From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1761683AbZFKJhr (ORCPT ); Thu, 11 Jun 2009 05:37:47 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1760825AbZFKJhj (ORCPT ); Thu, 11 Jun 2009 05:37:39 -0400 Received: from mail-gx0-f214.google.com ([209.85.217.214]:53384 "EHLO mail-gx0-f214.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1757008AbZFKJhj convert rfc822-to-8bit (ORCPT ); Thu, 11 Jun 2009 05:37:39 -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=VhH7R+ZXMTCFg+m+L/xLY/8WsJCVB4MIeZGX9yiUqg3oUDqE4jlEhsRjzPNYjXF56r N7E9M3XaRjKJ51rr8o0uCW8pTSXc1cgLHRwmKUgqLT/PCTn2CTgQoQ3KfTS4rZQtFP0y nKHNuDH0hATlosId4li0qNct+1jRvEtDMMOik= MIME-Version: 1.0 In-Reply-To: <20090611173819.0f76e431.kamezawa.hiroyu@jp.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> Date: Thu, 11 Jun 2009 18:37:40 +0900 Message-ID: <28c262360906110237u1f3d1877hae54a51575955549@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 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 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. 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) 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. I am not sure this is right in case of memcg. -- Kinds regards, Minchan Kim