This is the mail archive of the
gcc-regression@gcc.gnu.org
mailing list for the GCC project.
[patch] Re: new FAILs on HEAD
Michael Ritzert <gcc@ds217-115-141-84.dedicated.hosteurope.de> writes:
> FAIL: gcc.dg/wtr-func-def-1.c (test for excess errors)
Not sure how I missed this. Anyway, obvious thinko, easy fix.
zw
* c-decl.c (store_parm_decls_newstyle): Correct test for a
nested function.
===================================================================
Index: c-decl.c
--- c-decl.c 9 May 2004 21:26:58 -0000 1.497
+++ c-decl.c 10 May 2004 19:17:36 -0000
@@ -5828,8 +5828,7 @@ store_parm_decls_newstyle (tree fndecl,
warning if we got here because ARG_INFO_TYPES was error_mark_node
(this happens when a function definition has just an ellipsis in
its parameter list). */
- else if (warn_traditional && !in_system_header
- && !current_scope->outer_function
+ else if (warn_traditional && !in_system_header && !current_function_scope
&& ARG_INFO_TYPES (arg_info) != error_mark_node)
warning ("%Jtraditional C rejects ISO C style function definitions",
fndecl);