scsi: lpfc: Fix PLOGI S_ID of 0 on pt2pt config
authorJames Smart <jsmart2021@gmail.com>
Mon, 4 Jan 2021 18:02:26 +0000 (10:02 -0800)
committerMartin K. Petersen <martin.petersen@oracle.com>
Fri, 8 Jan 2021 04:02:35 +0000 (23:02 -0500)
Under some pt2pt situations, the other end of the link may issue a LOGO
after successfully completing PLOGI and assigning addresses to the port.
Thus the driver may attempt a new PLOGI to re-create the login, but the
LOGO handling cleared the address back to 0. Once this happens, the other
end, which may be address 0, gets all confused and this cannot be resolved
without an administrative action to bounce the link.

Fix by assuming that address assignment only occurs on the 1st PLOGI after
link up, and regardless of login state, the address assignment sticks.  The
FC standards aren't particularly clear in this situation (it only describes
initial PLOGI), but there is nothing that contradicts this and behaviors on
the devices tested appears to conform to the understanding.

Thus, don't reset the port address to 0 as part of LOGO handling. Port
addresses will only reset on link down.

Link: https://lore.kernel.org/r/20210104180240.46824-2-jsmart2021@gmail.com
Co-developed-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <jsmart2021@gmail.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
drivers/scsi/lpfc/lpfc_els.c

index 96c087b..e099caa 100644 (file)
@@ -2815,7 +2815,6 @@ lpfc_cmpl_els_logo(struct lpfc_hba *phba, struct lpfc_iocbq *cmdiocb,
        struct lpfc_nodelist *ndlp = (struct lpfc_nodelist *) cmdiocb->context1;
        struct lpfc_vport *vport = ndlp->vport;
        IOCB_t *irsp;
-       struct lpfcMboxq *mbox;
        unsigned long flags;
        uint32_t skip_recovery = 0;
 
@@ -2884,31 +2883,11 @@ out:
        lpfc_els_free_iocb(phba, cmdiocb);
        lpfc_nlp_put(ndlp);
 
-       /* If we are in pt2pt mode, we could rcv new S_ID on PLOGI */
-       if ((vport->fc_flag & FC_PT2PT) &&
-               !(vport->fc_flag & FC_PT2PT_PLOGI)) {
-               phba->pport->fc_myDID = 0;
-
-               if ((vport->cfg_enable_fc4_type == LPFC_ENABLE_BOTH) ||
-                   (vport->cfg_enable_fc4_type == LPFC_ENABLE_NVME)) {
-                       if (phba->nvmet_support)
-                               lpfc_nvmet_update_targetport(phba);
-                       else
-                               lpfc_nvme_update_localport(phba->pport);
-               }
-
-               mbox = mempool_alloc(phba->mbox_mem_pool, GFP_KERNEL);
-               if (mbox) {
-                       lpfc_config_link(phba, mbox);
-                       mbox->mbox_cmpl = lpfc_sli_def_mbox_cmpl;
-                       mbox->vport = vport;
-                       if (lpfc_sli_issue_mbox(phba, mbox, MBX_NOWAIT) ==
-                               MBX_NOT_FINISHED) {
-                               mempool_free(mbox, phba->mbox_mem_pool);
-                               skip_recovery = 1;
-                       }
-               }
-       }
+       /* At this point, the LOGO processing is complete. NOTE: For a
+        * pt2pt topology, we are assuming the NPortID will only change
+        * on link up processing. For a LOGO / PLOGI initiated by the
+        * Initiator, we are assuming the NPortID is not going to change.
+        */
 
        /*
         * If the node is a target, the handling attempts to recover the port.