From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <gcc-return-161216-listarch-gcc=gcc.gnu.org@gcc.gnu.org>
Received: (qmail 13553 invoked by alias); 27 Apr 2010 07:57:37 -0000
Received: (qmail 13539 invoked by uid 22791); 27 Apr 2010 07:57:34 -0000
X-SWARE-Spam-Status: No, hits=-1.1 required=5.0 	tests=BAYES_00,DKIM_SIGNED,DKIM_VALID,DKIM_VALID_AU,FREEMAIL_FROM,SARE_MSGID_LONG45
X-Spam-Check-By: sourceware.org
Received: from mail-ww0-f47.google.com (HELO mail-ww0-f47.google.com) (74.125.82.47)     by sourceware.org (qpsmtpd/0.43rc1) with ESMTP; Tue, 27 Apr 2010 07:57:29 +0000
Received: by wwb17 with SMTP id 17so1545540wwb.20         for <gcc@gcc.gnu.org>; Tue, 27 Apr 2010 00:57:27 -0700 (PDT)
MIME-Version: 1.0
Received: by 10.216.88.143 with SMTP id a15mr4872277wef.6.1272355046834; Tue,  	27 Apr 2010 00:57:26 -0700 (PDT)
Received: by 10.216.0.145 with HTTP; Tue, 27 Apr 2010 00:57:26 -0700 (PDT)
In-Reply-To: <m2s84fc9c001004260731s9a185937l10c2f703ac94846f@mail.gmail.com>
References: <t2vbba479b11004132348r93afd614ybbbf1e3fa5b8c3b4@mail.gmail.com> 	 <t2n84fc9c001004140228hd06b1a78n26b793a6ee50b73a@mail.gmail.com> 	 <w2qbba479b11004220904nd6de0b02r801eee25e9898de7@mail.gmail.com> 	 <g2k84fc9c001004230500ha7849565jba776509a9c72fda@mail.gmail.com> 	 <m2xbba479b11004260043i5fcd5c55z17bb6595acdafb51@mail.gmail.com> 	 <m2m84fc9c001004260259p21f5045dxb40a4fc7fda08c5a@mail.gmail.com> 	 <j2obba479b11004260642jb2d2769aya4e2f487be3d1f6d@mail.gmail.com> 	 <m2s84fc9c001004260731s9a185937l10c2f703ac94846f@mail.gmail.com>
Date: Tue, 27 Apr 2010 08:04:00 -0000
Message-ID: <o2zbba479b11004270057o57d4b9a2ndde26e75ec98b60@mail.gmail.com>
Subject: Re: vectorization, scheduling and aliasing
From: roy rosen <roy.1rosen@gmail.com>
To: Richard Guenther <richard.guenther@gmail.com>
Cc: gcc@gcc.gnu.org
Content-Type: text/plain; charset=ISO-8859-1
Mailing-List: contact gcc-help@gcc.gnu.org; run by ezmlm
Precedence: bulk
List-Id: <gcc.gcc.gnu.org>
List-Archive: <http://gcc.gnu.org/ml/gcc/>
List-Post: <mailto:gcc@gcc.gnu.org>
List-Help: <http://gcc.gnu.org/ml/>
Sender: gcc-owner@gcc.gnu.org
X-SW-Source: 2010-04/txt/msg00872.txt.bz2

Hi,

I have looked a bit more and tried also ia-64 and bfin and actually I
can't find a single example where vectorized code using __restrict__
variables would break the dependency between stores and loads.

for this simple program:

unsigned short xxx(unsigned short* __restrict__ a, unsigned short*
__restrict__ b)
{
   int i;
   for (i = 0; i < 8; i++)
   {
       a[i] = b[i];
   }
   return 0;
}

After unrolling and vectorization I get for all architectures that
there is a dependency between the first store (to a) and the second
load (from b).
since these are restrict variables I think this behavior is incorrect.

Can someone please show me such an example? Or is this feature not yet
supported?

Thanks, Roy.