net: better skb->sender_cpu and skb->napi_id cohabitation

skb->sender_cpu and skb->napi_id share a common storage,
and we had various bugs about this.

We had to call skb_sender_cpu_clear() in some places to
not leave a prior skb->napi_id and fool netdev_pick_tx()

As suggested by Alexei, we could split the space so that
these errors can not happen.

0 value being reserved as the common (not initialized) value,
let's reserve [1 .. NR_CPUS] range for valid sender_cpu,
and [NR_CPUS+1 .. ~0U] for valid napi_id.

This will allow proper busy polling support over tunnels.
Signed-off-by: NEric Dumazet <edumazet@google.com>
Suggested-by: NAlexei Starovoitov <ast@kernel.org>
Acked-by: NAlexei Starovoitov <ast@kernel.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

include/linux/skbuff.h

@@ -1082,9 +1082,6 @@ static inline void skb_copy_hash(struct sk_buff *to, const struct sk_buff *from)
 
 static inline void skb_sender_cpu_clear(struct sk_buff *skb)
 {
-#ifdef CONFIG_XPS
-	skb->sender_cpu = 0;
-#endif
 }
 
 #ifdef NET_SKBUFF_DATA_USES_OFFSET

net/core/dev.c

@@ -182,7 +182,7 @@ EXPORT_SYMBOL(dev_base_lock);
 /* protects napi_hash addition/deletion and napi_gen_id */
 static DEFINE_SPINLOCK(napi_hash_lock);
 
-static unsigned int napi_gen_id;
+static unsigned int napi_gen_id = NR_CPUS;
 static DEFINE_HASHTABLE(napi_hash, 8);
 
 static seqcount_t devnet_rename_seq;
@@ -3021,7 +3021,9 @@ struct netdev_queue *netdev_pick_tx(struct net_device *dev,
 	int queue_index = 0;
 
 #ifdef CONFIG_XPS
-	if (skb->sender_cpu == 0)
+	u32 sender_cpu = skb->sender_cpu - 1;
+
+	if (sender_cpu >= (u32)NR_CPUS)
 		skb->sender_cpu = raw_smp_processor_id() + 1;
 #endif
 
@@ -4676,25 +4678,22 @@ EXPORT_SYMBOL_GPL(napi_by_id);
 
 void napi_hash_add(struct napi_struct *napi)
 {
-	if (!test_and_set_bit(NAPI_STATE_HASHED, &napi->state)) {
+	if (test_and_set_bit(NAPI_STATE_HASHED, &napi->state))
+		return;
 
-		spin_lock(&napi_hash_lock);
+	spin_lock(&napi_hash_lock);
 
-		/* 0 is not a valid id, we also skip an id that is taken
-		 * we expect both events to be extremely rare
-		 */
-		napi->napi_id = 0;
-		while (!napi->napi_id) {
-			napi->napi_id = ++napi_gen_id;
-			if (napi_by_id(napi->napi_id))
-				napi->napi_id = 0;
-		}
+	/* 0..NR_CPUS+1 range is reserved for sender_cpu use */
+	do {
+		if (unlikely(++napi_gen_id < NR_CPUS + 1))
+			napi_gen_id = NR_CPUS + 1;
+	} while (napi_by_id(napi_gen_id));
+	napi->napi_id = napi_gen_id;
 
-		hlist_add_head_rcu(&napi->napi_hash_node,
-			&napi_hash[napi->napi_id % HASH_SIZE(napi_hash)]);
+	hlist_add_head_rcu(&napi->napi_hash_node,
+			   &napi_hash[napi->napi_id % HASH_SIZE(napi_hash)]);
 
-		spin_unlock(&napi_hash_lock);
-	}
+	spin_unlock(&napi_hash_lock);
 }
 EXPORT_SYMBOL_GPL(napi_hash_add);