1 files changed, 395 insertions, 180 deletions

diff --git a/net/vmw_vsock/virtio_transport.c b/net/vmw_vsock/virtio_transport.c
index dfbaf6bd8b1c..8c867023a2e5 100644
--- a/net/vmw_vsock/virtio_transport.c
+++ b/net/vmw_vsock/virtio_transport.c
@@ -22,8 +22,9 @@
 #include <net/af_vsock.h>
 
 static struct workqueue_struct *virtio_vsock_workqueue;
-static struct virtio_vsock *the_virtio_vsock;
+static struct virtio_vsock __rcu *the_virtio_vsock;
 static DEFINE_MUTEX(the_virtio_vsock_mutex); /* protects the_virtio_vsock */
+static struct virtio_transport virtio_transport; /* forward declaration */
 
 struct virtio_vsock {
 	struct virtio_device *vdev;
@@ -41,8 +42,7 @@ struct virtio_vsock {
 	bool tx_run;
 
 	struct work_struct send_pkt_work;
-	spinlock_t send_pkt_list_lock;
-	struct list_head send_pkt_list;
+	struct sk_buff_head send_pkt_queue;
 
 	atomic_t queued_replies;
 
@@ -62,6 +62,18 @@ struct virtio_vsock {
 	struct virtio_vsock_event event_list[8];
 
 	u32 guest_cid;
+	bool seqpacket_allow;
+
+	/* These fields are used only in tx path in function
+	 * 'virtio_transport_send_pkt_work()', so to save
+	 * stack space in it, place both of them here. Each
+	 * pointer from 'out_sgs' points to the corresponding
+	 * element in 'out_bufs' - this is initialized in
+	 * 'virtio_vsock_probe()'. Both fields are protected
+	 * by 'tx_lock'. +1 is needed for packet header.
+	 */
+	struct scatterlist *out_sgs[MAX_SKB_FRAGS + 1];
+	struct scatterlist out_bufs[MAX_SKB_FRAGS + 1];
 };
 
 static u32 virtio_transport_get_local_cid(void)
@@ -82,6 +94,63 @@ out_rcu:
 	return ret;
 }
 
+/* Caller need to hold vsock->tx_lock on vq */
+static int virtio_transport_send_skb(struct sk_buff *skb, struct virtqueue *vq,
+				     struct virtio_vsock *vsock, gfp_t gfp)
+{
+	int ret, in_sg = 0, out_sg = 0;
+	struct scatterlist **sgs;
+
+	sgs = vsock->out_sgs;
+	sg_init_one(sgs[out_sg], virtio_vsock_hdr(skb),
+		    sizeof(*virtio_vsock_hdr(skb)));
+	out_sg++;
+
+	if (!skb_is_nonlinear(skb)) {
+		if (skb->len > 0) {
+			sg_init_one(sgs[out_sg], skb->data, skb->len);
+			out_sg++;
+		}
+	} else {
+		struct skb_shared_info *si;
+		int i;
+
+		/* If skb is nonlinear, then its buffer must contain
+		 * only header and nothing more. Data is stored in
+		 * the fragged part.
+		 */
+		WARN_ON_ONCE(skb_headroom(skb) != sizeof(*virtio_vsock_hdr(skb)));
+
+		si = skb_shinfo(skb);
+
+		for (i = 0; i < si->nr_frags; i++) {
+			skb_frag_t *skb_frag = &si->frags[i];
+			void *va;
+
+			/* We will use 'page_to_virt()' for the userspace page
+			 * here, because virtio or dma-mapping layers will call
+			 * 'virt_to_phys()' later to fill the buffer descriptor.
+			 * We don't touch memory at "virtual" address of this page.
+			 */
+			va = page_to_virt(skb_frag_page(skb_frag));
+			sg_init_one(sgs[out_sg],
+				    va + skb_frag_off(skb_frag),
+				    skb_frag_size(skb_frag));
+			out_sg++;
+		}
+	}
+
+	ret = virtqueue_add_sgs(vq, sgs, out_sg, in_sg, skb, gfp);
+	/* Usually this means that there is no more space available in
+	 * the vq
+	 */
+	if (ret < 0)
+		return ret;
+
+	virtio_transport_deliver_tap_pkt(skb);
+	return 0;
+}
+
 static void
 virtio_transport_send_pkt_work(struct work_struct *work)
 {
@@ -99,41 +168,19 @@ virtio_transport_send_pkt_work(struct work_struct *work)
 	vq = vsock->vqs[VSOCK_VQ_TX];
 
 	for (;;) {
-		struct virtio_vsock_pkt *pkt;
-		struct scatterlist hdr, buf, *sgs[2];
-		int ret, in_sg = 0, out_sg = 0;
+		struct sk_buff *skb;
 		bool reply;
+		int ret;
 
-		spin_lock_bh(&vsock->send_pkt_list_lock);
-		if (list_empty(&vsock->send_pkt_list)) {
-			spin_unlock_bh(&vsock->send_pkt_list_lock);
+		skb = virtio_vsock_skb_dequeue(&vsock->send_pkt_queue);
+		if (!skb)
 			break;
-		}
 
-		pkt = list_first_entry(&vsock->send_pkt_list,
-				       struct virtio_vsock_pkt, list);
-		list_del_init(&pkt->list);
-		spin_unlock_bh(&vsock->send_pkt_list_lock);
+		reply = virtio_vsock_skb_reply(skb);
 
-		virtio_transport_deliver_tap_pkt(pkt);
-
-		reply = pkt->reply;
-
-		sg_init_one(&hdr, &pkt->hdr, sizeof(pkt->hdr));
-		sgs[out_sg++] = &hdr;
-		if (pkt->buf) {
-			sg_init_one(&buf, pkt->buf, pkt->len);
-			sgs[out_sg++] = &buf;
-		}
-
-		ret = virtqueue_add_sgs(vq, sgs, out_sg, in_sg, pkt, GFP_KERNEL);
-		/* Usually this means that there is no more space available in
-		 * the vq
-		 */
+		ret = virtio_transport_send_skb(skb, vq, vsock, GFP_KERNEL);
 		if (ret < 0) {
-			spin_lock_bh(&vsock->send_pkt_list_lock);
-			list_add(&pkt->list, &vsock->send_pkt_list);
-			spin_unlock_bh(&vsock->send_pkt_list_lock);
+			virtio_vsock_skb_queue_head(&vsock->send_pkt_queue, skb);
 			break;
 		}
 
@@ -161,34 +208,65 @@ out:
 		queue_work(virtio_vsock_workqueue, &vsock->rx_work);
 }
 
+/* Caller need to hold RCU for vsock.
+ * Returns 0 if the packet is successfully put on the vq.
+ */
+static int virtio_transport_send_skb_fast_path(struct virtio_vsock *vsock, struct sk_buff *skb)
+{
+	struct virtqueue *vq = vsock->vqs[VSOCK_VQ_TX];
+	int ret;
+
+	/* Inside RCU, can't sleep! */
+	ret = mutex_trylock(&vsock->tx_lock);
+	if (unlikely(ret == 0))
+		return -EBUSY;
+
+	ret = virtio_transport_send_skb(skb, vq, vsock, GFP_ATOMIC);
+	if (ret == 0)
+		virtqueue_kick(vq);
+
+	mutex_unlock(&vsock->tx_lock);
+
+	return ret;
+}
+
 static int
-virtio_transport_send_pkt(struct virtio_vsock_pkt *pkt)
+virtio_transport_send_pkt(struct sk_buff *skb)
 {
+	struct virtio_vsock_hdr *hdr;
 	struct virtio_vsock *vsock;
-	int len = pkt->len;
+	int len = skb->len;
+
+	hdr = virtio_vsock_hdr(skb);
 
 	rcu_read_lock();
 	vsock = rcu_dereference(the_virtio_vsock);
 	if (!vsock) {
-		virtio_transport_free_pkt(pkt);
+		kfree_skb(skb);
 		len = -ENODEV;
 		goto out_rcu;
 	}
 
-	if (le64_to_cpu(pkt->hdr.dst_cid) == vsock->guest_cid) {
-		virtio_transport_free_pkt(pkt);
+	if (le64_to_cpu(hdr->dst_cid) == vsock->guest_cid) {
+		kfree_skb(skb);
 		len = -ENODEV;
 		goto out_rcu;
 	}
 
-	if (pkt->reply)
-		atomic_inc(&vsock->queued_replies);
-
-	spin_lock_bh(&vsock->send_pkt_list_lock);
-	list_add_tail(&pkt->list, &vsock->send_pkt_list);
-	spin_unlock_bh(&vsock->send_pkt_list_lock);
+	/* If send_pkt_queue is empty, we can safely bypass this queue
+	 * because packet order is maintained and (try) to put the packet
+	 * on the virtqueue using virtio_transport_send_skb_fast_path.
+	 * If this fails we simply put the packet on the intermediate
+	 * queue and schedule the worker.
+	 */
+	if (!skb_queue_empty_lockless(&vsock->send_pkt_queue) ||
+	    virtio_transport_send_skb_fast_path(vsock, skb)) {
+		if (virtio_vsock_skb_reply(skb))
+			atomic_inc(&vsock->queued_replies);
 
-	queue_work(virtio_vsock_workqueue, &vsock->send_pkt_work);
+		virtio_vsock_skb_queue_tail(&vsock->send_pkt_queue, skb);
+		queue_work(virtio_vsock_workqueue, &vsock->send_pkt_work);
+	}
 
 out_rcu:
 	rcu_read_unlock();
@@ -199,9 +277,7 @@ static int
 virtio_transport_cancel_pkt(struct vsock_sock *vsk)
 {
 	struct virtio_vsock *vsock;
-	struct virtio_vsock_pkt *pkt, *n;
 	int cnt = 0, ret;
-	LIST_HEAD(freeme);
 
 	rcu_read_lock();
 	vsock = rcu_dereference(the_virtio_vsock);
@@ -210,20 +286,7 @@ virtio_transport_cancel_pkt(struct vsock_sock *vsk)
 		goto out_rcu;
 	}
 
-	spin_lock_bh(&vsock->send_pkt_list_lock);
-	list_for_each_entry_safe(pkt, n, &vsock->send_pkt_list, list) {
-		if (pkt->vsk != vsk)
-			continue;
-		list_move(&pkt->list, &freeme);
-	}
-	spin_unlock_bh(&vsock->send_pkt_list_lock);
-
-	list_for_each_entry_safe(pkt, n, &freeme, list) {
-		if (pkt->reply)
-			cnt++;
-		list_del(&pkt->list);
-		virtio_transport_free_pkt(pkt);
-	}
+	cnt = virtio_transport_purge_skbs(vsk, &vsock->send_pkt_queue);
 
 	if (cnt) {
 		struct virtqueue *rx_vq = vsock->vqs[VSOCK_VQ_RX];
@@ -244,38 +307,28 @@ out_rcu:
 
 static void virtio_vsock_rx_fill(struct virtio_vsock *vsock)
 {
-	int buf_len = VIRTIO_VSOCK_DEFAULT_RX_BUF_SIZE;
-	struct virtio_vsock_pkt *pkt;
-	struct scatterlist hdr, buf, *sgs[2];
+	int total_len = VIRTIO_VSOCK_DEFAULT_RX_BUF_SIZE;
+	struct scatterlist pkt, *p;
 	struct virtqueue *vq;
+	struct sk_buff *skb;
 	int ret;
 
 	vq = vsock->vqs[VSOCK_VQ_RX];
 
 	do {
-		pkt = kzalloc(sizeof(*pkt), GFP_KERNEL);
-		if (!pkt)
+		skb = virtio_vsock_alloc_linear_skb(total_len, GFP_KERNEL);
+		if (!skb)
 			break;
 
-		pkt->buf = kmalloc(buf_len, GFP_KERNEL);
-		if (!pkt->buf) {
-			virtio_transport_free_pkt(pkt);
+		memset(skb->head, 0, VIRTIO_VSOCK_SKB_HEADROOM);
+		sg_init_one(&pkt, virtio_vsock_hdr(skb), total_len);
+		p = &pkt;
+		ret = virtqueue_add_sgs(vq, &p, 0, 1, skb, GFP_KERNEL);
+		if (ret < 0) {
+			kfree_skb(skb);
 			break;
 		}
 
-		pkt->buf_len = buf_len;
-		pkt->len = buf_len;
-
-		sg_init_one(&hdr, &pkt->hdr, sizeof(pkt->hdr));
-		sgs[0] = &hdr;
-
-		sg_init_one(&buf, pkt->buf, buf_len);
-		sgs[1] = &buf;
-		ret = virtqueue_add_sgs(vq, sgs, 0, 2, pkt, GFP_KERNEL);
-		if (ret) {
-			virtio_transport_free_pkt(pkt);
-			break;
-		}
 		vsock->rx_buf_nr++;
 	} while (vq->num_free);
 	if (vsock->rx_buf_nr > vsock->rx_buf_max_nr)
@@ -297,12 +350,12 @@ static void virtio_transport_tx_work(struct work_struct *work)
 		goto out;
 
 	do {
-		struct virtio_vsock_pkt *pkt;
+		struct sk_buff *skb;
 		unsigned int len;
 
 		virtqueue_disable_cb(vq);
-		while ((pkt = virtqueue_get_buf(vq, &len)) != NULL) {
-			virtio_transport_free_pkt(pkt);
+		while ((skb = virtqueue_get_buf(vq, &len)) != NULL) {
+			virtio_transport_consume_skb_sent(skb, true);
 			added = true;
 		}
 	} while (!virtqueue_enable_cb(vq));
@@ -356,11 +409,14 @@ static void virtio_vsock_event_fill(struct virtio_vsock *vsock)
 
 static void virtio_vsock_reset_sock(struct sock *sk)
 {
-	lock_sock(sk);
+	/* vmci_transport.c doesn't take sk_lock here either.  At least we're
+	 * under vsock_table_lock so the sock cannot disappear while we're
+	 * executing.
+	 */
+
 	sk->sk_state = TCP_CLOSE;
 	sk->sk_err = ECONNRESET;
-	sk->sk_error_report(sk);
-	release_sock(sk);
+	sk_error_report(sk);
 }
 
 static void virtio_vsock_update_guest_cid(struct virtio_vsock *vsock)
@@ -380,7 +436,8 @@ static void virtio_vsock_event_handle(struct virtio_vsock *vsock,
 	switch (le32_to_cpu(event->id)) {
 	case VIRTIO_VSOCK_EVENT_TRANSPORT_RESET:
 		virtio_vsock_update_guest_cid(vsock);
-		vsock_for_each_connected_socket(virtio_vsock_reset_sock);
+		vsock_for_each_connected_socket(&virtio_transport.transport,
+						virtio_vsock_reset_sock);
 		break;
 	}
 }
@@ -443,6 +500,44 @@ static void virtio_vsock_rx_done(struct virtqueue *vq)
 	queue_work(virtio_vsock_workqueue, &vsock->rx_work);
 }
 
+static bool virtio_transport_can_msgzerocopy(int bufs_num)
+{
+	struct virtio_vsock *vsock;
+	bool res = false;
+
+	rcu_read_lock();
+
+	vsock = rcu_dereference(the_virtio_vsock);
+	if (vsock) {
+		struct virtqueue *vq = vsock->vqs[VSOCK_VQ_TX];
+
+		/* Check that tx queue is large enough to keep whole
+		 * data to send. This is needed, because when there is
+		 * not enough free space in the queue, current skb to
+		 * send will be reinserted to the head of tx list of
+		 * the socket to retry transmission later, so if skb
+		 * is bigger than whole queue, it will be reinserted
+		 * again and again, thus blocking other skbs to be sent.
+		 * Each page of the user provided buffer will be added
+		 * as a single buffer to the tx virtqueue, so compare
+		 * number of pages against maximum capacity of the queue.
+		 */
+		if (bufs_num <= vq->num_max)
+			res = true;
+	}
+
+	rcu_read_unlock();
+
+	return res;
+}
+
+static bool virtio_transport_msgzerocopy_allow(void)
+{
+	return true;
+}
+
+static bool virtio_transport_seqpacket_allow(u32 remote_cid);
+
 static struct virtio_transport virtio_transport = {
 	.transport = {
 		.module                   = THIS_MODULE,
@@ -469,6 +564,13 @@ static struct virtio_transport virtio_transport = {
 		.stream_is_active         = virtio_transport_stream_is_active,
 		.stream_allow             = virtio_transport_stream_allow,
 
+		.seqpacket_dequeue        = virtio_transport_seqpacket_dequeue,
+		.seqpacket_enqueue        = virtio_transport_seqpacket_enqueue,
+		.seqpacket_allow          = virtio_transport_seqpacket_allow,
+		.seqpacket_has_data       = virtio_transport_seqpacket_has_data,
+
+		.msgzerocopy_allow        = virtio_transport_msgzerocopy_allow,
+
 		.notify_poll_in           = virtio_transport_notify_poll_in,
 		.notify_poll_out          = virtio_transport_notify_poll_out,
 		.notify_recv_init         = virtio_transport_notify_recv_init,
@@ -480,11 +582,32 @@ static struct virtio_transport virtio_transport = {
 		.notify_send_pre_enqueue  = virtio_transport_notify_send_pre_enqueue,
 		.notify_send_post_enqueue = virtio_transport_notify_send_post_enqueue,
 		.notify_buffer_size       = virtio_transport_notify_buffer_size,
+		.notify_set_rcvlowat      = virtio_transport_notify_set_rcvlowat,
+
+		.unsent_bytes             = virtio_transport_unsent_bytes,
+
+		.read_skb = virtio_transport_read_skb,
 	},
 
 	.send_pkt = virtio_transport_send_pkt,
+	.can_msgzerocopy = virtio_transport_can_msgzerocopy,
 };
 
+static bool virtio_transport_seqpacket_allow(u32 remote_cid)
+{
+	struct virtio_vsock *vsock;
+	bool seqpacket_allow;
+
+	seqpacket_allow = false;
+	rcu_read_lock();
+	vsock = rcu_dereference(the_virtio_vsock);
+	if (vsock)
+		seqpacket_allow = vsock->seqpacket_allow;
+	rcu_read_unlock();
+
+	return seqpacket_allow;
+}
+
 static void virtio_transport_rx_work(struct work_struct *work)
 {
 	struct virtio_vsock *vsock =
@@ -501,8 +624,9 @@ static void virtio_transport_rx_work(struct work_struct *work)
 	do {
 		virtqueue_disable_cb(vq);
 		for (;;) {
-			struct virtio_vsock_pkt *pkt;
-			unsigned int len;
+			unsigned int len, payload_len;
+			struct virtio_vsock_hdr *hdr;
+			struct sk_buff *skb;
 
 			if (!virtio_transport_more_replies(vsock)) {
 				/* Stop rx until the device processes already
@@ -512,23 +636,31 @@ static void virtio_transport_rx_work(struct work_struct *work)
 				goto out;
 			}
 
-			pkt = virtqueue_get_buf(vq, &len);
-			if (!pkt) {
+			skb = virtqueue_get_buf(vq, &len);
+			if (!skb)
 				break;
-			}
 
 			vsock->rx_buf_nr--;
 
 			/* Drop short/long packets */
-			if (unlikely(len < sizeof(pkt->hdr) ||
-				     len > sizeof(pkt->hdr) + pkt->len)) {
-				virtio_transport_free_pkt(pkt);
+			if (unlikely(len < sizeof(*hdr) ||
+				     len > virtio_vsock_skb_len(skb))) {
+				kfree_skb(skb);
+				continue;
+			}
+
+			hdr = virtio_vsock_hdr(skb);
+			payload_len = le32_to_cpu(hdr->len);
+			if (unlikely(payload_len > len - sizeof(*hdr))) {
+				kfree_skb(skb);
 				continue;
 			}
 
-			pkt->len = len - sizeof(pkt->hdr);
-			virtio_transport_deliver_tap_pkt(pkt);
-			virtio_transport_recv_pkt(&virtio_transport, pkt);
+			if (payload_len)
+				virtio_vsock_skb_put(skb, payload_len);
+
+			virtio_transport_deliver_tap_pkt(skb);
+			virtio_transport_recv_pkt(&virtio_transport, skb);
 		}
 	} while (!virtqueue_enable_cb(vq));
 
@@ -538,62 +670,36 @@ out:
 	mutex_unlock(&vsock->rx_lock);
 }
 
-static int virtio_vsock_probe(struct virtio_device *vdev)
+static int virtio_vsock_vqs_init(struct virtio_vsock *vsock)
 {
-	vq_callback_t *callbacks[] = {
-		virtio_vsock_rx_done,
-		virtio_vsock_tx_done,
-		virtio_vsock_event_done,
-	};
-	static const char * const names[] = {
-		"rx",
-		"tx",
-		"event",
+	struct virtio_device *vdev = vsock->vdev;
+	struct virtqueue_info vqs_info[] = {
+		{ "rx", virtio_vsock_rx_done },
+		{ "tx", virtio_vsock_tx_done },
+		{ "event", virtio_vsock_event_done },
 	};
-	struct virtio_vsock *vsock = NULL;
 	int ret;
 
-	ret = mutex_lock_interruptible(&the_virtio_vsock_mutex);
-	if (ret)
-		return ret;
-
-	/* Only one virtio-vsock device per guest is supported */
-	if (rcu_dereference_protected(the_virtio_vsock,
-				lockdep_is_held(&the_virtio_vsock_mutex))) {
-		ret = -EBUSY;
-		goto out;
-	}
-
-	vsock = kzalloc(sizeof(*vsock), GFP_KERNEL);
-	if (!vsock) {
-		ret = -ENOMEM;
-		goto out;
-	}
+	mutex_lock(&vsock->rx_lock);
+	vsock->rx_buf_nr = 0;
+	vsock->rx_buf_max_nr = 0;
+	mutex_unlock(&vsock->rx_lock);
 
-	vsock->vdev = vdev;
+	atomic_set(&vsock->queued_replies, 0);
 
-	ret = virtio_find_vqs(vsock->vdev, VSOCK_VQ_MAX,
-			      vsock->vqs, callbacks, names,
-			      NULL);
+	ret = virtio_find_vqs(vdev, VSOCK_VQ_MAX, vsock->vqs, vqs_info, NULL);
 	if (ret < 0)
-		goto out;
+		return ret;
 
 	virtio_vsock_update_guest_cid(vsock);
 
-	vsock->rx_buf_nr = 0;
-	vsock->rx_buf_max_nr = 0;
-	atomic_set(&vsock->queued_replies, 0);
+	virtio_device_ready(vdev);
 
-	mutex_init(&vsock->tx_lock);
-	mutex_init(&vsock->rx_lock);
-	mutex_init(&vsock->event_lock);
-	spin_lock_init(&vsock->send_pkt_list_lock);
-	INIT_LIST_HEAD(&vsock->send_pkt_list);
-	INIT_WORK(&vsock->rx_work, virtio_transport_rx_work);
-	INIT_WORK(&vsock->tx_work, virtio_transport_tx_work);
-	INIT_WORK(&vsock->event_work, virtio_transport_event_work);
-	INIT_WORK(&vsock->send_pkt_work, virtio_transport_send_pkt_work);
+	return 0;
+}
 
+static void virtio_vsock_vqs_start(struct virtio_vsock *vsock)
+{
 	mutex_lock(&vsock->tx_lock);
 	vsock->tx_run = true;
 	mutex_unlock(&vsock->tx_lock);
@@ -608,34 +714,29 @@ static int virtio_vsock_probe(struct virtio_device *vdev)
 	vsock->event_run = true;
 	mutex_unlock(&vsock->event_lock);
 
-	vdev->priv = vsock;
-	rcu_assign_pointer(the_virtio_vsock, vsock);
-
-	mutex_unlock(&the_virtio_vsock_mutex);
-	return 0;
-
-out:
-	kfree(vsock);
-	mutex_unlock(&the_virtio_vsock_mutex);
-	return ret;
+	/* virtio_transport_send_pkt() can queue packets once
+	 * the_virtio_vsock is set, but they won't be processed until
+	 * vsock->tx_run is set to true. We queue vsock->send_pkt_work
+	 * when initialization finishes to send those packets queued
+	 * earlier.
+	 * We don't need to queue the other workers (rx, event) because
+	 * as long as we don't fill the queues with empty buffers, the
+	 * host can't send us any notification.
+	 */
+	queue_work(virtio_vsock_workqueue, &vsock->send_pkt_work);
 }
 
-static void virtio_vsock_remove(struct virtio_device *vdev)
+static void virtio_vsock_vqs_del(struct virtio_vsock *vsock)
 {
-	struct virtio_vsock *vsock = vdev->priv;
-	struct virtio_vsock_pkt *pkt;
-
-	mutex_lock(&the_virtio_vsock_mutex);
-
-	vdev->priv = NULL;
-	rcu_assign_pointer(the_virtio_vsock, NULL);
-	synchronize_rcu();
+	struct virtio_device *vdev = vsock->vdev;
+	struct sk_buff *skb;
 
-	/* Reset all connected sockets when the device disappear */
-	vsock_for_each_connected_socket(virtio_vsock_reset_sock);
+	/* Reset all connected sockets when the VQs disappear */
+	vsock_for_each_connected_socket(&virtio_transport.transport,
+					virtio_vsock_reset_sock);
 
 	/* Stop all work handlers to make sure no one is accessing the device,
-	 * so we can safely call vdev->config->reset().
+	 * so we can safely call virtio_reset_device().
 	 */
 	mutex_lock(&vsock->rx_lock);
 	vsock->rx_run = false;
@@ -652,29 +753,95 @@ static void virtio_vsock_remove(struct virtio_device *vdev)
 	/* Flush all device writes and interrupts, device will not use any
 	 * more buffers.
 	 */
-	vdev->config->reset(vdev);
+	virtio_reset_device(vdev);
 
 	mutex_lock(&vsock->rx_lock);
-	while ((pkt = virtqueue_detach_unused_buf(vsock->vqs[VSOCK_VQ_RX])))
-		virtio_transport_free_pkt(pkt);
+	while ((skb = virtqueue_detach_unused_buf(vsock->vqs[VSOCK_VQ_RX])))
+		kfree_skb(skb);
 	mutex_unlock(&vsock->rx_lock);
 
 	mutex_lock(&vsock->tx_lock);
-	while ((pkt = virtqueue_detach_unused_buf(vsock->vqs[VSOCK_VQ_TX])))
-		virtio_transport_free_pkt(pkt);
+	while ((skb = virtqueue_detach_unused_buf(vsock->vqs[VSOCK_VQ_TX])))
+		kfree_skb(skb);
 	mutex_unlock(&vsock->tx_lock);
 
-	spin_lock_bh(&vsock->send_pkt_list_lock);
-	while (!list_empty(&vsock->send_pkt_list)) {
-		pkt = list_first_entry(&vsock->send_pkt_list,
-				       struct virtio_vsock_pkt, list);
-		list_del(&pkt->list);
-		virtio_transport_free_pkt(pkt);
-	}
-	spin_unlock_bh(&vsock->send_pkt_list_lock);
+	virtio_vsock_skb_queue_purge(&vsock->send_pkt_queue);
 
 	/* Delete virtqueues and flush outstanding callbacks if any */
 	vdev->config->del_vqs(vdev);
+}
+
+static int virtio_vsock_probe(struct virtio_device *vdev)
+{
+	struct virtio_vsock *vsock = NULL;
+	int ret;
+	int i;
+
+	ret = mutex_lock_interruptible(&the_virtio_vsock_mutex);
+	if (ret)
+		return ret;
+
+	/* Only one virtio-vsock device per guest is supported */
+	if (rcu_dereference_protected(the_virtio_vsock,
+				lockdep_is_held(&the_virtio_vsock_mutex))) {
+		ret = -EBUSY;
+		goto out;
+	}
+
+	vsock = kzalloc(sizeof(*vsock), GFP_KERNEL);
+	if (!vsock) {
+		ret = -ENOMEM;
+		goto out;
+	}
+
+	vsock->vdev = vdev;
+
+
+	mutex_init(&vsock->tx_lock);
+	mutex_init(&vsock->rx_lock);
+	mutex_init(&vsock->event_lock);
+	skb_queue_head_init(&vsock->send_pkt_queue);
+	INIT_WORK(&vsock->rx_work, virtio_transport_rx_work);
+	INIT_WORK(&vsock->tx_work, virtio_transport_tx_work);
+	INIT_WORK(&vsock->event_work, virtio_transport_event_work);
+	INIT_WORK(&vsock->send_pkt_work, virtio_transport_send_pkt_work);
+
+	if (virtio_has_feature(vdev, VIRTIO_VSOCK_F_SEQPACKET))
+		vsock->seqpacket_allow = true;
+
+	vdev->priv = vsock;
+
+	ret = virtio_vsock_vqs_init(vsock);
+	if (ret < 0)
+		goto out;
+
+	for (i = 0; i < ARRAY_SIZE(vsock->out_sgs); i++)
+		vsock->out_sgs[i] = &vsock->out_bufs[i];
+
+	rcu_assign_pointer(the_virtio_vsock, vsock);
+	virtio_vsock_vqs_start(vsock);
+
+	mutex_unlock(&the_virtio_vsock_mutex);
+
+	return 0;
+
+out:
+	kfree(vsock);
+	mutex_unlock(&the_virtio_vsock_mutex);
+	return ret;
+}
+
+static void virtio_vsock_remove(struct virtio_device *vdev)
+{
+	struct virtio_vsock *vsock = vdev->priv;
+
+	mutex_lock(&the_virtio_vsock_mutex);
+
+	vdev->priv = NULL;
+	rcu_assign_pointer(the_virtio_vsock, NULL);
+	synchronize_rcu();
+
+	virtio_vsock_vqs_del(vsock);
 
 	/* Other works can be queued before 'config->del_vqs()', so we flush
 	 * all works before to free the vsock object to avoid use after free.
@@ -689,29 +856,77 @@ static void virtio_vsock_remove(struct virtio_device *vdev)
 	kfree(vsock);
 }
 
+#ifdef CONFIG_PM_SLEEP
+static int virtio_vsock_freeze(struct virtio_device *vdev)
+{
+	struct virtio_vsock *vsock = vdev->priv;
+
+	mutex_lock(&the_virtio_vsock_mutex);
+
+	rcu_assign_pointer(the_virtio_vsock, NULL);
+	synchronize_rcu();
+
+	virtio_vsock_vqs_del(vsock);
+
+	mutex_unlock(&the_virtio_vsock_mutex);
+
+	return 0;
+}
+
+static int virtio_vsock_restore(struct virtio_device *vdev)
+{
+	struct virtio_vsock *vsock = vdev->priv;
+	int ret;
+
+	mutex_lock(&the_virtio_vsock_mutex);
+
+	/* Only one virtio-vsock device per guest is supported */
+	if (rcu_dereference_protected(the_virtio_vsock,
+				lockdep_is_held(&the_virtio_vsock_mutex))) {
+		ret = -EBUSY;
+		goto out;
+	}
+
+	ret = virtio_vsock_vqs_init(vsock);
+	if (ret < 0)
+		goto out;
+
+	rcu_assign_pointer(the_virtio_vsock, vsock);
+	virtio_vsock_vqs_start(vsock);
+
+out:
+	mutex_unlock(&the_virtio_vsock_mutex);
+	return ret;
+}
+#endif /* CONFIG_PM_SLEEP */
+
 static struct virtio_device_id id_table[] = {
 	{ VIRTIO_ID_VSOCK, VIRTIO_DEV_ANY_ID },
 	{ 0 },
 };
 
 static unsigned int features[] = {
+	VIRTIO_VSOCK_F_SEQPACKET
 };
 
 static struct virtio_driver virtio_vsock_driver = {
 	.feature_table = features,
 	.feature_table_size = ARRAY_SIZE(features),
 	.driver.name = KBUILD_MODNAME,
-	.driver.owner = THIS_MODULE,
 	.id_table = id_table,
 	.probe = virtio_vsock_probe,
 	.remove = virtio_vsock_remove,
+#ifdef CONFIG_PM_SLEEP
+	.freeze = virtio_vsock_freeze,
+	.restore = virtio_vsock_restore,
+#endif
 };
 
 static int __init virtio_vsock_init(void)
 {
 	int ret;
 
-	virtio_vsock_workqueue = alloc_workqueue("virtio_vsock", 0, 0);
+	virtio_vsock_workqueue = alloc_workqueue("virtio_vsock", WQ_PERCPU, 0);
 	if (!virtio_vsock_workqueue)
 		return -ENOMEM;