From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id ; Wed, 22 May 2002 18:45:14 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id ; Wed, 22 May 2002 18:45:12 -0400 Received: from e21.nc.us.ibm.com ([32.97.136.227]:7133 "EHLO e21.nc.us.ibm.com") by vger.kernel.org with ESMTP id ; Wed, 22 May 2002 18:45:09 -0400 Date: Wed, 22 May 2002 15:44:01 -0700 From: "Martin J. Bligh" To: Linus Torvalds cc: William Lee Irwin III , Alan Cox , "M. Edward Borasky" , linux-kernel@vger.kernel.org, andrea@suse.de, riel@surriel.com, akpm@zip.com.au Subject: Re: Have the 2.4 kernel memory management problems on large machines been fixed? Message-ID: <386750000.1022107441@flay> In-Reply-To: X-Mailer: Mulberry/2.1.2 (Linux/x86) MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org >> If we could get the apps (well, Oracle) to co-operate, we could just use >> clone ;-) Having this transparent for shmem segments would be really nice. > > The thing is, we won't get Oracle to rewrite a lot for a completely > threaded system. And clone does _not_ come with a way to share only parts > of the VM, and never will - that's fundamentally against the way "struct > mm_struct" works. We're actually playing with Oracle apps - I'm told they already run on threaded mode on NT ... I personally get the feeling that Oracle's commitment to Linux is distinctly half-hearted. The whole support matrix debacle was pretty indicative, IMHO. All personal opinion, I speaketh not for IBM. > Oracle is apparently already used to magic shmem-like things, so doing > that is probably acceptable to them. We can but try, but I still think some transparent magic would be implementable. M.