From mboxrd@z Thu Jan 1 00:00:00 1970 Date: Sat, 31 Dec 2005 20:20:17 -0200 From: Marcelo Tosatti Subject: Re: [PATCH 1/9] clockpro-nonresident.patch Message-ID: <20051231222017.GC4024@dmt.cnet> References: <20051230223952.765.21096.sendpatchset@twins.localnet> <20051230224222.765.32499.sendpatchset@twins.localnet> <20051231011324.GB4913@dmt.cnet> <1136022886.17853.18.camel@twins> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: Sender: owner-linux-mm@kvack.org Return-Path: To: Rik van Riel Cc: Peter Zijlstra , linux-mm@kvack.org, linux-kernel@vger.kernel.org, Andrew Morton , Christoph Lameter , Wu Fengguang , Nick Piggin , Marijn Meijles List-ID: On Sat, Dec 31, 2005 at 09:53:11AM -0500, Rik van Riel wrote: > On Sat, 31 Dec 2005, Peter Zijlstra wrote: > > > > > +/* > > > > + * For interactive workloads, we remember about as many non-resident pages > > > > + * as we have actual memory pages. For server workloads with large inter- > > > > + * reference distances we could benefit from remembering more. > > > > + */ > > > > > > This comment is bogus. Interactive or server loads have nothing to do > > > with the inter reference distance. To the contrary, interactive loads > > > have a higher chance to contain large inter reference distances, and > > > many common server loads have strong locality. > > > > > > > > > > Happy to drop it, Rik? > > Sorry, but the comment is accurate. > > For interactive workloads you want to forget interreference > distances between two updatedbs, even if mozilla didn't get > used all weekend. > > OTOH, on NFS servers, or other systems with large interreference > distances, you may _need_ to remember a larger set of non-resident > pages in order to find the pages that are the hottest. > > In those workloads, the shortest inter-reference distance might > still be larger than the size of memory... Sure, for the few cases you describe here the comment is valid. Happy new year! -- 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