#include <unistd.h>
#include <pthread.h>
#include <stdbool.h>
-#ifdef RTE_LIBRTE_VHOST_NUMA
-#include <numaif.h>
-#endif
#include <rte_mbuf.h>
#include <rte_ethdev.h>
#include "rte_eth_vhost.h"
+enum {VIRTIO_RXQ, VIRTIO_TXQ, VIRTIO_QNUM};
+
#define ETH_VHOST_IFACE_ARG "iface"
#define ETH_VHOST_QUEUES_ARG "queues"
#define ETH_VHOST_CLIENT_ARG "client"
#define ETH_VHOST_DEQUEUE_ZERO_COPY "dequeue-zero-copy"
-
-static const char *drivername = "VHOST PMD";
+#define VHOST_MAX_PKT_BURST 32
static const char *valid_arguments[] = {
ETH_VHOST_IFACE_ARG,
};
struct pmd_internal {
+ rte_atomic32_t dev_attached;
char *dev_name;
char *iface_name;
uint16_t max_queues;
- uint64_t flags;
-
- volatile uint16_t once;
+ rte_atomic32_t started;
};
struct internal_list {
static pthread_mutex_t internal_list_lock = PTHREAD_MUTEX_INITIALIZER;
-static rte_atomic16_t nb_started_ports;
-static pthread_t session_th;
-
static struct rte_eth_link pmd_link = {
.link_speed = 10000,
.link_duplex = ETH_LINK_FULL_DUPLEX,
*(uint64_t *)(((char *)vq)
+ vhost_rxport_stat_strings[t].offset);
}
+ xstats[count].id = count;
count++;
}
for (t = 0; t < VHOST_NB_XSTATS_TXPORT; t++) {
*(uint64_t *)(((char *)vq)
+ vhost_txport_stat_strings[t].offset);
}
+ xstats[count].id = count;
count++;
}
return count;
{
struct vhost_queue *r = q;
uint16_t i, nb_rx = 0;
+ uint16_t nb_receive = nb_bufs;
if (unlikely(rte_atomic32_read(&r->allow_queuing) == 0))
return 0;
goto out;
/* Dequeue packets from guest TX queue */
- nb_rx = rte_vhost_dequeue_burst(r->vid,
- r->virtqueue_id, r->mb_pool, bufs, nb_bufs);
+ while (nb_receive) {
+ uint16_t nb_pkts;
+ uint16_t num = (uint16_t)RTE_MIN(nb_receive,
+ VHOST_MAX_PKT_BURST);
+
+ nb_pkts = rte_vhost_dequeue_burst(r->vid, r->virtqueue_id,
+ r->mb_pool, &bufs[nb_rx],
+ num);
+
+ nb_rx += nb_pkts;
+ nb_receive -= nb_pkts;
+ if (nb_pkts < num)
+ break;
+ }
r->stats.pkts += nb_rx;
{
struct vhost_queue *r = q;
uint16_t i, nb_tx = 0;
+ uint16_t nb_send = nb_bufs;
if (unlikely(rte_atomic32_read(&r->allow_queuing) == 0))
return 0;
goto out;
/* Enqueue packets to guest RX queue */
- nb_tx = rte_vhost_enqueue_burst(r->vid,
- r->virtqueue_id, bufs, nb_bufs);
+ while (nb_send) {
+ uint16_t nb_pkts;
+ uint16_t num = (uint16_t)RTE_MIN(nb_send,
+ VHOST_MAX_PKT_BURST);
+
+ nb_pkts = rte_vhost_enqueue_burst(r->vid, r->virtqueue_id,
+ &bufs[nb_tx], num);
+
+ nb_tx += nb_pkts;
+ nb_send -= nb_pkts;
+ if (nb_pkts < num)
+ break;
+ }
r->stats.pkts += nb_tx;
r->stats.missed_pkts += nb_bufs - nb_tx;
return list;
}
+static void
+update_queuing_status(struct rte_eth_dev *dev)
+{
+ struct pmd_internal *internal = dev->data->dev_private;
+ struct vhost_queue *vq;
+ unsigned int i;
+ int allow_queuing = 1;
+
+ if (rte_atomic32_read(&internal->started) == 0 ||
+ rte_atomic32_read(&internal->dev_attached) == 0)
+ allow_queuing = 0;
+
+ /* Wait until rx/tx_pkt_burst stops accessing vhost device */
+ for (i = 0; i < dev->data->nb_rx_queues; i++) {
+ vq = dev->data->rx_queues[i];
+ if (vq == NULL)
+ continue;
+ rte_atomic32_set(&vq->allow_queuing, allow_queuing);
+ while (rte_atomic32_read(&vq->while_queuing))
+ rte_pause();
+ }
+
+ for (i = 0; i < dev->data->nb_tx_queues; i++) {
+ vq = dev->data->tx_queues[i];
+ if (vq == NULL)
+ continue;
+ rte_atomic32_set(&vq->allow_queuing, allow_queuing);
+ while (rte_atomic32_read(&vq->while_queuing))
+ rte_pause();
+ }
+}
+
static int
new_device(int vid)
{
vq->port = eth_dev->data->port_id;
}
- for (i = 0; i < rte_vhost_get_queue_num(vid) * VIRTIO_QNUM; i++)
+ for (i = 0; i < rte_vhost_get_vring_num(vid); i++)
rte_vhost_enable_guest_notification(vid, i, 0);
+ rte_vhost_get_mtu(vid, ð_dev->data->mtu);
+
eth_dev->data->dev_link.link_status = ETH_LINK_UP;
- for (i = 0; i < eth_dev->data->nb_rx_queues; i++) {
- vq = eth_dev->data->rx_queues[i];
- if (vq == NULL)
- continue;
- rte_atomic32_set(&vq->allow_queuing, 1);
- }
- for (i = 0; i < eth_dev->data->nb_tx_queues; i++) {
- vq = eth_dev->data->tx_queues[i];
- if (vq == NULL)
- continue;
- rte_atomic32_set(&vq->allow_queuing, 1);
- }
+ rte_atomic32_set(&internal->dev_attached, 1);
+ update_queuing_status(eth_dev);
RTE_LOG(INFO, PMD, "New connection established\n");
destroy_device(int vid)
{
struct rte_eth_dev *eth_dev;
+ struct pmd_internal *internal;
struct vhost_queue *vq;
struct internal_list *list;
char ifname[PATH_MAX];
return;
}
eth_dev = list->eth_dev;
+ internal = eth_dev->data->dev_private;
- /* Wait until rx/tx_pkt_burst stops accessing vhost device */
- for (i = 0; i < eth_dev->data->nb_rx_queues; i++) {
- vq = eth_dev->data->rx_queues[i];
- if (vq == NULL)
- continue;
- rte_atomic32_set(&vq->allow_queuing, 0);
- while (rte_atomic32_read(&vq->while_queuing))
- rte_pause();
- }
- for (i = 0; i < eth_dev->data->nb_tx_queues; i++) {
- vq = eth_dev->data->tx_queues[i];
- if (vq == NULL)
- continue;
- rte_atomic32_set(&vq->allow_queuing, 0);
- while (rte_atomic32_read(&vq->while_queuing))
- rte_pause();
- }
+ rte_atomic32_set(&internal->dev_attached, 0);
+ update_queuing_status(eth_dev);
eth_dev->data->dev_link.link_status = ETH_LINK_DOWN;
return 0;
}
+static struct vhost_device_ops vhost_ops = {
+ .new_device = new_device,
+ .destroy_device = destroy_device,
+ .vring_state_changed = vring_state_changed,
+};
+
int
rte_eth_vhost_get_queue_event(uint8_t port_id,
struct rte_eth_vhost_queue_event *event)
return vid;
}
-static void *
-vhost_driver_session(void *param __rte_unused)
-{
- static struct virtio_net_device_ops vhost_ops;
-
- /* set vhost arguments */
- vhost_ops.new_device = new_device;
- vhost_ops.destroy_device = destroy_device;
- vhost_ops.vring_state_changed = vring_state_changed;
- if (rte_vhost_driver_callback_register(&vhost_ops) < 0)
- RTE_LOG(ERR, PMD, "Can't register callbacks\n");
-
- /* start event handling */
- rte_vhost_driver_session_start();
-
- return NULL;
-}
-
static int
-vhost_driver_session_start(void)
+eth_dev_start(struct rte_eth_dev *dev)
{
- int ret;
+ struct pmd_internal *internal = dev->data->dev_private;
- ret = pthread_create(&session_th,
- NULL, vhost_driver_session, NULL);
- if (ret)
- RTE_LOG(ERR, PMD, "Can't create a thread\n");
+ rte_atomic32_set(&internal->started, 1);
+ update_queuing_status(dev);
- return ret;
+ return 0;
}
static void
-vhost_driver_session_stop(void)
+eth_dev_stop(struct rte_eth_dev *dev)
{
- int ret;
-
- ret = pthread_cancel(session_th);
- if (ret)
- RTE_LOG(ERR, PMD, "Can't cancel the thread\n");
+ struct pmd_internal *internal = dev->data->dev_private;
- ret = pthread_join(session_th, NULL);
- if (ret)
- RTE_LOG(ERR, PMD, "Can't join the thread\n");
+ rte_atomic32_set(&internal->started, 0);
+ update_queuing_status(dev);
}
-static int
-eth_dev_start(struct rte_eth_dev *dev)
+static void
+eth_dev_close(struct rte_eth_dev *dev)
{
- struct pmd_internal *internal = dev->data->dev_private;
- int ret = 0;
-
- if (rte_atomic16_cmpset(&internal->once, 0, 1)) {
- ret = rte_vhost_driver_register(internal->iface_name,
- internal->flags);
- if (ret)
- return ret;
- }
+ struct pmd_internal *internal;
+ struct internal_list *list;
- /* We need only one message handling thread */
- if (rte_atomic16_add_return(&nb_started_ports, 1) == 1)
- ret = vhost_driver_session_start();
+ internal = dev->data->dev_private;
+ if (!internal)
+ return;
- return ret;
-}
+ rte_vhost_driver_unregister(internal->iface_name);
-static void
-eth_dev_stop(struct rte_eth_dev *dev)
-{
- struct pmd_internal *internal = dev->data->dev_private;
+ list = find_internal_resource(internal->iface_name);
+ if (!list)
+ return;
- if (rte_atomic16_cmpset(&internal->once, 1, 0))
- rte_vhost_driver_unregister(internal->iface_name);
+ pthread_mutex_lock(&internal_list_lock);
+ TAILQ_REMOVE(&internal_list, list, next);
+ pthread_mutex_unlock(&internal_list_lock);
+ rte_free(list);
- if (rte_atomic16_sub_return(&nb_started_ports, 1) == 0)
- vhost_driver_session_stop();
+ free(internal->dev_name);
+ free(internal->iface_name);
+ rte_free(internal);
}
static int
return;
}
- dev_info->driver_name = drivername;
dev_info->max_mac_addrs = 1;
dev_info->max_rx_pktlen = (uint32_t)-1;
dev_info->max_rx_queues = internal->max_queues;
}
static int
-eth_link_update(struct rte_eth_dev *dev __rte_unused,
- int wait_to_complete __rte_unused)
+eth_tx_done_cleanup(void *txq __rte_unused, uint32_t free_cnt __rte_unused)
{
+ /*
+ * vHost does not hang onto mbuf. eth_vhost_tx() copies packet data
+ * and releases mbuf, so nothing to cleanup.
+ */
return 0;
}
-/**
- * Disable features in feature_mask. Returns 0 on success.
- */
-int
-rte_eth_vhost_feature_disable(uint64_t feature_mask)
-{
- return rte_vhost_feature_disable(feature_mask);
-}
-
-/**
- * Enable features in feature_mask. Returns 0 on success.
- */
-int
-rte_eth_vhost_feature_enable(uint64_t feature_mask)
-{
- return rte_vhost_feature_enable(feature_mask);
-}
-
-/* Returns currently supported vhost features */
-uint64_t
-rte_eth_vhost_feature_get(void)
+static int
+eth_link_update(struct rte_eth_dev *dev __rte_unused,
+ int wait_to_complete __rte_unused)
{
- return rte_vhost_feature_get();
+ return 0;
}
static const struct eth_dev_ops ops = {
.dev_start = eth_dev_start,
.dev_stop = eth_dev_stop,
+ .dev_close = eth_dev_close,
.dev_configure = eth_dev_configure,
.dev_infos_get = eth_dev_info,
.rx_queue_setup = eth_rx_queue_setup,
.tx_queue_setup = eth_tx_queue_setup,
.rx_queue_release = eth_queue_release,
.tx_queue_release = eth_queue_release,
+ .tx_done_cleanup = eth_tx_done_cleanup,
.link_update = eth_link_update,
.stats_get = eth_stats_get,
.stats_reset = eth_stats_reset,
.xstats_get_names = vhost_dev_xstats_get_names,
};
+static struct rte_vdev_driver pmd_vhost_drv;
+
static int
eth_dev_vhost_create(const char *name, char *iface_name, int16_t queues,
const unsigned numa_node, uint64_t flags)
if (vring_state == NULL)
goto error;
- TAILQ_INIT(ð_dev->link_intr_cbs);
-
/* now put it all together
* - store queue data in internal,
* - store numa_node info in ethdev data
internal->iface_name = strdup(iface_name);
if (internal->iface_name == NULL)
goto error;
- internal->flags = flags;
list->eth_dev = eth_dev;
pthread_mutex_lock(&internal_list_lock);
data->dev_flags =
RTE_ETH_DEV_DETACHABLE | RTE_ETH_DEV_INTR_LSC;
data->kdrv = RTE_KDRV_NONE;
- data->drv_name = internal->dev_name;
+ data->drv_name = pmd_vhost_drv.driver.name;
data->numa_node = numa_node;
/* finally assign rx and tx ops */
eth_dev->rx_pkt_burst = eth_vhost_rx;
eth_dev->tx_pkt_burst = eth_vhost_tx;
+ if (rte_vhost_driver_register(iface_name, flags))
+ goto error;
+
+ if (rte_vhost_driver_callback_register(iface_name, &vhost_ops) < 0) {
+ RTE_LOG(ERR, PMD, "Can't register callbacks\n");
+ goto error;
+ }
+
+ if (rte_vhost_driver_start(iface_name) < 0) {
+ RTE_LOG(ERR, PMD, "Failed to start driver for %s\n",
+ iface_name);
+ goto error;
+ }
+
return data->port_id;
error:
rte_pmd_vhost_remove(const char *name)
{
struct rte_eth_dev *eth_dev = NULL;
- struct pmd_internal *internal;
- struct internal_list *list;
unsigned int i;
RTE_LOG(INFO, PMD, "Un-Initializing pmd_vhost for %s\n", name);
if (eth_dev == NULL)
return -ENODEV;
- internal = eth_dev->data->dev_private;
- if (internal == NULL)
- return -ENODEV;
-
- list = find_internal_resource(internal->iface_name);
- if (list == NULL)
- return -ENODEV;
-
- pthread_mutex_lock(&internal_list_lock);
- TAILQ_REMOVE(&internal_list, list, next);
- pthread_mutex_unlock(&internal_list_lock);
- rte_free(list);
-
eth_dev_stop(eth_dev);
+ eth_dev_close(eth_dev);
+
rte_free(vring_states[eth_dev->data->port_id]);
vring_states[eth_dev->data->port_id] = NULL;
- free(internal->dev_name);
- free(internal->iface_name);
-
for (i = 0; i < eth_dev->data->nb_rx_queues; i++)
rte_free(eth_dev->data->rx_queues[i]);
for (i = 0; i < eth_dev->data->nb_tx_queues; i++)
rte_free(eth_dev->data->mac_addrs);
rte_free(eth_dev->data);
- rte_free(internal);
rte_eth_dev_release_port(eth_dev);
};
RTE_PMD_REGISTER_VDEV(net_vhost, pmd_vhost_drv);
+RTE_PMD_REGISTER_ALIAS(net_vhost, eth_vhost);
RTE_PMD_REGISTER_PARAM_STRING(net_vhost,
"iface=<ifc> "
"queues=<int>");
git.droids-corp.org / dpdk.git / blobdiff