Merge tag 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mst/vhost

Pull virtio/vhost updates from Michael Tsirkin:

 - new vsock device support in host and guest

 - platform IOMMU support in host and guest, including compatibility
   quirks for legacy systems.

 - misc fixes and cleanups.

* tag 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mst/vhost:
  VSOCK: Use kvfree()
  vhost: split out vringh Kconfig
  vhost: detect 32 bit integer wrap around
  vhost: new device IOTLB API
  vhost: drop vringh dependency
  vhost: convert pre sorted vhost memory array to interval tree
  vhost: introduce vhost memory accessors
  VSOCK: Add Makefile and Kconfig
  VSOCK: Introduce vhost_vsock.ko
  VSOCK: Introduce virtio_transport.ko
  VSOCK: Introduce virtio_vsock_common.ko
  VSOCK: defer sock removal to transports
  VSOCK: transport-specific vsock_transport functions
  vhost: drop vringh dependency
  vop: pull in vhost Kconfig
  virtio: new feature to detect IOMMU device quirk
  balloon: check the number of available pages in leak balloon
  vhost: lockless enqueuing
  vhost: simplify work flushing

12 files changed, 1634 insertions, 194 deletions

diff --git a/drivers/Makefile b/drivers/Makefile
index 954824438fd1..53abb4a5f736 100644
--- a/drivers/Makefile
+++ b/drivers/Makefile
@@ -138,6 +138,7 @@ obj-$(CONFIG_OF)		+= of/
 obj-$(CONFIG_SSB)		+= ssb/
 obj-$(CONFIG_BCMA)		+= bcma/
 obj-$(CONFIG_VHOST_RING)	+= vhost/
+obj-$(CONFIG_VHOST)		+= vhost/
 obj-$(CONFIG_VLYNQ)		+= vlynq/
 obj-$(CONFIG_STAGING)		+= staging/
 obj-y				+= platform/
diff --git a/drivers/misc/mic/Kconfig b/drivers/misc/mic/Kconfig
index 89e5917e1c33..6fd9d367dea7 100644
--- a/drivers/misc/mic/Kconfig
+++ b/drivers/misc/mic/Kconfig
@@ -146,3 +146,7 @@ config VOP
 	  More information about the Intel MIC family as well as the Linux
 	  OS and tools for MIC to use with this driver are available from
 	  <http://software.intel.com/en-us/mic-developer>.
+
+if VOP
+source "drivers/vhost/Kconfig.vringh"
+endif
diff --git a/drivers/net/caif/Kconfig b/drivers/net/caif/Kconfig
index 547098086773..f81df91a9ce1 100644
--- a/drivers/net/caif/Kconfig
+++ b/drivers/net/caif/Kconfig
@@ -52,5 +52,5 @@ config CAIF_VIRTIO
 	The caif driver for CAIF over Virtio.
 
 if CAIF_VIRTIO
-source "drivers/vhost/Kconfig"
+source "drivers/vhost/Kconfig.vringh"
 endif
diff --git a/drivers/vhost/Kconfig b/drivers/vhost/Kconfig
index 533eaf04f12f..40764ecad9ce 100644
--- a/drivers/vhost/Kconfig
+++ b/drivers/vhost/Kconfig
@@ -2,7 +2,6 @@ config VHOST_NET
 	tristate "Host kernel accelerator for virtio net"
 	depends on NET && EVENTFD && (TUN || !TUN) && (MACVTAP || !MACVTAP)
 	select VHOST
-	select VHOST_RING
 	---help---
 	  This kernel module can be loaded in host kernel to accelerate
 	  guest networking with virtio_net. Not to be confused with virtio_net
@@ -15,17 +14,24 @@ config VHOST_SCSI
 	tristate "VHOST_SCSI TCM fabric driver"
 	depends on TARGET_CORE && EVENTFD && m
 	select VHOST
-	select VHOST_RING
 	default n
 	---help---
 	Say M here to enable the vhost_scsi TCM fabric module
 	for use with virtio-scsi guests
 
-config VHOST_RING
-	tristate
+config VHOST_VSOCK
+	tristate "vhost virtio-vsock driver"
+	depends on VSOCKETS && EVENTFD
+	select VIRTIO_VSOCKETS_COMMON
+	select VHOST
+	default n
 	---help---
-	  This option is selected by any driver which needs to access
-	  the host side of a virtio ring.
+	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
 	tristate
diff --git a/drivers/vhost/Kconfig.vringh b/drivers/vhost/Kconfig.vringh
new file mode 100644
index 000000000000..6a4490c09d7f
--- /dev/null
+++ b/drivers/vhost/Kconfig.vringh
@@ -0,0 +1,5 @@
+config VHOST_RING
+	tristate
+	---help---
+	  This option is selected by any driver which needs to access
+	  the host side of a virtio ring.
diff --git a/drivers/vhost/Makefile b/drivers/vhost/Makefile
index e0441c34db1c..6b012b986b57 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/drivers/vhost/net.c b/drivers/vhost/net.c
index e032ca397371..5dc128a8da83 100644
--- a/drivers/vhost/net.c
+++ b/drivers/vhost/net.c
@@ -61,7 +61,8 @@ MODULE_PARM_DESC(experimental_zcopytx, "Enable Zero Copy TX;"
 enum {
 	VHOST_NET_FEATURES = VHOST_FEATURES |
 			 (1ULL << VHOST_NET_F_VIRTIO_NET_HDR) |
-			 (1ULL << VIRTIO_NET_F_MRG_RXBUF)
+			 (1ULL << VIRTIO_NET_F_MRG_RXBUF) |
+			 (1ULL << VIRTIO_F_IOMMU_PLATFORM)
 };
 
 enum {
@@ -334,7 +335,7 @@ static int vhost_net_tx_get_vq_desc(struct vhost_net *net,
 {
 	unsigned long uninitialized_var(endtime);
 	int r = vhost_get_vq_desc(vq, vq->iov, ARRAY_SIZE(vq->iov),
-				    out_num, in_num, NULL, NULL);
+				  out_num, in_num, NULL, NULL);
 
 	if (r == vq->num && vq->busyloop_timeout) {
 		preempt_disable();
@@ -344,7 +345,7 @@ static int vhost_net_tx_get_vq_desc(struct vhost_net *net,
 			cpu_relax_lowlatency();
 		preempt_enable();
 		r = vhost_get_vq_desc(vq, vq->iov, ARRAY_SIZE(vq->iov),
-					out_num, in_num, NULL, NULL);
+				      out_num, in_num, NULL, NULL);
 	}
 
 	return r;
@@ -377,6 +378,9 @@ static void handle_tx(struct vhost_net *net)
 	if (!sock)
 		goto out;
 
+	if (!vq_iotlb_prefetch(vq))
+		goto out;
+
 	vhost_disable_notify(&net->dev, vq);
 
 	hdr_size = nvq->vhost_hlen;
@@ -652,6 +656,10 @@ static void handle_rx(struct vhost_net *net)
 	sock = vq->private_data;
 	if (!sock)
 		goto out;
+
+	if (!vq_iotlb_prefetch(vq))
+		goto out;
+
 	vhost_disable_notify(&net->dev, vq);
 	vhost_net_disable_vq(net, vq);
 
@@ -1052,20 +1060,20 @@ static long vhost_net_reset_owner(struct vhost_net *n)
 	struct socket *tx_sock = NULL;
 	struct socket *rx_sock = NULL;
 	long err;
-	struct vhost_memory *memory;
+	struct vhost_umem *umem;
 
 	mutex_lock(&n->dev.mutex);
 	err = vhost_dev_check_owner(&n->dev);
 	if (err)
 		goto done;
-	memory = vhost_dev_reset_owner_prepare();
-	if (!memory) {
+	umem = vhost_dev_reset_owner_prepare();
+	if (!umem) {
 		err = -ENOMEM;
 		goto done;
 	}
 	vhost_net_stop(n, &tx_sock, &rx_sock);
 	vhost_net_flush(n);
-	vhost_dev_reset_owner(&n->dev, memory);
+	vhost_dev_reset_owner(&n->dev, umem);
 	vhost_net_vq_reset(n);
 done:
 	mutex_unlock(&n->dev.mutex);
@@ -1096,10 +1104,14 @@ static int vhost_net_set_features(struct vhost_net *n, u64 features)
 	}
 	mutex_lock(&n->dev.mutex);
 	if ((features & (1 << VHOST_F_LOG_ALL)) &&
-	    !vhost_log_access_ok(&n->dev)) {
-		mutex_unlock(&n->dev.mutex);
-		return -EFAULT;
+	    !vhost_log_access_ok(&n->dev))
+		goto out_unlock;
+
+	if ((features & (1ULL << VIRTIO_F_IOMMU_PLATFORM))) {
+		if (vhost_init_device_iotlb(&n->dev, true))
+			goto out_unlock;
 	}
+
 	for (i = 0; i < VHOST_NET_VQ_MAX; ++i) {
 		mutex_lock(&n->vqs[i].vq.mutex);
 		n->vqs[i].vq.acked_features = features;
@@ -1109,6 +1121,10 @@ static int vhost_net_set_features(struct vhost_net *n, u64 features)
 	}
 	mutex_unlock(&n->dev.mutex);
 	return 0;
+
+out_unlock:
+	mutex_unlock(&n->dev.mutex);
+	return -EFAULT;
 }
 
 static long vhost_net_set_owner(struct vhost_net *n)
@@ -1182,9 +1198,40 @@ static long vhost_net_compat_ioctl(struct file *f, unsigned int ioctl,
 }
 #endif
 
+static ssize_t vhost_net_chr_read_iter(struct kiocb *iocb, struct iov_iter *to)
+{
+	struct file *file = iocb->ki_filp;
+	struct vhost_net *n = file->private_data;
+	struct vhost_dev *dev = &n->dev;
+	int noblock = file->f_flags & O_NONBLOCK;
+
+	return vhost_chr_read_iter(dev, to, noblock);
+}
+
+static ssize_t vhost_net_chr_write_iter(struct kiocb *iocb,
+					struct iov_iter *from)
+{
+	struct file *file = iocb->ki_filp;
+	struct vhost_net *n = file->private_data;
+	struct vhost_dev *dev = &n->dev;
+
+	return vhost_chr_write_iter(dev, from);
+}
+
+static unsigned int vhost_net_chr_poll(struct file *file, poll_table *wait)
+{
+	struct vhost_net *n = file->private_data;
+	struct vhost_dev *dev = &n->dev;
+
+	return vhost_chr_poll(file, dev, wait);
+}
+
 static const struct file_operations vhost_net_fops = {
 	.owner          = THIS_MODULE,
 	.release        = vhost_net_release,
+	.read_iter      = vhost_net_chr_read_iter,
+	.write_iter     = vhost_net_chr_write_iter,
+	.poll           = vhost_net_chr_poll,
 	.unlocked_ioctl = vhost_net_ioctl,
 #ifdef CONFIG_COMPAT
 	.compat_ioctl   = vhost_net_compat_ioctl,
diff --git a/drivers/vhost/vhost.c b/drivers/vhost/vhost.c
index 669fef1e2bb6..c6f2d89c0e97 100644
--- a/drivers/vhost/vhost.c
+++ b/drivers/vhost/vhost.c
@@ -27,6 +27,7 @@
 #include <linux/cgroup.h>
 #include <linux/module.h>
 #include <linux/sort.h>
+#include <linux/interval_tree_generic.h>
 
 #include "vhost.h"
 
@@ -34,6 +35,10 @@ static ushort max_mem_regions = 64;
 module_param(max_mem_regions, ushort, 0444);
 MODULE_PARM_DESC(max_mem_regions,
 	"Maximum number of memory regions in memory map. (default: 64)");
+static int max_iotlb_entries = 2048;
+module_param(max_iotlb_entries, int, 0444);
+MODULE_PARM_DESC(max_iotlb_entries,
+	"Maximum number of iotlb entries. (default: 2048)");
 
 enum {
 	VHOST_MEMORY_F_LOG = 0x1,
@@ -42,6 +47,10 @@ enum {
 #define vhost_used_event(vq) ((__virtio16 __user *)&vq->avail->ring[vq->num])
 #define vhost_avail_event(vq) ((__virtio16 __user *)&vq->used->ring[vq->num])
 
+INTERVAL_TREE_DEFINE(struct vhost_umem_node,
+		     rb, __u64, __subtree_last,
+		     START, LAST, , vhost_umem_interval_tree);
+
 #ifdef CONFIG_VHOST_CROSS_ENDIAN_LEGACY
 static void vhost_disable_cross_endian(struct vhost_virtqueue *vq)
 {
@@ -131,6 +140,19 @@ static void vhost_reset_is_le(struct vhost_virtqueue *vq)
 	vq->is_le = virtio_legacy_is_little_endian();
 }
 
+struct vhost_flush_struct {
+	struct vhost_work work;
+	struct completion wait_event;
+};
+
+static void vhost_flush_work(struct vhost_work *work)
+{
+	struct vhost_flush_struct *s;
+
+	s = container_of(work, struct vhost_flush_struct, work);
+	complete(&s->wait_event);
+}
+
 static void vhost_poll_func(struct file *file, wait_queue_head_t *wqh,
 			    poll_table *pt)
 {
@@ -155,11 +177,9 @@ static int vhost_poll_wakeup(wait_queue_t *wait, unsigned mode, int sync,
 
 void vhost_work_init(struct vhost_work *work, vhost_work_fn_t fn)
 {
-	INIT_LIST_HEAD(&work->node);
+	clear_bit(VHOST_WORK_QUEUED, &work->flags);
 	work->fn = fn;
 	init_waitqueue_head(&work->done);
-	work->flushing = 0;
-	work->queue_seq = work->done_seq = 0;
 }
 EXPORT_SYMBOL_GPL(vhost_work_init);
 
@@ -211,31 +231,17 @@ void vhost_poll_stop(struct vhost_poll *poll)
 }
 EXPORT_SYMBOL_GPL(vhost_poll_stop);
 
-static bool vhost_work_seq_done(struct vhost_dev *dev, struct vhost_work *work,
-				unsigned seq)
-{
-	int left;
-
-	spin_lock_irq(&dev->work_lock);
-	left = seq - work->done_seq;
-	spin_unlock_irq(&dev->work_lock);
-	return left <= 0;
-}
-
 void vhost_work_flush(struct vhost_dev *dev, struct vhost_work *work)
 {
-	unsigned seq;
-	int flushing;
+	struct vhost_flush_struct flush;
 
-	spin_lock_irq(&dev->work_lock);
-	seq = work->queue_seq;
-	work->flushing++;
-	spin_unlock_irq(&dev->work_lock);
-	wait_event(work->done, vhost_work_seq_done(dev, work, seq));
-	spin_lock_irq(&dev->work_lock);
-	flushing = --work->flushing;
-	spin_unlock_irq(&dev->work_lock);
-	BUG_ON(flushing < 0);
+	if (dev->worker) {
+		init_completion(&flush.wait_event);
+		vhost_work_init(&flush.work, vhost_flush_work);
+
+		vhost_work_queue(dev, &flush.work);
+		wait_for_completion(&flush.wait_event);
+	}
 }
 EXPORT_SYMBOL_GPL(vhost_work_flush);
 
@@ -249,16 +255,16 @@ EXPORT_SYMBOL_GPL(vhost_poll_flush);
 
 void vhost_work_queue(struct vhost_dev *dev, struct vhost_work *work)
 {
-	unsigned long flags;
+	if (!dev->worker)
+		return;
 
-	spin_lock_irqsave(&dev->work_lock, flags);
-	if (list_empty(&work->node)) {
-		list_add_tail(&work->node, &dev->work_list);
-		work->queue_seq++;
-		spin_unlock_irqrestore(&dev->work_lock, flags);
+	if (!test_and_set_bit(VHOST_WORK_QUEUED, &work->flags)) {
+		/* We can only add the work to the list after we're
+		 * sure it was not in the list.
+		 */
+		smp_mb();
+		llist_add(&work->node, &dev->work_list);
 		wake_up_process(dev->worker);
-	} else {
-		spin_unlock_irqrestore(&dev->work_lock, flags);
 	}
 }
 EXPORT_SYMBOL_GPL(vhost_work_queue);
@@ -266,7 +272,7 @@ EXPORT_SYMBOL_GPL(vhost_work_queue);
 /* A lockless hint for busy polling code to exit the loop */
 bool vhost_has_work(struct vhost_dev *dev)
 {
-	return !list_empty(&dev->work_list);
+	return !llist_empty(&dev->work_list);
 }
 EXPORT_SYMBOL_GPL(vhost_has_work);
 
@@ -300,17 +306,18 @@ static void vhost_vq_reset(struct vhost_dev *dev,
 	vq->call_ctx = NULL;
 	vq->call = NULL;
 	vq->log_ctx = NULL;
-	vq->memory = NULL;
 	vhost_reset_is_le(vq);
 	vhost_disable_cross_endian(vq);
 	vq->busyloop_timeout = 0;
+	vq->umem = NULL;
+	vq->iotlb = NULL;
 }
 
 static int vhost_worker(void *data)
 {
 	struct vhost_dev *dev = data;
-	struct vhost_work *work = NULL;
-	unsigned uninitialized_var(seq);
+	struct vhost_work *work, *work_next;
+	struct llist_node *node;
 	mm_segment_t oldfs = get_fs();
 
 	set_fs(USER_DS);
@@ -320,35 +327,25 @@ static int vhost_worker(void *data)
 		/* mb paired w/ kthread_stop */
 		set_current_state(TASK_INTERRUPTIBLE);
 
-		spin_lock_irq(&dev->work_lock);
-		if (work) {
-			work->done_seq = seq;
-			if (work->flushing)
-				wake_up_all(&work->done);
-		}
-
 		if (kthread_should_stop()) {
-			spin_unlock_irq(&dev->work_lock);
 			__set_current_state(TASK_RUNNING);
 			break;
 		}
-		if (!list_empty(&dev->work_list)) {
-			work = list_first_entry(&dev->work_list,
-						struct vhost_work, node);
-			list_del_init(&work->node);
-			seq = work->queue_seq;
-		} else
-			work = NULL;
-		spin_unlock_irq(&dev->work_lock);
 
-		if (work) {
+		node = llist_del_all(&dev->work_list);
+		if (!node)
+			schedule();
+
+		node = llist_reverse_order(node);
+		/* make sure flag is seen after deletion */
+		smp_wmb();
+		llist_for_each_entry_safe(work, work_next, node, node) {
+			clear_bit(VHOST_WORK_QUEUED, &work->flags);
 			__set_current_state(TASK_RUNNING);
 			work->fn(work);
 			if (need_resched())
 				schedule();
-		} else
-			schedule();
-
+		}
 	}
 	unuse_mm(dev->mm);
 	set_fs(oldfs);
@@ -407,11 +404,16 @@ void vhost_dev_init(struct vhost_dev *dev,
 	mutex_init(&dev->mutex);
 	dev->log_ctx = NULL;
 	dev->log_file = NULL;
-	dev->memory = NULL;
+	dev->umem = NULL;
+	dev->iotlb = NULL;
 	dev->mm = NULL;
-	spin_lock_init(&dev->work_lock);
-	INIT_LIST_HEAD(&dev->work_list);
 	dev->worker = NULL;
+	init_llist_head(&dev->work_list);
+	init_waitqueue_head(&dev->wait);
+	INIT_LIST_HEAD(&dev->read_list);
+	INIT_LIST_HEAD(&dev->pending_list);
+	spin_lock_init(&dev->iotlb_lock);
+
 
 	for (i = 0; i < dev->nvqs; ++i) {
 		vq = dev->vqs[i];
@@ -512,27 +514,36 @@ err_mm:
 }
 EXPORT_SYMBOL_GPL(vhost_dev_set_owner);
 
-struct vhost_memory *vhost_dev_reset_owner_prepare(void)
+static void *vhost_kvzalloc(unsigned long size)
 {
-	return kmalloc(offsetof(struct vhost_memory, regions), GFP_KERNEL);
+	void *n = kzalloc(size, GFP_KERNEL | __GFP_NOWARN | __GFP_REPEAT);
+
+	if (!n)
+		n = vzalloc(size);
+	return n;
+}
+
+struct vhost_umem *vhost_dev_reset_owner_prepare(void)
+{
+	return vhost_kvzalloc(sizeof(struct vhost_umem));
 }
 EXPORT_SYMBOL_GPL(vhost_dev_reset_owner_prepare);
 
 /* Caller should have device mutex */
-void vhost_dev_reset_owner(struct vhost_dev *dev, struct vhost_memory *memory)
+void vhost_dev_reset_owner(struct vhost_dev *dev, struct vhost_umem *umem)
 {
 	int i;
 
 	vhost_dev_cleanup(dev, true);
 
 	/* Restore memory to default empty mapping. */
-	memory->nregions = 0;
-	dev->memory = memory;
+	INIT_LIST_HEAD(&umem->umem_list);
+	dev->umem = umem;
 	/* We don't need VQ locks below since vhost_dev_cleanup makes sure
 	 * VQs aren't running.
 	 */
 	for (i = 0; i < dev->nvqs; ++i)
-		dev->vqs[i]->memory = memory;
+		dev->vqs[i]->umem = umem;
 }
 EXPORT_SYMBOL_GPL(vhost_dev_reset_owner);
 
@@ -549,6 +560,47 @@ void vhost_dev_stop(struct vhost_dev *dev)
 }
 EXPORT_SYMBOL_GPL(vhost_dev_stop);
 
+static void vhost_umem_free(struct vhost_umem *umem,
+			    struct vhost_umem_node *node)
+{
+	vhost_umem_interval_tree_remove(node, &umem->umem_tree);
+	list_del(&node->link);
+	kfree(node);
+	umem->numem--;
+}
+
+static void vhost_umem_clean(struct vhost_umem *umem)
+{
+	struct vhost_umem_node *node, *tmp;
+
+	if (!umem)
+		return;
+
+	list_for_each_entry_safe(node, tmp, &umem->umem_list, link)
+		vhost_umem_free(umem, node);
+
+	kvfree(umem);
+}
+
+static void vhost_clear_msg(struct vhost_dev *dev)
+{
+	struct vhost_msg_node *node, *n;
+
+	spin_lock(&dev->iotlb_lock);
+
+	list_for_each_entry_safe(node, n, &dev->read_list, node) {
+		list_del(&node->node);
+		kfree(node);
+	}
+
+	list_for_each_entry_safe(node, n, &dev->pending_list, node) {
+		list_del(&node->node);
+		kfree(node);
+	}
+
+	spin_unlock(&dev->iotlb_lock);
+}
+
 /* Caller should have device mutex if and only if locked is set */
 void vhost_dev_cleanup(struct vhost_dev *dev, bool locked)
 {
@@ -575,9 +627,13 @@ void vhost_dev_cleanup(struct vhost_dev *dev, bool locked)
 		fput(dev->log_file);
 	dev->log_file = NULL;
 	/* No one will access memory at this point */
-	kvfree(dev->memory);
-	dev->memory = NULL;
-	WARN_ON(!list_empty(&dev->work_list));
+	vhost_umem_clean(dev->umem);
+	dev->umem = NULL;
+	vhost_umem_clean(dev->iotlb);
+	dev->iotlb = NULL;
+	vhost_clear_msg(dev);
+	wake_up_interruptible_poll(&dev->wait, POLLIN | POLLRDNORM);
+	WARN_ON(!llist_empty(&dev->work_list));
 	if (dev->worker) {
 		kthread_stop(dev->worker);
 		dev->worker = NULL;
@@ -601,26 +657,34 @@ static int log_access_ok(void __user *log_base, u64 addr, unsigned long sz)
 			 (sz + VHOST_PAGE_SIZE * 8 - 1) / VHOST_PAGE_SIZE / 8);
 }
 
+static bool vhost_overflow(u64 uaddr, u64 size)
+{
+	/* Make sure 64 bit math will not overflow. */
+	return uaddr > ULONG_MAX || size > ULONG_MAX || uaddr > ULONG_MAX - size;
+}
+
 /* Caller should have vq mutex and device mutex. */
-static int vq_memory_access_ok(void __user *log_base, struct vhost_memory *mem,
+static int vq_memory_access_ok(void __user *log_base, struct vhost_umem *umem,
 			       int log_all)
 {
-	int i;
+	struct vhost_umem_node *node;
 
-	if (!mem)
+	if (!umem)
 		return 0;
 
-	for (i = 0; i < mem->nregions; ++i) {
-		struct vhost_memory_region *m = mem->regions + i;
-		unsigned long a = m->userspace_addr;
-		if (m->memory_size > ULONG_MAX)
+	list_for_each_entry(node, &umem->umem_list, link) {
+		unsigned long a = node->userspace_addr;
+
+		if (vhost_overflow(node->userspace_addr, node->size))
 			return 0;
-		else if (!access_ok(VERIFY_WRITE, (void __user *)a,
-				    m->memory_size))
+
+
+		if (!access_ok(VERIFY_WRITE, (void __user *)a,
+				    node->size))
 			return 0;
 		else if (log_all && !log_access_ok(log_base,
-						   m->guest_phys_addr,
-						   m->memory_size))
+						   node->start,
+						   node->size))
 			return 0;
 	}
 	return 1;
@@ -628,7 +692,7 @@ static int vq_memory_access_ok(void __user *log_base, struct vhost_memory *mem,
 
 /* Can we switch to this memory table? */
 /* Caller should have device mutex but not vq mutex */
-static int memory_access_ok(struct vhost_dev *d, struct vhost_memory *mem,
+static int memory_access_ok(struct vhost_dev *d, struct vhost_umem *umem,
 			    int log_all)
 {
 	int i;
@@ -641,7 +705,8 @@ static int memory_access_ok(struct vhost_dev *d, struct vhost_memory *mem,
 		log = log_all || vhost_has_feature(d->vqs[i], VHOST_F_LOG_ALL);
 		/* If ring is inactive, will check when it's enabled. */
 		if (d->vqs[i]->private_data)
-			ok = vq_memory_access_ok(d->vqs[i]->log_base, mem, log);
+			ok = vq_memory_access_ok(d->vqs[i]->log_base,
+						 umem, log);
 		else
 			ok = 1;
 		mutex_unlock(&d->vqs[i]->mutex);
@@ -651,12 +716,385 @@ static int memory_access_ok(struct vhost_dev *d, struct vhost_memory *mem,
 	return 1;
 }
 
+static int translate_desc(struct vhost_virtqueue *vq, u64 addr, u32 len,
+			  struct iovec iov[], int iov_size, int access);
+
+static int vhost_copy_to_user(struct vhost_virtqueue *vq, void *to,
+			      const void *from, unsigned size)
+{
+	int ret;
+
+	if (!vq->iotlb)
+		return __copy_to_user(to, from, size);
+	else {
+		/* This function should be called after iotlb
+		 * prefetch, which means we're sure that all vq
+		 * could be access through iotlb. So -EAGAIN should
+		 * not happen in this case.
+		 */
+		/* TODO: more fast path */
+		struct iov_iter t;
+		ret = translate_desc(vq, (u64)(uintptr_t)to, size, vq->iotlb_iov,
+				     ARRAY_SIZE(vq->iotlb_iov),
+				     VHOST_ACCESS_WO);
+		if (ret < 0)
+			goto out;
+		iov_iter_init(&t, WRITE, vq->iotlb_iov, ret, size);
+		ret = copy_to_iter(from, size, &t);
+		if (ret == size)
+			ret = 0;
+	}
+out:
+	return ret;
+}
+
+static int vhost_copy_from_user(struct vhost_virtqueue *vq, void *to,
+				void *from, unsigned size)
+{
+	int ret;
+
+	if (!vq->iotlb)
+		return __copy_from_user(to, from, size);
+	else {
+		/* This function should be called after iotlb
+		 * prefetch, which means we're sure that vq
+		 * could be access through iotlb. So -EAGAIN should
+		 * not happen in this case.
+		 */
+		/* TODO: more fast path */
+		struct iov_iter f;
+		ret = translate_desc(vq, (u64)(uintptr_t)from, size, vq->iotlb_iov,
+				     ARRAY_SIZE(vq->iotlb_iov),
+				     VHOST_ACCESS_RO);
+		if (ret < 0) {
+			vq_err(vq, "IOTLB translation failure: uaddr "
+			       "%p size 0x%llx\n", from,
+			       (unsigned long long) size);
+			goto out;
+		}
+		iov_iter_init(&f, READ, vq->iotlb_iov, ret, size);
+		ret = copy_from_iter(to, size, &f);
+		if (ret == size)
+			ret = 0;
+	}
+
+out:
+	return ret;
+}
+
+static void __user *__vhost_get_user(struct vhost_virtqueue *vq,
+				     void *addr, unsigned size)
+{
+	int ret;
+
+	/* This function should be called after iotlb
+	 * prefetch, which means we're sure that vq
+	 * could be access through iotlb. So -EAGAIN should
+	 * not happen in this case.
+	 */
+	/* TODO: more fast path */
+	ret = translate_desc(vq, (u64)(uintptr_t)addr, size, vq->iotlb_iov,
+			     ARRAY_SIZE(vq->iotlb_iov),
+			     VHOST_ACCESS_RO);
+	if (ret < 0) {
+		vq_err(vq, "IOTLB translation failure: uaddr "
+			"%p size 0x%llx\n", addr,
+			(unsigned long long) size);
+		return NULL;
+	}
+
+	if (ret != 1 || vq->iotlb_iov[0].iov_len != size) {
+		vq_err(vq, "Non atomic userspace memory access: uaddr "
+			"%p size 0x%llx\n", addr,
+			(unsigned long long) size);
+		return NULL;
+	}
+
+	return vq->iotlb_iov[0].iov_base;
+}
+
+#define vhost_put_user(vq, x, ptr) \
+({ \
+	int ret = -EFAULT; \
+	if (!vq->iotlb) { \
+		ret = __put_user(x, ptr); \
+	} else { \
+		__typeof__(ptr) to = \
+			(__typeof__(ptr)) __vhost_get_user(vq, ptr, sizeof(*ptr)); \
+		if (to != NULL) \
+			ret = __put_user(x, to); \
+		else \
+			ret = -EFAULT;	\
+	} \
+	ret; \
+})
+
+#define vhost_get_user(vq, x, ptr) \
+({ \
+	int ret; \
+	if (!vq->iotlb) { \
+		ret = __get_user(x, ptr); \
+	} else { \
+		__typeof__(ptr) from = \
+			(__typeof__(ptr)) __vhost_get_user(vq, ptr, sizeof(*ptr)); \
+		if (from != NULL) \
+			ret = __get_user(x, from); \
+		else \
+			ret = -EFAULT; \
+	} \
+	ret; \
+})
+
+static void vhost_dev_lock_vqs(struct vhost_dev *d)
+{
+	int i = 0;
+	for (i = 0; i < d->nvqs; ++i)
+		mutex_lock(&d->vqs[i]->mutex);
+}
+
+static void vhost_dev_unlock_vqs(struct vhost_dev *d)
+{
+	int i = 0;
+	for (i = 0; i < d->nvqs; ++i)
+		mutex_unlock(&d->vqs[i]->mutex);
+}
+
+static int vhost_new_umem_range(struct vhost_umem *umem,
+				u64 start, u64 size, u64 end,
+				u64 userspace_addr, int perm)
+{
+	struct vhost_umem_node *tmp, *node = kmalloc(sizeof(*node), GFP_ATOMIC);
+
+	if (!node)
+		return -ENOMEM;
+
+	if (umem->numem == max_iotlb_entries) {
+		tmp = list_first_entry(&umem->umem_list, typeof(*tmp), link);
+		vhost_umem_free(umem, tmp);
+	}
+
+	node->start = start;
+	node->size = size;
+	node->last = end;
+	node->userspace_addr = userspace_addr;
+	node->perm = perm;
+	INIT_LIST_HEAD(&node->link);
+	list_add_tail(&node->link, &umem->umem_list);
+	vhost_umem_interval_tree_insert(node, &umem->umem_tree);
+	umem->numem++;
+
+	return 0;
+}
+
+static void vhost_del_umem_range(struct vhost_umem *umem,
+				 u64 start, u64 end)
+{
+	struct vhost_umem_node *node;
+
+	while ((node = vhost_umem_interval_tree_iter_first(&umem->umem_tree,
+							   start, end)))
+		vhost_umem_free(umem, node);
+}
+
+static void vhost_iotlb_notify_vq(struct vhost_dev *d,
+				  struct vhost_iotlb_msg *msg)
+{
+	struct vhost_msg_node *node, *n;
+
+	spin_lock(&d->iotlb_lock);
+
+	list_for_each_entry_safe(node, n, &d->pending_list, node) {
+		struct vhost_iotlb_msg *vq_msg = &node->msg.iotlb;
+		if (msg->iova <= vq_msg->iova &&
+		    msg->iova + msg->size - 1 > vq_msg->iova &&
+		    vq_msg->type == VHOST_IOTLB_MISS) {
+			vhost_poll_queue(&node->vq->poll);
+			list_del(&node->node);
+			kfree(node);
+		}
+	}
+
+	spin_unlock(&d->iotlb_lock);
+}
+
+static int umem_access_ok(u64 uaddr, u64 size, int access)
+{
+	unsigned long a = uaddr;
+
+	/* Make sure 64 bit math will not overflow. */
+	if (vhost_overflow(uaddr, size))
+		return -EFAULT;
+
+	if ((access & VHOST_ACCESS_RO) &&
+	    !access_ok(VERIFY_READ, (void __user *)a, size))
+		return -EFAULT;
+	if ((access & VHOST_ACCESS_WO) &&
+	    !access_ok(VERIFY_WRITE, (void __user *)a, size))
+		return -EFAULT;
+	return 0;
+}
+
+int vhost_process_iotlb_msg(struct vhost_dev *dev,
+			    struct vhost_iotlb_msg *msg)
+{
+	int ret = 0;
+
+	vhost_dev_lock_vqs(dev);
+	switch (msg->type) {
+	case VHOST_IOTLB_UPDATE:
+		if (!dev->iotlb) {
+			ret = -EFAULT;
+			break;
+		}
+		if (umem_access_ok(msg->uaddr, msg->size, msg->perm)) {
+			ret = -EFAULT;
+			break;
+		}
+		if (vhost_new_umem_range(dev->iotlb, msg->iova, msg->size,
+					 msg->iova + msg->size - 1,
+					 msg->uaddr, msg->perm)) {
+			ret = -ENOMEM;
+			break;
+		}
+		vhost_iotlb_notify_vq(dev, msg);
+		break;
+	case VHOST_IOTLB_INVALIDATE:
+		vhost_del_umem_range(dev->iotlb, msg->iova,
+				     msg->iova + msg->size - 1);
+		break;
+	default:
+		ret = -EINVAL;
+		break;
+	}
+
+	vhost_dev_unlock_vqs(dev);
+	return ret;
+}
+ssize_t vhost_chr_write_iter(struct vhost_dev *dev,
+			     struct iov_iter *from)
+{
+	struct vhost_msg_node node;
+	unsigned size = sizeof(struct vhost_msg);
+	size_t ret;
+	int err;
+
+	if (iov_iter_count(from) < size)
+		return 0;
+	ret = copy_from_iter(&node.msg, size, from);
+	if (ret != size)
+		goto done;
+
+	switch (node.msg.type) {
+	case VHOST_IOTLB_MSG:
+		err = vhost_process_iotlb_msg(dev, &node.msg.iotlb);
+		if (err)
+			ret = err;
+		break;
+	default:
+		ret = -EINVAL;
+		break;
+	}
+
+done:
+	return ret;
+}
+EXPORT_SYMBOL(vhost_chr_write_iter);
+
+unsigned int vhost_chr_poll(struct file *file, struct vhost_dev *dev,
+			    poll_table *wait)
+{
+	unsigned int mask = 0;
+
+	poll_wait(file, &dev->wait, wait);
+
+	if (!list_empty(&dev->read_list))
+		mask |= POLLIN | POLLRDNORM;
+
+	return mask;
+}
+EXPORT_SYMBOL(vhost_chr_poll);
+
+ssize_t vhost_chr_read_iter(struct vhost_dev *dev, struct iov_iter *to,
+			    int noblock)
+{
+	DEFINE_WAIT(wait);
+	struct vhost_msg_node *node;
+	ssize_t ret = 0;
+	unsigned size = sizeof(struct vhost_msg);
+
+	if (iov_iter_count(to) < size)
+		return 0;
+
+	while (1) {
+		if (!noblock)
+			prepare_to_wait(&dev->wait, &wait,
+					TASK_INTERRUPTIBLE);
+
+		node = vhost_dequeue_msg(dev, &dev->read_list);
+		if (node)
+			break;
+		if (noblock) {
+			ret = -EAGAIN;
+			break;
+		}
+		if (signal_pending(current)) {
+			ret = -ERESTARTSYS;
+			break;
+		}
+		if (!dev->iotlb) {
+			ret = -EBADFD;
+			break;
+		}
+
+		schedule();
+	}
+
+	if (!noblock)
+		finish_wait(&dev->wait, &wait);
+
+	if (node) {
+		ret = copy_to_iter(&node->msg, size, to);
+
+		if (ret != size || node->msg.type != VHOST_IOTLB_MISS) {
+			kfree(node);
+			return ret;
+		}
+
+		vhost_enqueue_msg(dev, &dev->pending_list, node);
+	}
+
+	return ret;
+}
+EXPORT_SYMBOL_GPL(vhost_chr_read_iter);
+
+static int vhost_iotlb_miss(struct vhost_virtqueue *vq, u64 iova, int access)
+{
+	struct vhost_dev *dev = vq->dev;
+	struct vhost_msg_node *node;
+	struct vhost_iotlb_msg *msg;
+
+	node = vhost_new_msg(vq, VHOST_IOTLB_MISS);
+	if (!node)
+		return -ENOMEM;
+
+	msg = &node->msg.iotlb;
+	msg->type = VHOST_IOTLB_MISS;
+	msg->iova = iova;
+	msg->perm = access;
+
+	vhost_enqueue_msg(dev, &dev->read_list, node);
+
+	return 0;
+}
+
 static int vq_access_ok(struct vhost_virtqueue *vq, unsigned int num,
 			struct vring_desc __user *desc,
 			struct vring_avail __user *avail,
 			struct vring_used __user *used)
+
 {
 	size_t s = vhost_has_feature(vq, VIRTIO_RING_F_EVENT_IDX) ? 2 : 0;
+
 	return access_ok(VERIFY_READ, desc, num * sizeof *desc) &&
 	       access_ok(VERIFY_READ, avail,
 			 sizeof *avail + num * sizeof *avail->ring + s) &&
@@ -664,11 +1102,59 @@ static int vq_access_ok(struct vhost_virtqueue *vq, unsigned int num,
 			sizeof *used + num * sizeof *used->ring + s);
 }
 
+static int iotlb_access_ok(struct vhost_virtqueue *vq,
+			   int access, u64 addr, u64 len)
+{
+	const struct vhost_umem_node *node;
+	struct vhost_umem *umem = vq->iotlb;
+	u64 s = 0, size;
+
+	while (len > s) {
+		node = vhost_umem_interval_tree_iter_first(&umem->umem_tree,
+							   addr,
+							   addr + len - 1);
+		if (node == NULL || node->start > addr) {
+			vhost_iotlb_miss(vq, addr, access);
+			return false;
+		} else if (!(node->perm & access)) {
+			/* Report the possible access violation by
+			 * request another translation from userspace.
+			 */
+			return false;
+		}
+
+		size = node->size - addr + node->start;
+		s += size;
+		addr += size;
+	}
+
+	return true;
+}
+
+int vq_iotlb_prefetch(struct vhost_virtqueue *vq)
+{
+	size_t s = vhost_has_feature(vq, VIRTIO_RING_F_EVENT_IDX) ? 2 : 0;
+	unsigned int num = vq->num;
+
+	if (!vq->iotlb)
+		return 1;
+
+	return iotlb_access_ok(vq, VHOST_ACCESS_RO, (u64)(uintptr_t)vq->desc,
+			       num * sizeof *vq->desc) &&
+	       iotlb_access_ok(vq, VHOST_ACCESS_RO, (u64)(uintptr_t)vq->avail,
+			       sizeof *vq->avail +
+			       num * sizeof *vq->avail->ring + s) &&
+	       iotlb_access_ok(vq, VHOST_ACCESS_WO, (u64)(uintptr_t)vq->used,
+			       sizeof *vq->used +
+			       num * sizeof *vq->used->ring + s);
+}
+EXPORT_SYMBOL_GPL(vq_iotlb_prefetch);
+
 /* Can we log writes? */
 /* Caller should have device mutex but not vq mutex */
 int vhost_log_access_ok(struct vhost_dev *dev)
 {
-	return memory_access_ok(dev, dev->memory, 1);
+	return memory_access_ok(dev, dev->umem, 1);
 }
 EXPORT_SYMBOL_GPL(vhost_log_access_ok);
 
@@ -679,7 +1165,7 @@ static int vq_log_access_ok(struct vhost_virtqueue *vq,
 {
 	size_t s = vhost_has_feature(vq, VIRTIO_RING_F_EVENT_IDX) ? 2 : 0;
 
-	return vq_memory_access_ok(log_base, vq->memory,
+	return vq_memory_access_ok(log_base, vq->umem,
 				   vhost_has_feature(vq, VHOST_F_LOG_ALL)) &&
 		(!vq->log_used || log_access_ok(log_base, vq->log_addr,
 					sizeof *vq->used +
@@ -690,33 +1176,36 @@ static int vq_log_access_ok(struct vhost_virtqueue *vq,
 /* Caller should have vq mutex and device mutex */
 int vhost_vq_access_ok(struct vhost_virtqueue *vq)
 {
+	if (vq->iotlb) {
+		/* When device IOTLB was used, the access validation
+		 * will be validated during prefetching.
+		 */
+		return 1;
+	}
 	return vq_access_ok(vq, vq->num, vq->desc, vq->avail, vq->used) &&
 		vq_log_access_ok(vq, vq->log_base);
 }
 EXPORT_SYMBOL_GPL(vhost_vq_access_ok);
 
-static int vhost_memory_reg_sort_cmp(const void *p1, const void *p2)
+static struct vhost_umem *vhost_umem_alloc(void)
 {
-	const struct vhost_memory_region *r1 = p1, *r2 = p2;
-	if (r1->guest_phys_addr < r2->guest_phys_addr)
-		return 1;
-	if (r1->guest_phys_addr > r2->guest_phys_addr)
-		return -1;
-	return 0;
-}
+	struct vhost_umem *umem = vhost_kvzalloc(sizeof(*umem));
 
-static void *vhost_kvzalloc(unsigned long size)
-{
-	void *n = kzalloc(size, GFP_KERNEL | __GFP_NOWARN | __GFP_REPEAT);
+	if (!umem)
+		return NULL;
 
-	if (!n)
-		n = vzalloc(size);
-	return n;
+	umem->umem_tree = RB_ROOT;
+	umem->numem = 0;
+	INIT_LIST_HEAD(&umem->umem_list);
+
+	return umem;
 }
 
 static long vhost_set_memory(struct vhost_dev *d, struct vhost_memory __user *m)
 {
-	struct vhost_memory mem, *newmem, *oldmem;
+	struct vhost_memory mem, *newmem;
+	struct vhost_memory_region *region;
+	struct vhost_umem *newumem, *oldumem;
 	unsigned long size = offsetof(struct vhost_memory, regions);
 	int i;
 
@@ -736,24 +1225,47 @@ static long vhost_set_memory(struct vhost_dev *d, struct vhost_memory __user *m)
 		kvfree(newmem);
 		return -EFAULT;
 	}
-	sort(newmem->regions, newmem->nregions, sizeof(*newmem->regions),
-		vhost_memory_reg_sort_cmp, NULL);
 
-	if (!memory_access_ok(d, newmem, 0)) {
+	newumem = vhost_umem_alloc();
+	if (!newumem) {
 		kvfree(newmem);
-		return -EFAULT;
+		return -ENOMEM;
 	}
-	oldmem = d->memory;
-	d->memory = newmem;
+
+	for (region = newmem->regions;
+	     region < newmem->regions + mem.nregions;
+	     region++) {
+		if (vhost_new_umem_range(newumem,
+					 region->guest_phys_addr,
+					 region->memory_size,
+					 region->guest_phys_addr +
+					 region->memory_size - 1,
+					 region->userspace_addr,
+					 VHOST_ACCESS_RW))
+			goto err;
+	}
+
+	if (!memory_access_ok(d, newumem, 0))
+		goto err;
+
+	oldumem = d->umem;
+	d->umem = newumem;
 
 	/* All memory accesses are done under some VQ mutex. */
 	for (i = 0; i < d->nvqs; ++i) {
 		mutex_lock(&d->vqs[i]->mutex);
-		d->vqs[i]->memory = newmem;
+		d->vqs[i]->umem = newumem;
 		mutex_unlock(&d->vqs[i]->mutex);
 	}
-	kvfree(oldmem);
+
+	kvfree(newmem);
+	vhost_umem_clean(oldumem);
 	return 0;
+
+err:
+	vhost_umem_clean(newumem);
+	kvfree(newmem);
+	return -EFAULT;
 }
 
 long vhost_vring_ioctl(struct vhost_dev *d, int ioctl, void __user *argp)
@@ -974,6 +1486,30 @@ long vhost_vring_ioctl(struct vhost_dev *d, int ioctl, void __user *argp)
 }
 EXPORT_SYMBOL_GPL(vhost_vring_ioctl);
 
+int vhost_init_device_iotlb(struct vhost_dev *d, bool enabled)
+{
+	struct vhost_umem *niotlb, *oiotlb;
+	int i;
+
+	niotlb = vhost_umem_alloc();
+	if (!niotlb)
+		return -ENOMEM;
+
+	oiotlb = d->iotlb;
+	d->iotlb = niotlb;
+
+	for (i = 0; i < d->nvqs; ++i) {
+		mutex_lock(&d->vqs[i]->mutex);
+		d->vqs[i]->iotlb = niotlb;
+		mutex_unlock(&d->vqs[i]->mutex);
+	}
+
+	vhost_umem_clean(oiotlb);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(vhost_init_device_iotlb);
+
 /* Caller must have device mutex */
 long vhost_dev_ioctl(struct vhost_dev *d, unsigned int ioctl, void __user *argp)
 {
@@ -1056,28 +1592,6 @@ done:
 }
 EXPORT_SYMBOL_GPL(vhost_dev_ioctl);
 
-static const struct vhost_memory_region *find_region(struct vhost_memory *mem,
-						     __u64 addr, __u32 len)
-{
-	const struct vhost_memory_region *reg;
-	int start = 0, end = mem->nregions;
-
-	while (start < end) {
-		int slot = start + (end - start) / 2;
-		reg = mem->regions + slot;
-		if (addr >= reg->guest_phys_addr)
-			end = slot;
-		else
-			start = slot + 1;
-	}
-
-	reg = mem->regions + start;
-	if (addr >= reg->guest_phys_addr &&
-		reg->guest_phys_addr + reg->memory_size > addr)
-		return reg;
-	return NULL;
-}
-
 /* TODO: This is really inefficient.  We need something like get_user()
  * (instruction directly accesses the data, with an exception table entry
  * returning -EFAULT). See Documentation/x86/exception-tables.txt.
@@ -1156,7 +1670,8 @@ EXPORT_SYMBOL_GPL(vhost_log_write);
 static int vhost_update_used_flags(struct vhost_virtqueue *vq)
 {
 	void __user *used;
-	if (__put_user(cpu_to_vhost16(vq, vq->used_flags), &vq->used->flags) < 0)
+	if (vhost_put_user(vq, cpu_to_vhost16(vq, vq->used_flags),
+			   &vq->used->flags) < 0)
 		return -EFAULT;
 	if (unlikely(vq->log_used)) {
 		/* Make sure the flag is seen before log. */
@@ -1174,7 +1689,8 @@ static int vhost_update_used_flags(struct vhost_virtqueue *vq)
 
 static int vhost_update_avail_event(struct vhost_virtqueue *vq, u16 avail_event)
 {
-	if (__put_user(cpu_to_vhost16(vq, vq->avail_idx), vhost_avail_event(vq)))
+	if (vhost_put_user(vq, cpu_to_vhost16(vq, vq->avail_idx),
+			   vhost_avail_event(vq)))
 		return -EFAULT;
 	if (unlikely(vq->log_used)) {
 		void __user *used;
@@ -1208,15 +1724,20 @@ int vhost_vq_init_access(struct vhost_virtqueue *vq)
 	if (r)
 		goto err;
 	vq->signalled_used_valid = false;
-	if (!access_ok(VERIFY_READ, &vq->used->idx, sizeof vq->used->idx)) {
+	if (!vq->iotlb &&
+	    !access_ok(VERIFY_READ, &vq->used->idx, sizeof vq->used->idx)) {
 		r = -EFAULT;
 		goto err;
 	}
-	r = __get_user(last_used_idx, &vq->used->idx);
-	if (r)
+	r = vhost_get_user(vq, last_used_idx, &vq->used->idx);
+	if (r) {
+		vq_err(vq, "Can't access used idx at %p\n",
+		       &vq->used->idx);
 		goto err;
+	}
 	vq->last_used_idx = vhost16_to_cpu(vq, last_used_idx);
 	return 0;
+
 err:
 	vq->is_le = is_le;
 	return r;
@@ -1224,36 +1745,48 @@ err:
 EXPORT_SYMBOL_GPL(vhost_vq_init_access);
 
 static int translate_desc(struct vhost_virtqueue *vq, u64 addr, u32 len,
-			  struct iovec iov[], int iov_size)
+			  struct iovec iov[], int iov_size, int access)
 {
-	const struct vhost_memory_region *reg;
-	struct vhost_memory *mem;
+	const struct vhost_umem_node *node;
+	struct vhost_dev *dev = vq->dev;
+	struct vhost_umem *umem = dev->iotlb ? dev->iotlb : dev->umem;
 	struct iovec *_iov;
 	u64 s = 0;
 	int ret = 0;
 
-	mem = vq->memory;
 	while ((u64)len > s) {
 		u64 size;
 		if (unlikely(ret >= iov_size)) {
 			ret = -ENOBUFS;
 			break;
 		}
-		reg = find_region(mem, addr, len);
-		if (unlikely(!reg)) {
-			ret = -EFAULT;
+
+		node = vhost_umem_interval_tree_iter_first(&umem->umem_tree,
+							addr, addr + len - 1);
+		if (node == NULL || node->start > addr) {
+			if (umem != dev->iotlb) {
+				ret = -EFAULT;
+				break;
+			}
+			ret = -EAGAIN;
+			break;
+		} else if (!(node->perm & access)) {
+			ret = -EPERM;
 			break;
 		}
+
 		_iov = iov + ret;
-		size = reg->memory_size - addr + reg->guest_phys_addr;
+		size = node->size - addr + node->start;
 		_iov->iov_len = min((u64)len - s, size);
 		_iov->iov_base = (void __user *)(unsigned long)
-			(reg->userspace_addr + addr - reg->guest_phys_addr);
+			(node->userspace_addr + addr - node->start);
 		s += size;
 		addr += size;
 		++ret;
 	}
 
+	if (ret == -EAGAIN)
+		vhost_iotlb_miss(vq, addr, access);
 	return ret;
 }
 
@@ -1288,7 +1821,7 @@ static int get_indirect(struct vhost_virtqueue *vq,
 	unsigned int i = 0, count, found = 0;
 	u32 len = vhost32_to_cpu(vq, indirect->len);
 	struct iov_iter from;
-	int ret;
+	int ret, access;
 
 	/* Sanity check */
 	if (unlikely(len % sizeof desc)) {
@@ -1300,9 +1833,10 @@ static int get_indirect(struct vhost_virtqueue *vq,
 	}
 
 	ret = translate_desc(vq, vhost64_to_cpu(vq, indirect->addr), len, vq->indirect,
-			     UIO_MAXIOV);
+			     UIO_MAXIOV, VHOST_ACCESS_RO);
 	if (unlikely(ret < 0)) {
-		vq_err(vq, "Translation failure %d in indirect.\n", ret);
+		if (ret != -EAGAIN)
+			vq_err(vq, "Translation failure %d in indirect.\n", ret);
 		return ret;
 	}
 	iov_iter_init(&from, READ, vq->indirect, ret, len);
@@ -1340,16 +1874,22 @@ static int get_indirect(struct vhost_virtqueue *vq,
 			return -EINVAL;
 		}
 
+		if (desc.flags & cpu_to_vhost16(vq, VRING_DESC_F_WRITE))
+			access = VHOST_ACCESS_WO;
+		else
+			access = VHOST_ACCESS_RO;
+
 		ret = translate_desc(vq, vhost64_to_cpu(vq, desc.addr),
 				     vhost32_to_cpu(vq, desc.len), iov + iov_count,
-				     iov_size - iov_count);
+				     iov_size - iov_count, access);
 		if (unlikely(ret < 0)) {
-			vq_err(vq, "Translation failure %d indirect idx %d\n",
-			       ret, i);
+			if (ret != -EAGAIN)
+				vq_err(vq, "Translation failure %d indirect idx %d\n",
+					ret, i);
 			return ret;
 		}
 		/* If this is an input descriptor, increment that count. */
-		if (desc.flags & cpu_to_vhost16(vq, VRING_DESC_F_WRITE)) {
+		if (access == VHOST_ACCESS_WO) {
 			*in_num += ret;
 			if (unlikely(log)) {
 				log[*log_num].addr = vhost64_to_cpu(vq, desc.addr);
@@ -1388,11 +1928,11 @@ int vhost_get_vq_desc(struct vhost_virtqueue *vq,
 	u16 last_avail_idx;
 	__virtio16 avail_idx;
 	__virtio16 ring_head;
-	int ret;
+	int ret, access;
 
 	/* Check it isn't doing very strange things with descriptor numbers. */
 	last_avail_idx = vq->last_avail_idx;
-	if (unlikely(__get_user(avail_idx, &vq->avail->idx))) {
+	if (unlikely(vhost_get_user(vq, avail_idx, &vq->avail->idx))) {
 		vq_err(vq, "Failed to access avail idx at %p\n",
 		       &vq->avail->idx);
 		return -EFAULT;
@@ -1414,8 +1954,8 @@ int vhost_get_vq_desc(struct vhost_virtqueue *vq,
 
 	/* Grab the next descriptor number they're advertising, and increment
 	 * the index we've seen. */
-	if (unlikely(__get_user(ring_head,
-				&vq->avail->ring[last_avail_idx & (vq->num - 1)]))) {
+	if (unlikely(vhost_get_user(vq, ring_head,
+		     &vq->avail->ring[last_avail_idx & (vq->num - 1)]))) {
 		vq_err(vq, "Failed to read head: idx %d address %p\n",
 		       last_avail_idx,
 		       &vq->avail->ring[last_avail_idx % vq->num]);
@@ -1450,7 +1990,8 @@ int vhost_get_vq_desc(struct vhost_virtqueue *vq,
 			       i, vq->num, head);
 			return -EINVAL;
 		}
-		ret = __copy_from_user(&desc, vq->desc + i, sizeof desc);
+		ret = vhost_copy_from_user(vq, &desc, vq->desc + i,
+					   sizeof desc);
 		if (unlikely(ret)) {
 			vq_err(vq, "Failed to get descriptor: idx %d addr %p\n",
 			       i, vq->desc + i);
@@ -1461,22 +2002,28 @@ int vhost_get_vq_desc(struct vhost_virtqueue *vq,
 					   out_num, in_num,
 					   log, log_num, &desc);
 			if (unlikely(ret < 0)) {
-				vq_err(vq, "Failure detected "
-				       "in indirect descriptor at idx %d\n", i);
+				if (ret != -EAGAIN)
+					vq_err(vq, "Failure detected "
+						"in indirect descriptor at idx %d\n", i);
 				return ret;
 			}
 			continue;
 		}
 
+		if (desc.flags & cpu_to_vhost16(vq, VRING_DESC_F_WRITE))
+			access = VHOST_ACCESS_WO;
+		else
+			access = VHOST_ACCESS_RO;
 		ret = translate_desc(vq, vhost64_to_cpu(vq, desc.addr),
 				     vhost32_to_cpu(vq, desc.len), iov + iov_count,
-				     iov_size - iov_count);
+				     iov_size - iov_count, access);
 		if (unlikely(ret < 0)) {
-			vq_err(vq, "Translation failure %d descriptor idx %d\n",
-			       ret, i);
+			if (ret != -EAGAIN)
+				vq_err(vq, "Translation failure %d descriptor idx %d\n",
+					ret, i);
 			return ret;
 		}
-		if (desc.flags & cpu_to_vhost16(vq, VRING_DESC_F_WRITE)) {
+		if (access == VHOST_ACCESS_WO) {
 			/* If this is an input descriptor,
 			 * increment that count. */
 			*in_num += ret;
@@ -1538,15 +2085,15 @@ static int __vhost_add_used_n(struct vhost_virtqueue *vq,
 	start = vq->last_used_idx & (vq->num - 1);
 	used = vq->used->ring + start;
 	if (count == 1) {
-		if (__put_user(heads[0].id, &used->id)) {
+		if (vhost_put_user(vq, heads[0].id, &used->id)) {
 			vq_err(vq, "Failed to write used id");
 			return -EFAULT;
 		}
-		if (__put_user(heads[0].len, &used->len)) {
+		if (vhost_put_user(vq, heads[0].len, &used->len)) {
 			vq_err(vq, "Failed to write used len");
 			return -EFAULT;
 		}
-	} else if (__copy_to_user(used, heads, count * sizeof *used)) {
+	} else if (vhost_copy_to_user(vq, used, heads, count * sizeof *used)) {
 		vq_err(vq, "Failed to write used");
 		return -EFAULT;
 	}
@@ -1590,7 +2137,8 @@ int vhost_add_used_n(struct vhost_virtqueue *vq, struct vring_used_elem *heads,
 
 	/* Make sure buffer is written before we update index. */
 	smp_wmb();
-	if (__put_user(cpu_to_vhost16(vq, vq->last_used_idx), &vq->used->idx)) {
+	if (vhost_put_user(vq, cpu_to_vhost16(vq, vq->last_used_idx),
+			   &vq->used->idx)) {
 		vq_err(vq, "Failed to increment used idx");
 		return -EFAULT;
 	}
@@ -1622,7 +2170,7 @@ static bool vhost_notify(struct vhost_dev *dev, struct vhost_virtqueue *vq)
 
 	if (!vhost_has_feature(vq, VIRTIO_RING_F_EVENT_IDX)) {
 		__virtio16 flags;
-		if (__get_user(flags, &vq->avail->flags)) {
+		if (vhost_get_user(vq, flags, &vq->avail->flags)) {
 			vq_err(vq, "Failed to get flags");
 			return true;
 		}
@@ -1636,7 +2184,7 @@ static bool vhost_notify(struct vhost_dev *dev, struct vhost_virtqueue *vq)
 	if (unlikely(!v))
 		return true;
 
-	if (__get_user(event, vhost_used_event(vq))) {
+	if (vhost_get_user(vq, event, vhost_used_event(vq))) {
 		vq_err(vq, "Failed to get used event idx");
 		return true;
 	}
@@ -1678,7 +2226,7 @@ bool vhost_vq_avail_empty(struct vhost_dev *dev, struct vhost_virtqueue *vq)
 	__virtio16 avail_idx;
 	int r;
 
-	r = __get_user(avail_idx, &vq->avail->idx);
+	r = vhost_get_user(vq, avail_idx, &vq->avail->idx);
 	if (r)
 		return false;
 
@@ -1713,7 +2261,7 @@ bool vhost_enable_notify(struct vhost_dev *dev, struct vhost_virtqueue *vq)
 	/* They could have slipped one in as we were doing that: make
 	 * sure it's written, then check again. */
 	smp_mb();
-	r = __get_user(avail_idx, &vq->avail->idx);
+	r = vhost_get_user(vq, avail_idx, &vq->avail->idx);
 	if (r) {
 		vq_err(vq, "Failed to check avail idx at %p: %d\n",
 		       &vq->avail->idx, r);
@@ -1741,6 +2289,47 @@ void vhost_disable_notify(struct vhost_dev *dev, struct vhost_virtqueue *vq)
 }
 EXPORT_SYMBOL_GPL(vhost_disable_notify);
 
+/* Create a new message. */
+struct vhost_msg_node *vhost_new_msg(struct vhost_virtqueue *vq, int type)
+{
+	struct vhost_msg_node *node = kmalloc(sizeof *node, GFP_KERNEL);
+	if (!node)
+		return NULL;
+	node->vq = vq;
+	node->msg.type = type;
+	return node;
+}
+EXPORT_SYMBOL_GPL(vhost_new_msg);
+
+void vhost_enqueue_msg(struct vhost_dev *dev, struct list_head *head,
+		       struct vhost_msg_node *node)
+{
+	spin_lock(&dev->iotlb_lock);
+	list_add_tail(&node->node, head);
+	spin_unlock(&dev->iotlb_lock);
+
+	wake_up_interruptible_poll(&dev->wait, POLLIN | POLLRDNORM);
+}
+EXPORT_SYMBOL_GPL(vhost_enqueue_msg);
+
+struct vhost_msg_node *vhost_dequeue_msg(struct vhost_dev *dev,
+					 struct list_head *head)
+{
+	struct vhost_msg_node *node = NULL;
+
+	spin_lock(&dev->iotlb_lock);
+	if (!list_empty(head)) {
+		node = list_first_entry(head, struct vhost_msg_node,
+					node);
+		list_del(&node->node);
+	}
+	spin_unlock(&dev->iotlb_lock);
+
+	return node;
+}
+EXPORT_SYMBOL_GPL(vhost_dequeue_msg);
+
+
 static int __init vhost_init(void)
 {
 	return 0;
diff --git a/drivers/vhost/vhost.h b/drivers/vhost/vhost.h
index d36d8beb3351..78f3c5fc02e4 100644
--- a/drivers/vhost/vhost.h
+++ b/drivers/vhost/vhost.h
@@ -15,13 +15,15 @@
 struct vhost_work;
 typedef void (*vhost_work_fn_t)(struct vhost_work *work);
 
+#define VHOST_WORK_QUEUED 1
 struct vhost_work {
-	struct list_head	  node;
+	struct llist_node	  node;
 	vhost_work_fn_t		  fn;
 	wait_queue_head_t	  done;
 	int			  flushing;
 	unsigned		  queue_seq;
 	unsigned		  done_seq;
+	unsigned long		  flags;
 };
 
 /* Poll a file (eventfd or socket) */
@@ -53,6 +55,27 @@ struct vhost_log {
 	u64 len;
 };
 
+#define START(node) ((node)->start)
+#define LAST(node) ((node)->last)
+
+struct vhost_umem_node {
+	struct rb_node rb;
+	struct list_head link;
+	__u64 start;
+	__u64 last;
+	__u64 size;
+	__u64 userspace_addr;
+	__u32 perm;
+	__u32 flags_padding;
+	__u64 __subtree_last;
+};
+
+struct vhost_umem {
+	struct rb_root umem_tree;
+	struct list_head umem_list;
+	int numem;
+};
+
 /* The virtqueue structure describes a queue attached to a device. */
 struct vhost_virtqueue {
 	struct vhost_dev *dev;
@@ -98,10 +121,12 @@ struct vhost_virtqueue {
 	u64 log_addr;
 
 	struct iovec iov[UIO_MAXIOV];
+	struct iovec iotlb_iov[64];
 	struct iovec *indirect;
 	struct vring_used_elem *heads;
 	/* Protected by virtqueue mutex. */
-	struct vhost_memory *memory;
+	struct vhost_umem *umem;
+	struct vhost_umem *iotlb;
 	void *private_data;
 	u64 acked_features;
 	/* Log write descriptors */
@@ -118,25 +143,35 @@ struct vhost_virtqueue {
 	u32 busyloop_timeout;
 };
 
+struct vhost_msg_node {
+  struct vhost_msg msg;
+  struct vhost_virtqueue *vq;
+  struct list_head node;
+};
+
 struct vhost_dev {
-	struct vhost_memory *memory;
 	struct mm_struct *mm;
 	struct mutex mutex;
 	struct vhost_virtqueue **vqs;
 	int nvqs;
 	struct file *log_file;
 	struct eventfd_ctx *log_ctx;
-	spinlock_t work_lock;
-	struct list_head work_list;
+	struct llist_head work_list;
 	struct task_struct *worker;
+	struct vhost_umem *umem;
+	struct vhost_umem *iotlb;
+	spinlock_t iotlb_lock;
+	struct list_head read_list;
+	struct list_head pending_list;
+	wait_queue_head_t wait;
 };
 
 void vhost_dev_init(struct vhost_dev *, struct vhost_virtqueue **vqs, int nvqs);
 long vhost_dev_set_owner(struct vhost_dev *dev);
 bool vhost_dev_has_owner(struct vhost_dev *dev);
 long vhost_dev_check_owner(struct vhost_dev *);
-struct vhost_memory *vhost_dev_reset_owner_prepare(void);
-void vhost_dev_reset_owner(struct vhost_dev *, struct vhost_memory *);
+struct vhost_umem *vhost_dev_reset_owner_prepare(void);
+void vhost_dev_reset_owner(struct vhost_dev *, struct vhost_umem *);
 void vhost_dev_cleanup(struct vhost_dev *, bool locked);
 void vhost_dev_stop(struct vhost_dev *);
 long vhost_dev_ioctl(struct vhost_dev *, unsigned int ioctl, void __user *argp);
@@ -165,6 +200,21 @@ bool vhost_enable_notify(struct vhost_dev *, struct vhost_virtqueue *);
 
 int vhost_log_write(struct vhost_virtqueue *vq, struct vhost_log *log,
 		    unsigned int log_num, u64 len);
+int vq_iotlb_prefetch(struct vhost_virtqueue *vq);
+
+struct vhost_msg_node *vhost_new_msg(struct vhost_virtqueue *vq, int type);
+void vhost_enqueue_msg(struct vhost_dev *dev,
+		       struct list_head *head,
+		       struct vhost_msg_node *node);
+struct vhost_msg_node *vhost_dequeue_msg(struct vhost_dev *dev,
+					 struct list_head *head);
+unsigned int vhost_chr_poll(struct file *file, struct vhost_dev *dev,
+			    poll_table *wait);
+ssize_t vhost_chr_read_iter(struct vhost_dev *dev, struct iov_iter *to,
+			    int noblock);
+ssize_t vhost_chr_write_iter(struct vhost_dev *dev,
+			     struct iov_iter *from);
+int vhost_init_device_iotlb(struct vhost_dev *d, bool enabled);
 
 #define vq_err(vq, fmt, ...) do {                                  \
 		pr_debug(pr_fmt(fmt), ##__VA_ARGS__);       \
diff --git a/drivers/vhost/vsock.c b/drivers/vhost/vsock.c
new file mode 100644
index 000000000000..0ddf3a2dbfc4
--- /dev/null
+++ b/drivers/vhost/vsock.c
@@ -0,0 +1,719 @@
+/*
+ * vhost transport for vsock
+ *
+ * Copyright (C) 2013-2015 Red Hat, Inc.
+ * Author: Asias He <asias@redhat.com>
+ *         Stefan Hajnoczi <stefanha@redhat.com>
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2.
+ */
+#include <linux/miscdevice.h>
+#include <linux/atomic.h>
+#include <linux/module.h>
+#include <linux/mutex.h>
+#include <linux/vmalloc.h>
+#include <net/sock.h>
+#include <linux/virtio_vsock.h>
+#include <linux/vhost.h>
+
+#include <net/af_vsock.h>
+#include "vhost.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 DEFINE_SPINLOCK(vhost_vsock_lock);
+static LIST_HEAD(vhost_vsock_list);
+
+struct vhost_vsock {
+	struct vhost_dev dev;
+	struct vhost_virtqueue vqs[2];
+
+	/* Link to global vhost_vsock_list, protected by vhost_vsock_lock */
+	struct list_head list;
+
+	struct vhost_work send_pkt_work;
+	spinlock_t send_pkt_list_lock;
+	struct list_head send_pkt_list;	/* host->guest pending packets */
+
+	atomic_t queued_replies;
+
+	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;
+
+	spin_lock_bh(&vhost_vsock_lock);
+	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) {
+			spin_unlock_bh(&vhost_vsock_lock);
+			return vsock;
+		}
+	}
+	spin_unlock_bh(&vhost_vsock_lock);
+
+	return NULL;
+}
+
+static void
+vhost_transport_do_send_pkt(struct vhost_vsock *vsock,
+			    struct vhost_virtqueue *vq)
+{
+	struct vhost_virtqueue *tx_vq = &vsock->vqs[VSOCK_VQ_TX];
+	bool added = false;
+	bool restart_tx = false;
+
+	mutex_lock(&vq->mutex);
+
+	if (!vq->private_data)
+		goto out;
+
+	/* 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;
+
+		spin_lock_bh(&vsock->send_pkt_list_lock);
+		if (list_empty(&vsock->send_pkt_list)) {
+			spin_unlock_bh(&vsock->send_pkt_list_lock);
+			vhost_enable_notify(&vsock->dev, vq);
+			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);
+
+		head = vhost_get_vq_desc(vq, vq->iov, ARRAY_SIZE(vq->iov),
+					 &out, &in, NULL, NULL);
+		if (head < 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);
+			break;
+		}
+
+		if (head == vq->num) {
+			spin_lock_bh(&vsock->send_pkt_list_lock);
+			list_add(&pkt->list, &vsock->send_pkt_list);
+			spin_unlock_bh(&vsock->send_pkt_list_lock);
+
+			/* 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;
+		}
+
+		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;
+
+		if (pkt->reply) {
+			int val;
+
+			val = atomic_dec_return(&vsock->queued_replies);
+
+			/* Do we have resources to resume tx processing? */
+			if (val + 1 == tx_vq->num)
+				restart_tx = true;
+		}
+
+		virtio_transport_free_pkt(pkt);
+	}
+	if (added)
+		vhost_signal(&vsock->dev, vq);
+
+out:
+	mutex_unlock(&vq->mutex);
+
+	if (restart_tx)
+		vhost_poll_queue(&tx_vq->poll);
+}
+
+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_pkt(struct virtio_vsock_pkt *pkt)
+{
+	struct vhost_vsock *vsock;
+	struct vhost_virtqueue *vq;
+	int len = pkt->len;
+
+	/* Find the vhost_vsock according to guest context id  */
+	vsock = vhost_vsock_get(le64_to_cpu(pkt->hdr.dst_cid));
+	if (!vsock) {
+		virtio_transport_free_pkt(pkt);
+		return -ENODEV;
+	}
+
+	vq = &vsock->vqs[VSOCK_VQ_RX];
+
+	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);
+
+	vhost_work_queue(&vsock->dev, &vsock->send_pkt_work);
+	return len;
+}
+
+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;
+}
+
+/* Is there space left for replies to rx packets? */
+static bool vhost_vsock_more_replies(struct vhost_vsock *vsock)
+{
+	struct vhost_virtqueue *vq = &vsock->vqs[VSOCK_VQ_TX];
+	int val;
+
+	smp_rmb(); /* paired with atomic_inc() and atomic_dec_return() */
+	val = atomic_read(&vsock->queued_replies);
+
+	return val < vq->num;
+}
+
+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);
+
+	if (!vq->private_data)
+		goto out;
+
+	vhost_disable_notify(&vsock->dev, vq);
+	for (;;) {
+		if (!vhost_vsock_more_replies(vsock)) {
+			/* Stop tx until the device processes already
+			 * pending replies.  Leave tx virtqueue
+			 * callbacks disabled.
+			 */
+			goto no_more_replies;
+		}
+
+		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 (le64_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;
+	}
+
+no_more_replies:
+	if (added)
+		vhost_signal(&vsock->dev, vq);
+
+out:
+	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 int vhost_vsock_stop(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);
+		vq->private_data = NULL;
+		mutex_unlock(&vq->mutex);
+	}
+
+err:
+	mutex_unlock(&vsock->dev.mutex);
+	return ret;
+}
+
+static void vhost_vsock_free(struct vhost_vsock *vsock)
+{
+	kvfree(vsock);
+}
+
+static int vhost_vsock_dev_open(struct inode *inode, struct file *file)
+{
+	struct vhost_virtqueue **vqs;
+	struct vhost_vsock *vsock;
+	int ret;
+
+	/* This struct is large and allocation could fail, fall back to vmalloc
+	 * if there is no other way.
+	 */
+	vsock = kzalloc(sizeof(*vsock), GFP_KERNEL | __GFP_NOWARN | __GFP_REPEAT);
+	if (!vsock) {
+		vsock = vmalloc(sizeof(*vsock));
+		if (!vsock)
+			return -ENOMEM;
+	}
+
+	vqs = kmalloc_array(ARRAY_SIZE(vsock->vqs), sizeof(*vqs), GFP_KERNEL);
+	if (!vqs) {
+		ret = -ENOMEM;
+		goto out;
+	}
+
+	atomic_set(&vsock->queued_replies, 0);
+
+	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;
+	spin_lock_init(&vsock->send_pkt_list_lock);
+	INIT_LIST_HEAD(&vsock->send_pkt_list);
+	vhost_work_init(&vsock->send_pkt_work, vhost_transport_send_pkt_work);
+
+	spin_lock_bh(&vhost_vsock_lock);
+	list_add_tail(&vsock->list, &vhost_vsock_list);
+	spin_unlock_bh(&vhost_vsock_lock);
+	return 0;
+
+out:
+	vhost_vsock_free(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);
+
+	/* 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.
+	 */
+
+	if (!vhost_vsock_get(vsk->local_addr.svm_cid)) {
+		sock_set_flag(sk, SOCK_DONE);
+		vsk->peer_shutdown = SHUTDOWN_MASK;
+		sk->sk_state = SS_UNCONNECTED;
+		sk->sk_err = ECONNRESET;
+		sk->sk_error_report(sk);
+	}
+}
+
+static int vhost_vsock_dev_release(struct inode *inode, struct file *file)
+{
+	struct vhost_vsock *vsock = file->private_data;
+
+	spin_lock_bh(&vhost_vsock_lock);
+	list_del(&vsock->list);
+	spin_unlock_bh(&vhost_vsock_lock);
+
+	/* 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);
+
+	spin_lock_bh(&vsock->send_pkt_list_lock);
+	while (!list_empty(&vsock->send_pkt_list)) {
+		struct virtio_vsock_pkt *pkt;
+
+		pkt = list_first_entry(&vsock->send_pkt_list,
+				struct virtio_vsock_pkt, list);
+		list_del_init(&pkt->list);
+		virtio_transport_free_pkt(pkt);
+	}
+	spin_unlock_bh(&vsock->send_pkt_list_lock);
+
+	vhost_dev_cleanup(&vsock->dev, false);
+	kfree(vsock->dev.vqs);
+	vhost_vsock_free(vsock);
+	return 0;
+}
+
+static int vhost_vsock_set_cid(struct vhost_vsock *vsock, u64 guest_cid)
+{
+	struct vhost_vsock *other;
+
+	/* Refuse reserved CIDs */
+	if (guest_cid <= VMADDR_CID_HOST ||
+	    guest_cid == U32_MAX)
+		return -EINVAL;
+
+	/* 64-bit CIDs are not yet supported */
+	if (guest_cid > U32_MAX)
+		return -EINVAL;
+
+	/* Refuse if CID is already in use */
+	other = vhost_vsock_get(guest_cid);
+	if (other && other != vsock)
+		return -EADDRINUSE;
+
+	spin_lock_bh(&vhost_vsock_lock);
+	vsock->guest_cid = guest_cid;
+	spin_unlock_bh(&vhost_vsock_lock);
+
+	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 guest_cid;
+	u64 features;
+	int start;
+	int r;
+
+	switch (ioctl) {
+	case VHOST_VSOCK_SET_GUEST_CID:
+		if (copy_from_user(&guest_cid, argp, sizeof(guest_cid)))
+			return -EFAULT;
+		return vhost_vsock_set_cid(vsock, guest_cid);
+	case VHOST_VSOCK_SET_RUNNING:
+		if (copy_from_user(&start, argp, sizeof(start)))
+			return -EFAULT;
+		if (start)
+			return vhost_vsock_start(vsock);
+		else
+			return vhost_vsock_stop(vsock);
+	case VHOST_GET_FEATURES:
+		features = VHOST_VSOCK_FEATURES;
+		if (copy_to_user(argp, &features, sizeof(features)))
+			return -EFAULT;
+		return 0;
+	case VHOST_SET_FEATURES:
+		if (copy_from_user(&features, argp, 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,
+};
+
+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/virtio/virtio_balloon.c b/drivers/virtio/virtio_balloon.c
index 888d5f8322ce..4e7003db12c4 100644
--- a/drivers/virtio/virtio_balloon.c
+++ b/drivers/virtio/virtio_balloon.c
@@ -207,6 +207,8 @@ static unsigned leak_balloon(struct virtio_balloon *vb, size_t num)
 	num = min(num, ARRAY_SIZE(vb->pfns));
 
 	mutex_lock(&vb->balloon_lock);
+	/* We can't release more pages than taken */
+	num = min(num, (size_t)vb->num_pages);
 	for (vb->num_pfns = 0; vb->num_pfns < num;
 	     vb->num_pfns += VIRTIO_BALLOON_PAGES_PER_PAGE) {
 		page = balloon_page_dequeue(vb_dev_info);
diff --git a/drivers/virtio/virtio_ring.c b/drivers/virtio/virtio_ring.c
index ca6bfddaacad..114a0c88afb8 100644
--- a/drivers/virtio/virtio_ring.c
+++ b/drivers/virtio/virtio_ring.c
@@ -117,7 +117,10 @@ struct vring_virtqueue {
 #define to_vvq(_vq) container_of(_vq, struct vring_virtqueue, vq)
 
 /*
- * The interaction between virtio and a possible IOMMU is a mess.
+ * Modern virtio devices have feature bits to specify whether they need a
+ * quirk and bypass the IOMMU. If not there, just use the DMA API.
+ *
+ * If there, the interaction between virtio and DMA API is messy.
  *
  * On most systems with virtio, physical addresses match bus addresses,
  * and it doesn't particularly matter whether we use the DMA API.
@@ -133,10 +136,18 @@ struct vring_virtqueue {
  *
  * For the time being, we preserve historic behavior and bypass the DMA
  * API.
+ *
+ * TODO: install a per-device DMA ops structure that does the right thing
+ * taking into account all the above quirks, and use the DMA API
+ * unconditionally on data path.
  */
 
 static bool vring_use_dma_api(struct virtio_device *vdev)
 {
+	if (!virtio_has_iommu_quirk(vdev))
+		return true;
+
+	/* Otherwise, we are left to guess. */
 	/*
 	 * In theory, it's possible to have a buggy QEMU-supposed
 	 * emulated Q35 IOMMU and Xen enabled at the same time.  On
@@ -1099,6 +1110,8 @@ void vring_transport_features(struct virtio_device *vdev)
 			break;
 		case VIRTIO_F_VERSION_1:
 			break;
+		case VIRTIO_F_IOMMU_PLATFORM:
+			break;
 		default:
 			/* We don't understand this bit. */
 			__virtio_clear_bit(vdev, i);