2 files changed, 167 insertions, 45 deletions

diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c
index fe1610dde..ed458b315 100644
--- a/net/packet/af_packet.c
+++ b/net/packet/af_packet.c
@@ -543,15 +543,11 @@ static void prb_init_blk_timer(struct packet_sock *po,
 	pkc->retire_blk_timer.expires = jiffies;
 }
 
-static void prb_setup_retire_blk_timer(struct packet_sock *po, int tx_ring)
+static void prb_setup_retire_blk_timer(struct packet_sock *po)
 {
 	struct tpacket_kbdq_core *pkc;
 
-	if (tx_ring)
-		BUG();
-
-	pkc = tx_ring ? GET_PBDQC_FROM_RB(&po->tx_ring) :
-			GET_PBDQC_FROM_RB(&po->rx_ring);
+	pkc = GET_PBDQC_FROM_RB(&po->rx_ring);
 	prb_init_blk_timer(po, pkc, prb_retire_rx_blk_timer_expired);
 }
 
@@ -607,7 +603,7 @@ static void prb_init_ft_ops(struct tpacket_kbdq_core *p1,
 static void init_prb_bdqc(struct packet_sock *po,
 			struct packet_ring_buffer *rb,
 			struct pgv *pg_vec,
-			union tpacket_req_u *req_u, int tx_ring)
+			union tpacket_req_u *req_u)
 {
 	struct tpacket_kbdq_core *p1 = GET_PBDQC_FROM_RB(rb);
 	struct tpacket_block_desc *pbd;
@@ -634,7 +630,7 @@ static void init_prb_bdqc(struct packet_sock *po,
 
 	p1->max_frame_len = p1->kblk_size - BLK_PLUS_PRIV(p1->blk_sizeof_priv);
 	prb_init_ft_ops(p1, req_u);
-	prb_setup_retire_blk_timer(po, tx_ring);
+	prb_setup_retire_blk_timer(po);
 	prb_open_block(p1, pbd);
 }
 
@@ -1234,27 +1230,81 @@ static void packet_free_pending(struct packet_sock *po)
 	free_percpu(po->tx_ring.pending_refcnt);
 }
 
-static bool packet_rcv_has_room(struct packet_sock *po, struct sk_buff *skb)
+#define ROOM_POW_OFF	2
+#define ROOM_NONE	0x0
+#define ROOM_LOW	0x1
+#define ROOM_NORMAL	0x2
+
+static bool __tpacket_has_room(struct packet_sock *po, int pow_off)
+{
+	int idx, len;
+
+	len = po->rx_ring.frame_max + 1;
+	idx = po->rx_ring.head;
+	if (pow_off)
+		idx += len >> pow_off;
+	if (idx >= len)
+		idx -= len;
+	return packet_lookup_frame(po, &po->rx_ring, idx, TP_STATUS_KERNEL);
+}
+
+static bool __tpacket_v3_has_room(struct packet_sock *po, int pow_off)
+{
+	int idx, len;
+
+	len = po->rx_ring.prb_bdqc.knum_blocks;
+	idx = po->rx_ring.prb_bdqc.kactive_blk_num;
+	if (pow_off)
+		idx += len >> pow_off;
+	if (idx >= len)
+		idx -= len;
+	return prb_lookup_block(po, &po->rx_ring, idx, TP_STATUS_KERNEL);
+}
+
+static int __packet_rcv_has_room(struct packet_sock *po, struct sk_buff *skb)
 {
 	struct sock *sk = &po->sk;
-	bool has_room;
+	int ret = ROOM_NONE;
+
+	if (po->prot_hook.func != tpacket_rcv) {
+		int avail = sk->sk_rcvbuf - atomic_read(&sk->sk_rmem_alloc)
+					  - (skb ? skb->truesize : 0);
+		if (avail > (sk->sk_rcvbuf >> ROOM_POW_OFF))
+			return ROOM_NORMAL;
+		else if (avail > 0)
+			return ROOM_LOW;
+		else
+			return ROOM_NONE;
+	}
 
-	if (po->prot_hook.func != tpacket_rcv)
-		return (atomic_read(&sk->sk_rmem_alloc) + skb->truesize)
-			<= sk->sk_rcvbuf;
+	if (po->tp_version == TPACKET_V3) {
+		if (__tpacket_v3_has_room(po, ROOM_POW_OFF))
+			ret = ROOM_NORMAL;
+		else if (__tpacket_v3_has_room(po, 0))
+			ret = ROOM_LOW;
+	} else {
+		if (__tpacket_has_room(po, ROOM_POW_OFF))
+			ret = ROOM_NORMAL;
+		else if (__tpacket_has_room(po, 0))
+			ret = ROOM_LOW;
+	}
 
-	spin_lock(&sk->sk_receive_queue.lock);
-	if (po->tp_version == TPACKET_V3)
-		has_room = prb_lookup_block(po, &po->rx_ring,
-					    po->rx_ring.prb_bdqc.kactive_blk_num,
-					    TP_STATUS_KERNEL);
-	else
-		has_room = packet_lookup_frame(po, &po->rx_ring,
-					       po->rx_ring.head,
-					       TP_STATUS_KERNEL);
-	spin_unlock(&sk->sk_receive_queue.lock);
+	return ret;
+}
+
+static int packet_rcv_has_room(struct packet_sock *po, struct sk_buff *skb)
+{
+	int ret;
+	bool has_room;
+
+	spin_lock_bh(&po->sk.sk_receive_queue.lock);
+	ret = __packet_rcv_has_room(po, skb);
+	has_room = ret == ROOM_NORMAL;
+	if (po->pressure == has_room)
+		po->pressure = !has_room;
+	spin_unlock_bh(&po->sk.sk_receive_queue.lock);
 
-	return has_room;
+	return ret;
 }
 
 static void packet_sock_destruct(struct sock *sk)
@@ -1272,6 +1322,20 @@ static void packet_sock_destruct(struct sock *sk)
 	sk_refcnt_debug_dec(sk);
 }
 
+static bool fanout_flow_is_huge(struct packet_sock *po, struct sk_buff *skb)
+{
+	u32 rxhash;
+	int i, count = 0;
+
+	rxhash = skb_get_hash(skb);
+	for (i = 0; i < ROLLOVER_HLEN; i++)
+		if (po->rollover->history[i] == rxhash)
+			count++;
+
+	po->rollover->history[prandom_u32() % ROLLOVER_HLEN] = rxhash;
+	return count > (ROLLOVER_HLEN >> 1);
+}
+
 static unsigned int fanout_demux_hash(struct packet_fanout *f,
 				      struct sk_buff *skb,
 				      unsigned int num)
@@ -1304,22 +1368,40 @@ static unsigned int fanout_demux_rnd(struct packet_fanout *f,
 
 static unsigned int fanout_demux_rollover(struct packet_fanout *f,
 					  struct sk_buff *skb,
-					  unsigned int idx, unsigned int skip,
+					  unsigned int idx, bool try_self,
 					  unsigned int num)
 {
-	unsigned int i, j;
+	struct packet_sock *po, *po_next, *po_skip = NULL;
+	unsigned int i, j, room = ROOM_NONE;
+
+	po = pkt_sk(f->arr[idx]);
 
-	i = j = min_t(int, f->next[idx], num - 1);
+	if (try_self) {
+		room = packet_rcv_has_room(po, skb);
+		if (room == ROOM_NORMAL ||
+		    (room == ROOM_LOW && !fanout_flow_is_huge(po, skb)))
+			return idx;
+		po_skip = po;
+	}
+
+	i = j = min_t(int, po->rollover->sock, num - 1);
 	do {
-		if (i != skip && packet_rcv_has_room(pkt_sk(f->arr[i]), skb)) {
+		po_next = pkt_sk(f->arr[i]);
+		if (po_next != po_skip && !po_next->pressure &&
+		    packet_rcv_has_room(po_next, skb) == ROOM_NORMAL) {
 			if (i != j)
-				f->next[idx] = i;
+				po->rollover->sock = i;
+			atomic_long_inc(&po->rollover->num);
+			if (room == ROOM_LOW)
+				atomic_long_inc(&po->rollover->num_huge);
 			return i;
 		}
+
 		if (++i == num)
 			i = 0;
 	} while (i != j);
 
+	atomic_long_inc(&po->rollover->num_failed);
 	return idx;
 }
 
@@ -1372,17 +1454,14 @@ static int packet_rcv_fanout(struct sk_buff *skb, struct net_device *dev,
 		idx = fanout_demux_qm(f, skb, num);
 		break;
 	case PACKET_FANOUT_ROLLOVER:
-		idx = fanout_demux_rollover(f, skb, 0, (unsigned int) -1, num);
+		idx = fanout_demux_rollover(f, skb, 0, false, num);
 		break;
 	}
 
-	po = pkt_sk(f->arr[idx]);
-	if (fanout_has_flag(f, PACKET_FANOUT_FLAG_ROLLOVER) &&
-	    unlikely(!packet_rcv_has_room(po, skb))) {
-		idx = fanout_demux_rollover(f, skb, idx, idx, num);
-		po = pkt_sk(f->arr[idx]);
-	}
+	if (fanout_has_flag(f, PACKET_FANOUT_FLAG_ROLLOVER))
+		idx = fanout_demux_rollover(f, skb, idx, true, num);
 
+	po = pkt_sk(f->arr[idx]);
 	return po->prot_hook.func(skb, dev, &po->prot_hook, orig_dev);
 }
 
@@ -1453,6 +1532,16 @@ static int fanout_add(struct sock *sk, u16 id, u16 type_flags)
 	if (po->fanout)
 		return -EALREADY;
 
+	if (type == PACKET_FANOUT_ROLLOVER ||
+	    (type_flags & PACKET_FANOUT_FLAG_ROLLOVER)) {
+		po->rollover = kzalloc(sizeof(*po->rollover), GFP_KERNEL);
+		if (!po->rollover)
+			return -ENOMEM;
+		atomic_long_set(&po->rollover->num, 0);
+		atomic_long_set(&po->rollover->num_huge, 0);
+		atomic_long_set(&po->rollover->num_failed, 0);
+	}
+
 	mutex_lock(&fanout_mutex);
 	match = NULL;
 	list_for_each_entry(f, &fanout_list, list) {
@@ -1501,6 +1590,10 @@ static int fanout_add(struct sock *sk, u16 id, u16 type_flags)
 	}
 out:
 	mutex_unlock(&fanout_mutex);
+	if (err) {
+		kfree(po->rollover);
+		po->rollover = NULL;
+	}
 	return err;
 }
 
@@ -1522,6 +1615,9 @@ static void fanout_release(struct sock *sk)
 		kfree(f);
 	}
 	mutex_unlock(&fanout_mutex);
+
+	if (po->rollover)
+		kfree_rcu(po->rollover, rcu);
 }
 
 static const struct proto_ops packet_ops;
@@ -2307,7 +2403,8 @@ static int tpacket_snd(struct packet_sock *po, struct msghdr *msg)
 		}
 		tp_len = tpacket_fill_skb(po, skb, ph, dev, size_max, proto,
 					  addr, hlen);
-		if (tp_len > dev->mtu + dev->hard_header_len) {
+		if (likely(tp_len >= 0) &&
+		    tp_len > dev->mtu + dev->hard_header_len) {
 			struct ethhdr *ehdr;
 			/* Earlier code assumed this would be a VLAN pkt,
 			 * double-check this now that we have the actual
@@ -2688,7 +2785,7 @@ static int packet_release(struct socket *sock)
 static int packet_do_bind(struct sock *sk, struct net_device *dev, __be16 proto)
 {
 	struct packet_sock *po = pkt_sk(sk);
-	const struct net_device *dev_curr;
+	struct net_device *dev_curr;
 	__be16 proto_curr;
 	bool need_rehook;
 
@@ -2712,15 +2809,13 @@ static int packet_do_bind(struct sock *sk, struct net_device *dev, __be16 proto)
 
 		po->num = proto;
 		po->prot_hook.type = proto;
-
-		if (po->prot_hook.dev)
-			dev_put(po->prot_hook.dev);
-
 		po->prot_hook.dev = dev;
 
 		po->ifindex = dev ? dev->ifindex : 0;
 		packet_cached_dev_assign(po, dev);
 	}
+	if (dev_curr)
+		dev_put(dev_curr);
 
 	if (proto == 0 || !need_rehook)
 		goto out_unlock;
@@ -2821,7 +2916,7 @@ static int packet_create(struct net *net, struct socket *sock, int protocol,
 	sock->state = SS_UNCONNECTED;
 
 	err = -ENOBUFS;
-	sk = sk_alloc(net, PF_PACKET, GFP_KERNEL, &packet_proto);
+	sk = sk_alloc(net, PF_PACKET, GFP_KERNEL, &packet_proto, kern);
 	if (sk == NULL)
 		goto out;
 
@@ -2851,6 +2946,7 @@ static int packet_create(struct net *net, struct socket *sock, int protocol,
 
 	spin_lock_init(&po->bind_lock);
 	mutex_init(&po->pg_vec_lock);
+	po->rollover = NULL;
 	po->prot_hook.func = packet_rcv;
 
 	if (sock->type == SOCK_PACKET)
@@ -2928,6 +3024,9 @@ static int packet_recvmsg(struct socket *sock, struct msghdr *msg, size_t len,
 	if (skb == NULL)
 		goto out;
 
+	if (pkt_sk(sk)->pressure)
+		packet_rcv_has_room(pkt_sk(sk), NULL);
+
 	if (pkt_sk(sk)->has_vnet_hdr) {
 		struct virtio_net_hdr vnet_hdr = { 0 };
 
@@ -3471,6 +3570,7 @@ static int packet_getsockopt(struct socket *sock, int level, int optname,
 	struct packet_sock *po = pkt_sk(sk);
 	void *data = &val;
 	union tpacket_stats_u st;
+	struct tpacket_rollover_stats rstats;
 
 	if (level != SOL_PACKET)
 		return -ENOPROTOOPT;
@@ -3546,6 +3646,15 @@ static int packet_getsockopt(struct socket *sock, int level, int optname,
 			((u32)po->fanout->flags << 24)) :
 		       0);
 		break;
+	case PACKET_ROLLOVER_STATS:
+		if (!po->rollover)
+			return -EINVAL;
+		rstats.tp_all = atomic_long_read(&po->rollover->num);
+		rstats.tp_huge = atomic_long_read(&po->rollover->num_huge);
+		rstats.tp_failed = atomic_long_read(&po->rollover->num_failed);
+		data = &rstats;
+		lv = sizeof(rstats);
+		break;
 	case PACKET_TX_HAS_OFF:
 		val = po->tp_tx_has_off;
 		break;
@@ -3683,6 +3792,8 @@ static unsigned int packet_poll(struct file *file, struct socket *sock,
 			TP_STATUS_KERNEL))
 			mask |= POLLIN | POLLRDNORM;
 	}
+	if (po->pressure && __packet_rcv_has_room(po, NULL) == ROOM_NORMAL)
+		po->pressure = 0;
 	spin_unlock_bh(&sk->sk_receive_queue.lock);
 	spin_lock_bh(&sk->sk_write_queue.lock);
 	if (po->tx_ring.pg_vec) {
@@ -3872,7 +3983,7 @@ static int packet_set_ring(struct sock *sk, union tpacket_req_u *req_u,
 		 * it above but just being paranoid
 		 */
 			if (!tx_ring)
-				init_prb_bdqc(po, rb, pg_vec, req_u, tx_ring);
+				init_prb_bdqc(po, rb, pg_vec, req_u);
 			break;
 		default:
 			break;
diff --git a/net/packet/internal.h b/net/packet/internal.h
index fe6e20cae..e20b3e882 100644
--- a/net/packet/internal.h
+++ b/net/packet/internal.h
@@ -82,12 +82,21 @@ struct packet_fanout {
 	atomic_t		rr_cur;
 	struct list_head	list;
 	struct sock		*arr[PACKET_FANOUT_MAX];
-	int			next[PACKET_FANOUT_MAX];
 	spinlock_t		lock;
 	atomic_t		sk_ref;
 	struct packet_type	prot_hook ____cacheline_aligned_in_smp;
 };
 
+struct packet_rollover {
+	int			sock;
+	struct rcu_head		rcu;
+	atomic_long_t		num;
+	atomic_long_t		num_huge;
+	atomic_long_t		num_failed;
+#define ROLLOVER_HLEN	(L1_CACHE_BYTES / sizeof(u32))
+	u32			history[ROLLOVER_HLEN] ____cacheline_aligned;
+} ____cacheline_aligned_in_smp;
+
 struct packet_sock {
 	/* struct sock has to be the first member of packet_sock */
 	struct sock		sk;
@@ -102,8 +111,10 @@ struct packet_sock {
 				auxdata:1,
 				origdev:1,
 				has_vnet_hdr:1;
+	int			pressure;
 	int			ifindex;	/* bound device		*/
 	__be16			num;
+	struct packet_rollover	*rollover;
 	struct packet_mclist	*mclist;
 	atomic_t		mapped;
 	enum tpacket_versions	tp_version;