scsi: lpfc: Fix loss of remote port after devloss due to lack of RPIs

In tests with remote ports contantly logging out/logging coupled with
occassional local link bounce, if a remote port is disocnnected for longer
than devloss_tmo and then subsequently reconnected, eventually the test
will fail to login with the remote port and remote port connectivity is
lost.

When devloss_tmo expires, the driver does not free the node struct until
the port or npiv instances is being deleted. The node is left allocated but
the state set to UNUSED. If the node was in the process of logging in when
the local link drop occurred, meaning the RPI was allocated for the node in
order to send the ELS, but not yet registered which comes after successful
login, the node is moved to the NPR state, and if devloss expires, to
UNUSED state.  If the remote port comes back, the node associated with it
is restarted and this path happens to allocate a new RPI and overwrites the
prior RPI value. In the cases where the port was logged in and loggs out,
the path did release the RPI but did not set the node rpi value.  In the
cases where the remote port never finished logging in, the path never did
the call to release the rpi. In this latter case, when the node is
subsequently restore, the new rpi allocation overwrites the rpi that was
not released, and the rpi is now leaked.  Eventually the port will run out
of RPI resources to log into new remote ports.

Fix by following changes:

 - When an rpi is released, do so under locks and ensure the node rpi value
   is set to a non-allocated value (LPFC_RPI_ALLOC_ERROR).  Note:
   refactored to a small service routine to avoid indentation issues.

 - When re-enabling a node, check the rpi value to determine if a new
   allocation is necessary. If already set, use the prior rpi.

Enhanced logging to help in the future.

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>

3 files changed, 56 insertions, 26 deletions

diff --git a/drivers/scsi/lpfc/lpfc_ct.c b/drivers/scsi/lpfc/lpfc_ct.c
index 7649903d4134..c52d5edf4d44 100644
--- a/drivers/scsi/lpfc/lpfc_ct.c
+++ b/drivers/scsi/lpfc/lpfc_ct.c
@@ -462,6 +462,7 @@ lpfc_prep_node_fc4type(struct lpfc_vport *vport, uint32_t Did, uint8_t fc4_type)
 	struct lpfc_nodelist *ndlp;
 
 	if ((vport->port_type != LPFC_NPIV_PORT) ||
+	    (fc4_type == FC_TYPE_FCP) ||
 	    !(vport->ct_flags & FC_CT_RFF_ID) || !vport->cfg_restrict_login) {
 
 		ndlp = lpfc_setup_disc_node(vport, Did);
@@ -501,9 +502,9 @@ lpfc_prep_node_fc4type(struct lpfc_vport *vport, uint32_t Did, uint8_t fc4_type)
 
 			lpfc_printf_vlog(vport, KERN_INFO, LOG_DISCOVERY,
 					 "0239 Skip x%06x NameServer Rsp "
-					 "Data: x%x x%x\n", Did,
-					 vport->fc_flag,
-					 vport->fc_rscn_id_cnt);
+					 "Data: x%x x%x %p\n",
+					 Did, vport->fc_flag,
+					 vport->fc_rscn_id_cnt, ndlp);
 		}
 	} else {
 		if (!(vport->fc_flag & FC_RSCN_MODE) ||
diff --git a/drivers/scsi/lpfc/lpfc_hbadisc.c b/drivers/scsi/lpfc/lpfc_hbadisc.c
index c413f8c51aa8..22b6fd08e7f5 100644
--- a/drivers/scsi/lpfc/lpfc_hbadisc.c
+++ b/drivers/scsi/lpfc/lpfc_hbadisc.c
@@ -4480,9 +4480,21 @@ lpfc_enable_node(struct lpfc_vport *vport, struct lpfc_nodelist *ndlp,
 		return NULL;
 
 	if (phba->sli_rev == LPFC_SLI_REV4) {
-		rpi = lpfc_sli4_alloc_rpi(vport->phba);
-		if (rpi == LPFC_RPI_ALLOC_ERROR)
+		if (ndlp->nlp_rpi == LPFC_RPI_ALLOC_ERROR)
+			rpi = lpfc_sli4_alloc_rpi(vport->phba);
+		else
+			rpi = ndlp->nlp_rpi;
+
+		if (rpi == LPFC_RPI_ALLOC_ERROR) {
+			lpfc_printf_vlog(vport, KERN_WARNING, LOG_NODE,
+					 "0359 %s: ndlp:x%px "
+					 "usgmap:x%x refcnt:%d FAILED RPI "
+					 " ALLOC\n",
+					 __func__,
+					 (void *)ndlp, ndlp->nlp_usg_map,
+					 kref_read(&ndlp->kref));
 			return NULL;
+		}
 	}
 
 	spin_lock_irqsave(&phba->ndlp_lock, flags);
@@ -4541,6 +4553,14 @@ lpfc_enable_node(struct lpfc_vport *vport, struct lpfc_nodelist *ndlp,
 
 	if (state != NLP_STE_UNUSED_NODE)
 		lpfc_nlp_set_state(vport, ndlp, state);
+	else
+		lpfc_printf_vlog(vport, KERN_INFO, LOG_NODE,
+				 "0013 rpi:%x DID:%x flg:%x refcnt:%d "
+				 "map:%x x%px STATE=UNUSED\n",
+				 ndlp->nlp_rpi, ndlp->nlp_DID,
+				 ndlp->nlp_flag,
+				 kref_read(&ndlp->kref),
+				 ndlp->nlp_usg_map, ndlp);
 
 	lpfc_debugfs_disc_trc(vport, LPFC_DISC_TRC_NODE,
 		"node enable:       did:x%x",
@@ -5249,15 +5269,15 @@ __lpfc_findnode_did(struct lpfc_vport *vport, uint32_t did)
 
 	list_for_each_entry(ndlp, &vport->fc_nodes, nlp_listp) {
 		if (lpfc_matchdid(vport, ndlp, did)) {
-			data1 = (((uint32_t) ndlp->nlp_state << 24) |
-				 ((uint32_t) ndlp->nlp_xri << 16) |
-				 ((uint32_t) ndlp->nlp_type << 8) |
-				 ((uint32_t) ndlp->nlp_rpi & 0xff));
+			data1 = (((uint32_t)ndlp->nlp_state << 24) |
+				 ((uint32_t)ndlp->nlp_xri << 16) |
+				 ((uint32_t)ndlp->nlp_type << 8) |
+				 ((uint32_t)ndlp->nlp_usg_map & 0xff));
 			lpfc_printf_vlog(vport, KERN_INFO, LOG_NODE,
 					 "0929 FIND node DID "
-					 "Data: x%p x%x x%x x%x %p\n",
+					 "Data: x%px x%x x%x x%x x%x x%px\n",
 					 ndlp, ndlp->nlp_DID,
-					 ndlp->nlp_flag, data1,
+					 ndlp->nlp_flag, data1, ndlp->nlp_rpi,
 					 ndlp->active_rrqs_xri_bitmap);
 			return ndlp;
 		}
@@ -5342,8 +5362,11 @@ lpfc_setup_disc_node(struct lpfc_vport *vport, uint32_t did)
 		if (vport->phba->nvmet_support)
 			return NULL;
 		ndlp = lpfc_enable_node(vport, ndlp, NLP_STE_NPR_NODE);
-		if (!ndlp)
+		if (!ndlp) {
+			lpfc_printf_vlog(vport, KERN_WARNING, LOG_SLI,
+					 "0014 Could not enable ndlp\n");
 			return NULL;
+		}
 		spin_lock_irq(shost->host_lock);
 		ndlp->nlp_flag |= NLP_NPR_2B_DISC;
 		spin_unlock_irq(shost->host_lock);
diff --git a/drivers/scsi/lpfc/lpfc_sli.c b/drivers/scsi/lpfc/lpfc_sli.c
index 7c4bca47496a..8449f7f9cb64 100644
--- a/drivers/scsi/lpfc/lpfc_sli.c
+++ b/drivers/scsi/lpfc/lpfc_sli.c
@@ -2426,6 +2426,20 @@ lpfc_sli_wake_mbox_wait(struct lpfc_hba *phba, LPFC_MBOXQ_t *pmboxq)
 	return;
 }
 
+static void
+__lpfc_sli_rpi_release(struct lpfc_vport *vport, struct lpfc_nodelist *ndlp)
+{
+	unsigned long iflags;
+
+	if (ndlp->nlp_flag & NLP_RELEASE_RPI) {
+		lpfc_sli4_free_rpi(vport->phba, ndlp->nlp_rpi);
+		spin_lock_irqsave(&vport->phba->ndlp_lock, iflags);
+		ndlp->nlp_flag &= ~NLP_RELEASE_RPI;
+		ndlp->nlp_rpi = LPFC_RPI_ALLOC_ERROR;
+		spin_unlock_irqrestore(&vport->phba->ndlp_lock, iflags);
+	}
+	ndlp->nlp_flag &= ~NLP_UNREG_INP;
+}
 
 /**
  * lpfc_sli_def_mbox_cmpl - Default mailbox completion handler
@@ -2507,12 +2521,7 @@ lpfc_sli_def_mbox_cmpl(struct lpfc_hba *phba, LPFC_MBOXQ_t *pmb)
 				ndlp->nlp_defer_did = NLP_EVT_NOTHING_PENDING;
 				lpfc_issue_els_plogi(vport, ndlp->nlp_DID, 0);
 			} else {
-				if (ndlp->nlp_flag & NLP_RELEASE_RPI) {
-					lpfc_sli4_free_rpi(vport->phba,
-							   ndlp->nlp_rpi);
-					ndlp->nlp_flag &= ~NLP_RELEASE_RPI;
-				}
-				ndlp->nlp_flag &= ~NLP_UNREG_INP;
+				__lpfc_sli_rpi_release(vport, ndlp);
 			}
 			pmb->ctx_ndlp = NULL;
 		}
@@ -2587,14 +2596,7 @@ lpfc_sli4_unreg_rpi_cmpl_clr(struct lpfc_hba *phba, LPFC_MBOXQ_t *pmb)
 					lpfc_issue_els_plogi(
 						vport, ndlp->nlp_DID, 0);
 				} else {
-					if (ndlp->nlp_flag & NLP_RELEASE_RPI) {
-						lpfc_sli4_free_rpi(
-							vport->phba,
-							ndlp->nlp_rpi);
-						ndlp->nlp_flag &=
-							~NLP_RELEASE_RPI;
-					}
-					ndlp->nlp_flag &= ~NLP_UNREG_INP;
+					__lpfc_sli_rpi_release(vport, ndlp);
 				}
 			}
 		}
@@ -18225,6 +18227,10 @@ __lpfc_sli4_free_rpi(struct lpfc_hba *phba, int rpi)
 	if (test_and_clear_bit(rpi, phba->sli4_hba.rpi_bmask)) {
 		phba->sli4_hba.rpi_count--;
 		phba->sli4_hba.max_cfg_param.rpi_used--;
+	} else {
+		lpfc_printf_log(phba, KERN_INFO, LOG_SLI,
+				"2016 rpi %x not inuse\n",
+				rpi);
 	}
 }