n = 3
type(a) :: b(n)
real, parameter :: d1(3) = (/1., 2., 3./)
real, parameter :: d2(3) = (/4., 5., 6./)
real, parameter :: d3(3) = (/7., 8., 9./)
integer :: i, z(n)
data (b(i), i = 1, n) /a(d1), a(d2), a(d3)/
data (z(i), i = 1, n) / 1, 2, 3/
print *, z(1), b(1)
print *, z(2), b(2)
print *, z(3), b(3)
end
I get
REMOVE:kargl[248] gfc4x -o z pr41807.f90
offset: 0
offset: 1
offset: 2
offset: 0
offset: -1
offset: -1
REMOVE:kargl[249] ./z
1 7.0000000 8.0000000 9.0000000
2 1.0000000 2.0000000 3.0000000
3 0.0000000 0.0000000 0.0000000
The first 3 offset values is from the data statement for the
z(3) array. The next 3 are for the array b(3) of the derived
type a. My conclusion to this point is that the array spec for
an array of a derived type is not properly set, or we're looking
at the wrong array spec.
--
kargl at gcc dot gnu dot org changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |sgk at troutmask dot apl dot
| |washington dot edu
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=41807
