This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

back out last tree-inline change


Richard pointed out that this code duplicates code elsewhere, and I
can't make my testcase fail without it, so I guess it's unnecessary.
I'm leaving the testcase in Just In Case.

Tested with 'make quickstrap' on powerpc-darwin, and running that one
testcase.

-- 
- Geoffrey Keating <geoffk@apple.com>

===File ~/patches/gcc-novararginline-revert.patch===========
Index: ChangeLog
===================================================================
RCS file: /cvs/gcc/gcc/gcc/ChangeLog,v
retrieving revision 1.17651
diff -u -p -u -p -r1.17651 ChangeLog
--- ChangeLog	30 Apr 2003 21:44:19 -0000	1.17651
+++ ChangeLog	1 May 2003 01:20:03 -0000
@@ -1,3 +1,8 @@
+2003-04-30  Geoffrey Keating  <geoffk@apple.com>
+
+	* tree-inline.c (inlinable_function_p): Back out last change, it's
+	unnecessary.
+
 2003-04-30  Steven Bosscher  <steven@gcc.gnu.org>
 
 	* ggc-page.c (TREE_EXP_SIZE): Define.
Index: tree-inline.c
===================================================================
RCS file: /cvs/gcc/gcc/gcc/tree-inline.c,v
retrieving revision 1.55
diff -u -p -u -p -r1.55 tree-inline.c
--- tree-inline.c	29 Apr 2003 23:32:43 -0000	1.55
+++ tree-inline.c	1 May 2003 01:20:03 -0000
@@ -1008,13 +1008,6 @@ inlinable_function_p (fn, id, nolimit)
   else if (lookup_attribute ("always_inline", DECL_ATTRIBUTES (fn)) == NULL
 	   && find_alloca_call (DECL_SAVED_TREE (fn)))
     ;
-  /* Can't inline functions which use varargs.  It's not theoretically
-     impossible, we just don't do it yet; at least one problem is that
-     expand_builtin_next_arg needs to handle the situation.  */
-  else if (TYPE_ARG_TYPES (TREE_TYPE (fn)) != 0
-	   && (TREE_VALUE (tree_last (TYPE_ARG_TYPES (TREE_TYPE (fn))))
-	       != void_type_node))
-    ;
   /* All is well.  We can inline this function.  Traditionally, GCC
      has refused to inline functions using alloca, or functions whose
      values are returned in a PARALLEL, and a few other such obscure
============================================================


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]