Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752145AbZGPEWf (ORCPT ); Thu, 16 Jul 2009 00:22:35 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1752048AbZGPEWe (ORCPT ); Thu, 16 Jul 2009 00:22:34 -0400 Received: from mail-gx0-f213.google.com ([209.85.217.213]:54400 "EHLO mail-gx0-f213.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750840AbZGPEWd (ORCPT ); Thu, 16 Jul 2009 00:22:33 -0400 DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type:content-transfer-encoding; b=MRs+HBKeKcrWMeCfJOySOvlQpmU5PRVayesQjCdMJpbUjs8UpW4vKFVrWMNRLE1Jq+ /C7m6Ee+AQ67T7SDfv1QFec8mL55LTfZaYH9tV6VPV+DxsAMsX2MELWuapTgDt8EvUMj lutOX0J79Mx+0hKZ3QA2zL5Z/lFMcLIxJG8es= MIME-Version: 1.0 In-Reply-To: <20090715201657.b01edccd.akpm@linux-foundation.org> References: <20090716094619.9D07.A69D9226@jp.fujitsu.com> <20090716095344.9D10.A69D9226@jp.fujitsu.com> <20090715201657.b01edccd.akpm@linux-foundation.org> Date: Thu, 16 Jul 2009 13:22:32 +0900 Message-ID: <28c262360907152122v4f30594cr677b744df3aaefcc@mail.gmail.com> Subject: Re: [PATCH 3/3] add isolate pages vmstat From: Minchan Kim To: Andrew Morton Cc: KOSAKI Motohiro , LKML , linux-mm , Wu Fengguang , Rik van Riel , Christoph Lameter Content-Type: text/plain; charset=UTF-8 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from base64 to 8bit by alpha.home.local id n6G4NHJ7030532 Content-Length: 8635 Lines: 5 On Thu, Jul 16, 2009 at 12:16 PM, AndrewMorton wrote:> On Thu, 16 Jul 2009 09:55:47 +0900 (JST) KOSAKI Motohiro wrote:>>> ChangeLog>>   Since v5>>    - Rewrote the description>>    - Treat page migration>>   Since v4>>    - Changed displaing order in show_free_areas() (as Wu's suggested)>>   Since v3>>    - Fixed misaccount page bug when lumby reclaim occur>>   Since v2>>    - Separated IsolateLRU field to Isolated(anon) and Isolated(file)>>   Since v1>>    - Renamed IsolatePages to IsolatedLRU>>>> ==================================>> Subject: [PATCH] add isolate pages vmstat>>>> If the system is running a heavy load of processes then concurrent reclaim>> can isolate a large numbe of pages from the LRU. /proc/meminfo and the>> output generated for an OOM do not show how many pages were isolated.>>>> This patch shows the information about isolated pages.>>>>>> reproduce way>> ----------------------->> % ./hackbench 140 process 1000>>    => OOM occur>>>> active_anon:146 inactive_anon:0 isolated_anon:49245>>  active_file:79 inactive_file:18 isolated_file:113>>  unevictable:0 dirty:0 writeback:0 unstable:0 buffer:39>>  free:370 slab_reclaimable:309 slab_unreclaimable:5492>>  mapped:53 shmem:15 pagetables:28140 bounce:0>>>>>> Signed-off-by: KOSAKI Motohiro >> Acked-by: Rik van Riel >> Acked-by: Wu Fengguang >> Reviewed-by: Minchan Kim >> --->>  drivers/base/node.c    |    4 ++++>>  fs/proc/meminfo.c      |    4 ++++>>  include/linux/mmzone.h |    2 ++>>  mm/migrate.c           |   11 +++++++++++>>  mm/page_alloc.c        |   12 +++++++++--->>  mm/vmscan.c            |   12 +++++++++++->>  mm/vmstat.c            |    2 ++>>  7 files changed, 43 insertions(+), 4 deletions(-)>>>> Index: b/fs/proc/meminfo.c>> ===================================================================>> --- a/fs/proc/meminfo.c>> +++ b/fs/proc/meminfo.c>> @@ -65,6 +65,8 @@ static int meminfo_proc_show(struct seq_>>               "Active(file):   %8lu kB\n">>               "Inactive(file): %8lu kB\n">>               "Unevictable:    %8lu kB\n">> +             "Isolated(anon): %8lu kB\n">> +             "Isolated(file): %8lu kB\n">>               "Mlocked:        %8lu kB\n">> Are these counters really important enough to justify being present in> /proc/meminfo?  They seem fairly low-level developer-only details.> Perhaps relegate them to /proc/vmstat?>>>  #ifdef CONFIG_HIGHMEM>>               "HighTotal:      %8lu kB\n">> @@ -110,6 +112,8 @@ static int meminfo_proc_show(struct seq_>>               K(pages[LRU_ACTIVE_FILE]),>>               K(pages[LRU_INACTIVE_FILE]),>>               K(pages[LRU_UNEVICTABLE]),>> +             K(global_page_state(NR_ISOLATED_ANON)),>> +             K(global_page_state(NR_ISOLATED_FILE)),>>               K(global_page_state(NR_MLOCK)),>>  #ifdef CONFIG_HIGHMEM>>               K(i.totalhigh),>> Index: b/include/linux/mmzone.h>> ===================================================================>> --- a/include/linux/mmzone.h>> +++ b/include/linux/mmzone.h>> @@ -100,6 +100,8 @@ enum zone_stat_item {>>       NR_BOUNCE,>>       NR_VMSCAN_WRITE,>>       NR_WRITEBACK_TEMP,      /* Writeback using temporary buffers */>> +     NR_ISOLATED_ANON,       /* Temporary isolated pages from anon lru */>> +     NR_ISOLATED_FILE,       /* Temporary isolated pages from file lru */>>       NR_SHMEM,               /* shmem pages (included tmpfs/GEM pages) */>>  #ifdef CONFIG_NUMA>>       NUMA_HIT,               /* allocated in intended node */>> Index: b/mm/page_alloc.c>> ===================================================================>> --- a/mm/page_alloc.c>> +++ b/mm/page_alloc.c>> @@ -2115,16 +2115,18 @@ void show_free_areas(void)>>               }>>       }>>>> -     printk("Active_anon:%lu active_file:%lu inactive_anon:%lu\n">> -             " inactive_file:%lu">> +     printk("active_anon:%lu inactive_anon:%lu isolated_anon:%lu\n">> +             " active_file:%lu inactive_file:%lu isolated_file:%lu\n">>               " unevictable:%lu">>               " dirty:%lu writeback:%lu unstable:%lu buffer:%lu\n">>               " free:%lu slab_reclaimable:%lu slab_unreclaimable:%lu\n">>               " mapped:%lu shmem:%lu pagetables:%lu bounce:%lu\n",>>               global_page_state(NR_ACTIVE_ANON),>> -             global_page_state(NR_ACTIVE_FILE),>>               global_page_state(NR_INACTIVE_ANON),>> +             global_page_state(NR_ISOLATED_ANON),>> +             global_page_state(NR_ACTIVE_FILE),>>               global_page_state(NR_INACTIVE_FILE),>> +             global_page_state(NR_ISOLATED_FILE),>>               global_page_state(NR_UNEVICTABLE),>>               global_page_state(NR_FILE_DIRTY),>>               global_page_state(NR_WRITEBACK),>> @@ -2152,6 +2154,8 @@ void show_free_areas(void)>>                       " active_file:%lukB">>                       " inactive_file:%lukB">>                       " unevictable:%lukB">> +                     " isolated(anon):%lukB">> +                     " isolated(file):%lukB">>                       " present:%lukB">>                       " mlocked:%lukB">>                       " dirty:%lukB">> @@ -2178,6 +2182,8 @@ void show_free_areas(void)>>                       K(zone_page_state(zone, NR_ACTIVE_FILE)),>>                       K(zone_page_state(zone, NR_INACTIVE_FILE)),>>                       K(zone_page_state(zone, NR_UNEVICTABLE)),>> +                     K(zone_page_state(zone, NR_ISOLATED_ANON)),>> +                     K(zone_page_state(zone, NR_ISOLATED_FILE)),>>                       K(zone->present_pages),>>                       K(zone_page_state(zone, NR_MLOCK)),>>                       K(zone_page_state(zone, NR_FILE_DIRTY)),>> Index: b/mm/vmscan.c>> ===================================================================>> --- a/mm/vmscan.c>> +++ b/mm/vmscan.c>> @@ -1067,6 +1067,8 @@ static unsigned long shrink_inactive_lis>>               unsigned long nr_active;>>               unsigned int count[NR_LRU_LISTS] = { 0, };>>               int mode = lumpy_reclaim ? ISOLATE_BOTH : ISOLATE_INACTIVE;>> +             unsigned long nr_anon;>> +             unsigned long nr_file;>>>>               nr_taken = sc->isolate_pages(sc->swap_cluster_max,>>                            &page_list, &nr_scan, sc->order, mode,>> @@ -1097,6 +1099,10 @@ static unsigned long shrink_inactive_lis>>               __mod_zone_page_state(zone, NR_INACTIVE_ANON,>>                                               -count[LRU_INACTIVE_ANON]);>>>> +             nr_anon = count[LRU_ACTIVE_ANON] + count[LRU_INACTIVE_ANON];>> +             nr_file = count[LRU_ACTIVE_FILE] + count[LRU_INACTIVE_FILE];>> +             __mod_zone_page_state(zone, NR_ISOLATED_ANON, nr_anon);>> +             __mod_zone_page_state(zone, NR_ISOLATED_FILE, nr_file);>>>>               reclaim_stat->recent_scanned[0] += count[LRU_INACTIVE_ANON];>>               reclaim_stat->recent_scanned[0] += count[LRU_ACTIVE_ANON];>> @@ -1164,6 +1170,9 @@ static unsigned long shrink_inactive_lis>>                               spin_lock_irq(&zone->lru_lock);>>                       }>>               }>> +             __mod_zone_page_state(zone, NR_ISOLATED_ANON, -nr_anon);>> +             __mod_zone_page_state(zone, NR_ISOLATED_FILE, -nr_file);>> +>>       } while (nr_scanned < max_scan);>> This is a non-trivial amount of extra stuff.  Do we really need it?> I thought so.This patch results form process fork bomb(ex, mstctl11 in LTP).Too many isolated patches are based on isolation counter. So, I think we need this until now.If we can solve the problem with different method, then we can drop this. -- Kind regards,Minchan Kim????{.n?+???????+%?????ݶ??w??{.n?+????{??G?????{ay?ʇڙ?,j??f???h?????????z_??(?階?ݢj"???m??????G????????????&???~???iO???z??v?^?m???? ????????I?