Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1759536AbZFIIur (ORCPT ); Tue, 9 Jun 2009 04:50:47 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1754564AbZFIIuk (ORCPT ); Tue, 9 Jun 2009 04:50:40 -0400 Received: from gir.skynet.ie ([193.1.99.77]:46342 "EHLO gir.skynet.ie" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752668AbZFIIuj (ORCPT ); Tue, 9 Jun 2009 04:50:39 -0400 Date: Tue, 9 Jun 2009 09:50:38 +0100 From: Mel Gorman To: Wu Fengguang Cc: KOSAKI Motohiro , Rik van Riel , Christoph Lameter , "Zhang, Yanmin" , "linuxram@us.ibm.com" , linux-mm , LKML Subject: Re: [PATCH 3/3] Do not unconditionally treat zones that fail zone_reclaim() as full Message-ID: <20090609085037.GI18380@csn.ul.ie> References: <1244466090-10711-1-git-send-email-mel@csn.ul.ie> <1244466090-10711-4-git-send-email-mel@csn.ul.ie> <20090609031119.GB7875@localhost> MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-15 Content-Disposition: inline In-Reply-To: <20090609031119.GB7875@localhost> User-Agent: Mutt/1.5.17+20080114 (2008-01-14) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 2376 Lines: 56 On Tue, Jun 09, 2009 at 11:11:19AM +0800, Wu Fengguang wrote: > On Mon, Jun 08, 2009 at 09:01:30PM +0800, Mel Gorman wrote: > > On NUMA machines, the administrator can configure zone_reclaim_mode that > > is a more targetted form of direct reclaim. On machines with large NUMA > > distances for example, a zone_reclaim_mode defaults to 1 meaning that clean > > unmapped pages will be reclaimed if the zone watermarks are not being > > met. The problem is that zone_reclaim() failing at all means the zone > > gets marked full. > > > > This can cause situations where a zone is usable, but is being skipped > > because it has been considered full. Take a situation where a large tmpfs > > mount is occuping a large percentage of memory overall. The pages do not > > get cleaned or reclaimed by zone_reclaim(), but the zone gets marked full > > and the zonelist cache considers them not worth trying in the future. > > > > This patch makes zone_reclaim() return more fine-grained information about > > what occured when zone_reclaim() failued. The zone only gets marked full if > > it really is unreclaimable. If it's a case that the scan did not occur or > > if enough pages were not reclaimed with the limited reclaim_mode, then the > > zone is simply skipped. > > > > There is a side-effect to this patch. Currently, if zone_reclaim() > > successfully reclaimed SWAP_CLUSTER_MAX, an allocation attempt would > > go ahead. With this patch applied, zone watermarks are rechecked after > > zone_reclaim() does some work. > > > > Signed-off-by: Mel Gorman > > Thanks for making the code a lot more readable :) > > Reviewed-by: Wu Fengguang > Thanks. > > /* > > * Do not scan if the allocation should not be delayed. > > */ > > if (!(gfp_mask & __GFP_WAIT) || (current->flags & PF_MEMALLOC)) > > - return 0; > > + return ZONE_RECLAIM_NOSCAN; > > Why not kill the extra tab? > Why not indeed. Tab is now killed. -- Mel Gorman Part-time Phd Student Linux Technology Center University of Limerick IBM Dublin Software Lab -- 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/