Virtualization - Apr 2016 - [RFC v5 0/5] Add virtio transport for AF_VSOCK

This series is based on Michael Tsirkin's vhost branch (v4.5-rc6).

I'm about to process Claudio Imbrenda's locking fixes for virtio-vsock
but
first I want to share the latest version of the code.  Several people are
playing with vsock now so sharing the latest code should avoid duplicate work.

v5:
 * Transport reset event for live migration support
 * Reorder virtqueues, drop unused ctrl virtqueue
 * Switch to a free virtio device ID
 * More small changes, see patches for individual items

v4:
 * Addressed code review comments from Alex Bennee
 * MAINTAINERS file entries for new files
 * Trace events instead of pr_debug()
 * RST packet is sent when there is no listen socket
 * Allow guest->host connections again (began discussing netfilter support
with
   Matt Benjamin instead of hard-coding security policy in virtio-vsock code)
 * Many checkpatch.pl cleanups (will be 100% clean in v5)

v3:
 * Remove unnecessary 3-way handshake, just do REQUEST/RESPONSE instead
   of REQUEST/RESPONSE/ACK
 * Remove SOCK_DGRAM support and focus on SOCK_STREAM first
   (also drop v2 Patch 1, it's only needed for SOCK_DGRAM)
 * Only allow host->guest connections (same security model as latest
   VMware)
 * Don't put vhost vsock driver into staging
 * Add missing Kconfig dependencies (Arnd Bergmann <arnd at arndb.de>)
 * Remove unneeded variable used to store return value
   (Fengguang Wu <fengguang.wu at intel.com> and Julia Lawall
   <julia.lawall at lip6.fr>)

v2:
 * Rebased onto Linux v4.4-rc2
 * vhost: Refuse to assign reserved CIDs
 * vhost: Refuse guest CID if already in use
 * vhost: Only accept correctly addressed packets (no spoofing!)
 * vhost: Support flexible rx/tx descriptor layout
 * vhost: Add missing total_tx_buf decrement
 * virtio_transport: Fix total_tx_buf accounting
 * virtio_transport: Add virtio_transport global mutex to prevent races
 * common: Notify other side of SOCK_STREAM disconnect (fixes shutdown
   semantics)
 * common: Avoid recursive mutex_lock(tx_lock) for write_space (fixes deadlock)
 * common: Define VIRTIO_VSOCK_TYPE_STREAM/DGRAM hardware interface constants
 * common: Define VIRTIO_VSOCK_SHUTDOWN_RCV/SEND hardware interface constants
 * common: Fix peer_buf_alloc inheritance on child socket

This patch series adds a virtio transport for AF_VSOCK (net/vmw_vsock/).
AF_VSOCK is designed for communication between virtual machines and
hypervisors.  It is currently only implemented for VMware's VMCI transport.

Most of the work was done by Asias He and Gerd Hoffmann a while back.  I have
picked up the series again.

The QEMU userspace changes are here:
https://github.com/stefanha/qemu/commits/vsock

Why virtio-vsock?
-----------------
Guest<->host communication is currently done over the virtio-serial
device.
This makes it hard to port sockets API-based applications and is limited to
static ports.

virtio-vsock uses the sockets API so that applications can rely on familiar
SOCK_STREAM semantics.  Applications on the host can easily connect to guest
agents because the sockets API allows multiple connections to a listen socket
(unlike virtio-serial).  This simplifies the guest<->host communication
and
eliminates the need for extra processes on the host to arbitrate virtio-serial
ports.

Overview
--------
This series adds 3 pieces:

1. virtio_transport_common.ko - core virtio vsock code that uses vsock.ko

2. virtio_transport.ko - guest driver

3. drivers/vhost/vsock.ko - host driver

Howto
-----
The following kernel options are needed:
  CONFIG_VSOCKETS=y
  CONFIG_VIRTIO_VSOCKETS=y
  CONFIG_VIRTIO_VSOCKETS_COMMON=y
  CONFIG_VHOST_VSOCK=m

Launch QEMU as follows:
  # qemu ... -device vhost-vsock-pci,id=vhost-vsock-pci0,guest-cid=3

Guest and host can communicate via AF_VSOCK sockets.  The host's CID
(address)
is 2 and the guest must be assigned a CID (3 in the example above).

Asias He (4):
  VSOCK: Introduce virtio_vsock_common.ko
  VSOCK: Introduce virtio_transport.ko
  VSOCK: Introduce vhost_vsock.ko
  VSOCK: Add Makefile and Kconfig

Stefan Hajnoczi (1):
  VSOCK: transport-specific vsock_transport functions

 MAINTAINERS                                        |  13 +
 drivers/vhost/Kconfig                              |  15 +
 drivers/vhost/Makefile                             |   4 +
 drivers/vhost/vsock.c                              | 694 +++++++++++++++++
 drivers/vhost/vsock.h                              |   5 +
 include/linux/virtio_vsock.h                       | 167 ++++
 include/net/af_vsock.h                             |   3 +
 .../trace/events/vsock_virtio_transport_common.h   | 144 ++++
 include/uapi/linux/virtio_ids.h                    |   1 +
 include/uapi/linux/virtio_vsock.h                  |  94 +++
 net/vmw_vsock/Kconfig                              |  19 +
 net/vmw_vsock/Makefile                             |   2 +
 net/vmw_vsock/af_vsock.c                           |   9 +
 net/vmw_vsock/virtio_transport.c                   | 584 ++++++++++++++
 net/vmw_vsock/virtio_transport_common.c            | 838 +++++++++++++++++++++
 15 files changed, 2592 insertions(+)
 create mode 100644 drivers/vhost/vsock.c
 create mode 100644 drivers/vhost/vsock.h
 create mode 100644 include/linux/virtio_vsock.h
 create mode 100644 include/trace/events/vsock_virtio_transport_common.h
 create mode 100644 include/uapi/linux/virtio_vsock.h
 create mode 100644 net/vmw_vsock/virtio_transport.c
 create mode 100644 net/vmw_vsock/virtio_transport_common.c

-- 
2.5.5

struct vsock_transport contains function pointers called by AF_VSOCK
core code.  The transport may want its own transport-specific function
pointers and they can be added after struct vsock_transport.

Allow the transport to fetch vsock_transport.  It can downcast it to
access transport-specific function pointers.

The virtio transport will use this.

Signed-off-by: Stefan Hajnoczi <stefanha at redhat.com>
---
 include/net/af_vsock.h   | 3 +++
 net/vmw_vsock/af_vsock.c | 9 +++++++++
 2 files changed, 12 insertions(+)

diff --git a/include/net/af_vsock.h b/include/net/af_vsock.h
index e9eb2d6..23f5525 100644
--- a/include/net/af_vsock.h
+++ b/include/net/af_vsock.h
@@ -165,6 +165,9 @@ static inline int vsock_core_init(const struct
vsock_transport *t)
 }
 void vsock_core_exit(void);
 
+/* The transport may downcast this to access transport-specific functions */
+const struct vsock_transport *vsock_core_get_transport(void);
+
 /**** UTILS ****/
 
 void vsock_release_pending(struct sock *pending);
diff --git a/net/vmw_vsock/af_vsock.c b/net/vmw_vsock/af_vsock.c
index bbe65dc..1e5f5ed 100644
--- a/net/vmw_vsock/af_vsock.c
+++ b/net/vmw_vsock/af_vsock.c
@@ -1987,6 +1987,15 @@ void vsock_core_exit(void)
 }
 EXPORT_SYMBOL_GPL(vsock_core_exit);
 
+const struct vsock_transport *vsock_core_get_transport(void)
+{
+	/* vsock_register_mutex not taken since only the transport uses this
+	 * function and only while registered.
+	 */
+	return transport;
+}
+EXPORT_SYMBOL_GPL(vsock_core_get_transport);
+
 MODULE_AUTHOR("VMware, Inc.");
 MODULE_DESCRIPTION("VMware Virtual Socket Family");
 MODULE_VERSION("1.0.1.0-k");
-- 
2.5.5

From: Asias He <asias at redhat.com>

This module contains the common code and header files for the following
virtio_transporto and vhost_vsock kernel modules.

Signed-off-by: Asias He <asias at redhat.com>
Signed-off-by: Stefan Hajnoczi <stefanha at redhat.com>
---
v5:
 * Add event virtqueue, struct virtio_vsock_event, and transport reset
   event
 * Reorder virtqueue indices: rx, tx, event
 * Drop unused virtqueue_pairs config field
 * Drop unused ctrl virtqueue
 * Switch to a free virtio device ID, the previous one was reserved
v4:
 * Add MAINTAINERS file entry
 * checkpatch.pl cleanups
 * linux_vsock.h: drop wrong copy-pasted license header
 * Move tx sock refcounting to virtio_transport_alloc/free_pkt() to fix
   leaks in error paths
 * Add send_pkt_no_sock() to send RST packets with no listen socket
 * Rename per-socket state from virtio_transport to virtio_vsock_sock
 * Move send_pkt_ops to new virtio_transport struct
 * Drop dumppkt function, packet capture will be added in the future
 * Drop empty virtio_transport_dec_tx_pkt()
 * Allow guest->host connections again
 * Use trace events instead of pr_debug()
v3:
 * Remove unnecessary 3-way handshake, just do REQUEST/RESPONSE instead
   of REQUEST/RESPONSE/ACK
 * Remove SOCK_DGRAM support and focus on SOCK_STREAM first
 * Only allow host->guest connections (same security model as latest
   VMware)
v2:
 * Fix peer_buf_alloc inheritance on child socket
 * Notify other side of SOCK_STREAM disconnect (fixes shutdown
   semantics)
 * Avoid recursive mutex_lock(tx_lock) for write_space (fixes deadlock)
 * Define VIRTIO_VSOCK_TYPE_STREAM/DGRAM hardware interface constants
 * Define VIRTIO_VSOCK_SHUTDOWN_RCV/SEND hardware interface constants
---
 MAINTAINERS                                        |  10 +
 include/linux/virtio_vsock.h                       | 167 ++++
 .../trace/events/vsock_virtio_transport_common.h   | 144 ++++
 include/uapi/linux/virtio_ids.h                    |   1 +
 include/uapi/linux/virtio_vsock.h                  |  94 +++
 net/vmw_vsock/virtio_transport_common.c            | 838 +++++++++++++++++++++
 6 files changed, 1254 insertions(+)
 create mode 100644 include/linux/virtio_vsock.h
 create mode 100644 include/trace/events/vsock_virtio_transport_common.h
 create mode 100644 include/uapi/linux/virtio_vsock.h
 create mode 100644 net/vmw_vsock/virtio_transport_common.c

diff --git a/MAINTAINERS b/MAINTAINERS
index da3e4d8..fb9c47a 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -11602,6 +11602,16 @@ S:	Maintained
 F:	drivers/media/v4l2-core/videobuf2-*
 F:	include/media/videobuf2-*
 
+VIRTIO AND VHOST VSOCK DRIVER
+M:	Stefan Hajnoczi <stefanha at redhat.com>
+L:	kvm at vger.kernel.org
+L:	virtualization at lists.linux-foundation.org
+L:	netdev at vger.kernel.org
+S:	Maintained
+F:	include/linux/virtio_vsock.h
+F:	include/uapi/linux/virtio_vsock.h
+F:	net/vmw_vsock/virtio_transport_common.c
+
 VIRTUAL SERIO DEVICE DRIVER
 M:	Stephen Chandler Paul <thatslyude at gmail.com>
 S:	Maintained
diff --git a/include/linux/virtio_vsock.h b/include/linux/virtio_vsock.h
new file mode 100644
index 0000000..4c3d8e6
--- /dev/null
+++ b/include/linux/virtio_vsock.h
@@ -0,0 +1,167 @@
+#ifndef _LINUX_VIRTIO_VSOCK_H
+#define _LINUX_VIRTIO_VSOCK_H
+
+#include <uapi/linux/virtio_vsock.h>
+#include <linux/socket.h>
+#include <net/sock.h>
+#include <net/af_vsock.h>
+
+#define VIRTIO_VSOCK_DEFAULT_MIN_BUF_SIZE	128
+#define VIRTIO_VSOCK_DEFAULT_BUF_SIZE		(1024 * 256)
+#define VIRTIO_VSOCK_DEFAULT_MAX_BUF_SIZE	(1024 * 256)
+#define VIRTIO_VSOCK_DEFAULT_RX_BUF_SIZE	(1024 * 4)
+#define VIRTIO_VSOCK_MAX_BUF_SIZE		0xFFFFFFFFUL
+#define VIRTIO_VSOCK_MAX_PKT_BUF_SIZE		(1024 * 64)
+#define VIRTIO_VSOCK_MAX_TX_BUF_SIZE		(1024 * 1024 * 16)
+#define VIRTIO_VSOCK_MAX_DGRAM_SIZE		(1024 * 64)
+
+enum {
+	VSOCK_VQ_RX     = 0, /* for host to guest data */
+	VSOCK_VQ_TX     = 1, /* for guest to host data */
+	VSOCK_VQ_EVENT  = 2,
+	VSOCK_VQ_MAX    = 3,
+};
+
+/* Per-socket state (accessed via vsk->trans) */
+struct virtio_vsock_sock {
+	struct vsock_sock *vsk;
+
+	/* Protected by lock_sock(sk_vsock(trans->vsk)) */
+	u32 buf_size;
+	u32 buf_size_min;
+	u32 buf_size_max;
+
+	struct mutex tx_lock;
+	struct mutex rx_lock;
+
+	/* Protected by tx_lock */
+	u32 tx_cnt;
+	u32 buf_alloc;
+	u32 peer_fwd_cnt;
+	u32 peer_buf_alloc;
+
+	/* Protected by rx_lock */
+	u32 fwd_cnt;
+	u32 rx_bytes;
+	struct list_head rx_queue;
+};
+
+struct virtio_vsock_pkt {
+	struct virtio_vsock_hdr	hdr;
+	struct work_struct work;
+	struct list_head list;
+	void *buf;
+	u32 len;
+	u32 off;
+};
+
+struct virtio_vsock_pkt_info {
+	u32 remote_cid, remote_port;
+	struct msghdr *msg;
+	u32 pkt_len;
+	u16 type;
+	u16 op;
+	u32 flags;
+};
+
+struct virtio_transport {
+	/* This must be the first field */
+	struct vsock_transport transport;
+
+	/* Send packet for a specific socket */
+	int (*send_pkt)(struct vsock_sock *vsk,
+			struct virtio_vsock_pkt_info *info);
+
+	/* Send packet without a socket (e.g. RST).  Prefer send_pkt() over
+	 * send_pkt_no_sock() when a socket exists.
+	 */
+	int (*send_pkt_no_sock)(struct virtio_vsock_pkt *pkt);
+};
+
+struct virtio_vsock_pkt *
+virtio_transport_alloc_pkt(struct virtio_vsock_pkt_info *info,
+			   size_t len,
+			   u32 src_cid,
+			   u32 src_port,
+			   u32 dst_cid,
+			   u32 dst_port);
+ssize_t
+virtio_transport_stream_dequeue(struct vsock_sock *vsk,
+				struct msghdr *msg,
+				size_t len,
+				int type);
+int
+virtio_transport_dgram_dequeue(struct vsock_sock *vsk,
+			       struct msghdr *msg,
+			       size_t len, int flags);
+
+s64 virtio_transport_stream_has_data(struct vsock_sock *vsk);
+s64 virtio_transport_stream_has_space(struct vsock_sock *vsk);
+
+int virtio_transport_do_socket_init(struct vsock_sock *vsk,
+				 struct vsock_sock *psk);
+u64 virtio_transport_get_buffer_size(struct vsock_sock *vsk);
+u64 virtio_transport_get_min_buffer_size(struct vsock_sock *vsk);
+u64 virtio_transport_get_max_buffer_size(struct vsock_sock *vsk);
+void virtio_transport_set_buffer_size(struct vsock_sock *vsk, u64 val);
+void virtio_transport_set_min_buffer_size(struct vsock_sock *vsk, u64 val);
+void virtio_transport_set_max_buffer_size(struct vsock_sock *vs, u64 val);
+int
+virtio_transport_notify_poll_in(struct vsock_sock *vsk,
+				size_t target,
+				bool *data_ready_now);
+int
+virtio_transport_notify_poll_out(struct vsock_sock *vsk,
+				 size_t target,
+				 bool *space_available_now);
+
+int virtio_transport_notify_recv_init(struct vsock_sock *vsk,
+	size_t target, struct vsock_transport_recv_notify_data *data);
+int virtio_transport_notify_recv_pre_block(struct vsock_sock *vsk,
+	size_t target, struct vsock_transport_recv_notify_data *data);
+int virtio_transport_notify_recv_pre_dequeue(struct vsock_sock *vsk,
+	size_t target, struct vsock_transport_recv_notify_data *data);
+int virtio_transport_notify_recv_post_dequeue(struct vsock_sock *vsk,
+	size_t target, ssize_t copied, bool data_read,
+	struct vsock_transport_recv_notify_data *data);
+int virtio_transport_notify_send_init(struct vsock_sock *vsk,
+	struct vsock_transport_send_notify_data *data);
+int virtio_transport_notify_send_pre_block(struct vsock_sock *vsk,
+	struct vsock_transport_send_notify_data *data);
+int virtio_transport_notify_send_pre_enqueue(struct vsock_sock *vsk,
+	struct vsock_transport_send_notify_data *data);
+int virtio_transport_notify_send_post_enqueue(struct vsock_sock *vsk,
+	ssize_t written, struct vsock_transport_send_notify_data *data);
+
+u64 virtio_transport_stream_rcvhiwat(struct vsock_sock *vsk);
+bool virtio_transport_stream_is_active(struct vsock_sock *vsk);
+bool virtio_transport_stream_allow(u32 cid, u32 port);
+int virtio_transport_dgram_bind(struct vsock_sock *vsk,
+				struct sockaddr_vm *addr);
+bool virtio_transport_dgram_allow(u32 cid, u32 port);
+
+int virtio_transport_connect(struct vsock_sock *vsk);
+
+int virtio_transport_shutdown(struct vsock_sock *vsk, int mode);
+
+void virtio_transport_release(struct vsock_sock *vsk);
+
+ssize_t
+virtio_transport_stream_enqueue(struct vsock_sock *vsk,
+				struct msghdr *msg,
+				size_t len);
+int
+virtio_transport_dgram_enqueue(struct vsock_sock *vsk,
+			       struct sockaddr_vm *remote_addr,
+			       struct msghdr *msg,
+			       size_t len);
+
+void virtio_transport_destruct(struct vsock_sock *vsk);
+
+void virtio_transport_recv_pkt(struct virtio_vsock_pkt *pkt);
+void virtio_transport_free_pkt(struct virtio_vsock_pkt *pkt);
+void virtio_transport_inc_tx_pkt(struct virtio_vsock_sock *vvs, struct
virtio_vsock_pkt *pkt);
+u32 virtio_transport_get_credit(struct virtio_vsock_sock *vvs, u32 wanted);
+void virtio_transport_put_credit(struct virtio_vsock_sock *vvs, u32 credit);
+
+#endif /* _LINUX_VIRTIO_VSOCK_H */
diff --git a/include/trace/events/vsock_virtio_transport_common.h
b/include/trace/events/vsock_virtio_transport_common.h
new file mode 100644
index 0000000..b7f1d62
--- /dev/null
+++ b/include/trace/events/vsock_virtio_transport_common.h
@@ -0,0 +1,144 @@
+#undef TRACE_SYSTEM
+#define TRACE_SYSTEM vsock
+
+#if !defined(_TRACE_VSOCK_VIRTIO_TRANSPORT_COMMON_H) || \
+    defined(TRACE_HEADER_MULTI_READ)
+#define _TRACE_VSOCK_VIRTIO_TRANSPORT_COMMON_H
+
+#include <linux/tracepoint.h>
+
+TRACE_DEFINE_ENUM(VIRTIO_VSOCK_TYPE_STREAM);
+
+#define show_type(val) \
+	__print_symbolic(val, { VIRTIO_VSOCK_TYPE_STREAM, "STREAM" })
+
+TRACE_DEFINE_ENUM(VIRTIO_VSOCK_OP_INVALID);
+TRACE_DEFINE_ENUM(VIRTIO_VSOCK_OP_REQUEST);
+TRACE_DEFINE_ENUM(VIRTIO_VSOCK_OP_RESPONSE);
+TRACE_DEFINE_ENUM(VIRTIO_VSOCK_OP_RST);
+TRACE_DEFINE_ENUM(VIRTIO_VSOCK_OP_SHUTDOWN);
+TRACE_DEFINE_ENUM(VIRTIO_VSOCK_OP_RW);
+TRACE_DEFINE_ENUM(VIRTIO_VSOCK_OP_CREDIT_UPDATE);
+TRACE_DEFINE_ENUM(VIRTIO_VSOCK_OP_CREDIT_REQUEST);
+
+#define show_op(val) \
+	__print_symbolic(val, \
+			 { VIRTIO_VSOCK_OP_INVALID, "INVALID" }, \
+			 { VIRTIO_VSOCK_OP_REQUEST, "REQUEST" }, \
+			 { VIRTIO_VSOCK_OP_RESPONSE, "RESPONSE" }, \
+			 { VIRTIO_VSOCK_OP_RST, "RST" }, \
+			 { VIRTIO_VSOCK_OP_SHUTDOWN, "SHUTDOWN" }, \
+			 { VIRTIO_VSOCK_OP_RW, "RW" }, \
+			 { VIRTIO_VSOCK_OP_CREDIT_UPDATE, "CREDIT_UPDATE" }, \
+			 { VIRTIO_VSOCK_OP_CREDIT_REQUEST, "CREDIT_REQUEST" })
+
+TRACE_EVENT(virtio_transport_alloc_pkt,
+	TP_PROTO(
+		 __u32 src_cid, __u32 src_port,
+		 __u32 dst_cid, __u32 dst_port,
+		 __u32 len,
+		 __u16 type,
+		 __u16 op,
+		 __u32 flags
+	),
+	TP_ARGS(
+		src_cid, src_port,
+		dst_cid, dst_port,
+		len,
+		type,
+		op,
+		flags
+	),
+	TP_STRUCT__entry(
+		__field(__u32, src_cid)
+		__field(__u32, src_port)
+		__field(__u32, dst_cid)
+		__field(__u32, dst_port)
+		__field(__u32, len)
+		__field(__u16, type)
+		__field(__u16, op)
+		__field(__u32, flags)
+	),
+	TP_fast_assign(
+		__entry->src_cid = src_cid;
+		__entry->src_port = src_port;
+		__entry->dst_cid = dst_cid;
+		__entry->dst_port = dst_port;
+		__entry->len = len;
+		__entry->type = type;
+		__entry->op = op;
+		__entry->flags = flags;
+	),
+	TP_printk("%u:%u -> %u:%u len=%u type=%s op=%s flags=%#x",
+		  __entry->src_cid, __entry->src_port,
+		  __entry->dst_cid, __entry->dst_port,
+		  __entry->len,
+		  show_type(__entry->type),
+		  show_op(__entry->op),
+		  __entry->flags)
+);
+
+TRACE_EVENT(virtio_transport_recv_pkt,
+	TP_PROTO(
+		 __u32 src_cid, __u32 src_port,
+		 __u32 dst_cid, __u32 dst_port,
+		 __u32 len,
+		 __u16 type,
+		 __u16 op,
+		 __u32 flags,
+		 __u32 buf_alloc,
+		 __u32 fwd_cnt
+	),
+	TP_ARGS(
+		src_cid, src_port,
+		dst_cid, dst_port,
+		len,
+		type,
+		op,
+		flags,
+		buf_alloc,
+		fwd_cnt
+	),
+	TP_STRUCT__entry(
+		__field(__u32, src_cid)
+		__field(__u32, src_port)
+		__field(__u32, dst_cid)
+		__field(__u32, dst_port)
+		__field(__u32, len)
+		__field(__u16, type)
+		__field(__u16, op)
+		__field(__u32, flags)
+		__field(__u32, buf_alloc)
+		__field(__u32, fwd_cnt)
+	),
+	TP_fast_assign(
+		__entry->src_cid = src_cid;
+		__entry->src_port = src_port;
+		__entry->dst_cid = dst_cid;
+		__entry->dst_port = dst_port;
+		__entry->len = len;
+		__entry->type = type;
+		__entry->op = op;
+		__entry->flags = flags;
+		__entry->buf_alloc = buf_alloc;
+		__entry->fwd_cnt = fwd_cnt;
+	),
+	TP_printk("%u:%u -> %u:%u len=%u type=%s op=%s flags=%#x "
+		  "buf_alloc=%u fwd_cnt=%u",
+		  __entry->src_cid, __entry->src_port,
+		  __entry->dst_cid, __entry->dst_port,
+		  __entry->len,
+		  show_type(__entry->type),
+		  show_op(__entry->op),
+		  __entry->flags,
+		  __entry->buf_alloc,
+		  __entry->fwd_cnt)
+);
+
+#endif /* _TRACE_VSOCK_VIRTIO_TRANSPORT_COMMON_H */
+
+#undef TRACE_INCLUDE_FILE
+#define TRACE_INCLUDE_FILE vsock_virtio_transport_common
+
+/* This part must be outside protection */
+#include <trace/define_trace.h>
diff --git a/include/uapi/linux/virtio_ids.h b/include/uapi/linux/virtio_ids.h
index 77925f5..3228d58 100644
--- a/include/uapi/linux/virtio_ids.h
+++ b/include/uapi/linux/virtio_ids.h
@@ -41,5 +41,6 @@
 #define VIRTIO_ID_CAIF	       12 /* Virtio caif */
 #define VIRTIO_ID_GPU          16 /* virtio GPU */
 #define VIRTIO_ID_INPUT        18 /* virtio input */
+#define VIRTIO_ID_VSOCK        19 /* virtio vsock transport */
 
 #endif /* _LINUX_VIRTIO_IDS_H */
diff --git a/include/uapi/linux/virtio_vsock.h
b/include/uapi/linux/virtio_vsock.h
new file mode 100644
index 0000000..12946ab
--- /dev/null
+++ b/include/uapi/linux/virtio_vsock.h
@@ -0,0 +1,94 @@
+/*
+ * This header, excluding the #ifdef __KERNEL__ part, is BSD licensed so
+ * anyone can use the definitions to implement compatible drivers/servers:
+ *
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ * 1. Redistributions of source code must retain the above copyright
+ *    notice, this list of conditions and the following disclaimer.
+ * 2. Redistributions in binary form must reproduce the above copyright
+ *    notice, this list of conditions and the following disclaimer in the
+ *    documentation and/or other materials provided with the distribution.
+ * 3. Neither the name of IBM nor the names of its contributors
+ *    may be used to endorse or promote products derived from this software
+ *    without specific prior written permission.
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS ``AS
IS''
+ * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED.  IN NO EVENT SHALL IBM OR CONTRIBUTORS BE LIABLE
+ * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+ * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
+ * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
+ * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+ * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
+ * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
+ * SUCH DAMAGE.
+ *
+ * Copyright (C) Red Hat, Inc., 2013-2015
+ * Copyright (C) Asias He <asias at redhat.com>, 2013
+ * Copyright (C) Stefan Hajnoczi <stefanha at redhat.com>, 2015
+ */
+
+#ifndef _UAPI_LINUX_VIRTIO_VSOCK_H
+#define _UAPI_LINUX_VIRTIO_VOSCK_H
+
+#include <linux/types.h>
+#include <linux/virtio_ids.h>
+#include <linux/virtio_config.h>
+
+struct virtio_vsock_config {
+	__le32 guest_cid;
+};
+
+enum virtio_vsock_event_id {
+	VIRTIO_VSOCK_EVENT_TRANSPORT_RESET = 0,
+};
+
+struct virtio_vsock_event {
+	__le32 id;
+};
+
+struct virtio_vsock_hdr {
+	__le32	src_cid;
+	__le32	src_port;
+	__le32	dst_cid;
+	__le32	dst_port;
+	__le32	len;
+	__le16	type;		/* enum virtio_vsock_type */
+	__le16	op;		/* enum virtio_vsock_op */
+	__le32	flags;
+	__le32	buf_alloc;
+	__le32	fwd_cnt;
+};
+
+enum virtio_vsock_type {
+	VIRTIO_VSOCK_TYPE_STREAM = 1,
+};
+
+enum virtio_vsock_op {
+	VIRTIO_VSOCK_OP_INVALID = 0,
+
+	/* Connect operations */
+	VIRTIO_VSOCK_OP_REQUEST = 1,
+	VIRTIO_VSOCK_OP_RESPONSE = 2,
+	VIRTIO_VSOCK_OP_RST = 3,
+	VIRTIO_VSOCK_OP_SHUTDOWN = 4,
+
+	/* To send payload */
+	VIRTIO_VSOCK_OP_RW = 5,
+
+	/* Tell the peer our credit info */
+	VIRTIO_VSOCK_OP_CREDIT_UPDATE = 6,
+	/* Request the peer to send the credit info to us */
+	VIRTIO_VSOCK_OP_CREDIT_REQUEST = 7,
+};
+
+/* VIRTIO_VSOCK_OP_SHUTDOWN flags values */
+enum virtio_vsock_shutdown {
+	VIRTIO_VSOCK_SHUTDOWN_RCV = 1,
+	VIRTIO_VSOCK_SHUTDOWN_SEND = 2,
+};
+
+#endif /* _UAPI_LINUX_VIRTIO_VSOCK_H */
diff --git a/net/vmw_vsock/virtio_transport_common.c
b/net/vmw_vsock/virtio_transport_common.c
new file mode 100644
index 0000000..5b9e202
--- /dev/null
+++ b/net/vmw_vsock/virtio_transport_common.c
@@ -0,0 +1,838 @@
+/*
+ * common code for virtio vsock
+ *
+ * Copyright (C) 2013-2015 Red Hat, Inc.
+ * Author: Asias He <asias at redhat.com>
+ *         Stefan Hajnoczi <stefanha at redhat.com>
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2.
+ */
+#include <linux/module.h>
+#include <linux/ctype.h>
+#include <linux/list.h>
+#include <linux/virtio.h>
+#include <linux/virtio_ids.h>
+#include <linux/virtio_config.h>
+#include <linux/virtio_vsock.h>
+
+#include <net/sock.h>
+#include <net/af_vsock.h>
+
+#define CREATE_TRACE_POINTS
+#include <trace/events/vsock_virtio_transport_common.h>
+
+static const struct virtio_transport *virtio_transport_get_ops(void)
+{
+	const struct vsock_transport *t = vsock_core_get_transport();
+
+	return container_of(t, struct virtio_transport, transport);
+}
+
+static int virtio_transport_send_pkt(struct vsock_sock *vsk,
+				     struct virtio_vsock_pkt_info *info)
+{
+	return virtio_transport_get_ops()->send_pkt(vsk, info);
+}
+
+static int virtio_transport_send_pkt_no_sock(struct virtio_vsock_pkt *pkt)
+{
+	return virtio_transport_get_ops()->send_pkt_no_sock(pkt);
+}
+
+struct virtio_vsock_pkt *
+virtio_transport_alloc_pkt(struct virtio_vsock_pkt_info *info,
+			   size_t len,
+			   u32 src_cid,
+			   u32 src_port,
+			   u32 dst_cid,
+			   u32 dst_port)
+{
+	struct virtio_vsock_pkt *pkt;
+	int err;
+
+	pkt = kzalloc(sizeof(*pkt), GFP_KERNEL);
+	if (!pkt)
+		return NULL;
+
+	pkt->hdr.type		= cpu_to_le16(info->type);
+	pkt->hdr.op		= cpu_to_le16(info->op);
+	pkt->hdr.src_cid	= cpu_to_le32(src_cid);
+	pkt->hdr.src_port	= cpu_to_le32(src_port);
+	pkt->hdr.dst_cid	= cpu_to_le32(dst_cid);
+	pkt->hdr.dst_port	= cpu_to_le32(dst_port);
+	pkt->hdr.flags		= cpu_to_le32(info->flags);
+	pkt->len		= len;
+	pkt->hdr.len		= cpu_to_le32(len);
+
+	if (info->msg && len > 0) {
+		pkt->buf = kmalloc(len, GFP_KERNEL);
+		if (!pkt->buf)
+			goto out_pkt;
+		err = memcpy_from_msg(pkt->buf, info->msg, len);
+		if (err)
+			goto out;
+	}
+
+	trace_virtio_transport_alloc_pkt(src_cid, src_port,
+					 dst_cid, dst_port,
+					 len,
+					 info->type,
+					 info->op,
+					 info->flags);
+
+	return pkt;
+
+out:
+	kfree(pkt->buf);
+out_pkt:
+	kfree(pkt);
+	return NULL;
+}
+EXPORT_SYMBOL_GPL(virtio_transport_alloc_pkt);
+
+static void virtio_transport_inc_rx_pkt(struct virtio_vsock_sock *vvs,
+					struct virtio_vsock_pkt *pkt)
+{
+	vvs->rx_bytes += pkt->len;
+}
+
+static void virtio_transport_dec_rx_pkt(struct virtio_vsock_sock *vvs,
+					struct virtio_vsock_pkt *pkt)
+{
+	vvs->rx_bytes -= pkt->len;
+	vvs->fwd_cnt += pkt->len;
+}
+
+void virtio_transport_inc_tx_pkt(struct virtio_vsock_sock *vvs, struct
virtio_vsock_pkt *pkt)
+{
+	mutex_lock(&vvs->tx_lock);
+	pkt->hdr.fwd_cnt = cpu_to_le32(vvs->fwd_cnt);
+	pkt->hdr.buf_alloc = cpu_to_le32(vvs->buf_alloc);
+	mutex_unlock(&vvs->tx_lock);
+}
+EXPORT_SYMBOL_GPL(virtio_transport_inc_tx_pkt);
+
+u32 virtio_transport_get_credit(struct virtio_vsock_sock *vvs, u32 credit)
+{
+	u32 ret;
+
+	mutex_lock(&vvs->tx_lock);
+	ret = vvs->peer_buf_alloc - (vvs->tx_cnt - vvs->peer_fwd_cnt);
+	if (ret > credit)
+		ret = credit;
+	vvs->tx_cnt += ret;
+	mutex_unlock(&vvs->tx_lock);
+
+	return ret;
+}
+EXPORT_SYMBOL_GPL(virtio_transport_get_credit);
+
+void virtio_transport_put_credit(struct virtio_vsock_sock *vvs, u32 credit)
+{
+	mutex_lock(&vvs->tx_lock);
+	vvs->tx_cnt -= credit;
+	mutex_unlock(&vvs->tx_lock);
+}
+EXPORT_SYMBOL_GPL(virtio_transport_put_credit);
+
+static int virtio_transport_send_credit_update(struct vsock_sock *vsk,
+					       int type,
+					       struct virtio_vsock_hdr *hdr)
+{
+	struct virtio_vsock_pkt_info info = {
+		.op = VIRTIO_VSOCK_OP_CREDIT_UPDATE,
+		.type = type,
+	};
+
+	return virtio_transport_send_pkt(vsk, &info);
+}
+
+static ssize_t
+virtio_transport_stream_do_dequeue(struct vsock_sock *vsk,
+				   struct msghdr *msg,
+				   size_t len)
+{
+	struct virtio_vsock_sock *vvs = vsk->trans;
+	struct virtio_vsock_pkt *pkt;
+	size_t bytes, total = 0;
+	int err = -EFAULT;
+
+	mutex_lock(&vvs->rx_lock);
+	while (total < len &&
+	       vvs->rx_bytes > 0 &&
+	       !list_empty(&vvs->rx_queue)) {
+		pkt = list_first_entry(&vvs->rx_queue,
+				       struct virtio_vsock_pkt, list);
+
+		bytes = len - total;
+		if (bytes > pkt->len - pkt->off)
+			bytes = pkt->len - pkt->off;
+
+		err = memcpy_to_msg(msg, pkt->buf + pkt->off, bytes);
+		if (err)
+			goto out;
+		total += bytes;
+		pkt->off += bytes;
+		if (pkt->off == pkt->len) {
+			virtio_transport_dec_rx_pkt(vvs, pkt);
+			list_del(&pkt->list);
+			virtio_transport_free_pkt(pkt);
+		}
+	}
+	mutex_unlock(&vvs->rx_lock);
+
+	/* Send a credit pkt to peer */
+	virtio_transport_send_credit_update(vsk, VIRTIO_VSOCK_TYPE_STREAM,
+					    NULL);
+
+	return total;
+
+out:
+	mutex_unlock(&vvs->rx_lock);
+	if (total)
+		err = total;
+	return err;
+}
+
+ssize_t
+virtio_transport_stream_dequeue(struct vsock_sock *vsk,
+				struct msghdr *msg,
+				size_t len, int flags)
+{
+	if (flags & MSG_PEEK)
+		return -EOPNOTSUPP;
+
+	return virtio_transport_stream_do_dequeue(vsk, msg, len);
+}
+EXPORT_SYMBOL_GPL(virtio_transport_stream_dequeue);
+
+int
+virtio_transport_dgram_dequeue(struct vsock_sock *vsk,
+			       struct msghdr *msg,
+			       size_t len, int flags)
+{
+	return -EOPNOTSUPP;
+}
+EXPORT_SYMBOL_GPL(virtio_transport_dgram_dequeue);
+
+s64 virtio_transport_stream_has_data(struct vsock_sock *vsk)
+{
+	struct virtio_vsock_sock *vvs = vsk->trans;
+	s64 bytes;
+
+	mutex_lock(&vvs->rx_lock);
+	bytes = vvs->rx_bytes;
+	mutex_unlock(&vvs->rx_lock);
+
+	return bytes;
+}
+EXPORT_SYMBOL_GPL(virtio_transport_stream_has_data);
+
+static s64 virtio_transport_has_space(struct vsock_sock *vsk)
+{
+	struct virtio_vsock_sock *vvs = vsk->trans;
+	s64 bytes;
+
+	bytes = vvs->peer_buf_alloc - (vvs->tx_cnt - vvs->peer_fwd_cnt);
+	if (bytes < 0)
+		bytes = 0;
+
+	return bytes;
+}
+
+s64 virtio_transport_stream_has_space(struct vsock_sock *vsk)
+{
+	struct virtio_vsock_sock *vvs = vsk->trans;
+	s64 bytes;
+
+	mutex_lock(&vvs->tx_lock);
+	bytes = virtio_transport_has_space(vsk);
+	mutex_unlock(&vvs->tx_lock);
+
+	return bytes;
+}
+EXPORT_SYMBOL_GPL(virtio_transport_stream_has_space);
+
+int virtio_transport_do_socket_init(struct vsock_sock *vsk,
+				    struct vsock_sock *psk)
+{
+	struct virtio_vsock_sock *vvs;
+
+	vvs = kzalloc(sizeof(*vvs), GFP_KERNEL);
+	if (!vvs)
+		return -ENOMEM;
+
+	vsk->trans = vvs;
+	vvs->vsk = vsk;
+	if (psk) {
+		struct virtio_vsock_sock *ptrans = psk->trans;
+
+		vvs->buf_size	= ptrans->buf_size;
+		vvs->buf_size_min = ptrans->buf_size_min;
+		vvs->buf_size_max = ptrans->buf_size_max;
+		vvs->peer_buf_alloc = ptrans->peer_buf_alloc;
+	} else {
+		vvs->buf_size = VIRTIO_VSOCK_DEFAULT_BUF_SIZE;
+		vvs->buf_size_min = VIRTIO_VSOCK_DEFAULT_MIN_BUF_SIZE;
+		vvs->buf_size_max = VIRTIO_VSOCK_DEFAULT_MAX_BUF_SIZE;
+	}
+
+	vvs->buf_alloc = vvs->buf_size;
+
+	mutex_init(&vvs->rx_lock);
+	mutex_init(&vvs->tx_lock);
+	INIT_LIST_HEAD(&vvs->rx_queue);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(virtio_transport_do_socket_init);
+
+u64 virtio_transport_get_buffer_size(struct vsock_sock *vsk)
+{
+	struct virtio_vsock_sock *vvs = vsk->trans;
+
+	return vvs->buf_size;
+}
+EXPORT_SYMBOL_GPL(virtio_transport_get_buffer_size);
+
+u64 virtio_transport_get_min_buffer_size(struct vsock_sock *vsk)
+{
+	struct virtio_vsock_sock *vvs = vsk->trans;
+
+	return vvs->buf_size_min;
+}
+EXPORT_SYMBOL_GPL(virtio_transport_get_min_buffer_size);
+
+u64 virtio_transport_get_max_buffer_size(struct vsock_sock *vsk)
+{
+	struct virtio_vsock_sock *vvs = vsk->trans;
+
+	return vvs->buf_size_max;
+}
+EXPORT_SYMBOL_GPL(virtio_transport_get_max_buffer_size);
+
+void virtio_transport_set_buffer_size(struct vsock_sock *vsk, u64 val)
+{
+	struct virtio_vsock_sock *vvs = vsk->trans;
+
+	if (val > VIRTIO_VSOCK_MAX_BUF_SIZE)
+		val = VIRTIO_VSOCK_MAX_BUF_SIZE;
+	if (val < vvs->buf_size_min)
+		vvs->buf_size_min = val;
+	if (val > vvs->buf_size_max)
+		vvs->buf_size_max = val;
+	vvs->buf_size = val;
+	vvs->buf_alloc = val;
+}
+EXPORT_SYMBOL_GPL(virtio_transport_set_buffer_size);
+
+void virtio_transport_set_min_buffer_size(struct vsock_sock *vsk, u64 val)
+{
+	struct virtio_vsock_sock *vvs = vsk->trans;
+
+	if (val > VIRTIO_VSOCK_MAX_BUF_SIZE)
+		val = VIRTIO_VSOCK_MAX_BUF_SIZE;
+	if (val > vvs->buf_size)
+		vvs->buf_size = val;
+	vvs->buf_size_min = val;
+}
+EXPORT_SYMBOL_GPL(virtio_transport_set_min_buffer_size);
+
+void virtio_transport_set_max_buffer_size(struct vsock_sock *vsk, u64 val)
+{
+	struct virtio_vsock_sock *vvs = vsk->trans;
+
+	if (val > VIRTIO_VSOCK_MAX_BUF_SIZE)
+		val = VIRTIO_VSOCK_MAX_BUF_SIZE;
+	if (val < vvs->buf_size)
+		vvs->buf_size = val;
+	vvs->buf_size_max = val;
+}
+EXPORT_SYMBOL_GPL(virtio_transport_set_max_buffer_size);
+
+int
+virtio_transport_notify_poll_in(struct vsock_sock *vsk,
+				size_t target,
+				bool *data_ready_now)
+{
+	if (vsock_stream_has_data(vsk))
+		*data_ready_now = true;
+	else
+		*data_ready_now = false;
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(virtio_transport_notify_poll_in);
+
+int
+virtio_transport_notify_poll_out(struct vsock_sock *vsk,
+				 size_t target,
+				 bool *space_avail_now)
+{
+	s64 free_space;
+
+	free_space = vsock_stream_has_space(vsk);
+	if (free_space > 0)
+		*space_avail_now = true;
+	else if (free_space == 0)
+		*space_avail_now = false;
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(virtio_transport_notify_poll_out);
+
+int virtio_transport_notify_recv_init(struct vsock_sock *vsk,
+	size_t target, struct vsock_transport_recv_notify_data *data)
+{
+	return 0;
+}
+EXPORT_SYMBOL_GPL(virtio_transport_notify_recv_init);
+
+int virtio_transport_notify_recv_pre_block(struct vsock_sock *vsk,
+	size_t target, struct vsock_transport_recv_notify_data *data)
+{
+	return 0;
+}
+EXPORT_SYMBOL_GPL(virtio_transport_notify_recv_pre_block);
+
+int virtio_transport_notify_recv_pre_dequeue(struct vsock_sock *vsk,
+	size_t target, struct vsock_transport_recv_notify_data *data)
+{
+	return 0;
+}
+EXPORT_SYMBOL_GPL(virtio_transport_notify_recv_pre_dequeue);
+
+int virtio_transport_notify_recv_post_dequeue(struct vsock_sock *vsk,
+	size_t target, ssize_t copied, bool data_read,
+	struct vsock_transport_recv_notify_data *data)
+{
+	return 0;
+}
+EXPORT_SYMBOL_GPL(virtio_transport_notify_recv_post_dequeue);
+
+int virtio_transport_notify_send_init(struct vsock_sock *vsk,
+	struct vsock_transport_send_notify_data *data)
+{
+	return 0;
+}
+EXPORT_SYMBOL_GPL(virtio_transport_notify_send_init);
+
+int virtio_transport_notify_send_pre_block(struct vsock_sock *vsk,
+	struct vsock_transport_send_notify_data *data)
+{
+	return 0;
+}
+EXPORT_SYMBOL_GPL(virtio_transport_notify_send_pre_block);
+
+int virtio_transport_notify_send_pre_enqueue(struct vsock_sock *vsk,
+	struct vsock_transport_send_notify_data *data)
+{
+	return 0;
+}
+EXPORT_SYMBOL_GPL(virtio_transport_notify_send_pre_enqueue);
+
+int virtio_transport_notify_send_post_enqueue(struct vsock_sock *vsk,
+	ssize_t written, struct vsock_transport_send_notify_data *data)
+{
+	return 0;
+}
+EXPORT_SYMBOL_GPL(virtio_transport_notify_send_post_enqueue);
+
+u64 virtio_transport_stream_rcvhiwat(struct vsock_sock *vsk)
+{
+	struct virtio_vsock_sock *vvs = vsk->trans;
+
+	return vvs->buf_size;
+}
+EXPORT_SYMBOL_GPL(virtio_transport_stream_rcvhiwat);
+
+bool virtio_transport_stream_is_active(struct vsock_sock *vsk)
+{
+	return true;
+}
+EXPORT_SYMBOL_GPL(virtio_transport_stream_is_active);
+
+bool virtio_transport_stream_allow(u32 cid, u32 port)
+{
+	return true;
+}
+EXPORT_SYMBOL_GPL(virtio_transport_stream_allow);
+
+int virtio_transport_dgram_bind(struct vsock_sock *vsk,
+				struct sockaddr_vm *addr)
+{
+	return -EOPNOTSUPP;
+}
+EXPORT_SYMBOL_GPL(virtio_transport_dgram_bind);
+
+bool virtio_transport_dgram_allow(u32 cid, u32 port)
+{
+	return false;
+}
+EXPORT_SYMBOL_GPL(virtio_transport_dgram_allow);
+
+int virtio_transport_connect(struct vsock_sock *vsk)
+{
+	struct virtio_vsock_pkt_info info = {
+		.op = VIRTIO_VSOCK_OP_REQUEST,
+		.type = VIRTIO_VSOCK_TYPE_STREAM,
+	};
+
+	return virtio_transport_send_pkt(vsk, &info);
+}
+EXPORT_SYMBOL_GPL(virtio_transport_connect);
+
+int virtio_transport_shutdown(struct vsock_sock *vsk, int mode)
+{
+	struct virtio_vsock_pkt_info info = {
+		.op = VIRTIO_VSOCK_OP_SHUTDOWN,
+		.type = VIRTIO_VSOCK_TYPE_STREAM,
+		.flags = (mode & RCV_SHUTDOWN ?
+			  VIRTIO_VSOCK_SHUTDOWN_RCV : 0) |
+			 (mode & SEND_SHUTDOWN ?
+			  VIRTIO_VSOCK_SHUTDOWN_SEND : 0),
+	};
+
+	return virtio_transport_send_pkt(vsk, &info);
+}
+EXPORT_SYMBOL_GPL(virtio_transport_shutdown);
+
+void virtio_transport_release(struct vsock_sock *vsk)
+{
+	struct sock *sk = &vsk->sk;
+
+	/* Tell other side to terminate connection */
+	if (sk->sk_type == SOCK_STREAM &&
+	    vsk->peer_shutdown != SHUTDOWN_MASK &&
+	    sk->sk_state == SS_CONNECTED)
+		(void)virtio_transport_shutdown(vsk, SHUTDOWN_MASK);
+}
+EXPORT_SYMBOL_GPL(virtio_transport_release);
+
+int
+virtio_transport_dgram_enqueue(struct vsock_sock *vsk,
+			       struct sockaddr_vm *remote_addr,
+			       struct msghdr *msg,
+			       size_t dgram_len)
+{
+	return -EOPNOTSUPP;
+}
+EXPORT_SYMBOL_GPL(virtio_transport_dgram_enqueue);
+
+ssize_t
+virtio_transport_stream_enqueue(struct vsock_sock *vsk,
+				struct msghdr *msg,
+				size_t len)
+{
+	struct virtio_vsock_pkt_info info = {
+		.op = VIRTIO_VSOCK_OP_RW,
+		.type = VIRTIO_VSOCK_TYPE_STREAM,
+		.msg = msg,
+		.pkt_len = len,
+	};
+
+	return virtio_transport_send_pkt(vsk, &info);
+}
+EXPORT_SYMBOL_GPL(virtio_transport_stream_enqueue);
+
+void virtio_transport_destruct(struct vsock_sock *vsk)
+{
+	struct virtio_vsock_sock *vvs = vsk->trans;
+
+	kfree(vvs);
+}
+EXPORT_SYMBOL_GPL(virtio_transport_destruct);
+
+static int virtio_transport_send_reset(struct vsock_sock *vsk,
+				       struct virtio_vsock_pkt *pkt)
+{
+	struct virtio_vsock_pkt_info info = {
+		.op = VIRTIO_VSOCK_OP_RST,
+		.type = VIRTIO_VSOCK_TYPE_STREAM,
+	};
+
+	/* Send RST only if the original pkt is not a RST pkt */
+	if (le16_to_cpu(pkt->hdr.op) == VIRTIO_VSOCK_OP_RST)
+		return 0;
+
+	return virtio_transport_send_pkt(vsk, &info);
+}
+
+/* Normally packets are associated with a socket.  There may be no socket if an
+ * attempt was made to connect to a socket that does not exist.
+ */
+static int virtio_transport_send_reset_no_sock(struct virtio_vsock_pkt *pkt)
+{
+	struct virtio_vsock_pkt_info info = {
+		.op = VIRTIO_VSOCK_OP_RST,
+		.type = le16_to_cpu(pkt->hdr.type),
+	};
+
+	/* Send RST only if the original pkt is not a RST pkt */
+	if (le16_to_cpu(pkt->hdr.op) == VIRTIO_VSOCK_OP_RST)
+		return 0;
+
+	pkt = virtio_transport_alloc_pkt(&info, 0,
+					 le32_to_cpu(pkt->hdr.dst_cid),
+					 le32_to_cpu(pkt->hdr.dst_port),
+					 le32_to_cpu(pkt->hdr.src_cid),
+					 le32_to_cpu(pkt->hdr.src_port));
+	if (!pkt)
+		return -ENOMEM;
+
+	return virtio_transport_send_pkt_no_sock(pkt);
+}
+
+static int
+virtio_transport_recv_connecting(struct sock *sk,
+				 struct virtio_vsock_pkt *pkt)
+{
+	struct vsock_sock *vsk = vsock_sk(sk);
+	int err;
+	int skerr;
+
+	switch (le16_to_cpu(pkt->hdr.op)) {
+	case VIRTIO_VSOCK_OP_RESPONSE:
+		sk->sk_state = SS_CONNECTED;
+		sk->sk_socket->state = SS_CONNECTED;
+		vsock_insert_connected(vsk);
+		sk->sk_state_change(sk);
+		break;
+	case VIRTIO_VSOCK_OP_INVALID:
+		break;
+	case VIRTIO_VSOCK_OP_RST:
+		skerr = ECONNRESET;
+		err = 0;
+		goto destroy;
+	default:
+		skerr = EPROTO;
+		err = -EINVAL;
+		goto destroy;
+	}
+	return 0;
+
+destroy:
+	virtio_transport_send_reset(vsk, pkt);
+	sk->sk_state = SS_UNCONNECTED;
+	sk->sk_err = skerr;
+	sk->sk_error_report(sk);
+	return err;
+}
+
+static int
+virtio_transport_recv_connected(struct sock *sk,
+				struct virtio_vsock_pkt *pkt)
+{
+	struct vsock_sock *vsk = vsock_sk(sk);
+	struct virtio_vsock_sock *vvs = vsk->trans;
+	int err = 0;
+
+	switch (le16_to_cpu(pkt->hdr.op)) {
+	case VIRTIO_VSOCK_OP_RW:
+		pkt->len = le32_to_cpu(pkt->hdr.len);
+		pkt->off = 0;
+
+		mutex_lock(&vvs->rx_lock);
+		virtio_transport_inc_rx_pkt(vvs, pkt);
+		list_add_tail(&pkt->list, &vvs->rx_queue);
+		mutex_unlock(&vvs->rx_lock);
+
+		sk->sk_data_ready(sk);
+		return err;
+	case VIRTIO_VSOCK_OP_CREDIT_UPDATE:
+		sk->sk_write_space(sk);
+		break;
+	case VIRTIO_VSOCK_OP_SHUTDOWN:
+		if (le32_to_cpu(pkt->hdr.flags) & VIRTIO_VSOCK_SHUTDOWN_RCV)
+			vsk->peer_shutdown |= RCV_SHUTDOWN;
+		if (le32_to_cpu(pkt->hdr.flags) & VIRTIO_VSOCK_SHUTDOWN_SEND)
+			vsk->peer_shutdown |= SEND_SHUTDOWN;
+		if (vsk->peer_shutdown == SHUTDOWN_MASK &&
+		    vsock_stream_has_data(vsk) <= 0)
+			sk->sk_state = SS_DISCONNECTING;
+		if (le32_to_cpu(pkt->hdr.flags))
+			sk->sk_state_change(sk);
+		break;
+	case VIRTIO_VSOCK_OP_RST:
+		sock_set_flag(sk, SOCK_DONE);
+		vsk->peer_shutdown = SHUTDOWN_MASK;
+		if (vsock_stream_has_data(vsk) <= 0)
+			sk->sk_state = SS_DISCONNECTING;
+		sk->sk_state_change(sk);
+		break;
+	default:
+		err = -EINVAL;
+		break;
+	}
+
+	virtio_transport_free_pkt(pkt);
+	return err;
+}
+
+static int
+virtio_transport_send_response(struct vsock_sock *vsk,
+			       struct virtio_vsock_pkt *pkt)
+{
+	struct virtio_vsock_pkt_info info = {
+		.op = VIRTIO_VSOCK_OP_RESPONSE,
+		.type = VIRTIO_VSOCK_TYPE_STREAM,
+		.remote_cid = le32_to_cpu(pkt->hdr.src_cid),
+		.remote_port = le32_to_cpu(pkt->hdr.src_port),
+	};
+
+	return virtio_transport_send_pkt(vsk, &info);
+}
+
+/* Handle server socket */
+static int
+virtio_transport_recv_listen(struct sock *sk, struct virtio_vsock_pkt *pkt)
+{
+	struct vsock_sock *vsk = vsock_sk(sk);
+	struct vsock_sock *vchild;
+	struct sock *child;
+
+	if (le16_to_cpu(pkt->hdr.op) != VIRTIO_VSOCK_OP_REQUEST) {
+		virtio_transport_send_reset(vsk, pkt);
+		return -EINVAL;
+	}
+
+	if (sk_acceptq_is_full(sk)) {
+		virtio_transport_send_reset(vsk, pkt);
+		return -ENOMEM;
+	}
+
+	child = __vsock_create(sock_net(sk), NULL, sk, GFP_KERNEL,
+			       sk->sk_type, 0);
+	if (!child) {
+		virtio_transport_send_reset(vsk, pkt);
+		return -ENOMEM;
+	}
+
+	sk->sk_ack_backlog++;
+
+	lock_sock(child);
+
+	child->sk_state = SS_CONNECTED;
+
+	vchild = vsock_sk(child);
+	vsock_addr_init(&vchild->local_addr, le32_to_cpu(pkt->hdr.dst_cid),
+			le32_to_cpu(pkt->hdr.dst_port));
+	vsock_addr_init(&vchild->remote_addr, le32_to_cpu(pkt->hdr.src_cid),
+			le32_to_cpu(pkt->hdr.src_port));
+
+	vsock_insert_connected(vchild);
+	vsock_enqueue_accept(sk, child);
+	virtio_transport_send_response(vchild, pkt);
+
+	release_sock(child);
+
+	sk->sk_data_ready(sk);
+	return 0;
+}
+
+static void virtio_transport_space_update(struct sock *sk,
+					  struct virtio_vsock_pkt *pkt)
+{
+	struct vsock_sock *vsk = vsock_sk(sk);
+	struct virtio_vsock_sock *vvs = vsk->trans;
+	bool space_available;
+
+	/* buf_alloc and fwd_cnt is always included in the hdr */
+	mutex_lock(&vvs->tx_lock);
+	vvs->peer_buf_alloc = le32_to_cpu(pkt->hdr.buf_alloc);
+	vvs->peer_fwd_cnt = le32_to_cpu(pkt->hdr.fwd_cnt);
+	space_available = virtio_transport_has_space(vsk);
+	mutex_unlock(&vvs->tx_lock);
+
+	if (space_available)
+		sk->sk_write_space(sk);
+}
+
+/* We are under the virtio-vsock's vsock->rx_lock or vhost-vsock's
vq->mutex
+ * lock.
+ */
+void virtio_transport_recv_pkt(struct virtio_vsock_pkt *pkt)
+{
+	struct sockaddr_vm src, dst;
+	struct vsock_sock *vsk;
+	struct sock *sk;
+
+	vsock_addr_init(&src, le32_to_cpu(pkt->hdr.src_cid),
+			le32_to_cpu(pkt->hdr.src_port));
+	vsock_addr_init(&dst, le32_to_cpu(pkt->hdr.dst_cid),
+			le32_to_cpu(pkt->hdr.dst_port));
+
+	trace_virtio_transport_recv_pkt(src.svm_cid, src.svm_port,
+					dst.svm_cid, dst.svm_port,
+					le32_to_cpu(pkt->hdr.len),
+					le16_to_cpu(pkt->hdr.type),
+					le16_to_cpu(pkt->hdr.op),
+					le32_to_cpu(pkt->hdr.flags),
+					le32_to_cpu(pkt->hdr.buf_alloc),
+					le32_to_cpu(pkt->hdr.fwd_cnt));
+
+	if (le16_to_cpu(pkt->hdr.type) != VIRTIO_VSOCK_TYPE_STREAM) {
+		(void)virtio_transport_send_reset_no_sock(pkt);
+		goto free_pkt;
+	}
+
+	/* The socket must be in connected or bound table
+	 * otherwise send reset back
+	 */
+	sk = vsock_find_connected_socket(&src, &dst);
+	if (!sk) {
+		sk = vsock_find_bound_socket(&dst);
+		if (!sk) {
+			(void)virtio_transport_send_reset_no_sock(pkt);
+			goto free_pkt;
+		}
+	}
+
+	vsk = vsock_sk(sk);
+
+	virtio_transport_space_update(sk, pkt);
+
+	lock_sock(sk);
+
+	/* Update CID in case it has changed after a transport reset event */
+	vsk->local_addr.svm_cid = dst.svm_cid;
+
+	switch (sk->sk_state) {
+	case VSOCK_SS_LISTEN:
+		virtio_transport_recv_listen(sk, pkt);
+		virtio_transport_free_pkt(pkt);
+		break;
+	case SS_CONNECTING:
+		virtio_transport_recv_connecting(sk, pkt);
+		virtio_transport_free_pkt(pkt);
+		break;
+	case SS_CONNECTED:
+		virtio_transport_recv_connected(sk, pkt);
+		break;
+	default:
+		virtio_transport_free_pkt(pkt);
+		break;
+	}
+	release_sock(sk);
+
+	/* Release refcnt obtained when we fetched this socket out of the
+	 * bound or connected list.
+	 */
+	sock_put(sk);
+	return;
+
+free_pkt:
+	virtio_transport_free_pkt(pkt);
+}
+EXPORT_SYMBOL_GPL(virtio_transport_recv_pkt);
+
+void virtio_transport_free_pkt(struct virtio_vsock_pkt *pkt)
+{
+	kfree(pkt->buf);
+	kfree(pkt);
+}
+EXPORT_SYMBOL_GPL(virtio_transport_free_pkt);
+
+MODULE_LICENSE("GPL v2");
+MODULE_AUTHOR("Asias He");
+MODULE_DESCRIPTION("common code for virtio vsock");
-- 
2.5.5

From: Asias He <asias at redhat.com>

VM sockets virtio transport implementation.  This driver runs in the
guest.

Signed-off-by: Asias He <asias at redhat.com>
Signed-off-by: Stefan Hajnoczi <stefanha at redhat.com>
---
v5:
 * Add transport reset event handling
 * Drop ctrl virtqueue
v4:
 * Add MAINTAINERS file entry
 * Drop short/long rx packets
 * checkpatch.pl cleanups
 * Clarify locking in struct virtio_vsock
 * Narrow local variable scopes as suggested by Alex Bennee
 * Call wake_up() after decrementing total_tx_buf to avoid deadlock
v2:
 * Fix total_tx_buf accounting
 * Add virtio_transport global mutex to prevent races
---
 MAINTAINERS                      |   1 +
 net/vmw_vsock/virtio_transport.c | 584 +++++++++++++++++++++++++++++++++++++++
 2 files changed, 585 insertions(+)
 create mode 100644 net/vmw_vsock/virtio_transport.c

diff --git a/MAINTAINERS b/MAINTAINERS
index fb9c47a..1ed7364 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -11611,6 +11611,7 @@ S:	Maintained
 F:	include/linux/virtio_vsock.h
 F:	include/uapi/linux/virtio_vsock.h
 F:	net/vmw_vsock/virtio_transport_common.c
+F:	net/vmw_vsock/virtio_transport.c
 
 VIRTUAL SERIO DEVICE DRIVER
 M:	Stephen Chandler Paul <thatslyude at gmail.com>
diff --git a/net/vmw_vsock/virtio_transport.c b/net/vmw_vsock/virtio_transport.c
new file mode 100644
index 0000000..45472e0
--- /dev/null
+++ b/net/vmw_vsock/virtio_transport.c
@@ -0,0 +1,584 @@
+/*
+ * virtio transport for vsock
+ *
+ * Copyright (C) 2013-2015 Red Hat, Inc.
+ * Author: Asias He <asias at redhat.com>
+ *         Stefan Hajnoczi <stefanha at redhat.com>
+ *
+ * Some of the code is take from Gerd Hoffmann <kraxel at
redhat.com>'s
+ * early virtio-vsock proof-of-concept bits.
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2.
+ */
+#include <linux/spinlock.h>
+#include <linux/module.h>
+#include <linux/list.h>
+#include <linux/virtio.h>
+#include <linux/virtio_ids.h>
+#include <linux/virtio_config.h>
+#include <linux/virtio_vsock.h>
+#include <net/sock.h>
+#include <linux/mutex.h>
+#include <net/af_vsock.h>
+
+static struct workqueue_struct *virtio_vsock_workqueue;
+static struct virtio_vsock *the_virtio_vsock;
+static DEFINE_MUTEX(the_virtio_vsock_mutex); /* protects the_virtio_vsock */
+static void virtio_vsock_rx_fill(struct virtio_vsock *vsock);
+
+struct virtio_vsock {
+	struct virtio_device *vdev;
+	struct virtqueue *vqs[VSOCK_VQ_MAX];
+
+	/* Virtqueue processing is deferred to a workqueue */
+	struct work_struct tx_work;
+	struct work_struct rx_work;
+	struct work_struct event_work;
+
+	wait_queue_head_t tx_wait;	/* for waiting for tx resources */
+
+	/* The following fields are protected by tx_lock.  vqs[VSOCK_VQ_TX]
+	 * must be accessed with tx_lock held.
+	 */
+	struct mutex tx_lock;
+	u32 total_tx_buf;
+
+	/* The following fields are protected by rx_lock.  vqs[VSOCK_VQ_RX]
+	 * must be accessed with rx_lock held.
+	 */
+	struct mutex rx_lock;
+	int rx_buf_nr;
+	int rx_buf_max_nr;
+
+	/* The following fields are protected by event_lock.
+	 * vqs[VSOCK_VQ_EVENT] must be accessed with event_lock held.
+	 */
+	struct mutex event_lock;
+	struct virtio_vsock_event event_list[8];
+
+	u32 guest_cid;
+};
+
+static struct virtio_vsock *virtio_vsock_get(void)
+{
+	return the_virtio_vsock;
+}
+
+static u32 virtio_transport_get_local_cid(void)
+{
+	struct virtio_vsock *vsock = virtio_vsock_get();
+
+	return vsock->guest_cid;
+}
+
+static int
+virtio_transport_send_one_pkt(struct virtio_vsock *vsock,
+			      struct virtio_vsock_pkt *pkt)
+{
+	struct scatterlist hdr, buf, *sgs[2];
+	int ret, in_sg = 0, out_sg = 0;
+	struct virtqueue *vq;
+	DEFINE_WAIT(wait);
+
+	vq = vsock->vqs[VSOCK_VQ_TX];
+
+	/* Put pkt in the virtqueue */
+	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;
+	}
+
+	mutex_lock(&vsock->tx_lock);
+	while ((ret = virtqueue_add_sgs(vq, sgs, out_sg, in_sg, pkt,
+					GFP_KERNEL)) < 0) {
+		prepare_to_wait_exclusive(&vsock->tx_wait, &wait,
+					  TASK_UNINTERRUPTIBLE);
+		mutex_unlock(&vsock->tx_lock);
+		schedule();
+		mutex_lock(&vsock->tx_lock);
+		finish_wait(&vsock->tx_wait, &wait);
+	}
+	virtqueue_kick(vq);
+	mutex_unlock(&vsock->tx_lock);
+
+	return pkt->len;
+}
+
+static int
+virtio_transport_send_pkt_no_sock(struct virtio_vsock_pkt *pkt)
+{
+	struct virtio_vsock *vsock;
+
+	vsock = virtio_vsock_get();
+	if (!vsock) {
+		virtio_transport_free_pkt(pkt);
+		return -ENODEV;
+	}
+
+	return virtio_transport_send_one_pkt(vsock, pkt);
+}
+
+static int
+virtio_transport_send_pkt(struct vsock_sock *vsk,
+			  struct virtio_vsock_pkt_info *info)
+{
+	u32 src_cid, src_port, dst_cid, dst_port;
+	struct virtio_vsock_sock *vvs;
+	struct virtio_vsock_pkt *pkt;
+	struct virtio_vsock *vsock;
+	u32 pkt_len = info->pkt_len;
+	DEFINE_WAIT(wait);
+
+	vsock = virtio_vsock_get();
+	if (!vsock)
+		return -ENODEV;
+
+	src_cid	= virtio_transport_get_local_cid();
+	src_port = vsk->local_addr.svm_port;
+	if (!info->remote_cid) {
+		dst_cid	= vsk->remote_addr.svm_cid;
+		dst_port = vsk->remote_addr.svm_port;
+	} else {
+		dst_cid = info->remote_cid;
+		dst_port = info->remote_port;
+	}
+
+	vvs = vsk->trans;
+
+	if (pkt_len > VIRTIO_VSOCK_DEFAULT_RX_BUF_SIZE)
+		pkt_len = VIRTIO_VSOCK_DEFAULT_RX_BUF_SIZE;
+	pkt_len = virtio_transport_get_credit(vvs, pkt_len);
+	/* Do not send zero length OP_RW pkt*/
+	if (pkt_len == 0 && info->op == VIRTIO_VSOCK_OP_RW)
+		return pkt_len;
+
+	/* Respect global tx buf limitation */
+	mutex_lock(&vsock->tx_lock);
+	while (pkt_len + vsock->total_tx_buf > VIRTIO_VSOCK_MAX_TX_BUF_SIZE) {
+		prepare_to_wait_exclusive(&vsock->tx_wait, &wait,
+					  TASK_UNINTERRUPTIBLE);
+		mutex_unlock(&vsock->tx_lock);
+		schedule();
+		mutex_lock(&vsock->tx_lock);
+		finish_wait(&vsock->tx_wait, &wait);
+	}
+	vsock->total_tx_buf += pkt_len;
+	mutex_unlock(&vsock->tx_lock);
+
+	pkt = virtio_transport_alloc_pkt(info, pkt_len,
+					 src_cid, src_port,
+					 dst_cid, dst_port);
+	if (!pkt) {
+		mutex_lock(&vsock->tx_lock);
+		vsock->total_tx_buf -= pkt_len;
+		mutex_unlock(&vsock->tx_lock);
+		virtio_transport_put_credit(vvs, pkt_len);
+		wake_up(&vsock->tx_wait);
+		return -ENOMEM;
+	}
+
+	virtio_transport_inc_tx_pkt(vvs, pkt);
+
+	return virtio_transport_send_one_pkt(vsock, pkt);
+}
+
+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];
+	struct virtqueue *vq;
+	int ret;
+
+	vq = vsock->vqs[VSOCK_VQ_RX];
+
+	do {
+		pkt = kzalloc(sizeof(*pkt), GFP_KERNEL);
+		if (!pkt)
+			break;
+
+		pkt->buf = kmalloc(buf_len, GFP_KERNEL);
+		if (!pkt->buf) {
+			virtio_transport_free_pkt(pkt);
+			break;
+		}
+
+		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)
+		vsock->rx_buf_max_nr = vsock->rx_buf_nr;
+	virtqueue_kick(vq);
+}
+
+static void virtio_transport_send_pkt_work(struct work_struct *work)
+{
+	struct virtio_vsock *vsock +		container_of(work, struct virtio_vsock,
tx_work);
+	struct virtqueue *vq;
+	bool added = false;
+
+	vq = vsock->vqs[VSOCK_VQ_TX];
+	mutex_lock(&vsock->tx_lock);
+	do {
+		struct virtio_vsock_pkt *pkt;
+		unsigned int len;
+
+		virtqueue_disable_cb(vq);
+		while ((pkt = virtqueue_get_buf(vq, &len)) != NULL) {
+			vsock->total_tx_buf -= pkt->len;
+			virtio_transport_free_pkt(pkt);
+			added = true;
+		}
+	} while (!virtqueue_enable_cb(vq));
+	mutex_unlock(&vsock->tx_lock);
+
+	if (added)
+		wake_up(&vsock->tx_wait);
+}
+
+static void virtio_transport_recv_pkt_work(struct work_struct *work)
+{
+	struct virtio_vsock *vsock +		container_of(work, struct virtio_vsock,
rx_work);
+	struct virtqueue *vq;
+
+	vq = vsock->vqs[VSOCK_VQ_RX];
+	mutex_lock(&vsock->rx_lock);
+	do {
+		struct virtio_vsock_pkt *pkt;
+		unsigned int len;
+
+		virtqueue_disable_cb(vq);
+		while ((pkt = virtqueue_get_buf(vq, &len)) != NULL) {
+			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);
+				continue;
+			}
+
+			pkt->len = len - sizeof(pkt->hdr);
+			virtio_transport_recv_pkt(pkt);
+		}
+	} while (!virtqueue_enable_cb(vq));
+
+	if (vsock->rx_buf_nr < vsock->rx_buf_max_nr / 2)
+		virtio_vsock_rx_fill(vsock);
+	mutex_unlock(&vsock->rx_lock);
+}
+
+/* event_lock must be held */
+static int virtio_vsock_event_fill_one(struct virtio_vsock *vsock,
+				       struct virtio_vsock_event *event)
+{
+	struct scatterlist sg;
+	struct virtqueue *vq;
+
+	vq = vsock->vqs[VSOCK_VQ_EVENT];
+
+	sg_init_one(&sg, event, sizeof(*event));
+
+	return virtqueue_add_inbuf(vq, &sg, 1, event, GFP_KERNEL);
+}
+
+/* event_lock must be held */
+static void virtio_vsock_event_fill(struct virtio_vsock *vsock)
+{
+	size_t i;
+
+	for (i = 0; i < ARRAY_SIZE(vsock->event_list); i++) {
+		struct virtio_vsock_event *event = &vsock->event_list[i];
+
+		virtio_vsock_event_fill_one(vsock, event);
+	}
+
+	virtqueue_kick(vsock->vqs[VSOCK_VQ_EVENT]);
+}
+
+static void virtio_vsock_reset_sock(struct sock *sk)
+{
+	lock_sock(sk);
+	sk->sk_state = SS_UNCONNECTED;
+	sk->sk_err = ECONNRESET;
+	sk->sk_error_report(sk);
+	release_sock(sk);
+}
+
+static void virtio_vsock_update_guest_cid(struct virtio_vsock *vsock)
+{
+	struct virtio_device *vdev = vsock->vdev;
+	u32 guest_cid;
+
+	vdev->config->get(vdev, offsetof(struct virtio_vsock_config, guest_cid),
+			  &guest_cid, sizeof(guest_cid));
+	vsock->guest_cid = le32_to_cpu(guest_cid);
+}
+
+/* event_lock must be held */
+static void virtio_vsock_event_handle(struct virtio_vsock *vsock,
+				      struct virtio_vsock_event *event)
+{
+	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);
+		break;
+	}
+}
+
+static void virtio_transport_event_work(struct work_struct *work)
+{
+	struct virtio_vsock *vsock +		container_of(work, struct virtio_vsock,
event_work);
+	struct virtqueue *vq;
+
+	vq = vsock->vqs[VSOCK_VQ_EVENT];
+
+	mutex_lock(&vsock->event_lock);
+
+	do {
+		struct virtio_vsock_event *event;
+		unsigned int len;
+
+		virtqueue_disable_cb(vq);
+		while ((event = virtqueue_get_buf(vq, &len)) != NULL) {
+			if (len == sizeof(*event))
+				virtio_vsock_event_handle(vsock, event);
+
+			virtio_vsock_event_fill_one(vsock, event);
+		}
+	} while (!virtqueue_enable_cb(vq));
+
+	virtqueue_kick(vsock->vqs[VSOCK_VQ_EVENT]);
+
+	mutex_unlock(&vsock->event_lock);
+}
+
+static void virtio_vsock_event_done(struct virtqueue *vq)
+{
+	struct virtio_vsock *vsock = vq->vdev->priv;
+
+	if (!vsock)
+		return;
+	queue_work(virtio_vsock_workqueue, &vsock->event_work);
+}
+
+static void virtio_vsock_tx_done(struct virtqueue *vq)
+{
+	struct virtio_vsock *vsock = vq->vdev->priv;
+
+	if (!vsock)
+		return;
+	queue_work(virtio_vsock_workqueue, &vsock->tx_work);
+}
+
+static void virtio_vsock_rx_done(struct virtqueue *vq)
+{
+	struct virtio_vsock *vsock = vq->vdev->priv;
+
+	if (!vsock)
+		return;
+	queue_work(virtio_vsock_workqueue, &vsock->rx_work);
+}
+
+static struct virtio_transport virtio_transport = {
+	.transport = {
+		.get_local_cid            = virtio_transport_get_local_cid,
+
+		.init                     = virtio_transport_do_socket_init,
+		.destruct                 = virtio_transport_destruct,
+		.release                  = virtio_transport_release,
+		.connect                  = virtio_transport_connect,
+		.shutdown                 = virtio_transport_shutdown,
+
+		.dgram_bind               = virtio_transport_dgram_bind,
+		.dgram_dequeue            = virtio_transport_dgram_dequeue,
+		.dgram_enqueue            = virtio_transport_dgram_enqueue,
+		.dgram_allow              = virtio_transport_dgram_allow,
+
+		.stream_dequeue           = virtio_transport_stream_dequeue,
+		.stream_enqueue           = virtio_transport_stream_enqueue,
+		.stream_has_data          = virtio_transport_stream_has_data,
+		.stream_has_space         = virtio_transport_stream_has_space,
+		.stream_rcvhiwat          = virtio_transport_stream_rcvhiwat,
+		.stream_is_active         = virtio_transport_stream_is_active,
+		.stream_allow             = virtio_transport_stream_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,
+		.notify_recv_pre_block    = virtio_transport_notify_recv_pre_block,
+		.notify_recv_pre_dequeue  = virtio_transport_notify_recv_pre_dequeue,
+		.notify_recv_post_dequeue = virtio_transport_notify_recv_post_dequeue,
+		.notify_send_init         = virtio_transport_notify_send_init,
+		.notify_send_pre_block    = virtio_transport_notify_send_pre_block,
+		.notify_send_pre_enqueue  = virtio_transport_notify_send_pre_enqueue,
+		.notify_send_post_enqueue = virtio_transport_notify_send_post_enqueue,
+
+		.set_buffer_size          = virtio_transport_set_buffer_size,
+		.set_min_buffer_size      = virtio_transport_set_min_buffer_size,
+		.set_max_buffer_size      = virtio_transport_set_max_buffer_size,
+		.get_buffer_size          = virtio_transport_get_buffer_size,
+		.get_min_buffer_size      = virtio_transport_get_min_buffer_size,
+		.get_max_buffer_size      = virtio_transport_get_max_buffer_size,
+	},
+
+	.send_pkt		= virtio_transport_send_pkt,
+	.send_pkt_no_sock	= virtio_transport_send_pkt_no_sock,
+};
+
+static int virtio_vsock_probe(struct virtio_device *vdev)
+{
+	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_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 (the_virtio_vsock) {
+		ret = -EBUSY;
+		goto out;
+	}
+
+	vsock = kzalloc(sizeof(*vsock), GFP_KERNEL);
+	if (!vsock) {
+		ret = -ENOMEM;
+		goto out;
+	}
+
+	vsock->vdev = vdev;
+
+	ret = vsock->vdev->config->find_vqs(vsock->vdev, VSOCK_VQ_MAX,
+					    vsock->vqs, callbacks, names);
+	if (ret < 0)
+		goto out;
+
+	virtio_vsock_update_guest_cid(vsock);
+
+	ret = vsock_core_init(&virtio_transport.transport);
+	if (ret < 0)
+		goto out_vqs;
+
+	vsock->rx_buf_nr = 0;
+	vsock->rx_buf_max_nr = 0;
+
+	vdev->priv = vsock;
+	the_virtio_vsock = vsock;
+	init_waitqueue_head(&vsock->tx_wait);
+	mutex_init(&vsock->tx_lock);
+	mutex_init(&vsock->rx_lock);
+	mutex_init(&vsock->event_lock);
+	INIT_WORK(&vsock->rx_work, virtio_transport_recv_pkt_work);
+	INIT_WORK(&vsock->tx_work, virtio_transport_send_pkt_work);
+	INIT_WORK(&vsock->event_work, virtio_transport_event_work);
+
+	mutex_lock(&vsock->rx_lock);
+	virtio_vsock_rx_fill(vsock);
+	mutex_unlock(&vsock->rx_lock);
+
+	mutex_lock(&vsock->event_lock);
+	virtio_vsock_event_fill(vsock);
+	mutex_unlock(&vsock->event_lock);
+
+	mutex_unlock(&the_virtio_vsock_mutex);
+	return 0;
+
+out_vqs:
+	vsock->vdev->config->del_vqs(vsock->vdev);
+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;
+
+	flush_work(&vsock->rx_work);
+	flush_work(&vsock->tx_work);
+	flush_work(&vsock->event_work);
+
+	vdev->config->reset(vdev);
+
+	mutex_lock(&the_virtio_vsock_mutex);
+	the_virtio_vsock = NULL;
+	vsock_core_exit();
+	mutex_unlock(&the_virtio_vsock_mutex);
+
+	vdev->config->del_vqs(vdev);
+
+	kfree(vsock);
+}
+
+static struct virtio_device_id id_table[] = {
+	{ VIRTIO_ID_VSOCK, VIRTIO_DEV_ANY_ID },
+	{ 0 },
+};
+
+static unsigned int features[] = {
+};
+
+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,
+};
+
+static int __init virtio_vsock_init(void)
+{
+	int ret;
+
+	virtio_vsock_workqueue = alloc_workqueue("virtio_vsock", 0, 0);
+	if (!virtio_vsock_workqueue)
+		return -ENOMEM;
+	ret = register_virtio_driver(&virtio_vsock_driver);
+	if (ret)
+		destroy_workqueue(virtio_vsock_workqueue);
+	return ret;
+}
+
+static void __exit virtio_vsock_exit(void)
+{
+	unregister_virtio_driver(&virtio_vsock_driver);
+	destroy_workqueue(virtio_vsock_workqueue);
+}
+
+module_init(virtio_vsock_init);
+module_exit(virtio_vsock_exit);
+MODULE_LICENSE("GPL v2");
+MODULE_AUTHOR("Asias He");
+MODULE_DESCRIPTION("virtio transport for vsock");
+MODULE_DEVICE_TABLE(virtio, id_table);
-- 
2.5.5

From: Asias He <asias at redhat.com>

VM sockets vhost transport implementation.  This driver runs on the
host.

Signed-off-by: Asias He <asias at redhat.com>
Signed-off-by: Stefan Hajnoczi <stefanha at redhat.com>
---
v5:
 * Only take rx/tx virtqueues, userspace handles the other virtqueues
 * Explicitly skip instances without a CID when transferring packets
 * Add VHOST_VSOCK_START ioctl to being vhost virtqueue processing
 * Reset established connections when device is closed
v4:
 * Add MAINTAINERS file entry
 * virtqueue used len is now sizeof(pkt->hdr) + pkt->len instead of just
   pkt->len
 * checkpatch.pl cleanups
 * Clarify struct vhost_vsock locking
 * Add comments about optimization that disables virtqueue notify
 * Drop unused vhost_vsock_handle_ctl_kick()
 * Call wake_up() after decrementing total_tx_buf to prevent deadlock
v3:
 * Remove unneeded variable used to store return value
   (Fengguang Wu <fengguang.wu at intel.com> and Julia Lawall
   <julia.lawall at lip6.fr>)
v2:
 * Add missing total_tx_buf decrement
 * Support flexible rx/tx descriptor layout
 * Refuse to assign reserved CIDs
 * Refuse guest CID if already in use
 * Only accept correctly addressed packets
---
 MAINTAINERS           |   2 +
 drivers/vhost/vsock.c | 694 ++++++++++++++++++++++++++++++++++++++++++++++++++
 drivers/vhost/vsock.h |   5 +
 3 files changed, 701 insertions(+)
 create mode 100644 drivers/vhost/vsock.c
 create mode 100644 drivers/vhost/vsock.h

diff --git a/MAINTAINERS b/MAINTAINERS
index 1ed7364..7899c3c 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -11612,6 +11612,8 @@ F:	include/linux/virtio_vsock.h
 F:	include/uapi/linux/virtio_vsock.h
 F:	net/vmw_vsock/virtio_transport_common.c
 F:	net/vmw_vsock/virtio_transport.c
+F:	drivers/vhost/vsock.c
+F:	drivers/vhost/vsock.h
 
 VIRTUAL SERIO DEVICE DRIVER
 M:	Stephen Chandler Paul <thatslyude at gmail.com>
diff --git a/drivers/vhost/vsock.c b/drivers/vhost/vsock.c
new file mode 100644
index 0000000..8488d01
--- /dev/null
+++ b/drivers/vhost/vsock.c
@@ -0,0 +1,694 @@
+/*
+ * vhost transport for vsock
+ *
+ * Copyright (C) 2013-2015 Red Hat, Inc.
+ * Author: Asias He <asias at redhat.com>
+ *         Stefan Hajnoczi <stefanha at redhat.com>
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2.
+ */
+#include <linux/miscdevice.h>
+#include <linux/module.h>
+#include <linux/mutex.h>
+#include <net/sock.h>
+#include <linux/virtio_vsock.h>
+#include <linux/vhost.h>
+
+#include <net/af_vsock.h>
+#include "vhost.h"
+#include "vsock.h"
+
+#define VHOST_VSOCK_DEFAULT_HOST_CID	2
+
+enum {
+	VHOST_VSOCK_FEATURES = VHOST_FEATURES,
+};
+
+/* Used to track all the vhost_vsock instances on the system. */
+static LIST_HEAD(vhost_vsock_list);
+static DEFINE_MUTEX(vhost_vsock_mutex);
+
+struct vhost_vsock {
+	struct vhost_dev dev;
+	struct vhost_virtqueue vqs[2];
+
+	/* Link to global vhost_vsock_list, protected by vhost_vsock_mutex */
+	struct list_head list;
+
+	struct vhost_work send_pkt_work;
+	wait_queue_head_t send_wait;
+
+	/* Fields protected by vqs[VSOCK_VQ_RX].mutex */
+	struct list_head send_pkt_list;	/* host->guest pending packets */
+	u32 total_tx_buf;
+
+	u32 guest_cid;
+};
+
+static u32 vhost_transport_get_local_cid(void)
+{
+	return VHOST_VSOCK_DEFAULT_HOST_CID;
+}
+
+static struct vhost_vsock *vhost_vsock_get(u32 guest_cid)
+{
+	struct vhost_vsock *vsock;
+
+	mutex_lock(&vhost_vsock_mutex);
+	list_for_each_entry(vsock, &vhost_vsock_list, list) {
+		u32 other_cid = vsock->guest_cid;
+
+		/* Skip instances that have no CID yet */
+		if (other_cid == 0)
+			continue;
+
+		if (other_cid == guest_cid) {
+			mutex_unlock(&vhost_vsock_mutex);
+			return vsock;
+		}
+	}
+	mutex_unlock(&vhost_vsock_mutex);
+
+	return NULL;
+}
+
+static void
+vhost_transport_do_send_pkt(struct vhost_vsock *vsock,
+			    struct vhost_virtqueue *vq)
+{
+	bool added = false;
+
+	mutex_lock(&vq->mutex);
+
+	/* Avoid further vmexits, we're already processing the virtqueue */
+	vhost_disable_notify(&vsock->dev, vq);
+
+	for (;;) {
+		struct virtio_vsock_pkt *pkt;
+		struct iov_iter iov_iter;
+		unsigned out, in;
+		size_t nbytes;
+		size_t len;
+		int head;
+
+		if (list_empty(&vsock->send_pkt_list)) {
+			vhost_enable_notify(&vsock->dev, vq);
+			break;
+		}
+
+		head = vhost_get_vq_desc(vq, vq->iov, ARRAY_SIZE(vq->iov),
+					 &out, &in, NULL, NULL);
+		if (head < 0)
+			break;
+
+		if (head == vq->num) {
+			/* We cannot finish yet if more buffers snuck in while
+			 * re-enabling notify.
+			 */
+			if (unlikely(vhost_enable_notify(&vsock->dev, vq))) {
+				vhost_disable_notify(&vsock->dev, vq);
+				continue;
+			}
+			break;
+		}
+
+		pkt = list_first_entry(&vsock->send_pkt_list,
+				       struct virtio_vsock_pkt, list);
+		list_del_init(&pkt->list);
+
+		if (out) {
+			virtio_transport_free_pkt(pkt);
+			vq_err(vq, "Expected 0 output buffers, got %u\n", out);
+			break;
+		}
+
+		len = iov_length(&vq->iov[out], in);
+		iov_iter_init(&iov_iter, READ, &vq->iov[out], in, len);
+
+		nbytes = copy_to_iter(&pkt->hdr, sizeof(pkt->hdr), &iov_iter);
+		if (nbytes != sizeof(pkt->hdr)) {
+			virtio_transport_free_pkt(pkt);
+			vq_err(vq, "Faulted on copying pkt hdr\n");
+			break;
+		}
+
+		nbytes = copy_to_iter(pkt->buf, pkt->len, &iov_iter);
+		if (nbytes != pkt->len) {
+			virtio_transport_free_pkt(pkt);
+			vq_err(vq, "Faulted on copying pkt buf\n");
+			break;
+		}
+
+		vhost_add_used(vq, head, sizeof(pkt->hdr) + pkt->len);
+		added = true;
+
+		vsock->total_tx_buf -= pkt->len;
+
+		virtio_transport_free_pkt(pkt);
+	}
+	if (added)
+		vhost_signal(&vsock->dev, vq);
+	mutex_unlock(&vq->mutex);
+
+	if (added)
+		wake_up(&vsock->send_wait);
+}
+
+static void vhost_transport_send_pkt_work(struct vhost_work *work)
+{
+	struct vhost_virtqueue *vq;
+	struct vhost_vsock *vsock;
+
+	vsock = container_of(work, struct vhost_vsock, send_pkt_work);
+	vq = &vsock->vqs[VSOCK_VQ_RX];
+
+	vhost_transport_do_send_pkt(vsock, vq);
+}
+
+static int
+vhost_transport_send_one_pkt(struct vhost_vsock *vsock,
+			     struct virtio_vsock_pkt *pkt)
+{
+	struct vhost_virtqueue *vq = &vsock->vqs[VSOCK_VQ_RX];
+
+	/* Queue it up in vhost work */
+	mutex_lock(&vq->mutex);
+	list_add_tail(&pkt->list, &vsock->send_pkt_list);
+	vhost_work_queue(&vsock->dev, &vsock->send_pkt_work);
+	mutex_unlock(&vq->mutex);
+
+	return pkt->len;
+}
+
+static int
+vhost_transport_send_pkt_no_sock(struct virtio_vsock_pkt *pkt)
+{
+	struct vhost_vsock *vsock;
+
+	/* Find the vhost_vsock according to guest context id  */
+	vsock = vhost_vsock_get(le32_to_cpu(pkt->hdr.dst_cid));
+	if (!vsock) {
+		virtio_transport_free_pkt(pkt);
+		return -ENODEV;
+	}
+
+	return vhost_transport_send_one_pkt(vsock, pkt);
+}
+
+static int
+vhost_transport_send_pkt(struct vsock_sock *vsk,
+			 struct virtio_vsock_pkt_info *info)
+{
+	u32 src_cid, src_port, dst_cid, dst_port;
+	struct virtio_vsock_sock *vvs;
+	struct virtio_vsock_pkt *pkt;
+	struct vhost_virtqueue *vq;
+	struct vhost_vsock *vsock;
+	u32 pkt_len = info->pkt_len;
+	DEFINE_WAIT(wait);
+
+	src_cid = vhost_transport_get_local_cid();
+	src_port = vsk->local_addr.svm_port;
+	if (!info->remote_cid) {
+		dst_cid	= vsk->remote_addr.svm_cid;
+		dst_port = vsk->remote_addr.svm_port;
+	} else {
+		dst_cid = info->remote_cid;
+		dst_port = info->remote_port;
+	}
+
+	/* Find the vhost_vsock according to guest context id  */
+	vsock = vhost_vsock_get(dst_cid);
+	if (!vsock)
+		return -ENODEV;
+
+	vvs = vsk->trans;
+	vq = &vsock->vqs[VSOCK_VQ_RX];
+
+	/* we can send less than pkt_len bytes */
+	if (pkt_len > VIRTIO_VSOCK_DEFAULT_RX_BUF_SIZE)
+		pkt_len = VIRTIO_VSOCK_DEFAULT_RX_BUF_SIZE;
+
+	/* virtio_transport_get_credit might return less than pkt_len credit */
+	pkt_len = virtio_transport_get_credit(vvs, pkt_len);
+
+	/* Do not send zero length OP_RW pkt*/
+	if (pkt_len == 0 && info->op == VIRTIO_VSOCK_OP_RW)
+		return pkt_len;
+
+	/* Respect global tx buf limitation */
+	mutex_lock(&vq->mutex);
+	while (pkt_len + vsock->total_tx_buf > VIRTIO_VSOCK_MAX_TX_BUF_SIZE) {
+		prepare_to_wait_exclusive(&vsock->send_wait, &wait,
+					  TASK_UNINTERRUPTIBLE);
+		mutex_unlock(&vq->mutex);
+		schedule();
+		mutex_lock(&vq->mutex);
+		finish_wait(&vsock->send_wait, &wait);
+	}
+	vsock->total_tx_buf += pkt_len;
+	mutex_unlock(&vq->mutex);
+
+	pkt = virtio_transport_alloc_pkt(info, pkt_len,
+					 src_cid, src_port,
+					 dst_cid, dst_port);
+	if (!pkt) {
+		mutex_lock(&vq->mutex);
+		vsock->total_tx_buf -= pkt_len;
+		mutex_unlock(&vq->mutex);
+		virtio_transport_put_credit(vvs, pkt_len);
+		wake_up(&vsock->send_wait);
+		return -ENOMEM;
+	}
+
+	virtio_transport_inc_tx_pkt(vvs, pkt);
+
+	return vhost_transport_send_one_pkt(vsock, pkt);
+}
+
+static struct virtio_vsock_pkt *
+vhost_vsock_alloc_pkt(struct vhost_virtqueue *vq,
+		      unsigned int out, unsigned int in)
+{
+	struct virtio_vsock_pkt *pkt;
+	struct iov_iter iov_iter;
+	size_t nbytes;
+	size_t len;
+
+	if (in != 0) {
+		vq_err(vq, "Expected 0 input buffers, got %u\n", in);
+		return NULL;
+	}
+
+	pkt = kzalloc(sizeof(*pkt), GFP_KERNEL);
+	if (!pkt)
+		return NULL;
+
+	len = iov_length(vq->iov, out);
+	iov_iter_init(&iov_iter, WRITE, vq->iov, out, len);
+
+	nbytes = copy_from_iter(&pkt->hdr, sizeof(pkt->hdr), &iov_iter);
+	if (nbytes != sizeof(pkt->hdr)) {
+		vq_err(vq, "Expected %zu bytes for pkt->hdr, got %zu bytes\n",
+		       sizeof(pkt->hdr), nbytes);
+		kfree(pkt);
+		return NULL;
+	}
+
+	if (le16_to_cpu(pkt->hdr.type) == VIRTIO_VSOCK_TYPE_STREAM)
+		pkt->len = le32_to_cpu(pkt->hdr.len);
+
+	/* No payload */
+	if (!pkt->len)
+		return pkt;
+
+	/* The pkt is too big */
+	if (pkt->len > VIRTIO_VSOCK_MAX_PKT_BUF_SIZE) {
+		kfree(pkt);
+		return NULL;
+	}
+
+	pkt->buf = kmalloc(pkt->len, GFP_KERNEL);
+	if (!pkt->buf) {
+		kfree(pkt);
+		return NULL;
+	}
+
+	nbytes = copy_from_iter(pkt->buf, pkt->len, &iov_iter);
+	if (nbytes != pkt->len) {
+		vq_err(vq, "Expected %u byte payload, got %zu bytes\n",
+		       pkt->len, nbytes);
+		virtio_transport_free_pkt(pkt);
+		return NULL;
+	}
+
+	return pkt;
+}
+
+static void vhost_vsock_handle_tx_kick(struct vhost_work *work)
+{
+	struct vhost_virtqueue *vq = container_of(work, struct vhost_virtqueue,
+						  poll.work);
+	struct vhost_vsock *vsock = container_of(vq->dev, struct vhost_vsock,
+						 dev);
+	struct virtio_vsock_pkt *pkt;
+	int head;
+	unsigned int out, in;
+	bool added = false;
+
+	mutex_lock(&vq->mutex);
+	vhost_disable_notify(&vsock->dev, vq);
+	for (;;) {
+		head = vhost_get_vq_desc(vq, vq->iov, ARRAY_SIZE(vq->iov),
+					 &out, &in, NULL, NULL);
+		if (head < 0)
+			break;
+
+		if (head == vq->num) {
+			if (unlikely(vhost_enable_notify(&vsock->dev, vq))) {
+				vhost_disable_notify(&vsock->dev, vq);
+				continue;
+			}
+			break;
+		}
+
+		pkt = vhost_vsock_alloc_pkt(vq, out, in);
+		if (!pkt) {
+			vq_err(vq, "Faulted on pkt\n");
+			continue;
+		}
+
+		/* Only accept correctly addressed packets */
+		if (le32_to_cpu(pkt->hdr.src_cid) == vsock->guest_cid)
+			virtio_transport_recv_pkt(pkt);
+		else
+			virtio_transport_free_pkt(pkt);
+
+		vhost_add_used(vq, head, sizeof(pkt->hdr) + pkt->len);
+		added = true;
+	}
+	if (added)
+		vhost_signal(&vsock->dev, vq);
+	mutex_unlock(&vq->mutex);
+}
+
+static void vhost_vsock_handle_rx_kick(struct vhost_work *work)
+{
+	struct vhost_virtqueue *vq = container_of(work, struct vhost_virtqueue,
+						poll.work);
+	struct vhost_vsock *vsock = container_of(vq->dev, struct vhost_vsock,
+						 dev);
+
+	vhost_transport_do_send_pkt(vsock, vq);
+}
+
+static int vhost_vsock_start(struct vhost_vsock *vsock)
+{
+	size_t i;
+	int ret;
+
+	mutex_lock(&vsock->dev.mutex);
+
+	ret = vhost_dev_check_owner(&vsock->dev);
+	if (ret)
+		goto err;
+
+	for (i = 0; i < ARRAY_SIZE(vsock->vqs); i++) {
+		struct vhost_virtqueue *vq = &vsock->vqs[i];
+
+		mutex_lock(&vq->mutex);
+
+		if (!vhost_vq_access_ok(vq)) {
+			ret = -EFAULT;
+			mutex_unlock(&vq->mutex);
+			goto err_vq;
+		}
+
+		if (!vq->private_data) {
+			vq->private_data = vsock;
+			vhost_vq_init_access(vq);
+		}
+
+		mutex_unlock(&vq->mutex);
+	}
+
+	mutex_unlock(&vsock->dev.mutex);
+	return 0;
+
+err_vq:
+	for (i = 0; i < ARRAY_SIZE(vsock->vqs); i++) {
+		struct vhost_virtqueue *vq = &vsock->vqs[i];
+
+		mutex_lock(&vq->mutex);
+		vq->private_data = NULL;
+		mutex_unlock(&vq->mutex);
+	}
+err:
+	mutex_unlock(&vsock->dev.mutex);
+	return ret;
+}
+
+static void vhost_vsock_stop(struct vhost_vsock *vsock)
+{
+	size_t i;
+
+	mutex_lock(&vsock->dev.mutex);
+
+	for (i = 0; i < ARRAY_SIZE(vsock->vqs); i++) {
+		struct vhost_virtqueue *vq = &vsock->vqs[i];
+
+		mutex_lock(&vq->mutex);
+		vq->private_data = vsock;
+		mutex_unlock(&vq->mutex);
+	}
+
+	mutex_unlock(&vsock->dev.mutex);
+}
+
+static int vhost_vsock_dev_open(struct inode *inode, struct file *file)
+{
+	struct vhost_virtqueue **vqs;
+	struct vhost_vsock *vsock;
+	int ret;
+
+	vsock = kzalloc(sizeof(*vsock), GFP_KERNEL);
+	if (!vsock)
+		return -ENOMEM;
+
+	vqs = kmalloc_array(ARRAY_SIZE(vsock->vqs), sizeof(*vqs), GFP_KERNEL);
+	if (!vqs) {
+		ret = -ENOMEM;
+		goto out;
+	}
+
+	vqs[VSOCK_VQ_TX] = &vsock->vqs[VSOCK_VQ_TX];
+	vqs[VSOCK_VQ_RX] = &vsock->vqs[VSOCK_VQ_RX];
+	vsock->vqs[VSOCK_VQ_TX].handle_kick = vhost_vsock_handle_tx_kick;
+	vsock->vqs[VSOCK_VQ_RX].handle_kick = vhost_vsock_handle_rx_kick;
+
+	vhost_dev_init(&vsock->dev, vqs, ARRAY_SIZE(vsock->vqs));
+
+	file->private_data = vsock;
+	init_waitqueue_head(&vsock->send_wait);
+	INIT_LIST_HEAD(&vsock->send_pkt_list);
+	vhost_work_init(&vsock->send_pkt_work, vhost_transport_send_pkt_work);
+
+	mutex_lock(&vhost_vsock_mutex);
+	list_add_tail(&vsock->list, &vhost_vsock_list);
+	mutex_unlock(&vhost_vsock_mutex);
+	return 0;
+
+out:
+	kfree(vsock);
+	return ret;
+}
+
+static void vhost_vsock_flush(struct vhost_vsock *vsock)
+{
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(vsock->vqs); i++)
+		if (vsock->vqs[i].handle_kick)
+			vhost_poll_flush(&vsock->vqs[i].poll);
+	vhost_work_flush(&vsock->dev, &vsock->send_pkt_work);
+}
+
+static void vhost_vsock_reset_orphans(struct sock *sk)
+{
+	struct vsock_sock *vsk = vsock_sk(sk);
+
+	lock_sock(sk);
+	if (!vhost_vsock_get(vsk->local_addr.svm_cid)) {
+		sk->sk_state = SS_UNCONNECTED;
+		sk->sk_err = ECONNRESET;
+		sk->sk_error_report(sk);
+	}
+	release_sock(sk);
+}
+
+static int vhost_vsock_dev_release(struct inode *inode, struct file *file)
+{
+	struct vhost_vsock *vsock = file->private_data;
+
+	mutex_lock(&vhost_vsock_mutex);
+	list_del(&vsock->list);
+	mutex_unlock(&vhost_vsock_mutex);
+
+	/* Iterating over all connections for all CIDs to find orphans is
+	 * inefficient.  Room for improvement here. */
+	vsock_for_each_connected_socket(vhost_vsock_reset_orphans);
+
+	vhost_vsock_stop(vsock);
+	vhost_vsock_flush(vsock);
+	vhost_dev_stop(&vsock->dev);
+	vhost_dev_cleanup(&vsock->dev, false);
+	kfree(vsock->dev.vqs);
+	kfree(vsock);
+	return 0;
+}
+
+static int vhost_vsock_set_cid(struct vhost_vsock *vsock, u32 guest_cid)
+{
+	struct vhost_vsock *other;
+
+	/* Refuse reserved CIDs */
+	if (guest_cid <= VMADDR_CID_HOST)
+		return -EINVAL;
+
+	/* Refuse if CID is already in use */
+	other = vhost_vsock_get(guest_cid);
+	if (other && other != vsock)
+		return -EADDRINUSE;
+
+	mutex_lock(&vhost_vsock_mutex);
+	vsock->guest_cid = guest_cid;
+	mutex_unlock(&vhost_vsock_mutex);
+
+	return 0;
+}
+
+static int vhost_vsock_set_features(struct vhost_vsock *vsock, u64 features)
+{
+	struct vhost_virtqueue *vq;
+	int i;
+
+	if (features & ~VHOST_VSOCK_FEATURES)
+		return -EOPNOTSUPP;
+
+	mutex_lock(&vsock->dev.mutex);
+	if ((features & (1 << VHOST_F_LOG_ALL)) &&
+	    !vhost_log_access_ok(&vsock->dev)) {
+		mutex_unlock(&vsock->dev.mutex);
+		return -EFAULT;
+	}
+
+	for (i = 0; i < ARRAY_SIZE(vsock->vqs); i++) {
+		vq = &vsock->vqs[i];
+		mutex_lock(&vq->mutex);
+		vq->acked_features = features;
+		mutex_unlock(&vq->mutex);
+	}
+	mutex_unlock(&vsock->dev.mutex);
+	return 0;
+}
+
+static long vhost_vsock_dev_ioctl(struct file *f, unsigned int ioctl,
+				  unsigned long arg)
+{
+	struct vhost_vsock *vsock = f->private_data;
+	void __user *argp = (void __user *)arg;
+	u64 __user *featurep = argp;
+	u32 __user *cidp = argp;
+	u32 guest_cid;
+	u64 features;
+	int r;
+
+	switch (ioctl) {
+	case VHOST_VSOCK_SET_GUEST_CID:
+		if (get_user(guest_cid, cidp))
+			return -EFAULT;
+		return vhost_vsock_set_cid(vsock, guest_cid);
+	case VHOST_VSOCK_START:
+		return vhost_vsock_start(vsock);
+	case VHOST_GET_FEATURES:
+		features = VHOST_VSOCK_FEATURES;
+		if (copy_to_user(featurep, &features, sizeof(features)))
+			return -EFAULT;
+		return 0;
+	case VHOST_SET_FEATURES:
+		if (copy_from_user(&features, featurep, sizeof(features)))
+			return -EFAULT;
+		return vhost_vsock_set_features(vsock, features);
+	default:
+		mutex_lock(&vsock->dev.mutex);
+		r = vhost_dev_ioctl(&vsock->dev, ioctl, argp);
+		if (r == -ENOIOCTLCMD)
+			r = vhost_vring_ioctl(&vsock->dev, ioctl, argp);
+		else
+			vhost_vsock_flush(vsock);
+		mutex_unlock(&vsock->dev.mutex);
+		return r;
+	}
+}
+
+static const struct file_operations vhost_vsock_fops = {
+	.owner          = THIS_MODULE,
+	.open           = vhost_vsock_dev_open,
+	.release        = vhost_vsock_dev_release,
+	.llseek		= noop_llseek,
+	.unlocked_ioctl = vhost_vsock_dev_ioctl,
+};
+
+static struct miscdevice vhost_vsock_misc = {
+	.minor = MISC_DYNAMIC_MINOR,
+	.name = "vhost-vsock",
+	.fops = &vhost_vsock_fops,
+};
+
+static struct virtio_transport vhost_transport = {
+	.transport = {
+		.get_local_cid            = vhost_transport_get_local_cid,
+
+		.init                     = virtio_transport_do_socket_init,
+		.destruct                 = virtio_transport_destruct,
+		.release                  = virtio_transport_release,
+		.connect                  = virtio_transport_connect,
+		.shutdown                 = virtio_transport_shutdown,
+
+		.dgram_enqueue            = virtio_transport_dgram_enqueue,
+		.dgram_dequeue            = virtio_transport_dgram_dequeue,
+		.dgram_bind               = virtio_transport_dgram_bind,
+		.dgram_allow              = virtio_transport_dgram_allow,
+
+		.stream_enqueue           = virtio_transport_stream_enqueue,
+		.stream_dequeue           = virtio_transport_stream_dequeue,
+		.stream_has_data          = virtio_transport_stream_has_data,
+		.stream_has_space         = virtio_transport_stream_has_space,
+		.stream_rcvhiwat          = virtio_transport_stream_rcvhiwat,
+		.stream_is_active         = virtio_transport_stream_is_active,
+		.stream_allow             = virtio_transport_stream_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,
+		.notify_recv_pre_block    = virtio_transport_notify_recv_pre_block,
+		.notify_recv_pre_dequeue  = virtio_transport_notify_recv_pre_dequeue,
+		.notify_recv_post_dequeue = virtio_transport_notify_recv_post_dequeue,
+		.notify_send_init         = virtio_transport_notify_send_init,
+		.notify_send_pre_block    = virtio_transport_notify_send_pre_block,
+		.notify_send_pre_enqueue  = virtio_transport_notify_send_pre_enqueue,
+		.notify_send_post_enqueue = virtio_transport_notify_send_post_enqueue,
+
+		.set_buffer_size          = virtio_transport_set_buffer_size,
+		.set_min_buffer_size      = virtio_transport_set_min_buffer_size,
+		.set_max_buffer_size      = virtio_transport_set_max_buffer_size,
+		.get_buffer_size          = virtio_transport_get_buffer_size,
+		.get_min_buffer_size      = virtio_transport_get_min_buffer_size,
+		.get_max_buffer_size      = virtio_transport_get_max_buffer_size,
+	},
+
+	.send_pkt = vhost_transport_send_pkt,
+	.send_pkt_no_sock = vhost_transport_send_pkt_no_sock,
+};
+
+static int __init vhost_vsock_init(void)
+{
+	int ret;
+
+	ret = vsock_core_init(&vhost_transport.transport);
+	if (ret < 0)
+		return ret;
+	return misc_register(&vhost_vsock_misc);
+};
+
+static void __exit vhost_vsock_exit(void)
+{
+	misc_deregister(&vhost_vsock_misc);
+	vsock_core_exit();
+};
+
+module_init(vhost_vsock_init);
+module_exit(vhost_vsock_exit);
+MODULE_LICENSE("GPL v2");
+MODULE_AUTHOR("Asias He");
+MODULE_DESCRIPTION("vhost transport for vsock ");
diff --git a/drivers/vhost/vsock.h b/drivers/vhost/vsock.h
new file mode 100644
index 0000000..173f9fc
--- /dev/null
+++ b/drivers/vhost/vsock.h
@@ -0,0 +1,5 @@
+#ifndef VHOST_VSOCK_H
+#define VHOST_VSOCK_H
+#define VHOST_VSOCK_SET_GUEST_CID	_IOW(VHOST_VIRTIO, 0x60, __u32)
+#define VHOST_VSOCK_START		_IO(VHOST_VIRTIO, 0x61)
+#endif
-- 
2.5.5

From: Asias He <asias at redhat.com>

Enable virtio-vsock and vhost-vsock.

Signed-off-by: Asias He <asias at redhat.com>
Signed-off-by: Stefan Hajnoczi <stefanha at redhat.com>
---
v4:
 * Make checkpatch.pl happy with longer option description
 * Clarify dependency on virtio rather than QEMU as suggested by Alex
   Bennee
v3:
 * Don't put vhost vsock driver into staging
 * Add missing Kconfig dependencies (Arnd Bergmann <arnd at arndb.de>)
---
 drivers/vhost/Kconfig  | 15 +++++++++++++++
 drivers/vhost/Makefile |  4 ++++
 net/vmw_vsock/Kconfig  | 19 +++++++++++++++++++
 net/vmw_vsock/Makefile |  2 ++
 4 files changed, 40 insertions(+)

diff --git a/drivers/vhost/Kconfig b/drivers/vhost/Kconfig
index 533eaf0..d7aae9e 100644
--- a/drivers/vhost/Kconfig
+++ b/drivers/vhost/Kconfig
@@ -21,6 +21,21 @@ config VHOST_SCSI
 	Say M here to enable the vhost_scsi TCM fabric module
 	for use with virtio-scsi guests
 
+config VHOST_VSOCK
+	tristate "vhost virtio-vsock driver"
+	depends on VSOCKETS && EVENTFD
+	select VIRTIO_VSOCKETS_COMMON
+	select VHOST
+	select VHOST_RING
+	default n
+	---help---
+	This kernel module can be loaded in the host kernel to provide AF_VSOCK
+	sockets for communicating with guests.  The guests must have the
+	virtio_transport.ko driver loaded to use the virtio-vsock device.
+
+	To compile this driver as a module, choose M here: the module will be called
+	vhost_vsock.
+
 config VHOST_RING
 	tristate
 	---help---
diff --git a/drivers/vhost/Makefile b/drivers/vhost/Makefile
index e0441c3..6b012b9 100644
--- a/drivers/vhost/Makefile
+++ b/drivers/vhost/Makefile
@@ -4,5 +4,9 @@ vhost_net-y := net.o
 obj-$(CONFIG_VHOST_SCSI) += vhost_scsi.o
 vhost_scsi-y := scsi.o
 
+obj-$(CONFIG_VHOST_VSOCK) += vhost_vsock.o
+vhost_vsock-y := vsock.o
+
 obj-$(CONFIG_VHOST_RING) += vringh.o
+
 obj-$(CONFIG_VHOST)	+= vhost.o
diff --git a/net/vmw_vsock/Kconfig b/net/vmw_vsock/Kconfig
index 14810ab..f27e74b 100644
--- a/net/vmw_vsock/Kconfig
+++ b/net/vmw_vsock/Kconfig
@@ -26,3 +26,22 @@ config VMWARE_VMCI_VSOCKETS
 
 	  To compile this driver as a module, choose M here: the module
 	  will be called vmw_vsock_vmci_transport. If unsure, say N.
+
+config VIRTIO_VSOCKETS
+	tristate "virtio transport for Virtual Sockets"
+	depends on VSOCKETS && VIRTIO
+	select VIRTIO_VSOCKETS_COMMON
+	help
+	  This module implements a virtio transport for Virtual Sockets.
+
+	  Enable this transport if your Virtual Machine host supports Virtual
+	  Sockets over virtio.
+
+	  To compile this driver as a module, choose M here: the module
+	  will be called virtio_vsock_transport. If unsure, say N.
+
+config VIRTIO_VSOCKETS_COMMON
+       tristate
+       ---help---
+         This option is selected by any driver which needs to access
+         the virtio_vsock.
diff --git a/net/vmw_vsock/Makefile b/net/vmw_vsock/Makefile
index 2ce52d7..cf4c294 100644
--- a/net/vmw_vsock/Makefile
+++ b/net/vmw_vsock/Makefile
@@ -1,5 +1,7 @@
 obj-$(CONFIG_VSOCKETS) += vsock.o
 obj-$(CONFIG_VMWARE_VMCI_VSOCKETS) += vmw_vsock_vmci_transport.o
+obj-$(CONFIG_VIRTIO_VSOCKETS) += virtio_transport.o
+obj-$(CONFIG_VIRTIO_VSOCKETS_COMMON) += virtio_transport_common.o
 
 vsock-y += af_vsock.o vsock_addr.o
 
-- 
2.5.5

On Fri, 2016-04-01 at 15:23 +0100, Stefan Hajnoczi
wrote:
> This series is based on Michael Tsirkin's vhost branch (v4.5-rc6).
> 
> I'm about to process Claudio Imbrenda's locking fixes for
virtio-vsock but
> first I want to share the latest version of the code.??Several people are
> playing with vsock now so sharing the latest code should avoid duplicate
work.
Thanks for this, I've been using it in my project and it mostly seems
fine.

One wrinkle I came across, which I'm not sure if it is by design or a
problem is that I can see this sequence coming from the guest (with
other activity in between):

? ? 1) OP_SHUTDOWN w/ flags == SHUTDOWN_RX
????2) OP_SHUTDOWN w/ flags == SHUTDOWN_TX
????3) OP_SHUTDOWN w/ flags == SHUTDOWN_TX|SHUTDOWN_RX

I orignally had my backend close things down at #2, however this meant
that when #3 arrived it was for a non-existent socket (or, worse, an
active one if the ports got reused). I checked v5 of the spec
proposal[0] which says:
    If these bits are set and there are no more virtqueue buffers
    pending the socket is disconnected.

but I'm not entirely sure if this behaviour contradicts this or not
(the bits have both been set at #2, but not at the same time).

BTW, how does one tell if there are no more virtqueue buffers pending
or not while processing the op?

Another thing I noticed, which is really more to do with the generic
AF_VSOCK bits than anything to do with your patches is that there is no
limitations on which vsock ports a non-privileged user can bind to and
relatedly that there is no netns support so e.g. users in unproivileged
containers can bind to any vsock port and talk to the host, which might
be undesirable. For my use for now I just went with the big hammer
approach of denying access from anything other than init_net
namespace[1] while I consider what the right answer is.

Ian.

[0]?http://thread.gmane.org/gmane.comp.emulators.virtio.devel/1092
[1]
>From 366c9c42afb9bd54f92f72518470c09e46f12e88 Mon Sep 17 00:00:00 2001
From: Ian Campbell <ian.campbell at docker.com>
Date: Mon, 4 Apr 2016 14:50:10 +0100
Subject: [PATCH] VSOCK: Only allow host network namespace to use AF_VSOCK.

The VSOCK addressing schema does not really lend itself to simply creating an
alternative end point address within a namespace.

Signed-off-by: Ian Campbell <ian.campbell at docker.com>
---
?net/vmw_vsock/af_vsock.c | 3 +++
?1 file changed, 3 insertions(+)

diff --git a/net/vmw_vsock/af_vsock.c b/net/vmw_vsock/af_vsock.c
index 1e5f5ed..cdb3dd3 100644
--- a/net/vmw_vsock/af_vsock.c
+++ b/net/vmw_vsock/af_vsock.c
@@ -1840,6 +1840,9 @@ static const struct proto_ops vsock_stream_ops = {
?static int vsock_create(struct net *net, struct socket *sock,
?			int protocol, int kern)
?{
+	if (!net_eq(net, &init_net))
+		return -EAFNOSUPPORT;
+
?	if (!sock)
?		return -EINVAL;
?
--?
2.8.0.rc3

On Fri, Apr 08, 2016 at 04:35:05PM +0100, Ian Campbell
wrote:
> On Fri, 2016-04-01 at 15:23 +0100, Stefan Hajnoczi wrote:
> > This series is based on Michael Tsirkin's vhost branch (v4.5-rc6).
> > 
> > I'm about to process Claudio Imbrenda's locking fixes for
virtio-vsock but
> > first I want to share the latest version of the code.??Several people
are
> > playing with vsock now so sharing the latest code should avoid
duplicate work.
> 
> Thanks for this, I've been using it in my project and it mostly seems
> fine.
> 
> One wrinkle I came across, which I'm not sure if it is by design or a
> problem is that I can see this sequence coming from the guest (with
> other activity in between):
> 
> ? ? 1) OP_SHUTDOWN w/ flags == SHUTDOWN_RX
> ????2) OP_SHUTDOWN w/ flags == SHUTDOWN_TX
> ????3) OP_SHUTDOWN w/ flags == SHUTDOWN_TX|SHUTDOWN_RX
> 
> I orignally had my backend close things down at #2, however this meant
> that when #3 arrived it was for a non-existent socket (or, worse, an
> active one if the ports got reused). I checked v5 of the spec
> proposal[0] which says:
>     If these bits are set and there are no more virtqueue buffers
>     pending the socket is disconnected.
> 
> but I'm not entirely sure if this behaviour contradicts this or not
> (the bits have both been set at #2, but not at the same time).
> 
> BTW, how does one tell if there are no more virtqueue buffers pending
> or not while processing the op?
#2 is odd.  The shutdown bits are sticky so they cannot be cleared once
set.  I would have expected just #1 and #3.  The behavior you observe
look like a bug.

The spec text does not convey the meaning of OP_SHUTDOWN well.
OP_SHUTDOWN SHUTDOWN_TX|SHUTDOWN_RX means no further rx/tx is possible
for this connection.  "there are no more virtqueue buffers pending the
socket" really means that this isn't an immediate close from the
perspective of the application.  If the application still has unread rx
buffers then the socket stays readable until the rx data has been fully
read.
> Another thing I noticed, which is really more to do with the generic
> AF_VSOCK bits than anything to do with your patches is that there is no
> limitations on which vsock ports a non-privileged user can bind to and
> relatedly that there is no netns support so e.g. users in unproivileged
> containers can bind to any vsock port and talk to the host, which might
> be undesirable. For my use for now I just went with the big hammer
> approach of denying access from anything other than init_net
> namespace[1] while I consider what the right answer is.
From the vhost point of view each netns should have its own AF_VSOCK
namespace.  This way two containers could act as "the host" (CID 2)
for
their respective guests.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 473 bytes
Desc: not available
URL:
<http://lists.linuxfoundation.org/pipermail/virtualization/attachments/20160411/ccc32408/attachment.sig>