21 files changed, 192 insertions, 72 deletions

diff --git a/lib/librte_cryptodev/rte_crypto_sym.h b/lib/librte_cryptodev/rte_crypto_sym.h
index d3d38e4f..8f81d254 100644
--- a/lib/librte_cryptodev/rte_crypto_sym.h
+++ b/lib/librte_cryptodev/rte_crypto_sym.h
@@ -276,11 +276,10 @@ struct rte_crypto_auth_xform {
 	 * this specifies the length of the digest to be compared for the
 	 * session.
 	 *
+	 * It is the caller's responsibility to ensure that the
+	 * digest length is compliant with the hash algorithm being used.
 	 * If the value is less than the maximum length allowed by the hash,
-	 * the result shall be truncated.  If the value is greater than the
-	 * maximum length allowed by the hash then an error will be generated
-	 * by *rte_cryptodev_sym_session_create* or by the
-	 * *rte_cryptodev_sym_enqueue_burst* if using session-less APIs.
+	 * the result shall be truncated.
 	 */
 
 	uint32_t add_auth_data_length;
@@ -571,7 +570,9 @@ struct rte_crypto_sym_op {
 			phys_addr_t phys_addr;
 			/**< Physical address of digest */
 			uint16_t length;
-			/**< Length of digest */
+			/**< Length of digest. This must be the same value as
+			 * @ref rte_crypto_auth_xform.digest_length.
+			 */
 		} digest; /**< Digest parameters */
 
 		struct {
diff --git a/lib/librte_cryptodev/rte_cryptodev_pmd.h b/lib/librte_cryptodev/rte_cryptodev_pmd.h
index c6a57945..e82dcf07 100644
--- a/lib/librte_cryptodev/rte_cryptodev_pmd.h
+++ b/lib/librte_cryptodev/rte_cryptodev_pmd.h
@@ -57,14 +57,6 @@ extern "C" {
 #include "rte_crypto.h"
 #include "rte_cryptodev.h"
 
-
-#ifdef RTE_LIBRTE_CRYPTODEV_DEBUG
-#define RTE_PMD_DEBUG_TRACE(...) \
-	rte_pmd_debug_trace(__func__, __VA_ARGS__)
-#else
-#define RTE_PMD_DEBUG_TRACE(...)
-#endif
-
 struct rte_cryptodev_session {
 	RTE_STD_C11
 	struct {
diff --git a/lib/librte_eal/bsdapp/eal/eal_pci.c b/lib/librte_eal/bsdapp/eal/eal_pci.c
index 8b3ed881..f1de16ea 100644
--- a/lib/librte_eal/bsdapp/eal/eal_pci.c
+++ b/lib/librte_eal/bsdapp/eal/eal_pci.c
@@ -323,6 +323,7 @@ pci_scan_one(int dev_pci_fd, struct pci_conf *conf)
 
 	/* device is valid, add in list (sorted) */
 	if (TAILQ_EMPTY(&pci_device_list)) {
+		rte_eal_device_insert(&dev->device);
 		TAILQ_INSERT_TAIL(&pci_device_list, dev, next);
 	}
 	else {
@@ -335,7 +336,7 @@ pci_scan_one(int dev_pci_fd, struct pci_conf *conf)
 				continue;
 			else if (ret < 0) {
 				TAILQ_INSERT_BEFORE(dev2, dev, next);
-				return 0;
+				rte_eal_device_insert(&dev->device);
 			} else { /* already registered */
 				dev2->kdrv = dev->kdrv;
 				dev2->max_vfs = dev->max_vfs;
@@ -343,9 +344,10 @@ pci_scan_one(int dev_pci_fd, struct pci_conf *conf)
 					dev->mem_resource,
 					sizeof(dev->mem_resource));
 				free(dev);
-				return 0;
 			}
+			return 0;
 		}
+		rte_eal_device_insert(&dev->device);
 		TAILQ_INSERT_TAIL(&pci_device_list, dev, next);
 	}
 
diff --git a/lib/librte_eal/bsdapp/nic_uio/nic_uio.c b/lib/librte_eal/bsdapp/nic_uio/nic_uio.c
index 99a4975c..4bd7545a 100644
--- a/lib/librte_eal/bsdapp/nic_uio/nic_uio.c
+++ b/lib/librte_eal/bsdapp/nic_uio/nic_uio.c
@@ -180,6 +180,10 @@ nic_uio_probe (device_t dev)
 	unsigned int device = pci_get_slot(dev);
 	unsigned int function = pci_get_function(dev);
 
+	char bdf_str[256];
+	char *token, *remaining;
+
+	/* First check if we found this on load */
 	for (i = 0; i < num_detached; i++)
 		if (bus == pci_get_bus(detached_devices[i]) &&
 		    device == pci_get_slot(detached_devices[i]) &&
@@ -188,6 +192,45 @@ nic_uio_probe (device_t dev)
 			return BUS_PROBE_SPECIFIC;
 		}
 
+	/* otherwise check if it's a new device and if it matches the BDF */
+	memset(bdf_str, 0, sizeof(bdf_str));
+	TUNABLE_STR_FETCH("hw.nic_uio.bdfs", bdf_str, sizeof(bdf_str));
+	remaining = bdf_str;
+	while (1) {
+		if (remaining == NULL || remaining[0] == '\0')
+			break;
+		token = strsep(&remaining, ",:");
+		if (token == NULL)
+			break;
+		bus = strtol(token, NULL, 10);
+		token = strsep(&remaining, ",:");
+		if (token == NULL)
+			break;
+		device = strtol(token, NULL, 10);
+		token = strsep(&remaining, ",:");
+		if (token == NULL)
+			break;
+		function = strtol(token, NULL, 10);
+
+		if (bus == pci_get_bus(dev) &&
+				device == pci_get_slot(dev) &&
+				function == pci_get_function(dev)) {
+
+			if (num_detached < MAX_DETACHED_DEVICES) {
+				printf("%s: probed dev=%p\n",
+					       __func__, dev);
+				detached_devices[num_detached++] = dev;
+				device_set_desc(dev, "DPDK PCI Device");
+				return BUS_PROBE_SPECIFIC;
+			} else {
+				printf("%s: reached MAX_DETACHED_DEVICES=%d. dev=%p won't be reattached\n",
+						__func__, MAX_DETACHED_DEVICES,
+						dev);
+				break;
+			}
+		}
+	}
+
 	return ENXIO;
 }
 
@@ -248,6 +291,7 @@ nic_uio_load(void)
 	memset(bdf_str, 0, sizeof(bdf_str));
 	TUNABLE_STR_FETCH("hw.nic_uio.bdfs", bdf_str, sizeof(bdf_str));
 	remaining = bdf_str;
+	printf("nic_uio: hw.nic_uio.bdfs = '%s'\n", bdf_str);
 	/*
 	 * Users should specify PCI BDFs in the format "b:d:f,b:d:f,b:d:f".
 	 *  But the code below does not try differentiate between : and ,
diff --git a/lib/librte_eal/common/include/rte_dev.h b/lib/librte_eal/common/include/rte_dev.h
index 8840380d..dcf33d0e 100644
--- a/lib/librte_eal/common/include/rte_dev.h
+++ b/lib/librte_eal/common/include/rte_dev.h
@@ -49,6 +49,7 @@ extern "C" {
 #include <stdio.h>
 #include <sys/queue.h>
 
+#include <rte_config.h>
 #include <rte_log.h>
 
 __attribute__((format(printf, 2, 0)))
@@ -70,6 +71,19 @@ rte_pmd_debug_trace(const char *func_name, const char *fmt, ...)
 	rte_log(RTE_LOG_ERR, RTE_LOGTYPE_PMD, "%s: %s", func_name, buffer);
 }
 
+/*
+ * Enable RTE_PMD_DEBUG_TRACE() when at least one component relying on the
+ * RTE_*_RET() macros defined below is compiled in debug mode.
+ */
+#if defined(RTE_LIBRTE_ETHDEV_DEBUG) || \
+	defined(RTE_LIBRTE_CRYPTODEV_DEBUG) || \
+	defined(RTE_LIBRTE_EVENTDEV_DEBUG)
+#define RTE_PMD_DEBUG_TRACE(...) \
+	rte_pmd_debug_trace(__func__, __VA_ARGS__)
+#else
+#define RTE_PMD_DEBUG_TRACE(...) (void)0
+#endif
+
 /* Macros for checking for restricting functions to primary instance only */
 #define RTE_PROC_PRIMARY_OR_ERR_RET(retval) do { \
 	if (rte_eal_process_type() != RTE_PROC_PRIMARY) { \
diff --git a/lib/librte_eal/common/include/rte_version.h b/lib/librte_eal/common/include/rte_version.h
index 0de35fb7..5bbe906b 100644
--- a/lib/librte_eal/common/include/rte_version.h
+++ b/lib/librte_eal/common/include/rte_version.h
@@ -66,7 +66,7 @@ extern "C" {
 /**
  * Patch level number i.e. the z in yy.mm.z
  */
-#define RTE_VER_MINOR 1
+#define RTE_VER_MINOR 2
 
 /**
  * Extra string to be appended to version number
diff --git a/lib/librte_eal/linuxapp/eal/eal_interrupts.c b/lib/librte_eal/linuxapp/eal/eal_interrupts.c
index 47a3b20a..368863f9 100644
--- a/lib/librte_eal/linuxapp/eal/eal_interrupts.c
+++ b/lib/librte_eal/linuxapp/eal/eal_interrupts.c
@@ -194,14 +194,14 @@ vfio_disable_intx(struct rte_intr_handle *intr_handle) {
 	irq_set = (struct vfio_irq_set *) irq_set_buf;
 	irq_set->argsz = len;
 	irq_set->count = 1;
-	irq_set->flags = VFIO_IRQ_SET_DATA_NONE | VFIO_IRQ_SET_ACTION_UNMASK;
+	irq_set->flags = VFIO_IRQ_SET_DATA_NONE | VFIO_IRQ_SET_ACTION_MASK;
 	irq_set->index = VFIO_PCI_INTX_IRQ_INDEX;
 	irq_set->start = 0;
 
 	ret = ioctl(intr_handle->vfio_dev_fd, VFIO_DEVICE_SET_IRQS, irq_set);
 
 	if (ret) {
-		RTE_LOG(ERR, EAL, "Error unmasking INTx interrupts for fd %d\n",
+		RTE_LOG(ERR, EAL, "Error masking INTx interrupts for fd %d\n",
 						intr_handle->fd);
 		return -1;
 	}
diff --git a/lib/librte_eal/linuxapp/eal/eal_pci_uio.c b/lib/librte_eal/linuxapp/eal/eal_pci_uio.c
index 3e4ffb57..aac05d7a 100644
--- a/lib/librte_eal/linuxapp/eal/eal_pci_uio.c
+++ b/lib/librte_eal/linuxapp/eal/eal_pci_uio.c
@@ -38,6 +38,7 @@
 #include <inttypes.h>
 #include <sys/stat.h>
 #include <sys/mman.h>
+#include <sys/sysmacros.h>
 #include <linux/pci_regs.h>
 
 #if defined(RTE_ARCH_X86)
diff --git a/lib/librte_eal/linuxapp/eal/eal_pci_vfio.c b/lib/librte_eal/linuxapp/eal/eal_pci_vfio.c
index 5f478c59..7d8b9fb2 100644
--- a/lib/librte_eal/linuxapp/eal/eal_pci_vfio.c
+++ b/lib/librte_eal/linuxapp/eal/eal_pci_vfio.c
@@ -355,7 +355,8 @@ pci_vfio_map_resource(struct rte_pci_device *dev)
 	} else {
 		/* if we're in a secondary process, just find our tailq entry */
 		TAILQ_FOREACH(vfio_res, vfio_res_list, next) {
-			if (memcmp(&vfio_res->pci_addr, &dev->addr, sizeof(dev->addr)))
+			if (rte_eal_compare_pci_addr(&vfio_res->pci_addr,
+						     &dev->addr))
 				continue;
 			break;
 		}
diff --git a/lib/librte_eal/linuxapp/kni/compat.h b/lib/librte_eal/linuxapp/kni/compat.h
index 78da08e5..d96275af 100644
--- a/lib/librte_eal/linuxapp/kni/compat.h
+++ b/lib/librte_eal/linuxapp/kni/compat.h
@@ -2,6 +2,8 @@
  * Minimal wrappers to allow compiling kni on older kernels.
  */
 
+#include <linux/version.h>
+
 #ifndef RHEL_RELEASE_VERSION
 #define RHEL_RELEASE_VERSION(a, b) (((a) << 8) + (b))
 #endif
@@ -67,3 +69,7 @@
 	(LINUX_VERSION_CODE < KERNEL_VERSION(2, 6, 34)))
 #undef NET_NAME_UNKNOWN
 #endif
+
+#if LINUX_VERSION_CODE >= KERNEL_VERSION(4, 11, 0)
+#define HAVE_SIGNAL_FUNCTIONS_OWN_HEADER
+#endif
diff --git a/lib/librte_eal/linuxapp/kni/kni_dev.h b/lib/librte_eal/linuxapp/kni/kni_dev.h
index 58cbadd3..f0f6e61f 100644
--- a/lib/librte_eal/linuxapp/kni/kni_dev.h
+++ b/lib/librte_eal/linuxapp/kni/kni_dev.h
@@ -30,9 +30,15 @@
 #endif
 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
 
+#include "compat.h"
+
 #include <linux/if.h>
 #include <linux/wait.h>
+#ifdef HAVE_SIGNAL_FUNCTIONS_OWN_HEADER
+#include <linux/sched/signal.h>
+#else
 #include <linux/sched.h>
+#endif
 #include <linux/netdevice.h>
 #include <linux/spinlock.h>
 #include <linux/list.h>
diff --git a/lib/librte_ether/rte_ethdev.c b/lib/librte_ether/rte_ethdev.c
index 5a317594..ea545250 100644
--- a/lib/librte_ether/rte_ethdev.c
+++ b/lib/librte_ether/rte_ethdev.c
@@ -1051,8 +1051,10 @@ rte_eth_dev_close(uint8_t port_id)
 	dev->data->dev_started = 0;
 	(*dev->dev_ops->dev_close)(dev);
 
+	dev->data->nb_rx_queues = 0;
 	rte_free(dev->data->rx_queues);
 	dev->data->rx_queues = NULL;
+	dev->data->nb_tx_queues = 0;
 	rte_free(dev->data->tx_queues);
 	dev->data->tx_queues = NULL;
 }
diff --git a/lib/librte_ether/rte_ethdev.h b/lib/librte_ether/rte_ethdev.h
index 96781792..11ec1fa8 100644
--- a/lib/librte_ether/rte_ethdev.h
+++ b/lib/librte_ether/rte_ethdev.h
@@ -1001,15 +1001,6 @@ struct rte_eth_dev_callback;
 /** @internal Structure to keep track of registered callbacks */
 TAILQ_HEAD(rte_eth_dev_cb_list, rte_eth_dev_callback);
 
-
-#ifdef RTE_LIBRTE_ETHDEV_DEBUG
-#define RTE_PMD_DEBUG_TRACE(...) \
-	rte_pmd_debug_trace(__func__, __VA_ARGS__)
-#else
-#define RTE_PMD_DEBUG_TRACE(...)
-#endif
-
-
 /* Macros to check for valid port */
 #define RTE_ETH_VALID_PORTID_OR_ERR_RET(port_id, retval) do { \
 	if (!rte_eth_dev_is_valid_port(port_id)) { \
diff --git a/lib/librte_kni/rte_kni.c b/lib/librte_kni/rte_kni.c
index a80cefd2..c3f9208c 100644
--- a/lib/librte_kni/rte_kni.c
+++ b/lib/librte_kni/rte_kni.c
@@ -451,17 +451,35 @@ kni_free_fifo(struct rte_kni_fifo *fifo)
 	} while (ret);
 }
 
+static void *
+va2pa(struct rte_mbuf *m)
+{
+	return (void *)((unsigned long)m -
+			((unsigned long)m->buf_addr -
+			 (unsigned long)m->buf_physaddr));
+}
+
+static void
+obj_free(struct rte_mempool *mp __rte_unused, void *opaque, void *obj,
+		unsigned obj_idx __rte_unused)
+{
+	struct rte_mbuf *m = obj;
+	void *mbuf_phys = opaque;
+
+	if (va2pa(m) == mbuf_phys)
+		rte_pktmbuf_free(m);
+}
+
 static void
-kni_free_fifo_phy(struct rte_kni_fifo *fifo)
+kni_free_fifo_phy(struct rte_mempool *mp, struct rte_kni_fifo *fifo)
 {
 	void *mbuf_phys;
 	int ret;
 
 	do {
 		ret = kni_fifo_get(fifo, &mbuf_phys, 1);
-		/*
-		 * TODO: free mbufs
-		 */
+		if (ret)
+			rte_mempool_obj_iter(mp, obj_free, mbuf_phys);
 	} while (ret);
 }
 
@@ -470,6 +488,7 @@ rte_kni_release(struct rte_kni *kni)
 {
 	struct rte_kni_device_info dev_info;
 	uint32_t slot_id;
+	uint32_t retry = 5;
 
 	if (!kni || !kni->in_use)
 		return -1;
@@ -481,9 +500,16 @@ rte_kni_release(struct rte_kni *kni)
 	}
 
 	/* mbufs in all fifo should be released, except request/response */
+
+	/* wait until all rxq packets processed by kernel */
+	while (kni_fifo_count(kni->rx_q) && retry--)
+		usleep(1000);
+
+	if (kni_fifo_count(kni->rx_q))
+		RTE_LOG(ERR, KNI, "Fail to free all Rx-q items\n");
+
+	kni_free_fifo_phy(kni->pktmbuf_pool, kni->alloc_q);
 	kni_free_fifo(kni->tx_q);
-	kni_free_fifo_phy(kni->rx_q);
-	kni_free_fifo_phy(kni->alloc_q);
 	kni_free_fifo(kni->free_q);
 
 	slot_id = kni->slot_id;
@@ -549,14 +575,6 @@ rte_kni_handle_request(struct rte_kni *kni)
 	return 0;
 }
 
-static void *
-va2pa(struct rte_mbuf *m)
-{
-	return (void *)((unsigned long)m -
-			((unsigned long)m->buf_addr -
-			 (unsigned long)m->buf_physaddr));
-}
-
 unsigned
 rte_kni_tx_burst(struct rte_kni *kni, struct rte_mbuf **mbufs, unsigned num)
 {
diff --git a/lib/librte_kni/rte_kni_fifo.h b/lib/librte_kni/rte_kni_fifo.h
index 8cb85873..c7cd5c26 100644
--- a/lib/librte_kni/rte_kni_fifo.h
+++ b/lib/librte_kni/rte_kni_fifo.h
@@ -91,3 +91,12 @@ kni_fifo_get(struct rte_kni_fifo *fifo, void **data, unsigned num)
 	fifo->read = new_read;
 	return i;
 }
+
+/**
+ * Get the num of elements in the fifo
+ */
+static inline uint32_t
+kni_fifo_count(struct rte_kni_fifo *fifo)
+{
+	return (fifo->len + fifo->write - fifo->read) & (fifo->len - 1);
+}
diff --git a/lib/librte_mbuf/rte_mbuf_ptype.h b/lib/librte_mbuf/rte_mbuf_ptype.h
index ff6de9d1..a3269c4c 100644
--- a/lib/librte_mbuf/rte_mbuf_ptype.h
+++ b/lib/librte_mbuf/rte_mbuf_ptype.h
@@ -91,6 +91,9 @@
  * RTE_PTYPE_INNER_L4_UDP.
  */
 
+#include <stddef.h>
+#include <stdint.h>
+
 #ifdef __cplusplus
 extern "C" {
 #endif
diff --git a/lib/librte_net/rte_ether.h b/lib/librte_net/rte_ether.h
index ff3d0654..5edf66c3 100644
--- a/lib/librte_net/rte_ether.h
+++ b/lib/librte_net/rte_ether.h
@@ -357,7 +357,7 @@ static inline int rte_vlan_strip(struct rte_mbuf *m)
 		return -1;
 
 	struct vlan_hdr *vh = (struct vlan_hdr *)(eh + 1);
-	m->ol_flags |= PKT_RX_VLAN_PKT;
+	m->ol_flags |= PKT_RX_VLAN_PKT | PKT_RX_VLAN_STRIPPED;
 	m->vlan_tci = rte_be_to_cpu_16(vh->vlan_tci);
 
 	/* Copy ether header over rather than moving whole packet */
@@ -407,6 +407,8 @@ static inline int rte_vlan_insert(struct rte_mbuf **m)
 	vh = (struct vlan_hdr *) (nh + 1);
 	vh->vlan_tci = rte_cpu_to_be_16((*m)->vlan_tci);
 
+	(*m)->ol_flags &= ~PKT_RX_VLAN_STRIPPED;
+
 	return 0;
 }
 
diff --git a/lib/librte_vhost/socket.c b/lib/librte_vhost/socket.c
index aaa9c270..84e05951 100644
--- a/lib/librte_vhost/socket.c
+++ b/lib/librte_vhost/socket.c
@@ -52,14 +52,18 @@
 #include "vhost.h"
 #include "vhost_user.h"
 
+
+TAILQ_HEAD(vhost_user_connection_list, vhost_user_connection);
+
 /*
  * Every time rte_vhost_driver_register() is invoked, an associated
  * vhost_user_socket struct will be created.
  */
 struct vhost_user_socket {
+	struct vhost_user_connection_list conn_list;
+	pthread_mutex_t conn_mutex;
 	char *path;
 	int listenfd;
-	int connfd;
 	bool is_server;
 	bool reconnect;
 	bool dequeue_zero_copy;
@@ -67,7 +71,10 @@ struct vhost_user_socket {
 
 struct vhost_user_connection {
 	struct vhost_user_socket *vsocket;
+	int connfd;
 	int vid;
+
+	TAILQ_ENTRY(vhost_user_connection) next;
 };
 
 #define MAX_VHOST_SOCKET 1024
@@ -209,19 +216,24 @@ vhost_user_add_connection(int fd, struct vhost_user_socket *vsocket)
 
 	RTE_LOG(INFO, VHOST_CONFIG, "new device, handle is %d\n", vid);
 
-	vsocket->connfd = fd;
+	conn->connfd = fd;
 	conn->vsocket = vsocket;
 	conn->vid = vid;
 	ret = fdset_add(&vhost_user.fdset, fd, vhost_user_read_cb,
 			NULL, conn);
 	if (ret < 0) {
-		vsocket->connfd = -1;
+		conn->connfd = -1;
 		free(conn);
 		close(fd);
 		RTE_LOG(ERR, VHOST_CONFIG,
 			"failed to add fd %d into vhost server fdset\n",
 			fd);
+		return;
 	}
+
+	pthread_mutex_lock(&vsocket->conn_mutex);
+	TAILQ_INSERT_TAIL(&vsocket->conn_list, conn, next);
+	pthread_mutex_unlock(&vsocket->conn_mutex);
 }
 
 /* call back when there is new vhost-user connection from client  */
@@ -247,10 +259,14 @@ vhost_user_read_cb(int connfd, void *dat, int *remove)
 
 	ret = vhost_user_msg_handler(conn->vid, connfd);
 	if (ret < 0) {
-		vsocket->connfd = -1;
 		close(connfd);
 		*remove = 1;
 		vhost_destroy_device(conn->vid);
+
+		pthread_mutex_lock(&vsocket->conn_mutex);
+		TAILQ_REMOVE(&vsocket->conn_list, conn, next);
+		pthread_mutex_unlock(&vsocket->conn_mutex);
+
 		free(conn);
 
 		if (vsocket->reconnect)
@@ -448,7 +464,7 @@ vhost_user_create_client(struct vhost_user_socket *vsocket)
 		return 0;
 	}
 
-	RTE_LOG(ERR, VHOST_CONFIG,
+	RTE_LOG(WARNING, VHOST_CONFIG,
 		"failed to connect to %s: %s\n",
 		path, strerror(errno));
 
@@ -457,7 +473,7 @@ vhost_user_create_client(struct vhost_user_socket *vsocket)
 		return -1;
 	}
 
-	RTE_LOG(ERR, VHOST_CONFIG, "%s: reconnecting...\n", path);
+	RTE_LOG(INFO, VHOST_CONFIG, "%s: reconnecting...\n", path);
 	reconn = malloc(sizeof(*reconn));
 	if (reconn == NULL) {
 		RTE_LOG(ERR, VHOST_CONFIG,
@@ -502,7 +518,8 @@ rte_vhost_driver_register(const char *path, uint64_t flags)
 		goto out;
 	memset(vsocket, 0, sizeof(struct vhost_user_socket));
 	vsocket->path = strdup(path);
-	vsocket->connfd = -1;
+	TAILQ_INIT(&vsocket->conn_list);
+	pthread_mutex_init(&vsocket->conn_mutex, NULL);
 	vsocket->dequeue_zero_copy = flags & RTE_VHOST_USER_DEQUEUE_ZERO_COPY;
 
 	if ((flags & RTE_VHOST_USER_CLIENT) != 0) {
@@ -565,7 +582,7 @@ rte_vhost_driver_unregister(const char *path)
 {
 	int i;
 	int count;
-	struct vhost_user_connection *conn;
+	struct vhost_user_connection *conn, *next;
 
 	pthread_mutex_lock(&vhost_user.mutex);
 
@@ -581,15 +598,22 @@ rte_vhost_driver_unregister(const char *path)
 				vhost_user_remove_reconnect(vsocket);
 			}
 
-			conn = fdset_del(&vhost_user.fdset, vsocket->connfd);
-			if (conn) {
+			pthread_mutex_lock(&vsocket->conn_mutex);
+			for (conn = TAILQ_FIRST(&vsocket->conn_list);
+			     conn != NULL;
+			     conn = next) {
+				next = TAILQ_NEXT(conn, next);
+
+				fdset_del(&vhost_user.fdset, conn->connfd);
 				RTE_LOG(INFO, VHOST_CONFIG,
 					"free connfd = %d for device '%s'\n",
-					vsocket->connfd, path);
-				close(vsocket->connfd);
+					conn->connfd, path);
+				close(conn->connfd);
 				vhost_destroy_device(conn->vid);
+				TAILQ_REMOVE(&vsocket->conn_list, conn, next);
 				free(conn);
 			}
+			pthread_mutex_unlock(&vsocket->conn_mutex);
 
 			free(vsocket->path);
 			free(vsocket);
diff --git a/lib/librte_vhost/vhost.c b/lib/librte_vhost/vhost.c
index e4150934..3c3f6a42 100644
--- a/lib/librte_vhost/vhost.c
+++ b/lib/librte_vhost/vhost.c
@@ -56,7 +56,7 @@
 				(1ULL << VIRTIO_NET_F_CTRL_VQ) | \
 				(1ULL << VIRTIO_NET_F_CTRL_RX) | \
 				(1ULL << VIRTIO_NET_F_GUEST_ANNOUNCE) | \
-				(VHOST_SUPPORTS_MQ)            | \
+				(1ULL << VIRTIO_NET_F_MQ)      | \
 				(1ULL << VIRTIO_F_VERSION_1)   | \
 				(1ULL << VHOST_F_LOG_ALL)      | \
 				(1ULL << VHOST_USER_F_PROTOCOL_FEATURES) | \
diff --git a/lib/librte_vhost/vhost.h b/lib/librte_vhost/vhost.h
index 22564f1c..d97df1d8 100644
--- a/lib/librte_vhost/vhost.h
+++ b/lib/librte_vhost/vhost.h
@@ -110,25 +110,15 @@ struct vhost_virtqueue {
 	uint16_t                shadow_used_idx;
 } __rte_cache_aligned;
 
-/* Old kernels have no such macro defined */
+/* Old kernels have no such macros defined */
 #ifndef VIRTIO_NET_F_GUEST_ANNOUNCE
  #define VIRTIO_NET_F_GUEST_ANNOUNCE 21
 #endif
 
-
-/*
- * Make an extra wrapper for VIRTIO_NET_F_MQ and
- * VIRTIO_NET_CTRL_MQ_VQ_PAIRS_MAX as they are
- * introduced since kernel v3.8. This makes our
- * code buildable for older kernel.
- */
-#ifdef VIRTIO_NET_F_MQ
- #define VHOST_MAX_QUEUE_PAIRS	VIRTIO_NET_CTRL_MQ_VQ_PAIRS_MAX
- #define VHOST_SUPPORTS_MQ	(1ULL << VIRTIO_NET_F_MQ)
-#else
- #define VHOST_MAX_QUEUE_PAIRS	1
- #define VHOST_SUPPORTS_MQ	0
+#ifndef VIRTIO_NET_F_MQ
+ #define VIRTIO_NET_F_MQ		22
 #endif
+#define VHOST_MAX_QUEUE_PAIRS		0x80
 
 /*
  * Define virtio 1.0 for older kernels
diff --git a/lib/librte_vhost/virtio_net.c b/lib/librte_vhost/virtio_net.c
index 337470d6..ea027f14 100644
--- a/lib/librte_vhost/virtio_net.c
+++ b/lib/librte_vhost/virtio_net.c
@@ -905,6 +905,8 @@ copy_desc_to_mbuf(struct virtio_net *dev, struct vring_desc *descs,
 					"allocate memory for mbuf.\n");
 				return -1;
 			}
+			if (unlikely(dev->dequeue_zero_copy))
+				rte_mbuf_refcnt_update(cur, 1);
 
 			prev->next = cur;
 			prev->data_len = mbuf_offset;
@@ -1056,9 +1058,21 @@ rte_vhost_dequeue_burst(int vid, uint16_t queue_id,
 	 * array, to looks like that guest actually send such packet.
 	 *
 	 * Check user_send_rarp() for more information.
+	 *
+	 * broadcast_rarp shares a cacheline in the virtio_net structure
+	 * with some fields that are accessed during enqueue and
+	 * rte_atomic16_cmpset() causes a write if using cmpxchg. This could
+	 * result in false sharing between enqueue and dequeue.
+	 *
+	 * Prevent unnecessary false sharing by reading broadcast_rarp first
+	 * and only performing cmpset if the read indicates it is likely to
+	 * be set.
 	 */
-	if (unlikely(rte_atomic16_cmpset((volatile uint16_t *)
-					 &dev->broadcast_rarp.cnt, 1, 0))) {
+
+	if (unlikely(rte_atomic16_read(&dev->broadcast_rarp) &&
+			rte_atomic16_cmpset((volatile uint16_t *)
+				&dev->broadcast_rarp.cnt, 1, 0))) {
+
 		rarp_mbuf = rte_pktmbuf_alloc(mbuf_pool);
 		if (rarp_mbuf == NULL) {
 			RTE_LOG(ERR, VHOST_DATA,