IB/rxe: Move refcounting earlier in rxe_send()

The network stack will call nskb's destructor, rxe_skb_tx_dtor(), if the
packet gets dropped by ip_local_out()/ip6_local_out(). Thus we need to add
the QP ref before output to avoid extra dereferences during network
congestion. This could lead to unwanted destruction of the QP.

Fix up the skb_out accounting, too.

Fixes: fda85ce91240 ("IB/rxe: Fix kernel panic from skb destructor")
Signed-off-by: Andrew Boyer <andrew.boyer@dell.com>
Acked-by: Moni Shoua <monis@mellanox.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>

diff --git a/drivers/infiniband/sw/rxe/rxe_net.c b/drivers/infiniband/sw/rxe/rxe_net.c
index 08f3f90d29123e9f6840b34f72b8aa56b7054b7d..0810f38bb1fa971002a905031689013f22952872 100644 (file)
--- a/drivers/infiniband/sw/rxe/rxe_net.c
+++ b/drivers/infiniband/sw/rxe/rxe_net.c
@@ -460,12 +460,17 @@ int rxe_send(struct rxe_dev *rxe, struct rxe_pkt_info *pkt, struct sk_buff *skb)
        nskb->destructor = rxe_skb_tx_dtor;
        nskb->sk = pkt->qp->sk->sk;
 
+       rxe_add_ref(pkt->qp);
+       atomic_inc(&pkt->qp->skb_out);
+
        if (av->network_type == RDMA_NETWORK_IPV4) {
                err = ip_local_out(dev_net(skb_dst(skb)->dev), nskb->sk, nskb);
        } else if (av->network_type == RDMA_NETWORK_IPV6) {
                err = ip6_local_out(dev_net(skb_dst(skb)->dev), nskb->sk, nskb);
        } else {
                pr_err("Unknown layer 3 protocol: %d\n", av->network_type);
+               atomic_dec(&pkt->qp->skb_out);
+               rxe_drop_ref(pkt->qp);
                kfree_skb(nskb);
                return -EINVAL;
        }
@@ -475,10 +480,7 @@ int rxe_send(struct rxe_dev *rxe, struct rxe_pkt_info *pkt, struct sk_buff *skb)
                return -EAGAIN;
        }
 
-       rxe_add_ref(pkt->qp);
-       atomic_inc(&pkt->qp->skb_out);
        kfree_skb(skb);
-
        return 0;
 }