From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.43) id 1K95q7-0001eh-Cx for qemu-devel@nongnu.org; Wed, 18 Jun 2008 18:09:15 -0400 Received: from exim by lists.gnu.org with spam-scanned (Exim 4.43) id 1K95q6-0001eJ-1P for qemu-devel@nongnu.org; Wed, 18 Jun 2008 18:09:15 -0400 Received: from [199.232.76.173] (port=46652 helo=monty-python.gnu.org) by lists.gnu.org with esmtp (Exim 4.43) id 1K95q5-0001eG-RG for qemu-devel@nongnu.org; Wed, 18 Jun 2008 18:09:13 -0400 Received: from bangui.magic.fr ([195.154.194.245]:43280) by monty-python.gnu.org with esmtps (TLS-1.0:DHE_RSA_AES_256_CBC_SHA1:32) (Exim 4.60) (envelope-from ) id 1K95q5-0003rF-CE for qemu-devel@nongnu.org; Wed, 18 Jun 2008 18:09:13 -0400 Received: from [192.168.0.2] (ppp-36.net-123.static.magiconline.fr [80.118.184.36]) by bangui.magic.fr (8.13.1/8.13.1) with ESMTP id m5IM9BI3019196 for ; Thu, 19 Jun 2008 00:09:11 +0200 Subject: Re: [Qemu-devel] [PATCH] ppc32 guests: fix computation of XER.{CA, OV} in addme, subfme, mullwo From: "J. Mayer" In-Reply-To: <200806182339.32714.jseward@acm.org> References: <200805110204.47184.jseward@acm.org> <200806180123.49284.jseward@acm.org> <1213824735.19143.35.camel@localhost> <200806182339.32714.jseward@acm.org> Content-Type: text/plain Date: Thu, 19 Jun 2008 00:09:10 +0200 Message-Id: <1213826950.19143.47.camel@localhost> Mime-Version: 1.0 Content-Transfer-Encoding: 7bit Reply-To: qemu-devel@nongnu.org List-Id: qemu-devel.nongnu.org List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: qemu-devel@nongnu.org On Wed, 2008-06-18 at 23:39 +0200, Julian Seward wrote: > On Wednesday 18 June 2008 23:32, J. Mayer wrote: > > Looking at the (relativelly) recent changes in this code, I can see that > > I did an optimization that seems correct at first sight but... > > [ .. big patch deleted .. ] > > Do you feel like this patch is buggy ? > > Why don't you just commit the patch I sent? It's shorter, it > fixes the problem I reported and it does not cause any regressions. it's ugly, obfuscated code, adding useless temporary variables and tests that used not to be needed to achieve the same computation (where we want the code to be fast and would like to remove all tests in flags computations...). Furthermore, it's inconsistent as more operations use similar code then would need to be patched too if those one are false. Then, I got numerous test results which gives the correct result with the same arguments you use, running qemu on different hosts (x86, x86_64 and PPC32/64), then I must find time to redo all those tests to get convinced it's now broken, find what did broke it, then rollback the changes that broke this code. -- J. Mayer Never organized