Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751462AbZL1D5p (ORCPT ); Sun, 27 Dec 2009 22:57:45 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1751181AbZL1D5o (ORCPT ); Sun, 27 Dec 2009 22:57:44 -0500 Received: from e23smtp01.au.ibm.com ([202.81.31.143]:36746 "EHLO e23smtp01.au.ibm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751113AbZL1D5n (ORCPT ); Sun, 27 Dec 2009 22:57:43 -0500 Date: Mon, 28 Dec 2009 09:27:38 +0530 From: Balbir Singh To: Rik van Riel Cc: Minchan Kim , Andrew Morton , lkml , linux-mm , Hugh Dickins , KOSAKI Motohiro Subject: Re: [PATCH -mmotm-2009-12-10-17-19] Prevent churning of zero page in LRU list. Message-ID: <20091228035738.GH3601@balbir.in.ibm.com> Reply-To: balbir@linux.vnet.ibm.com References: <20091228115315.76b1ecd0.minchan.kim@barrios-desktop> <4B38246C.3020209@redhat.com> <20091228035639.GG3601@balbir.in.ibm.com> MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Disposition: inline In-Reply-To: <20091228035639.GG3601@balbir.in.ibm.com> User-Agent: Mutt/1.5.20 (2009-08-17) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1428 Lines: 40 * Balbir Singh [2009-12-28 09:26:39]: > * Rik van Riel [2009-12-27 22:22:20]: > > > On 12/27/2009 09:53 PM, Minchan Kim wrote: > > > > > >VM doesn't add zero page to LRU list. > > >It means zero page's churning in LRU list is pointless. > > > > > >As a matter of fact, zero page can't be promoted by mark_page_accessed > > >since it doesn't have PG_lru. > > > > > >This patch prevent unecessary mark_page_accessed call of zero page > > >alghouth caller want FOLL_TOUCH. > > > > > >Signed-off-by: Minchan Kim > > > > The code looks correct, but I wonder how frequently we run into > > the zero page in this code, vs. how much the added cost is of > > having this extra code in follow_page. > > > > What kind of problem were you running into that motivated you > > to write this patch? > > > > Frequent moving of zero page should ideally put it to the head of the > LRU list, leaving it untouched is likely to cause it to be scanned > often - no? Should this be moved to the unevictable list? > Sorry, I replied to wrong email, I should have been clearer that this question is for Minchan Kim. -- Balbir -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/