[Bug fortran/64921] [4.9/5/6 Regression] FAIL: gfortran.dg/class_allocate_18.f90

["rguenther at suse dot de" <gcc-bugzilla@gcc.gnu.org>]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=64921

--- Comment #21 from rguenther at suse dot de <rguenther at suse dot de> ---
On July 27, 2015 8:45:41 PM GMT+02:00, "mikael at gcc dot gnu.org"
<gcc-bugzilla@gcc.gnu.org> wrote:
>https://gcc.gnu.org/bugzilla/show_bug.cgi?id=64921
>
>Mikael Morin <mikael at gcc dot gnu.org> changed:
>
>           What    |Removed                     |Added
>----------------------------------------------------------------------------
>              CC|                            |mikael at gcc dot gnu.org
>
>--- Comment #20 from Mikael Morin <mikael at gcc dot gnu.org> ---
>(In reply to Richard Biener from comment #19)
>> So the question is whether the frontend emits the correct test
>against zero:
>> 
>>                     offset = offset * byte_stride;
>>                     D.3466 = (void *) array->data;
>>                     D.3467 = D.3466;
>>                     D.3465 = 8;
>>                     D.3469 = 8;
>>                     __builtin_memcpy ((void *) &transfer.4, (void *)
>> &D.3467, (unsigned long) MAX_EXPR <MIN_EXPR <D.3469, D.3465>, 0>);
>>                     ptr2 = (struct t4 *) (transfer.4 + offset);
>>                     if (ptr2 != 0B)
>>                       {
>>                         {
>>                           integer(kind=4) stat.5;
>> 
>>                           if (ptr2->k == 0B)
>> 
>> to me it looks like if we really want to test transfer.4
>(array->data)
>> against
>> zero.
>
>transfer.4 + offset calculates the address of an element of an array.
>I believe that if that code is executed, array.data is non-zero, and 
>of course
>array.data + offset as well.

Yes but if transfer.4 is null then transfer.4 + offset does not have to be.

Transfer.4 _is_ null in the case we segfault.  So the guard us clearly wrong.

>I think the the test should check for ptr2%j's nullness before
>deallocating
>ptr2%j, instead of testing ptr2.
>
>With a patch like this:
>
>diff --git a/gcc/fortran/class.c b/gcc/fortran/class.c
>index 218973d..1ff7437 100644
>--- a/gcc/fortran/class.c
>+++ b/gcc/fortran/class.c
>@@ -967,7 +967,7 @@ finalize_component (gfc_expr *expr, gfc_symbol
>*derived,
>gfc_component *comp,
>       cond->block->expr1->ts.kind = gfc_default_logical_kind;
> cond->block->expr1->value.function.isym = gfc_intrinsic_function_by_id
>(GFC_ISYM_ASSOCIATED);
> cond->block->expr1->value.function.actual = gfc_get_actual_arglist ();
>-      cond->block->expr1->value.function.actual->expr = gfc_copy_expr
>(expr);
>+      cond->block->expr1->value.function.actual->expr = gfc_copy_expr
>(e);
>cond->block->expr1->value.function.actual->next =
>gfc_get_actual_arglist
>();
>       cond->block->next = dealloc;
>
>
>Unfortunately, it doesn't seem to fix the problem.