From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 14970 invoked by alias); 17 May 2005 17:48:18 -0000 Mailing-List: contact gcc-bugs-help@gcc.gnu.org; run by ezmlm Precedence: bulk List-Archive: List-Post: List-Help: Sender: gcc-bugs-owner@gcc.gnu.org Received: (qmail 14848 invoked by uid 48); 17 May 2005 17:48:13 -0000 Date: Tue, 17 May 2005 17:48:00 -0000 Message-ID: <20050517174813.14847.qmail@sourceware.org> From: "pinskia at gcc dot gnu dot org" To: gcc-bugs@gcc.gnu.org In-Reply-To: <20050517061434.21614.yuri@tsoft.com> References: <20050517061434.21614.yuri@tsoft.com> Reply-To: gcc-bugzilla@gcc.gnu.org Subject: [Bug c++/21614] [4.0/4.1 Regression] invokation of undefined class'es method is ignored X-Bugzilla-Reason: CC X-SW-Source: 2005-05/txt/msg02357.txt.bz2 List-Id: ------- Additional Comments From pinskia at gcc dot gnu dot org 2005-05-17 17:48 ------- (In reply to comment #2) > (In reply to comment #1) > > Hmm, I think we get an error mark node but no error. > Why is this the wrong code? It's easy to execute it. No the wrong-code keyword says that GCC is emitting wrong code. But what I am saying with the error mark node is that the front-end should not be creating using error mark node in the internals unless there was an error emitted. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=21614