From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <gcc-return-106622-listarch-gcc=gcc.gnu.org@gcc.gnu.org>
Received: (qmail 11523 invoked by alias); 15 Dec 2004 16:55:03 -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 6007 invoked from network); 15 Dec 2004 16:45:01 -0000
Received: from unknown (HELO develer.com) (151.38.19.110)
  by sourceware.org with SMTP; 15 Dec 2004 16:45:01 -0000
Received: (qmail 771 invoked from network); 15 Dec 2004 16:44:59 -0000
Received: from mimas.trilan (HELO mimas) (10.3.3.245)
  by trinity.trilan with SMTP; 15 Dec 2004 16:44:59 -0000
Message-ID: <06c401c4e2c5$6a294220$f503030a@mimas>
From: "Giovanni Bajo" <rasky@develer.com>
To: "Sebastian Pop" <sebastian.pop@cri.ensmp.fr>
Cc: <gcc@gcc.gnu.org>
References: <Pine.LNX.4.44.0412141534000.4584-100000@alwazn.tat.physik.uni-tuebingen.de> <20041215154618.GA7519@cri.ensmp.fr>
Subject: Re: Strange IV choices?
Date: Wed, 15 Dec 2004 16:55:00 -0000
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: 7bit
X-SW-Source: 2004-12/txt/msg00584.txt.bz2

Sebastian Pop <sebastian.pop@cri.ensmp.fr> wrote:

> In the second case the analyzer produces a scev_not_known element:
>
> (analyze_scalar_evolution
>   (loop_nb = 1)
>   (scalar = ostride_192)
> (get_scalar_evolution
>   (scalar = ostride_192)
>   (scalar_evolution = ))
> (analyze_initial_condition
>   (loop_phi_node =
> ostride_192 = PHI <ostride_362(11), ostride_125(16)>;)
>   (init_cond = ostride_125))
> (analyze_evolution_in_loop
>   (loop_phi_node = ostride_192 = PHI <ostride_362(11), ostride_125(16)>;)
>   (evolution_function = scev_not_known))
> (set_scalar_evolution
>   (scalar = ostride_192)
>   (scalar_evolution = ostride_192))
> )
>
> This situation is generated by the following code:
>
> # ostrideD.165952_5 = PHI <1(0)>;
> D.166161_150 = (*dD.166162_151)[1];
> ostrideD.165952_125 = ostrideD.165952_5 * D.166161_150;
> loop
>   # ostrideD.165952_192 = PHI <ostrideD.165952_362(11),
>   ostrideD.165952_125(16)>; D.166161_358 = (*dD.166162_356)[1];
>   ostrideD.165952_362 = ostrideD.165952_192 * D.166161_358;
> endloop

Then, why does it recognize the evolution of the first testcase? They should
be identical.
-- 
Giovanni Bajo