Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751215AbWCAVfu (ORCPT ); Wed, 1 Mar 2006 16:35:50 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1751277AbWCAVfu (ORCPT ); Wed, 1 Mar 2006 16:35:50 -0500 Received: from main.gmane.org ([80.91.229.2]:32415 "EHLO ciao.gmane.org") by vger.kernel.org with ESMTP id S1751215AbWCAVft (ORCPT ); Wed, 1 Mar 2006 16:35:49 -0500 X-Injected-Via-Gmane: http://gmane.org/ To: linux-kernel@vger.kernel.org From: Giuseppe Bilotta Subject: Re: [PATCH] mm: implement swap prefetching (v26) Date: Wed, 1 Mar 2006 22:34:01 +0100 Message-ID: <1h8uuytzsq106$.k1fw2sw35o89$.dlg@40tude.net> References: <200602172235.40019.kernel@kolivas.org> <3b0ffc1f0602170618u7a1ad877s337de33c0a8f44f9@mail.gmail.com> <200602180126.58519.kernel@kolivas.org> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit X-Complaints-To: usenet@sea.gmane.org X-Gmane-NNTP-Posting-Host: host-84-220-48-150.cust-adsl.tiscali.it User-Agent: 40tude_Dialog/2.0.15.1 Cc: ck@vds.kolivas.org Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1543 Lines: 29 [Hope I managed to keep the CC list intact ... I'm using a newsreader to follow LKML from gmane] On Sat, 18 Feb 2006 01:26:57 +1100, Con Kolivas wrote: > On Saturday 18 February 2006 01:18, Kevin Radloff wrote: >> On 2/17/06, Con Kolivas wrote: >>> Added disabling of swap prefetching when laptop_mode is enabled. >> >> Why bother with this? As someone commented in a previous thread, >> wouldn't it be better to let the laptop_mode script handle it? > > The discussion was about what size to make the swap prefetching. Since the > size is not user tunable any more that is not the case. I had an offlist > discussion with Bart Samwel about it and basically if your drive spins down > at 5 seconds (which is what commonly happens with laptop mode) you will never > have an opportunity to prefetch. This means swap prefetch will basically > always spin up the drive nullifying laptop mode. On balance if you care about > power more than anything to actually set laptop mode I suspect you wont want > prefetch using any more power. Would it make any sense to just delay swap prefetch execution and/or analysis to "as soon as the HD is spun up/just before the HD is spun down", rather than completely disabling it? -- Giuseppe "Oblomov" Bilotta Hic manebimus optime - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/