net: call rcu_read_lock early in process_backlog

commit 2c17d27c upstream.

Incoming packet should be either in backlog queue or
in RCU read-side section. Otherwise, the final sequence of
flush_backlog() and synchronize_net() may miss packets
that can run without device reference:

CPU 1                  CPU 2
                       skb->dev: no reference
                       process_backlog:__skb_dequeue
                       process_backlog:local_irq_enable

on_each_cpu for
flush_backlog =>       IPI(hardirq): flush_backlog
                       - packet not found in backlog

                       CPU delayed ...
synchronize_net
- no ongoing RCU
read-side sections

netdev_run_todo,
rcu_barrier: no
ongoing callbacks
                       __netif_receive_skb_core:rcu_read_lock
                       - too late
free dev
                       process packet for freed dev

Fixes: 6e583ce5 ("net: eliminate refcounting in backlog queue")
Cc: Eric W. Biederman <ebiederm@xmission.com>
Cc: Stephen Hemminger <stephen@networkplumber.org>
Signed-off-by: Julian Anastasov <ja@ssi.bg>
Signed-off-by: David S. Miller <davem@davemloft.net>
[lizf: Backported to 3.4:
 - adjust context
 - no need to change "goto unlock" to "goto out"]
Signed-off-by: Zefan Li <lizefan@huawei.com>

net/core/dev.c

@@ -3191,8 +3191,6 @@ static int __netif_receive_skb(struct sk_buff *skb)
 
 	pt_prev = NULL;
 
-	rcu_read_lock();
-
 another_round:
 
 	__this_cpu_inc(softnet_data.processed);
@@ -3287,7 +3285,6 @@ static int __netif_receive_skb(struct sk_buff *skb)
 	}
 
 out:
-	rcu_read_unlock();
 	return ret;
 }
 
@@ -3308,29 +3305,30 @@ static int __netif_receive_skb(struct sk_buff *skb)
  */
 int netif_receive_skb(struct sk_buff *skb)
 {
+	int ret;
+
 	net_timestamp_check(netdev_tstamp_prequeue, skb);
 
 	if (skb_defer_rx_timestamp(skb))
 		return NET_RX_SUCCESS;
 
+	rcu_read_lock();
+
 #ifdef CONFIG_RPS
 	if (static_key_false(&rps_needed)) {
 		struct rps_dev_flow voidflow, *rflow = &voidflow;
-		int cpu, ret;
-
-		rcu_read_lock();
-
-		cpu = get_rps_cpu(skb->dev, skb, &rflow);
+		int cpu = get_rps_cpu(skb->dev, skb, &rflow);
 
 		if (cpu >= 0) {
 			ret = enqueue_to_backlog(skb, cpu, &rflow->last_qtail);
 			rcu_read_unlock();
 			return ret;
 		}
-		rcu_read_unlock();
 	}
 #endif
-	return __netif_receive_skb(skb);
+	ret = __netif_receive_skb(skb);
+	rcu_read_unlock();
+	return ret;
 }
 EXPORT_SYMBOL(netif_receive_skb);
 
@@ -3721,8 +3719,10 @@ static int process_backlog(struct napi_struct *napi, int quota)
 		unsigned int qlen;
 
 		while ((skb = __skb_dequeue(&sd->process_queue))) {
+			rcu_read_lock();
 			local_irq_enable();
 			__netif_receive_skb(skb);
+			rcu_read_unlock();
 			local_irq_disable();
 			input_queue_head_incr(sd);
 			if (++work >= quota) {