[SCSI] lpfc 8.3.32: Fix system panic due to node state change

Fix System Panic During IO Test using Medusa tool

Signed-off-by: Alex Iannicelli <alex.iannicelli@emulex.com>
Signed-off-by: James Smart <james.smart@emulex.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>

diff --git a/drivers/scsi/lpfc/lpfc_scsi.c b/drivers/scsi/lpfc/lpfc_scsi.c
index 169b77b82b487c8cea694793c493ddfa1330c5a5..925975d2d76579359ebe4181141f6e077e384f33 100644 (file)
--- a/drivers/scsi/lpfc/lpfc_scsi.c
+++ b/drivers/scsi/lpfc/lpfc_scsi.c
@@ -4275,10 +4275,8 @@ lpfc_queuecommand_lck(struct scsi_cmnd *cmnd, void (*done) (struct scsi_cmnd *))
         * Catch race where our node has transitioned, but the
         * transport is still transitioning.
         */
-       if (!ndlp || !NLP_CHK_NODE_ACT(ndlp)) {
-               cmnd->result = ScsiResult(DID_IMM_RETRY, 0);
-               goto out_fail_command;
-       }
+       if (!ndlp || !NLP_CHK_NODE_ACT(ndlp))
+               goto out_tgt_busy;
        if (atomic_read(&ndlp->cmd_pending) >= ndlp->cmd_qdepth)
                goto out_tgt_busy;