From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail138.messagelabs.com (mail138.messagelabs.com [216.82.249.35]) by kanga.kvack.org (Postfix) with ESMTP id B3F266B00F1 for ; Wed, 13 May 2009 07:28:37 -0400 (EDT) Date: Wed, 13 May 2009 13:26:40 +0200 From: Johannes Weiner Subject: Re: [PATCH 3/4] vmscan: zone_reclaim use may_swap Message-ID: <20090513112640.GC2254@cmpxchg.org> References: <20090513120155.5879.A69D9226@jp.fujitsu.com> <20090513120651.5882.A69D9226@jp.fujitsu.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20090513120651.5882.A69D9226@jp.fujitsu.com> Sender: owner-linux-mm@kvack.org To: KOSAKI Motohiro Cc: LKML , linux-mm , Andrew Morton , Rik van Riel , Christoph Lameter List-ID: On Wed, May 13, 2009 at 12:07:30PM +0900, KOSAKI Motohiro wrote: > Subject: [PATCH] vmscan: zone_reclaim use may_swap > > Documentation/sysctl/vm.txt says > > zone_reclaim_mode: > > Zone_reclaim_mode allows someone to set more or less aggressive approaches to > reclaim memory when a zone runs out of memory. If it is set to zero then no > zone reclaim occurs. Allocations will be satisfied from other zones / nodes > in the system. > > This is value ORed together of > > 1 = Zone reclaim on > 2 = Zone reclaim writes dirty pages out > 4 = Zone reclaim swaps pages > > > So, "(zone_reclaim_mode & RECLAIM_SWAP) == 0" mean we don't want to reclaim > swap-backed pages. not mapped file. > > Thus, may_swap is better than may_unmap. > > > Signed-off-by: KOSAKI Motohiro > Cc: Christoph Lameter > Cc: Rik van Riel Reviewed-by: Johannes Weiner -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@kvack.org. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: email@kvack.org