summaryrefslogtreecommitdiffstats
path: root/src/spdk/dpdk/examples/tep_termination
diff options
context:
space:
mode:
Diffstat (limited to 'src/spdk/dpdk/examples/tep_termination')
-rw-r--r--src/spdk/dpdk/examples/tep_termination/Makefile66
-rw-r--r--src/spdk/dpdk/examples/tep_termination/main.c1244
-rw-r--r--src/spdk/dpdk/examples/tep_termination/main.h93
-rw-r--r--src/spdk/dpdk/examples/tep_termination/meson.build16
-rw-r--r--src/spdk/dpdk/examples/tep_termination/vxlan.c236
-rw-r--r--src/spdk/dpdk/examples/tep_termination/vxlan.h57
-rw-r--r--src/spdk/dpdk/examples/tep_termination/vxlan_setup.c440
-rw-r--r--src/spdk/dpdk/examples/tep_termination/vxlan_setup.h58
8 files changed, 2210 insertions, 0 deletions
diff --git a/src/spdk/dpdk/examples/tep_termination/Makefile b/src/spdk/dpdk/examples/tep_termination/Makefile
new file mode 100644
index 00000000..8ec1a38e
--- /dev/null
+++ b/src/spdk/dpdk/examples/tep_termination/Makefile
@@ -0,0 +1,66 @@
+# SPDX-License-Identifier: BSD-3-Clause
+# Copyright(c) 2010-2015 Intel Corporation
+
+# binary name
+APP = tep_termination
+
+# all source are stored in SRCS-y
+SRCS-y := main.c vxlan_setup.c vxlan.c
+
+# Build using pkg-config variables if possible
+$(shell pkg-config --exists libdpdk)
+ifeq ($(.SHELLSTATUS),0)
+
+all: shared
+.PHONY: shared static
+shared: build/$(APP)-shared
+ ln -sf $(APP)-shared build/$(APP)
+static: build/$(APP)-static
+ ln -sf $(APP)-static build/$(APP)
+
+LDFLAGS += -pthread
+
+PC_FILE := $(shell pkg-config --path libdpdk)
+CFLAGS += -O3 $(shell pkg-config --cflags libdpdk)
+LDFLAGS_SHARED = $(shell pkg-config --libs libdpdk)
+LDFLAGS_STATIC = -Wl,-Bstatic $(shell pkg-config --static --libs libdpdk)
+
+CFLAGS += -DALLOW_EXPERIMENTAL_API
+
+build/$(APP)-shared: $(SRCS-y) Makefile $(PC_FILE) | build
+ $(CC) $(CFLAGS) $(SRCS-y) -o $@ $(LDFLAGS) $(LDFLAGS_SHARED)
+
+build/$(APP)-static: $(SRCS-y) Makefile $(PC_FILE) | build
+ $(CC) $(CFLAGS) $(SRCS-y) -o $@ $(LDFLAGS) $(LDFLAGS_STATIC)
+
+build:
+ @mkdir -p $@
+
+.PHONY: clean
+clean:
+ rm -f build/$(APP) build/$(APP)-static build/$(APP)-shared
+ rmdir --ignore-fail-on-non-empty build
+
+else # Build using legacy build system
+
+ifeq ($(RTE_SDK),)
+$(error "Please define RTE_SDK environment variable")
+endif
+
+# Default target, can be overridden by command line or environment
+RTE_TARGET ?= x86_64-native-linuxapp-gcc
+
+include $(RTE_SDK)/mk/rte.vars.mk
+
+ifneq ($(CONFIG_RTE_EXEC_ENV_LINUXAPP),y)
+$(error This application can only operate in a linuxapp environment, \
+please change the definition of the RTE_TARGET environment variable)
+endif
+
+CFLAGS += -DALLOW_EXPERIMENTAL_API
+CFLAGS += -O3
+CFLAGS += $(WERROR_FLAGS)
+CFLAGS += -D_GNU_SOURCE
+
+include $(RTE_SDK)/mk/rte.extapp.mk
+endif
diff --git a/src/spdk/dpdk/examples/tep_termination/main.c b/src/spdk/dpdk/examples/tep_termination/main.c
new file mode 100644
index 00000000..7795d089
--- /dev/null
+++ b/src/spdk/dpdk/examples/tep_termination/main.c
@@ -0,0 +1,1244 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(c) 2010-2015 Intel Corporation
+ */
+
+#include <arpa/inet.h>
+#include <getopt.h>
+#include <linux/if_ether.h>
+#include <linux/if_vlan.h>
+#include <linux/virtio_net.h>
+#include <linux/virtio_ring.h>
+#include <signal.h>
+#include <stdint.h>
+#include <sys/eventfd.h>
+#include <sys/param.h>
+#include <unistd.h>
+
+#include <rte_atomic.h>
+#include <rte_cycles.h>
+#include <rte_ethdev.h>
+#include <rte_log.h>
+#include <rte_string_fns.h>
+#include <rte_malloc.h>
+#include <rte_vhost.h>
+#include <rte_pause.h>
+
+#include "main.h"
+#include "vxlan.h"
+#include "vxlan_setup.h"
+
+/* the maximum number of external ports supported */
+#define MAX_SUP_PORTS 1
+
+/**
+ * Calculate the number of buffers needed per port
+ */
+#define NUM_MBUFS_PER_PORT ((MAX_QUEUES * RTE_TEST_RX_DESC_DEFAULT) +\
+ (nb_switching_cores * MAX_PKT_BURST) +\
+ (nb_switching_cores * \
+ RTE_TEST_TX_DESC_DEFAULT) +\
+ (nb_switching_cores * MBUF_CACHE_SIZE))
+
+#define MBUF_CACHE_SIZE 128
+#define MBUF_DATA_SIZE RTE_MBUF_DEFAULT_BUF_SIZE
+
+#define MAX_PKT_BURST 32 /* Max burst size for RX/TX */
+#define BURST_TX_DRAIN_US 100 /* TX drain every ~100us */
+
+/* Defines how long we wait between retries on RX */
+#define BURST_RX_WAIT_US 15
+
+#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
+#define DEVICE_SAFE_REMOVE 2
+
+/* Config_core_flag status definitions. */
+#define REQUEST_DEV_REMOVAL 1
+#define ACK_DEV_REMOVAL 0
+
+/* Configurable number of RX/TX ring descriptors */
+#define RTE_TEST_RX_DESC_DEFAULT 1024
+#define RTE_TEST_TX_DESC_DEFAULT 512
+
+/* Get first 4 bytes in mbuf headroom. */
+#define MBUF_HEADROOM_UINT32(mbuf) (*(uint32_t *)((uint8_t *)(mbuf) \
+ + sizeof(struct rte_mbuf)))
+
+#define INVALID_PORT_ID 0xFFFF
+
+/* Size of buffers used for snprintfs. */
+#define MAX_PRINT_BUFF 6072
+
+/* Maximum character device basename size. */
+#define MAX_BASENAME_SZ 20
+
+/* Maximum long option length for option parsing. */
+#define MAX_LONG_OPT_SZ 64
+
+/* Used to compare MAC addresses. */
+#define MAC_ADDR_CMP 0xFFFFFFFFFFFFULL
+
+#define CMD_LINE_OPT_NB_DEVICES "nb-devices"
+#define CMD_LINE_OPT_UDP_PORT "udp-port"
+#define CMD_LINE_OPT_TX_CHECKSUM "tx-checksum"
+#define CMD_LINE_OPT_TSO_SEGSZ "tso-segsz"
+#define CMD_LINE_OPT_FILTER_TYPE "filter-type"
+#define CMD_LINE_OPT_ENCAP "encap"
+#define CMD_LINE_OPT_DECAP "decap"
+#define CMD_LINE_OPT_RX_RETRY "rx-retry"
+#define CMD_LINE_OPT_RX_RETRY_DELAY "rx-retry-delay"
+#define CMD_LINE_OPT_RX_RETRY_NUM "rx-retry-num"
+#define CMD_LINE_OPT_STATS "stats"
+#define CMD_LINE_OPT_DEV_BASENAME "dev-basename"
+
+/* mask of enabled ports */
+static uint32_t enabled_port_mask;
+
+/*Number of switching cores enabled*/
+static uint32_t nb_switching_cores;
+
+/* number of devices/queues to support*/
+uint16_t nb_devices = 2;
+
+/* max ring descriptor, ixgbe, i40e, e1000 all are 4096. */
+#define MAX_RING_DESC 4096
+
+struct vpool {
+ struct rte_mempool *pool;
+ struct rte_ring *ring;
+ uint32_t buf_size;
+} vpool_array[MAX_QUEUES+MAX_QUEUES];
+
+/* UDP tunneling port */
+uint16_t udp_port = 4789;
+
+/* enable/disable inner TX checksum */
+uint8_t tx_checksum = 0;
+
+/* TCP segment size */
+uint16_t tso_segsz = 0;
+
+/* enable/disable decapsulation */
+uint8_t rx_decap = 1;
+
+/* enable/disable encapsulation */
+uint8_t tx_encap = 1;
+
+/* RX filter type for tunneling packet */
+uint8_t filter_idx = 1;
+
+/* overlay packet operation */
+struct ol_switch_ops overlay_options = {
+ .port_configure = vxlan_port_init,
+ .tunnel_setup = vxlan_link,
+ .tunnel_destroy = vxlan_unlink,
+ .tx_handle = vxlan_tx_pkts,
+ .rx_handle = vxlan_rx_pkts,
+ .param_handle = NULL,
+};
+
+/* Enable stats. */
+uint32_t enable_stats = 0;
+/* Enable retries on RX. */
+static uint32_t enable_retry = 1;
+/* Specify timeout (in useconds) between retries on RX. */
+static uint32_t burst_rx_delay_time = BURST_RX_WAIT_US;
+/* Specify the number of retries on RX. */
+static uint32_t burst_rx_retry_num = BURST_RX_RETRIES;
+
+/* Character device basename. Can be set by user. */
+static char dev_basename[MAX_BASENAME_SZ] = "vhost-net";
+
+static unsigned lcore_ids[RTE_MAX_LCORE];
+uint16_t ports[RTE_MAX_ETHPORTS];
+
+static unsigned nb_ports; /**< The number of ports specified in command line */
+
+/* ethernet addresses of ports */
+struct ether_addr ports_eth_addr[RTE_MAX_ETHPORTS];
+
+/* heads for the main used and free linked lists for the data path. */
+static struct virtio_net_data_ll *ll_root_used;
+static struct virtio_net_data_ll *ll_root_free;
+
+/**
+ * Array of data core structures containing information on
+ * individual core linked lists.
+ */
+static struct lcore_info lcore_info[RTE_MAX_LCORE];
+
+/* Used for queueing bursts of TX packets. */
+struct mbuf_table {
+ unsigned len;
+ unsigned txq_id;
+ struct rte_mbuf *m_table[MAX_PKT_BURST];
+};
+
+/* TX queue for each data core. */
+struct mbuf_table lcore_tx_queue[RTE_MAX_LCORE];
+
+struct device_statistics dev_statistics[MAX_DEVICES];
+
+/**
+ * Set character device basename.
+ */
+static int
+us_vhost_parse_basename(const char *q_arg)
+{
+ /* parse number string */
+ if (strlen(q_arg) >= MAX_BASENAME_SZ)
+ return -1;
+ else
+ snprintf((char *)&dev_basename, MAX_BASENAME_SZ, "%s", q_arg);
+
+ return 0;
+}
+
+/**
+ * Parse the portmask provided at run time.
+ */
+static int
+parse_portmask(const char *portmask)
+{
+ char *end = NULL;
+ unsigned long pm;
+
+ /* parse hexadecimal string */
+ pm = strtoul(portmask, &end, 16);
+ if ((portmask[0] == '\0') || (end == NULL) || (*end != '\0'))
+ return -1;
+
+ if (pm == 0)
+ return -1;
+
+ return pm;
+}
+
+/**
+ * Parse num options at run time.
+ */
+static int
+parse_num_opt(const char *q_arg, uint32_t max_valid_value)
+{
+ char *end = NULL;
+ unsigned long num;
+
+ /* parse unsigned int string */
+ num = strtoul(q_arg, &end, 10);
+ if ((q_arg[0] == '\0') || (end == NULL) || (*end != '\0'))
+ return -1;
+
+ if (num > max_valid_value)
+ return -1;
+
+ return num;
+}
+
+/**
+ * Display usage
+ */
+static void
+tep_termination_usage(const char *prgname)
+{
+ RTE_LOG(INFO, VHOST_CONFIG, "%s [EAL options] -- -p PORTMASK\n"
+ " --udp-port: UDP destination port for VXLAN packet\n"
+ " --nb-devices[1-64]: The number of virtIO device\n"
+ " --tx-checksum [0|1]: inner Tx checksum offload\n"
+ " --tso-segsz [0-N]: TCP segment size\n"
+ " --decap [0|1]: tunneling packet decapsulation\n"
+ " --encap [0|1]: tunneling packet encapsulation\n"
+ " --filter-type[1-3]: filter type for tunneling packet\n"
+ " 1: Inner MAC and tenent ID\n"
+ " 2: Inner MAC and VLAN, and tenent ID\n"
+ " 3: Outer MAC, Inner MAC and tenent ID\n"
+ " -p PORTMASK: Set mask for ports to be used by application\n"
+ " --rx-retry [0|1]: disable/enable(default) retries on rx."
+ " Enable retry if destintation queue is full\n"
+ " --rx-retry-delay [0-N]: timeout(in usecond) between retries on RX."
+ " This makes effect only if retries on rx enabled\n"
+ " --rx-retry-num [0-N]: the number of retries on rx."
+ " This makes effect only if retries on rx enabled\n"
+ " --stats [0-N]: 0: Disable stats, N: Time in seconds to print stats\n"
+ " --dev-basename: The basename to be used for the character device.\n",
+ prgname);
+}
+
+/**
+ * Parse the arguments given in the command line of the application.
+ */
+static int
+tep_termination_parse_args(int argc, char **argv)
+{
+ int opt, ret;
+ int option_index;
+ unsigned i;
+ const char *prgname = argv[0];
+ static struct option long_option[] = {
+ {CMD_LINE_OPT_NB_DEVICES, required_argument, NULL, 0},
+ {CMD_LINE_OPT_UDP_PORT, required_argument, NULL, 0},
+ {CMD_LINE_OPT_TX_CHECKSUM, required_argument, NULL, 0},
+ {CMD_LINE_OPT_TSO_SEGSZ, required_argument, NULL, 0},
+ {CMD_LINE_OPT_DECAP, required_argument, NULL, 0},
+ {CMD_LINE_OPT_ENCAP, required_argument, NULL, 0},
+ {CMD_LINE_OPT_FILTER_TYPE, required_argument, NULL, 0},
+ {CMD_LINE_OPT_RX_RETRY, required_argument, NULL, 0},
+ {CMD_LINE_OPT_RX_RETRY_DELAY, required_argument, NULL, 0},
+ {CMD_LINE_OPT_RX_RETRY_NUM, required_argument, NULL, 0},
+ {CMD_LINE_OPT_STATS, required_argument, NULL, 0},
+ {CMD_LINE_OPT_DEV_BASENAME, required_argument, NULL, 0},
+ {NULL, 0, 0, 0},
+ };
+
+ /* Parse command line */
+ while ((opt = getopt_long(argc, argv, "p:",
+ long_option, &option_index)) != EOF) {
+ switch (opt) {
+ /* Portmask */
+ case 'p':
+ enabled_port_mask = parse_portmask(optarg);
+ if (enabled_port_mask == 0) {
+ RTE_LOG(INFO, VHOST_CONFIG,
+ "Invalid portmask\n");
+ tep_termination_usage(prgname);
+ return -1;
+ }
+ break;
+ case 0:
+ if (!strncmp(long_option[option_index].name,
+ CMD_LINE_OPT_NB_DEVICES,
+ sizeof(CMD_LINE_OPT_NB_DEVICES))) {
+ ret = parse_num_opt(optarg, MAX_DEVICES);
+ if (ret == -1) {
+ RTE_LOG(INFO, VHOST_CONFIG,
+ "Invalid argument for nb-devices [0-%d]\n",
+ MAX_DEVICES);
+ tep_termination_usage(prgname);
+ return -1;
+ } else
+ nb_devices = ret;
+ }
+
+ /* Enable/disable retries on RX. */
+ if (!strncmp(long_option[option_index].name,
+ CMD_LINE_OPT_RX_RETRY,
+ sizeof(CMD_LINE_OPT_RX_RETRY))) {
+ ret = parse_num_opt(optarg, 1);
+ if (ret == -1) {
+ RTE_LOG(INFO, VHOST_CONFIG,
+ "Invalid argument for rx-retry [0|1]\n");
+ tep_termination_usage(prgname);
+ return -1;
+ } else
+ enable_retry = ret;
+ }
+
+ if (!strncmp(long_option[option_index].name,
+ CMD_LINE_OPT_TSO_SEGSZ,
+ sizeof(CMD_LINE_OPT_TSO_SEGSZ))) {
+ ret = parse_num_opt(optarg, INT16_MAX);
+ if (ret == -1) {
+ RTE_LOG(INFO, VHOST_CONFIG,
+ "Invalid argument for TCP segment size [0-N]\n");
+ tep_termination_usage(prgname);
+ return -1;
+ } else
+ tso_segsz = ret;
+ }
+
+ if (!strncmp(long_option[option_index].name,
+ CMD_LINE_OPT_UDP_PORT,
+ sizeof(CMD_LINE_OPT_UDP_PORT))) {
+ ret = parse_num_opt(optarg, INT16_MAX);
+ if (ret == -1) {
+ RTE_LOG(INFO, VHOST_CONFIG,
+ "Invalid argument for UDP port [0-N]\n");
+ tep_termination_usage(prgname);
+ return -1;
+ } else
+ udp_port = ret;
+ }
+
+ /* Specify the retries delay time (in useconds) on RX.*/
+ if (!strncmp(long_option[option_index].name,
+ CMD_LINE_OPT_RX_RETRY_DELAY,
+ sizeof(CMD_LINE_OPT_RX_RETRY_DELAY))) {
+ ret = parse_num_opt(optarg, INT32_MAX);
+ if (ret == -1) {
+ RTE_LOG(INFO, VHOST_CONFIG,
+ "Invalid argument for rx-retry-delay [0-N]\n");
+ tep_termination_usage(prgname);
+ return -1;
+ } else
+ burst_rx_delay_time = ret;
+ }
+
+ /* Specify the retries number on RX. */
+ if (!strncmp(long_option[option_index].name,
+ CMD_LINE_OPT_RX_RETRY_NUM,
+ sizeof(CMD_LINE_OPT_RX_RETRY_NUM))) {
+ ret = parse_num_opt(optarg, INT32_MAX);
+ if (ret == -1) {
+ RTE_LOG(INFO, VHOST_CONFIG,
+ "Invalid argument for rx-retry-num [0-N]\n");
+ tep_termination_usage(prgname);
+ return -1;
+ } else
+ burst_rx_retry_num = ret;
+ }
+
+ if (!strncmp(long_option[option_index].name,
+ CMD_LINE_OPT_TX_CHECKSUM,
+ sizeof(CMD_LINE_OPT_TX_CHECKSUM))) {
+ ret = parse_num_opt(optarg, 1);
+ if (ret == -1) {
+ RTE_LOG(INFO, VHOST_CONFIG,
+ "Invalid argument for tx-checksum [0|1]\n");
+ tep_termination_usage(prgname);
+ return -1;
+ } else
+ tx_checksum = ret;
+ }
+
+ if (!strncmp(long_option[option_index].name,
+ CMD_LINE_OPT_FILTER_TYPE,
+ sizeof(CMD_LINE_OPT_FILTER_TYPE))) {
+ ret = parse_num_opt(optarg, 3);
+ if ((ret == -1) || (ret == 0)) {
+ RTE_LOG(INFO, VHOST_CONFIG,
+ "Invalid argument for filter type [1-3]\n");
+ tep_termination_usage(prgname);
+ return -1;
+ } else
+ filter_idx = ret - 1;
+ }
+
+ /* Enable/disable encapsulation on RX. */
+ if (!strncmp(long_option[option_index].name,
+ CMD_LINE_OPT_DECAP,
+ sizeof(CMD_LINE_OPT_DECAP))) {
+ ret = parse_num_opt(optarg, 1);
+ if (ret == -1) {
+ RTE_LOG(INFO, VHOST_CONFIG,
+ "Invalid argument for decap [0|1]\n");
+ tep_termination_usage(prgname);
+ return -1;
+ } else
+ rx_decap = ret;
+ }
+
+ /* Enable/disable encapsulation on TX. */
+ if (!strncmp(long_option[option_index].name,
+ CMD_LINE_OPT_ENCAP,
+ sizeof(CMD_LINE_OPT_ENCAP))) {
+ ret = parse_num_opt(optarg, 1);
+ if (ret == -1) {
+ RTE_LOG(INFO, VHOST_CONFIG,
+ "Invalid argument for encap [0|1]\n");
+ tep_termination_usage(prgname);
+ return -1;
+ } else
+ tx_encap = ret;
+ }
+
+ /* Enable/disable stats. */
+ if (!strncmp(long_option[option_index].name,
+ CMD_LINE_OPT_STATS,
+ sizeof(CMD_LINE_OPT_STATS))) {
+ ret = parse_num_opt(optarg, INT32_MAX);
+ if (ret == -1) {
+ RTE_LOG(INFO, VHOST_CONFIG,
+ "Invalid argument for stats [0..N]\n");
+ tep_termination_usage(prgname);
+ return -1;
+ } else
+ enable_stats = ret;
+ }
+
+ /* Set character device basename. */
+ if (!strncmp(long_option[option_index].name,
+ CMD_LINE_OPT_DEV_BASENAME,
+ sizeof(CMD_LINE_OPT_DEV_BASENAME))) {
+ if (us_vhost_parse_basename(optarg) == -1) {
+ RTE_LOG(INFO, VHOST_CONFIG,
+ "Invalid argument for character "
+ "device basename (Max %d characters)\n",
+ MAX_BASENAME_SZ);
+ tep_termination_usage(prgname);
+ return -1;
+ }
+ }
+
+ break;
+
+ /* Invalid option - print options. */
+ default:
+ tep_termination_usage(prgname);
+ return -1;
+ }
+ }
+
+ for (i = 0; i < RTE_MAX_ETHPORTS; i++) {
+ if (enabled_port_mask & (1 << i))
+ ports[nb_ports++] = (uint8_t)i;
+ }
+
+ if ((nb_ports == 0) || (nb_ports > MAX_SUP_PORTS)) {
+ RTE_LOG(INFO, VHOST_PORT, "Current enabled port number is %u,"
+ "but only %u port can be enabled\n", nb_ports,
+ MAX_SUP_PORTS);
+ return -1;
+ }
+
+ return 0;
+}
+
+/**
+ * Update the global var NB_PORTS and array PORTS
+ * according to system ports number and return valid ports number
+ */
+static unsigned
+check_ports_num(unsigned max_nb_ports)
+{
+ unsigned valid_nb_ports = nb_ports;
+ unsigned portid;
+
+ if (nb_ports > max_nb_ports) {
+ RTE_LOG(INFO, VHOST_PORT, "\nSpecified port number(%u) "
+ " exceeds total system port number(%u)\n",
+ nb_ports, max_nb_ports);
+ nb_ports = max_nb_ports;
+ }
+
+ for (portid = 0; portid < nb_ports; portid++) {
+ if (!rte_eth_dev_is_valid_port(ports[portid])) {
+ RTE_LOG(INFO, VHOST_PORT,
+ "\nSpecified port ID(%u) is not valid\n",
+ ports[portid]);
+ ports[portid] = INVALID_PORT_ID;
+ valid_nb_ports--;
+ }
+ }
+ return valid_nb_ports;
+}
+
+/**
+ * This function routes the TX packet to the correct interface. This may be a local device
+ * or the physical port.
+ */
+static __rte_always_inline void
+virtio_tx_route(struct vhost_dev *vdev, struct rte_mbuf *m)
+{
+ struct mbuf_table *tx_q;
+ struct rte_mbuf **m_table;
+ unsigned len, ret = 0;
+ const uint16_t lcore_id = rte_lcore_id();
+
+ RTE_LOG_DP(DEBUG, VHOST_DATA, "(%d) TX: MAC address is external\n",
+ vdev->vid);
+
+ /* Add packet to the port tx queue */
+ tx_q = &lcore_tx_queue[lcore_id];
+ len = tx_q->len;
+
+ tx_q->m_table[len] = m;
+ len++;
+ if (enable_stats) {
+ dev_statistics[vdev->vid].tx_total++;
+ dev_statistics[vdev->vid].tx++;
+ }
+
+ if (unlikely(len == MAX_PKT_BURST)) {
+ m_table = (struct rte_mbuf **)tx_q->m_table;
+ ret = overlay_options.tx_handle(ports[0],
+ (uint16_t)tx_q->txq_id, m_table,
+ (uint16_t)tx_q->len);
+
+ /* Free any buffers not handled by TX and update
+ * the port stats.
+ */
+ if (unlikely(ret < len)) {
+ do {
+ rte_pktmbuf_free(m_table[ret]);
+ } while (++ret < len);
+ }
+
+ len = 0;
+ }
+
+ tx_q->len = len;
+ return;
+}
+
+/**
+ * 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
+ * with all devices in the main linked list.
+ */
+static int
+switch_worker(__rte_unused void *arg)
+{
+ struct rte_mempool *mbuf_pool = arg;
+ struct vhost_dev *vdev = NULL;
+ struct rte_mbuf *pkts_burst[MAX_PKT_BURST];
+ struct virtio_net_data_ll *dev_ll;
+ struct mbuf_table *tx_q;
+ volatile struct lcore_ll_info *lcore_ll;
+ const uint64_t drain_tsc = (rte_get_tsc_hz() + US_PER_S - 1)
+ / US_PER_S * BURST_TX_DRAIN_US;
+ uint64_t prev_tsc, diff_tsc, cur_tsc, ret_count = 0;
+ unsigned i, ret = 0;
+ const uint16_t lcore_id = rte_lcore_id();
+ const uint16_t num_cores = (uint16_t)rte_lcore_count();
+ uint16_t rx_count = 0;
+ uint16_t tx_count;
+ uint32_t retry = 0;
+
+ RTE_LOG(INFO, VHOST_DATA, "Procesing on Core %u started\n", lcore_id);
+ lcore_ll = lcore_info[lcore_id].lcore_ll;
+ prev_tsc = 0;
+
+ tx_q = &lcore_tx_queue[lcore_id];
+ for (i = 0; i < num_cores; i++) {
+ if (lcore_ids[i] == lcore_id) {
+ tx_q->txq_id = i;
+ break;
+ }
+ }
+
+ while (1) {
+ cur_tsc = rte_rdtsc();
+ /*
+ * TX burst queue drain
+ */
+ diff_tsc = cur_tsc - prev_tsc;
+ if (unlikely(diff_tsc > drain_tsc)) {
+
+ if (tx_q->len) {
+ RTE_LOG_DP(DEBUG, VHOST_DATA, "TX queue drained after "
+ "timeout with burst size %u\n",
+ tx_q->len);
+ ret = overlay_options.tx_handle(ports[0],
+ (uint16_t)tx_q->txq_id,
+ (struct rte_mbuf **)tx_q->m_table,
+ (uint16_t)tx_q->len);
+ if (unlikely(ret < tx_q->len)) {
+ do {
+ rte_pktmbuf_free(tx_q->m_table[ret]);
+ } while (++ret < tx_q->len);
+ }
+
+ tx_q->len = 0;
+ }
+
+ prev_tsc = cur_tsc;
+
+ }
+
+ rte_prefetch0(lcore_ll->ll_root_used);
+
+ /**
+ * Inform the configuration core that we have exited
+ * the linked list and that no devices are
+ * in use if requested.
+ */
+ if (lcore_ll->dev_removal_flag == REQUEST_DEV_REMOVAL)
+ lcore_ll->dev_removal_flag = ACK_DEV_REMOVAL;
+
+ /*
+ * Process devices
+ */
+ dev_ll = lcore_ll->ll_root_used;
+
+ while (dev_ll != NULL) {
+ vdev = dev_ll->vdev;
+
+ if (unlikely(vdev->remove)) {
+ dev_ll = dev_ll->next;
+ overlay_options.tunnel_destroy(vdev);
+ vdev->ready = DEVICE_SAFE_REMOVE;
+ continue;
+ }
+ if (likely(vdev->ready == DEVICE_RX)) {
+ /* Handle guest RX */
+ rx_count = rte_eth_rx_burst(ports[0],
+ vdev->rx_q, pkts_burst, MAX_PKT_BURST);
+
+ if (rx_count) {
+ /*
+ * Retry is enabled and the queue is
+ * full then we wait and retry to
+ * avoid packet loss. Here MAX_PKT_BURST
+ * must be less than virtio queue size
+ */
+ if (enable_retry && unlikely(rx_count >
+ rte_vhost_avail_entries(vdev->vid, VIRTIO_RXQ))) {
+ for (retry = 0; retry < burst_rx_retry_num;
+ retry++) {
+ rte_delay_us(burst_rx_delay_time);
+ if (rx_count <= rte_vhost_avail_entries(vdev->vid, VIRTIO_RXQ))
+ break;
+ }
+ }
+
+ ret_count = overlay_options.rx_handle(vdev->vid, pkts_burst, rx_count);
+ if (enable_stats) {
+ rte_atomic64_add(
+ &dev_statistics[vdev->vid].rx_total_atomic,
+ rx_count);
+ rte_atomic64_add(
+ &dev_statistics[vdev->vid].rx_atomic, ret_count);
+ }
+ while (likely(rx_count)) {
+ rx_count--;
+ rte_pktmbuf_free(pkts_burst[rx_count]);
+ }
+
+ }
+ }
+
+ if (likely(!vdev->remove)) {
+ /* Handle guest TX*/
+ tx_count = rte_vhost_dequeue_burst(vdev->vid,
+ VIRTIO_TXQ, mbuf_pool,
+ pkts_burst, MAX_PKT_BURST);
+ /* If this is the first received packet we need to learn the MAC */
+ if (unlikely(vdev->ready == DEVICE_MAC_LEARNING) && tx_count) {
+ if (vdev->remove ||
+ (overlay_options.tunnel_setup(vdev, pkts_burst[0]) == -1)) {
+ while (tx_count)
+ rte_pktmbuf_free(pkts_burst[--tx_count]);
+ }
+ }
+ while (tx_count)
+ virtio_tx_route(vdev, pkts_burst[--tx_count]);
+ }
+
+ /* move to the next device in the list */
+ dev_ll = dev_ll->next;
+ }
+ }
+
+ return 0;
+}
+
+/**
+ * Add an entry to a used linked list. A free entry must first be found
+ * in the free linked list using get_data_ll_free_entry();
+ */
+static void
+add_data_ll_entry(struct virtio_net_data_ll **ll_root_addr,
+ struct virtio_net_data_ll *ll_dev)
+{
+ struct virtio_net_data_ll *ll = *ll_root_addr;
+
+ /* Set next as NULL and use a compiler barrier to avoid reordering. */
+ ll_dev->next = NULL;
+ rte_compiler_barrier();
+
+ /* If ll == NULL then this is the first device. */
+ if (ll) {
+ /* Increment to the tail of the linked list. */
+ while (ll->next != NULL)
+ ll = ll->next;
+
+ ll->next = ll_dev;
+ } else {
+ *ll_root_addr = ll_dev;
+ }
+}
+
+/**
+ * Remove an entry from a used linked list. The entry must then be added to
+ * the free linked list using put_data_ll_free_entry().
+ */
+static void
+rm_data_ll_entry(struct virtio_net_data_ll **ll_root_addr,
+ struct virtio_net_data_ll *ll_dev,
+ struct virtio_net_data_ll *ll_dev_last)
+{
+ struct virtio_net_data_ll *ll = *ll_root_addr;
+
+ if (unlikely((ll == NULL) || (ll_dev == NULL)))
+ return;
+
+ if (ll_dev == ll)
+ *ll_root_addr = ll_dev->next;
+ else
+ if (likely(ll_dev_last != NULL))
+ ll_dev_last->next = ll_dev->next;
+ else
+ RTE_LOG(ERR, VHOST_CONFIG,
+ "Remove entry form ll failed.\n");
+}
+
+/**
+ * Find and return an entry from the free linked list.
+ */
+static struct virtio_net_data_ll *
+get_data_ll_free_entry(struct virtio_net_data_ll **ll_root_addr)
+{
+ struct virtio_net_data_ll *ll_free = *ll_root_addr;
+ struct virtio_net_data_ll *ll_dev;
+
+ if (ll_free == NULL)
+ return NULL;
+
+ ll_dev = ll_free;
+ *ll_root_addr = ll_free->next;
+
+ return ll_dev;
+}
+
+/**
+ * Place an entry back on to the free linked list.
+ */
+static void
+put_data_ll_free_entry(struct virtio_net_data_ll **ll_root_addr,
+ struct virtio_net_data_ll *ll_dev)
+{
+ struct virtio_net_data_ll *ll_free = *ll_root_addr;
+
+ if (ll_dev == NULL)
+ return;
+
+ ll_dev->next = ll_free;
+ *ll_root_addr = ll_dev;
+}
+
+/**
+ * Creates a linked list of a given size.
+ */
+static struct virtio_net_data_ll *
+alloc_data_ll(uint32_t size)
+{
+ struct virtio_net_data_ll *ll_new;
+ uint32_t i;
+
+ /* Malloc and then chain the linked list. */
+ ll_new = malloc(size * sizeof(struct virtio_net_data_ll));
+ if (ll_new == NULL) {
+ RTE_LOG(ERR, VHOST_CONFIG,
+ "Failed to allocate memory for ll_new.\n");
+ return NULL;
+ }
+
+ for (i = 0; i < size - 1; i++) {
+ ll_new[i].vdev = NULL;
+ ll_new[i].next = &ll_new[i+1];
+ }
+ ll_new[i].next = NULL;
+
+ return ll_new;
+}
+
+/**
+ * Create the main linked list along with each individual cores
+ * linked list. A used and a free list are created to manage entries.
+ */
+static int
+init_data_ll(void)
+{
+ int lcore;
+
+ RTE_LCORE_FOREACH_SLAVE(lcore) {
+ lcore_info[lcore].lcore_ll =
+ malloc(sizeof(struct lcore_ll_info));
+ if (lcore_info[lcore].lcore_ll == NULL) {
+ RTE_LOG(ERR, VHOST_CONFIG,
+ "Failed to allocate memory for lcore_ll.\n");
+ return -1;
+ }
+
+ lcore_info[lcore].lcore_ll->device_num = 0;
+ lcore_info[lcore].lcore_ll->dev_removal_flag = ACK_DEV_REMOVAL;
+ lcore_info[lcore].lcore_ll->ll_root_used = NULL;
+ if (nb_devices % nb_switching_cores)
+ lcore_info[lcore].lcore_ll->ll_root_free =
+ alloc_data_ll((nb_devices / nb_switching_cores)
+ + 1);
+ else
+ lcore_info[lcore].lcore_ll->ll_root_free =
+ alloc_data_ll(nb_devices / nb_switching_cores);
+ }
+
+ /* Allocate devices up to a maximum of MAX_DEVICES. */
+ ll_root_free = alloc_data_ll(MIN((nb_devices), MAX_DEVICES));
+
+ return 0;
+}
+
+/**
+ * Remove a device from the specific data core linked list and
+ * from the main linked list. Synchonization occurs through the use
+ * of the lcore dev_removal_flag.
+ */
+static void
+destroy_device(int vid)
+{
+ struct virtio_net_data_ll *ll_lcore_dev_cur;
+ struct virtio_net_data_ll *ll_main_dev_cur;
+ struct virtio_net_data_ll *ll_lcore_dev_last = NULL;
+ struct virtio_net_data_ll *ll_main_dev_last = NULL;
+ struct vhost_dev *vdev = NULL;
+ int lcore;
+
+ ll_main_dev_cur = ll_root_used;
+ while (ll_main_dev_cur != NULL) {
+ if (ll_main_dev_cur->vdev->vid == vid) {
+ vdev = ll_main_dev_cur->vdev;
+ break;
+ }
+ }
+ if (!vdev)
+ return;
+
+ /* set the remove flag. */
+ vdev->remove = 1;
+ while (vdev->ready != DEVICE_SAFE_REMOVE)
+ rte_pause();
+
+ /* Search for entry to be removed from lcore ll */
+ ll_lcore_dev_cur = lcore_info[vdev->coreid].lcore_ll->ll_root_used;
+ while (ll_lcore_dev_cur != NULL) {
+ if (ll_lcore_dev_cur->vdev == vdev) {
+ break;
+ } else {
+ ll_lcore_dev_last = ll_lcore_dev_cur;
+ ll_lcore_dev_cur = ll_lcore_dev_cur->next;
+ }
+ }
+
+ if (ll_lcore_dev_cur == NULL) {
+ RTE_LOG(ERR, VHOST_CONFIG,
+ "(%d) Failed to find the dev to be destroy.\n", vid);
+ return;
+ }
+
+ /* Search for entry to be removed from main ll */
+ ll_main_dev_cur = ll_root_used;
+ ll_main_dev_last = NULL;
+ while (ll_main_dev_cur != NULL) {
+ if (ll_main_dev_cur->vdev == vdev) {
+ break;
+ } else {
+ ll_main_dev_last = ll_main_dev_cur;
+ ll_main_dev_cur = ll_main_dev_cur->next;
+ }
+ }
+
+ /* Remove entries from the lcore and main ll. */
+ rm_data_ll_entry(&lcore_info[vdev->coreid].lcore_ll->ll_root_used,
+ ll_lcore_dev_cur, ll_lcore_dev_last);
+ rm_data_ll_entry(&ll_root_used, ll_main_dev_cur, ll_main_dev_last);
+
+ /* Set the dev_removal_flag on each lcore. */
+ RTE_LCORE_FOREACH_SLAVE(lcore) {
+ lcore_info[lcore].lcore_ll->dev_removal_flag =
+ REQUEST_DEV_REMOVAL;
+ }
+
+ /*
+ * Once each core has set the dev_removal_flag to
+ * ACK_DEV_REMOVAL we can be sure that they can no longer access
+ * the device removed from the linked lists and that the devices
+ * are no longer in use.
+ */
+ RTE_LCORE_FOREACH_SLAVE(lcore) {
+ while (lcore_info[lcore].lcore_ll->dev_removal_flag
+ != ACK_DEV_REMOVAL)
+ rte_pause();
+ }
+
+ /* Add the entries back to the lcore and main free ll.*/
+ put_data_ll_free_entry(&lcore_info[vdev->coreid].lcore_ll->ll_root_free,
+ ll_lcore_dev_cur);
+ put_data_ll_free_entry(&ll_root_free, ll_main_dev_cur);
+
+ /* Decrement number of device on the lcore. */
+ lcore_info[vdev->coreid].lcore_ll->device_num--;
+
+ RTE_LOG(INFO, VHOST_DATA, "(%d) Device has been removed "
+ "from data core\n", vid);
+
+ rte_free(vdev);
+
+}
+
+/**
+ * A new device is added to a data core. First the device is added
+ * to the main linked list and the allocated to a specific data core.
+ */
+static int
+new_device(int vid)
+{
+ struct virtio_net_data_ll *ll_dev;
+ int lcore, core_add = 0;
+ uint32_t device_num_min = nb_devices;
+ struct vhost_dev *vdev;
+
+ vdev = rte_zmalloc("vhost device", sizeof(*vdev), RTE_CACHE_LINE_SIZE);
+ if (vdev == NULL) {
+ RTE_LOG(INFO, VHOST_DATA,
+ "(%d) Couldn't allocate memory for vhost dev\n", vid);
+ return -1;
+ }
+ vdev->vid = vid;
+ /* Add device to main ll */
+ ll_dev = get_data_ll_free_entry(&ll_root_free);
+ if (ll_dev == NULL) {
+ RTE_LOG(INFO, VHOST_DATA, "(%d) No free entry found in"
+ " linked list Device limit of %d devices per core"
+ " has been reached\n", vid, nb_devices);
+ if (vdev->regions_hpa)
+ rte_free(vdev->regions_hpa);
+ rte_free(vdev);
+ return -1;
+ }
+ ll_dev->vdev = vdev;
+ add_data_ll_entry(&ll_root_used, ll_dev);
+ vdev->rx_q = vid;
+
+ /* reset ready flag */
+ vdev->ready = DEVICE_MAC_LEARNING;
+ vdev->remove = 0;
+
+ /* Find a suitable lcore to add the device. */
+ RTE_LCORE_FOREACH_SLAVE(lcore) {
+ if (lcore_info[lcore].lcore_ll->device_num < device_num_min) {
+ device_num_min = lcore_info[lcore].lcore_ll->device_num;
+ core_add = lcore;
+ }
+ }
+ /* Add device to lcore ll */
+ ll_dev = get_data_ll_free_entry(&lcore_info[core_add].lcore_ll->ll_root_free);
+ if (ll_dev == NULL) {
+ RTE_LOG(INFO, VHOST_DATA,
+ "(%d) Failed to add device to data core\n",
+ vid);
+ vdev->ready = DEVICE_SAFE_REMOVE;
+ destroy_device(vid);
+ rte_free(vdev->regions_hpa);
+ rte_free(vdev);
+ return -1;
+ }
+ ll_dev->vdev = vdev;
+ vdev->coreid = core_add;
+
+ add_data_ll_entry(&lcore_info[vdev->coreid].lcore_ll->ll_root_used,
+ ll_dev);
+
+ /* Initialize device stats */
+ memset(&dev_statistics[vid], 0,
+ sizeof(struct device_statistics));
+
+ /* Disable notifications. */
+ rte_vhost_enable_guest_notification(vid, VIRTIO_RXQ, 0);
+ rte_vhost_enable_guest_notification(vid, VIRTIO_TXQ, 0);
+ lcore_info[vdev->coreid].lcore_ll->device_num++;
+
+ RTE_LOG(INFO, VHOST_DATA, "(%d) Device has been added to data core %d\n",
+ vid, vdev->coreid);
+
+ return 0;
+}
+
+/**
+ * These callback allow devices to be added to the data core when configuration
+ * has been fully complete.
+ */
+static const struct vhost_device_ops virtio_net_device_ops = {
+ .new_device = new_device,
+ .destroy_device = destroy_device,
+};
+
+/**
+ * This is a thread will wake up after a period to print stats if the user has
+ * enabled them.
+ */
+static void *
+print_stats(__rte_unused void *arg)
+{
+ struct virtio_net_data_ll *dev_ll;
+ uint64_t tx_dropped, rx_dropped;
+ uint64_t tx, tx_total, rx, rx_total, rx_ip_csum, rx_l4_csum;
+ int vid;
+ const char clr[] = { 27, '[', '2', 'J', '\0' };
+ const char top_left[] = { 27, '[', '1', ';', '1', 'H', '\0' };
+
+ while (1) {
+ sleep(enable_stats);
+
+ /* Clear screen and move to top left */
+ printf("%s%s", clr, top_left);
+
+ printf("\nDevice statistics ================================");
+
+ dev_ll = ll_root_used;
+ while (dev_ll != NULL) {
+ vid = dev_ll->vdev->vid;
+ tx_total = dev_statistics[vid].tx_total;
+ tx = dev_statistics[vid].tx;
+ tx_dropped = tx_total - tx;
+
+ rx_total = rte_atomic64_read(
+ &dev_statistics[vid].rx_total_atomic);
+ rx = rte_atomic64_read(
+ &dev_statistics[vid].rx_atomic);
+ rx_dropped = rx_total - rx;
+ rx_ip_csum = rte_atomic64_read(
+ &dev_statistics[vid].rx_bad_ip_csum);
+ rx_l4_csum = rte_atomic64_read(
+ &dev_statistics[vid].rx_bad_l4_csum);
+
+ printf("\nStatistics for device %d ----------"
+ "\nTX total: %"PRIu64""
+ "\nTX dropped: %"PRIu64""
+ "\nTX successful: %"PRIu64""
+ "\nRX total: %"PRIu64""
+ "\nRX bad IP csum: %"PRIu64""
+ "\nRX bad L4 csum: %"PRIu64""
+ "\nRX dropped: %"PRIu64""
+ "\nRX successful: %"PRIu64"",
+ vid,
+ tx_total,
+ tx_dropped,
+ tx,
+ rx_total,
+ rx_ip_csum,
+ rx_l4_csum,
+ rx_dropped,
+ rx);
+
+ dev_ll = dev_ll->next;
+ }
+ printf("\n================================================\n");
+ }
+
+ return NULL;
+}
+
+/**
+ * Main function, does initialisation and calls the per-lcore functions.
+ */
+int
+main(int argc, char *argv[])
+{
+ struct rte_mempool *mbuf_pool = NULL;
+ unsigned lcore_id, core_id = 0;
+ unsigned nb_ports, valid_nb_ports;
+ int ret;
+ uint16_t portid;
+ uint16_t queue_id;
+ static pthread_t tid;
+
+ /* init EAL */
+ ret = rte_eal_init(argc, argv);
+ if (ret < 0)
+ rte_exit(EXIT_FAILURE, "Error with EAL initialization\n");
+ argc -= ret;
+ argv += ret;
+
+ /* parse app arguments */
+ ret = tep_termination_parse_args(argc, argv);
+ if (ret < 0)
+ rte_exit(EXIT_FAILURE, "Invalid argument\n");
+
+ for (lcore_id = 0; lcore_id < RTE_MAX_LCORE; lcore_id++)
+ if (rte_lcore_is_enabled(lcore_id))
+ lcore_ids[core_id++] = lcore_id;
+
+ /* set the number of swithcing cores available */
+ nb_switching_cores = rte_lcore_count()-1;
+
+ /* Get the number of physical ports. */
+ nb_ports = rte_eth_dev_count_avail();
+
+ /*
+ * Update the global var NB_PORTS and global array PORTS
+ * and get value of var VALID_NB_PORTS according to system ports number
+ */
+ valid_nb_ports = check_ports_num(nb_ports);
+
+ if ((valid_nb_ports == 0) || (valid_nb_ports > MAX_SUP_PORTS)) {
+ rte_exit(EXIT_FAILURE, "Current enabled port number is %u,"
+ "but only %u port can be enabled\n", nb_ports,
+ MAX_SUP_PORTS);
+ }
+ /* Create the mbuf pool. */
+ mbuf_pool = rte_pktmbuf_pool_create(
+ "MBUF_POOL",
+ NUM_MBUFS_PER_PORT * valid_nb_ports,
+ MBUF_CACHE_SIZE,
+ 0,
+ MBUF_DATA_SIZE,
+ rte_socket_id());
+ if (mbuf_pool == NULL)
+ rte_exit(EXIT_FAILURE, "Cannot create mbuf pool\n");
+
+ for (queue_id = 0; queue_id < MAX_QUEUES + 1; queue_id++)
+ vpool_array[queue_id].pool = mbuf_pool;
+
+ /* initialize all ports */
+ RTE_ETH_FOREACH_DEV(portid) {
+ /* skip ports that are not enabled */
+ if ((enabled_port_mask & (1 << portid)) == 0) {
+ RTE_LOG(INFO, VHOST_PORT,
+ "Skipping disabled port %d\n", portid);
+ continue;
+ }
+ if (overlay_options.port_configure(portid, mbuf_pool) != 0)
+ rte_exit(EXIT_FAILURE,
+ "Cannot initialize network ports\n");
+ }
+
+ /* Initialise all linked lists. */
+ if (init_data_ll() == -1)
+ rte_exit(EXIT_FAILURE, "Failed to initialize linked list\n");
+
+ /* Initialize device stats */
+ memset(&dev_statistics, 0, sizeof(dev_statistics));
+
+ /* Enable stats if the user option is set. */
+ if (enable_stats) {
+ ret = rte_ctrl_thread_create(&tid, "print-stats", NULL,
+ print_stats, NULL);
+ if (ret < 0)
+ rte_exit(EXIT_FAILURE, "Cannot create print-stats thread\n");
+ }
+
+ /* Launch all data cores. */
+ RTE_LCORE_FOREACH_SLAVE(lcore_id) {
+ rte_eal_remote_launch(switch_worker,
+ mbuf_pool, lcore_id);
+ }
+
+ ret = rte_vhost_driver_register((char *)&dev_basename, 0);
+ if (ret != 0)
+ rte_exit(EXIT_FAILURE, "failed to register vhost driver.\n");
+
+ rte_vhost_driver_disable_features(dev_basename,
+ 1ULL << VIRTIO_NET_F_MRG_RXBUF);
+
+ ret = rte_vhost_driver_callback_register(dev_basename,
+ &virtio_net_device_ops);
+ if (ret != 0) {
+ rte_exit(EXIT_FAILURE,
+ "failed to register vhost driver callbacks.\n");
+ }
+
+ if (rte_vhost_driver_start(dev_basename) < 0) {
+ rte_exit(EXIT_FAILURE,
+ "failed to start vhost driver.\n");
+ }
+
+ RTE_LCORE_FOREACH_SLAVE(lcore_id)
+ rte_eal_wait_lcore(lcore_id);
+
+ return 0;
+}
diff --git a/src/spdk/dpdk/examples/tep_termination/main.h b/src/spdk/dpdk/examples/tep_termination/main.h
new file mode 100644
index 00000000..966c63a5
--- /dev/null
+++ b/src/spdk/dpdk/examples/tep_termination/main.h
@@ -0,0 +1,93 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(c) 2010-2015 Intel Corporation
+ */
+
+#ifndef _MAIN_H_
+#define _MAIN_H_
+
+#include <rte_ether.h>
+
+/* Macros for printing using RTE_LOG */
+#define RTE_LOGTYPE_VHOST_CONFIG RTE_LOGTYPE_USER1
+#define RTE_LOGTYPE_VHOST_DATA RTE_LOGTYPE_USER2
+#define RTE_LOGTYPE_VHOST_PORT RTE_LOGTYPE_USER3
+
+/* State of virtio device. */
+#define DEVICE_MAC_LEARNING 0
+#define DEVICE_RX 1
+#define DEVICE_SAFE_REMOVE 2
+
+#define MAX_QUEUES 512
+
+/* Max burst size for RX/TX */
+#define MAX_PKT_BURST 32
+
+/* Max number of devices. Limited by the application. */
+#define MAX_DEVICES 64
+
+enum {VIRTIO_RXQ, VIRTIO_TXQ, VIRTIO_QNUM};
+
+/* Per-device statistics struct */
+struct device_statistics {
+ uint64_t tx_total;
+ rte_atomic64_t rx_total_atomic;
+ uint64_t rx_total;
+ uint64_t tx;
+ rte_atomic64_t rx_atomic;
+ /**< Bad inner IP csum for tunneling pkt */
+ rte_atomic64_t rx_bad_ip_csum;
+ /**< Bad inner L4 csum for tunneling pkt */
+ rte_atomic64_t rx_bad_l4_csum;
+} __rte_cache_aligned;
+
+/**
+ * Device linked list structure for data path.
+ */
+struct vhost_dev {
+ int vid;
+ /**< Number of memory regions for gpa to hpa translation. */
+ uint32_t nregions_hpa;
+ /**< Memory region information for gpa to hpa translation. */
+ struct virtio_memory_regions_hpa *regions_hpa;
+ /**< Device MAC address (Obtained on first TX packet). */
+ struct ether_addr mac_address;
+ /**< RX queue number. */
+ uint16_t rx_q;
+ /**< Data core that the device is added to. */
+ uint16_t coreid;
+ /**< A device is set as ready if the MAC address has been set. */
+ volatile uint8_t ready;
+ /**< Device is marked for removal from the data core. */
+ volatile uint8_t remove;
+} __rte_cache_aligned;
+
+/**
+ * Structure containing data core specific information.
+ */
+struct lcore_ll_info {
+ /**< Pointer to head in free linked list. */
+ struct virtio_net_data_ll *ll_root_free;
+ /**< Pointer to head of used linked list. */
+ struct virtio_net_data_ll *ll_root_used;
+ /**< Number of devices on lcore. */
+ uint32_t device_num;
+ /**< Flag to synchronize device removal. */
+ volatile uint8_t dev_removal_flag;
+};
+
+struct lcore_info {
+ /**< Pointer to data core specific lcore_ll_info struct */
+ struct lcore_ll_info *lcore_ll;
+};
+
+struct virtio_net_data_ll {
+ /**< Pointer to device created by configuration core. */
+ struct vhost_dev *vdev;
+ /**< Pointer to next device in linked list. */
+ struct virtio_net_data_ll *next;
+};
+
+uint32_t
+virtio_dev_rx(int vid, struct rte_mbuf **pkts, uint32_t count);
+
+#endif /* _MAIN_H_ */
diff --git a/src/spdk/dpdk/examples/tep_termination/meson.build b/src/spdk/dpdk/examples/tep_termination/meson.build
new file mode 100644
index 00000000..6d363624
--- /dev/null
+++ b/src/spdk/dpdk/examples/tep_termination/meson.build
@@ -0,0 +1,16 @@
+# SPDX-License-Identifier: BSD-3-Clause
+# Copyright(c) 2017 Intel Corporation
+
+# meson file, for building this example as part of a main DPDK build.
+#
+# To build this example as a standalone application with an already-installed
+# DPDK instance, use 'make'
+
+if host_machine.system() != 'linux'
+ build = false
+endif
+deps += ['hash', 'vhost']
+allow_experimental_apis = true
+sources = files(
+ 'main.c', 'vxlan.c', 'vxlan_setup.c'
+)
diff --git a/src/spdk/dpdk/examples/tep_termination/vxlan.c b/src/spdk/dpdk/examples/tep_termination/vxlan.c
new file mode 100644
index 00000000..7732821d
--- /dev/null
+++ b/src/spdk/dpdk/examples/tep_termination/vxlan.c
@@ -0,0 +1,236 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(c) 2010-2015 Intel Corporation
+ */
+
+#include <stdint.h>
+#include <rte_mbuf.h>
+#include <rte_hash_crc.h>
+#include <rte_byteorder.h>
+#include <rte_udp.h>
+#include <rte_tcp.h>
+#include <rte_sctp.h>
+
+#include "main.h"
+#include "vxlan.h"
+
+static uint16_t
+get_psd_sum(void *l3_hdr, uint16_t ethertype, uint64_t ol_flags)
+{
+ if (ethertype == ETHER_TYPE_IPv4)
+ return rte_ipv4_phdr_cksum(l3_hdr, ol_flags);
+ else /* assume ethertype == ETHER_TYPE_IPv6 */
+ return rte_ipv6_phdr_cksum(l3_hdr, ol_flags);
+}
+
+/**
+ * Parse an ethernet header to fill the ethertype, outer_l2_len, outer_l3_len and
+ * ipproto. This function is able to recognize IPv4/IPv6 with one optional vlan
+ * header.
+ */
+static void
+parse_ethernet(struct ether_hdr *eth_hdr, union tunnel_offload_info *info,
+ uint8_t *l4_proto)
+{
+ struct ipv4_hdr *ipv4_hdr;
+ struct ipv6_hdr *ipv6_hdr;
+ uint16_t ethertype;
+
+ info->outer_l2_len = sizeof(struct ether_hdr);
+ ethertype = rte_be_to_cpu_16(eth_hdr->ether_type);
+
+ if (ethertype == ETHER_TYPE_VLAN) {
+ struct vlan_hdr *vlan_hdr = (struct vlan_hdr *)(eth_hdr + 1);
+ info->outer_l2_len += sizeof(struct vlan_hdr);
+ ethertype = rte_be_to_cpu_16(vlan_hdr->eth_proto);
+ }
+
+ switch (ethertype) {
+ case ETHER_TYPE_IPv4:
+ ipv4_hdr = (struct ipv4_hdr *)
+ ((char *)eth_hdr + info->outer_l2_len);
+ info->outer_l3_len = sizeof(struct ipv4_hdr);
+ *l4_proto = ipv4_hdr->next_proto_id;
+ break;
+ case ETHER_TYPE_IPv6:
+ ipv6_hdr = (struct ipv6_hdr *)
+ ((char *)eth_hdr + info->outer_l2_len);
+ info->outer_l3_len = sizeof(struct ipv6_hdr);
+ *l4_proto = ipv6_hdr->proto;
+ break;
+ default:
+ info->outer_l3_len = 0;
+ *l4_proto = 0;
+ break;
+ }
+}
+
+/**
+ * Calculate the checksum of a packet in hardware
+ */
+static uint64_t
+process_inner_cksums(struct ether_hdr *eth_hdr, union tunnel_offload_info *info)
+{
+ void *l3_hdr = NULL;
+ uint8_t l4_proto;
+ uint16_t ethertype;
+ struct ipv4_hdr *ipv4_hdr;
+ struct ipv6_hdr *ipv6_hdr;
+ struct udp_hdr *udp_hdr;
+ struct tcp_hdr *tcp_hdr;
+ struct sctp_hdr *sctp_hdr;
+ uint64_t ol_flags = 0;
+
+ info->l2_len = sizeof(struct ether_hdr);
+ ethertype = rte_be_to_cpu_16(eth_hdr->ether_type);
+
+ if (ethertype == ETHER_TYPE_VLAN) {
+ struct vlan_hdr *vlan_hdr = (struct vlan_hdr *)(eth_hdr + 1);
+ info->l2_len += sizeof(struct vlan_hdr);
+ ethertype = rte_be_to_cpu_16(vlan_hdr->eth_proto);
+ }
+
+ l3_hdr = (char *)eth_hdr + info->l2_len;
+
+ if (ethertype == ETHER_TYPE_IPv4) {
+ ipv4_hdr = (struct ipv4_hdr *)l3_hdr;
+ ipv4_hdr->hdr_checksum = 0;
+ ol_flags |= PKT_TX_IPV4;
+ ol_flags |= PKT_TX_IP_CKSUM;
+ info->l3_len = sizeof(struct ipv4_hdr);
+ l4_proto = ipv4_hdr->next_proto_id;
+ } else if (ethertype == ETHER_TYPE_IPv6) {
+ ipv6_hdr = (struct ipv6_hdr *)l3_hdr;
+ info->l3_len = sizeof(struct ipv6_hdr);
+ l4_proto = ipv6_hdr->proto;
+ ol_flags |= PKT_TX_IPV6;
+ } else
+ return 0; /* packet type not supported, nothing to do */
+
+ if (l4_proto == IPPROTO_UDP) {
+ udp_hdr = (struct udp_hdr *)((char *)l3_hdr + info->l3_len);
+ ol_flags |= PKT_TX_UDP_CKSUM;
+ udp_hdr->dgram_cksum = get_psd_sum(l3_hdr,
+ ethertype, ol_flags);
+ } else if (l4_proto == IPPROTO_TCP) {
+ tcp_hdr = (struct tcp_hdr *)((char *)l3_hdr + info->l3_len);
+ /* Put PKT_TX_TCP_SEG bit setting before get_psd_sum(), because
+ * it depends on PKT_TX_TCP_SEG to calculate pseudo-header
+ * checksum.
+ */
+ if (tso_segsz != 0) {
+ ol_flags |= PKT_TX_TCP_SEG;
+ info->tso_segsz = tso_segsz;
+ info->l4_len = (tcp_hdr->data_off & 0xf0) >> 2;
+ }
+ ol_flags |= PKT_TX_TCP_CKSUM;
+ tcp_hdr->cksum = get_psd_sum(l3_hdr, ethertype, ol_flags);
+
+ } else if (l4_proto == IPPROTO_SCTP) {
+ sctp_hdr = (struct sctp_hdr *)((char *)l3_hdr + info->l3_len);
+ sctp_hdr->cksum = 0;
+ ol_flags |= PKT_TX_SCTP_CKSUM;
+ }
+
+ return ol_flags;
+}
+
+int
+decapsulation(struct rte_mbuf *pkt)
+{
+ uint8_t l4_proto = 0;
+ uint16_t outer_header_len;
+ struct udp_hdr *udp_hdr;
+ union tunnel_offload_info info = { .data = 0 };
+ struct ether_hdr *phdr = rte_pktmbuf_mtod(pkt, struct ether_hdr *);
+
+ parse_ethernet(phdr, &info, &l4_proto);
+
+ if (l4_proto != IPPROTO_UDP)
+ return -1;
+
+ udp_hdr = (struct udp_hdr *)((char *)phdr +
+ info.outer_l2_len + info.outer_l3_len);
+
+ /** check udp destination port, 4789 is the default vxlan port
+ * (rfc7348) or that the rx offload flag is set (i40e only
+ * currently)*/
+ if (udp_hdr->dst_port != rte_cpu_to_be_16(DEFAULT_VXLAN_PORT) &&
+ (pkt->packet_type & RTE_PTYPE_TUNNEL_MASK) == 0)
+ return -1;
+ outer_header_len = info.outer_l2_len + info.outer_l3_len
+ + sizeof(struct udp_hdr) + sizeof(struct vxlan_hdr);
+
+ rte_pktmbuf_adj(pkt, outer_header_len);
+
+ return 0;
+}
+
+void
+encapsulation(struct rte_mbuf *m, uint8_t queue_id)
+{
+ uint vport_id;
+ uint64_t ol_flags = 0;
+ uint32_t old_len = m->pkt_len, hash;
+ union tunnel_offload_info tx_offload = { .data = 0 };
+ struct ether_hdr *phdr = rte_pktmbuf_mtod(m, struct ether_hdr *);
+
+ /*Allocate space for new ethernet, IPv4, UDP and VXLAN headers*/
+ struct ether_hdr *pneth = (struct ether_hdr *) rte_pktmbuf_prepend(m,
+ sizeof(struct ether_hdr) + sizeof(struct ipv4_hdr)
+ + sizeof(struct udp_hdr) + sizeof(struct vxlan_hdr));
+
+ struct ipv4_hdr *ip = (struct ipv4_hdr *) &pneth[1];
+ struct udp_hdr *udp = (struct udp_hdr *) &ip[1];
+ struct vxlan_hdr *vxlan = (struct vxlan_hdr *) &udp[1];
+
+ /* convert TX queue ID to vport ID */
+ vport_id = queue_id - 1;
+
+ /* replace original Ethernet header with ours */
+ pneth = rte_memcpy(pneth, &app_l2_hdr[vport_id],
+ sizeof(struct ether_hdr));
+
+ /* copy in IP header */
+ ip = rte_memcpy(ip, &app_ip_hdr[vport_id],
+ sizeof(struct ipv4_hdr));
+ ip->total_length = rte_cpu_to_be_16(m->pkt_len
+ - sizeof(struct ether_hdr));
+
+ /* outer IP checksum */
+ ol_flags |= PKT_TX_OUTER_IP_CKSUM;
+ ip->hdr_checksum = 0;
+
+ /* inner IP checksum offload */
+ if (tx_checksum) {
+ ol_flags |= process_inner_cksums(phdr, &tx_offload);
+ m->l2_len = tx_offload.l2_len;
+ m->l3_len = tx_offload.l3_len;
+ m->l4_len = tx_offload.l4_len;
+ m->l2_len += ETHER_VXLAN_HLEN;
+ }
+
+ m->outer_l2_len = sizeof(struct ether_hdr);
+ m->outer_l3_len = sizeof(struct ipv4_hdr);
+
+ ol_flags |= PKT_TX_TUNNEL_VXLAN;
+
+ m->ol_flags |= ol_flags;
+ m->tso_segsz = tx_offload.tso_segsz;
+
+ /*VXLAN HEADER*/
+ vxlan->vx_flags = rte_cpu_to_be_32(VXLAN_HF_VNI);
+ vxlan->vx_vni = rte_cpu_to_be_32(vxdev.out_key << 8);
+
+ /*UDP HEADER*/
+ udp->dgram_cksum = 0;
+ udp->dgram_len = rte_cpu_to_be_16(old_len
+ + sizeof(struct udp_hdr)
+ + sizeof(struct vxlan_hdr));
+
+ udp->dst_port = rte_cpu_to_be_16(vxdev.dst_port);
+ hash = rte_hash_crc(phdr, 2 * ETHER_ADDR_LEN, phdr->ether_type);
+ udp->src_port = rte_cpu_to_be_16((((uint64_t) hash * PORT_RANGE) >> 32)
+ + PORT_MIN);
+
+ return;
+}
diff --git a/src/spdk/dpdk/examples/tep_termination/vxlan.h b/src/spdk/dpdk/examples/tep_termination/vxlan.h
new file mode 100644
index 00000000..bff786a2
--- /dev/null
+++ b/src/spdk/dpdk/examples/tep_termination/vxlan.h
@@ -0,0 +1,57 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(c) 2010-2015 Intel Corporation
+ */
+
+#ifndef _VXLAN_H_
+#define _VXLAN_H_
+
+#include <rte_ether.h>
+#include <rte_ip.h>
+
+#define PORT_MIN 49152
+#define PORT_MAX 65535
+#define PORT_RANGE ((PORT_MAX - PORT_MIN) + 1)
+
+#define VXLAN_N_PORTS 2
+#define VXLAN_HF_VNI 0x08000000
+#define DEFAULT_VXLAN_PORT 4789
+
+extern struct ipv4_hdr app_ip_hdr[VXLAN_N_PORTS];
+extern struct ether_hdr app_l2_hdr[VXLAN_N_PORTS];
+extern uint8_t tx_checksum;
+extern uint16_t tso_segsz;
+
+struct vxlan_port {
+ uint32_t vport_id; /**< VirtIO port id */
+ uint32_t peer_ip; /**< remote VTEP IP address */
+ struct ether_addr peer_mac; /**< remote VTEP MAC address */
+ struct ether_addr vport_mac; /**< VirtIO port MAC address */
+} __rte_cache_aligned;
+
+struct vxlan_conf {
+ uint16_t dst_port; /**< VXLAN UDP destination port */
+ uint32_t port_ip; /**< DPDK port IP address*/
+ uint32_t in_key; /**< VLAN ID */
+ uint32_t out_key; /**< VXLAN VNI */
+ struct vxlan_port port[VXLAN_N_PORTS]; /**< VXLAN configuration */
+} __rte_cache_aligned;
+
+extern struct vxlan_conf vxdev;
+
+/* structure that caches offload info for the current packet */
+union tunnel_offload_info {
+ uint64_t data;
+ struct {
+ uint64_t l2_len:7; /**< L2 (MAC) Header Length. */
+ uint64_t l3_len:9; /**< L3 (IP) Header Length. */
+ uint64_t l4_len:8; /**< L4 Header Length. */
+ uint64_t tso_segsz:16; /**< TCP TSO segment size */
+ uint64_t outer_l2_len:7; /**< outer L2 Header Length */
+ uint64_t outer_l3_len:16; /**< outer L3 Header Length */
+ };
+} __rte_cache_aligned;
+
+int decapsulation(struct rte_mbuf *pkt);
+void encapsulation(struct rte_mbuf *m, uint8_t queue_id);
+
+#endif /* _VXLAN_H_ */
diff --git a/src/spdk/dpdk/examples/tep_termination/vxlan_setup.c b/src/spdk/dpdk/examples/tep_termination/vxlan_setup.c
new file mode 100644
index 00000000..b99ab97d
--- /dev/null
+++ b/src/spdk/dpdk/examples/tep_termination/vxlan_setup.c
@@ -0,0 +1,440 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(c) 2010-2015 Intel Corporation
+ */
+
+#include <getopt.h>
+#include <linux/if_ether.h>
+#include <linux/if_vlan.h>
+#include <linux/virtio_net.h>
+#include <linux/virtio_ring.h>
+#include <sys/param.h>
+#include <unistd.h>
+
+#include <rte_ethdev.h>
+#include <rte_log.h>
+#include <rte_string_fns.h>
+#include <rte_mbuf.h>
+#include <rte_malloc.h>
+#include <rte_ip.h>
+#include <rte_udp.h>
+#include <rte_tcp.h>
+
+#include "main.h"
+#include "rte_vhost.h"
+#include "vxlan.h"
+#include "vxlan_setup.h"
+
+#define IPV4_HEADER_LEN 20
+#define UDP_HEADER_LEN 8
+#define VXLAN_HEADER_LEN 8
+
+#define IP_VERSION 0x40
+#define IP_HDRLEN 0x05 /* default IP header length == five 32-bits words. */
+#define IP_DEFTTL 64 /* from RFC 1340. */
+#define IP_VHL_DEF (IP_VERSION | IP_HDRLEN)
+
+#define IP_DN_FRAGMENT_FLAG 0x0040
+
+/* Used to compare MAC addresses. */
+#define MAC_ADDR_CMP 0xFFFFFFFFFFFFULL
+
+/* Configurable number of RX/TX ring descriptors */
+#define RTE_TEST_RX_DESC_DEFAULT 1024
+#define RTE_TEST_TX_DESC_DEFAULT 512
+
+/* Default inner VLAN ID */
+#define INNER_VLAN_ID 100
+
+/* VXLAN device */
+struct vxlan_conf vxdev;
+
+struct ipv4_hdr app_ip_hdr[VXLAN_N_PORTS];
+struct ether_hdr app_l2_hdr[VXLAN_N_PORTS];
+
+/* local VTEP IP address */
+uint8_t vxlan_multicast_ips[2][4] = { {239, 1, 1, 1 }, {239, 1, 2, 1 } };
+
+/* Remote VTEP IP address */
+uint8_t vxlan_overlay_ips[2][4] = { {192, 168, 10, 1}, {192, 168, 30, 1} };
+
+/* Remote VTEP MAC address */
+uint8_t peer_mac[6] = {0x00, 0x11, 0x01, 0x00, 0x00, 0x01};
+
+/* VXLAN RX filter type */
+uint8_t tep_filter_type[] = {RTE_TUNNEL_FILTER_IMAC_TENID,
+ RTE_TUNNEL_FILTER_IMAC_IVLAN_TENID,
+ RTE_TUNNEL_FILTER_OMAC_TENID_IMAC,};
+
+/* Options for configuring ethernet port */
+static struct rte_eth_conf port_conf = {
+ .rxmode = {
+ .split_hdr_size = 0,
+ .offloads = DEV_RX_OFFLOAD_CRC_STRIP,
+ },
+ .txmode = {
+ .mq_mode = ETH_MQ_TX_NONE,
+ .offloads = (DEV_TX_OFFLOAD_IPV4_CKSUM |
+ DEV_TX_OFFLOAD_UDP_CKSUM |
+ DEV_TX_OFFLOAD_TCP_CKSUM |
+ DEV_TX_OFFLOAD_SCTP_CKSUM |
+ DEV_TX_OFFLOAD_OUTER_IPV4_CKSUM |
+ DEV_TX_OFFLOAD_TCP_TSO |
+ DEV_TX_OFFLOAD_MULTI_SEGS |
+ DEV_TX_OFFLOAD_VXLAN_TNL_TSO),
+ },
+};
+
+/**
+ * The one or two device(s) that belongs to the same tenant ID can
+ * be assigned in a VM.
+ */
+const uint16_t tenant_id_conf[] = {
+ 1000, 1000, 1001, 1001, 1002, 1002, 1003, 1003,
+ 1004, 1004, 1005, 1005, 1006, 1006, 1007, 1007,
+ 1008, 1008, 1009, 1009, 1010, 1010, 1011, 1011,
+ 1012, 1012, 1013, 1013, 1014, 1014, 1015, 1015,
+ 1016, 1016, 1017, 1017, 1018, 1018, 1019, 1019,
+ 1020, 1020, 1021, 1021, 1022, 1022, 1023, 1023,
+ 1024, 1024, 1025, 1025, 1026, 1026, 1027, 1027,
+ 1028, 1028, 1029, 1029, 1030, 1030, 1031, 1031,
+};
+
+/**
+ * Initialises a given port using global settings and with the rx buffers
+ * coming from the mbuf_pool passed as parameter
+ */
+int
+vxlan_port_init(uint16_t port, struct rte_mempool *mbuf_pool)
+{
+ int retval;
+ uint16_t q;
+ struct rte_eth_dev_info dev_info;
+ uint16_t rx_rings, tx_rings = (uint16_t)rte_lcore_count();
+ uint16_t rx_ring_size = RTE_TEST_RX_DESC_DEFAULT;
+ uint16_t tx_ring_size = RTE_TEST_TX_DESC_DEFAULT;
+ struct rte_eth_udp_tunnel tunnel_udp;
+ struct rte_eth_rxconf *rxconf;
+ struct rte_eth_txconf *txconf;
+ struct vxlan_conf *pconf = &vxdev;
+ struct rte_eth_conf local_port_conf = port_conf;
+
+ pconf->dst_port = udp_port;
+
+ rte_eth_dev_info_get(port, &dev_info);
+
+ if (dev_info.max_rx_queues > MAX_QUEUES) {
+ rte_exit(EXIT_FAILURE,
+ "please define MAX_QUEUES no less than %u in %s\n",
+ dev_info.max_rx_queues, __FILE__);
+ }
+
+ rxconf = &dev_info.default_rxconf;
+ txconf = &dev_info.default_txconf;
+
+ if (!rte_eth_dev_is_valid_port(port))
+ return -1;
+
+ rx_rings = nb_devices;
+ if (dev_info.tx_offload_capa & DEV_TX_OFFLOAD_MBUF_FAST_FREE)
+ local_port_conf.txmode.offloads |=
+ DEV_TX_OFFLOAD_MBUF_FAST_FREE;
+ /* Configure ethernet device. */
+ retval = rte_eth_dev_configure(port, rx_rings, tx_rings,
+ &local_port_conf);
+ if (retval != 0)
+ return retval;
+
+ retval = rte_eth_dev_adjust_nb_rx_tx_desc(port, &rx_ring_size,
+ &tx_ring_size);
+ if (retval != 0)
+ return retval;
+
+ /* Setup the queues. */
+ rxconf->offloads = local_port_conf.rxmode.offloads;
+ for (q = 0; q < rx_rings; q++) {
+ retval = rte_eth_rx_queue_setup(port, q, rx_ring_size,
+ rte_eth_dev_socket_id(port),
+ rxconf,
+ mbuf_pool);
+ if (retval < 0)
+ return retval;
+ }
+ txconf->offloads = local_port_conf.txmode.offloads;
+ for (q = 0; q < tx_rings; q++) {
+ retval = rte_eth_tx_queue_setup(port, q, tx_ring_size,
+ rte_eth_dev_socket_id(port),
+ txconf);
+ if (retval < 0)
+ return retval;
+ }
+
+ /* Start the device. */
+ retval = rte_eth_dev_start(port);
+ if (retval < 0)
+ return retval;
+
+ /* Configure UDP port for UDP tunneling */
+ tunnel_udp.udp_port = udp_port;
+ tunnel_udp.prot_type = RTE_TUNNEL_TYPE_VXLAN;
+ retval = rte_eth_dev_udp_tunnel_port_add(port, &tunnel_udp);
+ if (retval < 0)
+ return retval;
+ rte_eth_macaddr_get(port, &ports_eth_addr[port]);
+ RTE_LOG(INFO, PORT, "Port %u MAC: %02"PRIx8" %02"PRIx8" %02"PRIx8
+ " %02"PRIx8" %02"PRIx8" %02"PRIx8"\n",
+ port,
+ ports_eth_addr[port].addr_bytes[0],
+ ports_eth_addr[port].addr_bytes[1],
+ ports_eth_addr[port].addr_bytes[2],
+ ports_eth_addr[port].addr_bytes[3],
+ ports_eth_addr[port].addr_bytes[4],
+ ports_eth_addr[port].addr_bytes[5]);
+
+ if (tso_segsz != 0) {
+ struct rte_eth_dev_info dev_info;
+ rte_eth_dev_info_get(port, &dev_info);
+ if ((dev_info.tx_offload_capa & DEV_TX_OFFLOAD_TCP_TSO) == 0)
+ RTE_LOG(WARNING, PORT,
+ "hardware TSO offload is not supported\n");
+ }
+ return 0;
+}
+
+static int
+vxlan_rx_process(struct rte_mbuf *pkt)
+{
+ int ret = 0;
+
+ if (rx_decap)
+ ret = decapsulation(pkt);
+
+ return ret;
+}
+
+static void
+vxlan_tx_process(uint8_t queue_id, struct rte_mbuf *pkt)
+{
+ if (tx_encap)
+ encapsulation(pkt, queue_id);
+
+ return;
+}
+
+/*
+ * This function learns the MAC address of the device and set init
+ * L2 header and L3 header info.
+ */
+int
+vxlan_link(struct vhost_dev *vdev, struct rte_mbuf *m)
+{
+ int i, ret;
+ struct ether_hdr *pkt_hdr;
+ uint64_t portid = vdev->vid;
+ struct ipv4_hdr *ip;
+
+ struct rte_eth_tunnel_filter_conf tunnel_filter_conf;
+
+ if (unlikely(portid >= VXLAN_N_PORTS)) {
+ RTE_LOG(INFO, VHOST_DATA,
+ "(%d) WARNING: Not configuring device,"
+ "as already have %d ports for VXLAN.",
+ vdev->vid, VXLAN_N_PORTS);
+ return -1;
+ }
+
+ /* Learn MAC address of guest device from packet */
+ pkt_hdr = rte_pktmbuf_mtod(m, struct ether_hdr *);
+ if (is_same_ether_addr(&(pkt_hdr->s_addr), &vdev->mac_address)) {
+ RTE_LOG(INFO, VHOST_DATA,
+ "(%d) WARNING: This device is using an existing"
+ " MAC address and has not been registered.\n",
+ vdev->vid);
+ return -1;
+ }
+
+ for (i = 0; i < ETHER_ADDR_LEN; i++) {
+ vdev->mac_address.addr_bytes[i] =
+ vxdev.port[portid].vport_mac.addr_bytes[i] =
+ pkt_hdr->s_addr.addr_bytes[i];
+ vxdev.port[portid].peer_mac.addr_bytes[i] = peer_mac[i];
+ }
+
+ memset(&tunnel_filter_conf, 0,
+ sizeof(struct rte_eth_tunnel_filter_conf));
+
+ ether_addr_copy(&ports_eth_addr[0], &tunnel_filter_conf.outer_mac);
+ tunnel_filter_conf.filter_type = tep_filter_type[filter_idx];
+
+ /* inner MAC */
+ ether_addr_copy(&vdev->mac_address, &tunnel_filter_conf.inner_mac);
+
+ tunnel_filter_conf.queue_id = vdev->rx_q;
+ tunnel_filter_conf.tenant_id = tenant_id_conf[vdev->rx_q];
+
+ if (tep_filter_type[filter_idx] == RTE_TUNNEL_FILTER_IMAC_IVLAN_TENID)
+ tunnel_filter_conf.inner_vlan = INNER_VLAN_ID;
+
+ tunnel_filter_conf.tunnel_type = RTE_TUNNEL_TYPE_VXLAN;
+
+ ret = rte_eth_dev_filter_ctrl(ports[0],
+ RTE_ETH_FILTER_TUNNEL,
+ RTE_ETH_FILTER_ADD,
+ &tunnel_filter_conf);
+ if (ret) {
+ RTE_LOG(ERR, VHOST_DATA,
+ "%d Failed to add device MAC address to cloud filter\n",
+ vdev->rx_q);
+ return -1;
+ }
+
+ /* Print out inner MAC and VNI info. */
+ RTE_LOG(INFO, VHOST_DATA,
+ "(%d) MAC_ADDRESS %02x:%02x:%02x:%02x:%02x:%02x and VNI %d registered\n",
+ vdev->rx_q,
+ vdev->mac_address.addr_bytes[0],
+ vdev->mac_address.addr_bytes[1],
+ vdev->mac_address.addr_bytes[2],
+ vdev->mac_address.addr_bytes[3],
+ vdev->mac_address.addr_bytes[4],
+ vdev->mac_address.addr_bytes[5],
+ tenant_id_conf[vdev->rx_q]);
+
+ vxdev.port[portid].vport_id = portid;
+
+ for (i = 0; i < 4; i++) {
+ /* Local VTEP IP */
+ vxdev.port_ip |= vxlan_multicast_ips[portid][i] << (8 * i);
+ /* Remote VTEP IP */
+ vxdev.port[portid].peer_ip |=
+ vxlan_overlay_ips[portid][i] << (8 * i);
+ }
+
+ vxdev.out_key = tenant_id_conf[vdev->rx_q];
+ ether_addr_copy(&vxdev.port[portid].peer_mac,
+ &app_l2_hdr[portid].d_addr);
+ ether_addr_copy(&ports_eth_addr[0],
+ &app_l2_hdr[portid].s_addr);
+ app_l2_hdr[portid].ether_type = rte_cpu_to_be_16(ETHER_TYPE_IPv4);
+
+ ip = &app_ip_hdr[portid];
+ ip->version_ihl = IP_VHL_DEF;
+ ip->type_of_service = 0;
+ ip->total_length = 0;
+ ip->packet_id = 0;
+ ip->fragment_offset = IP_DN_FRAGMENT_FLAG;
+ ip->time_to_live = IP_DEFTTL;
+ ip->next_proto_id = IPPROTO_UDP;
+ ip->hdr_checksum = 0;
+ ip->src_addr = vxdev.port_ip;
+ ip->dst_addr = vxdev.port[portid].peer_ip;
+
+ /* Set device as ready for RX. */
+ vdev->ready = DEVICE_RX;
+
+ return 0;
+}
+
+/**
+ * Removes cloud filter. Ensures that nothing is adding buffers to the RX
+ * queue before disabling RX on the device.
+ */
+void
+vxlan_unlink(struct vhost_dev *vdev)
+{
+ unsigned i = 0, rx_count;
+ int ret;
+ struct rte_mbuf *pkts_burst[MAX_PKT_BURST];
+ struct rte_eth_tunnel_filter_conf tunnel_filter_conf;
+
+ if (vdev->ready == DEVICE_RX) {
+ memset(&tunnel_filter_conf, 0,
+ sizeof(struct rte_eth_tunnel_filter_conf));
+
+ ether_addr_copy(&ports_eth_addr[0], &tunnel_filter_conf.outer_mac);
+ ether_addr_copy(&vdev->mac_address, &tunnel_filter_conf.inner_mac);
+ tunnel_filter_conf.tenant_id = tenant_id_conf[vdev->rx_q];
+ tunnel_filter_conf.filter_type = tep_filter_type[filter_idx];
+
+ if (tep_filter_type[filter_idx] ==
+ RTE_TUNNEL_FILTER_IMAC_IVLAN_TENID)
+ tunnel_filter_conf.inner_vlan = INNER_VLAN_ID;
+
+ tunnel_filter_conf.queue_id = vdev->rx_q;
+ tunnel_filter_conf.tunnel_type = RTE_TUNNEL_TYPE_VXLAN;
+
+ ret = rte_eth_dev_filter_ctrl(ports[0],
+ RTE_ETH_FILTER_TUNNEL,
+ RTE_ETH_FILTER_DELETE,
+ &tunnel_filter_conf);
+ if (ret) {
+ RTE_LOG(ERR, VHOST_DATA,
+ "%d Failed to add device MAC address to cloud filter\n",
+ vdev->rx_q);
+ return;
+ }
+ for (i = 0; i < ETHER_ADDR_LEN; i++)
+ vdev->mac_address.addr_bytes[i] = 0;
+
+ /* Clear out the receive buffers */
+ rx_count = rte_eth_rx_burst(ports[0],
+ (uint16_t)vdev->rx_q,
+ pkts_burst, MAX_PKT_BURST);
+
+ while (rx_count) {
+ for (i = 0; i < rx_count; i++)
+ rte_pktmbuf_free(pkts_burst[i]);
+
+ rx_count = rte_eth_rx_burst(ports[0],
+ (uint16_t)vdev->rx_q,
+ pkts_burst, MAX_PKT_BURST);
+ }
+ vdev->ready = DEVICE_MAC_LEARNING;
+ }
+}
+
+/* Transmit packets after encapsulating */
+int
+vxlan_tx_pkts(uint16_t port_id, uint16_t queue_id,
+ struct rte_mbuf **tx_pkts, uint16_t nb_pkts) {
+ int ret = 0;
+ uint16_t i;
+
+ for (i = 0; i < nb_pkts; i++)
+ vxlan_tx_process(queue_id, tx_pkts[i]);
+
+ ret = rte_eth_tx_burst(port_id, queue_id, tx_pkts, nb_pkts);
+
+ return ret;
+}
+
+/* Check for decapsulation and pass packets directly to VIRTIO device */
+int
+vxlan_rx_pkts(int vid, struct rte_mbuf **pkts_burst, uint32_t rx_count)
+{
+ uint32_t i = 0;
+ uint32_t count = 0;
+ int ret;
+ struct rte_mbuf *pkts_valid[rx_count];
+
+ for (i = 0; i < rx_count; i++) {
+ if (enable_stats) {
+ rte_atomic64_add(
+ &dev_statistics[vid].rx_bad_ip_csum,
+ (pkts_burst[i]->ol_flags & PKT_RX_IP_CKSUM_BAD)
+ != 0);
+ rte_atomic64_add(
+ &dev_statistics[vid].rx_bad_ip_csum,
+ (pkts_burst[i]->ol_flags & PKT_RX_L4_CKSUM_BAD)
+ != 0);
+ }
+ ret = vxlan_rx_process(pkts_burst[i]);
+ if (unlikely(ret < 0))
+ continue;
+
+ pkts_valid[count] = pkts_burst[i];
+ count++;
+ }
+
+ ret = rte_vhost_enqueue_burst(vid, VIRTIO_RXQ, pkts_valid, count);
+ return ret;
+}
diff --git a/src/spdk/dpdk/examples/tep_termination/vxlan_setup.h b/src/spdk/dpdk/examples/tep_termination/vxlan_setup.h
new file mode 100644
index 00000000..2c20e2e3
--- /dev/null
+++ b/src/spdk/dpdk/examples/tep_termination/vxlan_setup.h
@@ -0,0 +1,58 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(c) 2010-2015 Intel Corporation
+ */
+
+#ifndef VXLAN_SETUP_H_
+#define VXLAN_SETUP_H_
+
+extern uint16_t nb_devices;
+extern uint16_t udp_port;
+extern uint8_t filter_idx;
+extern uint16_t ports[RTE_MAX_ETHPORTS];
+extern struct ether_addr ports_eth_addr[RTE_MAX_ETHPORTS];
+extern uint32_t enable_stats;
+extern struct device_statistics dev_statistics[MAX_DEVICES];
+extern uint8_t rx_decap;
+extern uint8_t tx_encap;
+
+typedef int (*ol_port_configure_t)(uint16_t port,
+ struct rte_mempool *mbuf_pool);
+
+typedef int (*ol_tunnel_setup_t)(struct vhost_dev *vdev,
+ struct rte_mbuf *m);
+
+typedef void (*ol_tunnel_destroy_t)(struct vhost_dev *vdev);
+
+typedef int (*ol_tx_handle_t)(uint16_t port_id, uint16_t queue_id,
+ struct rte_mbuf **tx_pkts, uint16_t nb_pkts);
+
+typedef int (*ol_rx_handle_t)(int vid, struct rte_mbuf **pkts,
+ uint32_t count);
+
+typedef int (*ol_param_handle)(int vid);
+
+struct ol_switch_ops {
+ ol_port_configure_t port_configure;
+ ol_tunnel_setup_t tunnel_setup;
+ ol_tunnel_destroy_t tunnel_destroy;
+ ol_tx_handle_t tx_handle;
+ ol_rx_handle_t rx_handle;
+ ol_param_handle param_handle;
+};
+
+int
+vxlan_port_init(uint16_t port, struct rte_mempool *mbuf_pool);
+
+int
+vxlan_link(struct vhost_dev *vdev, struct rte_mbuf *m);
+
+void
+vxlan_unlink(struct vhost_dev *vdev);
+
+int
+vxlan_tx_pkts(uint16_t port_id, uint16_t queue_id,
+ struct rte_mbuf **tx_pkts, uint16_t nb_pkts);
+int
+vxlan_rx_pkts(int vid, struct rte_mbuf **pkts, uint32_t count);
+
+#endif /* VXLAN_SETUP_H_ */