X-Git-Url: http://git.droids-corp.org/?a=blobdiff_plain;f=examples%2Fvhost%2Fmain.c;h=85ee8b8f11abbc8281d00cc0bbf38d80392aefe5;hb=08b563ffb19d8baf59dd84200f25bc85031d18a7;hp=e0fc2c98c21f04eb639228fcf9f7395bee3dd06b;hpb=7b79b2718f0d028cc0f38351dfbf36c0f5a1a48b;p=dpdk.git

diff --git a/examples/vhost/main.c b/examples/vhost/main.c
index e0fc2c98c2..85ee8b8f11 100644
--- a/examples/vhost/main.c
+++ b/examples/vhost/main.c
@@ -106,6 +106,8 @@
 #define BURST_RX_WAIT_US 15 	/* Defines how long we wait between retries on RX */
 #define BURST_RX_RETRIES 4		/* Number of retries on RX. */
 
+#define JUMBO_FRAME_MAX_SIZE    0x2600
+
 /* State of virtio device. */
 #define DEVICE_MAC_LEARNING 0
 #define DEVICE_RX			1
@@ -143,7 +145,7 @@
 /* Max number of devices. Limited by vmdq. */
 #define MAX_DEVICES 64
 
-/* Size of buffers used for rte_snprintfs. */
+/* Size of buffers used for snprintfs. */
 #define MAX_PRINT_BUFF 6072
 
 /* Maximum character device basename size. */
@@ -498,7 +500,7 @@ us_vhost_parse_basename(const char *q_arg)
 	if (strnlen(q_arg, MAX_BASENAME_SZ) > MAX_BASENAME_SZ)
 		return -1;
 	else
-		rte_snprintf((char*)&dev_basename, MAX_BASENAME_SZ, "%s", q_arg);
+		snprintf((char*)&dev_basename, MAX_BASENAME_SZ, "%s", q_arg);
 
 	return 0;
 }
@@ -676,8 +678,12 @@ us_vhost_parse_args(int argc, char **argv)
 					us_vhost_usage(prgname);
 					return -1;
 				} else {
-					if (ret)
+					if (ret) {
+						vmdq_conf_default.rxmode.jumbo_frame = 1;
+						vmdq_conf_default.rxmode.max_rx_pkt_len
+							= JUMBO_FRAME_MAX_SIZE;
 						VHOST_FEATURES = (1ULL << VIRTIO_NET_F_MRG_RXBUF);
+					}
 				}
 			}
 
@@ -727,10 +733,10 @@ us_vhost_parse_args(int argc, char **argv)
 					zero_copy = ret;
 
 				if (zero_copy) {
-#ifdef RTE_MBUF_SCATTER_GATHER
+#ifdef RTE_MBUF_REFCNT
 					RTE_LOG(ERR, VHOST_CONFIG, "Before running "
 					"zero copy vhost APP, please "
-					"disable RTE_MBUF_SCATTER_GATHER\n"
+					"disable RTE_MBUF_REFCNT\n"
 					"in config file and then rebuild DPDK "
 					"core lib!\n"
 					"Otherwise please disable zero copy "
@@ -797,6 +803,14 @@ us_vhost_parse_args(int argc, char **argv)
 		return -1;
 	}
 
+	if ((zero_copy == 1) && (vmdq_conf_default.rxmode.jumbo_frame == 1)) {
+		RTE_LOG(INFO, VHOST_PORT,
+			"Vhost zero copy doesn't support jumbo frame,"
+			"please specify '--mergeable 0' to disable the "
+			"mergeable feature.\n");
+		return -1;
+	}
+
 	return 0;
 }
 
@@ -837,14 +851,14 @@ static unsigned check_ports_num(unsigned nb_ports)
 	char packet[MAX_PRINT_BUFF];																					\
 																													\
 	if ((header))																									\
-		rte_snprintf(packet, MAX_PRINT_BUFF, "(%"PRIu64") Header size %d: ", (device->device_fh), (size));				\
+		snprintf(packet, MAX_PRINT_BUFF, "(%"PRIu64") Header size %d: ", (device->device_fh), (size));				\
 	else																											\
-		rte_snprintf(packet, MAX_PRINT_BUFF, "(%"PRIu64") Packet size %d: ", (device->device_fh), (size));				\
+		snprintf(packet, MAX_PRINT_BUFF, "(%"PRIu64") Packet size %d: ", (device->device_fh), (size));				\
 	for (index = 0; index < (size); index++) {																		\
-		rte_snprintf(packet + strnlen(packet, MAX_PRINT_BUFF), MAX_PRINT_BUFF - strnlen(packet, MAX_PRINT_BUFF),	\
+		snprintf(packet + strnlen(packet, MAX_PRINT_BUFF), MAX_PRINT_BUFF - strnlen(packet, MAX_PRINT_BUFF),	\
 			"%02hhx ", pkt_addr[index]);																			\
 	}																												\
-	rte_snprintf(packet + strnlen(packet, MAX_PRINT_BUFF), MAX_PRINT_BUFF - strnlen(packet, MAX_PRINT_BUFF), "\n");	\
+	snprintf(packet + strnlen(packet, MAX_PRINT_BUFF), MAX_PRINT_BUFF - strnlen(packet, MAX_PRINT_BUFF), "\n");	\
 																													\
 	LOG_DEBUG(VHOST_DATA, "%s", packet);																					\
 } while(0)
@@ -916,7 +930,7 @@ gpa_to_hpa(struct virtio_net *dev, uint64_t guest_pa,
  * This function adds buffers to the virtio devices RX virtqueue. Buffers can
  * be received from the physical port or from another virtio device. A packet
  * count is returned to indicate the number of packets that were succesfully
- * added to the RX queue.
+ * added to the RX queue. This function works when mergeable is disabled.
  */
 static inline uint32_t __attribute__((always_inline))
 virtio_dev_rx(struct virtio_net *dev, struct rte_mbuf **pkts, uint32_t count)
@@ -930,7 +944,6 @@ virtio_dev_rx(struct virtio_net *dev, struct rte_mbuf **pkts, uint32_t count)
 	uint64_t buff_hdr_addr = 0;
 	uint32_t head[MAX_PKT_BURST], packet_len = 0;
 	uint32_t head_idx, packet_success = 0;
-	uint32_t mergeable, mrg_count = 0;
 	uint32_t retry = 0;
 	uint16_t avail_idx, res_cur_idx;
 	uint16_t res_base_idx, res_end_idx;
@@ -940,6 +953,7 @@ virtio_dev_rx(struct virtio_net *dev, struct rte_mbuf **pkts, uint32_t count)
 	LOG_DEBUG(VHOST_DATA, "(%"PRIu64") virtio_dev_rx()\n", dev->device_fh);
 	vq = dev->virtqueue[VIRTIO_RXQ];
 	count = (count > MAX_PKT_BURST) ? MAX_PKT_BURST : count;
+
 	/* As many data cores may want access to available buffers, they need to be reserved. */
 	do {
 		res_base_idx = vq->last_used_idx_res;
@@ -976,9 +990,6 @@ virtio_dev_rx(struct virtio_net *dev, struct rte_mbuf **pkts, uint32_t count)
 	/* Prefetch available ring to retrieve indexes. */
 	rte_prefetch0(&vq->avail->ring[res_cur_idx & (vq->size - 1)]);
 
-	/* Check if the VIRTIO_NET_F_MRG_RXBUF feature is enabled. */
-	mergeable = dev->features & (1 << VIRTIO_NET_F_MRG_RXBUF);
-
 	/* Retrieve all of the head indexes first to avoid caching issues. */
 	for (head_idx = 0; head_idx < count; head_idx++)
 		head[head_idx] = vq->avail->ring[(res_cur_idx + head_idx) & (vq->size - 1)];
@@ -997,56 +1008,44 @@ virtio_dev_rx(struct virtio_net *dev, struct rte_mbuf **pkts, uint32_t count)
 		/* Prefetch buffer address. */
 		rte_prefetch0((void*)(uintptr_t)buff_addr);
 
-		if (mergeable && (mrg_count != 0)) {
-			desc->len = packet_len = rte_pktmbuf_data_len(buff);
-		} else {
-			/* Copy virtio_hdr to packet and increment buffer address */
-			buff_hdr_addr = buff_addr;
-			packet_len = rte_pktmbuf_data_len(buff) + vq->vhost_hlen;
+		/* Copy virtio_hdr to packet and increment buffer address */
+		buff_hdr_addr = buff_addr;
+		packet_len = rte_pktmbuf_data_len(buff) + vq->vhost_hlen;
 
-			/*
-			 * If the descriptors are chained the header and data are placed in
-			 * separate buffers.
-			 */
-			if (desc->flags & VRING_DESC_F_NEXT) {
-				desc->len = vq->vhost_hlen;
-				desc = &vq->desc[desc->next];
-				/* Buffer address translation. */
-				buff_addr = gpa_to_vva(dev, desc->addr);
-				desc->len = rte_pktmbuf_data_len(buff);
-			} else {
-				buff_addr += vq->vhost_hlen;
-				desc->len = packet_len;
-			}
+		/*
+		 * If the descriptors are chained the header and data are
+		 * placed in separate buffers.
+		 */
+		if (desc->flags & VRING_DESC_F_NEXT) {
+			desc->len = vq->vhost_hlen;
+			desc = &vq->desc[desc->next];
+			/* Buffer address translation. */
+			buff_addr = gpa_to_vva(dev, desc->addr);
+			desc->len = rte_pktmbuf_data_len(buff);
+		} else {
+			buff_addr += vq->vhost_hlen;
+			desc->len = packet_len;
 		}
 
-		PRINT_PACKET(dev, (uintptr_t)buff_addr, rte_pktmbuf_data_len(buff), 0);
-
 		/* Update used ring with desc information */
 		vq->used->ring[res_cur_idx & (vq->size - 1)].id = head[packet_success];
 		vq->used->ring[res_cur_idx & (vq->size - 1)].len = packet_len;
 
 		/* Copy mbuf data to buffer */
-		rte_memcpy((void *)(uintptr_t)buff_addr, (const void*)buff->pkt.data, rte_pktmbuf_data_len(buff));
+		rte_memcpy((void *)(uintptr_t)buff_addr,
+			rte_pktmbuf_mtod(buff, const void *),
+			rte_pktmbuf_data_len(buff));
+		PRINT_PACKET(dev, (uintptr_t)buff_addr,
+			rte_pktmbuf_data_len(buff), 0);
 
 		res_cur_idx++;
 		packet_success++;
 
-		/* If mergeable is disabled then a header is required per buffer. */
-		if (!mergeable) {
-			rte_memcpy((void *)(uintptr_t)buff_hdr_addr, (const void*)&virtio_hdr, vq->vhost_hlen);
-			PRINT_PACKET(dev, (uintptr_t)buff_hdr_addr, vq->vhost_hlen, 1);
-		} else {
-			mrg_count++;
-			/* Merge buffer can only handle so many buffers at a time. Tell the guest if this limit is reached. */
-			if ((mrg_count == MAX_MRG_PKT_BURST) || (res_cur_idx == res_end_idx)) {
-				virtio_hdr.num_buffers = mrg_count;
-				LOG_DEBUG(VHOST_DATA, "(%"PRIu64") RX: Num merge buffers %d\n", dev->device_fh, virtio_hdr.num_buffers);
-				rte_memcpy((void *)(uintptr_t)buff_hdr_addr, (const void*)&virtio_hdr, vq->vhost_hlen);
-				PRINT_PACKET(dev, (uintptr_t)buff_hdr_addr, vq->vhost_hlen, 1);
-				mrg_count = 0;
-			}
-		}
+		rte_memcpy((void *)(uintptr_t)buff_hdr_addr,
+			(const void *)&virtio_hdr, vq->vhost_hlen);
+
+		PRINT_PACKET(dev, (uintptr_t)buff_hdr_addr, vq->vhost_hlen, 1);
+
 		if (res_cur_idx < res_end_idx) {
 			/* Prefetch descriptor index. */
 			rte_prefetch0(&vq->desc[head[packet_success]]);
@@ -1068,6 +1067,356 @@ virtio_dev_rx(struct virtio_net *dev, struct rte_mbuf **pkts, uint32_t count)
 	return count;
 }
 
+static inline uint32_t __attribute__((always_inline))
+copy_from_mbuf_to_vring(struct virtio_net *dev,
+	uint16_t res_base_idx, uint16_t res_end_idx,
+	struct rte_mbuf *pkt)
+{
+	uint32_t vec_idx = 0;
+	uint32_t entry_success = 0;
+	struct vhost_virtqueue *vq;
+	/* The virtio_hdr is initialised to 0. */
+	struct virtio_net_hdr_mrg_rxbuf virtio_hdr = {
+		{0, 0, 0, 0, 0, 0}, 0};
+	uint16_t cur_idx = res_base_idx;
+	uint64_t vb_addr = 0;
+	uint64_t vb_hdr_addr = 0;
+	uint32_t seg_offset = 0;
+	uint32_t vb_offset = 0;
+	uint32_t seg_avail;
+	uint32_t vb_avail;
+	uint32_t cpy_len, entry_len;
+
+	if (pkt == NULL)
+		return 0;
+
+	LOG_DEBUG(VHOST_DATA, "(%"PRIu64") Current Index %d| "
+		"End Index %d\n",
+		dev->device_fh, cur_idx, res_end_idx);
+
+	/*
+	 * Convert from gpa to vva
+	 * (guest physical addr -> vhost virtual addr)
+	 */
+	vq = dev->virtqueue[VIRTIO_RXQ];
+	vb_addr =
+		gpa_to_vva(dev, vq->buf_vec[vec_idx].buf_addr);
+	vb_hdr_addr = vb_addr;
+
+	/* Prefetch buffer address. */
+	rte_prefetch0((void *)(uintptr_t)vb_addr);
+
+	virtio_hdr.num_buffers = res_end_idx - res_base_idx;
+
+	LOG_DEBUG(VHOST_DATA, "(%"PRIu64") RX: Num merge buffers %d\n",
+		dev->device_fh, virtio_hdr.num_buffers);
+
+	rte_memcpy((void *)(uintptr_t)vb_hdr_addr,
+		(const void *)&virtio_hdr, vq->vhost_hlen);
+
+	PRINT_PACKET(dev, (uintptr_t)vb_hdr_addr, vq->vhost_hlen, 1);
+
+	seg_avail = rte_pktmbuf_data_len(pkt);
+	vb_offset = vq->vhost_hlen;
+	vb_avail =
+		vq->buf_vec[vec_idx].buf_len - vq->vhost_hlen;
+
+	entry_len = vq->vhost_hlen;
+
+	if (vb_avail == 0) {
+		uint32_t desc_idx =
+			vq->buf_vec[vec_idx].desc_idx;
+		vq->desc[desc_idx].len = vq->vhost_hlen;
+
+		if ((vq->desc[desc_idx].flags
+			& VRING_DESC_F_NEXT) == 0) {
+			/* Update used ring with desc information */
+			vq->used->ring[cur_idx & (vq->size - 1)].id
+				= vq->buf_vec[vec_idx].desc_idx;
+			vq->used->ring[cur_idx & (vq->size - 1)].len
+				= entry_len;
+
+			entry_len = 0;
+			cur_idx++;
+			entry_success++;
+		}
+
+		vec_idx++;
+		vb_addr =
+			gpa_to_vva(dev, vq->buf_vec[vec_idx].buf_addr);
+
+		/* Prefetch buffer address. */
+		rte_prefetch0((void *)(uintptr_t)vb_addr);
+		vb_offset = 0;
+		vb_avail = vq->buf_vec[vec_idx].buf_len;
+	}
+
+	cpy_len = RTE_MIN(vb_avail, seg_avail);
+
+	while (cpy_len > 0) {
+		/* Copy mbuf data to vring buffer */
+		rte_memcpy((void *)(uintptr_t)(vb_addr + vb_offset),
+			(const void *)(rte_pktmbuf_mtod(pkt, char*) + seg_offset),
+			cpy_len);
+
+		PRINT_PACKET(dev,
+			(uintptr_t)(vb_addr + vb_offset),
+			cpy_len, 0);
+
+		seg_offset += cpy_len;
+		vb_offset += cpy_len;
+		seg_avail -= cpy_len;
+		vb_avail -= cpy_len;
+		entry_len += cpy_len;
+
+		if (seg_avail != 0) {
+			/*
+			 * The virtio buffer in this vring
+			 * entry reach to its end.
+			 * But the segment doesn't complete.
+			 */
+			if ((vq->desc[vq->buf_vec[vec_idx].desc_idx].flags &
+				VRING_DESC_F_NEXT) == 0) {
+				/* Update used ring with desc information */
+				vq->used->ring[cur_idx & (vq->size - 1)].id
+					= vq->buf_vec[vec_idx].desc_idx;
+				vq->used->ring[cur_idx & (vq->size - 1)].len
+					= entry_len;
+				entry_len = 0;
+				cur_idx++;
+				entry_success++;
+			}
+
+			vec_idx++;
+			vb_addr = gpa_to_vva(dev,
+				vq->buf_vec[vec_idx].buf_addr);
+			vb_offset = 0;
+			vb_avail = vq->buf_vec[vec_idx].buf_len;
+			cpy_len = RTE_MIN(vb_avail, seg_avail);
+		} else {
+			/*
+			 * This current segment complete, need continue to
+			 * check if the whole packet complete or not.
+			 */
+			pkt = pkt->next;
+			if (pkt != NULL) {
+				/*
+				 * There are more segments.
+				 */
+				if (vb_avail == 0) {
+					/*
+					 * This current buffer from vring is
+					 * used up, need fetch next buffer
+					 * from buf_vec.
+					 */
+					uint32_t desc_idx =
+						vq->buf_vec[vec_idx].desc_idx;
+					vq->desc[desc_idx].len = vb_offset;
+
+					if ((vq->desc[desc_idx].flags &
+						VRING_DESC_F_NEXT) == 0) {
+						uint16_t wrapped_idx =
+							cur_idx & (vq->size - 1);
+						/*
+						 * Update used ring with the
+						 * descriptor information
+						 */
+						vq->used->ring[wrapped_idx].id
+							= desc_idx;
+						vq->used->ring[wrapped_idx].len
+							= entry_len;
+						entry_success++;
+						entry_len = 0;
+						cur_idx++;
+					}
+
+					/* Get next buffer from buf_vec. */
+					vec_idx++;
+					vb_addr = gpa_to_vva(dev,
+						vq->buf_vec[vec_idx].buf_addr);
+					vb_avail =
+						vq->buf_vec[vec_idx].buf_len;
+					vb_offset = 0;
+				}
+
+				seg_offset = 0;
+				seg_avail = rte_pktmbuf_data_len(pkt);
+				cpy_len = RTE_MIN(vb_avail, seg_avail);
+			} else {
+				/*
+				 * This whole packet completes.
+				 */
+				uint32_t desc_idx =
+					vq->buf_vec[vec_idx].desc_idx;
+				vq->desc[desc_idx].len = vb_offset;
+
+				while (vq->desc[desc_idx].flags &
+					VRING_DESC_F_NEXT) {
+					desc_idx = vq->desc[desc_idx].next;
+					 vq->desc[desc_idx].len = 0;
+				}
+
+				/* Update used ring with desc information */
+				vq->used->ring[cur_idx & (vq->size - 1)].id
+					= vq->buf_vec[vec_idx].desc_idx;
+				vq->used->ring[cur_idx & (vq->size - 1)].len
+					= entry_len;
+				entry_len = 0;
+				cur_idx++;
+				entry_success++;
+				seg_avail = 0;
+				cpy_len = RTE_MIN(vb_avail, seg_avail);
+			}
+		}
+	}
+
+	return entry_success;
+}
+
+/*
+ * This function adds buffers to the virtio devices RX virtqueue. Buffers can
+ * be received from the physical port or from another virtio device. A packet
+ * count is returned to indicate the number of packets that were succesfully
+ * added to the RX queue. This function works for mergeable RX.
+ */
+static inline uint32_t __attribute__((always_inline))
+virtio_dev_merge_rx(struct virtio_net *dev, struct rte_mbuf **pkts,
+	uint32_t count)
+{
+	struct vhost_virtqueue *vq;
+	uint32_t pkt_idx = 0, entry_success = 0;
+	uint32_t retry = 0;
+	uint16_t avail_idx, res_cur_idx;
+	uint16_t res_base_idx, res_end_idx;
+	uint8_t success = 0;
+
+	LOG_DEBUG(VHOST_DATA, "(%"PRIu64") virtio_dev_merge_rx()\n",
+		dev->device_fh);
+	vq = dev->virtqueue[VIRTIO_RXQ];
+	count = RTE_MIN((uint32_t)MAX_PKT_BURST, count);
+
+	if (count == 0)
+		return 0;
+
+	for (pkt_idx = 0; pkt_idx < count; pkt_idx++) {
+		uint32_t secure_len = 0;
+		uint16_t need_cnt;
+		uint32_t vec_idx = 0;
+		uint32_t pkt_len = pkts[pkt_idx]->pkt_len + vq->vhost_hlen;
+		uint16_t i, id;
+
+		do {
+			/*
+			 * As many data cores may want access to available
+			 * buffers, they need to be reserved.
+			 */
+			res_base_idx = vq->last_used_idx_res;
+			res_cur_idx = res_base_idx;
+
+			do {
+				avail_idx = *((volatile uint16_t *)&vq->avail->idx);
+				if (unlikely(res_cur_idx == avail_idx)) {
+					/*
+					 * If retry is enabled and the queue is
+					 * full then we wait and retry to avoid
+					 * packet loss.
+					 */
+					if (enable_retry) {
+						uint8_t cont = 0;
+						for (retry = 0; retry < burst_rx_retry_num; retry++) {
+							rte_delay_us(burst_rx_delay_time);
+							avail_idx =
+								*((volatile uint16_t *)&vq->avail->idx);
+							if (likely(res_cur_idx != avail_idx)) {
+								cont = 1;
+								break;
+							}
+						}
+						if (cont == 1)
+							continue;
+					}
+
+					LOG_DEBUG(VHOST_DATA,
+						"(%"PRIu64") Failed "
+						"to get enough desc from "
+						"vring\n",
+						dev->device_fh);
+					return pkt_idx;
+				} else {
+					uint16_t wrapped_idx =
+						(res_cur_idx) & (vq->size - 1);
+					uint32_t idx =
+						vq->avail->ring[wrapped_idx];
+					uint8_t next_desc;
+
+					do {
+						next_desc = 0;
+						secure_len += vq->desc[idx].len;
+						if (vq->desc[idx].flags &
+							VRING_DESC_F_NEXT) {
+							idx = vq->desc[idx].next;
+							next_desc = 1;
+						}
+					} while (next_desc);
+
+					res_cur_idx++;
+				}
+			} while (pkt_len > secure_len);
+
+			/* vq->last_used_idx_res is atomically updated. */
+			success = rte_atomic16_cmpset(&vq->last_used_idx_res,
+							res_base_idx,
+							res_cur_idx);
+		} while (success == 0);
+
+		id = res_base_idx;
+		need_cnt = res_cur_idx - res_base_idx;
+
+		for (i = 0; i < need_cnt; i++, id++) {
+			uint16_t wrapped_idx = id & (vq->size - 1);
+			uint32_t idx = vq->avail->ring[wrapped_idx];
+			uint8_t next_desc;
+			do {
+				next_desc = 0;
+				vq->buf_vec[vec_idx].buf_addr =
+					vq->desc[idx].addr;
+				vq->buf_vec[vec_idx].buf_len =
+					vq->desc[idx].len;
+				vq->buf_vec[vec_idx].desc_idx = idx;
+				vec_idx++;
+
+				if (vq->desc[idx].flags & VRING_DESC_F_NEXT) {
+					idx = vq->desc[idx].next;
+					next_desc = 1;
+				}
+			} while (next_desc);
+		}
+
+		res_end_idx = res_cur_idx;
+
+		entry_success = copy_from_mbuf_to_vring(dev, res_base_idx,
+			res_end_idx, pkts[pkt_idx]);
+
+		rte_compiler_barrier();
+
+		/*
+		 * Wait until it's our turn to add our buffer
+		 * to the used ring.
+		 */
+		while (unlikely(vq->last_used_idx != res_base_idx))
+			rte_pause();
+
+		*(volatile uint16_t *)&vq->used->idx += entry_success;
+		vq->last_used_idx = res_end_idx;
+
+		/* Kick the guest if necessary. */
+		if (!(vq->avail->flags & VRING_AVAIL_F_NO_INTERRUPT))
+			eventfd_write((int)vq->kickfd, 1);
+	}
+
+	return count;
+}
+
 /*
  * Compares a packet destination MAC address to a device MAC address.
  */
@@ -1089,7 +1438,7 @@ link_vmdq(struct virtio_net *dev, struct rte_mbuf *m)
 	int i, ret;
 
 	/* Learn MAC address of guest device from packet */
-	pkt_hdr = (struct ether_hdr *)m->pkt.data;
+	pkt_hdr = rte_pktmbuf_mtod(m, struct ether_hdr *);
 
 	dev_ll = ll_root_used;
 
@@ -1176,7 +1525,7 @@ virtio_tx_local(struct virtio_net *dev, struct rte_mbuf *m)
 	struct ether_hdr *pkt_hdr;
 	uint64_t ret = 0;
 
-	pkt_hdr = (struct ether_hdr *)m->pkt.data;
+	pkt_hdr = rte_pktmbuf_mtod(m, struct ether_hdr *);
 
 	/*get the used devices list*/
 	dev_ll = ll_root_used;
@@ -1199,8 +1548,17 @@ virtio_tx_local(struct virtio_net *dev, struct rte_mbuf *m)
 				/*drop the packet if the device is marked for removal*/
 				LOG_DEBUG(VHOST_DATA, "(%"PRIu64") Device is marked for removal\n", dev_ll->dev->device_fh);
 			} else {
+				uint32_t mergeable =
+					dev_ll->dev->features &
+					(1 << VIRTIO_NET_F_MRG_RXBUF);
+
 				/*send the packet to the local virtio device*/
-				ret = virtio_dev_rx(dev_ll->dev, &m, 1);
+				if (likely(mergeable == 0))
+					ret = virtio_dev_rx(dev_ll->dev, &m, 1);
+				else
+					ret = virtio_dev_merge_rx(dev_ll->dev,
+						&m, 1);
+
 				if (enable_stats) {
 					rte_atomic64_add(
 					&dev_statistics[dev_ll->dev->device_fh].rx_total_atomic,
@@ -1231,11 +1589,11 @@ virtio_tx_route(struct virtio_net* dev, struct rte_mbuf *m, struct rte_mempool *
 	struct mbuf_table *tx_q;
 	struct vlan_ethhdr *vlan_hdr;
 	struct rte_mbuf **m_table;
-	struct rte_mbuf *mbuf;
+	struct rte_mbuf *mbuf, *prev;
 	unsigned len, ret, offset = 0;
 	const uint16_t lcore_id = rte_lcore_id();
 	struct virtio_net_data_ll *dev_ll = ll_root_used;
-	struct ether_hdr *pkt_hdr = (struct ether_hdr *)m->pkt.data;
+	struct ether_hdr *pkt_hdr = rte_pktmbuf_mtod(m, struct ether_hdr *);
 
 	/*check if destination is local VM*/
 	if ((vm2vm_mode == VM2VM_SOFTWARE) && (virtio_tx_local(dev, m) == 0))
@@ -1284,26 +1642,54 @@ virtio_tx_route(struct virtio_net* dev, struct rte_mbuf *m, struct rte_mempool *
 	/* Allocate an mbuf and populate the structure. */
 	mbuf = rte_pktmbuf_alloc(mbuf_pool);
 	if (unlikely(mbuf == NULL)) {
-		RTE_LOG(ERR, VHOST_DATA, "Failed to allocate memory for mbuf.\n");
+		RTE_LOG(ERR, VHOST_DATA,
+			"Failed to allocate memory for mbuf.\n");
 		return;
 	}
 
-	mbuf->pkt.data_len = m->pkt.data_len + VLAN_HLEN + offset;
-	mbuf->pkt.pkt_len = mbuf->pkt.data_len;
+	mbuf->data_len = m->data_len + VLAN_HLEN + offset;
+	mbuf->pkt_len = m->pkt_len + VLAN_HLEN + offset;
+	mbuf->nb_segs = m->nb_segs;
 
 	/* Copy ethernet header to mbuf. */
-	rte_memcpy((void*)mbuf->pkt.data, (const void*)m->pkt.data, ETH_HLEN);
+	rte_memcpy(rte_pktmbuf_mtod(mbuf, void *),
+		rte_pktmbuf_mtod(m, const void *),
+		ETH_HLEN);
 
 
 	/* Setup vlan header. Bytes need to be re-ordered for network with htons()*/
-	vlan_hdr = (struct vlan_ethhdr *) mbuf->pkt.data;
+	vlan_hdr = rte_pktmbuf_mtod(mbuf, struct vlan_ethhdr *);
 	vlan_hdr->h_vlan_encapsulated_proto = vlan_hdr->h_vlan_proto;
 	vlan_hdr->h_vlan_proto = htons(ETH_P_8021Q);
 	vlan_hdr->h_vlan_TCI = htons(vlan_tag);
 
 	/* Copy the remaining packet contents to the mbuf. */
-	rte_memcpy((void*) ((uint8_t*)mbuf->pkt.data + VLAN_ETH_HLEN),
-		(const void*) ((uint8_t*)m->pkt.data + ETH_HLEN), (m->pkt.data_len - ETH_HLEN));
+	rte_memcpy((void *)(rte_pktmbuf_mtod(mbuf, uint8_t *) + VLAN_ETH_HLEN),
+		(const void *)(rte_pktmbuf_mtod(m, uint8_t *) + ETH_HLEN),
+		(m->data_len - ETH_HLEN));
+
+	/* Copy the remaining segments for the whole packet. */
+	prev = mbuf;
+	while (m->next) {
+		/* Allocate an mbuf and populate the structure. */
+		struct rte_mbuf *next_mbuf = rte_pktmbuf_alloc(mbuf_pool);
+		if (unlikely(next_mbuf == NULL)) {
+			rte_pktmbuf_free(mbuf);
+			RTE_LOG(ERR, VHOST_DATA,
+				"Failed to allocate memory for mbuf.\n");
+			return;
+		}
+
+		m = m->next;
+		prev->next = next_mbuf;
+		prev = next_mbuf;
+		next_mbuf->data_len = m->data_len;
+
+		/* Copy data to next mbuf. */
+		rte_memcpy(rte_pktmbuf_mtod(next_mbuf, void *),
+			rte_pktmbuf_mtod(m, const void *), m->data_len);
+	}
+
 	tx_q->m_table[len] = mbuf;
 	len++;
 	if (enable_stats) {
@@ -1393,8 +1779,9 @@ virtio_dev_tx(struct virtio_net* dev, struct rte_mempool *mbuf_pool)
 		vq->used->ring[used_idx].len = 0;
 
 		/* Setup dummy mbuf. This is copied to a real mbuf if transmitted out the physical port. */
-		m.pkt.data_len = desc->len;
-		m.pkt.data = (void*)(uintptr_t)buff_addr;
+		m.data_len = desc->len;
+		m.pkt_len = desc->len;
+		m.data_off = 0;
 
 		PRINT_PACKET(dev, (uintptr_t)buff_addr, desc->len, 0);
 
@@ -1420,6 +1807,227 @@ virtio_dev_tx(struct virtio_net* dev, struct rte_mempool *mbuf_pool)
 		eventfd_write((int)vq->kickfd, 1);
 }
 
+/* This function works for TX packets with mergeable feature enabled. */
+static inline void __attribute__((always_inline))
+virtio_dev_merge_tx(struct virtio_net *dev, struct rte_mempool *mbuf_pool)
+{
+	struct rte_mbuf *m, *prev;
+	struct vhost_virtqueue *vq;
+	struct vring_desc *desc;
+	uint64_t vb_addr = 0;
+	uint32_t head[MAX_PKT_BURST];
+	uint32_t used_idx;
+	uint32_t i;
+	uint16_t free_entries, entry_success = 0;
+	uint16_t avail_idx;
+	uint32_t buf_size = MBUF_SIZE - (sizeof(struct rte_mbuf)
+			+ RTE_PKTMBUF_HEADROOM);
+
+	vq = dev->virtqueue[VIRTIO_TXQ];
+	avail_idx =  *((volatile uint16_t *)&vq->avail->idx);
+
+	/* If there are no available buffers then return. */
+	if (vq->last_used_idx == avail_idx)
+		return;
+
+	LOG_DEBUG(VHOST_DATA, "(%"PRIu64") virtio_dev_merge_tx()\n",
+		dev->device_fh);
+
+	/* Prefetch available ring to retrieve head indexes. */
+	rte_prefetch0(&vq->avail->ring[vq->last_used_idx & (vq->size - 1)]);
+
+	/*get the number of free entries in the ring*/
+	free_entries = (avail_idx - vq->last_used_idx);
+
+	/* Limit to MAX_PKT_BURST. */
+	free_entries = RTE_MIN(free_entries, MAX_PKT_BURST);
+
+	LOG_DEBUG(VHOST_DATA, "(%"PRIu64") Buffers available %d\n",
+		dev->device_fh, free_entries);
+	/* Retrieve all of the head indexes first to avoid caching issues. */
+	for (i = 0; i < free_entries; i++)
+		head[i] = vq->avail->ring[(vq->last_used_idx + i) & (vq->size - 1)];
+
+	/* Prefetch descriptor index. */
+	rte_prefetch0(&vq->desc[head[entry_success]]);
+	rte_prefetch0(&vq->used->ring[vq->last_used_idx & (vq->size - 1)]);
+
+	while (entry_success < free_entries) {
+		uint32_t vb_avail, vb_offset;
+		uint32_t seg_avail, seg_offset;
+		uint32_t cpy_len;
+		uint32_t seg_num = 0;
+		struct rte_mbuf *cur;
+		uint8_t alloc_err = 0;
+
+		desc = &vq->desc[head[entry_success]];
+
+		/* Discard first buffer as it is the virtio header */
+		desc = &vq->desc[desc->next];
+
+		/* Buffer address translation. */
+		vb_addr = gpa_to_vva(dev, desc->addr);
+		/* Prefetch buffer address. */
+		rte_prefetch0((void *)(uintptr_t)vb_addr);
+
+		used_idx = vq->last_used_idx & (vq->size - 1);
+
+		if (entry_success < (free_entries - 1)) {
+			/* Prefetch descriptor index. */
+			rte_prefetch0(&vq->desc[head[entry_success+1]]);
+			rte_prefetch0(&vq->used->ring[(used_idx + 1) & (vq->size - 1)]);
+		}
+
+		/* Update used index buffer information. */
+		vq->used->ring[used_idx].id = head[entry_success];
+		vq->used->ring[used_idx].len = 0;
+
+		vb_offset = 0;
+		vb_avail = desc->len;
+		seg_offset = 0;
+		seg_avail = buf_size;
+		cpy_len = RTE_MIN(vb_avail, seg_avail);
+
+		PRINT_PACKET(dev, (uintptr_t)vb_addr, desc->len, 0);
+
+		/* Allocate an mbuf and populate the structure. */
+		m = rte_pktmbuf_alloc(mbuf_pool);
+		if (unlikely(m == NULL)) {
+			RTE_LOG(ERR, VHOST_DATA,
+				"Failed to allocate memory for mbuf.\n");
+			return;
+		}
+
+		seg_num++;
+		cur = m;
+		prev = m;
+		while (cpy_len != 0) {
+			rte_memcpy((void *)(rte_pktmbuf_mtod(cur, char *) + seg_offset),
+				(void *)((uintptr_t)(vb_addr + vb_offset)),
+				cpy_len);
+
+			seg_offset += cpy_len;
+			vb_offset += cpy_len;
+			vb_avail -= cpy_len;
+			seg_avail -= cpy_len;
+
+			if (vb_avail != 0) {
+				/*
+				 * The segment reachs to its end,
+				 * while the virtio buffer in TX vring has
+				 * more data to be copied.
+				 */
+				cur->data_len = seg_offset;
+				m->pkt_len += seg_offset;
+				/* Allocate mbuf and populate the structure. */
+				cur = rte_pktmbuf_alloc(mbuf_pool);
+				if (unlikely(cur == NULL)) {
+					RTE_LOG(ERR, VHOST_DATA, "Failed to "
+						"allocate memory for mbuf.\n");
+					rte_pktmbuf_free(m);
+					alloc_err = 1;
+					break;
+				}
+
+				seg_num++;
+				prev->next = cur;
+				prev = cur;
+				seg_offset = 0;
+				seg_avail = buf_size;
+			} else {
+				if (desc->flags & VRING_DESC_F_NEXT) {
+					/*
+					 * There are more virtio buffers in
+					 * same vring entry need to be copied.
+					 */
+					if (seg_avail == 0) {
+						/*
+						 * The current segment hasn't
+						 * room to accomodate more
+						 * data.
+						 */
+						cur->data_len = seg_offset;
+						m->pkt_len += seg_offset;
+						/*
+						 * Allocate an mbuf and
+						 * populate the structure.
+						 */
+						cur = rte_pktmbuf_alloc(mbuf_pool);
+						if (unlikely(cur == NULL)) {
+							RTE_LOG(ERR,
+								VHOST_DATA,
+								"Failed to "
+								"allocate memory "
+								"for mbuf\n");
+							rte_pktmbuf_free(m);
+							alloc_err = 1;
+							break;
+						}
+						seg_num++;
+						prev->next = cur;
+						prev = cur;
+						seg_offset = 0;
+						seg_avail = buf_size;
+					}
+
+					desc = &vq->desc[desc->next];
+
+					/* Buffer address translation. */
+					vb_addr = gpa_to_vva(dev, desc->addr);
+					/* Prefetch buffer address. */
+					rte_prefetch0((void *)(uintptr_t)vb_addr);
+					vb_offset = 0;
+					vb_avail = desc->len;
+
+					PRINT_PACKET(dev, (uintptr_t)vb_addr,
+						desc->len, 0);
+				} else {
+					/* The whole packet completes. */
+					cur->data_len = seg_offset;
+					m->pkt_len += seg_offset;
+					vb_avail = 0;
+				}
+			}
+
+			cpy_len = RTE_MIN(vb_avail, seg_avail);
+		}
+
+		if (unlikely(alloc_err == 1))
+			break;
+
+		m->nb_segs = seg_num;
+
+		/*
+		 * If this is the first received packet we need to learn
+		 * the MAC and setup VMDQ
+		 */
+		if (dev->ready == DEVICE_MAC_LEARNING) {
+			if (dev->remove || (link_vmdq(dev, m) == -1)) {
+				/*
+				 * Discard frame if device is scheduled for
+				 * removal or a duplicate MAC address is found.
+				 */
+				entry_success = free_entries;
+				vq->last_used_idx += entry_success;
+				rte_pktmbuf_free(m);
+				break;
+			}
+		}
+
+		virtio_tx_route(dev, m, mbuf_pool, (uint16_t)dev->device_fh);
+		vq->last_used_idx++;
+		entry_success++;
+		rte_pktmbuf_free(m);
+	}
+
+	rte_compiler_barrier();
+	vq->used->idx += entry_success;
+	/* Kick guest if required. */
+	if (!(vq->avail->flags & VRING_AVAIL_F_NO_INTERRUPT))
+		eventfd_write((int)vq->kickfd, 1);
+
+}
+
 /*
  * This function is called by each data core. It handles all RX/TX registered with the
  * core. For TX the specific lcore linked list is used. For RX, MAC addresses are compared
@@ -1440,8 +2048,9 @@ switch_worker(__attribute__((unused)) void *arg)
 	const uint16_t lcore_id = rte_lcore_id();
 	const uint16_t num_cores = (uint16_t)rte_lcore_count();
 	uint16_t rx_count = 0;
+	uint32_t mergeable = 0;
 
-	RTE_LOG(INFO, VHOST_DATA, "Procesing on Core %u started \n", lcore_id);
+	RTE_LOG(INFO, VHOST_DATA, "Procesing on Core %u started\n", lcore_id);
 	lcore_ll = lcore_info[lcore_id].lcore_ll;
 	prev_tsc = 0;
 
@@ -1497,6 +2106,8 @@ switch_worker(__attribute__((unused)) void *arg)
 		while (dev_ll != NULL) {
 			/*get virtio device ID*/
 			dev = dev_ll->dev;
+			mergeable =
+				dev->features & (1 << VIRTIO_NET_F_MRG_RXBUF);
 
 			if (dev->remove) {
 				dev_ll = dev_ll->next;
@@ -1510,7 +2121,15 @@ switch_worker(__attribute__((unused)) void *arg)
 					(uint16_t)dev->vmdq_rx_q, pkts_burst, MAX_PKT_BURST);
 
 				if (rx_count) {
-					ret_count = virtio_dev_rx(dev, pkts_burst, rx_count);
+					if (likely(mergeable == 0))
+						ret_count =
+							virtio_dev_rx(dev,
+							pkts_burst, rx_count);
+					else
+						ret_count =
+							virtio_dev_merge_rx(dev,
+							pkts_burst, rx_count);
+
 					if (enable_stats) {
 						rte_atomic64_add(
 						&dev_statistics[dev_ll->dev->device_fh].rx_total_atomic,
@@ -1520,15 +2139,19 @@ switch_worker(__attribute__((unused)) void *arg)
 					}
 					while (likely(rx_count)) {
 						rx_count--;
-						rte_pktmbuf_free_seg(pkts_burst[rx_count]);
+						rte_pktmbuf_free(pkts_burst[rx_count]);
 					}
 
 				}
 			}
 
-			if (!dev->remove)
+			if (!dev->remove) {
 				/*Handle guest TX*/
-				virtio_dev_tx(dev, mbuf_pool);
+				if (likely(mergeable == 0))
+					virtio_dev_tx(dev, mbuf_pool);
+				else
+					virtio_dev_merge_tx(dev, mbuf_pool);
+			}
 
 			/*move to the next device in the list*/
 			dev_ll = dev_ll->next;
@@ -1713,9 +2336,9 @@ attach_rxmbuf_zcp(struct virtio_net *dev)
 	}
 
 	mbuf->buf_addr = (void *)(uintptr_t)(buff_addr - RTE_PKTMBUF_HEADROOM);
-	mbuf->pkt.data = (void *)(uintptr_t)(buff_addr);
+	mbuf->data_off = RTE_PKTMBUF_HEADROOM;
 	mbuf->buf_physaddr = phys_addr - RTE_PKTMBUF_HEADROOM;
-	mbuf->pkt.data_len = desc->len;
+	mbuf->data_len = desc->len;
 	MBUF_HEADROOM_UINT32(mbuf) = (uint32_t)desc_idx;
 
 	LOG_DEBUG(VHOST_DATA,
@@ -1750,9 +2373,9 @@ static inline void pktmbuf_detach_zcp(struct rte_mbuf *m)
 
 	buf_ofs = (RTE_PKTMBUF_HEADROOM <= m->buf_len) ?
 			RTE_PKTMBUF_HEADROOM : m->buf_len;
-	m->pkt.data = (char *) m->buf_addr + buf_ofs;
+	m->data_off = buf_ofs;
 
-	m->pkt.data_len = 0;
+	m->data_len = 0;
 }
 
 /*
@@ -1984,7 +2607,7 @@ virtio_tx_route_zcp(struct virtio_net *dev, struct rte_mbuf *m,
 	unsigned len, ret, offset = 0;
 	struct vpool *vpool;
 	struct virtio_net_data_ll *dev_ll = ll_root_used;
-	struct ether_hdr *pkt_hdr = (struct ether_hdr *)m->pkt.data;
+	struct ether_hdr *pkt_hdr = rte_pktmbuf_mtod(m, struct ether_hdr *);
 	uint16_t vlan_tag = (uint16_t)vlan_tags[(uint16_t)dev->device_fh];
 
 	/*Add packet to the port tx queue*/
@@ -2055,24 +2678,24 @@ virtio_tx_route_zcp(struct virtio_net *dev, struct rte_mbuf *m,
 		}
 	}
 
-	mbuf->pkt.nb_segs = m->pkt.nb_segs;
-	mbuf->pkt.next = m->pkt.next;
-	mbuf->pkt.data_len = m->pkt.data_len + offset;
-	mbuf->pkt.pkt_len = mbuf->pkt.data_len;
+	mbuf->nb_segs = m->nb_segs;
+	mbuf->next = m->next;
+	mbuf->data_len = m->data_len + offset;
+	mbuf->pkt_len = mbuf->data_len;
 	if (unlikely(need_copy)) {
 		/* Copy the packet contents to the mbuf. */
-		rte_memcpy((void *)((uint8_t *)mbuf->pkt.data),
-			(const void *) ((uint8_t *)m->pkt.data),
-			m->pkt.data_len);
+		rte_memcpy(rte_pktmbuf_mtod(mbuf, void *),
+			rte_pktmbuf_mtod(m, void *),
+			m->data_len);
 	} else {
-		mbuf->pkt.data = m->pkt.data;
+		mbuf->data_off = m->data_off;
 		mbuf->buf_physaddr = m->buf_physaddr;
 		mbuf->buf_addr = m->buf_addr;
 	}
 	mbuf->ol_flags = PKT_TX_VLAN_PKT;
-	mbuf->pkt.vlan_macip.f.vlan_tci = vlan_tag;
-	mbuf->pkt.vlan_macip.f.l2_len = sizeof(struct ether_hdr);
-	mbuf->pkt.vlan_macip.f.l3_len = sizeof(struct ipv4_hdr);
+	mbuf->vlan_tci = vlan_tag;
+	mbuf->l2_len = sizeof(struct ether_hdr);
+	mbuf->l3_len = sizeof(struct ipv4_hdr);
 	MBUF_HEADROOM_UINT32(mbuf) = (uint32_t)desc_idx;
 
 	tx_q->m_table[len] = mbuf;
@@ -2081,8 +2704,8 @@ virtio_tx_route_zcp(struct virtio_net *dev, struct rte_mbuf *m,
 	LOG_DEBUG(VHOST_DATA,
 		"(%"PRIu64") in tx_route_zcp: pkt: nb_seg: %d, next:%s\n",
 		dev->device_fh,
-		mbuf->pkt.nb_segs,
-		(mbuf->pkt.next == NULL) ? "null" : "non-null");
+		mbuf->nb_segs,
+		(mbuf->next == NULL) ? "null" : "non-null");
 
 	if (enable_stats) {
 		dev_statistics[dev->device_fh].tx_total++;
@@ -2196,11 +2819,11 @@ virtio_dev_tx_zcp(struct virtio_net *dev)
 		 * Setup dummy mbuf. This is copied to a real mbuf if
 		 * transmitted out the physical port.
 		 */
-		m.pkt.data_len = desc->len;
-		m.pkt.nb_segs = 1;
-		m.pkt.next = NULL;
-		m.pkt.data = (void *)(uintptr_t)buff_addr;
-		m.buf_addr = m.pkt.data;
+		m.data_len = desc->len;
+		m.nb_segs = 1;
+		m.next = NULL;
+		m.data_off = 0;
+		m.buf_addr = (void *)(uintptr_t)buff_addr;
 		m.buf_physaddr = phys_addr;
 
 		/*
@@ -2992,9 +3615,9 @@ MAIN(int argc, char *argv[])
 			+ num_switching_cores * MAX_PKT_BURST;
 
 		for (queue_id = 0; queue_id < MAX_QUEUES; queue_id++) {
-			rte_snprintf(pool_name, sizeof(pool_name),
+			snprintf(pool_name, sizeof(pool_name),
 				"rxmbuf_pool_%u", queue_id);
-			rte_snprintf(ring_name, sizeof(ring_name),
+			snprintf(ring_name, sizeof(ring_name),
 				"rxmbuf_ring_%u", queue_id);
 			setup_mempool_tbl(rte_socket_id(), queue_id,
 				pool_name, ring_name, nb_mbuf);
@@ -3005,9 +3628,9 @@ MAIN(int argc, char *argv[])
 				+ num_switching_cores * MAX_PKT_BURST;
 
 		for (queue_id = 0; queue_id < MAX_QUEUES; queue_id++) {
-			rte_snprintf(pool_name, sizeof(pool_name),
+			snprintf(pool_name, sizeof(pool_name),
 				"txmbuf_pool_%u", queue_id);
-			rte_snprintf(ring_name, sizeof(ring_name),
+			snprintf(ring_name, sizeof(ring_name),
 				"txmbuf_ring_%u", queue_id);
 			setup_mempool_tbl(rte_socket_id(),
 				(queue_id + MAX_QUEUES),