https://gcc.gnu.org/bugzilla/show_bug.cgi?id=64921
--- Comment #17 from Uroš Bizjak <ubizjak at gmail dot com> ---
(In reply to Mat Cross from comment #14)
> For the record, perhaps it is of interest for me to note that we are running
> into this (cf. PR64230 comment 9) on code like
>
> Program test
> Implicit None
> Type :: t1
> Integer, Allocatable :: i
> End Type
> Type :: t2
> Integer, Allocatable :: i
> End Type
> Type, Extends (t1) :: t3
> Type (t2) :: j
> End Type
> Type, Extends (t3) :: t4
> Integer, Allocatable :: k
> End Type
> Call s
> Print *, 'ok'
> Contains
> Subroutine s
> Class (t1), Allocatable :: x
> Allocate (t4 :: x)
> End Subroutine
> End Program
>
> Since the crash is in bad compiler-generated finalization code (since 4.9),
> and given that (if I recall correctly) gfortran is using the Fortran 2008
> semantics for entities declared in a main program being implicitly saved,
> this is why removing the Deallocate (in the comment 12 example) works - the
> finalizer is never called then.
No wonder this test crashes. Tree-optimizers (-O2) on x86_64 produce:
--cut here--
test ()
{
integer(kind=8)[0:D.4089] * restrict sizes;
integer(kind=8)[0:D.4068] * restrict sizes;
void * _13;
integer(kind=4) * _63;
integer(kind=4) * _121;
<bb 2>:
_13 = __builtin_malloc (24);
if (_13 == 0B)
goto <bb 3>;
else
goto <bb 4>;
<bb 3>:
_gfortran_os_error (&"Allocation would exceed memory limit"[1]{lb: 1 sz: 1});
<bb 4>:
MEM[(c_char * {ref-all})_13] = MEM[(c_char * {ref-all})&__def_init_test_T4];
sizes_22 = __builtin_malloc (8);
MEM[(integer(kind=8)[0:D.3483] *)sizes_22][0] = 1;
_63 = MEM[(struct t4 *)_13].k;
if (_63 == 0B)
goto <bb 6>;
else
goto <bb 5>;
<bb 5>:
__builtin_free (_63);
<bb 6>:
sizes_79 = __builtin_malloc (8);
_121 ={v} MEM[(struct t3 *)0B].j.i;
__builtin_trap ();
}
--cut here--
The <bb 6>: part reads from address 0x0+, and if this doesn't crash, trap insn
surely crashes program.
