Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752074Ab1D0EfC (ORCPT ); Wed, 27 Apr 2011 00:35:02 -0400 Received: from fgwmail5.fujitsu.co.jp ([192.51.44.35]:51176 "EHLO fgwmail5.fujitsu.co.jp" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751618Ab1D0EfA (ORCPT ); Wed, 27 Apr 2011 00:35:00 -0400 X-SecurityPolicyCheck-FJ: OK by FujitsuOutboundMailChecker v1.3.1 Date: Wed, 27 Apr 2011 13:28:14 +0900 From: KAMEZAWA Hiroyuki To: Ying Han Cc: "linux-mm@kvack.org" , "linux-kernel@vger.kernel.org" , "akpm@linux-foundation.org" , "balbir@linux.vnet.ibm.com" , Daisuke Nishimura , "kosaki.motohiro@jp.fujitsu.com" Subject: Re: [PATCH] memcg: reclaim memory from nodes in round robin Message-Id: <20110427132814.be22bab0.kamezawa.hiroyu@jp.fujitsu.com> In-Reply-To: References: <20110427115718.ab6c55ae.kamezawa.hiroyu@jp.fujitsu.com> Organization: FUJITSU Co. LTD. X-Mailer: Sylpheed 3.1.0 (GTK+ 2.10.14; i686-pc-mingw32) Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 5614 Lines: 129 On Tue, 26 Apr 2011 20:52:39 -0700 Ying Han wrote: > On Tue, Apr 26, 2011 at 7:57 PM, KAMEZAWA Hiroyuki > wrote: > > Now, memory cgroup's direct reclaim frees memory from the current node. > > But this has some troubles. In usual, when a set of threads works in > > cooperative way, they are tend to on the same node. So, if they hit > > limits under memcg, it will reclaim memory from themselves, it may be > > active working set. > > > > For example, assume 2 node system which has Node 0 and Node 1 > > and a memcg which has 1G limit. After some work, file cacne remains and > > and usages are > >   Node 0:  1M > >   Node 1:  998M. > > > > and run an application on Node 0, it will eats its foot before freeing > > unnecessary file caches. > > > > This patch adds round-robin for NUMA and adds equal pressure to each > > node. When using cpuset's spread memory feature, this will work very well. > > > > But yes, better algorithm is appreciated. > > > > From: Ying Han > > Signed-off-by: Ying Han > > Signed-off-by: KAMEZAWA Hiroyuki > > --- > >  include/linux/memcontrol.h |    1 + > >  mm/memcontrol.c            |   25 +++++++++++++++++++++++++ > >  mm/vmscan.c                |    9 ++++++++- > >  3 files changed, 34 insertions(+), 1 deletion(-) > > > > Index: memcg/include/linux/memcontrol.h > > =================================================================== > > --- memcg.orig/include/linux/memcontrol.h > > +++ memcg/include/linux/memcontrol.h > > @@ -108,6 +108,7 @@ extern void mem_cgroup_end_migration(str > >  */ > >  int mem_cgroup_inactive_anon_is_low(struct mem_cgroup *memcg); > >  int mem_cgroup_inactive_file_is_low(struct mem_cgroup *memcg); > > +int mem_cgroup_select_victim_node(struct mem_cgroup *memcg); > >  unsigned long mem_cgroup_zone_nr_pages(struct mem_cgroup *memcg, > >                                       struct zone *zone, > >                                       enum lru_list lru); > > Index: memcg/mm/memcontrol.c > > =================================================================== > > --- memcg.orig/mm/memcontrol.c > > +++ memcg/mm/memcontrol.c > > @@ -237,6 +237,7 @@ struct mem_cgroup { > >         * reclaimed from. > >         */ > >        int last_scanned_child; > > +       int last_scanned_node; > >        /* > >         * Should the accounting and control be hierarchical, per subtree? > >         */ > > @@ -1472,6 +1473,29 @@ mem_cgroup_select_victim(struct mem_cgro > >  } > > > >  /* > > + * Selecting a node where we start reclaim from. Because what we need is just > > + * reducing usage counter, start from anywhere is O,K. When considering > > + * memory reclaim from current node, there are pros. and cons. > > + * Freeing memory from current node means freeing memory from a node which > > + * we'll use or we've used. So, it may make LRU bad. And if several threads > > + * hit limits, it will see a contention on a node. But freeing from remote > > + * node mean more costs for memory reclaim because of memory latency. > > + * > > + * Now, we use round-robin. Better algorithm is welcomed. > > + */ > > +int mem_cgroup_select_victim_node(struct mem_cgroup *mem) > > +{ > > +       int node; > > + > > +       node = next_node(mem->last_scanned_node, node_states[N_HIGH_MEMORY]); > > +       if (node == MAX_NUMNODES) > > +               node = first_node(node_states[N_HIGH_MEMORY]); > > + > > +       mem->last_scanned_node = node; > > +       return node; > > +} > > + > > +/* > >  * Scan the hierarchy if needed to reclaim memory. We remember the last child > >  * we reclaimed from, so that we don't end up penalizing one child extensively > >  * based on its position in the children list. > > @@ -4678,6 +4702,7 @@ mem_cgroup_create(struct cgroup_subsys * > >                res_counter_init(&mem->memsw, NULL); > >        } > >        mem->last_scanned_child = 0; > > +       mem->last_scanned_node = MAX_NUMNODES; > >        INIT_LIST_HEAD(&mem->oom_notify); > > > >        if (parent) > > Index: memcg/mm/vmscan.c > > =================================================================== > > --- memcg.orig/mm/vmscan.c > > +++ memcg/mm/vmscan.c > > @@ -2198,6 +2198,7 @@ unsigned long try_to_free_mem_cgroup_pag > >  { > >        struct zonelist *zonelist; > >        unsigned long nr_reclaimed; > > +       int nid; > >        struct scan_control sc = { > >                .may_writepage = !laptop_mode, > >                .may_unmap = 1, > > @@ -2208,10 +2209,16 @@ unsigned long try_to_free_mem_cgroup_pag > >                .mem_cgroup = mem_cont, > >                .nodemask = NULL, /* we don't care the placement */ > >        }; > > +       /* > > +        * Unlike direct reclaim via allo_pages(), memcg's reclaim > > +        * don't take care from where we get free resouce. So, the node where > > +        * we need to start scan is not need to be current node. > > +        */ > Sorry, some typos. alloc_pages() instead of alloc_pages(). And "free resource". > ok, will fix. Thank you for pointing out. Thanks, -Kame -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/