From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1763515AbZFKL7f (ORCPT ); Thu, 11 Jun 2009 07:59:35 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1753363AbZFKL71 (ORCPT ); Thu, 11 Jun 2009 07:59:27 -0400 Received: from mail-yx0-f203.google.com ([209.85.210.203]:35021 "EHLO mail-yx0-f203.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752547AbZFKL70 convert rfc822-to-8bit (ORCPT ); Thu, 11 Jun 2009 07:59:26 -0400 X-Greylist: delayed 2325 seconds by postgrey-1.27 at vger.kernel.org; Thu, 11 Jun 2009 07:59:26 EDT 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=XI9TF1ur0r18jLyYmNPsrZQarscCAePQtTjgRAFYbBRJZpHQpjbVEHYkrp04Com11m GKx6UrHEpyUuHkFVeMVeG9UMsdXta3NAA4pSUQsjX2Mv+dC5CxvOKzpdEzXWfvLnRjee uUbZBKhtG6SGMh6q+nu5Excu1ASC83iOgDt1E= MIME-Version: 1.0 In-Reply-To: <9d4a7c0691aa5e13247f694f2dfe55ad.squirrel@webmail-b.css.fujitsu.com> References: <20090611165535.cf46bf29.kamezawa.hiroyu@jp.fujitsu.com> <20090611170152.7a43b13b.kamezawa.hiroyu@jp.fujitsu.com> <20090611172249.6D3C.A69D9226@jp.fujitsu.com> <20090611173819.0f76e431.kamezawa.hiroyu@jp.fujitsu.com> <28c262360906110237u1f3d1877hae54a51575955549@mail.gmail.com> <9d4a7c0691aa5e13247f694f2dfe55ad.squirrel@webmail-b.css.fujitsu.com> Date: Thu, 11 Jun 2009 20:59:28 +0900 Message-ID: <28c262360906110459s923d7a6p4e555344e8bbd265@mail.gmail.com> Subject: Re: [PATCH 2/3] check unevictable flag in lumy reclaim v2 From: Minchan Kim To: KAMEZAWA Hiroyuki Cc: KOSAKI Motohiro , "linux-mm@kvack.org" , "linux-kernel@vger.kernel.org" , "nishimura@mxp.nes.nec.co.jp" , "balbir@linux.vnet.ibm.com" , "akpm@linux-foundation.org" , apw@canonical.com, riel@redhat.com, mel@csn.ul.ie, Lee Schermerhorn 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 2009/6/11 KAMEZAWA Hiroyuki : > Minchan Kim さん wrote: >> On Thu, Jun 11, 2009 at 5:38 PM, KAMEZAWA >> Hiroyuki wrote: >>> How about this ? >>> >>> From: KAMEZAWA Hiroyuki >>> >>> Lumpy reclaim check pages from their pfn. Then, it can find unevictable >>> pages >>> in its loop. >>> Abort lumpy reclaim when we find Unevictable page, we never get a lump >>> of pages for requested order. >>> >>> Changelog: v1->v2 >>> ?- rewrote commet. >>> >>> Signed-off-by: KAMEZAWA Hiroyuki >>> --- >>> ?mm/vmscan.c | ? ?9 +++++++++ >>> ?1 file changed, 9 insertions(+) >>> >>> Index: lumpy-reclaim-trial/mm/vmscan.c >>> =================================================================== >>> --- lumpy-reclaim-trial.orig/mm/vmscan.c >>> +++ lumpy-reclaim-trial/mm/vmscan.c >>> @@ -936,6 +936,15 @@ static unsigned long isolate_lru_pages(u >>> ? ? ? ? ? ? ? ? ? ? ? ?/* Check that we have not crossed a zone >>> boundary. */ >>> ? ? ? ? ? ? ? ? ? ? ? ?if (unlikely(page_zone_id(cursor_page) != >>> zone_id)) >>> ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?continue; >>> + ? ? ? ? ? ? ? ? ? ? ? /* >>> + ? ? ? ? ? ? ? ? ? ? ? ?* We tries to free all pages in this range to >>> create >>> + ? ? ? ? ? ? ? ? ? ? ? ?* a free large page. Then, if the range >>> includes a page >>> + ? ? ? ? ? ? ? ? ? ? ? ?* never be reclaimed, we have no reason to do >>> more. >>> + ? ? ? ? ? ? ? ? ? ? ? ?* PageUnevictable page is not a page which can >>> be >>> + ? ? ? ? ? ? ? ? ? ? ? ?* easily freed. Abort this scan now. >>> + ? ? ? ? ? ? ? ? ? ? ? ?*/ >>> + ? ? ? ? ? ? ? ? ? ? ? if (unlikely(PageUnevictable(cursor_page))) >>> + ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? break; >> >> __isolate_lru_pages already checked PageUnevictable to return error. >> I want to remove repeated check although it is trivial. >> >> By your patch, It seems to remove PageUnevictable check in >> __isolate_lru_pages. >> > yes. > >> But I know that. If we remove PageUnevictable check in >> __isolate_lru_pages, it can't go into BUG in non-lumpy case. ( I >> mentioned following as code) >> > In non-lumpy case, we'll never see Unevictable, maybe. I think so if it doesn't happen RAM failure. AFAIK, Unevictable check didn't related with RAM failure. > >>                 case -EBUSY: >>                         /* else it is being freed elsewhere */ >>                         list_move(&page->lru, src); >>                         continue; >> >>                 default: >>                         BUG(); >>                 } >> >> >> It means we can remove BUG in non-lumpy case and then add BUG into >> __isolate_lru_pages directly. >> >> If we can do it, we can remove unnecessary PageUnevictable check in >> __isolate_lru_page. >> > Hmm, but Unevicable check had tons of troubles at its implementation > and I don't want to do it at once. I think it's not a big problem. As comment said, the check's goal is to prevent in lumpy case. /* * When this function is being called for lumpy reclaim, we * initially look into all LRU pages, active, inactive and * unevictable; only give shrink_page_list evictable pages. */ if (PageUnevictable(page)) return ret; So I think we can remove this check. >> I am not sure this is right in case of memcg. >> > I think we don't see Unevictable in memcg's path if my memcg-lru code > works as designed. > > I'll postpone this patch for a while until my brain works well. If you have a concern about that, how about this ? (This code will be hunk since gmail webserver always mangle. Pz,forgive me) Also, we can CC original authors. --- a/mm/vmscan.c ++ b/mm/vmscan.c @@ -936,19 +936,20 @@ static unsigned long isolate_lru_pages(unsigned long nr_to_scan, /* Check that we have not crossed a zone boundary. */ if (unlikely(page_zone_id(cursor_page) != zone_id)) continue; - switch (__isolate_lru_page(cursor_page, mode, file)) { - case 0: + if (__isolate_lru_page(cursor_page, mode, file) == 0) { list_move(&cursor_page->lru, dst); nr_taken++; scan++; - break; - - case -EBUSY: - /* else it is being freed elsewhere */ - list_move(&cursor_page->lru, src); - default: - break; /* ! on LRU or wrong list */ } + else if (PageUnevictable(cursor_page)) + /* + * We tries to free all pages in this range to create + * a free large page. Then, if the range includes a page + * never be reclaimed, we have no reason to do more. + * PageUnevictable page is not a page which can be + * easily freed. Abort this scan now. + */ + break } } > Thanks, > -Kame > > -- Kinds regards, Minchan Kim