A different reg-stack.c patch

[hjl at valinux dot com (H.J. Lu)]

Hi,

Jeff's patch to stack_result should work fine. But it may cost more
for most cases since it is most likely that aggregate_value_p () will
return false. I think the patch here is safe and faster since in most
cases, no function calls are needed. Please consider it.

Thanks.

-- 
H.J. Lu (hjl@gnu.org)
---
--- ../../../import/gcc-2.95.x/egcs/gcc/reg-stack.c	Tue Oct 19 07:38:49 1999
+++ ./reg-stack.c	Tue Oct 19 08:27:58 1999
@@ -1186,20 +1186,19 @@ static rtx
 stack_result (decl)
      tree decl;
 {
-  rtx result;
+  rtx result = DECL_RTL (DECL_RESULT (decl));
 
-  /* If the value is supposed to be returned in memory, then clearly
-     it is not returned in a stack register.  */
-  if (aggregate_value_p (DECL_RESULT (decl)))
-    return 0;
-
-  result = DECL_RTL (DECL_RESULT (decl));
-  /* ?!?  What is this code supposed to do?  Can this code actually
-     trigger if we kick out aggregates above?  */
   if (result != 0
       && ! (GET_CODE (result) == REG
 	    && REGNO (result) < FIRST_PSEUDO_REGISTER))
     {
+      /* If the value is supposed to be returned in memory, then
+         clearly it is not returned in a stack register. */
+      if (aggregate_value_p (DECL_RESULT (decl)))
+	return 0;
+
+      /* ?!?  What is this code supposed to do?  Can this code actually
+         trigger if we kick out aggregates above?  */
 #ifdef FUNCTION_OUTGOING_VALUE
       result
         = FUNCTION_OUTGOING_VALUE (TREE_TYPE (DECL_RESULT (decl)), decl);