From mboxrd@z Thu Jan 1 00:00:00 1970 Date: Thu, 16 Feb 2006 15:25:15 +0900 From: IWAMOTO Toshihiro Subject: Re: [PATCH 6/9] clockpro-clockpro.patch In-Reply-To: <1139985308.6722.4.camel@localhost.localdomain> References: <20051230223952.765.21096.sendpatchset@twins.localnet> <20051230224312.765.58575.sendpatchset@twins.localnet> <20060214072917.926A97402D@sv1.valinux.co.jp> <1139985308.6722.4.camel@localhost.localdomain> MIME-Version: 1.0 (generated by SEMI 1.14.6 - "Maruoka") Content-Type: text/plain; charset=US-ASCII Message-Id: <20060216062515.1832374044@sv1.valinux.co.jp> Sender: owner-linux-mm@kvack.org Return-Path: To: Peter Zijlstra Cc: IWAMOTO Toshihiro , linux-mm@kvack.org, Andrew Morton , Christoph Lameter , Wu Fengguang , Nick Piggin , Marijn Meijles , Rik van Riel , Marcelo Tosatti List-ID: At Wed, 15 Feb 2006 07:35:08 +0100, Peter Zijlstra wrote: > > On Tue, 2006-02-14 at 16:29 +0900, IWAMOTO Toshihiro wrote: > > Why do you put non-pagecache pages into hot lists? > > For hot pages, it means less time allowed to get their reference bits > > set. For cold test pages, test bits will be just cleared when Hhot > > passes. > > My assumption was that since that is the position just vacated, it will > be the position for new pages, as in CLOCK. I've just reread the paper > once again, but fail to see any part dealing with the insertion of new > pages (might be my bad, it's way too early - please point me to the > relevant section if you do find it). Look at the last paragraph of the section 4.3. -- IWAMOTO Toshihiro -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@kvack.org. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: email@kvack.org