[PATCH 03/10] lpfc: Log mcqe contents for mbox commands with no context

[Justin Tee <justintee8345@gmail.com>]

Update log message to display the entirety of an MCQE for which there is no
submission context.  This log message is not expected to occur and hence is
tagged as a LOG_TRACE_EVENT.  As such, move the hbalock release to before
this log message so that the trace event process does not hold the hbalock
for too long.

Signed-off-by: Justin Tee <justintee8345@gmail.com>
---
 drivers/scsi/lpfc/lpfc_sli.c | 8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

diff --git a/drivers/scsi/lpfc/lpfc_sli.c b/drivers/scsi/lpfc/lpfc_sli.c
index bd71292e7480..b32a1870eec2 100644
--- a/drivers/scsi/lpfc/lpfc_sli.c
+++ b/drivers/scsi/lpfc/lpfc_sli.c
@@ -14337,13 +14337,15 @@ lpfc_sli4_sp_handle_mbox_event(struct lpfc_hba *phba, struct lpfc_mcqe *mcqe)
 	/* Get the reference to the active mbox command */
 	spin_lock_irqsave(&phba->hbalock, iflags);
 	pmb = phba->sli.mbox_active;
+	spin_unlock_irqrestore(&phba->hbalock, iflags);
 	if (unlikely(!pmb)) {
 		lpfc_printf_log(phba, KERN_ERR, LOG_TRACE_EVENT,
-				"1832 No pending MBOX command to handle\n");
-		spin_unlock_irqrestore(&phba->hbalock, iflags);
+				"1832 No pending MBOX command to handle, "
+				"mcqe: x%08x x%08x x%08x x%08x\n",
+				mcqe->word0, mcqe->mcqe_tag0,
+				mcqe->mcqe_tag1, mcqe->trailer);
 		goto out_no_mqe_complete;
 	}
-	spin_unlock_irqrestore(&phba->hbalock, iflags);
 	mqe = &pmb->u.mqe;
 	pmbox = (MAILBOX_t *)&pmb->u.mqe;
 	mbox = phba->mbox;
-- 
2.38.0