[EXTERNAL] Re: [Patch, F03, RFC] FINAL support in 4.8

[Tobias Burnus]

Janus Weil wrote:
> unfortunately the current version of the patch ICEs on your test case:

And the ICE happens when creating the finalizer for "universal", which 
contains the (finalizable) component "type(ref_counter) :: counter" 
(here, "comp->ts.u.derived" is ref_counter):

       vtab = gfc_get_derived_vtab (comp->ts.u.derived);
       for (c = vtab->ts.u.derived->components; c; c = c->next)
         if (strcmp (c->name, "_final") == 0)
           break;
...
       final_wrap->symtree = c->initializer->symtree;

The problem is that "c->initializer" == NULL. For some reason, all 
initializer components of _vtab are NULL.

In that code line, one really needs the symtree as one otherwise doesn't 
have the proper interface of the wrapper. I don't see ad-hoc why there 
is no symtree. In principle, 
__final_ref_counter_implementation_Ref_counter exists in 
ref_counter_implementation.mod as does "(52 '_final' (INTEGER 4 53 0 0 
INTEGER ())" where 53 is the id of the wrapper function.

(If one moves all type declarations into the same mode, one no longer 
runs into this issue.)

I don't quickly see why the "_vtab" lacks the initializer. Janus, do 
have an idea?

  * * *

After moving all derived type decs into the ref_counter module, one gets 
past that ICE, however, one runs into another ICE as "fini->proc_tree" 
is NULL in:
           if (fini->proc_tree->n.sym->attr.elemental)

It works with the following patch, but I don't know whether it makes sense:


--- a/gcc/fortran/class.c
+++ b/gcc/fortran/class.c
@@ -1248,3 +1248,5 @@ generate_finalization_wrapper (gfc_symbol 
*derived, gfc_namespace *ns,
         {
-         if (fini->proc_tree->n.sym->attr.elemental)
+         gfc_symbol *fsym;
+         fsym = fini->proc_tree ? fini->proc_tree->n.sym : fini->proc_sym;
+         if (fsym->attr.elemental)
             {
@@ -1269,6 +1271,6 @@ generate_finalization_wrapper (gfc_symbol 
*derived, gfc_namespace *ns,
           block->ext.block.case_list->where = gfc_current_locus;
-         if (fini->proc_tree->n.sym->attr.dimension)
+         if (fsym->attr.dimension)
             block->ext.block.case_list->low
              = gfc_get_int_expr (gfc_default_integer_kind, NULL,
- fini->proc_tree->n.sym->formal->sym->as->rank);
+ fsym->formal->sym->as->rank);
           else
@@ -1284,3 +1286,3 @@ generate_finalization_wrapper (gfc_symbol 
*derived, gfc_namespace *ns,
           block->next->symtree = fini->proc_tree;
-         block->next->resolved_sym = fini->proc_tree->n.sym;
+         block->next->resolved_sym = fsym;
           block->next->ext.actual = gfc_get_actual_arglist ();

  * * *

Finally, one runs into an ICE for conv_parent_component_references. That 
happens in the finalization wrapper for the derived type "foo" where one 
accesses "ptr%component%universal%counter", where the latter is part of 
an abstract type. I haven't investigated why it fails.

Tobias