From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail137.messagelabs.com (mail137.messagelabs.com [216.82.249.19]) by kanga.kvack.org (Postfix) with ESMTP id 062536B004D for ; Sun, 16 Aug 2009 02:00:07 -0400 (EDT) Received: from d23relay01.au.ibm.com (d23relay01.au.ibm.com [202.81.31.243]) by e23smtp05.au.ibm.com (8.14.3/8.13.1) with ESMTP id n7G5vZq8009814 for ; Sun, 16 Aug 2009 15:57:35 +1000 Received: from d23av03.au.ibm.com (d23av03.au.ibm.com [9.190.234.97]) by d23relay01.au.ibm.com (8.13.8/8.13.8/NCO v10.0) with ESMTP id n7G604Uo418190 for ; Sun, 16 Aug 2009 16:00:04 +1000 Received: from d23av03.au.ibm.com (loopback [127.0.0.1]) by d23av03.au.ibm.com (8.12.11.20060308/8.13.3) with ESMTP id n7G602L0031452 for ; Sun, 16 Aug 2009 16:00:03 +1000 Date: Sun, 16 Aug 2009 11:29:57 +0530 From: Balbir Singh Subject: Re: [RFC] respect the referenced bit of KVM guest pages? Message-ID: <20090816055957.GS5087@balbir.in.ibm.com> Reply-To: balbir@linux.vnet.ibm.com References: <20090806100824.GO23385@random.random> <4A7AAE07.1010202@redhat.com> <20090806102057.GQ23385@random.random> <20090806105932.GA1569@localhost> <4A7AC201.4010202@redhat.com> <20090806130631.GB6162@localhost> <4A7AD79E.4020604@redhat.com> <20090816032822.GB6888@localhost> <4A878377.70502@redhat.com> <20090816045522.GA13740@localhost> MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Disposition: inline In-Reply-To: <20090816045522.GA13740@localhost> Sender: owner-linux-mm@kvack.org To: Wu Fengguang Cc: Rik van Riel , Avi Kivity , Andrea Arcangeli , "Dike, Jeffrey G" , "Yu, Wilfred" , "Kleen, Andi" , Hugh Dickins , Andrew Morton , Christoph Lameter , KOSAKI Motohiro , Mel Gorman , LKML , linux-mm List-ID: * Wu Fengguang [2009-08-16 12:55:22]: > On Sun, Aug 16, 2009 at 11:56:39AM +0800, Rik van Riel wrote: > > Wu Fengguang wrote: > > > > > Right, but I meant busty page allocations and accesses on them, which > > > can make a large continuous segment of referenced pages in LRU list, > > > say 50MB. They may or may not be valuable as a whole, however a local > > > algorithm may keep the first 4MB and drop the remaining 46MB. > > > > I wonder if the problem is that we simply do not keep a large > > enough inactive list in Jeff's test. If we do not, pages do > > not have a chance to be referenced again before the reclaim > > code comes in. > > Exactly, that's the case I call the list FIFO. > > > The cgroup stats should show how many active anon and inactive > > anon pages there are in the cgroup. > > Jeff, can you have a look at these stats? Thanks! Another experiment would be to toy with memory.swappiness (although defaults should work well). Could you compare the in-guest values of nr_*active* with the cgroup values as seen by the host? -- Balbir -- 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