On Thu, Jan 28, 2021 at 05:33:19PM -0600, Bob Pearson wrote:
> rxe_udp_encap_recv() drops the reference to rxe->ib_dev taken by
> rxe_get_dev_from_net() which should be held until each received
> skb is freed. This patch moves the calls to ib_device_put() to
> each place a received skb is freed. It also takes references to
> the ib_device for each cloned skb created to process received
> multicast packets.
>
> Fixes: 4c173f596b3ff ("RDMA/rxe: Use ib_device_get_by_netdev()
> instead of open coding")
Do not line wrap fixes lines
> diff --git a/drivers/infiniband/sw/rxe/rxe_recv.c b/drivers/infiniband/sw/rxe/rxe_recv.c
> index c9984a28eecc..8e60d9eaf79a 100644
> --- a/drivers/infiniband/sw/rxe/rxe_recv.c
> +++ b/drivers/infiniband/sw/rxe/rxe_recv.c
> @@ -266,10 +266,19 @@ static void rxe_rcv_mcast_pkt(struct rxe_dev *rxe, struct sk_buff *skb)
> /* for all but the last qp create a new clone of the
> * skb and pass to the qp.
> */
> - if (mce->qp_list.next != &mcg->qp_list)
> + if (mce->qp_list.next != &mcg->qp_list) {
> per_qp_skb = skb_clone(skb, GFP_ATOMIC);
> - else
> + if (!ib_device_try_get(&rxe->ib_dev)) {
> + /* shouldn't happen we already have
> + * one ref for skb.
> + */
> + pr_warn("ib_device_try_get failed\n");
> + kfree_skb(per_qp_skb);
I fixed this to just
+ if (WARN_ON(!ib_device_try_get(&rxe->ib_dev))) {
+ kfree_skb(per_qp_skb);
+ continue;
+ }
> + continue;
> + }
> + } else {
> per_qp_skb = skb;
> + }
And this had a merge collision with the mcast patch, I fixed it up
Applied to for-next, thanks
Jason
