Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id ; Sat, 19 Jan 2002 00:46:54 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id ; Sat, 19 Jan 2002 00:46:43 -0500 Received: from khan.acc.umu.se ([130.239.18.139]:1209 "EHLO khan.acc.umu.se") by vger.kernel.org with ESMTP id ; Sat, 19 Jan 2002 00:46:28 -0500 Date: Sat, 19 Jan 2002 06:46:14 +0100 From: David Weinehall To: David Luyer Cc: "'Oliver Xymoron'" , linux-kernel@vger.kernel.org Subject: Re: vm philosophising Message-ID: <20020119064613.C18492@khan.acc.umu.se> In-Reply-To: <007601c1a0a4$9ddc5d70$46943ecb@pacific.net.au> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.5.1i In-Reply-To: <007601c1a0a4$9ddc5d70$46943ecb@pacific.net.au>; from david_luyer@pacific.net.au on Sat, Jan 19, 2002 at 03:49:02PM +1100 Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Sat, Jan 19, 2002 at 03:49:02PM +1100, David Luyer wrote: > I wrote: > > Alan Cox wrote: > > > > There is another VM that has a property that people would like: > > > > deterministically handling memory exhaustion. > > > > Unfortunately, that VM > > > > probably can't co-exist with over-commit and the > > > > performance gains that > > > > affords. > > > > > > It can definitely co-exist. Overcommit control is just a > > > book keeping > > > exercise on address space commits. > > [...] > > and the comment I somehow missed putting on the end: > > If you want to philosophise about VM strategies, think of > overcommit as "ethernet" and precommit as "token ring". You mean, that while technically superior, precommit suffers from a topological problem and the fact that a very expensive concentrator is needed?! ;-) Token Ring still lives in the spirit, though it's called FDDI nowadays... /David _ _ // David Weinehall /> Northern lights wander \\ // Maintainer of the v2.0 kernel // Dance across the winter sky // \> http://www.acc.umu.se/~tao/