c++: New module mangling ABI

This implements a new module mangling ABI as the original one has a
few issues:

a) it was not demangleable (oops)

b) implemented a weak ownership model.

This implements a strong ownership model, so that exported entities
from named modules are mangled to include their module attachment.
This gives more informative linker diagnostics and better module
isolation.  Weak ownership was hoped to allow backwards compatibility
with non-modular code, but in practice was very brittle, and C++20
added new semantics for linkage declarations that cover the needed
functionality.

FAOD Clang is also moving to this ABI and documentation will be added
to the Itanium ABI specification.

gcc/cp/
	* cp-tree.h (mangle_identifier): Replace with ...
	(mangle_module_component): ... this.
	* mangle.cc (dump_substitution_candidates): Adjust.
	(add_substitution): Likewise.
	(find_substitution): Likewise.
	(unmangled_name_p): Likewise.
	(mangle_module_substitution): Reimplement.
	(mangle_module_component): New.
	(write_module, maybe_write_module): Adjust.
	(write_name): Drop modules here.
	(write_unqualified): Do them here instead.
	(mangle_global_init): Adjust.
	* module.cc (module_state::mangle): Adjust.
	(mangle_module): Likewise.
	(get_originating_module): Adjust.

gcc/testsuite/
	* g++.dg/modules/fn-inline-1_b.C: Adjust.
	* g++.dg/modules/fn-inline-1_c.C: Adjust.
	* g++.dg/modules/imp-inline-1_a.C: Adjust.
	* g++.dg/modules/imp-inline-1_b.C: Adjust.
	* g++.dg/modules/init-2_a.C: Adjust.
	* g++.dg/modules/init-2_b.C: Adjust.
	* g++.dg/modules/init-2_c.C: Adjust.
	* g++.dg/modules/member-def-2_d.C: Adjust.
	* g++.dg/modules/mod-sym-1.C: Adjust.
	* g++.dg/modules/mod-sym-2.C: Adjust.
	* g++.dg/modules/mod-sym-3.C: Adjust.
	* g++.dg/modules/sym-subst-1.C: Adjust.
	* g++.dg/modules/sym-subst-2_b.C: Adjust.
	* g++.dg/modules/sym-subst-3_a.C: Adjust.
	* g++.dg/modules/sym-subst-3_b.C: Adjust.
	* g++.dg/modules/sym-subst-4.C: Adjust.
	* g++.dg/modules/sym-subst-5.C: Adjust.
	* g++.dg/modules/sym-subst-6.C: Adjust.
	* g++.dg/modules/tpl-spec-1_a.C: Adjust.
	* g++.dg/modules/tpl-spec-2_b.C: Adjust.
	* g++.dg/modules/tpl-spec-2_d.C: Adjust.
	* g++.dg/modules/tpl-spec-3_a.C: Adjust.
	* g++.dg/modules/virt-1_a.C: Adjust.
	* g++.dg/modules/virt-2_a.C: Adjust.
	* g++.dg/modules/virt-2_b.C: Adjust.
	* g++.dg/modules/virt-2_c.C: Adjust.
	* g++.dg/modules/vtt-1_a.C: Adjust.
	* g++.dg/modules/vtt-1_b.C: Adjust.

3 files changed, 88 insertions, 71 deletions

diff --git a/gcc/cp/cp-tree.h b/gcc/cp/cp-tree.h
index 80994e9..cf08e16 100644
--- a/gcc/cp/cp-tree.h
+++ b/gcc/cp/cp-tree.h
@@ -2829,6 +2829,12 @@ struct GTY(()) lang_decl_base {
 
   /* The following apply to VAR, FUNCTION, TYPE, CONCEPT, & NAMESPACE
      decls.  */
+  // FIXME: Purview and Attachment are not the same thing, due to
+  // linkage-declarations.  The modules code presumes they are the
+  // same.  (For context, linkage-decl semantics was a very late
+  // change). We need a module_attachment_p flag, and this will allow
+  // some simplification of how we handle header unit entities.
+  // Hurrah!
   unsigned module_purview_p : 1;	   /* in module purview (not GMF) */
   unsigned module_import_p : 1;     	   /* from an import */
   unsigned module_entity_p : 1;		   /* is in the entitity ary &
@@ -8180,7 +8186,7 @@ extern char *get_mangled_vtable_map_var_name    (tree);
 extern bool mangle_return_type_p		(tree);
 extern tree mangle_decomp			(tree, vec<tree> &);
 extern void mangle_module_substitution		(int);
-extern void mangle_identifier			(char, tree);
+extern int mangle_module_component		(tree id, bool partition);
 extern tree mangle_module_global_init		(int);
 
 /* in dump.cc */
diff --git a/gcc/cp/mangle.cc b/gcc/cp/mangle.cc
index a20f0e0..6657ce4 100644
--- a/gcc/cp/mangle.cc
+++ b/gcc/cp/mangle.cc
@@ -102,7 +102,8 @@ along with GCC; see the file COPYING3.  If not see
 /* Things we only need one of.  This module is not reentrant.  */
 struct GTY(()) globals {
   /* An array of the current substitution candidates, in the order
-     we've seen them.  */
+     we've seen them.  Contains NULLS, which correspond to module
+     substitutions.  */
   vec<tree, va_gc> *substitutions;
 
   /* The entity that is being mangled.  */
@@ -318,20 +319,26 @@ dump_substitution_candidates (void)
 
       if (i > 0)
 	fprintf (stderr, "                    ");
-      if (DECL_P (el))
+      if (!el)
+	name = "module";
+      else if (DECL_P (el))
 	name = IDENTIFIER_POINTER (DECL_NAME (el));
       else if (TREE_CODE (el) == TREE_LIST)
 	name = IDENTIFIER_POINTER (DECL_NAME (TREE_VALUE (el)));
       else if (TYPE_NAME (el))
 	name = TYPE_NAME_STRING (el);
       fprintf (stderr, " S%d_ = ", i - 1);
-      if (TYPE_P (el) &&
-	  (CP_TYPE_RESTRICT_P (el)
-	   || CP_TYPE_VOLATILE_P (el)
-	   || CP_TYPE_CONST_P (el)))
-	fprintf (stderr, "CV-");
-      fprintf (stderr, "%s (%s at %p)\n",
-	       name, get_tree_code_name (TREE_CODE (el)), (void *) el);
+      if (el)
+	{
+	  if (TYPE_P (el) &&
+	      (CP_TYPE_RESTRICT_P (el)
+	       || CP_TYPE_VOLATILE_P (el)
+	       || CP_TYPE_CONST_P (el)))
+	    fprintf (stderr, "CV-");
+	  fprintf (stderr, "%s (%s at %p)",
+		   name, get_tree_code_name (TREE_CODE (el)), (void *) el);
+	}
+      fprintf (stderr, "\n");
     }
 }
 
@@ -443,11 +450,12 @@ add_substitution (tree node)
       tree candidate;
 
       FOR_EACH_VEC_SAFE_ELT (G.substitutions, i, candidate)
-	{
-	  gcc_assert (!(DECL_P (node) && node == candidate));
-	  gcc_assert (!(TYPE_P (node) && TYPE_P (candidate)
-		      && same_type_p (node, candidate)));
-	}
+	if (candidate)
+	  {
+	    gcc_assert (!(DECL_P (node) && node == candidate));
+	    gcc_assert (!(TYPE_P (node) && TYPE_P (candidate)
+			  && same_type_p (node, candidate)));
+	  }
     }
 
   /* Put the decl onto the varray of substitution candidates.  */
@@ -669,20 +677,21 @@ find_substitution (tree node)
     tags = get_abi_tags (type);
   /* Now check the list of available substitutions for this mangling
      operation.  */
-  if (!abbr || tags) for (i = 0; i < size; ++i)
-    {
-      tree candidate = (*G.substitutions)[i];
-      /* NODE is a matched to a candidate if it's the same decl node or
-	 if it's the same type.  */
-      if (decl == candidate
-	  || (TYPE_P (candidate) && type && TYPE_P (node)
-	      && same_type_p (type, candidate))
-	  || NESTED_TEMPLATE_MATCH (node, candidate))
+  if (!abbr || tags)
+    for (i = 0; i < size; ++i)
+      if (tree candidate = (*G.substitutions)[i])
 	{
-	  write_substitution (i);
-	  return 1;
+	  /* NODE is a matched to a candidate if it's the same decl node or
+	     if it's the same type.  */
+	  if (decl == candidate
+	      || (TYPE_P (candidate) && type && TYPE_P (node)
+		  && same_type_p (type, candidate))
+	      || NESTED_TEMPLATE_MATCH (node, candidate))
+	    {
+	      write_substitution (i);
+	      return 1;
+	    }
 	}
-    }
 
   if (!abbr)
     /* No substitution found.  */
@@ -735,6 +744,10 @@ unmangled_name_p (const tree decl)
       if (get_abi_tags (decl))
 	return false;
 
+      // Declarations attached to a named module are mangled
+      if (modules_p () && get_originating_module (decl, true) >= 0)
+	return false;
+
       /* The names of non-static global variables aren't mangled.  */
       return true;
     }
@@ -853,51 +866,49 @@ write_encoding (const tree decl)
 void
 mangle_module_substitution (int v)
 {
-  if (v < 10)
-    {
-      write_char ('_');
-      write_char ('0' + v);
-    }
-  else
-    {
-      write_char ('W');
-      write_unsigned_number (v - 10);
-      write_char ('_');
-    }
+  write_substitution (v - 1);
 }
 
-void
-mangle_identifier (char c, tree id)
+int
+mangle_module_component (tree comp, bool partition_p)
 {
-  if (c)
-    write_char (c);
-  write_source_name (id);
+  write_char ('W');
+  if (partition_p)
+    write_char ('P');
+  write_source_name (comp);
+
+  // Module substitutions use the same number-space as entity
+  // substitutions, but are orthogonal.
+  vec_safe_push (G.substitutions, NULL_TREE);
+  return G.substitutions->length ();
 }
 
 /* If the outermost non-namespace context (including DECL itself) is
    a module-linkage decl, mangle the module information.  For module
    global initializers we need to include the partition part.
 
-   <module-name> ::= W <module-id>+ E
-   <module-id> :: <unqualified-name>
-               || _ <digit>  ;; short backref
-	       || W <number> _  ;; long backref
-               || P <module-id> ;; partition introducer
+   <module-name> ::= <module-sub>
+		 || <subst>
+                 || <module-name> <module-sub>
+   <module-sub> :: W [P] <unqualified-name>
 */
 
 static void
 write_module (int m, bool include_partition)
 {
   G.mod = true;
-
-  write_char ('W');
   mangle_module (m, include_partition);
-  write_char ('E');
 }
 
 static void
 maybe_write_module (tree decl)
 {
+  if (!DECL_NAMESPACE_SCOPE_P (decl))
+    return;
+
+  if (TREE_CODE (decl) == NAMESPACE_DECL && DECL_NAME (decl))
+    return;
+
   int m = get_originating_module (decl, true);
   if (m >= 0)
     write_module (m, false);
@@ -965,9 +976,6 @@ write_name (tree decl, const int ignore_local_scope)
       decl = TYPE_NAME (TYPE_MAIN_VARIANT (TREE_TYPE (decl)));
     }
 
-  if (modules_p ())
-    maybe_write_module (decl);
-
   context = decl_mangling_context (decl);
 
   gcc_assert (context != NULL_TREE);
@@ -1356,10 +1364,10 @@ find_decomp_unqualified_name (tree decl, size_t *len)
 /* We don't need to handle thunks, vtables, or VTTs here.  Those are
    mangled through special entry points.
 
-    <unqualified-name>  ::= <operator-name>
+    <unqualified-name>  ::= [<module-name>] <operator-name>
 			::= <special-name>
-			::= <source-name>
-			::= <unnamed-type-name>
+			::= [<module-name>] <source-name>
+			::= [<module-name>] <unnamed-type-name>
 			::= <local-source-name> 
 
     <local-source-name>	::= L <source-name> <discriminator> */
@@ -1385,6 +1393,9 @@ write_unqualified_name (tree decl)
 {
   MANGLE_TRACE_TREE ("unqualified-name", decl);
 
+  if (modules_p ())
+    maybe_write_module (decl);
+
   if (identifier_p (decl))
     {
       write_unqualified_id (decl);
@@ -3996,7 +4007,6 @@ mangle_module_global_init (int module)
 
   write_string ("_ZGI");
   write_module (module, true);
-  write_char ('v');
 
   return finish_mangling_get_identifier ();
 }
diff --git a/gcc/cp/module.cc b/gcc/cp/module.cc
index 6e6b008..ddaf9c2 100644
--- a/gcc/cp/module.cc
+++ b/gcc/cp/module.cc
@@ -13893,20 +13893,18 @@ void
 module_state::mangle (bool include_partition)
 {
   if (subst)
-    mangle_module_substitution (subst - 1);
+    mangle_module_substitution (subst);
   else
     {
       if (parent)
 	parent->mangle (include_partition);
-      if (include_partition || !is_partition ())
+      if (include_partition || !is_partition ()) 
 	{
-	  char p = 0;
-	  // Partitions are significant for global initializer functions
-	  if (is_partition () && !parent->is_partition ())
-	    p = 'P';
+	  // Partitions are significant for global initializer
+	  // functions
+	  bool partition = is_partition () && !parent->is_partition ();
+	  subst = mangle_module_component (name, partition);
 	  substs.safe_push (this);
-	  subst = substs.length ();
-	  mangle_identifier (p, name);
 	}
     }
 }
@@ -13916,6 +13914,8 @@ mangle_module (int mod, bool include_partition)
 {
   module_state *imp = (*modules)[mod];
 
+  gcc_checking_assert (!imp->is_header ());
+
   if (!imp->name)
     /* Set when importing the primary module interface.  */
     imp = imp->parent;
@@ -18391,14 +18391,15 @@ get_originating_module (tree decl, bool for_mangle)
   if (!DECL_LANG_SPECIFIC (not_tmpl))
     return for_mangle ? -1 : 0;
 
-  if (for_mangle
-      && (DECL_MODULE_EXPORT_P (owner) || !DECL_MODULE_PURVIEW_P (not_tmpl)))
+  if (for_mangle && !DECL_MODULE_PURVIEW_P (not_tmpl))
     return -1;
 
-  if (!DECL_MODULE_IMPORT_P (not_tmpl))
-    return 0;
+  int mod = !DECL_MODULE_IMPORT_P (not_tmpl) ? 0 : get_importing_module (owner);
 
-  return get_importing_module (owner);
+  if (for_mangle && (*modules)[mod]->is_header ())
+    return -1;
+
+  return mod;
 }
 
 unsigned