Hi Bryant, (Btw, your response didn't make it to target-devel) On Thu, 2016-03-10 at 12:21 -0600, Bryant G Ly wrote: > Hello Nick, > > "Nicholas A. Bellinger" <nab@xxxxxxxxxxxxxxx> wrote on 03/07/2016 > 07:39:34 PM: > > > From: "Nicholas A. Bellinger" <nab@xxxxxxxxxxxxxxx> > > To: Bryant G Ly/Rochester/IBM@IBMUS > > Cc: target-devel@xxxxxxxxxxxxxxx > > Date: 03/07/2016 07:39 PM > > Subject: Re: Reviving Ibmvscsi target/Questions > > > > Hey Bryant, > > > > Any updates wrt getting your WIP code posted in a public repo..? > > > > Yes, it finally has been approved for me to post code on a public > github. > The link is: github.com/powervm/ibmvscsis > Great. I noticed https://github.com/powervm/ibmvscsis is stand-lone files, and not a clone of upstream linux.git. So you'll want to go ahead and create a fork from: https://github.com/torvalds/linux and then use target-pending/for-next (eg: the target code that will become v4.6-rc1) as a starting point for further ibmvscsis development. Also, please let me know if you have any questions wrt to git. > > Are there any code items blocking you from making forward > progress..? > > > Nothing blocking, but I'd like to hear your thoughts on porting libsrp > to use target. > So after a initial grok on your WIP code, here is a few quick comments: https://github.com/powervm/ibmvscsis/blob/master/drivers/scsi/ibmvscsi/ibmvscsis.c#L406 transport_init_session_tags() should be passing TARGET_PROT_NORMAL, as AFAIK IBM vscsi doesn't support T10 protection information (PI). https://github.com/powervm/ibmvscsis/blob/master/drivers/scsi/ibmvscsi/ibmvscsis.c#L424 You can drop cmd->sgl, cmd->upages and cmd->prot_sgl. Both are related to vhost-scsi, and aren't used by ibmvscsis. https://github.com/powervm/ibmvscsis/blob/master/drivers/scsi/ibmvscsi/ibmvscsis.c#L630 Just use SCSI_PROTOCOL_SRP by default, and drop the extra tport->tport_proto_id related stuff. https://github.com/powervm/ibmvscsis/blob/master/drivers/scsi/ibmvscsi/ibmvscsis.c#L1109 For tcm_queuecommand() dispatch code, it should be using modern target_submit_cmd() for dispatching incoming SCSI CDBs + payloads into target-core. https://github.com/powervm/ibmvscsis/blob/master/drivers/scsi/ibmvscsi/ibmvscsis.c#L1194 For the original internal ibmvscsis INQUIRY + MODE_SENSE, and REPORT_LUNS emulation, it would be useful to see if we can (at least) make REPORT_LUNs use common code. This could be done easily in spc_emulate_report_luns(), with a ibmvscsis provided callback via target_core_fabric_ops using the existing make_lun() encoding format. https://github.com/powervm/ibmvscsis/blob/master/drivers/scsi/ibmvscsi/ibmvscsis.c#L1980 Since ibmviscsis uses a single internal se_session (eg: SCSI I_T Nexus) per /sys/kernel/config/target/ibmvscsis/$WWN/$TPGT/ endpoint, there is no need for an extra non-zero .node_acl_size. So let's get your code into a proper linux.git based repo, and we can continue the discussion to get up and running from there. -- To unsubscribe from this list: send the line "unsubscribe target-devel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
