4.8-stable review patch.  If anyone has any objections, please let me know.

------------------

From: Mike Marciniszyn <redacted>

commit 5a648dfad033bf5f945795c4e56ee7577f24f3e9 upstream.

The call is misplaced in the reset calldown function
and causes issues with lockdep assertions that are to
be added.

Fixes: Commit a2c2d608957c ("staging/rdma/hfi1: Remove create_qp functionality")
Reviewed-by: Ira Weiny <redacted>
Reviewed-by: Dennis Dalessandro <redacted>
Signed-off-by: Mike Marciniszyn <redacted>
Signed-off-by: Dennis Dalessandro <redacted>
Signed-off-by: Doug Ledford <redacted>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

---
 drivers/infiniband/hw/hfi1/qp.c |   14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)

--- a/drivers/infiniband/hw/hfi1/qp.c
+++ b/drivers/infiniband/hw/hfi1/qp.c

@@ -808,6 +808,13 @@ void *qp_priv_alloc(struct rvt_dev_info
 		kfree(priv);
 		return ERR_PTR(-ENOMEM);
 	}
+	iowait_init(
+		&priv->s_iowait,
+		1,
+		_hfi1_do_send,
+		iowait_sleep,
+		iowait_wakeup,
+		iowait_sdma_drained);
 	setup_timer(&priv->s_rnr_timer, hfi1_rc_rnr_retry, (unsigned long)qp);
 	qp->s_timer.function = hfi1_rc_timeout;
 	return priv;

@@ -873,13 +880,6 @@ void notify_qp_reset(struct rvt_qp *qp)
 {
 	struct hfi1_qp_priv *priv = qp->priv;
 
-	iowait_init(
-		&priv->s_iowait,
-		1,
-		_hfi1_do_send,
-		iowait_sleep,
-		iowait_wakeup,
-		iowait_sdma_drained);
 	priv->r_adefered = 0;
 	clear_ahg(qp);
 }