From mboxrd@z Thu Jan 1 00:00:00 1970 From: Christoph Hellwig Subject: Re: [PATCH v2] scsi: storvsc: Add support for FC rport. Date: Mon, 3 Apr 2017 23:53:31 -0700 Message-ID: <20170404065331.GA2626@infradead.org> References: <1491219195-3692-1-git-send-email-cavery@redhat.com> <20170403121718.GA13130@infradead.org> <58E24A9A.30200@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Content-Disposition: inline In-Reply-To: <58E24A9A.30200@redhat.com> Sender: linux-kernel-owner@vger.kernel.org To: Cathy Avery Cc: Christoph Hellwig , kys@microsoft.com, haiyangz@microsoft.com, jejb@linux.vnet.ibm.com, martin.petersen@oracle.com, stephen@networkplumber.org, dan.carpenter@oracle.com, devel@linuxdriverproject.org, linux-kernel@vger.kernel.org, linux-scsi@vger.kernel.org List-Id: linux-scsi@vger.kernel.org On Mon, Apr 03, 2017 at 09:14:02AM -0400, Cathy Avery wrote: > On 04/03/2017 08:17 AM, Christoph Hellwig wrote: > > > if (host->transportt == fc_transport_template) { > > > + struct fc_rport_identifiers ids = { > > > + .roles = FC_PORT_ROLE_FCP_TARGET, > > > + }; > > I don't think storvsc ever acts as FCP target. > > In order to implement the work around so that the scsi scan works indicating > FC_PORT_ROLE_FCP_TARGET as a role was necessary due to its test in > fc_scsi_scan_rport. The idea here is to avoid making any changes to the > fc_transport driver which was of some concern. Please add a FC_PORT_ROLE_FCP_DUMMY_INITIATOR role instead.