From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <gcc-return-106322-listarch-gcc=gcc.gnu.org@gcc.gnu.org>
Received: (qmail 13275 invoked by alias); 7 Dec 2004 13:59:18 -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 13159 invoked from network); 7 Dec 2004 13:59:10 -0000
Received: from unknown (HELO Cantor.suse.de) (195.135.220.2)
  by sourceware.org with SMTP; 7 Dec 2004 13:59:10 -0000
Received: from hermes.suse.de (hermes-ext.suse.de [195.135.221.8])
	(using TLSv1 with cipher EDH-RSA-DES-CBC3-SHA (168/168 bits))
	(No client certificate requested)
	by Cantor.suse.de (Postfix) with ESMTP id A4DB011DBEE4
	for <gcc@gcc.gnu.org>; Tue,  7 Dec 2004 14:59:09 +0100 (CET)
Message-ID: <41B5B781.2010602@suse.de>
Date: Tue, 07 Dec 2004 13:59:00 -0000
From: Paolo Carlini <pcarlini@suse.de>
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.7.3) Gecko/20040913
MIME-Version: 1.0
To: gcc@gcc.gnu.org
Subject: Gcc complex division (default) algorithm
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
X-SW-Source: 2004-12/txt/msg00284.txt.bz2

Hi everyone,

I would ask you to have a brief look to the recent messages in gcc-bugs
named *complex numbers: basically, it turns out that, by default (that
is, when flag_complex_divide_method == 0) we are using a ""naive"" algo
for complex division.

Now, I can definitely understand that the "improved" algorithm, already
implemented (see expand_complex_div_wide in gcc/tree-complex.c) is more
costly, but would like to know the rationale of the choice, its history,
future plans (C99?): comments are very welcome!

Thank you in advance,
Paolo.