From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.43) id 1NXqK5-0000ST-7Y for qemu-devel@nongnu.org; Thu, 21 Jan 2010 01:15:17 -0500 Received: from exim by lists.gnu.org with spam-scanned (Exim 4.43) id 1NXqK0-0000SD-RC for qemu-devel@nongnu.org; Thu, 21 Jan 2010 01:15:16 -0500 Received: from [199.232.76.173] (port=57780 helo=monty-python.gnu.org) by lists.gnu.org with esmtp (Exim 4.43) id 1NXqK0-0000SA-Mp for qemu-devel@nongnu.org; Thu, 21 Jan 2010 01:15:12 -0500 Received: from rio.cs.utah.edu ([155.98.64.241]:34195 helo=mail-svr1.cs.utah.edu) by monty-python.gnu.org with esmtp (Exim 4.60) (envelope-from ) id 1NXqJz-0007zD-Ff for qemu-devel@nongnu.org; Thu, 21 Jan 2010 01:15:12 -0500 Received: from localhost (localhost [127.0.0.1]) by mail-svr1.cs.utah.edu (Postfix) with ESMTP id BDD6A6500CB for ; Wed, 20 Jan 2010 23:15:08 -0700 (MST) Received: from mail-svr1.cs.utah.edu ([127.0.0.1]) by localhost (mail-svr1.cs.utah.edu [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id QsU-Nt-ZlBOM for ; Wed, 20 Jan 2010 23:15:08 -0700 (MST) Received: from [192.168.15.5] (75-169-48-51.slkc.qwest.net [75.169.48.51]) by smtps.cs.utah.edu (Postfix) with ESMTPSA id 61F4F6500C3 for ; Wed, 20 Jan 2010 23:15:08 -0700 (MST) Message-ID: <4B57F0EB.1050905@cs.utah.edu> Date: Wed, 20 Jan 2010 23:15:07 -0700 From: John Regehr MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Subject: [Qemu-devel] possible qemu miscompilation by latest gcc List-Id: qemu-devel.nongnu.org List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: qemu-devel@nongnu.org Hi folks- Just wanted to let you know that perhaps the function helper_neon_rshl_s8() is being miscompiled by the latest gcc. I'm using qemu 0.12.2 and gcc rev 156103, which is a pre-version of gcc 4.5. This is on an x86 machine running Ubuntu 9.10. At -O2 or higher this is the resulting object code: 00002060 : 2060: 31 c0 xor %eax,%eax 2062: c3 ret If this is not the intended result, then either the function has a latent bug or else the compiler is doing something bad. Hope this is helpful, John Regehr