From mboxrd@z Thu Jan  1 00:00:00 1970
Received: from list by monty-python.gnu.org with tmda-scanned (Exim 4.24)
	id 1AEbvX-0001SW-HY
	for qemu-devel@nongnu.org; Tue, 28 Oct 2003 17:02:59 -0500
Received: from mail by monty-python.gnu.org with spam-scanned (Exim 4.24)
	id 1AEbv1-0001IJ-B3
	for qemu-devel@nongnu.org; Tue, 28 Oct 2003 17:02:58 -0500
Received: from [193.252.22.29] (helo=mwinf0202.wanadoo.fr)
	by monty-python.gnu.org with esmtp (Exim 4.24) id 1AEbWZ-0000RV-9E
	for qemu-devel@nongnu.org; Tue, 28 Oct 2003 16:37:11 -0500
Received: from free.fr (ATuileries-112-1-1-209.w80-11.abo.wanadoo.fr
	[80.11.167.209])
	by mwinf0202.wanadoo.fr (SMTP Server) with ESMTP id B8C90A4007CF
	for <qemu-devel@nongnu.org>; Tue, 28 Oct 2003 22:37:05 +0100 (CET)
Message-ID: <3F9EE1A5.4000401@free.fr>
Date: Tue, 28 Oct 2003 22:37:41 +0100
From: Fabrice Bellard <fabrice.bellard@free.fr>
MIME-Version: 1.0
Subject: Re: [Qemu-devel] Math/FP problems in 0.5.0?
References: <1067319999.1257.9.camel@shawn.local>
In-Reply-To: <1067319999.1257.9.camel@shawn.local>
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
Reply-To: qemu-devel@nongnu.org
List-Id: <qemu-devel.nongnu.org>
List-Unsubscribe: <http://mail.nongnu.org/mailman/listinfo/qemu-devel>,
	<mailto:qemu-devel-request@nongnu.org?subject=unsubscribe>
List-Archive: <http://mail.gnu.org/pipermail/qemu-devel>
List-Post: <mailto:qemu-devel@nongnu.org>
List-Help: <mailto:qemu-devel-request@nongnu.org?subject=help>
List-Subscribe: <http://mail.nongnu.org/mailman/listinfo/qemu-devel>,
	<mailto:qemu-devel-request@nongnu.org?subject=subscribe>
To: qemu-devel@nongnu.org

Leigh Dyer wrote:
> Hi,
> 
> I've been playing with both CVS and the 0.5.0 release of qemu today,
> using a debian system that I installed under bochs. The system boots and
> seems to run quite well, but I noticed some numerical problems showing
> up - e2fsck's percentage meter switched between 0.0% and nan%, and
> running nbench produced similar results, with either 0 or nan for all
> reported values in the output.
> 
> I thought that it might be a problem with my toolchain (gcc 3.3 in
> Debian unstable messes things up, so I use unstable's gcc 3.2), so I
> grabbed the qemu binary and linux test image from the website. Running
> df seems to show the problem as well: the Capacity of the filesystem is
> shown as 0%, and if you use df -h, Size, Used and Avail all show as 0G.

I am going to try to reproduce the problem. Maybe the memory macros for 
floating point access are incorrect.

Fabrice.