From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753226AbYLCQPu (ORCPT ); Wed, 3 Dec 2008 11:15:50 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1751305AbYLCQPk (ORCPT ); Wed, 3 Dec 2008 11:15:40 -0500 Received: from fgwmail7.fujitsu.co.jp ([192.51.44.37]:39957 "EHLO fgwmail7.fujitsu.co.jp" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751005AbYLCQPj (ORCPT ); Wed, 3 Dec 2008 11:15:39 -0500 Message-ID: <17466.10.75.179.62.1228320932.squirrel@webmail-b.css.fujitsu.com> In-Reply-To: <20081203141931.GH17701@balbir.in.ibm.com> References: <20081201205810.1CCA.KOSAKI.MOTOHIRO@jp.fujitsu.com> <20081201211905.1CEB.KOSAKI.MOTOHIRO@jp.fujitsu.com> <20081202180525.2023892c.kamezawa.hiroyu@jp.fujitsu.com> <20081203141931.GH17701@balbir.in.ibm.com> Date: Thu, 4 Dec 2008 01:15:32 +0900 (JST) Subject: Re: [PATCH 11/11] memcg: show reclaim_stat From: "KAMEZAWA Hiroyuki" To: balbir@linux.vnet.ibm.com Cc: "KAMEZAWA Hiroyuki" , "KOSAKI Motohiro" , "LKML" , "linux-mm" , "Andrew Morton" , "Rik van Riel" User-Agent: SquirrelMail/1.4.3a X-Mailer: SquirrelMail/1.4.3a MIME-Version: 1.0 Content-Type: text/plain;charset=us-ascii Content-Transfer-Encoding: 8bit X-Priority: 3 (Normal) Importance: Normal Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Balbir Singh said: > * KAMEZAWA Hiroyuki [2008-12-02 > 18:05:25]: > >> On Mon, 1 Dec 2008 21:19:49 +0900 (JST) >> KOSAKI Motohiro wrote: >> >> > added following four field to memory.stat file. >> > >> > - recent_rotated_anon >> > - recent_rotated_file >> > - recent_scanned_anon >> > - recent_scanned_file >> > >> > it is useful for memcg reclaim debugging. >> > >> I'll put this under CONFIG_DEBUG_VM. >> > > I think they'll be useful even outside for tasks that need to take > decisions, it will be nice to see what sort of reclaim is going on. There are already pgin/pgout value. > I > would like to see them outside, there is no cost associated with them > and assuming we'll not change the LRU logic very frequently, we don't > need to be afraid of breaking ABI either :) > There are 2 reasons to put this under DEBUG 1. This is not exported as this value under /proc by global VM management. 2. Few people can explain what this really means. No documentation in Docunemtation/ directory. I can't add precise explanation by myself. As Kosaki wrote, this is for his debug, IMHO. If you want to show this, please add above two first. Thanks, -Kame