From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 14849 invoked by alias); 29 Jun 2011 09:15:25 -0000 Received: (qmail 14839 invoked by uid 22791); 29 Jun 2011 09:15:23 -0000 X-SWARE-Spam-Status: No, hits=-2.4 required=5.0 tests=AWL,BAYES_00,DKIM_SIGNED,DKIM_VALID,DKIM_VALID_AU,FREEMAIL_FROM,RCVD_IN_DNSWL_LOW X-Spam-Check-By: sourceware.org Received: from mail-ww0-f41.google.com (HELO mail-ww0-f41.google.com) (74.125.82.41) by sourceware.org (qpsmtpd/0.43rc1) with ESMTP; Wed, 29 Jun 2011 09:15:09 +0000 Received: by wwi14 with SMTP id 14so3627455wwi.2 for ; Wed, 29 Jun 2011 02:15:08 -0700 (PDT) MIME-Version: 1.0 Received: by 10.227.58.71 with SMTP id f7mr470086wbh.110.1309338908325; Wed, 29 Jun 2011 02:15:08 -0700 (PDT) Received: by 10.227.36.212 with HTTP; Wed, 29 Jun 2011 02:15:08 -0700 (PDT) In-Reply-To: <201106282236.53994.ebotcazou@adacore.com> References: <201106282236.53994.ebotcazou@adacore.com> Date: Wed, 29 Jun 2011 09:38:00 -0000 Message-ID: Subject: Re: [patch] Fix oversight in tuplification of DOM From: Richard Guenther To: Eric Botcazou Cc: gcc-patches@gcc.gnu.org Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable X-IsSubscribed: yes Mailing-List: contact gcc-patches-help@gcc.gnu.org; run by ezmlm Precedence: bulk List-Id: List-Archive: List-Post: List-Help: Sender: gcc-patches-owner@gcc.gnu.org X-SW-Source: 2011-06/txt/msg02209.txt.bz2 On Tue, Jun 28, 2011 at 10:36 PM, Eric Botcazou wro= te: > Hi, > > the attached testcase triggers an ICE when compiled at -O or above, on al= l the > open branches. =A0This is a regression introduced with the tuplification.= =A0The > problem is that 2 ARRAY_RANGE_REFs are recognized as equivalent, although= they > don't have the same number of elements. =A0This is so because their type = isn't > taken into account by the hash equality function as it simply isn't recor= ded > in initialize_hash_element (GIMPLE_SINGLE_RHS case). =A0Now in all the ot= her > cases it is recorded so this very likely is an oversight. > > Tested on x86_64-suse-linux, OK for all branches? Oops. This would also make two MEM-REFs equivalent which access different amount of memory. Ok. Thanks, Richard. > > 2011-06-28 =A0Eric Botcazou =A0 > > =A0 =A0 =A0 =A0* tree-ssa-dom.c (initialize_hash_element): Fix oversight. > > > 2011-06-28 =A0Eric Botcazou =A0 > > =A0 =A0 =A0 =A0* gnat.dg/opt17.ad[sb]: New test. > > > -- > Eric Botcazou >