From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail202.messagelabs.com (mail202.messagelabs.com [216.82.254.227]) by kanga.kvack.org (Postfix) with ESMTP id 1511F6B005D for ; Mon, 16 Mar 2009 12:51:06 -0400 (EDT) Date: Mon, 16 Mar 2009 16:51:03 +0000 From: Mel Gorman Subject: Re: [PATCH 30/35] Skip the PCP list search by counting the order and type of pages on list Message-ID: <20090316165103.GN24293@csn.ul.ie> References: <1237196790-7268-1-git-send-email-mel@csn.ul.ie> <1237196790-7268-31-git-send-email-mel@csn.ul.ie> MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-15 Content-Disposition: inline In-Reply-To: Sender: owner-linux-mm@kvack.org To: Christoph Lameter Cc: Linux Memory Management List , Pekka Enberg , Rik van Riel , KOSAKI Motohiro , Johannes Weiner , Nick Piggin , Linux Kernel Mailing List , Lin Ming , Zhang Yanmin , Peter Zijlstra List-ID: On Mon, Mar 16, 2009 at 12:31:44PM -0400, Christoph Lameter wrote: > On Mon, 16 Mar 2009, Mel Gorman wrote: > > > > > struct per_cpu_pages { > > - int count; /* number of pages in the list */ > > + /* The total number of pages on the PCP lists */ > > + int count; > > + > > + /* Count of each migratetype and order */ > > + u8 mocount[MIGRATE_PCPTYPES][PAGE_ALLOC_COSTLY_ORDER+1]; > > What about overflow? You could have more than 255 pages of a given type in > a pcp. > Are you sure? The maximum size I would expect is pcp->high which is 6 * pcp->batch which is limited in size. I might have missed something though. -- Mel Gorman Part-time Phd Student Linux Technology Center University of Limerick IBM Dublin Software Lab -- 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