https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66089
Thomas Koenig <tkoenig at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Keywords| |ice-on-valid-code
--- Comment #32 from Thomas Koenig <tkoenig at gcc dot gnu.org> ---
The problem is a bit deeper.
First, something I forgot to mention: We now ICE on
type :: t
integer :: c
end type t
class(t), dimension(:), allocatable :: b,c
allocate (b(5), source=t(7))
allocate(c(5), source=t(13))
c = plus(c(1), b)
print *, c%c
if (any(c%c /= 20)) call abort
contains
elemental function plus(lhs, rhs)
class(t), intent(in) :: lhs, rhs
type(t) :: plus
plus%c = lhs%c + rhs%c
end function plus
end
Going up from the ICE, it can be seen that gfc_trans_assignment_1
gets the arguments
(gdb) call debug(expr1)
MAIN__:c % _data(FULL) (CLASS __class_MAIN___T_1_0a)
(gdb) call debug(expr2)
plus[[((MAIN__:c(1)) (MAIN__:b % _data(FULL)))]] (DERIVED t)
so there is no _data for c(1). I suspect this is the cause
of the trouble.
In gfc_conv_resolve_dependencies, we then see
(gdb) call debug(dest->info->expr)
MAIN__:c % _data(FULL) (CLASS __class_MAIN___T_1_0a)
(gdb) call debug(rss->info->expr)
MAIN__:c(1) (CLASS __class_MAIN___T_1_0a)
which gets handed further down, leading to the problem.
