scsi: lpfc: Fix discovery failure when PLOGI is defered

When a target's link dropped, an RSCN was received to communicate the
change. The driver detected the loss of the target and issued and UNREG_RPI
mailbox command.  While that was being processed, another RSCN was received
to communicate the port coming back.  The driver deferred the PLOGI to the
port until the mailbox command finishes. When the mailbox command completed
it saw the pending port and called the routines to issue the
PLOGI. However, it forgot to clear the UNREG_INP state flag, so the PLOGI
xmt routine nooped the PLOGI request assuming it needed to wait for the
mailbox command.  At this point, login would never be re-attempted.

Clear UNREG_INP before issuing the deferred PLOGI.

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>

1 files changed, 2 insertions, 0 deletions

diff --git a/drivers/scsi/lpfc/lpfc_hbadisc.c b/drivers/scsi/lpfc/lpfc_hbadisc.c
index 4c1ba88b678a..b183b882d506 100644
--- a/drivers/scsi/lpfc/lpfc_hbadisc.c
+++ b/drivers/scsi/lpfc/lpfc_hbadisc.c
@@ -4802,6 +4802,8 @@ lpfc_nlp_logo_unreg(struct lpfc_hba *phba, LPFC_MBOXQ_t *pmb)
 		ndlp->nlp_flag &= ~NLP_UNREG_INP;
 		ndlp->nlp_defer_did = NLP_EVT_NOTHING_PENDING;
 		lpfc_issue_els_plogi(vport, ndlp->nlp_DID, 0);
+	} else {
+		ndlp->nlp_flag &= ~NLP_UNREG_INP;
 	}
 }