From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <gcc-return-61164-listarch-gcc=gcc.gnu.org@gcc.gnu.org>
Received: (qmail 15757 invoked by alias); 6 Oct 2002 14:29:44 -0000
Mailing-List: contact gcc-help@gcc.gnu.org; run by ezmlm
Precedence: bulk
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
Received: (qmail 15749 invoked from network); 6 Oct 2002 14:29:43 -0000
Received: from unknown (HELO tornado.toronto.redhat.com) (216.138.202.10)
  by sources.redhat.com with SMTP; 6 Oct 2002 14:29:43 -0000
Received: from tornado.toronto.redhat.com (localhost [127.0.0.1])
	by tornado.toronto.redhat.com (8.12.5/8.12.5) with ESMTP id g96ETgMu025630;
	Sun, 6 Oct 2002 10:29:42 -0400
Received: (from dnovillo@localhost)
	by tornado.toronto.redhat.com (8.12.5/8.12.5/Submit) id g96ETfj3025628;
	Sun, 6 Oct 2002 10:29:41 -0400
Date: Sun, 06 Oct 2002 08:38:00 -0000
From: Diego Novillo <dnovillo@acm.org>
To: zhongda lu <ludaxd@yahoo.com.cn>
Cc: gcc@gcc.gnu.org
Subject: Re: [tree-ssa-branch] Can I utilize the CFG?
Message-ID: <20021006142941.GB25559@tornado.toronto.redhat.com>
References: <20021006075115.53840.qmail@web15107.mail.bjs.yahoo.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
In-Reply-To: <20021006075115.53840.qmail@web15107.mail.bjs.yahoo.com>
User-Agent: Mutt/1.4i
X-SW-Source: 2002-10/txt/msg00388.txt.bz2

On Sun, 06 Oct 2002, zhongda lu wrote:

> Maybe this is not hopeful for me,because I would like the
> original AST being transformed to CFG.How could I control your
> code to achieve this? Another
>
You can't.  The CFG is built after simplification.  If you feed
it an unsimplified program it will either build the wrong CFG or
abort.  The flowgraph builder is not prepared to deal with front
end trees.


> thing  is whether the simplification of the AST will change the
> control flow of a function and will affect the analysis on the
> CFG.
>
Nope.  The only thing simplification does is break appart
convoluted expressions to expose side-effects and make the code
easier to analyze and transform.  The resulting program is
semantically equivalent to the original.


>If I generate the data dependence graph of a  function,can I
>utilize the reaching definition the tree-ssa branch has
>completed .thanks a lot.
> 
Yes.


Diego.