From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 21078 invoked by alias); 1 Apr 2004 07:18:46 -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 21060 invoked by alias); 1 Apr 2004 07:18:45 -0000 Date: Thu, 01 Apr 2004 07:18:00 -0000 Message-ID: <20040401071845.21059.qmail@sources.redhat.com> From: "bje at au1 dot ibm dot com" To: gcc-bugs@gcc.gnu.org In-Reply-To: <20040325051437.14730.pinskia@gcc.gnu.org> References: <20040325051437.14730.pinskia@gcc.gnu.org> Reply-To: gcc-bugzilla@gcc.gnu.org Subject: [Bug middle-end/14730] [tree-ssa] ICE on a switch statement with a case which is bigger than the type of switcher X-Bugzilla-Reason: CC X-SW-Source: 2004-04/txt/msg00028.txt.bz2 List-Id: ------- Additional Comments From bje at au1 dot ibm dot com 2004-04-01 07:18 ------- Subject: PR 14730 The bug originator says that he expects that the out-of-range case label would be optimised away, however I feel that we should focus on getting tree-ssa to not regress from the mainline at this stage (which retains the case 256 label and promotes the controlling expression to int). The dce2 pass is removing the cast: Eliminating unnecessary statements: Deleting : T.0_2 = (int)i_1; Ben -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=14730