[RFA, C++] Symbol table 11/many: Avoid C++ FE from fiddling with TREE_SYMBOL_REFERENCED

Jan Hubicka hubicka@ucw.cz
Fri Apr 20 16:03:00 GMT 2012


Hi,
there are only two places left where frontend is fiddling with TREE_SYMBOL_REFERENCED.
The first should not be needed because alias code now properly analyze the aliases
in callgraph.
The second can be replaced by TREE_USED (and I think should be).  It is only
needed to the following decl_needed_p knows about it.

Bootstrapped/regtested x86_64-linux, OK?

Honza

	* method.c (make_alias_for): Do not set TREE_SYMBOL_REFERENCED.
	* decl2.c (mark_needed): Likewise.
	(decl_needed_p): Do not test TREE_SYMBOL_REFERENCED.
Index: method.c
===================================================================
--- method.c	(revision 186623)
+++ method.c	(working copy)
@@ -244,7 +244,6 @@ make_alias_for (tree target, tree newid)
   TREE_ADDRESSABLE (alias) = 1;
   TREE_USED (alias) = 1;
   SET_DECL_ASSEMBLER_NAME (alias, DECL_NAME (alias));
-  TREE_SYMBOL_REFERENCED (DECL_ASSEMBLER_NAME (alias)) = 1;
   return alias;
 }
 
Index: decl2.c
===================================================================
--- decl2.c	(revision 186627)
+++ decl2.c	(working copy)
@@ -1782,10 +1782,7 @@ var_finalized_p (tree var)
 void
 mark_needed (tree decl)
 {
-  /* It's possible that we no longer need to set
-     TREE_SYMBOL_REFERENCED here directly, but doing so is
-     harmless.  */
-  TREE_SYMBOL_REFERENCED (DECL_ASSEMBLER_NAME (decl)) = 1;
+  TREE_USED (decl) = 1;
   mark_decl_referenced (decl);
 }
 
@@ -1811,9 +1808,7 @@ decl_needed_p (tree decl)
     return true;
   /* If this entity was used, let the back end see it; it will decide
      whether or not to emit it into the object file.  */
-  if (TREE_USED (decl)
-      || (DECL_ASSEMBLER_NAME_SET_P (decl)
-	  && TREE_SYMBOL_REFERENCED (DECL_ASSEMBLER_NAME (decl))))
+  if (TREE_USED (decl))
       return true;
   /* Functions marked "dllexport" must be emitted so that they are
      visible to other DLLs.  */



More information about the Gcc-patches mailing list