Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S261176AbTH2M5y (ORCPT ); Fri, 29 Aug 2003 08:57:54 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S261186AbTH2M5y (ORCPT ); Fri, 29 Aug 2003 08:57:54 -0400 Received: from shark.pro-futura.com ([161.58.178.219]:30910 "EHLO shark.pro-futura.com") by vger.kernel.org with ESMTP id S261176AbTH2M5u (ORCPT ); Fri, 29 Aug 2003 08:57:50 -0400 From: "Tvrtko A. =?iso-8859-2?q?Ur=B9ulin?=" To: linux-kernel@vger.kernel.org Subject: Pagecache going out of control (2.4) Date: Fri, 29 Aug 2003 14:59:10 +0200 User-Agent: KMail/1.5.1 MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-2" Content-Transfer-Encoding: 7bit Content-Disposition: inline Message-Id: <200308291459.10983.tvrtko@croadria.com> Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1165 Lines: 28 Hello everyone, For some time now, the situation with 2.4 kernels is the following: Try to gzip -d very large file (few times larger than physical RAM) in the background and do some other work. You will find out that your system gets very slow. All RAM is used up for pagecache, and even more, your application data is getting swapped out to make more room for pagecache! I think this is a bad behaviour, and would pretty much like to be able to control the maximum memory used for pagecache. Recently, I found a patch by indou.takao@jp.fujitsu.com which implements a pgcache-max sysctl for 2.5.x. I tried to backport it to 2.4.21 but so far with no success (many differences in VM system). I don't understand why not to give such a option to users, I know a lot of them irritated with current pagecache behaviour... options and configurability should be a good thing, no? Best regards, Tvrtko A. Ursulin - 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/