From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S269131AbTGJJ3V (ORCPT ); Thu, 10 Jul 2003 05:29:21 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S269134AbTGJJ3V (ORCPT ); Thu, 10 Jul 2003 05:29:21 -0400 Received: from [213.171.53.133] ([213.171.53.133]:15108 "EHLO gulipin.miee.ru") by vger.kernel.org with ESMTP id S269131AbTGJJ3S (ORCPT ); Thu, 10 Jul 2003 05:29:18 -0400 Date: Thu, 10 Jul 2003 12:45:39 +0400 From: Samium Gromoff To: akpm@osdl.org, mbligh@aracnet.com Cc: linux-kernel@vger.kernel.org Subject: Re: [Bug 890] New: performance regression compared to 2.4.20 under Message-Id: <20030710124539.1d9dab9f.deepfire@ibe.miee.ru> X-Mailer: Sylpheed version 0.9.0 (GTK+ 1.2.10; i386-debian-linux-gnu) Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Andrew Morton wrote: > "Martin J. Bligh" wrote: > > > > As can be seen, the differences are quite significant, about three seconds on > > average, which I believe may be related to the increased swapping time I have > > encountered. > > The 2.4 VM's virtual scan has the effect of swapping out one process at a > time. 2.5's physical(ish) scan doesn't have that side-effect. > > It means that in 2.4, the lucky processes can make decent progress. In > 2.5, everyone makes equal progress and everyone thrashes everyone else to > bits. I have precisely the same problem on my p166-16M RAM. Disk/controller are fast, and dma is on. 2.4 runs much much faster. Remote shell logins are more than twice as fast. Needless to say it runs 2.4 ;-) -- Cheers, Samium Gromoff regards, Samium Gromoff