pkt_sched: Fix OOPS on ingress qdisc add.

Bug report from Steven Jan Springl:

	Issuing the following command causes a kernel oops:
		tc qdisc add dev eth0 handle ffff: ingress

The problem mostly stems from all of the special case handling of
ingress qdiscs.

So, to fix this, do the grafting operation the same way we do for TX
qdiscs.  Which means that dev_activate() and dev_deactivate() now do
the "qdisc_sleeping <--> qdisc" transitions on dev->rx_queue too.

Future simplifications are possible now, mainly because it is
impossible for dev_queue->{qdisc,qdisc_sleeping} to be NULL.  There
are NULL checks all over to handle the ingress qdisc special case
that used to exist before this commit.

Signed-off-by: David S. Miller <davem@davemloft.net>

1 files changed, 2 insertions, 2 deletions

diff --git a/net/core/dev.c b/net/core/dev.c
index 8d13a9b9f1df..63d6bcddbf46 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -2100,7 +2100,7 @@ static int ing_filter(struct sk_buff *skb)
         rxq = &dev->rx_queue;
 
         q = rxq->qdisc;
-        if (q) {
+        if (q != &noop_qdisc) {
                 spin_lock(qdisc_lock(q));
                 result = qdisc_enqueue_root(skb, q);
                 spin_unlock(qdisc_lock(q));
@@ -2113,7 +2113,7 @@ static inline struct sk_buff *handle_ing(struct sk_buff *skb,
                                          struct packet_type **pt_prev,
                                          int *ret, struct net_device *orig_dev)
 {
-        if (!skb->dev->rx_queue.qdisc)
+        if (skb->dev->rx_queue.qdisc == &noop_qdisc)
                 goto out;
 
         if (*pt_prev) {