From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <gcc-patches-return-185201-listarch-gcc-patches=gcc.gnu.org@gcc.gnu.org>
Received: (qmail 14224 invoked by alias); 4 Jan 2007 19:54:00 -0000
Received: (qmail 14216 invoked by uid 22791); 4 Jan 2007 19:54:00 -0000
X-Spam-Check-By: sourceware.org
Received: from ug-out-1314.google.com (HELO ug-out-1314.google.com) (66.249.92.172)     by sourceware.org (qpsmtpd/0.31) with ESMTP; Thu, 04 Jan 2007 19:53:56 +0000
Received: by ug-out-1314.google.com with SMTP id s2so5426762uge         for <gcc-patches@gcc.gnu.org>; Thu, 04 Jan 2007 11:53:54 -0800 (PST)
Received: by 10.82.162.14 with SMTP id k14mr1642826bue.1167940433947;         Thu, 04 Jan 2007 11:53:53 -0800 (PST)
Received: by 10.82.150.13 with HTTP; Thu, 4 Jan 2007 11:53:53 -0800 (PST)
Message-ID: <84fc9c000701041153h64404483l207bc4d7333101f2@mail.gmail.com>
Date: Thu, 04 Jan 2007 19:54:00 -0000
From: "Richard Guenther" <richard.guenther@gmail.com>
To: "Robert Kennedy" <jimbob@google.com>
Subject: Re: [PATCH] Fold more aggressively for trip counts
Cc: gcc-patches@gcc.gnu.org
In-Reply-To: <17821.23092.925585.601518@whippen.corp.google.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
Content-Disposition: inline
References: <17819.5994.111024.985123@whippen.corp.google.com> 	 <84fc9c000701030557y771149b5kb19643b0a50dd502@mail.gmail.com> 	 <17821.22245.42440.581996@whippen.corp.google.com> 	 <84fc9c000701041145x79441d61o89b42aef4003b516@mail.gmail.com> 	 <17821.23092.925585.601518@whippen.corp.google.com>
X-IsSubscribed: yes
Mailing-List: contact gcc-patches-help@gcc.gnu.org; run by ezmlm
Precedence: bulk
List-Id: <gcc-patches.gcc.gnu.org>
List-Archive: <http://gcc.gnu.org/ml/gcc-patches/>
List-Post: <mailto:gcc-patches@gcc.gnu.org>
List-Help: <mailto:gcc-patches-help@gcc.gnu.org>
Sender: gcc-patches-owner@gcc.gnu.org
X-SW-Source: 2007-01/txt/msg00261.txt.bz2

On 1/4/07, Robert Kennedy <jimbob@google.com> wrote:
> > Well, a testcase looks like
> >
> > int foo(int a)
> > {
> >   return a * 1000 > 0;
> > }
> >
> > and then check the gimple tree dump.
>
> I tried that. Somehow (I'm not sure how) it gets folded anyway without
> my change. Whatever is doing it in that context does not apply during
> the loop optimizations.

I see the condition survive until asm even with -O2 on i686.

Richard.