From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1757315Ab2BPG5V (ORCPT ); Thu, 16 Feb 2012 01:57:21 -0500 Received: from mail-bk0-f46.google.com ([209.85.214.46]:33167 "EHLO mail-bk0-f46.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1757249Ab2BPG5S (ORCPT ); Thu, 16 Feb 2012 01:57:18 -0500 Message-ID: <4F3CA8CA.8020004@openvz.org> Date: Thu, 16 Feb 2012 10:57:14 +0400 From: Konstantin Khlebnikov User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.1.19) Gecko/20120201 Iceape/2.0.14 MIME-Version: 1.0 To: KAMEZAWA Hiroyuki CC: "linux-mm@kvack.org" , Andrew Morton , Johannes Weiner , "cgroups@vger.kernel.org" , "linux-kernel@vger.kernel.org" Subject: Re: [PATCH] memcg: rework inactive_ratio logic References: <20120215162442.13588.21790.stgit@zurg> <20120216103842.0c3e9258.kamezawa.hiroyu@jp.fujitsu.com> In-Reply-To: <20120216103842.0c3e9258.kamezawa.hiroyu@jp.fujitsu.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org KAMEZAWA Hiroyuki wrote: > On Wed, 15 Feb 2012 20:24:42 +0400 > Konstantin Khlebnikov wrote: > >> This patch adds mem_cgroup->inactive_ratio calculated from hierarchical memory limit. >> It updated at each limit change before shrinking cgroup to this new limit. >> Ratios for all child cgroups are updated too, because parent limit can affect them. >> Update precedure can be greatly optimized if its performance becomes the problem. >> Inactive ratio for unlimited or huge limit does not matter, because we'll never hit it. >> >> At global reclaim always use global ratio from zone->inactive_ratio. >> At mem-cgroup reclaim use inactive_ratio from target memory cgroup, >> this is cgroup which hit its limit and cause this reclaimer invocation. >> >> Thus, global memory reclaimer will try to keep ratio for all lru lists in zone >> above one mark, this guarantee that total ratio in this zone will be above too. >> Meanwhile mem-cgroup will do the same thing for its lru lists in all zones, and >> for all lru lists in all sub-cgroups in hierarchy. >> >> Also this patch removes some redundant code. >> >> Signed-off-by: Konstantin Khlebnikov > > Hmm, the main purpose of this patch is to remove calculation per get_scan_ratio() ? Technically, it was preparation for "mm: unify inactive_list_is_low()" from "memory book keeping" patchset. So, actually its main purpose is moving all active/inactive size calculation to mm/vmscan.c Also I trying to figure out most sane logic for inactive_ratio calculation, currently global memory reclaimer sometimes uses memcg-calculated ratio, it looks strange. >> --- >> include/linux/memcontrol.h | 16 ++------ >> mm/memcontrol.c | 85 ++++++++++++++++++++++++-------------------- >> mm/vmscan.c | 82 +++++++++++++++++++++++------------------- >> 3 files changed, 93 insertions(+), 90 deletions(-) >> static int mem_cgroup_resize_limit(struct mem_cgroup *memcg, >> unsigned long long val) >> { >> @@ -3422,6 +3416,7 @@ static int mem_cgroup_resize_limit(struct mem_cgroup *memcg, >> else >> memcg->memsw_is_minimum = false; >> } >> + mem_cgroup_update_inactive_ratio(memcg, val); >> mutex_unlock(&set_limit_mutex); >> >> if (!ret) >> @@ -3439,6 +3434,12 @@ static int mem_cgroup_resize_limit(struct mem_cgroup *memcg, >> if (!ret&& enlarge) >> memcg_oom_recover(memcg); >> >> + if (ret) { >> + mutex_lock(&set_limit_mutex); >> + mem_cgroup_update_inactive_ratio(memcg, RESOURCE_MAX); >> + mutex_unlock(&set_limit_mutex); >> + } > > Why RESOUECE_MAX ? resize was failed, so we return back normal value calculated from the current limit. target == RESOURCE_MAX isn't clip limit: min(RESOURCE_MAX, limit) == limit > > Thanks, > -Kame >