From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752571AbZHTFRF (ORCPT ); Thu, 20 Aug 2009 01:17:05 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1751500AbZHTFRE (ORCPT ); Thu, 20 Aug 2009 01:17:04 -0400 Received: from e28smtp04.in.ibm.com ([59.145.155.4]:60187 "EHLO e28smtp04.in.ibm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751181AbZHTFRC (ORCPT ); Thu, 20 Aug 2009 01:17:02 -0400 Date: Thu, 20 Aug 2009 10:46:56 +0530 From: Balbir Singh To: Wu Fengguang Cc: KAMEZAWA Hiroyuki , Andrew Morton , KOSAKI Motohiro , Rik van Riel , Johannes Weiner , Avi Kivity , Andrea Arcangeli , "Dike, Jeffrey G" , Hugh Dickins , Christoph Lameter , Mel Gorman , LKML , linux-mm , "nishimura@mxp.nes.nec.co.jp" , "lizf@cn.fujitsu.com" , "menage@google.com" Subject: Re: [PATCH -v2] mm: do batched scans for mem_cgroup Message-ID: <20090820051656.GB26265@balbir.in.ibm.com> Reply-To: balbir@linux.vnet.ibm.com References: <20090820024929.GA19793@localhost> <20090820121347.8a886e4b.kamezawa.hiroyu@jp.fujitsu.com> <20090820040533.GA27540@localhost> MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Disposition: inline In-Reply-To: <20090820040533.GA27540@localhost> User-Agent: Mutt/1.5.18 (2008-05-17) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org * Wu Fengguang [2009-08-20 12:05:33]: > On Thu, Aug 20, 2009 at 11:13:47AM +0800, KAMEZAWA Hiroyuki wrote: > > On Thu, 20 Aug 2009 10:49:29 +0800 > > Wu Fengguang wrote: > > > > > For mem_cgroup, shrink_zone() may call shrink_list() with nr_to_scan=1, > > > in which case shrink_list() _still_ calls isolate_pages() with the much > > > larger SWAP_CLUSTER_MAX. It effectively scales up the inactive list > > > scan rate by up to 32 times. > > > > > > For example, with 16k inactive pages and DEF_PRIORITY=12, (16k >> 12)=4. > > > So when shrink_zone() expects to scan 4 pages in the active/inactive > > > list, it will be scanned SWAP_CLUSTER_MAX=32 pages in effect. > > > > > > The accesses to nr_saved_scan are not lock protected and so not 100% > > > accurate, however we can tolerate small errors and the resulted small > > > imbalanced scan rates between zones. > > > > > > This batching won't blur up the cgroup limits, since it is driven by > > > "pages reclaimed" rather than "pages scanned". When shrink_zone() > > > decides to cancel (and save) one smallish scan, it may well be called > > > again to accumulate up nr_saved_scan. > > > > > > It could possibly be a problem for some tiny mem_cgroup (which may be > > > _full_ scanned too much times in order to accumulate up nr_saved_scan). > > > > > > CC: Rik van Riel > > > CC: Minchan Kim > > > CC: Balbir Singh > > > CC: KOSAKI Motohiro > > > CC: KAMEZAWA Hiroyuki > > > Signed-off-by: Wu Fengguang > > > --- > > > > Hmm, how about this ? > > == > > Now, nr_saved_scan is tied to zone's LRU. > > But, considering how vmscan works, it should be tied to reclaim_stat. > > > > By this, memcg can make use of nr_saved_scan information seamlessly. > > Good idea, full patch updated with your signed-off-by :) > > Thanks, > Fengguang > --- > mm: do batched scans for mem_cgroup > > For mem_cgroup, shrink_zone() may call shrink_list() with nr_to_scan=1, > in which case shrink_list() _still_ calls isolate_pages() with the much > larger SWAP_CLUSTER_MAX. It effectively scales up the inactive list > scan rate by up to 32 times. > > For example, with 16k inactive pages and DEF_PRIORITY=12, (16k >> 12)=4. > So when shrink_zone() expects to scan 4 pages in the active/inactive > list, it will be scanned SWAP_CLUSTER_MAX=32 pages in effect. > > The accesses to nr_saved_scan are not lock protected and so not 100% > accurate, however we can tolerate small errors and the resulted small > imbalanced scan rates between zones. > > This batching won't blur up the cgroup limits, since it is driven by > "pages reclaimed" rather than "pages scanned". When shrink_zone() > decides to cancel (and save) one smallish scan, it may well be called > again to accumulate up nr_saved_scan. > > It could possibly be a problem for some tiny mem_cgroup (which may be > _full_ scanned too much times in order to accumulate up nr_saved_scan). > Looks good to me, how did you test it? Acked-by: Balbir Singh -- Balbir