#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 "main.h"
#ifndef MAX_QUEUES
#define MAX_QUEUES 128
#endif
#define MAX_SUP_PORTS 1
#define MBUF_CACHE_SIZE 128
#define MBUF_DATA_SIZE RTE_MBUF_DEFAULT_BUF_SIZE
#define BURST_TX_DRAIN_US 100
#define BURST_RX_WAIT_US 15
#define BURST_RX_RETRIES 4
#define JUMBO_FRAME_MAX_SIZE 0x2600
#define DEVICE_MAC_LEARNING 0
#define DEVICE_RX 1
#define DEVICE_SAFE_REMOVE 2
#define RTE_TEST_RX_DESC_DEFAULT 1024
#define RTE_TEST_TX_DESC_DEFAULT 512
#define INVALID_PORT_ID 0xFF
#define MAX_DEVICES 64
#define MAX_PRINT_BUFF 6072
#define MAX_LONG_OPT_SZ 64
static uint32_t enabled_port_mask = 0;
static uint32_t promiscuous;
static uint32_t num_queues = 0;
static uint32_t num_devices;
static int mergeable;
typedef enum {
VM2VM_DISABLED = 0,
VM2VM_SOFTWARE = 1,
VM2VM_HARDWARE = 2,
VM2VM_LAST
} vm2vm_type;
static vm2vm_type vm2vm_mode = VM2VM_SOFTWARE;
static uint32_t enable_stats = 0;
static uint32_t enable_retry = 1;
static uint32_t enable_tx_csum;
static uint32_t enable_tso;
static int client_mode;
static int dequeue_zero_copy;
static int builtin_net_driver;
static uint32_t burst_rx_delay_time = BURST_RX_WAIT_US;
static uint32_t burst_rx_retry_num = BURST_RX_RETRIES;
static char *socket_files;
static int nb_sockets;
.split_hdr_size = 0,
.header_split = 0,
.hw_ip_checksum = 0,
.hw_vlan_filter = 0,
.hw_vlan_strip = 1,
.jumbo_frame = 0,
.hw_strip_crc = 1,
},
.txmode = {
},
.rx_adv_conf = {
.vmdq_rx_conf = {
.enable_default_pool = 0,
.default_pool = 0,
.nb_pool_maps = 0,
.pool_map = {{0, 0},},
},
},
};
static unsigned lcore_ids[RTE_MAX_LCORE];
static uint8_t ports[RTE_MAX_ETHPORTS];
static unsigned num_ports = 0;
static uint16_t num_pf_queues, num_vmdq_queues;
static uint16_t vmdq_pool_base, vmdq_queue_base;
static uint16_t queues_per_pool;
const uint16_t vlan_tags[] = {
1000, 1001, 1002, 1003, 1004, 1005, 1006, 1007,
1008, 1009, 1010, 1011, 1012, 1013, 1014, 1015,
1016, 1017, 1018, 1019, 1020, 1021, 1022, 1023,
1024, 1025, 1026, 1027, 1028, 1029, 1030, 1031,
1032, 1033, 1034, 1035, 1036, 1037, 1038, 1039,
1040, 1041, 1042, 1043, 1044, 1045, 1046, 1047,
1048, 1049, 1050, 1051, 1052, 1053, 1054, 1055,
1056, 1057, 1058, 1059, 1060, 1061, 1062, 1063,
};
static struct ether_addr vmdq_ports_eth_addr[RTE_MAX_ETHPORTS];
static struct vhost_dev_tailq_list vhost_dev_list =
TAILQ_HEAD_INITIALIZER(vhost_dev_list);
static struct lcore_info lcore_info[RTE_MAX_LCORE];
struct mbuf_table {
unsigned len;
unsigned txq_id;
struct rte_mbuf *m_table[MAX_PKT_BURST];
};
struct mbuf_table lcore_tx_queue[RTE_MAX_LCORE];
#define MBUF_TABLE_DRAIN_TSC ((rte_get_tsc_hz() + US_PER_S - 1) \
/ US_PER_S * BURST_TX_DRAIN_US)
#define VLAN_HLEN 4
static inline int
get_eth_conf(
struct rte_eth_conf *eth_conf, uint32_t num_devices)
{
struct rte_eth_vmdq_rx_conf conf;
struct rte_eth_vmdq_rx_conf *def_conf =
unsigned i;
memset(&conf, 0, sizeof(conf));
conf.nb_pool_maps = num_devices;
conf.enable_loop_back = def_conf->enable_loop_back;
conf.rx_mode = def_conf->rx_mode;
for (i = 0; i < conf.nb_pool_maps; i++) {
conf.pool_map[i].vlan_id = vlan_tags[ i ];
conf.pool_map[i].pools = (1UL << i);
}
(void)(
rte_memcpy(eth_conf, &vmdq_conf_default,
sizeof(*eth_conf)));
return 0;
}
static inline int
validate_num_devices(uint32_t max_nb_devices)
{
if (num_devices > max_nb_devices) {
RTE_LOG(ERR, VHOST_PORT,
"invalid number of devices\n");
return -1;
}
return 0;
}
static inline int
{
int16_t rx_rings, tx_rings;
uint16_t rx_ring_size, tx_ring_size;
int retval;
uint16_t q;
if (dev_info.max_rx_queues > MAX_QUEUES) {
"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;
num_devices = dev_info.max_vmdq_pools;
rx_ring_size = RTE_TEST_RX_DESC_DEFAULT;
tx_ring_size = RTE_TEST_TX_DESC_DEFAULT;
if (dequeue_zero_copy)
tx_ring_size = 64;
retval = validate_num_devices(MAX_DEVICES);
if (retval < 0)
return retval;
retval = get_eth_conf(&port_conf, num_devices);
if (retval < 0)
return retval;
num_pf_queues = dev_info.max_rx_queues - dev_info.vmdq_queue_num;
queues_per_pool = dev_info.vmdq_queue_num / dev_info.max_vmdq_pools;
num_vmdq_queues = num_devices * queues_per_pool;
num_queues = num_pf_queues + num_vmdq_queues;
vmdq_queue_base = dev_info.vmdq_queue_base;
vmdq_pool_base = dev_info.vmdq_pool_base;
printf("pf queue num: %u, configured vmdq pool num: %u, each vmdq pool has %u queues\n",
num_pf_queues, num_devices, queues_per_pool);
rx_rings = (uint16_t)dev_info.max_rx_queues;
if (retval != 0) {
RTE_LOG(ERR, VHOST_PORT,
"Failed to configure port %u: %s.\n",
port, strerror(-retval));
return retval;
}
for (q = 0; q < rx_rings; q ++) {
rxconf,
mbuf_pool);
if (retval < 0) {
"Failed to setup rx queue %u of port %u: %s.\n",
q, port, strerror(-retval));
return retval;
}
}
for (q = 0; q < tx_rings; q ++) {
txconf);
if (retval < 0) {
"Failed to setup tx queue %u of port %u: %s.\n",
q, port, strerror(-retval));
return retval;
}
}
if (retval < 0) {
RTE_LOG(ERR, VHOST_PORT,
"Failed to start port %u: %s\n",
port, strerror(-retval));
return retval;
}
if (promiscuous)
RTE_LOG(INFO, VHOST_PORT,
"Max virtio devices supported: %u\n", num_devices);
RTE_LOG(INFO, VHOST_PORT,
"Port %u MAC: %02"PRIx8
" %02"PRIx8
" %02"PRIx8
" %02"PRIx8" %02"PRIx8" %02"PRIx8"\n",
(unsigned)port,
vmdq_ports_eth_addr[port].addr_bytes[0],
vmdq_ports_eth_addr[port].addr_bytes[1],
vmdq_ports_eth_addr[port].addr_bytes[2],
vmdq_ports_eth_addr[port].addr_bytes[3],
vmdq_ports_eth_addr[port].addr_bytes[4],
vmdq_ports_eth_addr[port].addr_bytes[5]);
return 0;
}
static int
us_vhost_parse_socket_path(const char *q_arg)
{
if (strnlen(q_arg, PATH_MAX) == PATH_MAX)
return -1;
socket_files = realloc(socket_files, PATH_MAX * (nb_sockets + 1));
snprintf(socket_files + nb_sockets * PATH_MAX, PATH_MAX, "%s", q_arg);
nb_sockets++;
return 0;
}
static int
parse_portmask(const char *portmask)
{
char *end = NULL;
unsigned long pm;
errno = 0;
pm = strtoul(portmask, &end, 16);
if ((portmask[0] == '\0') || (end == NULL) || (*end != '\0') || (errno != 0))
return -1;
if (pm == 0)
return -1;
return pm;
}
static int
parse_num_opt(const char *q_arg, uint32_t max_valid_value)
{
char *end = NULL;
unsigned long num;
errno = 0;
num = strtoul(q_arg, &end, 10);
if ((q_arg[0] == '\0') || (end == NULL) || (*end != '\0') || (errno != 0))
return -1;
if (num > max_valid_value)
return -1;
return num;
}
static void
us_vhost_usage(const char *prgname)
{
RTE_LOG(INFO, VHOST_CONFIG,
"%s [EAL options] -- -p PORTMASK\n"
" --vm2vm [0|1|2]\n"
" --rx_retry [0|1] --mergeable [0|1] --stats [0-N]\n"
" --socket-file <path>\n"
" --nb-devices ND\n"
" -p PORTMASK: Set mask for ports to be used by application\n"
" --vm2vm [0|1|2]: disable/software(default)/hardware vm2vm comms\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"
" --mergeable [0|1]: disable(default)/enable RX mergeable buffers\n"
" --stats [0-N]: 0: Disable stats, N: Time in seconds to print stats\n"
" --socket-file: The path of the socket file.\n"
" --tx-csum [0|1] disable/enable TX checksum offload.\n"
" --tso [0|1] disable/enable TCP segment offload.\n"
" --client register a vhost-user socket as client mode.\n"
" --dequeue-zero-copy enables dequeue zero copy\n",
prgname);
}
static int
us_vhost_parse_args(int argc, char **argv)
{
int opt, ret;
int option_index;
unsigned i;
const char *prgname = argv[0];
static struct option long_option[] = {
{"vm2vm", required_argument, NULL, 0},
{"rx-retry", required_argument, NULL, 0},
{"rx-retry-delay", required_argument, NULL, 0},
{"rx-retry-num", required_argument, NULL, 0},
{"mergeable", required_argument, NULL, 0},
{"stats", required_argument, NULL, 0},
{"socket-file", required_argument, NULL, 0},
{"tx-csum", required_argument, NULL, 0},
{"tso", required_argument, NULL, 0},
{"client", no_argument, &client_mode, 1},
{"dequeue-zero-copy", no_argument, &dequeue_zero_copy, 1},
{"builtin-net-driver", no_argument, &builtin_net_driver, 1},
{NULL, 0, 0, 0},
};
while ((opt = getopt_long(argc, argv, "p:P",
long_option, &option_index)) != EOF) {
switch (opt) {
case 'p':
enabled_port_mask = parse_portmask(optarg);
if (enabled_port_mask == 0) {
RTE_LOG(INFO, VHOST_CONFIG,
"Invalid portmask\n");
us_vhost_usage(prgname);
return -1;
}
break;
case 'P':
promiscuous = 1;
break;
case 0:
if (!strncmp(long_option[option_index].name, "vm2vm",
MAX_LONG_OPT_SZ)) {
ret = parse_num_opt(optarg, (VM2VM_LAST - 1));
if (ret == -1) {
"Invalid argument for "
"vm2vm [0|1|2]\n");
us_vhost_usage(prgname);
return -1;
} else {
vm2vm_mode = (vm2vm_type)ret;
}
}
if (!strncmp(long_option[option_index].name, "rx-retry", MAX_LONG_OPT_SZ)) {
ret = parse_num_opt(optarg, 1);
if (ret == -1) {
RTE_LOG(INFO, VHOST_CONFIG,
"Invalid argument for rx-retry [0|1]\n");
us_vhost_usage(prgname);
return -1;
} else {
enable_retry = ret;
}
}
if (!strncmp(long_option[option_index].name, "tx-csum", MAX_LONG_OPT_SZ)) {
ret = parse_num_opt(optarg, 1);
if (ret == -1) {
RTE_LOG(INFO, VHOST_CONFIG,
"Invalid argument for tx-csum [0|1]\n");
us_vhost_usage(prgname);
return -1;
} else
enable_tx_csum = ret;
}
if (!strncmp(long_option[option_index].name, "tso", MAX_LONG_OPT_SZ)) {
ret = parse_num_opt(optarg, 1);
if (ret == -1) {
RTE_LOG(INFO, VHOST_CONFIG,
"Invalid argument for tso [0|1]\n");
us_vhost_usage(prgname);
return -1;
} else
enable_tso = ret;
}
if (!strncmp(long_option[option_index].name, "rx-retry-delay", MAX_LONG_OPT_SZ)) {
ret = parse_num_opt(optarg, INT32_MAX);
if (ret == -1) {
RTE_LOG(INFO, VHOST_CONFIG,
"Invalid argument for rx-retry-delay [0-N]\n");
us_vhost_usage(prgname);
return -1;
} else {
burst_rx_delay_time = ret;
}
}
if (!strncmp(long_option[option_index].name, "rx-retry-num", MAX_LONG_OPT_SZ)) {
ret = parse_num_opt(optarg, INT32_MAX);
if (ret == -1) {
RTE_LOG(INFO, VHOST_CONFIG,
"Invalid argument for rx-retry-num [0-N]\n");
us_vhost_usage(prgname);
return -1;
} else {
burst_rx_retry_num = ret;
}
}
if (!strncmp(long_option[option_index].name, "mergeable", MAX_LONG_OPT_SZ)) {
ret = parse_num_opt(optarg, 1);
if (ret == -1) {
RTE_LOG(INFO, VHOST_CONFIG,
"Invalid argument for mergeable [0|1]\n");
us_vhost_usage(prgname);
return -1;
} else {
mergeable = !!ret;
if (ret) {
= JUMBO_FRAME_MAX_SIZE;
}
}
}
if (!strncmp(long_option[option_index].name, "stats", MAX_LONG_OPT_SZ)) {
ret = parse_num_opt(optarg, INT32_MAX);
if (ret == -1) {
"Invalid argument for stats [0..N]\n");
us_vhost_usage(prgname);
return -1;
} else {
enable_stats = ret;
}
}
if (!strncmp(long_option[option_index].name,
"socket-file", MAX_LONG_OPT_SZ)) {
if (us_vhost_parse_socket_path(optarg) == -1) {
"Invalid argument for socket name (Max %d characters)\n",
PATH_MAX);
us_vhost_usage(prgname);
return -1;
}
}
break;
default:
us_vhost_usage(prgname);
return -1;
}
}
for (i = 0; i < RTE_MAX_ETHPORTS; i++) {
if (enabled_port_mask & (1 << i))
ports[num_ports++] = (uint8_t)i;
}
if ((num_ports == 0) || (num_ports > MAX_SUP_PORTS)) {
RTE_LOG(INFO, VHOST_PORT,
"Current enabled port number is %u,"
"but only %u port can be enabled\n",num_ports, MAX_SUP_PORTS);
return -1;
}
return 0;
}
static unsigned check_ports_num(unsigned nb_ports)
{
unsigned valid_num_ports = num_ports;
unsigned portid;
if (num_ports > nb_ports) {
RTE_LOG(INFO, VHOST_PORT,
"\nSpecified port number(%u) exceeds total system port number(%u)\n",
num_ports, nb_ports);
num_ports = nb_ports;
}
for (portid = 0; portid < num_ports; portid ++) {
if (ports[portid] >= nb_ports) {
RTE_LOG(INFO, VHOST_PORT,
"\nSpecified port ID(%u) exceeds max system port ID(%u)\n",
ports[portid], (nb_ports - 1));
ports[portid] = INVALID_PORT_ID;
valid_num_ports--;
}
}
return valid_num_ports;
}
static inline struct vhost_dev *__attribute__((always_inline))
{
struct vhost_dev *vdev;
TAILQ_FOREACH(vdev, &vhost_dev_list, global_vdev_entry) {
if (vdev->ready == DEVICE_RX &&
return vdev;
}
return NULL;
}
static int
link_vmdq(
struct vhost_dev *vdev,
struct rte_mbuf *m)
{
int i, ret;
if (find_vhost_dev(&pkt_hdr->
s_addr)) {
"(%d) device is using a registered MAC!\n",
vdev->vid);
return -1;
}
vdev->vlan_tag = vlan_tags[vdev->vid];
"(%d) mac %02x:%02x:%02x:%02x:%02x:%02x and vlan %d registered\n",
vdev->vid,
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],
vdev->vlan_tag);
(uint32_t)vdev->vid + vmdq_pool_base);
if (ret)
"(%d) failed to add device MAC address to VMDQ\n",
vdev->vid);
vdev->ready = DEVICE_RX;
return 0;
}
static inline void
unlink_vmdq(struct vhost_dev *vdev)
{
unsigned i = 0;
unsigned rx_count;
struct rte_mbuf *pkts_burst[MAX_PKT_BURST];
if (vdev->ready == DEVICE_RX) {
for (i = 0; i < 6; i++)
vdev->mac_address.addr_bytes[i] = 0;
vdev->vlan_tag = 0;
(uint16_t)vdev->vmdq_rx_q, pkts_burst, MAX_PKT_BURST);
while (rx_count) {
for (i = 0; i < rx_count; i++)
(uint16_t)vdev->vmdq_rx_q, pkts_burst, MAX_PKT_BURST);
}
vdev->ready = DEVICE_MAC_LEARNING;
}
}
static inline void __attribute__((always_inline))
virtio_xmit(struct vhost_dev *dst_vdev, struct vhost_dev *src_vdev,
{
uint16_t ret;
if (builtin_net_driver) {
ret = vs_enqueue_pkts(dst_vdev, VIRTIO_RXQ, &m, 1);
} else {
}
if (enable_stats) {
src_vdev->stats.tx_total++;
src_vdev->stats.tx += ret;
}
}
static inline int __attribute__((always_inline))
virtio_tx_local(struct vhost_dev *vdev, struct
rte_mbuf *m)
{
struct vhost_dev *dst_vdev;
dst_vdev = find_vhost_dev(&pkt_hdr->
d_addr);
if (!dst_vdev)
return -1;
if (vdev->vid == dst_vdev->vid) {
"(%d) TX: src and dst MAC is same. Dropping packet.\n",
vdev->vid);
return 0;
}
"(%d) TX: MAC address is local\n", dst_vdev->vid);
"(%d) device is marked for removal\n", dst_vdev->vid);
return 0;
}
virtio_xmit(dst_vdev, vdev, m);
return 0;
}
static inline int __attribute__((always_inline))
find_local_dest(struct vhost_dev *vdev, struct
rte_mbuf *m,
uint32_t *offset, uint16_t *vlan_tag)
{
struct vhost_dev *dst_vdev;
dst_vdev = find_vhost_dev(&pkt_hdr->
d_addr);
if (!dst_vdev)
return 0;
if (vdev->vid == dst_vdev->vid) {
"(%d) TX: src and dst MAC is same. Dropping packet.\n",
vdev->vid);
return -1;
}
*offset = VLAN_HLEN;
*vlan_tag = vlan_tags[vdev->vid];
"(%d) TX: pkt to local VM device id: (%d), vlan tag: %u.\n",
vdev->vid, dst_vdev->vid, *vlan_tag);
return 0;
}
static uint16_t
get_psd_sum(void *l3_hdr, uint64_t ol_flags)
{
else
}
static void virtio_tx_offload(
struct rte_mbuf *m)
{
void *l3_hdr;
l3_hdr = (
char *)eth_hdr + m->
l2_len;
ipv4_hdr = l3_hdr;
}
tcp_hdr = (
struct tcp_hdr *)((
char *)l3_hdr + m->
l3_len);
}
static inline void
free_pkts(
struct rte_mbuf **pkts, uint16_t n)
{
while (n--)
}
static inline void __attribute__((always_inline))
do_drain_mbuf_table(struct mbuf_table *tx_q)
{
uint16_t count;
tx_q->m_table, tx_q->len);
free_pkts(&tx_q->m_table[count], tx_q->len - count);
tx_q->len = 0;
}
static inline void __attribute__((always_inline))
virtio_tx_route(struct vhost_dev *vdev, struct
rte_mbuf *m, uint16_t vlan_tag)
{
struct mbuf_table *tx_q;
unsigned offset = 0;
struct vhost_dev *vdev2;
TAILQ_FOREACH(vdev2, &vhost_dev_list, global_vdev_entry) {
virtio_xmit(vdev2, vdev, m);
}
goto queue2nic;
}
if ((vm2vm_mode == VM2VM_SOFTWARE) && (virtio_tx_local(vdev, m) == 0)) {
return;
}
if (
unlikely(vm2vm_mode == VM2VM_HARDWARE)) {
if (
unlikely(find_local_dest(vdev, m, &offset,
&vlan_tag) != 0)) {
return;
}
}
"(%d) TX: MAC address is external\n", vdev->vid);
queue2nic:
tx_q = &lcore_tx_queue[lcore_id];
if ((vm2vm_mode == VM2VM_HARDWARE) &&
} else {
if (
unlikely(vm2vm_mode == VM2VM_HARDWARE)) {
m->data_len += offset;
else {
while ((seg->
next != NULL) &&
}
m->pkt_len += offset;
}
m->vlan_tci = vlan_tag;
}
virtio_tx_offload(m);
tx_q->m_table[tx_q->len++] = m;
if (enable_stats) {
vdev->stats.tx_total++;
vdev->stats.tx++;
}
if (
unlikely(tx_q->len == MAX_PKT_BURST))
do_drain_mbuf_table(tx_q);
}
static inline void __attribute__((always_inline))
drain_mbuf_table(struct mbuf_table *tx_q)
{
static uint64_t prev_tsc;
uint64_t cur_tsc;
if (tx_q->len == 0)
return;
cur_tsc = rte_rdtsc();
if (
unlikely(cur_tsc - prev_tsc > MBUF_TABLE_DRAIN_TSC)) {
prev_tsc = cur_tsc;
"TX queue drained after timeout with burst size %u\n",
tx_q->len);
do_drain_mbuf_table(tx_q);
}
}
static inline void __attribute__((always_inline))
drain_eth_rx(struct vhost_dev *vdev)
{
uint16_t rx_count, enqueue_count;
pkts, MAX_PKT_BURST);
if (!rx_count)
return;
if (enable_retry &&
VIRTIO_RXQ))) {
uint32_t retry;
for (retry = 0; retry < burst_rx_retry_num; retry++) {
VIRTIO_RXQ))
break;
}
}
if (builtin_net_driver) {
enqueue_count = vs_enqueue_pkts(vdev, VIRTIO_RXQ,
pkts, rx_count);
} else {
pkts, rx_count);
}
if (enable_stats) {
}
free_pkts(pkts, rx_count);
}
static inline void __attribute__((always_inline))
drain_virtio_tx(struct vhost_dev *vdev)
{
uint16_t count;
uint16_t i;
if (builtin_net_driver) {
count = vs_dequeue_pkts(vdev, VIRTIO_TXQ, mbuf_pool,
pkts, MAX_PKT_BURST);
} else {
mbuf_pool, pkts, MAX_PKT_BURST);
}
if (
unlikely(vdev->ready == DEVICE_MAC_LEARNING) && count) {
if (vdev->remove || link_vmdq(vdev, pkts[0]) == -1)
free_pkts(pkts, count);
}
for (i = 0; i < count; ++i)
virtio_tx_route(vdev, pkts[i], vlan_tags[vdev->vid]);
}
static int
{
unsigned i;
struct vhost_dev *vdev;
struct mbuf_table *tx_q;
RTE_LOG(INFO, VHOST_DATA,
"Procesing on Core %u started\n", lcore_id);
tx_q = &lcore_tx_queue[lcore_id];
if (lcore_ids[i] == lcore_id) {
tx_q->txq_id = i;
break;
}
}
while(1) {
drain_mbuf_table(tx_q);
if (lcore_info[lcore_id].dev_removal_flag == REQUEST_DEV_REMOVAL)
lcore_info[lcore_id].dev_removal_flag = ACK_DEV_REMOVAL;
TAILQ_FOREACH(vdev, &lcore_info[lcore_id].vdev_list,
lcore_vdev_entry) {
unlink_vmdq(vdev);
vdev->ready = DEVICE_SAFE_REMOVE;
continue;
}
if (
likely(vdev->ready == DEVICE_RX))
drain_eth_rx(vdev);
drain_virtio_tx(vdev);
}
}
return 0;
}
static void
destroy_device(int vid)
{
struct vhost_dev *vdev = NULL;
int lcore;
TAILQ_FOREACH(vdev, &vhost_dev_list, global_vdev_entry) {
if (vdev->vid == vid)
break;
}
if (!vdev)
return;
vdev->remove = 1;
while(vdev->ready != DEVICE_SAFE_REMOVE) {
rte_pause();
}
if (builtin_net_driver)
vs_vhost_net_remove(vdev);
TAILQ_REMOVE(&lcore_info[vdev->coreid].vdev_list, vdev,
lcore_vdev_entry);
TAILQ_REMOVE(&vhost_dev_list, vdev, global_vdev_entry);
lcore_info[lcore].dev_removal_flag = REQUEST_DEV_REMOVAL;
while (lcore_info[lcore].dev_removal_flag != ACK_DEV_REMOVAL)
rte_pause();
}
lcore_info[vdev->coreid].device_num--;
"(%d) device has been removed from data core\n",
vdev->vid);
}
static int
new_device(int vid)
{
int lcore, core_add = 0;
uint32_t device_num_min = num_devices;
struct vhost_dev *vdev;
vdev =
rte_zmalloc(
"vhost device",
sizeof(*vdev), RTE_CACHE_LINE_SIZE);
if (vdev == NULL) {
"(%d) couldn't allocate memory for vhost dev\n",
vid);
return -1;
}
vdev->vid = vid;
if (builtin_net_driver)
vs_vhost_net_setup(vdev);
TAILQ_INSERT_TAIL(&vhost_dev_list, vdev, global_vdev_entry);
vdev->vmdq_rx_q = vid * queues_per_pool + vmdq_queue_base;
vdev->ready = DEVICE_MAC_LEARNING;
vdev->remove = 0;
if (lcore_info[lcore].device_num < device_num_min) {
device_num_min = lcore_info[lcore].device_num;
core_add = lcore;
}
}
vdev->coreid = core_add;
TAILQ_INSERT_TAIL(&lcore_info[vdev->coreid].vdev_list, vdev,
lcore_vdev_entry);
lcore_info[vdev->coreid].device_num++;
rte_vhost_enable_guest_notification(vid, VIRTIO_RXQ, 0);
rte_vhost_enable_guest_notification(vid, VIRTIO_TXQ, 0);
"(%d) device has been added to data core %d\n",
vid, vdev->coreid);
return 0;
}
{
};
static void
print_stats(void)
{
struct vhost_dev *vdev;
uint64_t tx_dropped, rx_dropped;
uint64_t tx, tx_total, rx, rx_total;
const char clr[] = { 27, '[', '2', 'J', '\0' };
const char top_left[] = { 27, '[', '1', ';', '1', 'H','\0' };
while(1) {
sleep(enable_stats);
printf("%s%s\n", clr, top_left);
printf("Device statistics =================================\n");
TAILQ_FOREACH(vdev, &vhost_dev_list, global_vdev_entry) {
tx_total = vdev->stats.tx_total;
tx = vdev->stats.tx;
tx_dropped = tx_total - tx;
rx_dropped = rx_total - rx;
printf("Statistics for device %d\n"
"-----------------------\n"
"TX total: %" PRIu64 "\n"
"TX dropped: %" PRIu64 "\n"
"TX successful: %" PRIu64 "\n"
"RX total: %" PRIu64 "\n"
"RX dropped: %" PRIu64 "\n"
"RX successful: %" PRIu64 "\n",
vdev->vid,
tx_total, tx_dropped, tx,
rx_total, rx_dropped, rx);
}
printf("===================================================\n");
}
}
static void
unregister_drivers(int socket_num)
{
int i, ret;
for (i = 0; i < socket_num; i++) {
ret = rte_vhost_driver_unregister(socket_files + i * PATH_MAX);
if (ret != 0)
"Fail to unregister vhost driver for %s.\n",
socket_files + i * PATH_MAX);
}
}
static void
sigint_handler(__rte_unused int signum)
{
unregister_drivers(nb_sockets);
exit(0);
}
static void
create_mbuf_pool(uint16_t nr_port, uint32_t nr_switch_core, uint32_t mbuf_size,
uint32_t nr_queues, uint32_t nr_rx_desc, uint32_t nr_mbuf_cache)
{
uint32_t nr_mbufs;
uint32_t nr_mbufs_per_core;
uint32_t mtu = 1500;
if (mergeable)
mtu = 9000;
if (enable_tso)
mtu = 64 * 1024;
nr_mbufs_per_core = (mtu + mbuf_size) * MAX_PKT_BURST /
(mbuf_size - RTE_PKTMBUF_HEADROOM);
nr_mbufs_per_core += nr_rx_desc;
nr_mbufs_per_core =
RTE_MAX(nr_mbufs_per_core, nr_mbuf_cache);
nr_mbufs = nr_queues * nr_rx_desc;
nr_mbufs += nr_mbufs_per_core * nr_switch_core;
nr_mbufs *= nr_port;
nr_mbuf_cache, 0, mbuf_size,
if (mbuf_pool == NULL)
rte_exit(EXIT_FAILURE,
"Cannot create mbuf pool\n");
}
int
main(int argc, char *argv[])
{
unsigned lcore_id, core_id = 0;
unsigned nb_ports, valid_num_ports;
int ret, i;
uint8_t portid;
static pthread_t tid;
char thread_name[RTE_MAX_THREAD_NAME_LEN];
uint64_t flags = 0;
signal(SIGINT, sigint_handler);
if (ret < 0)
rte_exit(EXIT_FAILURE,
"Error with EAL initialization\n");
argc -= ret;
argv += ret;
ret = us_vhost_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++) {
TAILQ_INIT(&lcore_info[lcore_id].vdev_list);
lcore_ids[core_id++] = lcore_id;
}
rte_exit(EXIT_FAILURE,
"Not enough cores\n");
valid_num_ports = check_ports_num(nb_ports);
if ((valid_num_ports == 0) || (valid_num_ports > MAX_SUP_PORTS)) {
RTE_LOG(INFO, VHOST_PORT,
"Current enabled port number is %u,"
"but only %u port can be enabled\n",num_ports, MAX_SUP_PORTS);
return -1;
}
MAX_QUEUES, RTE_TEST_RX_DESC_DEFAULT, MBUF_CACHE_SIZE);
if (vm2vm_mode == VM2VM_HARDWARE) {
"Enable loop back for L2 switch in vmdq.\n");
}
for (portid = 0; portid < nb_ports; portid++) {
if ((enabled_port_mask & (1 << portid)) == 0) {
"Skipping disabled port %d\n", portid);
continue;
}
if (port_init(portid) != 0)
"Cannot initialize network ports\n");
}
if (enable_stats) {
ret = pthread_create(&tid, NULL, (void *)print_stats, NULL);
if (ret != 0)
"Cannot create print-stats thread\n");
snprintf(thread_name, RTE_MAX_THREAD_NAME_LEN, "print-stats");
if (ret != 0)
"Cannot set print-stats name\n");
}
if (client_mode)
flags |= RTE_VHOST_USER_CLIENT;
if (dequeue_zero_copy)
flags |= RTE_VHOST_USER_DEQUEUE_ZERO_COPY;
for (i = 0; i < nb_sockets; i++) {
char *file = socket_files + i * PATH_MAX;
if (ret != 0) {
unregister_drivers(i);
"vhost driver register failure.\n");
}
if (builtin_net_driver)
if (mergeable == 0) {
1ULL << VIRTIO_NET_F_MRG_RXBUF);
}
if (enable_tx_csum == 0) {
1ULL << VIRTIO_NET_F_CSUM);
}
if (enable_tso == 0) {
1ULL << VIRTIO_NET_F_HOST_TSO4);
1ULL << VIRTIO_NET_F_HOST_TSO6);
1ULL << VIRTIO_NET_F_GUEST_TSO4);
1ULL << VIRTIO_NET_F_GUEST_TSO6);
}
if (promiscuous) {
1ULL << VIRTIO_NET_F_CTRL_RX);
}
ret = rte_vhost_driver_callback_register(file,
&virtio_net_device_ops);
if (ret != 0) {
"failed to register vhost driver callbacks.\n");
}
"failed to start vhost driver.\n");
}
}
return 0;
}