Re: [Fortran, Patch, PR70842, v1] [4.9/5/6/7 Regression] internal compiler error with character members within a polymorphic pointer

[Andre Vehreschild <vehre at gmx dot de>]

Hi Jerry,

thanks for the fast review. Committed to trunk as r238347. Will wait
one week before applying to the other branches.

Regards,
	Andre

On Thu, 14 Jul 2016 09:12:27 -0700
Jerry DeLisle <jvdelisle@charter.net> wrote:

> On 07/14/2016 07:56 AM, Andre Vehreschild wrote:
> > Hi all,
> >
> > attached patch fixes the ICE and the wrong char array length
> > mentioned in comment #3 of the PR. The issue was that getting the
> > _len component is valid online for unlimited polymorphic pointers.
> > But here this was tried for polymorphic entity, which had to fail.
> >
> > Bootstrapped and regtests ok on trunk, gcc-6-branch, gcc-5-branch,
> > gcc-4_9-branch. Ok for trunk and one week later for the other
> > branches?
> >
> > Regards,
> > 	Andre
> >  
> 
> Yes, OK and thanks for patch. (Maybe add a blank line just before the 
> existing comment shown in the patch.)
> 
> Regards,
> 
> Jerry


-- 
Andre Vehreschild * Email: vehre ad gmx dot de 

Index: gcc/fortran/ChangeLog
===================================================================
--- gcc/fortran/ChangeLog	(Revision 238344)
+++ gcc/fortran/ChangeLog	(Arbeitskopie)
@@ -1,3 +1,9 @@
+2016-07-14  Andre Vehreschild  <vehre@gcc.gnu.org>
+
+	PR fortran/70842
+	* simplify.c (gfc_simplify_len): Only for unlimited polymorphic
+	types replace the expression's _data ref with a _len ref.
+
 2016-07-09  Thomas Koenig  <tkoenig@gcc.gnu.org>
 
 	PR fortran/71783
Index: gcc/fortran/simplify.c
===================================================================
--- gcc/fortran/simplify.c	(Revision 238344)
+++ gcc/fortran/simplify.c	(Arbeitskopie)
@@ -3816,8 +3816,12 @@
     }
   else if (e->expr_type == EXPR_VARIABLE && e->ts.type == BT_CHARACTER
 	   && e->symtree->n.sym
+	   && e->symtree->n.sym->ts.type != BT_DERIVED
 	   && e->symtree->n.sym->assoc && e->symtree->n.sym->assoc->target
-	   && e->symtree->n.sym->assoc->target->ts.type == BT_DERIVED)
+	   && e->symtree->n.sym->assoc->target->ts.type == BT_DERIVED
+	   && e->symtree->n.sym->assoc->target->symtree->n.sym
+	   && UNLIMITED_POLY (e->symtree->n.sym->assoc->target->symtree->n.sym))
+
     /* The expression in assoc->target points to a ref to the _data component
        of the unlimited polymorphic entity.  To get the _len component the last
        _data ref needs to be stripped and a ref to the _len component added.  */
Index: gcc/testsuite/ChangeLog
===================================================================
--- gcc/testsuite/ChangeLog	(Revision 238344)
+++ gcc/testsuite/ChangeLog	(Arbeitskopie)
@@ -1,3 +1,8 @@
+2016-07-14  Andre Vehreschild  <vehre@gcc.gnu.org>
+
+	PR fortran/70842
+	* gfortran.dg/select_type_35.f03: New test.
+
 2016-07-14  Kyrylo Tkachov  <kyrylo.tkachov@arm.com>
 
 	PR target/65951
Index: gcc/testsuite/gfortran.dg/select_type_35.f03
===================================================================
--- gcc/testsuite/gfortran.dg/select_type_35.f03	(nicht existent)
+++ gcc/testsuite/gfortran.dg/select_type_35.f03	(Arbeitskopie)
@@ -0,0 +1,41 @@
+! { dg-do run }
+!
+! Contributed by Nathanael Huebbe
+! Check fix for PR/70842
+
+program foo
+
+  TYPE, ABSTRACT :: t_Intermediate
+  END TYPE t_Intermediate
+
+  type, extends(t_Intermediate) :: t_Foo
+    character(:), allocatable :: string
+  end type t_Foo
+
+  class(t_Foo), allocatable :: obj
+
+  allocate(obj)
+  obj%string = "blabarfoo"
+
+  call bar(obj)
+
+  deallocate(obj)
+contains
+  subroutine bar(me)
+    class(t_Intermediate), target :: me
+
+    class(*), pointer :: alias
+
+    select type(me)
+      type is(t_Foo)
+      if (len(me%string) /= 9) call abort()
+    end select
+
+    alias => me
+    select type(alias)
+      type is(t_Foo)
+        if (len(alias%string) /= 9) call abort()
+    end select
+  end subroutine bar
+end program foo
+