From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S262386AbUBXSa7 (ORCPT ); Tue, 24 Feb 2004 13:30:59 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S262363AbUBXS3K (ORCPT ); Tue, 24 Feb 2004 13:29:10 -0500 Received: from mx1.redhat.com ([66.187.233.31]:54912 "EHLO mx1.redhat.com") by vger.kernel.org with ESMTP id S262396AbUBXS2Q (ORCPT ); Tue, 24 Feb 2004 13:28:16 -0500 From: Daniel Phillips To: Lars Marowsky-Bree Subject: Re: GFS requirements (was: Non-GPL export of invalidate_mmap_range) Date: Tue, 24 Feb 2004 13:26:04 -0500 User-Agent: KMail/1.5.4 Cc: linux-kernel References: <20040216190927.GA2969@us.ibm.com> <200402211409.13203.phillips@arcor.de> <20040222103731.GA19437@marowsky-bree.de> In-Reply-To: <20040222103731.GA19437@marowsky-bree.de> MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Content-Disposition: inline Message-Id: <200402241326.05698.phillips@arcor.de> Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Sunday 22 February 2004 05:37, Lars Marowsky-Bree wrote: > > > 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. > > That doesn't answer my question how you are doing the user-space / > kernel communication ;-) Hi Lars, OpenGFS uses sockets, see: http://cvs.sourceforge.net/viewcvs.py/opengfs/opengfs/src/locking/modules/memexp/moduleops.c?view=markup > And will the user-space infrastructure to go with GFS be open source > too? Questions over questions. Yes, it's all to be opened up. If you don't mind, I'd prefer to hold off discussing the infrastructure details until the code lands. Regards, Daniel