From mboxrd@z Thu Jan 1 00:00:00 1970 From: James Bottomley Subject: RE: [PATCH] exposing sdev_target in sysfs Date: 07 Sep 2004 10:11:28 -0400 Sender: linux-scsi-owner@vger.kernel.org Message-ID: <1094566289.2069.16.camel@mulgrave> References: <0B1E13B586976742A7599D71A6AC733C02D946@xbl3.ma.emulex.com> Mime-Version: 1.0 Content-Type: text/plain Content-Transfer-Encoding: 7bit Return-path: Received: from stat16.steeleye.com ([209.192.50.48]:45703 "EHLO hancock.sc.steeleye.com") by vger.kernel.org with ESMTP id S268074AbUIGOLa (ORCPT ); Tue, 7 Sep 2004 10:11:30 -0400 In-Reply-To: <0B1E13B586976742A7599D71A6AC733C02D946@xbl3.ma.emulex.com> List-Id: linux-scsi@vger.kernel.org To: James.Smart@Emulex.Com Cc: SCSI Mailing List On Tue, 2004-09-07 at 10:05, James.Smart@Emulex.Com wrote: > Did you give any consideration to the fact that FC would like to see all FC nodes - not just those that are FCP targets ? I assume we could augment this patch to deal with the transport allocating the target (or node) structures rather than it being a byproduct of lun attachment. I would also assume we'd still allow lun discovery to create targets as well for compatibility w/ LLDDs that don't call out to the transport to alloc targets. Actually, no. I'm playing purely with SPI here. I assumed people with FC stuff would be able to fix up the FC class after I'm done. By nodes, I assume you mean initiator or target? In which case, just exposing them and attaching a processor device would seem to be appropriate. James