Merge branch 'next' into for-linus

Merge first round of changes for 3.13 merge window.

7 files changed, 590 insertions, 450 deletions

diff --git a/drivers/vhost/Kconfig b/drivers/vhost/Kconfig
index 8b9226da3f5..017a1e8a8f6 100644
--- a/drivers/vhost/Kconfig
+++ b/drivers/vhost/Kconfig
@@ -1,6 +1,7 @@
 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
@@ -13,6 +14,7 @@ config VHOST_NET
 config VHOST_SCSI
 	tristate "VHOST_SCSI TCM fabric driver"
 	depends on TARGET_CORE && EVENTFD && m
+	select VHOST
 	select VHOST_RING
 	default n
 	---help---
@@ -24,3 +26,9 @@ config VHOST_RING
 	---help---
 	  This option is selected by any driver which needs to access
 	  the host side of a virtio ring.
+
+config VHOST
+	tristate
+	---help---
+	  This option is selected by any driver which needs to access
+	  the core of vhost.
diff --git a/drivers/vhost/Makefile b/drivers/vhost/Makefile
index 654e9afb11f..e0441c34db1 100644
--- a/drivers/vhost/Makefile
+++ b/drivers/vhost/Makefile
@@ -1,7 +1,8 @@
 obj-$(CONFIG_VHOST_NET) += vhost_net.o
-vhost_net-y := vhost.o net.o
+vhost_net-y := net.o
 
 obj-$(CONFIG_VHOST_SCSI) += vhost_scsi.o
 vhost_scsi-y := scsi.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 2b51e2336aa..831eb4fd197 100644
--- a/drivers/vhost/net.c
+++ b/drivers/vhost/net.c
@@ -15,7 +15,6 @@
 #include <linux/moduleparam.h>
 #include <linux/mutex.h>
 #include <linux/workqueue.h>
-#include <linux/rcupdate.h>
 #include <linux/file.h>
 #include <linux/slab.h>
 
@@ -150,28 +149,30 @@ static void vhost_net_ubuf_put_and_wait(struct vhost_net_ubuf_ref *ubufs)
 {
 	kref_put(&ubufs->kref, vhost_net_zerocopy_done_signal);
 	wait_event(ubufs->wait, !atomic_read(&ubufs->kref.refcount));
+}
+
+static void vhost_net_ubuf_put_wait_and_free(struct vhost_net_ubuf_ref *ubufs)
+{
+	vhost_net_ubuf_put_and_wait(ubufs);
 	kfree(ubufs);
 }
 
 static void vhost_net_clear_ubuf_info(struct vhost_net *n)
 {
-
-	bool zcopy;
 	int i;
 
-	for (i = 0; i < n->dev.nvqs; ++i) {
-		zcopy = vhost_net_zcopy_mask & (0x1 << i);
-		if (zcopy)
-			kfree(n->vqs[i].ubuf_info);
+	for (i = 0; i < VHOST_NET_VQ_MAX; ++i) {
+		kfree(n->vqs[i].ubuf_info);
+		n->vqs[i].ubuf_info = NULL;
 	}
 }
 
-int vhost_net_set_ubuf_info(struct vhost_net *n)
+static int vhost_net_set_ubuf_info(struct vhost_net *n)
 {
 	bool zcopy;
 	int i;
 
-	for (i = 0; i < n->dev.nvqs; ++i) {
+	for (i = 0; i < VHOST_NET_VQ_MAX; ++i) {
 		zcopy = vhost_net_zcopy_mask & (0x1 << i);
 		if (!zcopy)
 			continue;
@@ -183,25 +184,20 @@ int vhost_net_set_ubuf_info(struct vhost_net *n)
 	return 0;
 
 err:
-	while (i--) {
-		zcopy = vhost_net_zcopy_mask & (0x1 << i);
-		if (!zcopy)
-			continue;
-		kfree(n->vqs[i].ubuf_info);
-	}
+	vhost_net_clear_ubuf_info(n);
 	return -ENOMEM;
 }
 
-void vhost_net_vq_reset(struct vhost_net *n)
+static void vhost_net_vq_reset(struct vhost_net *n)
 {
 	int i;
 
+	vhost_net_clear_ubuf_info(n);
+
 	for (i = 0; i < VHOST_NET_VQ_MAX; i++) {
 		n->vqs[i].done_idx = 0;
 		n->vqs[i].upend_idx = 0;
 		n->vqs[i].ubufs = NULL;
-		kfree(n->vqs[i].ubuf_info);
-		n->vqs[i].ubuf_info = NULL;
 		n->vqs[i].vhost_hlen = 0;
 		n->vqs[i].sock_hlen = 0;
 	}
@@ -280,12 +276,12 @@ static void copy_iovec_hdr(const struct iovec *from, struct iovec *to,
  * of used idx. Once lower device DMA done contiguously, we will signal KVM
  * guest used idx.
  */
-static int vhost_zerocopy_signal_used(struct vhost_net *net,
-				      struct vhost_virtqueue *vq)
+static void vhost_zerocopy_signal_used(struct vhost_net *net,
+				       struct vhost_virtqueue *vq)
 {
 	struct vhost_net_virtqueue *nvq =
 		container_of(vq, struct vhost_net_virtqueue, vq);
-	int i;
+	int i, add;
 	int j = 0;
 
 	for (i = nvq->done_idx; i != nvq->upend_idx; i = (i + 1) % UIO_MAXIOV) {
@@ -293,15 +289,17 @@ static int vhost_zerocopy_signal_used(struct vhost_net *net,
 			vhost_net_tx_err(net);
 		if (VHOST_DMA_IS_DONE(vq->heads[i].len)) {
 			vq->heads[i].len = VHOST_DMA_CLEAR_LEN;
-			vhost_add_used_and_signal(vq->dev, vq,
-						  vq->heads[i].id, 0);
 			++j;
 		} else
 			break;
 	}
-	if (j)
-		nvq->done_idx = i;
-	return j;
+	while (j) {
+		add = min(UIO_MAXIOV - nvq->done_idx, j);
+		vhost_add_used_and_signal_n(vq->dev, vq,
+					    &vq->heads[nvq->done_idx], add);
+		nvq->done_idx = (nvq->done_idx + add) % UIO_MAXIOV;
+		j -= add;
+	}
 }
 
 static void vhost_zerocopy_callback(struct ubuf_info *ubuf, bool success)
@@ -310,6 +308,11 @@ static void vhost_zerocopy_callback(struct ubuf_info *ubuf, bool success)
 	struct vhost_virtqueue *vq = ubufs->vq;
 	int cnt = atomic_read(&ubufs->kref.refcount);
 
+	/* set len to mark this desc buffers done DMA */
+	vq->heads[ubuf->desc].len = success ?
+		VHOST_DMA_DONE_LEN : VHOST_DMA_FAILED_LEN;
+	vhost_net_ubuf_put(ubufs);
+
 	/*
 	 * Trigger polling thread if guest stopped submitting new buffers:
 	 * in this case, the refcount after decrement will eventually reach 1
@@ -320,10 +323,6 @@ static void vhost_zerocopy_callback(struct ubuf_info *ubuf, bool success)
 	 */
 	if (cnt <= 2 || !(cnt % 16))
 		vhost_poll_queue(&vq->poll);
-	/* set len to mark this desc buffers done DMA */
-	vq->heads[ubuf->desc].len = success ?
-		VHOST_DMA_DONE_LEN : VHOST_DMA_FAILED_LEN;
-	vhost_net_ubuf_put(ubufs);
 }
 
 /* Expects to be always run from workqueue - which acts as
@@ -349,12 +348,11 @@ static void handle_tx(struct vhost_net *net)
 	struct vhost_net_ubuf_ref *uninitialized_var(ubufs);
 	bool zcopy, zcopy_used;
 
-	/* TODO: check that we are running from vhost_worker? */
-	sock = rcu_dereference_check(vq->private_data, 1);
+	mutex_lock(&vq->mutex);
+	sock = vq->private_data;
 	if (!sock)
-		return;
+		goto out;
 
-	mutex_lock(&vq->mutex);
 	vhost_disable_notify(&net->dev, vq);
 
 	hdr_size = nvq->vhost_hlen;
@@ -365,6 +363,13 @@ static void handle_tx(struct vhost_net *net)
 		if (zcopy)
 			vhost_zerocopy_signal_used(net, vq);
 
+		/* If more outstanding DMAs, queue the work.
+		 * Handle upend_idx wrap around
+		 */
+		if (unlikely((nvq->upend_idx + vq->num - VHOST_MAX_PEND)
+			      % UIO_MAXIOV == nvq->done_idx))
+			break;
+
 		head = vhost_get_vq_desc(&net->dev, vq, vq->iov,
 					 ARRAY_SIZE(vq->iov),
 					 &out, &in,
@@ -374,17 +379,6 @@ static void handle_tx(struct vhost_net *net)
 			break;
 		/* Nothing new?  Wait for eventfd to tell us they refilled. */
 		if (head == vq->num) {
-			int num_pends;
-
-			/* If more outstanding DMAs, queue the work.
-			 * Handle upend_idx wrap around
-			 */
-			num_pends = likely(nvq->upend_idx >= nvq->done_idx) ?
-				    (nvq->upend_idx - nvq->done_idx) :
-				    (nvq->upend_idx + UIO_MAXIOV -
-				     nvq->done_idx);
-			if (unlikely(num_pends > VHOST_MAX_PEND))
-				break;
 			if (unlikely(vhost_enable_notify(&net->dev, vq))) {
 				vhost_disable_notify(&net->dev, vq);
 				continue;
@@ -407,42 +401,36 @@ static void handle_tx(struct vhost_net *net)
 			       iov_length(nvq->hdr, s), hdr_size);
 			break;
 		}
-		zcopy_used = zcopy && (len >= VHOST_GOODCOPY_LEN ||
-				       nvq->upend_idx != nvq->done_idx);
+
+		zcopy_used = zcopy && len >= VHOST_GOODCOPY_LEN
+				   && (nvq->upend_idx + 1) % UIO_MAXIOV !=
+				      nvq->done_idx
+				   && vhost_net_tx_select_zcopy(net);
 
 		/* use msg_control to pass vhost zerocopy ubuf info to skb */
 		if (zcopy_used) {
+			struct ubuf_info *ubuf;
+			ubuf = nvq->ubuf_info + nvq->upend_idx;
+
 			vq->heads[nvq->upend_idx].id = head;
-			if (!vhost_net_tx_select_zcopy(net) ||
-			    len < VHOST_GOODCOPY_LEN) {
-				/* copy don't need to wait for DMA done */
-				vq->heads[nvq->upend_idx].len =
-							VHOST_DMA_DONE_LEN;
-				msg.msg_control = NULL;
-				msg.msg_controllen = 0;
-				ubufs = NULL;
-			} else {
-				struct ubuf_info *ubuf;
-				ubuf = nvq->ubuf_info + nvq->upend_idx;
-
-				vq->heads[nvq->upend_idx].len =
-					VHOST_DMA_IN_PROGRESS;
-				ubuf->callback = vhost_zerocopy_callback;
-				ubuf->ctx = nvq->ubufs;
-				ubuf->desc = nvq->upend_idx;
-				msg.msg_control = ubuf;
-				msg.msg_controllen = sizeof(ubuf);
-				ubufs = nvq->ubufs;
-				kref_get(&ubufs->kref);
-			}
+			vq->heads[nvq->upend_idx].len = VHOST_DMA_IN_PROGRESS;
+			ubuf->callback = vhost_zerocopy_callback;
+			ubuf->ctx = nvq->ubufs;
+			ubuf->desc = nvq->upend_idx;
+			msg.msg_control = ubuf;
+			msg.msg_controllen = sizeof(ubuf);
+			ubufs = nvq->ubufs;
+			kref_get(&ubufs->kref);
 			nvq->upend_idx = (nvq->upend_idx + 1) % UIO_MAXIOV;
+		} else {
+			msg.msg_control = NULL;
+			ubufs = NULL;
 		}
 		/* TODO: Check specific error and bomb out unless ENOBUFS? */
 		err = sock->ops->sendmsg(NULL, sock, &msg, len);
 		if (unlikely(err < 0)) {
 			if (zcopy_used) {
-				if (ubufs)
-					vhost_net_ubuf_put(ubufs);
+				vhost_net_ubuf_put(ubufs);
 				nvq->upend_idx = ((unsigned)nvq->upend_idx - 1)
 					% UIO_MAXIOV;
 			}
@@ -463,7 +451,7 @@ static void handle_tx(struct vhost_net *net)
 			break;
 		}
 	}
-
+out:
 	mutex_unlock(&vq->mutex);
 }
 
@@ -572,14 +560,14 @@ static void handle_rx(struct vhost_net *net)
 	s16 headcount;
 	size_t vhost_hlen, sock_hlen;
 	size_t vhost_len, sock_len;
-	/* TODO: check that we are running from vhost_worker? */
-	struct socket *sock = rcu_dereference_check(vq->private_data, 1);
-
-	if (!sock)
-		return;
+	struct socket *sock;
 
 	mutex_lock(&vq->mutex);
+	sock = vq->private_data;
+	if (!sock)
+		goto out;
 	vhost_disable_notify(&net->dev, vq);
+
 	vhost_hlen = nvq->vhost_hlen;
 	sock_hlen = nvq->sock_hlen;
 
@@ -654,7 +642,7 @@ static void handle_rx(struct vhost_net *net)
 			break;
 		}
 	}
-
+out:
 	mutex_unlock(&vq->mutex);
 }
 
@@ -752,8 +740,7 @@ static int vhost_net_enable_vq(struct vhost_net *n,
 	struct vhost_poll *poll = n->poll + (nvq - n->vqs);
 	struct socket *sock;
 
-	sock = rcu_dereference_protected(vq->private_data,
-					 lockdep_is_held(&vq->mutex));
+	sock = vq->private_data;
 	if (!sock)
 		return 0;
 
@@ -766,10 +753,9 @@ static struct socket *vhost_net_stop_vq(struct vhost_net *n,
 	struct socket *sock;
 
 	mutex_lock(&vq->mutex);
-	sock = rcu_dereference_protected(vq->private_data,
-					 lockdep_is_held(&vq->mutex));
+	sock = vq->private_data;
 	vhost_net_disable_vq(n, vq);
-	rcu_assign_pointer(vq->private_data, NULL);
+	vq->private_data = NULL;
 	mutex_unlock(&vq->mutex);
 	return sock;
 }
@@ -925,8 +911,7 @@ static long vhost_net_set_backend(struct vhost_net *n, unsigned index, int fd)
 	}
 
 	/* start polling new socket */
-	oldsock = rcu_dereference_protected(vq->private_data,
-					    lockdep_is_held(&vq->mutex));
+	oldsock = vq->private_data;
 	if (sock != oldsock) {
 		ubufs = vhost_net_ubuf_alloc(vq,
 					     sock && vhost_sock_zcopy(sock));
@@ -936,7 +921,7 @@ static long vhost_net_set_backend(struct vhost_net *n, unsigned index, int fd)
 		}
 
 		vhost_net_disable_vq(n, vq);
-		rcu_assign_pointer(vq->private_data, sock);
+		vq->private_data = sock;
 		r = vhost_init_used(vq);
 		if (r)
 			goto err_used;
@@ -955,7 +940,7 @@ static long vhost_net_set_backend(struct vhost_net *n, unsigned index, int fd)
 	mutex_unlock(&vq->mutex);
 
 	if (oldubufs) {
-		vhost_net_ubuf_put_and_wait(oldubufs);
+		vhost_net_ubuf_put_wait_and_free(oldubufs);
 		mutex_lock(&vq->mutex);
 		vhost_zerocopy_signal_used(n, vq);
 		mutex_unlock(&vq->mutex);
@@ -970,10 +955,10 @@ static long vhost_net_set_backend(struct vhost_net *n, unsigned index, int fd)
 	return 0;
 
 err_used:
-	rcu_assign_pointer(vq->private_data, oldsock);
+	vq->private_data = oldsock;
 	vhost_net_enable_vq(n, vq);
 	if (ubufs)
-		vhost_net_ubuf_put_and_wait(ubufs);
+		vhost_net_ubuf_put_wait_and_free(ubufs);
 err_ubufs:
 	fput(sock->file);
 err_vq:
@@ -1053,6 +1038,10 @@ static long vhost_net_set_owner(struct vhost_net *n)
 	int r;
 
 	mutex_lock(&n->dev.mutex);
+	if (vhost_dev_has_owner(&n->dev)) {
+		r = -EBUSY;
+		goto out;
+	}
 	r = vhost_net_set_ubuf_info(n);
 	if (r)
 		goto out;
diff --git a/drivers/vhost/scsi.c b/drivers/vhost/scsi.c
index 70142029722..ce5221fa393 100644
--- a/drivers/vhost/scsi.c
+++ b/drivers/vhost/scsi.c
@@ -1,12 +1,12 @@
 /*******************************************************************************
  * Vhost kernel TCM fabric driver for virtio SCSI initiators
  *
- * (C) Copyright 2010-2012 RisingTide Systems LLC.
+ * (C) Copyright 2010-2013 Datera, Inc.
  * (C) Copyright 2010-2012 IBM Corp.
  *
  * Licensed to the Linux Foundation under the General Public License (GPL) version 2.
  *
- * Authors: Nicholas A. Bellinger <nab@risingtidesystems.com>
+ * Authors: Nicholas A. Bellinger <nab@daterainc.com>
  *          Stefan Hajnoczi <stefanha@linux.vnet.ibm.com>
  *
  * This program is free software; you can redistribute it and/or modify
@@ -48,13 +48,16 @@
 #include <linux/virtio_scsi.h>
 #include <linux/llist.h>
 #include <linux/bitmap.h>
+#include <linux/percpu_ida.h>
 
-#include "vhost.c"
 #include "vhost.h"
 
 #define TCM_VHOST_VERSION  "v0.1"
 #define TCM_VHOST_NAMELEN 256
 #define TCM_VHOST_MAX_CDB_SIZE 32
+#define TCM_VHOST_DEFAULT_TAGS 256
+#define TCM_VHOST_PREALLOC_SGLS 2048
+#define TCM_VHOST_PREALLOC_PAGES 2048
 
 struct vhost_scsi_inflight {
 	/* Wait for the flush operation to finish */
@@ -80,6 +83,7 @@ struct tcm_vhost_cmd {
 	u32 tvc_lun;
 	/* Pointer to the SGL formatted memory from virtio-scsi */
 	struct scatterlist *tvc_sgl;
+	struct page **tvc_upages;
 	/* Pointer to response */
 	struct virtio_scsi_cmd_resp __user *tvc_resp;
 	/* Pointer to vhost_scsi for our device */
@@ -116,7 +120,6 @@ struct tcm_vhost_nacl {
 	struct se_node_acl se_node_acl;
 };
 
-struct vhost_scsi;
 struct tcm_vhost_tpg {
 	/* Vhost port target portal group tag for TCM */
 	u16 tport_tpgt;
@@ -218,7 +221,7 @@ static int iov_num_pages(struct iovec *iov)
 	       ((unsigned long)iov->iov_base & PAGE_MASK)) >> PAGE_SHIFT;
 }
 
-void tcm_vhost_done_inflight(struct kref *kref)
+static void tcm_vhost_done_inflight(struct kref *kref)
 {
 	struct vhost_scsi_inflight *inflight;
 
@@ -329,11 +332,12 @@ static u32 tcm_vhost_get_default_depth(struct se_portal_group *se_tpg)
 	return 1;
 }
 
-static u32 tcm_vhost_get_pr_transport_id(struct se_portal_group *se_tpg,
-	struct se_node_acl *se_nacl,
-	struct t10_pr_registration *pr_reg,
-	int *format_code,
-	unsigned char *buf)
+static u32
+tcm_vhost_get_pr_transport_id(struct se_portal_group *se_tpg,
+			      struct se_node_acl *se_nacl,
+			      struct t10_pr_registration *pr_reg,
+			      int *format_code,
+			      unsigned char *buf)
 {
 	struct tcm_vhost_tpg *tpg = container_of(se_tpg,
 				struct tcm_vhost_tpg, se_tpg);
@@ -359,10 +363,11 @@ static u32 tcm_vhost_get_pr_transport_id(struct se_portal_group *se_tpg,
 			format_code, buf);
 }
 
-static u32 tcm_vhost_get_pr_transport_id_len(struct se_portal_group *se_tpg,
-	struct se_node_acl *se_nacl,
-	struct t10_pr_registration *pr_reg,
-	int *format_code)
+static u32
+tcm_vhost_get_pr_transport_id_len(struct se_portal_group *se_tpg,
+				  struct se_node_acl *se_nacl,
+				  struct t10_pr_registration *pr_reg,
+				  int *format_code)
 {
 	struct tcm_vhost_tpg *tpg = container_of(se_tpg,
 				struct tcm_vhost_tpg, se_tpg);
@@ -388,10 +393,11 @@ static u32 tcm_vhost_get_pr_transport_id_len(struct se_portal_group *se_tpg,
 			format_code);
 }
 
-static char *tcm_vhost_parse_pr_out_transport_id(struct se_portal_group *se_tpg,
-	const char *buf,
-	u32 *out_tid_len,
-	char **port_nexus_ptr)
+static char *
+tcm_vhost_parse_pr_out_transport_id(struct se_portal_group *se_tpg,
+				    const char *buf,
+				    u32 *out_tid_len,
+				    char **port_nexus_ptr)
 {
 	struct tcm_vhost_tpg *tpg = container_of(se_tpg,
 				struct tcm_vhost_tpg, se_tpg);
@@ -417,8 +423,8 @@ static char *tcm_vhost_parse_pr_out_transport_id(struct se_portal_group *se_tpg,
 			port_nexus_ptr);
 }
 
-static struct se_node_acl *tcm_vhost_alloc_fabric_acl(
-	struct se_portal_group *se_tpg)
+static struct se_node_acl *
+tcm_vhost_alloc_fabric_acl(struct se_portal_group *se_tpg)
 {
 	struct tcm_vhost_nacl *nacl;
 
@@ -431,8 +437,9 @@ static struct se_node_acl *tcm_vhost_alloc_fabric_acl(
 	return &nacl->se_node_acl;
 }
 
-static void tcm_vhost_release_fabric_acl(struct se_portal_group *se_tpg,
-	struct se_node_acl *se_nacl)
+static void
+tcm_vhost_release_fabric_acl(struct se_portal_group *se_tpg,
+			     struct se_node_acl *se_nacl)
 {
 	struct tcm_vhost_nacl *nacl = container_of(se_nacl,
 			struct tcm_vhost_nacl, se_node_acl);
@@ -446,7 +453,18 @@ static u32 tcm_vhost_tpg_get_inst_index(struct se_portal_group *se_tpg)
 
 static void tcm_vhost_release_cmd(struct se_cmd *se_cmd)
 {
-	return;
+	struct tcm_vhost_cmd *tv_cmd = container_of(se_cmd,
+				struct tcm_vhost_cmd, tvc_se_cmd);
+	struct se_session *se_sess = se_cmd->se_sess;
+
+	if (tv_cmd->tvc_sgl_count) {
+		u32 i;
+		for (i = 0; i < tv_cmd->tvc_sgl_count; i++)
+			put_page(sg_page(&tv_cmd->tvc_sgl[i]));
+	}
+
+	tcm_vhost_put_inflight(tv_cmd->inflight);
+	percpu_ida_free(&se_sess->sess_tag_pool, se_cmd->map_tag);
 }
 
 static int tcm_vhost_shutdown_session(struct se_session *se_sess)
@@ -491,34 +509,34 @@ static int tcm_vhost_get_cmd_state(struct se_cmd *se_cmd)
 	return 0;
 }
 
-static void vhost_scsi_complete_cmd(struct tcm_vhost_cmd *tv_cmd)
+static void vhost_scsi_complete_cmd(struct tcm_vhost_cmd *cmd)
 {
-	struct vhost_scsi *vs = tv_cmd->tvc_vhost;
+	struct vhost_scsi *vs = cmd->tvc_vhost;
 
-	llist_add(&tv_cmd->tvc_completion_list, &vs->vs_completion_list);
+	llist_add(&cmd->tvc_completion_list, &vs->vs_completion_list);
 
 	vhost_work_queue(&vs->dev, &vs->vs_completion_work);
 }
 
 static int tcm_vhost_queue_data_in(struct se_cmd *se_cmd)
 {
-	struct tcm_vhost_cmd *tv_cmd = container_of(se_cmd,
+	struct tcm_vhost_cmd *cmd = container_of(se_cmd,
 				struct tcm_vhost_cmd, tvc_se_cmd);
-	vhost_scsi_complete_cmd(tv_cmd);
+	vhost_scsi_complete_cmd(cmd);
 	return 0;
 }
 
 static int tcm_vhost_queue_status(struct se_cmd *se_cmd)
 {
-	struct tcm_vhost_cmd *tv_cmd = container_of(se_cmd,
+	struct tcm_vhost_cmd *cmd = container_of(se_cmd,
 				struct tcm_vhost_cmd, tvc_se_cmd);
-	vhost_scsi_complete_cmd(tv_cmd);
+	vhost_scsi_complete_cmd(cmd);
 	return 0;
 }
 
-static int tcm_vhost_queue_tm_rsp(struct se_cmd *se_cmd)
+static void tcm_vhost_queue_tm_rsp(struct se_cmd *se_cmd)
 {
-	return 0;
+	return;
 }
 
 static void tcm_vhost_free_evt(struct vhost_scsi *vs, struct tcm_vhost_evt *evt)
@@ -527,8 +545,9 @@ static void tcm_vhost_free_evt(struct vhost_scsi *vs, struct tcm_vhost_evt *evt)
 	kfree(evt);
 }
 
-static struct tcm_vhost_evt *tcm_vhost_allocate_evt(struct vhost_scsi *vs,
-	u32 event, u32 reason)
+static struct tcm_vhost_evt *
+tcm_vhost_allocate_evt(struct vhost_scsi *vs,
+		       u32 event, u32 reason)
 {
 	struct vhost_virtqueue *vq = &vs->vqs[VHOST_SCSI_VQ_EVT].vq;
 	struct tcm_vhost_evt *evt;
@@ -552,28 +571,22 @@ static struct tcm_vhost_evt *tcm_vhost_allocate_evt(struct vhost_scsi *vs,
 	return evt;
 }
 
-static void vhost_scsi_free_cmd(struct tcm_vhost_cmd *tv_cmd)
+static void vhost_scsi_free_cmd(struct tcm_vhost_cmd *cmd)
 {
-	struct se_cmd *se_cmd = &tv_cmd->tvc_se_cmd;
+	struct se_cmd *se_cmd = &cmd->tvc_se_cmd;
 
 	/* TODO locking against target/backend threads? */
-	transport_generic_free_cmd(se_cmd, 1);
+	transport_generic_free_cmd(se_cmd, 0);
 
-	if (tv_cmd->tvc_sgl_count) {
-		u32 i;
-		for (i = 0; i < tv_cmd->tvc_sgl_count; i++)
-			put_page(sg_page(&tv_cmd->tvc_sgl[i]));
-
-		kfree(tv_cmd->tvc_sgl);
-	}
-
-	tcm_vhost_put_inflight(tv_cmd->inflight);
+}
 
-	kfree(tv_cmd);
+static int vhost_scsi_check_stop_free(struct se_cmd *se_cmd)
+{
+	return target_put_sess_cmd(se_cmd->se_sess, se_cmd);
 }
 
-static void tcm_vhost_do_evt_work(struct vhost_scsi *vs,
-	struct tcm_vhost_evt *evt)
+static void
+tcm_vhost_do_evt_work(struct vhost_scsi *vs, struct tcm_vhost_evt *evt)
 {
 	struct vhost_virtqueue *vq = &vs->vqs[VHOST_SCSI_VQ_EVT].vq;
 	struct virtio_scsi_event *event = &evt->event;
@@ -652,7 +665,7 @@ static void vhost_scsi_complete_cmd_work(struct vhost_work *work)
 					vs_completion_work);
 	DECLARE_BITMAP(signal, VHOST_SCSI_MAX_VQ);
 	struct virtio_scsi_cmd_resp v_rsp;
-	struct tcm_vhost_cmd *tv_cmd;
+	struct tcm_vhost_cmd *cmd;
 	struct llist_node *llnode;
 	struct se_cmd *se_cmd;
 	int ret, vq;
@@ -660,32 +673,32 @@ static void vhost_scsi_complete_cmd_work(struct vhost_work *work)
 	bitmap_zero(signal, VHOST_SCSI_MAX_VQ);
 	llnode = llist_del_all(&vs->vs_completion_list);
 	while (llnode) {
-		tv_cmd = llist_entry(llnode, struct tcm_vhost_cmd,
+		cmd = llist_entry(llnode, struct tcm_vhost_cmd,
 				     tvc_completion_list);
 		llnode = llist_next(llnode);
-		se_cmd = &tv_cmd->tvc_se_cmd;
+		se_cmd = &cmd->tvc_se_cmd;
 
 		pr_debug("%s tv_cmd %p resid %u status %#02x\n", __func__,
-			tv_cmd, se_cmd->residual_count, se_cmd->scsi_status);
+			cmd, se_cmd->residual_count, se_cmd->scsi_status);
 
 		memset(&v_rsp, 0, sizeof(v_rsp));
 		v_rsp.resid = se_cmd->residual_count;
 		/* TODO is status_qualifier field needed? */
 		v_rsp.status = se_cmd->scsi_status;
 		v_rsp.sense_len = se_cmd->scsi_sense_length;
-		memcpy(v_rsp.sense, tv_cmd->tvc_sense_buf,
+		memcpy(v_rsp.sense, cmd->tvc_sense_buf,
 		       v_rsp.sense_len);
-		ret = copy_to_user(tv_cmd->tvc_resp, &v_rsp, sizeof(v_rsp));
+		ret = copy_to_user(cmd->tvc_resp, &v_rsp, sizeof(v_rsp));
 		if (likely(ret == 0)) {
 			struct vhost_scsi_virtqueue *q;
-			vhost_add_used(tv_cmd->tvc_vq, tv_cmd->tvc_vq_desc, 0);
-			q = container_of(tv_cmd->tvc_vq, struct vhost_scsi_virtqueue, vq);
+			vhost_add_used(cmd->tvc_vq, cmd->tvc_vq_desc, 0);
+			q = container_of(cmd->tvc_vq, struct vhost_scsi_virtqueue, vq);
 			vq = q - vs->vqs;
 			__set_bit(vq, signal);
 		} else
 			pr_err("Faulted on virtio_scsi_cmd_resp\n");
 
-		vhost_scsi_free_cmd(tv_cmd);
+		vhost_scsi_free_cmd(cmd);
 	}
 
 	vq = -1;
@@ -694,35 +707,49 @@ static void vhost_scsi_complete_cmd_work(struct vhost_work *work)
 		vhost_signal(&vs->dev, &vs->vqs[vq].vq);
 }
 
-static struct tcm_vhost_cmd *vhost_scsi_allocate_cmd(
-	struct vhost_virtqueue *vq,
-	struct tcm_vhost_tpg *tv_tpg,
-	struct virtio_scsi_cmd_req *v_req,
-	u32 exp_data_len,
-	int data_direction)
+static struct tcm_vhost_cmd *
+vhost_scsi_get_tag(struct vhost_virtqueue *vq,
+			struct tcm_vhost_tpg *tpg,
+			struct virtio_scsi_cmd_req *v_req,
+			u32 exp_data_len,
+			int data_direction)
 {
-	struct tcm_vhost_cmd *tv_cmd;
+	struct tcm_vhost_cmd *cmd;
 	struct tcm_vhost_nexus *tv_nexus;
+	struct se_session *se_sess;
+	struct scatterlist *sg;
+	struct page **pages;
+	int tag;
 
-	tv_nexus = tv_tpg->tpg_nexus;
+	tv_nexus = tpg->tpg_nexus;
 	if (!tv_nexus) {
 		pr_err("Unable to locate active struct tcm_vhost_nexus\n");
 		return ERR_PTR(-EIO);
 	}
+	se_sess = tv_nexus->tvn_se_sess;
 
-	tv_cmd = kzalloc(sizeof(struct tcm_vhost_cmd), GFP_ATOMIC);
-	if (!tv_cmd) {
-		pr_err("Unable to allocate struct tcm_vhost_cmd\n");
+	tag = percpu_ida_alloc(&se_sess->sess_tag_pool, GFP_ATOMIC);
+	if (tag < 0) {
+		pr_err("Unable to obtain tag for tcm_vhost_cmd\n");
 		return ERR_PTR(-ENOMEM);
 	}
-	tv_cmd->tvc_tag = v_req->tag;
-	tv_cmd->tvc_task_attr = v_req->task_attr;
-	tv_cmd->tvc_exp_data_len = exp_data_len;
-	tv_cmd->tvc_data_direction = data_direction;
-	tv_cmd->tvc_nexus = tv_nexus;
-	tv_cmd->inflight = tcm_vhost_get_inflight(vq);
 
-	return tv_cmd;
+	cmd = &((struct tcm_vhost_cmd *)se_sess->sess_cmd_map)[tag];
+	sg = cmd->tvc_sgl;
+	pages = cmd->tvc_upages;
+	memset(cmd, 0, sizeof(struct tcm_vhost_cmd));
+
+	cmd->tvc_sgl = sg;
+	cmd->tvc_upages = pages;
+	cmd->tvc_se_cmd.map_tag = tag;
+	cmd->tvc_tag = v_req->tag;
+	cmd->tvc_task_attr = v_req->task_attr;
+	cmd->tvc_exp_data_len = exp_data_len;
+	cmd->tvc_data_direction = data_direction;
+	cmd->tvc_nexus = tv_nexus;
+	cmd->inflight = tcm_vhost_get_inflight(vq);
+
+	return cmd;
 }
 
 /*
@@ -730,8 +757,12 @@ static struct tcm_vhost_cmd *vhost_scsi_allocate_cmd(
  *
  * Returns the number of scatterlist entries used or -errno on error.
  */
-static int vhost_scsi_map_to_sgl(struct scatterlist *sgl,
-	unsigned int sgl_count, struct iovec *iov, int write)
+static int
+vhost_scsi_map_to_sgl(struct tcm_vhost_cmd *tv_cmd,
+		      struct scatterlist *sgl,
+		      unsigned int sgl_count,
+		      struct iovec *iov,
+		      int write)
 {
 	unsigned int npages = 0, pages_nr, offset, nbytes;
 	struct scatterlist *sg = sgl;
@@ -740,13 +771,25 @@ static int vhost_scsi_map_to_sgl(struct scatterlist *sgl,
 	struct page **pages;
 	int ret, i;
 
+	if (sgl_count > TCM_VHOST_PREALLOC_SGLS) {
+		pr_err("vhost_scsi_map_to_sgl() psgl_count: %u greater than"
+		       " preallocated TCM_VHOST_PREALLOC_SGLS: %u\n",
+			sgl_count, TCM_VHOST_PREALLOC_SGLS);
+		return -ENOBUFS;
+	}
+
 	pages_nr = iov_num_pages(iov);
 	if (pages_nr > sgl_count)
 		return -ENOBUFS;
 
-	pages = kmalloc(pages_nr * sizeof(struct page *), GFP_KERNEL);
-	if (!pages)
-		return -ENOMEM;
+	if (pages_nr > TCM_VHOST_PREALLOC_PAGES) {
+		pr_err("vhost_scsi_map_to_sgl() pages_nr: %u greater than"
+		       " preallocated TCM_VHOST_PREALLOC_PAGES: %u\n",
+			pages_nr, TCM_VHOST_PREALLOC_PAGES);
+		return -ENOBUFS;
+	}
+
+	pages = tv_cmd->tvc_upages;
 
 	ret = get_user_pages_fast((unsigned long)ptr, pages_nr, write, pages);
 	/* No pages were pinned */
@@ -771,12 +814,14 @@ static int vhost_scsi_map_to_sgl(struct scatterlist *sgl,
 	}
 
 out:
-	kfree(pages);
 	return ret;
 }
 
-static int vhost_scsi_map_iov_to_sgl(struct tcm_vhost_cmd *tv_cmd,
-	struct iovec *iov, unsigned int niov, int write)
+static int
+vhost_scsi_map_iov_to_sgl(struct tcm_vhost_cmd *cmd,
+			  struct iovec *iov,
+			  unsigned int niov,
+			  int write)
 {
 	int ret;
 	unsigned int i;
@@ -792,25 +837,21 @@ static int vhost_scsi_map_iov_to_sgl(struct tcm_vhost_cmd *tv_cmd,
 
 	/* TODO overflow checking */
 
-	sg = kmalloc(sizeof(tv_cmd->tvc_sgl[0]) * sgl_count, GFP_ATOMIC);
-	if (!sg)
-		return -ENOMEM;
-	pr_debug("%s sg %p sgl_count %u is_err %d\n", __func__,
-	       sg, sgl_count, !sg);
+	sg = cmd->tvc_sgl;
+	pr_debug("%s sg %p sgl_count %u\n", __func__, sg, sgl_count);
 	sg_init_table(sg, sgl_count);
 
-	tv_cmd->tvc_sgl = sg;
-	tv_cmd->tvc_sgl_count = sgl_count;
+	cmd->tvc_sgl_count = sgl_count;
 
 	pr_debug("Mapping %u iovecs for %u pages\n", niov, sgl_count);
 	for (i = 0; i < niov; i++) {
-		ret = vhost_scsi_map_to_sgl(sg, sgl_count, &iov[i], write);
+		ret = vhost_scsi_map_to_sgl(cmd, sg, sgl_count, &iov[i],
+					    write);
 		if (ret < 0) {
-			for (i = 0; i < tv_cmd->tvc_sgl_count; i++)
-				put_page(sg_page(&tv_cmd->tvc_sgl[i]));
-			kfree(tv_cmd->tvc_sgl);
-			tv_cmd->tvc_sgl = NULL;
-			tv_cmd->tvc_sgl_count = 0;
+			for (i = 0; i < cmd->tvc_sgl_count; i++)
+				put_page(sg_page(&cmd->tvc_sgl[i]));
+
+			cmd->tvc_sgl_count = 0;
 			return ret;
 		}
 
@@ -822,15 +863,15 @@ static int vhost_scsi_map_iov_to_sgl(struct tcm_vhost_cmd *tv_cmd,
 
 static void tcm_vhost_submission_work(struct work_struct *work)
 {
-	struct tcm_vhost_cmd *tv_cmd =
+	struct tcm_vhost_cmd *cmd =
 		container_of(work, struct tcm_vhost_cmd, work);
 	struct tcm_vhost_nexus *tv_nexus;
-	struct se_cmd *se_cmd = &tv_cmd->tvc_se_cmd;
+	struct se_cmd *se_cmd = &cmd->tvc_se_cmd;
 	struct scatterlist *sg_ptr, *sg_bidi_ptr = NULL;
 	int rc, sg_no_bidi = 0;
 
-	if (tv_cmd->tvc_sgl_count) {
-		sg_ptr = tv_cmd->tvc_sgl;
+	if (cmd->tvc_sgl_count) {
+		sg_ptr = cmd->tvc_sgl;
 /* FIXME: Fix BIDI operation in tcm_vhost_submission_work() */
 #if 0
 		if (se_cmd->se_cmd_flags & SCF_BIDI) {
@@ -841,13 +882,13 @@ static void tcm_vhost_submission_work(struct work_struct *work)
 	} else {
 		sg_ptr = NULL;
 	}
-	tv_nexus = tv_cmd->tvc_nexus;
+	tv_nexus = cmd->tvc_nexus;
 
 	rc = target_submit_cmd_map_sgls(se_cmd, tv_nexus->tvn_se_sess,
-			tv_cmd->tvc_cdb, &tv_cmd->tvc_sense_buf[0],
-			tv_cmd->tvc_lun, tv_cmd->tvc_exp_data_len,
-			tv_cmd->tvc_task_attr, tv_cmd->tvc_data_direction,
-			0, sg_ptr, tv_cmd->tvc_sgl_count,
+			cmd->tvc_cdb, &cmd->tvc_sense_buf[0],
+			cmd->tvc_lun, cmd->tvc_exp_data_len,
+			cmd->tvc_task_attr, cmd->tvc_data_direction,
+			TARGET_SCF_ACK_KREF, sg_ptr, cmd->tvc_sgl_count,
 			sg_bidi_ptr, sg_no_bidi);
 	if (rc < 0) {
 		transport_send_check_condition_and_sense(se_cmd,
@@ -856,8 +897,10 @@ static void tcm_vhost_submission_work(struct work_struct *work)
 	}
 }
 
-static void vhost_scsi_send_bad_target(struct vhost_scsi *vs,
-	struct vhost_virtqueue *vq, int head, unsigned out)
+static void
+vhost_scsi_send_bad_target(struct vhost_scsi *vs,
+			   struct vhost_virtqueue *vq,
+			   int head, unsigned out)
 {
 	struct virtio_scsi_cmd_resp __user *resp;
 	struct virtio_scsi_cmd_resp rsp;
@@ -873,31 +916,27 @@ static void vhost_scsi_send_bad_target(struct vhost_scsi *vs,
 		pr_err("Faulted on virtio_scsi_cmd_resp\n");
 }
 
-static void vhost_scsi_handle_vq(struct vhost_scsi *vs,
-	struct vhost_virtqueue *vq)
+static void
+vhost_scsi_handle_vq(struct vhost_scsi *vs, struct vhost_virtqueue *vq)
 {
 	struct tcm_vhost_tpg **vs_tpg;
 	struct virtio_scsi_cmd_req v_req;
-	struct tcm_vhost_tpg *tv_tpg;
-	struct tcm_vhost_cmd *tv_cmd;
+	struct tcm_vhost_tpg *tpg;
+	struct tcm_vhost_cmd *cmd;
 	u32 exp_data_len, data_first, data_num, data_direction;
 	unsigned out, in, i;
 	int head, ret;
 	u8 target;
 
+	mutex_lock(&vq->mutex);
 	/*
 	 * We can handle the vq only after the endpoint is setup by calling the
 	 * VHOST_SCSI_SET_ENDPOINT ioctl.
-	 *
-	 * TODO: Check that we are running from vhost_worker which acts
-	 * as read-side critical section for vhost kind of RCU.
-	 * See the comments in struct vhost_virtqueue in drivers/vhost/vhost.h
 	 */
-	vs_tpg = rcu_dereference_check(vq->private_data, 1);
+	vs_tpg = vq->private_data;
 	if (!vs_tpg)
-		return;
+		goto out;
 
-	mutex_lock(&vq->mutex);
 	vhost_disable_notify(&vs->dev, vq);
 
 	for (;;) {
@@ -964,10 +1003,10 @@ static void vhost_scsi_handle_vq(struct vhost_scsi *vs,
 
 		/* Extract the tpgt */
 		target = v_req.lun[1];
-		tv_tpg = ACCESS_ONCE(vs_tpg[target]);
+		tpg = ACCESS_ONCE(vs_tpg[target]);
 
 		/* Target does not exist, fail the request */
-		if (unlikely(!tv_tpg)) {
+		if (unlikely(!tpg)) {
 			vhost_scsi_send_bad_target(vs, vq, head, out);
 			continue;
 		}
@@ -976,46 +1015,46 @@ static void vhost_scsi_handle_vq(struct vhost_scsi *vs,
 		for (i = 0; i < data_num; i++)
 			exp_data_len += vq->iov[data_first + i].iov_len;
 
-		tv_cmd = vhost_scsi_allocate_cmd(vq, tv_tpg, &v_req,
-					exp_data_len, data_direction);
-		if (IS_ERR(tv_cmd)) {
-			vq_err(vq, "vhost_scsi_allocate_cmd failed %ld\n",
-					PTR_ERR(tv_cmd));
+		cmd = vhost_scsi_get_tag(vq, tpg, &v_req,
+					 exp_data_len, data_direction);
+		if (IS_ERR(cmd)) {
+			vq_err(vq, "vhost_scsi_get_tag failed %ld\n",
+					PTR_ERR(cmd));
 			goto err_cmd;
 		}
 		pr_debug("Allocated tv_cmd: %p exp_data_len: %d, data_direction"
-			": %d\n", tv_cmd, exp_data_len, data_direction);
+			": %d\n", cmd, exp_data_len, data_direction);
 
-		tv_cmd->tvc_vhost = vs;
-		tv_cmd->tvc_vq = vq;
-		tv_cmd->tvc_resp = vq->iov[out].iov_base;
+		cmd->tvc_vhost = vs;
+		cmd->tvc_vq = vq;
+		cmd->tvc_resp = vq->iov[out].iov_base;
 
 		/*
-		 * Copy in the recieved CDB descriptor into tv_cmd->tvc_cdb
+		 * Copy in the recieved CDB descriptor into cmd->tvc_cdb
 		 * that will be used by tcm_vhost_new_cmd_map() and down into
 		 * target_setup_cmd_from_cdb()
 		 */
-		memcpy(tv_cmd->tvc_cdb, v_req.cdb, TCM_VHOST_MAX_CDB_SIZE);
+		memcpy(cmd->tvc_cdb, v_req.cdb, TCM_VHOST_MAX_CDB_SIZE);
 		/*
 		 * Check that the recieved CDB size does not exceeded our
 		 * hardcoded max for tcm_vhost
 		 */
 		/* TODO what if cdb was too small for varlen cdb header? */
-		if (unlikely(scsi_command_size(tv_cmd->tvc_cdb) >
+		if (unlikely(scsi_command_size(cmd->tvc_cdb) >
 					TCM_VHOST_MAX_CDB_SIZE)) {
 			vq_err(vq, "Received SCSI CDB with command_size: %d that"
 				" exceeds SCSI_MAX_VARLEN_CDB_SIZE: %d\n",
-				scsi_command_size(tv_cmd->tvc_cdb),
+				scsi_command_size(cmd->tvc_cdb),
 				TCM_VHOST_MAX_CDB_SIZE);
 			goto err_free;
 		}
-		tv_cmd->tvc_lun = ((v_req.lun[2] << 8) | v_req.lun[3]) & 0x3FFF;
+		cmd->tvc_lun = ((v_req.lun[2] << 8) | v_req.lun[3]) & 0x3FFF;
 
 		pr_debug("vhost_scsi got command opcode: %#02x, lun: %d\n",
-			tv_cmd->tvc_cdb[0], tv_cmd->tvc_lun);
+			cmd->tvc_cdb[0], cmd->tvc_lun);
 
 		if (data_direction != DMA_NONE) {
-			ret = vhost_scsi_map_iov_to_sgl(tv_cmd,
+			ret = vhost_scsi_map_iov_to_sgl(cmd,
 					&vq->iov[data_first], data_num,
 					data_direction == DMA_TO_DEVICE);
 			if (unlikely(ret)) {
@@ -1029,24 +1068,25 @@ static void vhost_scsi_handle_vq(struct vhost_scsi *vs,
 		 * complete the virtio-scsi request in TCM callback context via
 		 * tcm_vhost_queue_data_in() and tcm_vhost_queue_status()
 		 */
-		tv_cmd->tvc_vq_desc = head;
+		cmd->tvc_vq_desc = head;
 		/*
 		 * Dispatch tv_cmd descriptor for cmwq execution in process
 		 * context provided by tcm_vhost_workqueue.  This also ensures
 		 * tv_cmd is executed on the same kworker CPU as this vhost
 		 * thread to gain positive L2 cache locality effects..
 		 */
-		INIT_WORK(&tv_cmd->work, tcm_vhost_submission_work);
-		queue_work(tcm_vhost_workqueue, &tv_cmd->work);
+		INIT_WORK(&cmd->work, tcm_vhost_submission_work);
+		queue_work(tcm_vhost_workqueue, &cmd->work);
 	}
 
 	mutex_unlock(&vq->mutex);
 	return;
 
 err_free:
-	vhost_scsi_free_cmd(tv_cmd);
+	vhost_scsi_free_cmd(cmd);
 err_cmd:
 	vhost_scsi_send_bad_target(vs, vq, head, out);
+out:
 	mutex_unlock(&vq->mutex);
 }
 
@@ -1055,8 +1095,12 @@ static void vhost_scsi_ctl_handle_kick(struct vhost_work *work)
 	pr_debug("%s: The handling func for control queue.\n", __func__);
 }
 
-static void tcm_vhost_send_evt(struct vhost_scsi *vs, struct tcm_vhost_tpg *tpg,
-	struct se_lun *lun, u32 event, u32 reason)
+static void
+tcm_vhost_send_evt(struct vhost_scsi *vs,
+		   struct tcm_vhost_tpg *tpg,
+		   struct se_lun *lun,
+		   u32 event,
+		   u32 reason)
 {
 	struct tcm_vhost_evt *evt;
 
@@ -1146,12 +1190,12 @@ static void vhost_scsi_flush(struct vhost_scsi *vs)
  *  The lock nesting rule is:
  *    tcm_vhost_mutex -> vs->dev.mutex -> tpg->tv_tpg_mutex -> vq->mutex
  */
-static int vhost_scsi_set_endpoint(
-	struct vhost_scsi *vs,
-	struct vhost_scsi_target *t)
+static int
+vhost_scsi_set_endpoint(struct vhost_scsi *vs,
+			struct vhost_scsi_target *t)
 {
 	struct tcm_vhost_tport *tv_tport;
-	struct tcm_vhost_tpg *tv_tpg;
+	struct tcm_vhost_tpg *tpg;
 	struct tcm_vhost_tpg **vs_tpg;
 	struct vhost_virtqueue *vq;
 	int index, ret, i, len;
@@ -1178,32 +1222,32 @@ static int vhost_scsi_set_endpoint(
 	if (vs->vs_tpg)
 		memcpy(vs_tpg, vs->vs_tpg, len);
 
-	list_for_each_entry(tv_tpg, &tcm_vhost_list, tv_tpg_list) {
-		mutex_lock(&tv_tpg->tv_tpg_mutex);
-		if (!tv_tpg->tpg_nexus) {
-			mutex_unlock(&tv_tpg->tv_tpg_mutex);
+	list_for_each_entry(tpg, &tcm_vhost_list, tv_tpg_list) {
+		mutex_lock(&tpg->tv_tpg_mutex);
+		if (!tpg->tpg_nexus) {
+			mutex_unlock(&tpg->tv_tpg_mutex);
 			continue;
 		}
-		if (tv_tpg->tv_tpg_vhost_count != 0) {
-			mutex_unlock(&tv_tpg->tv_tpg_mutex);
+		if (tpg->tv_tpg_vhost_count != 0) {
+			mutex_unlock(&tpg->tv_tpg_mutex);
 			continue;
 		}
-		tv_tport = tv_tpg->tport;
+		tv_tport = tpg->tport;
 
 		if (!strcmp(tv_tport->tport_name, t->vhost_wwpn)) {
-			if (vs->vs_tpg && vs->vs_tpg[tv_tpg->tport_tpgt]) {
+			if (vs->vs_tpg && vs->vs_tpg[tpg->tport_tpgt]) {
 				kfree(vs_tpg);
-				mutex_unlock(&tv_tpg->tv_tpg_mutex);
+				mutex_unlock(&tpg->tv_tpg_mutex);
 				ret = -EEXIST;
 				goto out;
 			}
-			tv_tpg->tv_tpg_vhost_count++;
-			tv_tpg->vhost_scsi = vs;
-			vs_tpg[tv_tpg->tport_tpgt] = tv_tpg;
+			tpg->tv_tpg_vhost_count++;
+			tpg->vhost_scsi = vs;
+			vs_tpg[tpg->tport_tpgt] = tpg;
 			smp_mb__after_atomic_inc();
 			match = true;
 		}
-		mutex_unlock(&tv_tpg->tv_tpg_mutex);
+		mutex_unlock(&tpg->tv_tpg_mutex);
 	}
 
 	if (match) {
@@ -1211,9 +1255,8 @@ static int vhost_scsi_set_endpoint(
 		       sizeof(vs->vs_vhost_wwpn));
 		for (i = 0; i < VHOST_SCSI_MAX_VQ; i++) {
 			vq = &vs->vqs[i].vq;
-			/* Flushing the vhost_work acts as synchronize_rcu */
 			mutex_lock(&vq->mutex);
-			rcu_assign_pointer(vq->private_data, vs_tpg);
+			vq->private_data = vs_tpg;
 			vhost_init_used(vq);
 			mutex_unlock(&vq->mutex);
 		}
@@ -1236,12 +1279,12 @@ out:
 	return ret;
 }
 
-static int vhost_scsi_clear_endpoint(
-	struct vhost_scsi *vs,
-	struct vhost_scsi_target *t)
+static int
+vhost_scsi_clear_endpoint(struct vhost_scsi *vs,
+			  struct vhost_scsi_target *t)
 {
 	struct tcm_vhost_tport *tv_tport;
-	struct tcm_vhost_tpg *tv_tpg;
+	struct tcm_vhost_tpg *tpg;
 	struct vhost_virtqueue *vq;
 	bool match = false;
 	int index, ret, i;
@@ -1264,37 +1307,36 @@ static int vhost_scsi_clear_endpoint(
 
 	for (i = 0; i < VHOST_SCSI_MAX_TARGET; i++) {
 		target = i;
-		tv_tpg = vs->vs_tpg[target];
-		if (!tv_tpg)
+		tpg = vs->vs_tpg[target];
+		if (!tpg)
 			continue;
 
-		mutex_lock(&tv_tpg->tv_tpg_mutex);
-		tv_tport = tv_tpg->tport;
+		mutex_lock(&tpg->tv_tpg_mutex);
+		tv_tport = tpg->tport;
 		if (!tv_tport) {
 			ret = -ENODEV;
 			goto err_tpg;
 		}
 
 		if (strcmp(tv_tport->tport_name, t->vhost_wwpn)) {
-			pr_warn("tv_tport->tport_name: %s, tv_tpg->tport_tpgt: %hu"
+			pr_warn("tv_tport->tport_name: %s, tpg->tport_tpgt: %hu"
 				" does not match t->vhost_wwpn: %s, t->vhost_tpgt: %hu\n",
-				tv_tport->tport_name, tv_tpg->tport_tpgt,
+				tv_tport->tport_name, tpg->tport_tpgt,
 				t->vhost_wwpn, t->vhost_tpgt);
 			ret = -EINVAL;
 			goto err_tpg;
 		}
-		tv_tpg->tv_tpg_vhost_count--;
-		tv_tpg->vhost_scsi = NULL;
+		tpg->tv_tpg_vhost_count--;
+		tpg->vhost_scsi = NULL;
 		vs->vs_tpg[target] = NULL;
 		match = true;
-		mutex_unlock(&tv_tpg->tv_tpg_mutex);
+		mutex_unlock(&tpg->tv_tpg_mutex);
 	}
 	if (match) {
 		for (i = 0; i < VHOST_SCSI_MAX_VQ; i++) {
 			vq = &vs->vqs[i].vq;
-			/* Flushing the vhost_work acts as synchronize_rcu */
 			mutex_lock(&vq->mutex);
-			rcu_assign_pointer(vq->private_data, NULL);
+			vq->private_data = NULL;
 			mutex_unlock(&vq->mutex);
 		}
 	}
@@ -1311,7 +1353,7 @@ static int vhost_scsi_clear_endpoint(
 	return 0;
 
 err_tpg:
-	mutex_unlock(&tv_tpg->tv_tpg_mutex);
+	mutex_unlock(&tpg->tv_tpg_mutex);
 err_dev:
 	mutex_unlock(&vs->dev.mutex);
 	mutex_unlock(&tcm_vhost_mutex);
@@ -1336,70 +1378,85 @@ static int vhost_scsi_set_features(struct vhost_scsi *vs, u64 features)
 	return 0;
 }
 
+static void vhost_scsi_free(struct vhost_scsi *vs)
+{
+	if (is_vmalloc_addr(vs))
+		vfree(vs);
+	else
+		kfree(vs);
+}
+
 static int vhost_scsi_open(struct inode *inode, struct file *f)
 {
-	struct vhost_scsi *s;
+	struct vhost_scsi *vs;
 	struct vhost_virtqueue **vqs;
-	int r, i;
+	int r = -ENOMEM, i;
 
-	s = kzalloc(sizeof(*s), GFP_KERNEL);
-	if (!s)
-		return -ENOMEM;
+	vs = kzalloc(sizeof(*vs), GFP_KERNEL | __GFP_NOWARN | __GFP_REPEAT);
+	if (!vs) {
+		vs = vzalloc(sizeof(*vs));
+		if (!vs)
+			goto err_vs;
+	}
 
 	vqs = kmalloc(VHOST_SCSI_MAX_VQ * sizeof(*vqs), GFP_KERNEL);
-	if (!vqs) {
-		kfree(s);
-		return -ENOMEM;
-	}
+	if (!vqs)
+		goto err_vqs;
 
-	vhost_work_init(&s->vs_completion_work, vhost_scsi_complete_cmd_work);
-	vhost_work_init(&s->vs_event_work, tcm_vhost_evt_work);
+	vhost_work_init(&vs->vs_completion_work, vhost_scsi_complete_cmd_work);
+	vhost_work_init(&vs->vs_event_work, tcm_vhost_evt_work);
 
-	s->vs_events_nr = 0;
-	s->vs_events_missed = false;
+	vs->vs_events_nr = 0;
+	vs->vs_events_missed = false;
 
-	vqs[VHOST_SCSI_VQ_CTL] = &s->vqs[VHOST_SCSI_VQ_CTL].vq;
-	vqs[VHOST_SCSI_VQ_EVT] = &s->vqs[VHOST_SCSI_VQ_EVT].vq;
-	s->vqs[VHOST_SCSI_VQ_CTL].vq.handle_kick = vhost_scsi_ctl_handle_kick;
-	s->vqs[VHOST_SCSI_VQ_EVT].vq.handle_kick = vhost_scsi_evt_handle_kick;
+	vqs[VHOST_SCSI_VQ_CTL] = &vs->vqs[VHOST_SCSI_VQ_CTL].vq;
+	vqs[VHOST_SCSI_VQ_EVT] = &vs->vqs[VHOST_SCSI_VQ_EVT].vq;
+	vs->vqs[VHOST_SCSI_VQ_CTL].vq.handle_kick = vhost_scsi_ctl_handle_kick;
+	vs->vqs[VHOST_SCSI_VQ_EVT].vq.handle_kick = vhost_scsi_evt_handle_kick;
 	for (i = VHOST_SCSI_VQ_IO; i < VHOST_SCSI_MAX_VQ; i++) {
-		vqs[i] = &s->vqs[i].vq;
-		s->vqs[i].vq.handle_kick = vhost_scsi_handle_kick;
+		vqs[i] = &vs->vqs[i].vq;
+		vs->vqs[i].vq.handle_kick = vhost_scsi_handle_kick;
 	}
-	r = vhost_dev_init(&s->dev, vqs, VHOST_SCSI_MAX_VQ);
+	r = vhost_dev_init(&vs->dev, vqs, VHOST_SCSI_MAX_VQ);
 
-	tcm_vhost_init_inflight(s, NULL);
+	tcm_vhost_init_inflight(vs, NULL);
 
-	if (r < 0) {
-		kfree(vqs);
-		kfree(s);
-		return r;
-	}
+	if (r < 0)
+		goto err_init;
 
-	f->private_data = s;
+	f->private_data = vs;
 	return 0;
+
+err_init:
+	kfree(vqs);
+err_vqs:
+	vhost_scsi_free(vs);
+err_vs:
+	return r;
 }
 
 static int vhost_scsi_release(struct inode *inode, struct file *f)
 {
-	struct vhost_scsi *s = f->private_data;
+	struct vhost_scsi *vs = f->private_data;
 	struct vhost_scsi_target t;
 
-	mutex_lock(&s->dev.mutex);
-	memcpy(t.vhost_wwpn, s->vs_vhost_wwpn, sizeof(t.vhost_wwpn));
-	mutex_unlock(&s->dev.mutex);
-	vhost_scsi_clear_endpoint(s, &t);
-	vhost_dev_stop(&s->dev);
-	vhost_dev_cleanup(&s->dev, false);
+	mutex_lock(&vs->dev.mutex);
+	memcpy(t.vhost_wwpn, vs->vs_vhost_wwpn, sizeof(t.vhost_wwpn));
+	mutex_unlock(&vs->dev.mutex);
+	vhost_scsi_clear_endpoint(vs, &t);
+	vhost_dev_stop(&vs->dev);
+	vhost_dev_cleanup(&vs->dev, false);
 	/* Jobs can re-queue themselves in evt kick handler. Do extra flush. */
-	vhost_scsi_flush(s);
-	kfree(s->dev.vqs);
-	kfree(s);
+	vhost_scsi_flush(vs);
+	kfree(vs->dev.vqs);
+	vhost_scsi_free(vs);
 	return 0;
 }
 
-static long vhost_scsi_ioctl(struct file *f, unsigned int ioctl,
-				unsigned long arg)
+static long
+vhost_scsi_ioctl(struct file *f,
+		 unsigned int ioctl,
+		 unsigned long arg)
 {
 	struct vhost_scsi *vs = f->private_data;
 	struct vhost_scsi_target backend;
@@ -1515,8 +1572,9 @@ static char *tcm_vhost_dump_proto_id(struct tcm_vhost_tport *tport)
 	return "Unknown";
 }
 
-static void tcm_vhost_do_plug(struct tcm_vhost_tpg *tpg,
-	struct se_lun *lun, bool plug)
+static void
+tcm_vhost_do_plug(struct tcm_vhost_tpg *tpg,
+		  struct se_lun *lun, bool plug)
 {
 
 	struct vhost_scsi *vs = tpg->vhost_scsi;
@@ -1556,18 +1614,18 @@ static void tcm_vhost_hotunplug(struct tcm_vhost_tpg *tpg, struct se_lun *lun)
 }
 
 static int tcm_vhost_port_link(struct se_portal_group *se_tpg,
-	struct se_lun *lun)
+			       struct se_lun *lun)
 {
-	struct tcm_vhost_tpg *tv_tpg = container_of(se_tpg,
+	struct tcm_vhost_tpg *tpg = container_of(se_tpg,
 				struct tcm_vhost_tpg, se_tpg);
 
 	mutex_lock(&tcm_vhost_mutex);
 
-	mutex_lock(&tv_tpg->tv_tpg_mutex);
-	tv_tpg->tv_tpg_port_count++;
-	mutex_unlock(&tv_tpg->tv_tpg_mutex);
+	mutex_lock(&tpg->tv_tpg_mutex);
+	tpg->tv_tpg_port_count++;
+	mutex_unlock(&tpg->tv_tpg_mutex);
 
-	tcm_vhost_hotplug(tv_tpg, lun);
+	tcm_vhost_hotplug(tpg, lun);
 
 	mutex_unlock(&tcm_vhost_mutex);
 
@@ -1575,26 +1633,26 @@ static int tcm_vhost_port_link(struct se_portal_group *se_tpg,
 }
 
 static void tcm_vhost_port_unlink(struct se_portal_group *se_tpg,
-	struct se_lun *lun)
+				  struct se_lun *lun)
 {
-	struct tcm_vhost_tpg *tv_tpg = container_of(se_tpg,
+	struct tcm_vhost_tpg *tpg = container_of(se_tpg,
 				struct tcm_vhost_tpg, se_tpg);
 
 	mutex_lock(&tcm_vhost_mutex);
 
-	mutex_lock(&tv_tpg->tv_tpg_mutex);
-	tv_tpg->tv_tpg_port_count--;
-	mutex_unlock(&tv_tpg->tv_tpg_mutex);
+	mutex_lock(&tpg->tv_tpg_mutex);
+	tpg->tv_tpg_port_count--;
+	mutex_unlock(&tpg->tv_tpg_mutex);
 
-	tcm_vhost_hotunplug(tv_tpg, lun);
+	tcm_vhost_hotunplug(tpg, lun);
 
 	mutex_unlock(&tcm_vhost_mutex);
 }
 
-static struct se_node_acl *tcm_vhost_make_nodeacl(
-	struct se_portal_group *se_tpg,
-	struct config_group *group,
-	const char *name)
+static struct se_node_acl *
+tcm_vhost_make_nodeacl(struct se_portal_group *se_tpg,
+		       struct config_group *group,
+		       const char *name)
 {
 	struct se_node_acl *se_nacl, *se_nacl_new;
 	struct tcm_vhost_nacl *nacl;
@@ -1635,35 +1693,78 @@ static void tcm_vhost_drop_nodeacl(struct se_node_acl *se_acl)
 	kfree(nacl);
 }
 
-static int tcm_vhost_make_nexus(struct tcm_vhost_tpg *tv_tpg,
-	const char *name)
+static void tcm_vhost_free_cmd_map_res(struct tcm_vhost_nexus *nexus,
+				       struct se_session *se_sess)
+{
+	struct tcm_vhost_cmd *tv_cmd;
+	unsigned int i;
+
+	if (!se_sess->sess_cmd_map)
+		return;
+
+	for (i = 0; i < TCM_VHOST_DEFAULT_TAGS; i++) {
+		tv_cmd = &((struct tcm_vhost_cmd *)se_sess->sess_cmd_map)[i];
+
+		kfree(tv_cmd->tvc_sgl);
+		kfree(tv_cmd->tvc_upages);
+	}
+}
+
+static int tcm_vhost_make_nexus(struct tcm_vhost_tpg *tpg,
+				const char *name)
 {
 	struct se_portal_group *se_tpg;
+	struct se_session *se_sess;
 	struct tcm_vhost_nexus *tv_nexus;
+	struct tcm_vhost_cmd *tv_cmd;
+	unsigned int i;
 
-	mutex_lock(&tv_tpg->tv_tpg_mutex);
-	if (tv_tpg->tpg_nexus) {
-		mutex_unlock(&tv_tpg->tv_tpg_mutex);
-		pr_debug("tv_tpg->tpg_nexus already exists\n");
+	mutex_lock(&tpg->tv_tpg_mutex);
+	if (tpg->tpg_nexus) {
+		mutex_unlock(&tpg->tv_tpg_mutex);
+		pr_debug("tpg->tpg_nexus already exists\n");
 		return -EEXIST;
 	}
-	se_tpg = &tv_tpg->se_tpg;
+	se_tpg = &tpg->se_tpg;
 
 	tv_nexus = kzalloc(sizeof(struct tcm_vhost_nexus), GFP_KERNEL);
 	if (!tv_nexus) {
-		mutex_unlock(&tv_tpg->tv_tpg_mutex);
+		mutex_unlock(&tpg->tv_tpg_mutex);
 		pr_err("Unable to allocate struct tcm_vhost_nexus\n");
 		return -ENOMEM;
 	}
 	/*
-	 *  Initialize the struct se_session pointer
+	 *  Initialize the struct se_session pointer and setup tagpool
+	 *  for struct tcm_vhost_cmd descriptors
 	 */
-	tv_nexus->tvn_se_sess = transport_init_session();
+	tv_nexus->tvn_se_sess = transport_init_session_tags(
+					TCM_VHOST_DEFAULT_TAGS,
+					sizeof(struct tcm_vhost_cmd));
 	if (IS_ERR(tv_nexus->tvn_se_sess)) {
-		mutex_unlock(&tv_tpg->tv_tpg_mutex);
+		mutex_unlock(&tpg->tv_tpg_mutex);
 		kfree(tv_nexus);
 		return -ENOMEM;
 	}
+	se_sess = tv_nexus->tvn_se_sess;
+	for (i = 0; i < TCM_VHOST_DEFAULT_TAGS; i++) {
+		tv_cmd = &((struct tcm_vhost_cmd *)se_sess->sess_cmd_map)[i];
+
+		tv_cmd->tvc_sgl = kzalloc(sizeof(struct scatterlist) *
+					TCM_VHOST_PREALLOC_SGLS, GFP_KERNEL);
+		if (!tv_cmd->tvc_sgl) {
+			mutex_unlock(&tpg->tv_tpg_mutex);
+			pr_err("Unable to allocate tv_cmd->tvc_sgl\n");
+			goto out;
+		}
+
+		tv_cmd->tvc_upages = kzalloc(sizeof(struct page *) *
+					TCM_VHOST_PREALLOC_PAGES, GFP_KERNEL);
+		if (!tv_cmd->tvc_upages) {
+			mutex_unlock(&tpg->tv_tpg_mutex);
+			pr_err("Unable to allocate tv_cmd->tvc_upages\n");
+			goto out;
+		}
+	}
 	/*
 	 * Since we are running in 'demo mode' this call with generate a
 	 * struct se_node_acl for the tcm_vhost struct se_portal_group with
@@ -1672,12 +1773,10 @@ static int tcm_vhost_make_nexus(struct tcm_vhost_tpg *tv_tpg,
 	tv_nexus->tvn_se_sess->se_node_acl = core_tpg_check_initiator_node_acl(
 				se_tpg, (unsigned char *)name);
 	if (!tv_nexus->tvn_se_sess->se_node_acl) {
-		mutex_unlock(&tv_tpg->tv_tpg_mutex);
+		mutex_unlock(&tpg->tv_tpg_mutex);
 		pr_debug("core_tpg_check_initiator_node_acl() failed"
 				" for %s\n", name);
-		transport_free_session(tv_nexus->tvn_se_sess);
-		kfree(tv_nexus);
-		return -ENOMEM;
+		goto out;
 	}
 	/*
 	 * Now register the TCM vhost virtual I_T Nexus as active with the
@@ -1685,10 +1784,16 @@ static int tcm_vhost_make_nexus(struct tcm_vhost_tpg *tv_tpg,
 	 */
 	__transport_register_session(se_tpg, tv_nexus->tvn_se_sess->se_node_acl,
 			tv_nexus->tvn_se_sess, tv_nexus);
-	tv_tpg->tpg_nexus = tv_nexus;
+	tpg->tpg_nexus = tv_nexus;
 
-	mutex_unlock(&tv_tpg->tv_tpg_mutex);
+	mutex_unlock(&tpg->tv_tpg_mutex);
 	return 0;
+
+out:
+	tcm_vhost_free_cmd_map_res(tv_nexus, se_sess);
+	transport_free_session(se_sess);
+	kfree(tv_nexus);
+	return -ENOMEM;
 }
 
 static int tcm_vhost_drop_nexus(struct tcm_vhost_tpg *tpg)
@@ -1728,6 +1833,8 @@ static int tcm_vhost_drop_nexus(struct tcm_vhost_tpg *tpg)
 	pr_debug("TCM_vhost_ConfigFS: Removing I_T Nexus to emulated"
 		" %s Initiator Port: %s\n", tcm_vhost_dump_proto_id(tpg->tport),
 		tv_nexus->tvn_se_sess->se_node_acl->initiatorname);
+
+	tcm_vhost_free_cmd_map_res(tv_nexus, se_sess);
 	/*
 	 * Release the SCSI I_T Nexus to the emulated vhost Target Port
 	 */
@@ -1740,40 +1847,40 @@ static int tcm_vhost_drop_nexus(struct tcm_vhost_tpg *tpg)
 }
 
 static ssize_t tcm_vhost_tpg_show_nexus(struct se_portal_group *se_tpg,
-	char *page)
+					char *page)
 {
-	struct tcm_vhost_tpg *tv_tpg = container_of(se_tpg,
+	struct tcm_vhost_tpg *tpg = container_of(se_tpg,
 				struct tcm_vhost_tpg, se_tpg);
 	struct tcm_vhost_nexus *tv_nexus;
 	ssize_t ret;
 
-	mutex_lock(&tv_tpg->tv_tpg_mutex);
-	tv_nexus = tv_tpg->tpg_nexus;
+	mutex_lock(&tpg->tv_tpg_mutex);
+	tv_nexus = tpg->tpg_nexus;
 	if (!tv_nexus) {
-		mutex_unlock(&tv_tpg->tv_tpg_mutex);
+		mutex_unlock(&tpg->tv_tpg_mutex);
 		return -ENODEV;
 	}
 	ret = snprintf(page, PAGE_SIZE, "%s\n",
 			tv_nexus->tvn_se_sess->se_node_acl->initiatorname);
-	mutex_unlock(&tv_tpg->tv_tpg_mutex);
+	mutex_unlock(&tpg->tv_tpg_mutex);
 
 	return ret;
 }
 
 static ssize_t tcm_vhost_tpg_store_nexus(struct se_portal_group *se_tpg,
-	const char *page,
-	size_t count)
+					 const char *page,
+					 size_t count)
 {
-	struct tcm_vhost_tpg *tv_tpg = container_of(se_tpg,
+	struct tcm_vhost_tpg *tpg = container_of(se_tpg,
 				struct tcm_vhost_tpg, se_tpg);
-	struct tcm_vhost_tport *tport_wwn = tv_tpg->tport;
+	struct tcm_vhost_tport *tport_wwn = tpg->tport;
 	unsigned char i_port[TCM_VHOST_NAMELEN], *ptr, *port_ptr;
 	int ret;
 	/*
 	 * Shutdown the active I_T nexus if 'NULL' is passed..
 	 */
 	if (!strncmp(page, "NULL", 4)) {
-		ret = tcm_vhost_drop_nexus(tv_tpg);
+		ret = tcm_vhost_drop_nexus(tpg);
 		return (!ret) ? count : ret;
 	}
 	/*
@@ -1831,7 +1938,7 @@ check_newline:
 	if (i_port[strlen(i_port)-1] == '\n')
 		i_port[strlen(i_port)-1] = '\0';
 
-	ret = tcm_vhost_make_nexus(tv_tpg, port_ptr);
+	ret = tcm_vhost_make_nexus(tpg, port_ptr);
 	if (ret < 0)
 		return ret;
 
@@ -1845,9 +1952,10 @@ static struct configfs_attribute *tcm_vhost_tpg_attrs[] = {
 	NULL,
 };
 
-static struct se_portal_group *tcm_vhost_make_tpg(struct se_wwn *wwn,
-	struct config_group *group,
-	const char *name)
+static struct se_portal_group *
+tcm_vhost_make_tpg(struct se_wwn *wwn,
+		   struct config_group *group,
+		   const char *name)
 {
 	struct tcm_vhost_tport *tport = container_of(wwn,
 			struct tcm_vhost_tport, tport_wwn);
@@ -1903,9 +2011,10 @@ static void tcm_vhost_drop_tpg(struct se_portal_group *se_tpg)
 	kfree(tpg);
 }
 
-static struct se_wwn *tcm_vhost_make_tport(struct target_fabric_configfs *tf,
-	struct config_group *group,
-	const char *name)
+static struct se_wwn *
+tcm_vhost_make_tport(struct target_fabric_configfs *tf,
+		     struct config_group *group,
+		     const char *name)
 {
 	struct tcm_vhost_tport *tport;
 	char *ptr;
@@ -1975,9 +2084,9 @@ static void tcm_vhost_drop_tport(struct se_wwn *wwn)
 	kfree(tport);
 }
 
-static ssize_t tcm_vhost_wwn_show_attr_version(
-	struct target_fabric_configfs *tf,
-	char *page)
+static ssize_t
+tcm_vhost_wwn_show_attr_version(struct target_fabric_configfs *tf,
+				char *page)
 {
 	return sprintf(page, "TCM_VHOST fabric module %s on %s/%s"
 		"on "UTS_RELEASE"\n", TCM_VHOST_VERSION, utsname()->sysname,
@@ -2008,6 +2117,7 @@ static struct target_core_fabric_ops tcm_vhost_ops = {
 	.tpg_release_fabric_acl		= tcm_vhost_release_fabric_acl,
 	.tpg_get_inst_index		= tcm_vhost_tpg_get_inst_index,
 	.release_cmd			= tcm_vhost_release_cmd,
+	.check_stop_free		= vhost_scsi_check_stop_free,
 	.shutdown_session		= tcm_vhost_shutdown_session,
 	.close_session			= tcm_vhost_close_session,
 	.sess_get_index			= tcm_vhost_sess_get_index,
diff --git a/drivers/vhost/test.c b/drivers/vhost/test.c
index 1ee45bc85f6..339eae85859 100644
--- a/drivers/vhost/test.c
+++ b/drivers/vhost/test.c
@@ -13,12 +13,11 @@
 #include <linux/module.h>
 #include <linux/mutex.h>
 #include <linux/workqueue.h>
-#include <linux/rcupdate.h>
 #include <linux/file.h>
 #include <linux/slab.h>
 
 #include "test.h"
-#include "vhost.c"
+#include "vhost.h"
 
 /* Max number of bytes transferred before requeueing the job.
  * Using this limit prevents one virtqueue from starving others. */
@@ -38,17 +37,19 @@ struct vhost_test {
  * read-size critical section for our kind of RCU. */
 static void handle_vq(struct vhost_test *n)
 {
-	struct vhost_virtqueue *vq = &n->dev.vqs[VHOST_TEST_VQ];
+	struct vhost_virtqueue *vq = &n->vqs[VHOST_TEST_VQ];
 	unsigned out, in;
 	int head;
 	size_t len, total_len = 0;
 	void *private;
 
-	private = rcu_dereference_check(vq->private_data, 1);
-	if (!private)
+	mutex_lock(&vq->mutex);
+	private = vq->private_data;
+	if (!private) {
+		mutex_unlock(&vq->mutex);
 		return;
+	}
 
-	mutex_lock(&vq->mutex);
 	vhost_disable_notify(&n->dev, vq);
 
 	for (;;) {
@@ -102,15 +103,23 @@ static int vhost_test_open(struct inode *inode, struct file *f)
 {
 	struct vhost_test *n = kmalloc(sizeof *n, GFP_KERNEL);
 	struct vhost_dev *dev;
+	struct vhost_virtqueue **vqs;
 	int r;
 
 	if (!n)
 		return -ENOMEM;
+	vqs = kmalloc(VHOST_TEST_VQ_MAX * sizeof(*vqs), GFP_KERNEL);
+	if (!vqs) {
+		kfree(n);
+		return -ENOMEM;
+	}
 
 	dev = &n->dev;
+	vqs[VHOST_TEST_VQ] = &n->vqs[VHOST_TEST_VQ];
 	n->vqs[VHOST_TEST_VQ].handle_kick = handle_vq_kick;
-	r = vhost_dev_init(dev, n->vqs, VHOST_TEST_VQ_MAX);
+	r = vhost_dev_init(dev, vqs, VHOST_TEST_VQ_MAX);
 	if (r < 0) {
+		kfree(vqs);
 		kfree(n);
 		return r;
 	}
@@ -126,9 +135,8 @@ static void *vhost_test_stop_vq(struct vhost_test *n,
 	void *private;
 
 	mutex_lock(&vq->mutex);
-	private = rcu_dereference_protected(vq->private_data,
-					 lockdep_is_held(&vq->mutex));
-	rcu_assign_pointer(vq->private_data, NULL);
+	private = vq->private_data;
+	vq->private_data = NULL;
 	mutex_unlock(&vq->mutex);
 	return private;
 }
@@ -140,7 +148,7 @@ static void vhost_test_stop(struct vhost_test *n, void **privatep)
 
 static void vhost_test_flush_vq(struct vhost_test *n, int index)
 {
-	vhost_poll_flush(&n->dev.vqs[index].poll);
+	vhost_poll_flush(&n->vqs[index].poll);
 }
 
 static void vhost_test_flush(struct vhost_test *n)
@@ -191,9 +199,8 @@ static long vhost_test_run(struct vhost_test *n, int test)
 		priv = test ? n : NULL;
 
 		/* start polling new socket */
-		oldpriv = rcu_dereference_protected(vq->private_data,
-						    lockdep_is_held(&vq->mutex));
-		rcu_assign_pointer(vq->private_data, priv);
+		oldpriv = vq->private_data;
+		vq->private_data = priv;
 
 		r = vhost_init_used(&n->vqs[index]);
 
@@ -268,14 +275,14 @@ static long vhost_test_ioctl(struct file *f, unsigned int ioctl,
 			return -EFAULT;
 		return vhost_test_run(n, test);
 	case VHOST_GET_FEATURES:
-		features = VHOST_NET_FEATURES;
+		features = VHOST_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;
-		if (features & ~VHOST_NET_FEATURES)
+		if (features & ~VHOST_FEATURES)
 			return -EOPNOTSUPP;
 		return vhost_test_set_features(n, features);
 	case VHOST_RESET_OWNER:
diff --git a/drivers/vhost/vhost.c b/drivers/vhost/vhost.c
index beee7f5787e..69068e0d8f3 100644
--- a/drivers/vhost/vhost.c
+++ b/drivers/vhost/vhost.c
@@ -13,7 +13,7 @@
 
 #include <linux/eventfd.h>
 #include <linux/vhost.h>
-#include <linux/socket.h> /* memcpy_fromiovec */
+#include <linux/uio.h>
 #include <linux/mm.h>
 #include <linux/mmu_context.h>
 #include <linux/miscdevice.h>
@@ -25,6 +25,7 @@
 #include <linux/slab.h>
 #include <linux/kthread.h>
 #include <linux/cgroup.h>
+#include <linux/module.h>
 
 #include "vhost.h"
 
@@ -66,6 +67,7 @@ void vhost_work_init(struct vhost_work *work, vhost_work_fn_t fn)
 	work->flushing = 0;
 	work->queue_seq = work->done_seq = 0;
 }
+EXPORT_SYMBOL_GPL(vhost_work_init);
 
 /* Init poll structure */
 void vhost_poll_init(struct vhost_poll *poll, vhost_work_fn_t fn,
@@ -79,6 +81,7 @@ void vhost_poll_init(struct vhost_poll *poll, vhost_work_fn_t fn,
 
 	vhost_work_init(&poll->work, fn);
 }
+EXPORT_SYMBOL_GPL(vhost_poll_init);
 
 /* Start polling a file. We add ourselves to file's wait queue. The caller must
  * keep a reference to a file until after vhost_poll_stop is called. */
@@ -101,6 +104,7 @@ int vhost_poll_start(struct vhost_poll *poll, struct file *file)
 
 	return ret;
 }
+EXPORT_SYMBOL_GPL(vhost_poll_start);
 
 /* Stop polling a file. After this function returns, it becomes safe to drop the
  * file reference. You must also flush afterwards. */
@@ -111,6 +115,7 @@ void vhost_poll_stop(struct vhost_poll *poll)
 		poll->wqh = NULL;
 	}
 }
+EXPORT_SYMBOL_GPL(vhost_poll_stop);
 
 static bool vhost_work_seq_done(struct vhost_dev *dev, struct vhost_work *work,
 				unsigned seq)
@@ -123,7 +128,7 @@ static bool vhost_work_seq_done(struct vhost_dev *dev, struct vhost_work *work,
 	return left <= 0;
 }
 
-static void vhost_work_flush(struct vhost_dev *dev, struct vhost_work *work)
+void vhost_work_flush(struct vhost_dev *dev, struct vhost_work *work)
 {
 	unsigned seq;
 	int flushing;
@@ -138,6 +143,7 @@ static void vhost_work_flush(struct vhost_dev *dev, struct vhost_work *work)
 	spin_unlock_irq(&dev->work_lock);
 	BUG_ON(flushing < 0);
 }
+EXPORT_SYMBOL_GPL(vhost_work_flush);
 
 /* Flush any work that has been scheduled. When calling this, don't hold any
  * locks that are also used by the callback. */
@@ -145,6 +151,7 @@ void vhost_poll_flush(struct vhost_poll *poll)
 {
 	vhost_work_flush(poll->dev, &poll->work);
 }
+EXPORT_SYMBOL_GPL(vhost_poll_flush);
 
 void vhost_work_queue(struct vhost_dev *dev, struct vhost_work *work)
 {
@@ -154,15 +161,19 @@ void vhost_work_queue(struct vhost_dev *dev, struct vhost_work *work)
 	if (list_empty(&work->node)) {
 		list_add_tail(&work->node, &dev->work_list);
 		work->queue_seq++;
+		spin_unlock_irqrestore(&dev->work_lock, flags);
 		wake_up_process(dev->worker);
+	} else {
+		spin_unlock_irqrestore(&dev->work_lock, flags);
 	}
-	spin_unlock_irqrestore(&dev->work_lock, flags);
 }
+EXPORT_SYMBOL_GPL(vhost_work_queue);
 
 void vhost_poll_queue(struct vhost_poll *poll)
 {
 	vhost_work_queue(poll->dev, &poll->work);
 }
+EXPORT_SYMBOL_GPL(vhost_poll_queue);
 
 static void vhost_vq_reset(struct vhost_dev *dev,
 			   struct vhost_virtqueue *vq)
@@ -251,17 +262,16 @@ static void vhost_vq_free_iovecs(struct vhost_virtqueue *vq)
 /* Helper to allocate iovec buffers for all vqs. */
 static long vhost_dev_alloc_iovecs(struct vhost_dev *dev)
 {
+	struct vhost_virtqueue *vq;
 	int i;
 
 	for (i = 0; i < dev->nvqs; ++i) {
-		dev->vqs[i]->indirect = kmalloc(sizeof *dev->vqs[i]->indirect *
-					       UIO_MAXIOV, GFP_KERNEL);
-		dev->vqs[i]->log = kmalloc(sizeof *dev->vqs[i]->log * UIO_MAXIOV,
-					  GFP_KERNEL);
-		dev->vqs[i]->heads = kmalloc(sizeof *dev->vqs[i]->heads *
-					    UIO_MAXIOV, GFP_KERNEL);
-		if (!dev->vqs[i]->indirect || !dev->vqs[i]->log ||
-			!dev->vqs[i]->heads)
+		vq = dev->vqs[i];
+		vq->indirect = kmalloc(sizeof *vq->indirect * UIO_MAXIOV,
+				       GFP_KERNEL);
+		vq->log = kmalloc(sizeof *vq->log * UIO_MAXIOV, GFP_KERNEL);
+		vq->heads = kmalloc(sizeof *vq->heads * UIO_MAXIOV, GFP_KERNEL);
+		if (!vq->indirect || !vq->log || !vq->heads)
 			goto err_nomem;
 	}
 	return 0;
@@ -283,6 +293,7 @@ static void vhost_dev_free_iovecs(struct vhost_dev *dev)
 long vhost_dev_init(struct vhost_dev *dev,
 		    struct vhost_virtqueue **vqs, int nvqs)
 {
+	struct vhost_virtqueue *vq;
 	int i;
 
 	dev->vqs = vqs;
@@ -297,19 +308,21 @@ long vhost_dev_init(struct vhost_dev *dev,
 	dev->worker = NULL;
 
 	for (i = 0; i < dev->nvqs; ++i) {
-		dev->vqs[i]->log = NULL;
-		dev->vqs[i]->indirect = NULL;
-		dev->vqs[i]->heads = NULL;
-		dev->vqs[i]->dev = dev;
-		mutex_init(&dev->vqs[i]->mutex);
-		vhost_vq_reset(dev, dev->vqs[i]);
-		if (dev->vqs[i]->handle_kick)
-			vhost_poll_init(&dev->vqs[i]->poll,
-					dev->vqs[i]->handle_kick, POLLIN, dev);
+		vq = dev->vqs[i];
+		vq->log = NULL;
+		vq->indirect = NULL;
+		vq->heads = NULL;
+		vq->dev = dev;
+		mutex_init(&vq->mutex);
+		vhost_vq_reset(dev, vq);
+		if (vq->handle_kick)
+			vhost_poll_init(&vq->poll, vq->handle_kick,
+					POLLIN, dev);
 	}
 
 	return 0;
 }
+EXPORT_SYMBOL_GPL(vhost_dev_init);
 
 /* Caller should have device mutex */
 long vhost_dev_check_owner(struct vhost_dev *dev)
@@ -317,6 +330,7 @@ long vhost_dev_check_owner(struct vhost_dev *dev)
 	/* Are you the owner? If not, I don't think you mean to do that */
 	return dev->mm == current->mm ? 0 : -EPERM;
 }
+EXPORT_SYMBOL_GPL(vhost_dev_check_owner);
 
 struct vhost_attach_cgroups_struct {
 	struct vhost_work work;
@@ -344,13 +358,20 @@ static int vhost_attach_cgroups(struct vhost_dev *dev)
 }
 
 /* Caller should have device mutex */
+bool vhost_dev_has_owner(struct vhost_dev *dev)
+{
+	return dev->mm;
+}
+EXPORT_SYMBOL_GPL(vhost_dev_has_owner);
+
+/* Caller should have device mutex */
 long vhost_dev_set_owner(struct vhost_dev *dev)
 {
 	struct task_struct *worker;
 	int err;
 
 	/* Is there an owner already? */
-	if (dev->mm) {
+	if (vhost_dev_has_owner(dev)) {
 		err = -EBUSY;
 		goto err_mm;
 	}
@@ -385,11 +406,13 @@ err_worker:
 err_mm:
 	return err;
 }
+EXPORT_SYMBOL_GPL(vhost_dev_set_owner);
 
 struct vhost_memory *vhost_dev_reset_owner_prepare(void)
 {
 	return kmalloc(offsetof(struct vhost_memory, regions), GFP_KERNEL);
 }
+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)
@@ -400,6 +423,7 @@ void vhost_dev_reset_owner(struct vhost_dev *dev, struct vhost_memory *memory)
 	memory->nregions = 0;
 	RCU_INIT_POINTER(dev->memory, memory);
 }
+EXPORT_SYMBOL_GPL(vhost_dev_reset_owner);
 
 void vhost_dev_stop(struct vhost_dev *dev)
 {
@@ -412,6 +436,7 @@ void vhost_dev_stop(struct vhost_dev *dev)
 		}
 	}
 }
+EXPORT_SYMBOL_GPL(vhost_dev_stop);
 
 /* Caller should have device mutex if and only if locked is set */
 void vhost_dev_cleanup(struct vhost_dev *dev, bool locked)
@@ -452,6 +477,7 @@ void vhost_dev_cleanup(struct vhost_dev *dev, bool locked)
 		mmput(dev->mm);
 	dev->mm = NULL;
 }
+EXPORT_SYMBOL_GPL(vhost_dev_cleanup);
 
 static int log_access_ok(void __user *log_base, u64 addr, unsigned long sz)
 {
@@ -537,6 +563,7 @@ int vhost_log_access_ok(struct vhost_dev *dev)
 				       lockdep_is_held(&dev->mutex));
 	return memory_access_ok(dev, mp, 1);
 }
+EXPORT_SYMBOL_GPL(vhost_log_access_ok);
 
 /* Verify access for write logging. */
 /* Caller should have vq mutex and device mutex */
@@ -562,6 +589,7 @@ int vhost_vq_access_ok(struct vhost_virtqueue *vq)
 	return vq_access_ok(vq->dev, vq->num, vq->desc, vq->avail, vq->used) &&
 		vq_log_access_ok(vq->dev, vq, vq->log_base);
 }
+EXPORT_SYMBOL_GPL(vhost_vq_access_ok);
 
 static long vhost_set_memory(struct vhost_dev *d, struct vhost_memory __user *m)
 {
@@ -791,6 +819,7 @@ long vhost_vring_ioctl(struct vhost_dev *d, int ioctl, void __user *argp)
 		vhost_poll_flush(&vq->poll);
 	return r;
 }
+EXPORT_SYMBOL_GPL(vhost_vring_ioctl);
 
 /* Caller must have device mutex */
 long vhost_dev_ioctl(struct vhost_dev *d, unsigned int ioctl, void __user *argp)
@@ -871,6 +900,7 @@ long vhost_dev_ioctl(struct vhost_dev *d, unsigned int ioctl, void __user *argp)
 done:
 	return r;
 }
+EXPORT_SYMBOL_GPL(vhost_dev_ioctl);
 
 static const struct vhost_memory_region *find_region(struct vhost_memory *mem,
 						     __u64 addr, __u32 len)
@@ -962,6 +992,7 @@ int vhost_log_write(struct vhost_virtqueue *vq, struct vhost_log *log,
 	BUG();
 	return 0;
 }
+EXPORT_SYMBOL_GPL(vhost_log_write);
 
 static int vhost_update_used_flags(struct vhost_virtqueue *vq)
 {
@@ -1013,6 +1044,7 @@ int vhost_init_used(struct vhost_virtqueue *vq)
 	vq->signalled_used_valid = false;
 	return get_user(vq->last_used_idx, &vq->used->idx);
 }
+EXPORT_SYMBOL_GPL(vhost_init_used);
 
 static int translate_desc(struct vhost_dev *dev, u64 addr, u32 len,
 			  struct iovec iov[], int iov_size)
@@ -1289,60 +1321,24 @@ int vhost_get_vq_desc(struct vhost_dev *dev, struct vhost_virtqueue *vq,
 	BUG_ON(!(vq->used_flags & VRING_USED_F_NO_NOTIFY));
 	return head;
 }
+EXPORT_SYMBOL_GPL(vhost_get_vq_desc);
 
 /* Reverse the effect of vhost_get_vq_desc. Useful for error handling. */
 void vhost_discard_vq_desc(struct vhost_virtqueue *vq, int n)
 {
 	vq->last_avail_idx -= n;
 }
+EXPORT_SYMBOL_GPL(vhost_discard_vq_desc);
 
 /* After we've used one of their buffers, we tell them about it.  We'll then
  * want to notify the guest, using eventfd. */
 int vhost_add_used(struct vhost_virtqueue *vq, unsigned int head, int len)
 {
-	struct vring_used_elem __user *used;
+	struct vring_used_elem heads = { head, len };
 
-	/* The virtqueue contains a ring of used buffers.  Get a pointer to the
-	 * next entry in that used ring. */
-	used = &vq->used->ring[vq->last_used_idx % vq->num];
-	if (__put_user(head, &used->id)) {
-		vq_err(vq, "Failed to write used id");
-		return -EFAULT;
-	}
-	if (__put_user(len, &used->len)) {
-		vq_err(vq, "Failed to write used len");
-		return -EFAULT;
-	}
-	/* Make sure buffer is written before we update index. */
-	smp_wmb();
-	if (__put_user(vq->last_used_idx + 1, &vq->used->idx)) {
-		vq_err(vq, "Failed to increment used idx");
-		return -EFAULT;
-	}
-	if (unlikely(vq->log_used)) {
-		/* Make sure data is seen before log. */
-		smp_wmb();
-		/* Log used ring entry write. */
-		log_write(vq->log_base,
-			  vq->log_addr +
-			   ((void __user *)used - (void __user *)vq->used),
-			  sizeof *used);
-		/* Log used index update. */
-		log_write(vq->log_base,
-			  vq->log_addr + offsetof(struct vring_used, idx),
-			  sizeof vq->used->idx);
-		if (vq->log_ctx)
-			eventfd_signal(vq->log_ctx, 1);
-	}
-	vq->last_used_idx++;
-	/* If the driver never bothers to signal in a very long while,
-	 * used index might wrap around. If that happens, invalidate
-	 * signalled_used index we stored. TODO: make sure driver
-	 * signals at least once in 2^16 and remove this. */
-	if (unlikely(vq->last_used_idx == vq->signalled_used))
-		vq->signalled_used_valid = false;
-	return 0;
+	return vhost_add_used_n(vq, &heads, 1);
 }
+EXPORT_SYMBOL_GPL(vhost_add_used);
 
 static int __vhost_add_used_n(struct vhost_virtqueue *vq,
 			    struct vring_used_elem *heads,
@@ -1354,7 +1350,16 @@ static int __vhost_add_used_n(struct vhost_virtqueue *vq,
 
 	start = vq->last_used_idx % vq->num;
 	used = vq->used->ring + start;
-	if (__copy_to_user(used, heads, count * sizeof *used)) {
+	if (count == 1) {
+		if (__put_user(heads[0].id, &used->id)) {
+			vq_err(vq, "Failed to write used id");
+			return -EFAULT;
+		}
+		if (__put_user(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)) {
 		vq_err(vq, "Failed to write used");
 		return -EFAULT;
 	}
@@ -1412,6 +1417,7 @@ int vhost_add_used_n(struct vhost_virtqueue *vq, struct vring_used_elem *heads,
 	}
 	return r;
 }
+EXPORT_SYMBOL_GPL(vhost_add_used_n);
 
 static bool vhost_notify(struct vhost_dev *dev, struct vhost_virtqueue *vq)
 {
@@ -1456,6 +1462,7 @@ void vhost_signal(struct vhost_dev *dev, struct vhost_virtqueue *vq)
 	if (vq->call_ctx && vhost_notify(dev, vq))
 		eventfd_signal(vq->call_ctx, 1);
 }
+EXPORT_SYMBOL_GPL(vhost_signal);
 
 /* And here's the combo meal deal.  Supersize me! */
 void vhost_add_used_and_signal(struct vhost_dev *dev,
@@ -1465,6 +1472,7 @@ void vhost_add_used_and_signal(struct vhost_dev *dev,
 	vhost_add_used(vq, head, len);
 	vhost_signal(dev, vq);
 }
+EXPORT_SYMBOL_GPL(vhost_add_used_and_signal);
 
 /* multi-buffer version of vhost_add_used_and_signal */
 void vhost_add_used_and_signal_n(struct vhost_dev *dev,
@@ -1474,6 +1482,7 @@ void vhost_add_used_and_signal_n(struct vhost_dev *dev,
 	vhost_add_used_n(vq, heads, count);
 	vhost_signal(dev, vq);
 }
+EXPORT_SYMBOL_GPL(vhost_add_used_and_signal_n);
 
 /* OK, now we need to know about added descriptors. */
 bool vhost_enable_notify(struct vhost_dev *dev, struct vhost_virtqueue *vq)
@@ -1511,6 +1520,7 @@ bool vhost_enable_notify(struct vhost_dev *dev, struct vhost_virtqueue *vq)
 
 	return avail_idx != vq->avail_idx;
 }
+EXPORT_SYMBOL_GPL(vhost_enable_notify);
 
 /* We don't need to be notified again. */
 void vhost_disable_notify(struct vhost_dev *dev, struct vhost_virtqueue *vq)
@@ -1527,3 +1537,21 @@ void vhost_disable_notify(struct vhost_dev *dev, struct vhost_virtqueue *vq)
 			       &vq->used->flags, r);
 	}
 }
+EXPORT_SYMBOL_GPL(vhost_disable_notify);
+
+static int __init vhost_init(void)
+{
+	return 0;
+}
+
+static void __exit vhost_exit(void)
+{
+}
+
+module_init(vhost_init);
+module_exit(vhost_exit);
+
+MODULE_VERSION("0.0.1");
+MODULE_LICENSE("GPL v2");
+MODULE_AUTHOR("Michael S. Tsirkin");
+MODULE_DESCRIPTION("Host kernel accelerator for virtio");
diff --git a/drivers/vhost/vhost.h b/drivers/vhost/vhost.h
index a7ad6359298..4465ed5f316 100644
--- a/drivers/vhost/vhost.h
+++ b/drivers/vhost/vhost.h
@@ -46,6 +46,8 @@ int vhost_poll_start(struct vhost_poll *poll, struct file *file);
 void vhost_poll_stop(struct vhost_poll *poll);
 void vhost_poll_flush(struct vhost_poll *poll);
 void vhost_poll_queue(struct vhost_poll *poll);
+void vhost_work_flush(struct vhost_dev *dev, struct vhost_work *work);
+long vhost_vring_ioctl(struct vhost_dev *d, int ioctl, void __user *argp);
 
 struct vhost_log {
 	u64 addr;
@@ -101,14 +103,8 @@ struct vhost_virtqueue {
 	struct iovec iov[UIO_MAXIOV];
 	struct iovec *indirect;
 	struct vring_used_elem *heads;
-	/* We use a kind of RCU to access private pointer.
-	 * All readers access it from worker, which makes it possible to
-	 * flush the vhost_work instead of synchronize_rcu. Therefore readers do
-	 * not need to call rcu_read_lock/rcu_read_unlock: the beginning of
-	 * vhost_work execution acts instead of rcu_read_lock() and the end of
-	 * vhost_work execution acts instead of rcu_read_unlock().
-	 * Writers use virtqueue mutex. */
-	void __rcu *private_data;
+	/* Protected by virtqueue mutex. */
+	void *private_data;
 	/* Log write descriptors */
 	void __user *log_base;
 	struct vhost_log *log;
@@ -133,6 +129,7 @@ struct vhost_dev {
 
 long 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 *);