int
rte_eth_bond_create(const char *name, uint8_t mode, uint8_t socket_id)
{
struct rte_pci_device *pci_dev = NULL;
struct bond_dev_private *internals = NULL;
struct rte_eth_dev *eth_dev = NULL;
struct eth_driver *eth_drv = NULL;
struct rte_pci_driver *pci_drv = NULL;
struct rte_pci_id *pci_id_table = NULL;
/* now do all data allocation - for eth_dev structure, dummy pci driver
* and internal (private) data
*/
if (name == NULL) {
RTE_BOND_LOG(ERR, "Invalid name specified");
goto err;
}
if (socket_id >= number_of_sockets()) {
RTE_BOND_LOG(ERR,
"Invalid socket id specified to create bonded device on.");
goto err;
}
pci_dev = rte_zmalloc_socket(name, sizeof(*pci_dev), 0, socket_id);
if (pci_dev == NULL) {
RTE_BOND_LOG(ERR, "Unable to malloc pci dev on socket");
goto err;
}
eth_drv = rte_zmalloc_socket(name, sizeof(*eth_drv), 0, socket_id);
if (eth_drv == NULL) {
RTE_BOND_LOG(ERR, "Unable to malloc eth_drv on socket");
goto err;
}
pci_drv = ð_drv->pci_drv;
pci_id_table = rte_zmalloc_socket(name, sizeof(*pci_id_table), 0, socket_id);
if (pci_id_table == NULL) {
RTE_BOND_LOG(ERR, "Unable to malloc pci_id_table on socket");
goto err;
}
pci_drv->id_table = pci_id_table;
pci_drv->id_table->device_id = PCI_ANY_ID;
pci_drv->id_table->subsystem_device_id = PCI_ANY_ID;
pci_drv->id_table->vendor_id = PCI_ANY_ID;
pci_drv->id_table->subsystem_vendor_id = PCI_ANY_ID;
pci_drv->drv_flags = RTE_PCI_DRV_INTR_LSC;
internals = rte_zmalloc_socket(name, sizeof(*internals), 0, socket_id);
if (internals == NULL) {
RTE_BOND_LOG(ERR, "Unable to malloc internals on socket");
goto err;
}
/* reserve an ethdev entry */
eth_dev = rte_eth_dev_allocate(name, RTE_ETH_DEV_VIRTUAL);
if (eth_dev == NULL) {
RTE_BOND_LOG(ERR, "Unable to allocate rte_eth_dev");
goto err;
}
pci_dev->numa_node = socket_id;
pci_drv->name = driver_name;
eth_dev->driver = eth_drv;
eth_dev->data->dev_private = internals;
eth_dev->data->nb_rx_queues = (uint16_t)1;
eth_dev->data->nb_tx_queues = (uint16_t)1;
TAILQ_INIT(&(eth_dev->link_intr_cbs));
eth_dev->data->dev_link.link_status = 0;
eth_dev->data->mac_addrs = rte_zmalloc_socket(name, ETHER_ADDR_LEN, 0,
socket_id);
eth_dev->data->dev_started = 0;
eth_dev->data->promiscuous = 0;
eth_dev->data->scattered_rx = 0;
eth_dev->data->all_multicast = 0;
eth_dev->dev_ops = &default_dev_ops;
eth_dev->pci_dev = pci_dev;
rte_spinlock_init(&internals->lock);
internals->port_id = eth_dev->data->port_id;
internals->mode = BONDING_MODE_INVALID;
internals->current_primary_port = 0;
internals->balance_xmit_policy = BALANCE_XMIT_POLICY_LAYER2;
internals->xmit_hash = xmit_l2_hash;
internals->user_defined_mac = 0;
internals->link_props_set = 0;
internals->link_status_polling_enabled = 0;
internals->link_status_polling_interval_ms = DEFAULT_POLLING_INTERVAL_10_MS;
internals->link_down_delay_ms = 0;
internals->link_up_delay_ms = 0;
internals->slave_count = 0;
internals->active_slave_count = 0;
internals->rx_offload_capa = 0;
internals->tx_offload_capa = 0;
memset(internals->active_slaves, 0, sizeof(internals->active_slaves));
memset(internals->slaves, 0, sizeof(internals->slaves));
/* Set mode 4 default configuration */
bond_mode_8023ad_setup(eth_dev, NULL);
if (bond_ethdev_mode_set(eth_dev, mode)) {
RTE_BOND_LOG(ERR, "Failed to set bonded device %d mode too %d",
eth_dev->data->port_id, mode);
goto err;
}
return eth_dev->data->port_id;
err:
if (pci_dev)
rte_free(pci_dev);
if (pci_id_table)
rte_free(pci_id_table);
if (eth_drv)
rte_free(eth_drv);
if (internals)
rte_free(internals);
return -1;
}
/**
* DPDK callback to register a PCI device.
*
* This function creates an Ethernet device for each port of a given
* PCI device.
*
* @param[in] pci_drv
* PCI driver structure (mlx5_driver).
* @param[in] pci_dev
* PCI device information.
*
* @return
* 0 on success, negative errno value on failure.
*/
static int
mlx5_pci_devinit(struct rte_pci_driver *pci_drv, struct rte_pci_device *pci_dev)
{
struct ibv_device **list;
struct ibv_device *ibv_dev;
int err = 0;
struct ibv_context *attr_ctx = NULL;
struct ibv_device_attr device_attr;
unsigned int vf;
int idx;
int i;
(void)pci_drv;
assert(pci_drv == &mlx5_driver.pci_drv);
/* Get mlx5_dev[] index. */
idx = mlx5_dev_idx(&pci_dev->addr);
if (idx == -1) {
ERROR("this driver cannot support any more adapters");
return -ENOMEM;
}
DEBUG("using driver device index %d", idx);
/* Save PCI address. */
mlx5_dev[idx].pci_addr = pci_dev->addr;
list = ibv_get_device_list(&i);
if (list == NULL) {
assert(errno);
if (errno == ENOSYS) {
WARN("cannot list devices, is ib_uverbs loaded?");
return 0;
}
return -errno;
}
assert(i >= 0);
/*
* For each listed device, check related sysfs entry against
* the provided PCI ID.
*/
while (i != 0) {
struct rte_pci_addr pci_addr;
--i;
DEBUG("checking device \"%s\"", list[i]->name);
if (mlx5_ibv_device_to_pci_addr(list[i], &pci_addr))
continue;
if ((pci_dev->addr.domain != pci_addr.domain) ||
(pci_dev->addr.bus != pci_addr.bus) ||
(pci_dev->addr.devid != pci_addr.devid) ||
(pci_dev->addr.function != pci_addr.function))
continue;
vf = ((pci_dev->id.device_id ==
PCI_DEVICE_ID_MELLANOX_CONNECTX4VF) ||
(pci_dev->id.device_id ==
PCI_DEVICE_ID_MELLANOX_CONNECTX4LXVF));
INFO("PCI information matches, using device \"%s\" (VF: %s)",
list[i]->name, (vf ? "true" : "false"));
attr_ctx = ibv_open_device(list[i]);
err = errno;
break;
}
if (attr_ctx == NULL) {
ibv_free_device_list(list);
switch (err) {
case 0:
WARN("cannot access device, is mlx5_ib loaded?");
return 0;
case EINVAL:
WARN("cannot use device, are drivers up to date?");
return 0;
}
assert(err > 0);
return -err;
}
ibv_dev = list[i];
DEBUG("device opened");
if (ibv_query_device(attr_ctx, &device_attr))
goto error;
INFO("%u port(s) detected", device_attr.phys_port_cnt);
for (i = 0; i < device_attr.phys_port_cnt; i++) {
uint32_t port = i + 1; /* ports are indexed from one */
uint32_t test = (1 << i);
struct ibv_context *ctx = NULL;
struct ibv_port_attr port_attr;
struct ibv_pd *pd = NULL;
struct priv *priv = NULL;
struct rte_eth_dev *eth_dev;
#ifdef HAVE_EXP_QUERY_DEVICE
struct ibv_exp_device_attr exp_device_attr;
#endif /* HAVE_EXP_QUERY_DEVICE */
struct ether_addr mac;
#ifdef HAVE_EXP_QUERY_DEVICE
exp_device_attr.comp_mask =
IBV_EXP_DEVICE_ATTR_EXP_CAP_FLAGS |
IBV_EXP_DEVICE_ATTR_RX_HASH;
#endif /* HAVE_EXP_QUERY_DEVICE */
DEBUG("using port %u (%08" PRIx32 ")", port, test);
ctx = ibv_open_device(ibv_dev);
if (ctx == NULL)
goto port_error;
/* Check port status. */
err = ibv_query_port(ctx, port, &port_attr);
if (err) {
ERROR("port query failed: %s", strerror(err));
goto port_error;
}
if (port_attr.state != IBV_PORT_ACTIVE)
DEBUG("port %d is not active: \"%s\" (%d)",
port, ibv_port_state_str(port_attr.state),
port_attr.state);
/* Allocate protection domain. */
pd = ibv_alloc_pd(ctx);
if (pd == NULL) {
ERROR("PD allocation failure");
err = ENOMEM;
goto port_error;
}
mlx5_dev[idx].ports |= test;
/* from rte_ethdev.c */
priv = rte_zmalloc("ethdev private structure",
sizeof(*priv),
RTE_CACHE_LINE_SIZE);
if (priv == NULL) {
ERROR("priv allocation failure");
err = ENOMEM;
goto port_error;
}
priv->ctx = ctx;
priv->device_attr = device_attr;
priv->port = port;
priv->pd = pd;
priv->mtu = ETHER_MTU;
#ifdef HAVE_EXP_QUERY_DEVICE
if (ibv_exp_query_device(ctx, &exp_device_attr)) {
ERROR("ibv_exp_query_device() failed");
goto port_error;
}
priv->hw_csum =
((exp_device_attr.exp_device_cap_flags &
IBV_EXP_DEVICE_RX_CSUM_TCP_UDP_PKT) &&
(exp_device_attr.exp_device_cap_flags &
IBV_EXP_DEVICE_RX_CSUM_IP_PKT));
DEBUG("checksum offloading is %ssupported",
(priv->hw_csum ? "" : "not "));
priv->hw_csum_l2tun = !!(exp_device_attr.exp_device_cap_flags &
IBV_EXP_DEVICE_VXLAN_SUPPORT);
DEBUG("L2 tunnel checksum offloads are %ssupported",
(priv->hw_csum_l2tun ? "" : "not "));
priv->ind_table_max_size = exp_device_attr.rx_hash_caps.max_rwq_indirection_table_size;
DEBUG("maximum RX indirection table size is %u",
priv->ind_table_max_size);
#else /* HAVE_EXP_QUERY_DEVICE */
priv->ind_table_max_size = RSS_INDIRECTION_TABLE_SIZE;
#endif /* HAVE_EXP_QUERY_DEVICE */
priv->vf = vf;
/* Allocate and register default RSS hash keys. */
priv->rss_conf = rte_calloc(__func__, hash_rxq_init_n,
sizeof((*priv->rss_conf)[0]), 0);
if (priv->rss_conf == NULL) {
err = ENOMEM;
goto port_error;
}
err = rss_hash_rss_conf_new_key(priv,
rss_hash_default_key,
rss_hash_default_key_len,
ETH_RSS_PROTO_MASK);
if (err)
goto port_error;
/* Configure the first MAC address by default. */
if (priv_get_mac(priv, &mac.addr_bytes)) {
ERROR("cannot get MAC address, is mlx5_en loaded?"
" (errno: %s)", strerror(errno));
goto port_error;
}
INFO("port %u MAC address is %02x:%02x:%02x:%02x:%02x:%02x",
priv->port,
mac.addr_bytes[0], mac.addr_bytes[1],
mac.addr_bytes[2], mac.addr_bytes[3],
mac.addr_bytes[4], mac.addr_bytes[5]);
/* Register MAC and broadcast addresses. */
claim_zero(priv_mac_addr_add(priv, 0,
(const uint8_t (*)[ETHER_ADDR_LEN])
mac.addr_bytes));
claim_zero(priv_mac_addr_add(priv, (RTE_DIM(priv->mac) - 1),
&(const uint8_t [ETHER_ADDR_LEN])
{ "\xff\xff\xff\xff\xff\xff" }));
#ifndef NDEBUG
{
char ifname[IF_NAMESIZE];
if (priv_get_ifname(priv, &ifname) == 0)
DEBUG("port %u ifname is \"%s\"",
priv->port, ifname);
else
DEBUG("port %u ifname is unknown", priv->port);
}
#endif
/* Get actual MTU if possible. */
priv_get_mtu(priv, &priv->mtu);
DEBUG("port %u MTU is %u", priv->port, priv->mtu);
/* from rte_ethdev.c */
{
char name[RTE_ETH_NAME_MAX_LEN];
snprintf(name, sizeof(name), "%s port %u",
ibv_get_device_name(ibv_dev), port);
eth_dev = rte_eth_dev_allocate(name, RTE_ETH_DEV_PCI);
}
if (eth_dev == NULL) {
ERROR("can not allocate rte ethdev");
err = ENOMEM;
goto port_error;
}
eth_dev->data->dev_private = priv;
eth_dev->pci_dev = pci_dev;
eth_dev->driver = &mlx5_driver;
eth_dev->data->rx_mbuf_alloc_failed = 0;
eth_dev->data->mtu = ETHER_MTU;
priv->dev = eth_dev;
eth_dev->dev_ops = &mlx5_dev_ops;
eth_dev->data->mac_addrs = priv->mac;
TAILQ_INIT(ð_dev->link_intr_cbs);
/* Bring Ethernet device up. */
DEBUG("forcing Ethernet interface up");
priv_set_flags(priv, ~IFF_UP, IFF_UP);
continue;
port_error:
rte_free(priv->rss_conf);
rte_free(priv);
if (pd)
claim_zero(ibv_dealloc_pd(pd));
if (ctx)
claim_zero(ibv_close_device(ctx));
break;
}
int
rte_eth_bond_create(const char *name, uint8_t mode, uint8_t socket_id)
{
struct bond_dev_private *internals = NULL;
struct rte_eth_dev *eth_dev = NULL;
uint32_t vlan_filter_bmp_size;
/* now do all data allocation - for eth_dev structure, dummy pci driver
* and internal (private) data
*/
if (name == NULL) {
RTE_BOND_LOG(ERR, "Invalid name specified");
goto err;
}
if (socket_id >= number_of_sockets()) {
RTE_BOND_LOG(ERR,
"Invalid socket id specified to create bonded device on.");
goto err;
}
internals = rte_zmalloc_socket(name, sizeof(*internals), 0, socket_id);
if (internals == NULL) {
RTE_BOND_LOG(ERR, "Unable to malloc internals on socket");
goto err;
}
/* reserve an ethdev entry */
eth_dev = rte_eth_dev_allocate(name);
if (eth_dev == NULL) {
RTE_BOND_LOG(ERR, "Unable to allocate rte_eth_dev");
goto err;
}
eth_dev->data->dev_private = internals;
eth_dev->data->nb_rx_queues = (uint16_t)1;
eth_dev->data->nb_tx_queues = (uint16_t)1;
eth_dev->data->mac_addrs = rte_zmalloc_socket(name, ETHER_ADDR_LEN, 0,
socket_id);
if (eth_dev->data->mac_addrs == NULL) {
RTE_BOND_LOG(ERR, "Unable to malloc mac_addrs");
goto err;
}
eth_dev->dev_ops = &default_dev_ops;
eth_dev->data->dev_flags = RTE_ETH_DEV_INTR_LSC |
RTE_ETH_DEV_DETACHABLE;
eth_dev->driver = NULL;
eth_dev->data->kdrv = RTE_KDRV_NONE;
eth_dev->data->drv_name = pmd_bond_drv.driver.name;
eth_dev->data->numa_node = socket_id;
rte_spinlock_init(&internals->lock);
internals->port_id = eth_dev->data->port_id;
internals->mode = BONDING_MODE_INVALID;
internals->current_primary_port = RTE_MAX_ETHPORTS + 1;
internals->balance_xmit_policy = BALANCE_XMIT_POLICY_LAYER2;
internals->xmit_hash = xmit_l2_hash;
internals->user_defined_mac = 0;
internals->link_props_set = 0;
internals->link_status_polling_enabled = 0;
internals->link_status_polling_interval_ms = DEFAULT_POLLING_INTERVAL_10_MS;
internals->link_down_delay_ms = 0;
internals->link_up_delay_ms = 0;
internals->slave_count = 0;
internals->active_slave_count = 0;
internals->rx_offload_capa = 0;
internals->tx_offload_capa = 0;
internals->candidate_max_rx_pktlen = 0;
internals->max_rx_pktlen = 0;
/* Initially allow to choose any offload type */
internals->flow_type_rss_offloads = ETH_RSS_PROTO_MASK;
memset(internals->active_slaves, 0, sizeof(internals->active_slaves));
memset(internals->slaves, 0, sizeof(internals->slaves));
/* Set mode 4 default configuration */
bond_mode_8023ad_setup(eth_dev, NULL);
if (bond_ethdev_mode_set(eth_dev, mode)) {
RTE_BOND_LOG(ERR, "Failed to set bonded device %d mode too %d",
eth_dev->data->port_id, mode);
goto err;
}
vlan_filter_bmp_size =
rte_bitmap_get_memory_footprint(ETHER_MAX_VLAN_ID + 1);
internals->vlan_filter_bmpmem = rte_malloc(name, vlan_filter_bmp_size,
RTE_CACHE_LINE_SIZE);
if (internals->vlan_filter_bmpmem == NULL) {
RTE_BOND_LOG(ERR,
"Failed to allocate vlan bitmap for bonded device %u\n",
eth_dev->data->port_id);
goto err;
}
internals->vlan_filter_bmp = rte_bitmap_init(ETHER_MAX_VLAN_ID + 1,
internals->vlan_filter_bmpmem, vlan_filter_bmp_size);
if (internals->vlan_filter_bmp == NULL) {
RTE_BOND_LOG(ERR,
"Failed to init vlan bitmap for bonded device %u\n",
eth_dev->data->port_id);
rte_free(internals->vlan_filter_bmpmem);
goto err;
}
return eth_dev->data->port_id;
err:
rte_free(internals);
if (eth_dev != NULL) {
rte_free(eth_dev->data->mac_addrs);
rte_eth_dev_release_port(eth_dev);
}
return -1;
}
