From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756715Ab0DBJNO (ORCPT ); Fri, 2 Apr 2010 05:13:14 -0400 Received: from fgwmail6.fujitsu.co.jp ([192.51.44.36]:33804 "EHLO fgwmail6.fujitsu.co.jp" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753014Ab0DBJNI convert rfc822-to-8bit (ORCPT ); Fri, 2 Apr 2010 05:13:08 -0400 X-SecurityPolicyCheck-FJ: OK by FujitsuOutboundMailChecker v1.3.1 From: KOSAKI Motohiro To: Andrew Morton Subject: Re: [PATCH]vmscan: handle underflow for get_scan_ratio Cc: kosaki.motohiro@jp.fujitsu.com, Wu Fengguang , "Li, Shaohua" , "linux-mm@kvack.org" , "linux-kernel@vger.kernel.org" In-Reply-To: <20100401151639.a030fb10.akpm@linux-foundation.org> References: <20100331145602.03A7.A69D9226@jp.fujitsu.com> <20100401151639.a030fb10.akpm@linux-foundation.org> Message-Id: <20100402180812.646D.A69D9226@jp.fujitsu.com> MIME-Version: 1.0 Content-Type: text/plain; charset="US-ASCII" Content-Transfer-Encoding: 8BIT X-Mailer: Becky! ver. 2.50.07 [ja] Date: Fri, 2 Apr 2010 18:13:04 +0900 (JST) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org > > Yeah, I don't want ignore .33-stable too. if I can't find the root cause > > in 2-3 days, I'll revert guilty patch anyway. > > > > It's a good idea to avoid fixing a bug one-way-in-stable, > other-way-in-mainline. Because then we have new code in both trees > which is different. And the -stable guys sensibly like to see code get > a bit of a shakedown in mainline before backporting it. > > So it would be better to merge the "simple" patch into mainline, tagged > for -stable backporting. Then we can later implement the larger fix in > mainline, perhaps starting by reverting the "simple" fix. .....ok. I don't have to prevent your code maintainship. although I still think we need to fix the issue completely. =================================================================== >>From 52358cbccdfe94e0381974cd6e937bcc6b1c608b Mon Sep 17 00:00:00 2001 From: KOSAKI Motohiro Date: Fri, 2 Apr 2010 17:13:48 +0900 Subject: [PATCH] Revert "vmscan: get_scan_ratio() cleanup" Shaohua Li reported his tmpfs streaming I/O test can lead to make oom. The test uses a 6G tmpfs in a system with 3G memory. In the tmpfs, there are 6 copies of kernel source and the test does kbuild for each copy. His investigation shows the test has a lot of rotated anon pages and quite few file pages, so get_scan_ratio calculates percent[0] (i.e. scanning percent for anon) to be zero. Actually the percent[0] shoule be a big value, but our calculation round it to zero. Although before commit 84b18490, we have the same sick too. but the old logic can rescue percent[0]==0 case only when priority==0. It had hided the real issue. I didn't think merely streaming io can makes percent[0]==0 && priority==0 situation. but I was wrong. So, definitely we have to fix such tmpfs streaming io issue. but anyway I revert the regression commit at first. This reverts commit 84b18490d1f1bc7ed5095c929f78bc002eb70f26. Reported-by: Shaohua Li Signed-off-by: KOSAKI Motohiro --- mm/vmscan.c | 23 +++++++++-------------- 1 files changed, 9 insertions(+), 14 deletions(-) diff --git a/mm/vmscan.c b/mm/vmscan.c index 79c8098..cb3947e 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -1535,13 +1535,6 @@ static void get_scan_ratio(struct zone *zone, struct scan_control *sc, unsigned long ap, fp; struct zone_reclaim_stat *reclaim_stat = get_reclaim_stat(zone, sc); - /* If we have no swap space, do not bother scanning anon pages. */ - if (!sc->may_swap || (nr_swap_pages <= 0)) { - percent[0] = 0; - percent[1] = 100; - return; - } - anon = zone_nr_lru_pages(zone, sc, LRU_ACTIVE_ANON) + zone_nr_lru_pages(zone, sc, LRU_INACTIVE_ANON); file = zone_nr_lru_pages(zone, sc, LRU_ACTIVE_FILE) + @@ -1639,20 +1632,22 @@ static void shrink_zone(int priority, struct zone *zone, unsigned long nr_reclaimed = sc->nr_reclaimed; unsigned long nr_to_reclaim = sc->nr_to_reclaim; struct zone_reclaim_stat *reclaim_stat = get_reclaim_stat(zone, sc); + int noswap = 0; - get_scan_ratio(zone, sc, percent); + /* If we have no swap space, do not bother scanning anon pages. */ + if (!sc->may_swap || (nr_swap_pages <= 0)) { + noswap = 1; + percent[0] = 0; + percent[1] = 100; + } else + get_scan_ratio(zone, sc, percent); for_each_evictable_lru(l) { int file = is_file_lru(l); unsigned long scan; - if (percent[file] == 0) { - nr[l] = 0; - continue; - } - scan = zone_nr_lru_pages(zone, sc, l); - if (priority) { + if (priority || noswap) { scan >>= priority; scan = (scan * percent[file]) / 100; } -- 1.6.5.2