IB/hfi1: Add additional checks when handling TID RDMA WRITE DATA packet

In a congested fabric with adaptive routing enabled, traces show that
packets could be delivered out of order, which could cause incorrect
processing of stale packets. For stale TID RDMA WRITE DATA packets that
cause KDETH EFLAGS errors, this patch adds additional checks before
processing the packets.

Fixes: d72fe7d5008b ("IB/hfi1: Add a function to receive TID RDMA WRITE DATA packet")
Cc: <stable@vger.kernel.org>
Reviewed-by: Mike Marciniszyn <mike.marciniszyn@intel.com>
Signed-off-by: Kaike Wan <kaike.wan@intel.com>
Signed-off-by: Dennis Dalessandro <dennis.dalessandro@intel.com>
Link: https://lore.kernel.org/r/20190815192051.105923.69979.stgit@awfm-01.aw.intel.com
Signed-off-by: Doug Ledford <dledford@redhat.com>

diff --git a/drivers/infiniband/hw/hfi1/tid_rdma.c b/drivers/infiniband/hw/hfi1/tid_rdma.c
index 23bb249..7bccb59 100644 (file)
--- a/drivers/infiniband/hw/hfi1/tid_rdma.c
+++ b/drivers/infiniband/hw/hfi1/tid_rdma.c
@@ -2945,8 +2945,15 @@ bool hfi1_handle_kdeth_eflags(struct hfi1_ctxtdata *rcd,
         */
        spin_lock(&qp->s_lock);
        qpriv = qp->priv;
+       if (qpriv->r_tid_tail == HFI1_QP_WQE_INVALID ||
+           qpriv->r_tid_tail == qpriv->r_tid_head)
+               goto unlock;
        e = &qp->s_ack_queue[qpriv->r_tid_tail];
+       if (e->opcode != TID_OP(WRITE_REQ))
+               goto unlock;
        req = ack_to_tid_req(e);
+       if (req->comp_seg == req->cur_seg)
+               goto unlock;
        flow = &req->flows[req->clear_tail];
        trace_hfi1_eflags_err_write(qp, rcv_type, rte, psn);
        trace_hfi1_rsp_handle_kdeth_eflags(qp, psn);