aboutsummaryrefslogtreecommitdiffstats
path: root/drivers/net/bonding/rte_eth_bond_pmd.c
diff options
context:
space:
mode:
authorLuca Boccassi <luca.boccassi@gmail.com>2018-11-01 11:59:50 +0000
committerLuca Boccassi <luca.boccassi@gmail.com>2018-11-01 12:00:19 +0000
commit8d01b9cd70a67cdafd5b965a70420c3bd7fb3f82 (patch)
tree208e3bc33c220854d89d010e3abf720a2e62e546 /drivers/net/bonding/rte_eth_bond_pmd.c
parentb63264c8342e6a1b6971c79550d2af2024b6a4de (diff)
New upstream version 18.11-rc1upstream/18.11-rc1
Change-Id: Iaa71986dd6332e878d8f4bf493101b2bbc6313bb Signed-off-by: Luca Boccassi <luca.boccassi@gmail.com>
Diffstat (limited to 'drivers/net/bonding/rte_eth_bond_pmd.c')
-rw-r--r--drivers/net/bonding/rte_eth_bond_pmd.c275
1 files changed, 128 insertions, 147 deletions
diff --git a/drivers/net/bonding/rte_eth_bond_pmd.c b/drivers/net/bonding/rte_eth_bond_pmd.c
index 58f7377c..156f31c6 100644
--- a/drivers/net/bonding/rte_eth_bond_pmd.c
+++ b/drivers/net/bonding/rte_eth_bond_pmd.c
@@ -37,7 +37,8 @@ get_vlan_offset(struct ether_hdr *eth_hdr, uint16_t *proto)
{
size_t vlan_offset = 0;
- if (rte_cpu_to_be_16(ETHER_TYPE_VLAN) == *proto) {
+ if (rte_cpu_to_be_16(ETHER_TYPE_VLAN) == *proto ||
+ rte_cpu_to_be_16(ETHER_TYPE_QINQ) == *proto) {
struct vlan_hdr *vlan_hdr = (struct vlan_hdr *)(eth_hdr + 1);
vlan_offset = sizeof(struct vlan_hdr);
@@ -57,28 +58,34 @@ bond_ethdev_rx_burst(void *queue, struct rte_mbuf **bufs, uint16_t nb_pkts)
{
struct bond_dev_private *internals;
- uint16_t num_rx_slave = 0;
uint16_t num_rx_total = 0;
-
+ uint16_t slave_count;
+ uint16_t active_slave;
int i;
/* Cast to structure, containing bonded device's port id and queue id */
struct bond_rx_queue *bd_rx_q = (struct bond_rx_queue *)queue;
-
internals = bd_rx_q->dev_private;
+ slave_count = internals->active_slave_count;
+ active_slave = internals->active_slave;
+ for (i = 0; i < slave_count && nb_pkts; i++) {
+ uint16_t num_rx_slave;
- for (i = 0; i < internals->active_slave_count && nb_pkts; i++) {
/* Offset of pointer to *bufs increases as packets are received
* from other slaves */
- num_rx_slave = rte_eth_rx_burst(internals->active_slaves[i],
- bd_rx_q->queue_id, bufs + num_rx_total, nb_pkts);
- if (num_rx_slave) {
- num_rx_total += num_rx_slave;
- nb_pkts -= num_rx_slave;
- }
+ num_rx_slave =
+ rte_eth_rx_burst(internals->active_slaves[active_slave],
+ bd_rx_q->queue_id,
+ bufs + num_rx_total, nb_pkts);
+ num_rx_total += num_rx_slave;
+ nb_pkts -= num_rx_slave;
+ if (++active_slave == slave_count)
+ active_slave = 0;
}
+ if (++internals->active_slave == slave_count)
+ internals->active_slave = 0;
return num_rx_total;
}
@@ -257,25 +264,32 @@ bond_ethdev_rx_burst_8023ad_fast_queue(void *queue, struct rte_mbuf **bufs,
uint16_t num_rx_total = 0; /* Total number of received packets */
uint16_t slaves[RTE_MAX_ETHPORTS];
uint16_t slave_count;
-
- uint16_t i, idx;
+ uint16_t active_slave;
+ uint16_t i;
/* Copy slave list to protect against slave up/down changes during tx
* bursting */
slave_count = internals->active_slave_count;
+ active_slave = internals->active_slave;
memcpy(slaves, internals->active_slaves,
sizeof(internals->active_slaves[0]) * slave_count);
- for (i = 0, idx = internals->active_slave;
- i < slave_count && num_rx_total < nb_pkts; i++, idx++) {
- idx = idx % slave_count;
+ for (i = 0; i < slave_count && nb_pkts; i++) {
+ uint16_t num_rx_slave;
/* Read packets from this slave */
- num_rx_total += rte_eth_rx_burst(slaves[idx], bd_rx_q->queue_id,
- &bufs[num_rx_total], nb_pkts - num_rx_total);
+ num_rx_slave = rte_eth_rx_burst(slaves[active_slave],
+ bd_rx_q->queue_id,
+ bufs + num_rx_total, nb_pkts);
+ num_rx_total += num_rx_slave;
+ nb_pkts -= num_rx_slave;
+
+ if (++active_slave == slave_count)
+ active_slave = 0;
}
- internals->active_slave = idx;
+ if (++internals->active_slave == slave_count)
+ internals->active_slave = 0;
return num_rx_total;
}
@@ -300,10 +314,10 @@ bond_ethdev_tx_burst_8023ad_fast_queue(void *queue, struct rte_mbuf **bufs,
/* Mapping array generated by hash function to map mbufs to slaves */
uint16_t bufs_slave_port_idxs[RTE_MAX_ETHPORTS] = { 0 };
- uint16_t slave_tx_count, slave_tx_fail_count[RTE_MAX_ETHPORTS] = { 0 };
+ uint16_t slave_tx_count;
uint16_t total_tx_count = 0, total_tx_fail_count = 0;
- uint16_t i, j;
+ uint16_t i;
if (unlikely(nb_bufs == 0))
return 0;
@@ -320,7 +334,7 @@ bond_ethdev_tx_burst_8023ad_fast_queue(void *queue, struct rte_mbuf **bufs,
dist_slave_count = 0;
for (i = 0; i < slave_count; i++) {
- struct port *port = &mode_8023ad_ports[slave_port_ids[i]];
+ struct port *port = &bond_mode_8023ad_ports[slave_port_ids[i]];
if (ACTOR_STATE(port, DISTRIBUTING))
dist_slave_port_ids[dist_slave_count++] =
@@ -358,34 +372,12 @@ bond_ethdev_tx_burst_8023ad_fast_queue(void *queue, struct rte_mbuf **bufs,
/* If tx burst fails move packets to end of bufs */
if (unlikely(slave_tx_count < slave_nb_bufs[i])) {
- slave_tx_fail_count[i] = slave_nb_bufs[i] -
+ int slave_tx_fail_count = slave_nb_bufs[i] -
slave_tx_count;
- total_tx_fail_count += slave_tx_fail_count[i];
-
- /*
- * Shift bufs to beginning of array to allow reordering
- * later
- */
- for (j = 0; j < slave_tx_fail_count[i]; j++) {
- slave_bufs[i][j] =
- slave_bufs[i][(slave_tx_count - 1) + j];
- }
- }
- }
-
- /*
- * If there are tx burst failures we move packets to end of bufs to
- * preserve expected PMD behaviour of all failed transmitted being
- * at the end of the input mbuf array
- */
- if (unlikely(total_tx_fail_count > 0)) {
- int bufs_idx = nb_bufs - total_tx_fail_count - 1;
-
- for (i = 0; i < slave_count; i++) {
- if (slave_tx_fail_count[i] > 0) {
- for (j = 0; j < slave_tx_fail_count[i]; j++)
- bufs[bufs_idx++] = slave_bufs[i][j];
- }
+ total_tx_fail_count += slave_tx_fail_count;
+ memcpy(&bufs[nb_bufs - total_tx_fail_count],
+ &slave_bufs[i][slave_tx_count],
+ slave_tx_fail_count * sizeof(bufs[0]));
}
}
@@ -400,8 +392,9 @@ bond_ethdev_rx_burst_8023ad(void *queue, struct rte_mbuf **bufs,
/* Cast to structure, containing bonded device's port id and queue id */
struct bond_rx_queue *bd_rx_q = (struct bond_rx_queue *)queue;
struct bond_dev_private *internals = bd_rx_q->dev_private;
- struct ether_addr bond_mac;
-
+ struct rte_eth_dev *bonded_eth_dev =
+ &rte_eth_devices[internals->port_id];
+ struct ether_addr *bond_mac = bonded_eth_dev->data->mac_addrs;
struct ether_hdr *hdr;
const uint16_t ether_type_slow_be = rte_be_to_cpu_16(ETHER_TYPE_SLOW);
@@ -414,7 +407,6 @@ bond_ethdev_rx_burst_8023ad(void *queue, struct rte_mbuf **bufs,
uint8_t i, j, k;
uint8_t subtype;
- rte_eth_macaddr_get(internals->port_id, &bond_mac);
/* Copy slave list to protect against slave up/down changes during tx
* bursting */
slave_count = internals->active_slave_count;
@@ -428,7 +420,7 @@ bond_ethdev_rx_burst_8023ad(void *queue, struct rte_mbuf **bufs,
}
for (i = 0; i < slave_count && num_rx_total < nb_pkts; i++) {
j = num_rx_total;
- collecting = ACTOR_STATE(&mode_8023ad_ports[slaves[idx]],
+ collecting = ACTOR_STATE(&bond_mode_8023ad_ports[slaves[idx]],
COLLECTING);
/* Read packets from this slave */
@@ -457,9 +449,11 @@ bond_ethdev_rx_burst_8023ad(void *queue, struct rte_mbuf **bufs,
* in collecting state or bonding interface is not in promiscuous
* mode and packet address does not match. */
if (unlikely(is_lacp_packets(hdr->ether_type, subtype, bufs[j]) ||
- !collecting || (!promisc &&
- !is_multicast_ether_addr(&hdr->d_addr) &&
- !is_same_ether_addr(&bond_mac, &hdr->d_addr)))) {
+ !collecting ||
+ (!promisc &&
+ !is_multicast_ether_addr(&hdr->d_addr) &&
+ !is_same_ether_addr(bond_mac,
+ &hdr->d_addr)))) {
if (hdr->ether_type == ether_type_slow_be) {
bond_mode_8023ad_handle_slow_pkt(
@@ -480,7 +474,9 @@ bond_ethdev_rx_burst_8023ad(void *queue, struct rte_mbuf **bufs,
idx = 0;
}
- internals->active_slave = idx;
+ if (++internals->active_slave == slave_count)
+ internals->active_slave = 0;
+
return num_rx_total;
}
@@ -715,8 +711,8 @@ bond_ethdev_tx_burst_round_robin(void *queue, struct rte_mbuf **bufs,
tx_fail_total += tx_fail_slave;
memcpy(&bufs[nb_pkts - tx_fail_total],
- &slave_bufs[i][num_tx_slave],
- tx_fail_slave * sizeof(bufs[0]));
+ &slave_bufs[i][num_tx_slave],
+ tx_fail_slave * sizeof(bufs[0]));
}
num_tx_total += num_tx_slave;
}
@@ -1221,10 +1217,10 @@ bond_ethdev_tx_burst_balance(void *queue, struct rte_mbuf **bufs,
/* Mapping array generated by hash function to map mbufs to slaves */
uint16_t bufs_slave_port_idxs[nb_bufs];
- uint16_t slave_tx_count, slave_tx_fail_count[RTE_MAX_ETHPORTS] = { 0 };
+ uint16_t slave_tx_count;
uint16_t total_tx_count = 0, total_tx_fail_count = 0;
- uint16_t i, j;
+ uint16_t i;
if (unlikely(nb_bufs == 0))
return 0;
@@ -1265,34 +1261,12 @@ bond_ethdev_tx_burst_balance(void *queue, struct rte_mbuf **bufs,
/* If tx burst fails move packets to end of bufs */
if (unlikely(slave_tx_count < slave_nb_bufs[i])) {
- slave_tx_fail_count[i] = slave_nb_bufs[i] -
+ int slave_tx_fail_count = slave_nb_bufs[i] -
slave_tx_count;
- total_tx_fail_count += slave_tx_fail_count[i];
-
- /*
- * Shift bufs to beginning of array to allow reordering
- * later
- */
- for (j = 0; j < slave_tx_fail_count[i]; j++) {
- slave_bufs[i][j] =
- slave_bufs[i][(slave_tx_count - 1) + j];
- }
- }
- }
-
- /*
- * If there are tx burst failures we move packets to end of bufs to
- * preserve expected PMD behaviour of all failed transmitted being
- * at the end of the input mbuf array
- */
- if (unlikely(total_tx_fail_count > 0)) {
- int bufs_idx = nb_bufs - total_tx_fail_count - 1;
-
- for (i = 0; i < slave_count; i++) {
- if (slave_tx_fail_count[i] > 0) {
- for (j = 0; j < slave_tx_fail_count[i]; j++)
- bufs[bufs_idx++] = slave_bufs[i][j];
- }
+ total_tx_fail_count += slave_tx_fail_count;
+ memcpy(&bufs[nb_bufs - total_tx_fail_count],
+ &slave_bufs[i][slave_tx_count],
+ slave_tx_fail_count * sizeof(bufs[0]));
}
}
@@ -1319,10 +1293,10 @@ bond_ethdev_tx_burst_8023ad(void *queue, struct rte_mbuf **bufs,
/* Mapping array generated by hash function to map mbufs to slaves */
uint16_t bufs_slave_port_idxs[RTE_MAX_ETHPORTS] = { 0 };
- uint16_t slave_tx_count, slave_tx_fail_count[RTE_MAX_ETHPORTS] = { 0 };
+ uint16_t slave_tx_count;
uint16_t total_tx_count = 0, total_tx_fail_count = 0;
- uint16_t i, j;
+ uint16_t i;
if (unlikely(nb_bufs == 0))
return 0;
@@ -1338,7 +1312,7 @@ bond_ethdev_tx_burst_8023ad(void *queue, struct rte_mbuf **bufs,
dist_slave_count = 0;
for (i = 0; i < slave_count; i++) {
- struct port *port = &mode_8023ad_ports[slave_port_ids[i]];
+ struct port *port = &bond_mode_8023ad_ports[slave_port_ids[i]];
if (ACTOR_STATE(port, DISTRIBUTING))
dist_slave_port_ids[dist_slave_count++] =
@@ -1380,46 +1354,20 @@ bond_ethdev_tx_burst_8023ad(void *queue, struct rte_mbuf **bufs,
/* If tx burst fails move packets to end of bufs */
if (unlikely(slave_tx_count < slave_nb_bufs[i])) {
- slave_tx_fail_count[i] = slave_nb_bufs[i] -
+ int slave_tx_fail_count = slave_nb_bufs[i] -
slave_tx_count;
- total_tx_fail_count += slave_tx_fail_count[i];
-
- /*
- * Shift bufs to beginning of array to allow
- * reordering later
- */
- for (j = 0; j < slave_tx_fail_count[i]; j++)
- slave_bufs[i][j] =
- slave_bufs[i]
- [(slave_tx_count - 1)
- + j];
- }
- }
+ total_tx_fail_count += slave_tx_fail_count;
- /*
- * If there are tx burst failures we move packets to end of
- * bufs to preserve expected PMD behaviour of all failed
- * transmitted being at the end of the input mbuf array
- */
- if (unlikely(total_tx_fail_count > 0)) {
- int bufs_idx = nb_bufs - total_tx_fail_count - 1;
-
- for (i = 0; i < slave_count; i++) {
- if (slave_tx_fail_count[i] > 0) {
- for (j = 0;
- j < slave_tx_fail_count[i];
- j++) {
- bufs[bufs_idx++] =
- slave_bufs[i][j];
- }
- }
+ memcpy(&bufs[nb_bufs - total_tx_fail_count],
+ &slave_bufs[i][slave_tx_count],
+ slave_tx_fail_count * sizeof(bufs[0]));
}
}
}
/* Check for LACP control packets and send if available */
for (i = 0; i < slave_count; i++) {
- struct port *port = &mode_8023ad_ports[slave_port_ids[i]];
+ struct port *port = &bond_mode_8023ad_ports[slave_port_ids[i]];
struct rte_mbuf *ctrl_pkt = NULL;
if (likely(rte_ring_empty(port->tx_ring)))
@@ -1770,7 +1718,7 @@ slave_configure_slow_queue(struct rte_eth_dev *bonded_eth_dev,
int errval = 0;
struct bond_dev_private *internals = (struct bond_dev_private *)
bonded_eth_dev->data->dev_private;
- struct port *port = &mode_8023ad_ports[slave_eth_dev->data->port_id];
+ struct port *port = &bond_mode_8023ad_ports[slave_eth_dev->data->port_id];
if (port->slow_pool == NULL) {
char mem_name[256];
@@ -1847,12 +1795,11 @@ slave_configure(struct rte_eth_dev *bonded_eth_dev,
/* If RSS is enabled for bonding, try to enable it for slaves */
if (bonded_eth_dev->data->dev_conf.rxmode.mq_mode & ETH_MQ_RX_RSS_FLAG) {
- if (bonded_eth_dev->data->dev_conf.rx_adv_conf.rss_conf.rss_key_len
- != 0) {
+ if (internals->rss_key_len != 0) {
slave_eth_dev->data->dev_conf.rx_adv_conf.rss_conf.rss_key_len =
- bonded_eth_dev->data->dev_conf.rx_adv_conf.rss_conf.rss_key_len;
+ internals->rss_key_len;
slave_eth_dev->data->dev_conf.rx_adv_conf.rss_conf.rss_key =
- bonded_eth_dev->data->dev_conf.rx_adv_conf.rss_conf.rss_key;
+ internals->rss_key;
} else {
slave_eth_dev->data->dev_conf.rx_adv_conf.rss_conf.rss_key = NULL;
}
@@ -2210,7 +2157,7 @@ bond_ethdev_stop(struct rte_eth_dev *eth_dev)
/* Discard all messages to/from mode 4 state machines */
for (i = 0; i < internals->active_slave_count; i++) {
- port = &mode_8023ad_ports[internals->active_slaves[i]];
+ port = &bond_mode_8023ad_ports[internals->active_slaves[i]];
RTE_ASSERT(port->rx_ring != NULL);
while (rte_ring_dequeue(port->rx_ring, &pkt) != -ENOENT)
@@ -2229,12 +2176,15 @@ bond_ethdev_stop(struct rte_eth_dev *eth_dev)
tlb_last_obytets[internals->active_slaves[i]] = 0;
}
- internals->link_status_polling_enabled = 0;
- for (i = 0; i < internals->slave_count; i++)
- internals->slaves[i].last_link_status = 0;
-
eth_dev->data->dev_link.link_status = ETH_LINK_DOWN;
eth_dev->data->dev_started = 0;
+
+ internals->link_status_polling_enabled = 0;
+ for (i = 0; i < internals->slave_count; i++) {
+ internals->slaves[i].last_link_status = 0;
+ rte_eth_dev_stop(internals->slaves[i].port_id);
+ deactivate_slave(eth_dev, internals->slaves[i].port_id);
+ }
}
void
@@ -2303,6 +2253,16 @@ bond_ethdev_info(struct rte_eth_dev *dev, struct rte_eth_dev_info *dev_info)
dev_info->max_rx_queues = max_nb_rx_queues;
dev_info->max_tx_queues = max_nb_tx_queues;
+ memcpy(&dev_info->default_rxconf, &internals->default_rxconf,
+ sizeof(dev_info->default_rxconf));
+ memcpy(&dev_info->default_txconf, &internals->default_txconf,
+ sizeof(dev_info->default_txconf));
+
+ memcpy(&dev_info->rx_desc_lim, &internals->rx_desc_lim,
+ sizeof(dev_info->rx_desc_lim));
+ memcpy(&dev_info->tx_desc_lim, &internals->tx_desc_lim,
+ sizeof(dev_info->tx_desc_lim));
+
/**
* If dedicated hw queues enabled for link bonding device in LACP mode
* then we need to reduce the maximum number of data path queues by 1.
@@ -3123,6 +3083,14 @@ bond_alloc(struct rte_vdev_device *dev, uint8_t mode)
/* Initially allow to choose any offload type */
internals->flow_type_rss_offloads = ETH_RSS_PROTO_MASK;
+ memset(&internals->default_rxconf, 0,
+ sizeof(internals->default_rxconf));
+ memset(&internals->default_txconf, 0,
+ sizeof(internals->default_txconf));
+
+ memset(&internals->rx_desc_lim, 0, sizeof(internals->rx_desc_lim));
+ memset(&internals->tx_desc_lim, 0, sizeof(internals->tx_desc_lim));
+
memset(internals->active_slaves, 0, sizeof(internals->active_slaves));
memset(internals->slaves, 0, sizeof(internals->slaves));
@@ -3162,10 +3130,9 @@ bond_alloc(struct rte_vdev_device *dev, uint8_t mode)
err:
rte_free(internals);
- if (eth_dev != NULL) {
- rte_free(eth_dev->data->mac_addrs);
- rte_eth_dev_release_port(eth_dev);
- }
+ if (eth_dev != NULL)
+ eth_dev->data->dev_private = NULL;
+ rte_eth_dev_release_port(eth_dev);
return -1;
}
@@ -3186,8 +3153,7 @@ bond_probe(struct rte_vdev_device *dev)
name = rte_vdev_device_name(dev);
RTE_BOND_LOG(INFO, "Initializing pmd_bond for %s", name);
- if (rte_eal_process_type() == RTE_PROC_SECONDARY &&
- strlen(rte_vdev_device_args(dev)) == 0) {
+ if (rte_eal_process_type() == RTE_PROC_SECONDARY) {
eth_dev = rte_eth_dev_attach_secondary(name);
if (!eth_dev) {
RTE_BOND_LOG(ERR, "Failed to probe %s", name);
@@ -3302,6 +3268,9 @@ bond_remove(struct rte_vdev_device *dev)
if (eth_dev == NULL)
return -ENODEV;
+ if (rte_eal_process_type() != RTE_PROC_PRIMARY)
+ return rte_eth_dev_release_port(eth_dev);
+
RTE_ASSERT(eth_dev->device == &dev->device);
internals = eth_dev->data->dev_private;
@@ -3324,8 +3293,6 @@ bond_remove(struct rte_vdev_device *dev)
rte_mempool_free(internals->mode6.mempool);
rte_bitmap_free(internals->vlan_filter_bmp);
rte_free(internals->vlan_filter_bmpmem);
- rte_free(eth_dev->data->dev_private);
- rte_free(eth_dev->data->mac_addrs);
rte_eth_dev_release_port(eth_dev);
@@ -3353,16 +3320,30 @@ bond_ethdev_configure(struct rte_eth_dev *dev)
unsigned i, j;
- /* If RSS is enabled, fill table and key with default values */
+ /*
+ * If RSS is enabled, fill table with default values and
+ * set key to the the value specified in port RSS configuration.
+ * Fall back to default RSS key if the key is not specified
+ */
if (dev->data->dev_conf.rxmode.mq_mode & ETH_MQ_RX_RSS) {
- dev->data->dev_conf.rx_adv_conf.rss_conf.rss_key = internals->rss_key;
- dev->data->dev_conf.rx_adv_conf.rss_conf.rss_key_len = 0;
- memcpy(internals->rss_key, default_rss_key, 40);
+ if (dev->data->dev_conf.rx_adv_conf.rss_conf.rss_key != NULL) {
+ internals->rss_key_len =
+ dev->data->dev_conf.rx_adv_conf.rss_conf.rss_key_len;
+ memcpy(internals->rss_key,
+ dev->data->dev_conf.rx_adv_conf.rss_conf.rss_key,
+ internals->rss_key_len);
+ } else {
+ internals->rss_key_len = sizeof(default_rss_key);
+ memcpy(internals->rss_key, default_rss_key,
+ internals->rss_key_len);
+ }
for (i = 0; i < RTE_DIM(internals->reta_conf); i++) {
internals->reta_conf[i].mask = ~0LL;
for (j = 0; j < RTE_RETA_GROUP_SIZE; j++)
- internals->reta_conf[i].reta[j] = j % dev->data->nb_rx_queues;
+ internals->reta_conf[i].reta[j] =
+ (i * RTE_RETA_GROUP_SIZE + j) %
+ dev->data->nb_rx_queues;
}
}
@@ -3618,7 +3599,7 @@ int bond_logtype;
RTE_INIT(bond_init_log)
{
- bond_logtype = rte_log_register("pmd.net.bon");
+ bond_logtype = rte_log_register("pmd.net.bond");
if (bond_logtype >= 0)
rte_log_set_level(bond_logtype, RTE_LOG_NOTICE);
}