From mboxrd@z Thu Jan 1 00:00:00 1970 Date: Fri, 2 May 2008 03:23:50 +0200 From: Nick Piggin Subject: Re: Warning on memory offline (and possible in usual migration?) Message-ID: <20080502012350.GF30768@wotan.suse.de> References: <20080423004804.GA14134@wotan.suse.de> <20080429162016.961aa59d.kamezawa.hiroyu@jp.fujitsu.com> <20080430065611.GH27652@wotan.suse.de> <20080430001249.c07ff5c8.akpm@linux-foundation.org> <20080430072620.GI27652@wotan.suse.de> <20080501014418.GB15179@wotan.suse.de> <20080502004445.GB30768@wotan.suse.de> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: Sender: owner-linux-mm@kvack.org Return-Path: To: Christoph Lameter Cc: Andrew Morton , KAMEZAWA Hiroyuki , "linux-mm@kvack.org" , GOTO List-ID: On Thu, May 01, 2008 at 06:07:43PM -0700, Christoph Lameter wrote: > On Fri, 2 May 2008, Nick Piggin wrote: > > > On Thu, May 01, 2008 at 12:25:54PM -0700, Christoph Lameter wrote: > > > On Thu, 1 May 2008, Nick Piggin wrote: > > > > > > > Yes if PageUptodate and the page is locked, then I don't believe > > > > any read IO should happen. > > > > > > Ok so page migration should check for that and not migrate a page that is > > > !Uptodate? > > > > Buffer migration seems to work OK now, why do you need to add the > > restriction? > > Because we have to protect against read I/O. We cannot migrate > a page that is under I/O and free the memory that is being written to by a > device. The buffer migration path protects against read IO on buffers. -- 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