From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S263219AbTDLJnK (for ); Sat, 12 Apr 2003 05:43:10 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S263220AbTDLJnK (for ); Sat, 12 Apr 2003 05:43:10 -0400 Received: from siaag2ae.compuserve.com ([149.174.40.135]:15539 "EHLO siaag2ae.compuserve.com") by vger.kernel.org with ESMTP id S263219AbTDLJnI (for ); Sat, 12 Apr 2003 05:43:08 -0400 Date: Sat, 12 Apr 2003 05:52:43 -0400 From: Chuck Ebbert <76306.1226@compuserve.com> Subject: Re: Completely new idea to virtual memory To: John Bradford Cc: linux-kernel Message-ID: <200304120554_MC3-1-341A-4160@compuserve.com> MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org >> Why not also write pages out to swap before it's really necessary? >> If they were left mapped but marked as having up-to-date copies >> on swap, they could be discarded immediately if the system needed >> memory. (Of course if they got written to they would have to be >> paged out again.) > >Hmmm, interesting... The gains from this might be largely illusion but you could get some theoretical benefits: o could write out lots of pages to swap without having to flush TLBs like you would if doing normal swap, since page remains present o if swsusp does not need to write already-backed pages it could reduce suspend time Maybe something like this? if about_to_spin_down_disks || swap_disk_idle if lots_of_memory_free page_in if little_memory_free copy_pages_to_swap -- "Let's fight till six, and then have dinner," said Tweedledum. --Lewis Carroll, _Through the Looking Glass_