#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_net.h>
#include <rte_vhost.h>
#include <rte_ip.h>
#include <rte_tcp.h>
#include <rte_pause.h>
+#include "ioat.h"
#include "main.h"
#ifndef MAX_QUEUES
#define BURST_RX_RETRIES 4 /* Number of retries on RX. */
#define JUMBO_FRAME_MAX_SIZE 0x2600
+#define MAX_MTU (JUMBO_FRAME_MAX_SIZE - (RTE_ETHER_HDR_LEN + RTE_ETHER_CRC_LEN))
/* State of virtio device. */
#define DEVICE_MAC_LEARNING 0
#define INVALID_PORT_ID 0xFF
-/* Maximum long option length for option parsing. */
-#define MAX_LONG_OPT_SZ 64
-
/* mask of enabled ports */
static uint32_t enabled_port_mask = 0;
static uint32_t enable_tso;
static int client_mode;
-static int dequeue_zero_copy;
static int builtin_net_driver;
+static int async_vhost_driver;
+
+static char *dma_type;
+
/* 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. */
/* empty vmdq configuration structure. Filled in programatically */
static struct rte_eth_conf vmdq_conf_default = {
.rxmode = {
- .mq_mode = ETH_MQ_RX_VMDQ_ONLY,
+ .mq_mode = RTE_ETH_MQ_RX_VMDQ_ONLY,
.split_hdr_size = 0,
/*
* VLAN strip is necessary for 1G NIC such as I350,
* this fixes bug of ipv4 forwarding in guest can't
* forward pakets from one virtio dev to another virtio dev.
*/
- .offloads = DEV_RX_OFFLOAD_VLAN_STRIP,
+ .offloads = RTE_ETH_RX_OFFLOAD_VLAN_STRIP,
},
.txmode = {
- .mq_mode = ETH_MQ_TX_NONE,
- .offloads = (DEV_TX_OFFLOAD_IPV4_CKSUM |
- DEV_TX_OFFLOAD_TCP_CKSUM |
- DEV_TX_OFFLOAD_VLAN_INSERT |
- DEV_TX_OFFLOAD_MULTI_SEGS |
- DEV_TX_OFFLOAD_TCP_TSO),
+ .mq_mode = RTE_ETH_MQ_TX_NONE,
+ .offloads = (RTE_ETH_TX_OFFLOAD_IPV4_CKSUM |
+ RTE_ETH_TX_OFFLOAD_TCP_CKSUM |
+ RTE_ETH_TX_OFFLOAD_VLAN_INSERT |
+ RTE_ETH_TX_OFFLOAD_MULTI_SEGS |
+ RTE_ETH_TX_OFFLOAD_TCP_TSO),
},
.rx_adv_conf = {
/*
* appropriate values
*/
.vmdq_rx_conf = {
- .nb_queue_pools = ETH_8_POOLS,
+ .nb_queue_pools = RTE_ETH_8_POOLS,
.enable_default_pool = 0,
.default_pool = 0,
.nb_pool_maps = 0,
struct rte_mbuf *m_table[MAX_PKT_BURST];
};
+struct vhost_bufftable {
+ uint32_t len;
+ uint64_t pre_tsc;
+ struct rte_mbuf *m_table[MAX_PKT_BURST];
+};
+
/* TX queue for each data core. */
struct mbuf_table lcore_tx_queue[RTE_MAX_LCORE];
+/*
+ * Vhost TX buffer for each data core.
+ * Every data core maintains a TX buffer for every vhost device,
+ * which is used for batch pkts enqueue for higher performance.
+ */
+struct vhost_bufftable *vhost_txbuff[RTE_MAX_LCORE * MAX_VHOST_DEVICE];
+
#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
+open_dma(const char *value)
+{
+ if (dma_type != NULL && strncmp(dma_type, "ioat", 4) == 0)
+ return open_ioat(value);
+
+ return -1;
+}
+
/*
* Builds up the correct configuration for VMDQ VLAN pool map
* according to the pool & queue limits.
rx_ring_size = RTE_TEST_RX_DESC_DEFAULT;
tx_ring_size = RTE_TEST_TX_DESC_DEFAULT;
- /*
- * When dequeue zero copy is enabled, guest Tx used vring will be
- * updated only when corresponding mbuf is freed. Thus, the nb_tx_desc
- * (tx_ring_size here) must be small enough so that the driver will
- * hit the free threshold easily and free mbufs timely. Otherwise,
- * guest Tx vring would be starved.
- */
- if (dequeue_zero_copy)
- tx_ring_size = 64;
-
tx_rings = (uint16_t)rte_lcore_count();
/* Get port configuration. */
return -1;
rx_rings = (uint16_t)dev_info.max_rx_queues;
- if (dev_info.tx_offload_capa & DEV_TX_OFFLOAD_MBUF_FAST_FREE)
+ if (dev_info.tx_offload_capa & RTE_ETH_TX_OFFLOAD_MBUF_FAST_FREE)
port_conf.txmode.offloads |=
- DEV_TX_OFFLOAD_MBUF_FAST_FREE;
+ RTE_ETH_TX_OFFLOAD_MBUF_FAST_FREE;
/* Configure ethernet device. */
retval = rte_eth_dev_configure(port, rx_rings, tx_rings, &port_conf);
if (retval != 0) {
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",
- 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]);
+ " %02"PRIx8" %02"PRIx8" %02"PRIx8"\n",
+ port, RTE_ETHER_ADDR_BYTES(&vmdq_ports_eth_addr[port]));
return 0;
}
/* parse hexadecimal string */
pm = strtoul(portmask, &end, 16);
if ((portmask[0] == '\0') || (end == NULL) || (*end != '\0') || (errno != 0))
- return -1;
-
- if (pm == 0)
- return -1;
+ return 0;
return pm;
" --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",
+ " --dma-type register dma type for your vhost async driver. For example \"ioat\" for now.\n"
+ " --dmas register dma channel for specific vhost device.\n",
prgname);
}
+enum {
+#define OPT_VM2VM "vm2vm"
+ OPT_VM2VM_NUM = 256,
+#define OPT_RX_RETRY "rx-retry"
+ OPT_RX_RETRY_NUM,
+#define OPT_RX_RETRY_DELAY "rx-retry-delay"
+ OPT_RX_RETRY_DELAY_NUM,
+#define OPT_RX_RETRY_NUMB "rx-retry-num"
+ OPT_RX_RETRY_NUMB_NUM,
+#define OPT_MERGEABLE "mergeable"
+ OPT_MERGEABLE_NUM,
+#define OPT_STATS "stats"
+ OPT_STATS_NUM,
+#define OPT_SOCKET_FILE "socket-file"
+ OPT_SOCKET_FILE_NUM,
+#define OPT_TX_CSUM "tx-csum"
+ OPT_TX_CSUM_NUM,
+#define OPT_TSO "tso"
+ OPT_TSO_NUM,
+#define OPT_CLIENT "client"
+ OPT_CLIENT_NUM,
+#define OPT_BUILTIN_NET_DRIVER "builtin-net-driver"
+ OPT_BUILTIN_NET_DRIVER_NUM,
+#define OPT_DMA_TYPE "dma-type"
+ OPT_DMA_TYPE_NUM,
+#define OPT_DMAS "dmas"
+ OPT_DMAS_NUM,
+};
+
/*
* Parse the arguments given in the command line of the application.
*/
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},
+ {OPT_VM2VM, required_argument,
+ NULL, OPT_VM2VM_NUM},
+ {OPT_RX_RETRY, required_argument,
+ NULL, OPT_RX_RETRY_NUM},
+ {OPT_RX_RETRY_DELAY, required_argument,
+ NULL, OPT_RX_RETRY_DELAY_NUM},
+ {OPT_RX_RETRY_NUMB, required_argument,
+ NULL, OPT_RX_RETRY_NUMB_NUM},
+ {OPT_MERGEABLE, required_argument,
+ NULL, OPT_MERGEABLE_NUM},
+ {OPT_STATS, required_argument,
+ NULL, OPT_STATS_NUM},
+ {OPT_SOCKET_FILE, required_argument,
+ NULL, OPT_SOCKET_FILE_NUM},
+ {OPT_TX_CSUM, required_argument,
+ NULL, OPT_TX_CSUM_NUM},
+ {OPT_TSO, required_argument,
+ NULL, OPT_TSO_NUM},
+ {OPT_CLIENT, no_argument,
+ NULL, OPT_CLIENT_NUM},
+ {OPT_BUILTIN_NET_DRIVER, no_argument,
+ NULL, OPT_BUILTIN_NET_DRIVER_NUM},
+ {OPT_DMA_TYPE, required_argument,
+ NULL, OPT_DMA_TYPE_NUM},
+ {OPT_DMAS, required_argument,
+ NULL, OPT_DMAS_NUM},
{NULL, 0, 0, 0},
};
case 'P':
promiscuous = 1;
vmdq_conf_default.rx_adv_conf.vmdq_rx_conf.rx_mode =
- ETH_VMDQ_ACCEPT_BROADCAST |
- ETH_VMDQ_ACCEPT_MULTICAST;
+ RTE_ETH_VMDQ_ACCEPT_BROADCAST |
+ RTE_ETH_VMDQ_ACCEPT_MULTICAST;
+ break;
+ case OPT_VM2VM_NUM:
+ ret = parse_num_opt(optarg, (VM2VM_LAST - 1));
+ if (ret == -1) {
+ RTE_LOG(INFO, VHOST_CONFIG,
+ "Invalid argument for "
+ "vm2vm [0|1|2]\n");
+ us_vhost_usage(prgname);
+ return -1;
+ }
+ vm2vm_mode = (vm2vm_type)ret;
break;
- case 0:
- /* Enable/disable vm2vm comms. */
- if (!strncmp(long_option[option_index].name, "vm2vm",
- MAX_LONG_OPT_SZ)) {
- ret = parse_num_opt(optarg, (VM2VM_LAST - 1));
- if (ret == -1) {
- RTE_LOG(INFO, VHOST_CONFIG,
- "Invalid argument for "
- "vm2vm [0|1|2]\n");
- us_vhost_usage(prgname);
- return -1;
- } else {
- vm2vm_mode = (vm2vm_type)ret;
- }
+ case OPT_RX_RETRY_NUM:
+ 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;
}
+ enable_retry = ret;
+ break;
- /* Enable/disable retries on RX. */
- 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;
- }
+ case OPT_TX_CSUM_NUM:
+ 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;
}
+ enable_tx_csum = ret;
+ break;
- /* Enable/disable TX checksum offload. */
- 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;
+ case OPT_TSO_NUM:
+ 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;
}
+ enable_tso = ret;
+ break;
- /* Enable/disable TSO offload. */
- 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;
+ case OPT_RX_RETRY_DELAY_NUM:
+ 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;
}
+ burst_rx_delay_time = ret;
+ break;
- /* Specify the retries delay time (in useconds) on RX. */
- 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;
- }
+ case OPT_RX_RETRY_NUMB_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");
+ us_vhost_usage(prgname);
+ return -1;
}
+ burst_rx_retry_num = ret;
+ break;
- /* Specify the retries number on RX. */
- 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;
- }
+ case OPT_MERGEABLE_NUM:
+ 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;
}
+ mergeable = !!ret;
+ if (ret)
+ vmdq_conf_default.rxmode.mtu = MAX_MTU;
+ break;
- /* Enable/disable RX mergeable buffers. */
- 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) {
- vmdq_conf_default.rxmode.offloads |=
- DEV_RX_OFFLOAD_JUMBO_FRAME;
- vmdq_conf_default.rxmode.max_rx_pkt_len
- = JUMBO_FRAME_MAX_SIZE;
- }
- }
+ case OPT_STATS_NUM:
+ ret = parse_num_opt(optarg, INT32_MAX);
+ if (ret == -1) {
+ RTE_LOG(INFO, VHOST_CONFIG,
+ "Invalid argument for stats [0..N]\n");
+ us_vhost_usage(prgname);
+ return -1;
}
+ enable_stats = ret;
+ break;
- /* Enable/disable stats. */
- if (!strncmp(long_option[option_index].name, "stats", MAX_LONG_OPT_SZ)) {
- ret = parse_num_opt(optarg, INT32_MAX);
- if (ret == -1) {
- RTE_LOG(INFO, VHOST_CONFIG,
- "Invalid argument for stats [0..N]\n");
- us_vhost_usage(prgname);
- return -1;
- } else {
- enable_stats = ret;
- }
+ /* Set socket file path. */
+ case OPT_SOCKET_FILE_NUM:
+ if (us_vhost_parse_socket_path(optarg) == -1) {
+ RTE_LOG(INFO, VHOST_CONFIG,
+ "Invalid argument for socket name (Max %d characters)\n",
+ PATH_MAX);
+ us_vhost_usage(prgname);
+ return -1;
}
+ break;
+
+ case OPT_DMA_TYPE_NUM:
+ dma_type = optarg;
+ break;
- /* Set socket file path. */
- if (!strncmp(long_option[option_index].name,
- "socket-file", MAX_LONG_OPT_SZ)) {
- if (us_vhost_parse_socket_path(optarg) == -1) {
- RTE_LOG(INFO, VHOST_CONFIG,
- "Invalid argument for socket name (Max %d characters)\n",
- PATH_MAX);
- us_vhost_usage(prgname);
- return -1;
- }
+ case OPT_DMAS_NUM:
+ if (open_dma(optarg) == -1) {
+ RTE_LOG(INFO, VHOST_CONFIG,
+ "Wrong DMA args\n");
+ us_vhost_usage(prgname);
+ return -1;
}
+ async_vhost_driver = 1;
+ break;
+
+ case OPT_CLIENT_NUM:
+ client_mode = 1;
+ break;
+ case OPT_BUILTIN_NET_DRIVER_NUM:
+ builtin_net_driver = 1;
break;
- /* Invalid option - print options. */
+ /* Invalid option - print options. */
default:
us_vhost_usage(prgname);
return -1;
/* Learn MAC address of guest device from packet */
pkt_hdr = rte_pktmbuf_mtod(m, struct rte_ether_hdr *);
- if (find_vhost_dev(&pkt_hdr->s_addr)) {
+ if (find_vhost_dev(&pkt_hdr->src_addr)) {
RTE_LOG(ERR, VHOST_DATA,
"(%d) device is using a registered MAC!\n",
vdev->vid);
}
for (i = 0; i < RTE_ETHER_ADDR_LEN; i++)
- vdev->mac_address.addr_bytes[i] = pkt_hdr->s_addr.addr_bytes[i];
+ vdev->mac_address.addr_bytes[i] =
+ pkt_hdr->src_addr.addr_bytes[i];
/* vlan_tag currently uses the device_id. */
vdev->vlan_tag = vlan_tags[vdev->vid];
/* Print out VMDQ registration info. */
RTE_LOG(INFO, VHOST_DATA,
- "(%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],
+ "(%d) mac " RTE_ETHER_ADDR_PRT_FMT " and vlan %d registered\n",
+ vdev->vid, RTE_ETHER_ADDR_BYTES(&vdev->mac_address),
vdev->vlan_tag);
/* Register the MAC address. */
}
}
+static inline void
+free_pkts(struct rte_mbuf **pkts, uint16_t n)
+{
+ while (n--)
+ rte_pktmbuf_free(pkts[n]);
+}
+
+static __rte_always_inline void
+complete_async_pkts(struct vhost_dev *vdev)
+{
+ struct rte_mbuf *p_cpl[MAX_PKT_BURST];
+ uint16_t complete_count;
+
+ complete_count = rte_vhost_poll_enqueue_completed(vdev->vid,
+ VIRTIO_RXQ, p_cpl, MAX_PKT_BURST);
+ if (complete_count) {
+ free_pkts(p_cpl, complete_count);
+ __atomic_sub_fetch(&vdev->pkts_inflight, complete_count, __ATOMIC_SEQ_CST);
+ }
+
+}
+
static __rte_always_inline void
-virtio_xmit(struct vhost_dev *dst_vdev, struct vhost_dev *src_vdev,
+sync_virtio_xmit(struct vhost_dev *dst_vdev, struct vhost_dev *src_vdev,
struct rte_mbuf *m)
{
uint16_t ret;
}
if (enable_stats) {
- rte_atomic64_inc(&dst_vdev->stats.rx_total_atomic);
- rte_atomic64_add(&dst_vdev->stats.rx_atomic, ret);
+ __atomic_add_fetch(&dst_vdev->stats.rx_total_atomic, 1,
+ __ATOMIC_SEQ_CST);
+ __atomic_add_fetch(&dst_vdev->stats.rx_atomic, ret,
+ __ATOMIC_SEQ_CST);
src_vdev->stats.tx_total++;
src_vdev->stats.tx += ret;
}
}
+static __rte_always_inline void
+drain_vhost(struct vhost_dev *vdev)
+{
+ uint16_t ret;
+ uint32_t buff_idx = rte_lcore_id() * MAX_VHOST_DEVICE + vdev->vid;
+ uint16_t nr_xmit = vhost_txbuff[buff_idx]->len;
+ struct rte_mbuf **m = vhost_txbuff[buff_idx]->m_table;
+
+ if (builtin_net_driver) {
+ ret = vs_enqueue_pkts(vdev, VIRTIO_RXQ, m, nr_xmit);
+ } else if (async_vhost_driver) {
+ uint16_t enqueue_fail = 0;
+
+ complete_async_pkts(vdev);
+ ret = rte_vhost_submit_enqueue_burst(vdev->vid, VIRTIO_RXQ, m, nr_xmit);
+ __atomic_add_fetch(&vdev->pkts_inflight, ret, __ATOMIC_SEQ_CST);
+
+ enqueue_fail = nr_xmit - ret;
+ if (enqueue_fail)
+ free_pkts(&m[ret], nr_xmit - ret);
+ } else {
+ ret = rte_vhost_enqueue_burst(vdev->vid, VIRTIO_RXQ,
+ m, nr_xmit);
+ }
+
+ if (enable_stats) {
+ __atomic_add_fetch(&vdev->stats.rx_total_atomic, nr_xmit,
+ __ATOMIC_SEQ_CST);
+ __atomic_add_fetch(&vdev->stats.rx_atomic, ret,
+ __ATOMIC_SEQ_CST);
+ }
+
+ if (!async_vhost_driver)
+ free_pkts(m, nr_xmit);
+}
+
+static __rte_always_inline void
+drain_vhost_table(void)
+{
+ uint16_t lcore_id = rte_lcore_id();
+ struct vhost_bufftable *vhost_txq;
+ struct vhost_dev *vdev;
+ uint64_t cur_tsc;
+
+ TAILQ_FOREACH(vdev, &vhost_dev_list, global_vdev_entry) {
+ if (unlikely(vdev->remove == 1))
+ continue;
+
+ vhost_txq = vhost_txbuff[lcore_id * MAX_VHOST_DEVICE
+ + vdev->vid];
+
+ cur_tsc = rte_rdtsc();
+ if (unlikely(cur_tsc - vhost_txq->pre_tsc
+ > MBUF_TABLE_DRAIN_TSC)) {
+ RTE_LOG_DP(DEBUG, VHOST_DATA,
+ "Vhost TX queue drained after timeout with burst size %u\n",
+ vhost_txq->len);
+ drain_vhost(vdev);
+ vhost_txq->len = 0;
+ vhost_txq->pre_tsc = cur_tsc;
+ }
+ }
+}
+
/*
* Check if the packet destination MAC address is for a local device. If so then put
* the packet on that devices RX queue. If not then return.
{
struct rte_ether_hdr *pkt_hdr;
struct vhost_dev *dst_vdev;
-
+ struct vhost_bufftable *vhost_txq;
+ uint16_t lcore_id = rte_lcore_id();
pkt_hdr = rte_pktmbuf_mtod(m, struct rte_ether_hdr *);
- dst_vdev = find_vhost_dev(&pkt_hdr->d_addr);
+ dst_vdev = find_vhost_dev(&pkt_hdr->dst_addr);
if (!dst_vdev)
return -1;
return 0;
}
- virtio_xmit(dst_vdev, vdev, m);
+ vhost_txq = vhost_txbuff[lcore_id * MAX_VHOST_DEVICE + dst_vdev->vid];
+ vhost_txq->m_table[vhost_txq->len++] = m;
+
+ if (enable_stats) {
+ vdev->stats.tx_total++;
+ vdev->stats.tx++;
+ }
+
+ if (unlikely(vhost_txq->len == MAX_PKT_BURST)) {
+ drain_vhost(dst_vdev);
+ vhost_txq->len = 0;
+ vhost_txq->pre_tsc = rte_rdtsc();
+ }
return 0;
}
struct rte_ether_hdr *pkt_hdr =
rte_pktmbuf_mtod(m, struct rte_ether_hdr *);
- dst_vdev = find_vhost_dev(&pkt_hdr->d_addr);
+ dst_vdev = find_vhost_dev(&pkt_hdr->dst_addr);
if (!dst_vdev)
return 0;
return 0;
}
-static uint16_t
-get_psd_sum(void *l3_hdr, uint64_t ol_flags)
-{
- if (ol_flags & PKT_TX_IPV4)
- return rte_ipv4_phdr_cksum(l3_hdr, ol_flags);
- else /* assume ethertype == RTE_ETHER_TYPE_IPV6 */
- return rte_ipv6_phdr_cksum(l3_hdr, ol_flags);
-}
-
static void virtio_tx_offload(struct rte_mbuf *m)
{
+ struct rte_net_hdr_lens hdr_lens;
+ struct rte_ipv4_hdr *ipv4_hdr;
+ struct rte_tcp_hdr *tcp_hdr;
+ uint32_t ptype;
void *l3_hdr;
- struct rte_ipv4_hdr *ipv4_hdr = NULL;
- struct rte_tcp_hdr *tcp_hdr = NULL;
- struct rte_ether_hdr *eth_hdr =
- rte_pktmbuf_mtod(m, struct rte_ether_hdr *);
- l3_hdr = (char *)eth_hdr + m->l2_len;
+ ptype = rte_net_get_ptype(m, &hdr_lens, RTE_PTYPE_ALL_MASK);
+ m->l2_len = hdr_lens.l2_len;
+ m->l3_len = hdr_lens.l3_len;
+ m->l4_len = hdr_lens.l4_len;
- if (m->ol_flags & PKT_TX_IPV4) {
+ l3_hdr = rte_pktmbuf_mtod_offset(m, void *, m->l2_len);
+ tcp_hdr = rte_pktmbuf_mtod_offset(m, struct rte_tcp_hdr *,
+ m->l2_len + m->l3_len);
+
+ m->ol_flags |= PKT_TX_TCP_SEG;
+ if ((ptype & RTE_PTYPE_L3_MASK) == RTE_PTYPE_L3_IPV4) {
+ m->ol_flags |= PKT_TX_IPV4;
+ m->ol_flags |= PKT_TX_IP_CKSUM;
ipv4_hdr = l3_hdr;
ipv4_hdr->hdr_checksum = 0;
- m->ol_flags |= PKT_TX_IP_CKSUM;
+ tcp_hdr->cksum = rte_ipv4_phdr_cksum(l3_hdr, m->ol_flags);
+ } else { /* assume ethertype == RTE_ETHER_TYPE_IPV6 */
+ m->ol_flags |= PKT_TX_IPV6;
+ tcp_hdr->cksum = rte_ipv6_phdr_cksum(l3_hdr, m->ol_flags);
}
-
- tcp_hdr = (struct rte_tcp_hdr *)((char *)l3_hdr + m->l3_len);
- tcp_hdr->cksum = get_psd_sum(l3_hdr, m->ol_flags);
-}
-
-static inline void
-free_pkts(struct rte_mbuf **pkts, uint16_t n)
-{
- while (n--)
- rte_pktmbuf_free(pkts[n]);
}
static __rte_always_inline void
nh = rte_pktmbuf_mtod(m, struct rte_ether_hdr *);
- if (unlikely(rte_is_broadcast_ether_addr(&nh->d_addr))) {
+ if (unlikely(rte_is_broadcast_ether_addr(&nh->dst_addr))) {
struct vhost_dev *vdev2;
TAILQ_FOREACH(vdev2, &vhost_dev_list, global_vdev_entry) {
if (vdev2 != vdev)
- virtio_xmit(vdev2, vdev, m);
+ sync_virtio_xmit(vdev2, vdev, m);
}
goto queue2nic;
}
/*check if destination is local VM*/
- if ((vm2vm_mode == VM2VM_SOFTWARE) && (virtio_tx_local(vdev, m) == 0)) {
- rte_pktmbuf_free(m);
+ 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,
m->vlan_tci = vlan_tag;
}
- if (m->ol_flags & PKT_TX_TCP_SEG)
+ if (m->ol_flags & PKT_RX_LRO)
virtio_tx_offload(m);
tx_q->m_table[tx_q->len++] = m;
rx_count = rte_eth_rx_burst(ports[0], vdev->vmdq_rx_q,
pkts, MAX_PKT_BURST);
+
if (!rx_count)
return;
if (builtin_net_driver) {
enqueue_count = vs_enqueue_pkts(vdev, VIRTIO_RXQ,
pkts, rx_count);
+ } else if (async_vhost_driver) {
+ uint16_t enqueue_fail = 0;
+
+ complete_async_pkts(vdev);
+ enqueue_count = rte_vhost_submit_enqueue_burst(vdev->vid,
+ VIRTIO_RXQ, pkts, rx_count);
+ __atomic_add_fetch(&vdev->pkts_inflight, enqueue_count, __ATOMIC_SEQ_CST);
+
+ enqueue_fail = rx_count - enqueue_count;
+ if (enqueue_fail)
+ free_pkts(&pkts[enqueue_count], enqueue_fail);
+
} else {
enqueue_count = rte_vhost_enqueue_burst(vdev->vid, VIRTIO_RXQ,
pkts, rx_count);
}
+
if (enable_stats) {
- rte_atomic64_add(&vdev->stats.rx_total_atomic, rx_count);
- rte_atomic64_add(&vdev->stats.rx_atomic, enqueue_count);
+ __atomic_add_fetch(&vdev->stats.rx_total_atomic, rx_count,
+ __ATOMIC_SEQ_CST);
+ __atomic_add_fetch(&vdev->stats.rx_atomic, enqueue_count,
+ __ATOMIC_SEQ_CST);
}
- free_pkts(pkts, rx_count);
+ if (!async_vhost_driver)
+ free_pkts(pkts, rx_count);
}
static __rte_always_inline void
while(1) {
drain_mbuf_table(tx_q);
-
+ drain_vhost_table();
/*
* Inform the configuration core that we have exited the
* linked list and that no devices are in use if requested.
{
struct vhost_dev *vdev = NULL;
int lcore;
+ uint16_t i;
TAILQ_FOREACH(vdev, &vhost_dev_list, global_vdev_entry) {
if (vdev->vid == vid)
rte_pause();
}
+ for (i = 0; i < RTE_MAX_LCORE; i++)
+ rte_free(vhost_txbuff[i * MAX_VHOST_DEVICE + vid]);
+
if (builtin_net_driver)
vs_vhost_net_remove(vdev);
/* Set the dev_removal_flag on each lcore. */
- RTE_LCORE_FOREACH_SLAVE(lcore)
+ RTE_LCORE_FOREACH_WORKER(lcore)
lcore_info[lcore].dev_removal_flag = REQUEST_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) {
+ RTE_LCORE_FOREACH_WORKER(lcore) {
while (lcore_info[lcore].dev_removal_flag != ACK_DEV_REMOVAL)
rte_pause();
}
"(%d) device has been removed from data core\n",
vdev->vid);
+ if (async_vhost_driver) {
+ uint16_t n_pkt = 0;
+ struct rte_mbuf *m_cpl[vdev->pkts_inflight];
+
+ while (vdev->pkts_inflight) {
+ n_pkt = rte_vhost_clear_queue_thread_unsafe(vid, VIRTIO_RXQ,
+ m_cpl, vdev->pkts_inflight);
+ free_pkts(m_cpl, n_pkt);
+ __atomic_sub_fetch(&vdev->pkts_inflight, n_pkt, __ATOMIC_SEQ_CST);
+ }
+
+ rte_vhost_async_channel_unregister(vid, VIRTIO_RXQ);
+ }
+
rte_free(vdev);
}
new_device(int vid)
{
int lcore, core_add = 0;
+ uint16_t i;
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) {
RTE_LOG(INFO, VHOST_DATA,
}
vdev->vid = vid;
+ for (i = 0; i < RTE_MAX_LCORE; i++) {
+ vhost_txbuff[i * MAX_VHOST_DEVICE + vid]
+ = rte_zmalloc("vhost bufftable",
+ sizeof(struct vhost_bufftable),
+ RTE_CACHE_LINE_SIZE);
+
+ if (vhost_txbuff[i * MAX_VHOST_DEVICE + vid] == NULL) {
+ RTE_LOG(INFO, VHOST_DATA,
+ "(%d) couldn't allocate memory for vhost TX\n", vid);
+ return -1;
+ }
+ }
+
if (builtin_net_driver)
vs_vhost_net_setup(vdev);
vdev->remove = 0;
/* Find a suitable lcore to add the device. */
- RTE_LCORE_FOREACH_SLAVE(lcore) {
+ RTE_LCORE_FOREACH_WORKER(lcore) {
if (lcore_info[lcore].device_num < device_num_min) {
device_num_min = lcore_info[lcore].device_num;
core_add = lcore;
"(%d) device has been added to data core %d\n",
vid, vdev->coreid);
+ if (async_vhost_driver) {
+ struct rte_vhost_async_config config = {0};
+ struct rte_vhost_async_channel_ops channel_ops;
+
+ if (dma_type != NULL && strncmp(dma_type, "ioat", 4) == 0) {
+ channel_ops.transfer_data = ioat_transfer_data_cb;
+ channel_ops.check_completed_copies =
+ ioat_check_completed_copies_cb;
+
+ config.features = RTE_VHOST_ASYNC_INORDER;
+
+ return rte_vhost_async_channel_register(vid, VIRTIO_RXQ,
+ config, &channel_ops);
+ }
+ }
+
+ return 0;
+}
+
+static int
+vring_state_changed(int vid, uint16_t queue_id, int enable)
+{
+ struct vhost_dev *vdev = NULL;
+
+ TAILQ_FOREACH(vdev, &vhost_dev_list, global_vdev_entry) {
+ if (vdev->vid == vid)
+ break;
+ }
+ if (!vdev)
+ return -1;
+
+ if (queue_id != VIRTIO_RXQ)
+ return 0;
+
+ if (async_vhost_driver) {
+ if (!enable) {
+ uint16_t n_pkt = 0;
+ struct rte_mbuf *m_cpl[vdev->pkts_inflight];
+
+ while (vdev->pkts_inflight) {
+ n_pkt = rte_vhost_clear_queue_thread_unsafe(vid, queue_id,
+ m_cpl, vdev->pkts_inflight);
+ free_pkts(m_cpl, n_pkt);
+ __atomic_sub_fetch(&vdev->pkts_inflight, n_pkt, __ATOMIC_SEQ_CST);
+ }
+ }
+ }
+
return 0;
}
{
.new_device = new_device,
.destroy_device = destroy_device,
+ .vring_state_changed = vring_state_changed,
};
/*
tx = vdev->stats.tx;
tx_dropped = tx_total - tx;
- rx_total = rte_atomic64_read(&vdev->stats.rx_total_atomic);
- rx = rte_atomic64_read(&vdev->stats.rx_atomic);
+ rx_total = __atomic_load_n(&vdev->stats.rx_total_atomic,
+ __ATOMIC_SEQ_CST);
+ rx = __atomic_load_n(&vdev->stats.rx_atomic,
+ __ATOMIC_SEQ_CST);
rx_dropped = rx_total - rx;
printf("Statistics for device %d\n"
}
printf("===================================================\n");
+
+ fflush(stdout);
}
return NULL;
int ret, i;
uint16_t portid;
static pthread_t tid;
- uint64_t flags = 0;
+ uint64_t flags = RTE_VHOST_USER_NET_COMPLIANT_OL_FLAGS;
signal(SIGINT, sigint_handler);
}
/* Launch all data cores. */
- RTE_LCORE_FOREACH_SLAVE(lcore_id)
+ RTE_LCORE_FOREACH_WORKER(lcore_id)
rte_eal_remote_launch(switch_worker, NULL, lcore_id);
if (client_mode)
flags |= RTE_VHOST_USER_CLIENT;
- if (dequeue_zero_copy)
- flags |= RTE_VHOST_USER_DEQUEUE_ZERO_COPY;
-
/* Register vhost user driver to handle vhost messages. */
for (i = 0; i < nb_sockets; i++) {
char *file = socket_files + i * PATH_MAX;
+
+ if (async_vhost_driver)
+ flags = flags | RTE_VHOST_USER_ASYNC_COPY;
+
ret = rte_vhost_driver_register(file, flags);
if (ret != 0) {
unregister_drivers(i);
}
}
- RTE_LCORE_FOREACH_SLAVE(lcore_id)
+ RTE_LCORE_FOREACH_WORKER(lcore_id)
rte_eal_wait_lcore(lcore_id);
- return 0;
+ /* clean up the EAL */
+ rte_eal_cleanup();
+ return 0;
}
git.droids-corp.org / dpdk.git / blobdiff