From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1757923AbZLFVBo (ORCPT ); Sun, 6 Dec 2009 16:01:44 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S932165AbZLFVBn (ORCPT ); Sun, 6 Dec 2009 16:01:43 -0500 Received: from mx1.redhat.com ([209.132.183.28]:16210 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1757743AbZLFVBn (ORCPT ); Sun, 6 Dec 2009 16:01:43 -0500 Message-ID: <4B1C1BB8.9080301@redhat.com> Date: Sun, 06 Dec 2009 16:01:44 -0500 From: Rik van Riel Organization: Red Hat, Inc User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.1.4pre) Gecko/20090922 Fedora/3.0-3.9.b4.fc12 Lightning/1.0pre Thunderbird/3.0b4 MIME-Version: 1.0 To: KOSAKI Motohiro CC: LKML , linux-mm , Andrea Arcangeli , Larry Woodman Subject: Re: [PATCH 6/7] wipe_page_reference return SWAP_AGAIN if VM pressulre is low and lock contention is detected. References: <20091204173233.5891.A69D9226@jp.fujitsu.com> <20091204174439.58A3.A69D9226@jp.fujitsu.com> In-Reply-To: <20091204174439.58A3.A69D9226@jp.fujitsu.com> Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 12/04/2009 03:45 AM, KOSAKI Motohiro wrote: > From 3fb2a585729a37e205c5ea42ac6c48d4a6c0a29c Mon Sep 17 00:00:00 2001 > From: KOSAKI Motohiro > Date: Fri, 4 Dec 2009 12:54:37 +0900 > Subject: [PATCH 6/7] wipe_page_reference return SWAP_AGAIN if VM pressulre is low and lock contention is detected. > > Larry Woodman reported AIM7 makes serious ptelock and anon_vma_lock > contention on current VM. because SplitLRU VM (since 2.6.28) remove > calc_reclaim_mapped() test, then shrink_active_list() always call > page_referenced() against mapped page although VM pressure is low. > Lightweight VM pressure is very common situation and it easily makes > ptelock contention with page fault. then, anon_vma_lock is holding > long time and it makes another lock contention. then, fork/exit > throughput decrease a lot. It looks good to me. Larry, does this patch series resolve your issue? Acked-by: Rik van Riel -- All rights reversed.