Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S262796AbUCKAZU (ORCPT ); Wed, 10 Mar 2004 19:25:20 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S262901AbUCKAZU (ORCPT ); Wed, 10 Mar 2004 19:25:20 -0500 Received: from mail-09.iinet.net.au ([203.59.3.41]:61128 "HELO mail.iinet.net.au") by vger.kernel.org with SMTP id S262796AbUCKAZJ (ORCPT ); Wed, 10 Mar 2004 19:25:09 -0500 Message-ID: <404FB1DF.3070605@cyberone.com.au> Date: Thu, 11 Mar 2004 11:25:03 +1100 From: Nick Piggin User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.6) Gecko/20040122 Debian/1.6-1 X-Accept-Language: en MIME-Version: 1.0 To: Mike Fedyk CC: Andrew Morton , linux-kernel@vger.kernel.org Subject: Re: VM patches in 2.6.4-rc1-mm2 References: <20040302201536.52c4e467.akpm@osdl.org> <40469E50.6090401@matchmail.com> <20040303193025.68a16dc4.akpm@osdl.org> <404ECFE5.7040005@matchmail.com> <404ED388.5050905@cyberone.com.au> <404F651B.1030202@matchmail.com> <404FAA04.1020300@cyberone.com.au> <404FAFFE.9010403@matchmail.com> In-Reply-To: <404FAFFE.9010403@matchmail.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1747 Lines: 72 Mike Fedyk wrote: > Nick Piggin wrote: > >> >> >> Mike Fedyk wrote: >> >>> Nick Piggin wrote: >>> >>>> Mainline doesn't put enough pressure on slab with highmem systems. >>>> This >>>> creates a lot more ZONE_NORMAL pressure and that causes swapping. >>>> >>> >>> Yep, saw that. Especially with 128MB Highmem (eg, 1G RAM) >>> >>>> Now with the 2.6 VM, you don't do any mapped memory scaning at all >>> >>> >>> >>> >>> You mean 2.6-mm? >>> >> >> Yes, either mm or linus. >> > > Have there been any VM patches merged into mainline? Or are you > saying that the imbalance in mainline would be enough to overcome to > lack of scanning of mapped pages? > There have been no VM patches merged into mainline. I just mean that neither mm or mainline does any mapped memory scanning when memory pressure is low. >> >> If you get a lot of pressure at one time it should push out your >> inactive mapped pages. Will get most of the really inactive ones, >> but it won't help pages becoming inactive in future. >> > > Ok, I see. This might be happening, since it is steadily getting more > into swap. > For the one that is swapping, yes this would be happening. >> >> Hasn't looked at it much. Probably not until some of the more basic >> VM patches can get merged into -linus. > > > Yes, I wonder if the VM patches helped -mm in the reaim tests... > > Let's get the fsfaz (free slab for all zones) into mainline asap! :-D > Well all the ones in -mm now are probably right to go to 2.6.5 I hope. - 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/