[SCSI] isci: Leave requests alone if already terminating.

Instead of immediately completing any request that has a second
termination call made on it, wait for the TC done/abort HW event.

Signed-off-by: Jeff Skirvin <jeffrey.d.skirvin@intel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
diff --git a/drivers/scsi/isci/request.c b/drivers/scsi/isci/request.c
index b4cf998..b5d3a8c 100644
--- a/drivers/scsi/isci/request.c
+++ b/drivers/scsi/isci/request.c
@@ -732,12 +732,20 @@
 		sci_change_state(&ireq->sm, SCI_REQ_ABORTING);
 		return SCI_SUCCESS;
 	case SCI_REQ_TASK_WAIT_TC_RESP:
+		/* The task frame was already confirmed to have been
+		 * sent by the SCU HW.  Since the state machine is
+		 * now only waiting for the task response itself,
+		 * abort the request and complete it immediately
+		 * and don't wait for the task response.
+		 */
 		sci_change_state(&ireq->sm, SCI_REQ_ABORTING);
 		sci_change_state(&ireq->sm, SCI_REQ_COMPLETED);
 		return SCI_SUCCESS;
 	case SCI_REQ_ABORTING:
-		sci_change_state(&ireq->sm, SCI_REQ_COMPLETED);
-		return SCI_SUCCESS;
+		/* If a request has a termination requested twice, return
+		 * a failure indication, since HW confirmation of the first
+		 * abort is still outstanding.
+		 */
 	case SCI_REQ_COMPLETED:
 	default:
 		dev_warn(&ireq->owning_controller->pdev->dev,