From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S261607AbUBUTLc (ORCPT ); Sat, 21 Feb 2004 14:11:32 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S261603AbUBUTLb (ORCPT ); Sat, 21 Feb 2004 14:11:31 -0500 Received: from mx1.redhat.com ([66.187.233.31]:44979 "EHLO mx1.redhat.com") by vger.kernel.org with ESMTP id S261604AbUBUTLa (ORCPT ); Sat, 21 Feb 2004 14:11:30 -0500 From: Daniel Phillips To: Lars Marowsky-Bree Subject: Re: GFS requirements (was: Non-GPL export of invalidate_mmap_range) Date: Sat, 21 Feb 2004 14:09:13 -0500 User-Agent: KMail/1.5.4 Cc: linux-kernel References: <20040216190927.GA2969@us.ibm.com> <200402202216.09908.phillips@arcor.de> <20040221141724.GH6280@marowsky-bree.de> In-Reply-To: <20040221141724.GH6280@marowsky-bree.de> MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Content-Disposition: inline Message-Id: <200402211409.13203.phillips@arcor.de> Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Saturday 21 February 2004 09:17, Lars Marowsky-Bree wrote: > On 2004-02-20T22:16:09, I said: > > Each DFS is free to implement its own infrastructure, possibly involving > > kernel extensions. > > Yes. Though I do reserve the right to find this highly silly, that we > might end up with multiple hooks for clustering infrastructure in the > kernel... But the one true clustering infrastructure hasn't been developed yet. The upcoming crop of designs need a chance to evolve before a framework is cast in stone. Perhaps we will eventually end up with a generic harness, something like a vfs for cluster infrastructure, but in my opinion, we're far from being able to define that sensibly now. It's better to implement exactly what a given DFS needs for the time being. > So, how does OpenGFS/GFS achieve the communication? How does it interact > with the infrastructure (which, I infere from your above comments, is > meant to reside in user-space)? It's done both ways, actually. No new kernel hooks are used in either case. Regards, Daniel