RDMA/cxgb4: Don't re-init wait object in init/fini paths

Re-initializing the wait object in rdma_init()/rdma_fini() causes a
timing window which can lead to a deadlock during close.  Once this
deadlock hits, all RDMA activity over the T4 device will be stuck.

There's no need to re-init the wait object, so remove it.

Signed-off-by: Steve Wise <swise@opengridcomputing.com>
Cc: <stable@kernel.org>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

diff --git a/drivers/infiniband/hw/cxgb4/qp.c b/drivers/infiniband/hw/cxgb4/qp.c
index 565a061..2080090 100644 (file)
--- a/drivers/infiniband/hw/cxgb4/qp.c
+++ b/drivers/infiniband/hw/cxgb4/qp.c
@@ -999,7 +999,6 @@ static int rdma_fini(struct c4iw_dev *rhp, struct c4iw_qp *qhp,
        wqe->cookie = (unsigned long) &ep->com.wr_wait;
 
        wqe->u.fini.type = FW_RI_TYPE_FINI;
-       c4iw_init_wr_wait(&ep->com.wr_wait);
        ret = c4iw_ofld_send(&rhp->rdev, skb);
        if (ret)
                goto out;
@@ -1095,7 +1094,6 @@ static int rdma_init(struct c4iw_dev *rhp, struct c4iw_qp *qhp)
        if (qhp->attr.mpa_attr.initiator)
                build_rtr_msg(qhp->attr.mpa_attr.p2p_type, &wqe->u.init);
 
-       c4iw_init_wr_wait(&qhp->ep->com.wr_wait);
        ret = c4iw_ofld_send(&rhp->rdev, skb);
        if (ret)
                goto out;