From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <gcc-bugs-return-297493-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org>
Received: (qmail 29354 invoked by alias); 23 Oct 2009 19:54:21 -0000
Received: (qmail 29335 invoked by uid 22791); 23 Oct 2009 19:54:21 -0000
X-SWARE-Spam-Status: No, hits=-1.8 required=5.0 	tests=AWL,BAYES_00,SARE_MSGID_LONG40,SPF_PASS
X-Spam-Check-By: sourceware.org
Received: from mail-ew0-f221.google.com (HELO mail-ew0-f221.google.com) (209.85.219.221)     by sourceware.org (qpsmtpd/0.43rc1) with ESMTP; Fri, 23 Oct 2009 19:54:14 +0000
Received: by ewy21 with SMTP id 21so7866087ewy.8         for <multiple recipients>; Fri, 23 Oct 2009 12:54:12 -0700 (PDT)
MIME-Version: 1.0
Received: by 10.210.106.12 with SMTP id e12mr3331632ebc.11.1256327652114; Fri,  	23 Oct 2009 12:54:12 -0700 (PDT)
In-Reply-To: <20091023194615.26530.qmail@sourceware.org>
References: <bug-41811-7780@http.gcc.gnu.org/bugzilla/> <20091023194615.26530.qmail@sourceware.org>
From: Sebastian Pop <sebpop@gmail.com>
Date: Fri, 23 Oct 2009 19:54:00 -0000
Message-ID: <cb9d34b20910231253m173c36b1j36bd7cbab7998b71@mail.gmail.com>
Subject: Re: [Bug tree-optimization/41811] graphite miscompiles 454.calculix  	of the SPEC 2k6
To: gcc-bugzilla@gcc.gnu.org
Cc: gcc-bugs@gcc.gnu.org
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable
X-IsSubscribed: yes
Mailing-List: contact gcc-bugs-help@gcc.gnu.org; run by ezmlm
Precedence: bulk
List-Id: <gcc-bugs.gcc.gnu.org>
List-Archive: <http://gcc.gnu.org/ml/gcc-bugs/>
List-Post: <mailto:gcc-bugs@gcc.gnu.org>
List-Help: <mailto:gcc-bugs-help@gcc.gnu.org>
Sender: gcc-bugs-owner@gcc.gnu.org
X-SW-Source: 2009-10/txt/msg02055.txt.bz2

On Fri, Oct 23, 2009 at 14:46, spop at gcc dot gnu dot org
<gcc-bugzilla@gcc.gnu.org> wrote:
> and the code generated by CLooG for the interchange looks like this:
>
> for (scat_1=3D0;scat_1<=3D2;scat_1++) {
> =C2=A0for (scat_3=3D0;scat_3<=3D2;scat_3++) {
> =C2=A0 =C2=A0S4(scat_1,scat_3) ;
> =C2=A0 =C2=A0for (scat_5=3D0;scat_5<=3D2;scat_5++) {
> =C2=A0 =C2=A0 =C2=A0S5(scat_1,scat_5,scat_3) ;
> =C2=A0 =C2=A0}
> =C2=A0 =C2=A0S7(scat_1,scat_3) ;
> =C2=A0 =C2=A0S18(scat_1,scat_3) ;
> =C2=A0}

S7 and S18 should not be generated before S5 finishes to execute over
all the iterations of the original innermost loop (do k=3D1,20).

S7 and S18 contain the end of the reduction and the write in the array
xs(i,j) that is independent of the k loop.

> =C2=A0for (scat_3=3D3;scat_3<=3D19;scat_3++) {
> =C2=A0 =C2=A0for (scat_5=3D0;scat_5<=3D2;scat_5++) {
> =C2=A0 =C2=A0 =C2=A0S5(scat_1,scat_5,scat_3) ;
> =C2=A0 =C2=A0}
> =C2=A0}
> }