Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S262557AbUCJIgk (ORCPT ); Wed, 10 Mar 2004 03:36:40 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S262556AbUCJIgk (ORCPT ); Wed, 10 Mar 2004 03:36:40 -0500 Received: from mail-01.iinet.net.au ([203.59.3.33]:29313 "HELO mail.iinet.net.au") by vger.kernel.org with SMTP id S262549AbUCJIg3 (ORCPT ); Wed, 10 Mar 2004 03:36:29 -0500 Message-ID: <404ED388.5050905@cyberone.com.au> Date: Wed, 10 Mar 2004 19:36:24 +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> In-Reply-To: <404ECFE5.7040005@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: 1609 Lines: 49 Mike Fedyk wrote: > Andrew Morton wrote: > >> Mike Fedyk wrote: >> >>> Most of the previous 2.6 kernels I was running on these servers >>> would be lightly hitting swap by now. This definitely looks better >>> to me. >> >> >> >> It sounds worse to me. "Lightly hitting swap" is good. It gets rid >> of stuff, >> freeing up physical memory. > > > Andrew, it looks like you're right. This[1] server doesn't seem to be > hitting swap enough. But my other[2] file server is doing great with > it on the other hand (though, it hasn't swapped at all). > Just curious, what makes you say [1] isn't hitting swap enough and [2] is OK? The graphs are better now, by the way. Thank you. > Maybe a little tuning is in order? > > Any patches I should try? > Mainline doesn't put enough pressure on slab with highmem systems. This creates a lot more ZONE_NORMAL pressure and that causes swapping. Now with the 2.6 VM, you don't do any mapped memory scaning at all while you only have a small amount of memory pressure. This means that truely inactive mapped pages never get reclaimed. The patches you are using do not address this. My split active list patches should do so. Alternatively you can increase /proc/sys/vm/swappiness, but that isn't a complete solution, and might make things too swappy. It is a difficult beast to control. - 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/