From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 3323 invoked by alias); 14 Oct 2004 03:47:20 -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 3314 invoked by uid 48); 14 Oct 2004 03:47:19 -0000 Date: Thu, 14 Oct 2004 03:47:00 -0000 Message-ID: <20041014034719.3313.qmail@sourceware.org> From: "pinskia at gcc dot gnu dot org" To: gcc-bugs@gcc.gnu.org In-Reply-To: <20041009192304.17913.jsm28@gcc.gnu.org> References: <20041009192304.17913.jsm28@gcc.gnu.org> Reply-To: gcc-bugzilla@gcc.gnu.org Subject: [Bug c/17913] [4.0 Regression] ICE jumping into statement expression X-Bugzilla-Reason: CC X-SW-Source: 2004-10/txt/msg01884.txt.bz2 List-Id: ------- Additional Comments From pinskia at gcc dot gnu dot org 2004-10-14 03:47 ------- Yes we remove it because ({ a : 1; 1; }) is marked as no has SIDE_EFFECTS which fold changes COND_EXPR ; to just true; which removes the label. The ICE is a regression but before in both 3.1 and 3.3, we don't produce the local label at all, I have not tested 2.95.3 at all. -- What |Removed |Added ---------------------------------------------------------------------------- Keywords| |ice-on-valid-code, wrong- | |code http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17913