From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1759705AbXEIWQu (ORCPT ); Wed, 9 May 2007 18:16:50 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1759727AbXEIWQY (ORCPT ); Wed, 9 May 2007 18:16:24 -0400 Received: from mx1.redhat.com ([66.187.233.31]:34830 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1759592AbXEIWQW convert rfc822-to-8bit (ORCPT ); Wed, 9 May 2007 18:16:22 -0400 Message-ID: <4642477A.3000801@redhat.com> Date: Wed, 09 May 2007 18:13:14 -0400 From: =?ISO-8859-1?Q?Kristian_H=F8gsberg?= User-Agent: Thunderbird 1.5.0.10 (X11/20070302) MIME-Version: 1.0 To: Stefan Richter CC: Christoph Hellwig , linux-kernel@vger.kernel.org, Linus Torvalds , Andrew Morton , linux1394-devel Subject: Re: [PATCH 5/6] firewire: SBP-2 highlevel driver References: <4637A29F.6070302@redhat.com> <20070502090007.GA28174@infradead.org> <20070502194408.GD1248@infradead.org> <464237A9.6060301@redhat.com> <464243C8.8090503@s5r6.in-berlin.de> In-Reply-To: <464243C8.8090503@s5r6.in-berlin.de> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 8BIT Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Stefan Richter wrote: > Kristian Høgsberg wrote: >> I was trying to be clever and only allocate the host once the device had >> been discovered and initialized. I have now changed the code to just >> allocate the host up front and use the hostdata mechanism for the >> sbp2_device struct, which also addresses the host life cycle comments >> below. > > I have doubts. IMO the previous code is 100% correct as long as 1 SBP-2 > target LU maps to 1 Scsi_Host. > > The lifetime of the Scsi_Host would only be longer than that of the LU > if all LUs (or all LUs at the same initiator port) would be added beneath > the same instance of Scsi_Host. Then the lifetime of the Scsi_Host would > be that of the fw-sbp2 driver, or that of fw-sbp2's representation of a > FireWire bus. In the patch, the sbp2_device is now allocated with the scsi_host and is the hostdata part of the host struct. This mean we have to add all the LUs from the unit directory corresponding to the sbp2_device struct to that host. Is that a problem? I think we had this discussion before, but I still don't understand why this approach isn't feasible. Kristian