From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756973Ab0IGOX6 (ORCPT ); Tue, 7 Sep 2010 10:23:58 -0400 Received: from smtp109.prem.mail.ac4.yahoo.com ([76.13.13.92]:30793 "HELO smtp109.prem.mail.ac4.yahoo.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with SMTP id S1751152Ab0IGOXx (ORCPT ); Tue, 7 Sep 2010 10:23:53 -0400 X-Yahoo-SMTP: _Dag8S.swBC1p4FJKLCXbs8NQzyse1SYSgnAbY0- X-YMail-OSG: Hrc_3PYVM1lW4LsQMjDQqnEbxUMf3e.A595KKJ5hn0Og6M2 _e2ekFJjq70qHexQ7jOEn9k.4kpubAMBm.iZDv0dPC1_GInZTWliqyUlpVbf RgdCNTMpEeeKJbfQkvDS0H_x2HARbKTiNA6acWBvb62WqCi1124ML8E2Oont GzMpRjkUTSY9SmAD88.WhJEle_XGLJy_cUV9tkTxa0elpKdq5dkjKcrjn60g x X-Yahoo-Newman-Property: ymail-3 Date: Tue, 7 Sep 2010 09:23:48 -0500 (CDT) From: Christoph Lameter X-X-Sender: cl@router.home To: Dave Chinner cc: Wu Fengguang , Andrew Morton , Mel Gorman , Linux Kernel List , "linux-mm@kvack.org" , Rik van Riel , Johannes Weiner , Minchan Kim , KAMEZAWA Hiroyuki , KOSAKI Motohiro , David Rientjes Subject: Re: [PATCH 3/3] mm: page allocator: Drain per-cpu lists after direct reclaim allocation fails In-Reply-To: <20100906040243.GA7362@dastard> Message-ID: References: <20100903160026.564fdcc9.akpm@linux-foundation.org> <20100904022545.GD705@dastard> <20100903202101.f937b0bb.akpm@linux-foundation.org> <20100904075840.GE705@dastard> <20100904081414.GF705@dastard> <20100905015400.GA10714@localhost> <20100905021555.GG705@dastard> <20100905060539.GA17450@localhost> <20100905131447.GJ705@dastard> <20100905134554.GA7083@localhost> <20100906040243.GA7362@dastard> User-Agent: Alpine 2.00 (DEB 1167 2008-08-23) MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Mon, 6 Sep 2010, Dave Chinner wrote: > [ 596.628086] [] ? drain_all_pages+0x1c/0x20 > [ 596.628086] [] ? __alloc_pages_nodemask+0x42d/0x700 > [ 596.628086] [] ? kmem_getpages+0x62/0x160 > [ 596.628086] [] ? fallback_alloc+0x196/0x240 fallback_alloc() showing up here means that one page allocator call from SLAB has already failed. SLAB then did an expensive search through all object caches on all nodes to find some available object. There were no objects in queues at all therefore SLAB called the page allocator again (kmem_getpages()). As soon as memory is available (on any node or any cpu, they are all empty) SLAB will repopulate its queues(!).