From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S264359AbTLYUwU (ORCPT ); Thu, 25 Dec 2003 15:52:20 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S264361AbTLYUwT (ORCPT ); Thu, 25 Dec 2003 15:52:19 -0500 Received: from citrine.spiritone.com ([216.99.193.133]:54410 "EHLO citrine.spiritone.com") by vger.kernel.org with ESMTP id S264359AbTLYUwS (ORCPT ); Thu, 25 Dec 2003 15:52:18 -0500 Date: Thu, 25 Dec 2003 12:52:09 -0800 From: "Martin J. Bligh" To: "Marcos D. Marado Torres" cc: linux-kernel , lse-tech Subject: Re: 2.6.0-mjb1 Message-ID: <175850000.1072385528@[10.10.2.4]> In-Reply-To: References: <165810000.1072370137@[10.10.2.4]> X-Mailer: Mulberry/2.2.1 (Linux/x86) MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org >> I'd be very interested in feedback from anyone willing to test on any >> platform, however large or small. > > It's working perfectly for me: in fact it's what I'm running right now. > I'm running it on an Asus M3700N laptop... > Any tests you want me to do, just e-mail asking them. Cool - thanks. It probably doesn't get as much testing under memory pressure as it should on laptop-sized systems (though I do my best on my own boxes). I'd be interested to see how it performs for people on desktop/laptops compared to virgin 2.6.0 - objrmap pays a little during pageout for what it gains during regular usage. Just regular desktop workload stuff, but measuring performance can be somewhat subjective unless you have a more formal (and artificial) test. As long as it's tangibly worse, that's fine. Or if people are bored, and want something functional to test ... ;-) Fiddle with some of the new config options, eg turn on 4/4 split or 2/2 split, or HZ=100, kgdb, kcg, schedstat, etc. Early printk is interesting too, but harder to test - you really need to make it lock up before console_init somehow (adding a BUG() or something). Thanks, M.