re PR target/34435 (SSE2 intrinsics - emmintrin with optimisations off and type conversion error)

2007-12-06  Paul Thomas  <pault@gcc.gnu.org>

	PR fortran/34435
	* module.c (find_symbol): Do not return symtrees with unique
	names, which shows that they are private.

2007-12-06  Paul Thomas  <pault@gcc.gnu.org>

	PR fortran/34435
	* gfortran.dg/used_types_19.f90: New test.

From-SVN: r130642

gcc/fortran/ChangeLog

+2007-12-06  Paul Thomas  <pault@gcc.gnu.org>
+
+	PR fortran/34435
+	* module.c (find_symbol): Do not return symtrees with unique
+	names, which shows that they are private.
+
 2007-12-05  Jakub Jelinek  <jakub@redhat.com>
 
 	PR debug/33739

gcc/fortran/module.c

@@ -3153,7 +3153,8 @@ find_symbol (gfc_symtree *st, const char *name,
 
   c = strcmp (name, st->n.sym->name);
   if (c == 0 && st->n.sym->module
-	     && strcmp (module, st->n.sym->module) == 0)
+	     && strcmp (module, st->n.sym->module) == 0
+	     && !check_unique_name (st->name))
     {
       if ((!generic && !st->n.sym->attr.generic)
 	     || (generic && st->n.sym->attr.generic))

gcc/testsuite/ChangeLog

+2007-12-06  Paul Thomas  <pault@gcc.gnu.org>
+
+	PR fortran/34435
+	* gfortran.dg/used_types_19.f90: New test.
+
 2007-12-06  Kaz Kojima  <kkojima@gcc.gnu.org>
 
 	* gfortran.dg/nan_2.f90: Add -mieee for sh.

gcc/testsuite/gfortran.dg/used_types_19.f90

+! { dg-do compile }
+! Tests the fix for PR34335 a regression in which the PRIVATE attribute
+! of type(a) in module b would be ignored and would prevent it being
+! loaded in the main program.
+!
+! Contributed by Janus Weil <jaydub66@gmail.com>
+!
+module A
+  type A_type
+    real comp
+  end type
+end module A
+
+module B
+  use A
+  private
+  type(A_type) :: B_var
+  public:: B_var
+end module B
+
+program C
+  use B
+  use A
+  type(A_type):: A_var
+end program C
+! { dg-final { cleanup-modules "a b" } }