-/*-
- * BSD LICENSE
- *
- * Copyright(c) 2010-2017 Intel Corporation. All rights reserved.
- * All rights reserved.
- *
- * Redistribution and use in source and binary forms, with or without
- * modification, are permitted provided that the following conditions
- * are met:
- *
- * * Redistributions of source code must retain the above copyright
- * notice, this list of conditions and the following disclaimer.
- * * Redistributions in binary form must reproduce the above copyright
- * notice, this list of conditions and the following disclaimer in
- * the documentation and/or other materials provided with the
- * distribution.
- * * Neither the name of Intel Corporation nor the names of its
- * contributors may be used to endorse or promote products derived
- * from this software without specific prior written permission.
- *
- * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
- * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
- * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
- * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
- * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
- * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
- * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
- * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
- * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
- * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
- * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(c) 2010-2017 Intel Corporation
*/
#include <stdlib.h>
#include <netinet/in.h>
#include <rte_mbuf.h>
#include <rte_malloc.h>
-#include <rte_ethdev.h>
+#include <rte_ethdev_driver.h>
+#include <rte_ethdev_vdev.h>
#include <rte_tcp.h>
#include <rte_udp.h>
#include <rte_ip.h>
#include <rte_ip_frag.h>
#include <rte_devargs.h>
#include <rte_kvargs.h>
-#include <rte_vdev.h>
+#include <rte_bus_vdev.h>
#include <rte_alarm.h>
#include <rte_cycles.h>
+#include <rte_string_fns.h>
#include "rte_eth_bond.h"
#include "rte_eth_bond_private.h"
#include "rte_eth_bond_8023ad_private.h"
#define REORDER_PERIOD_MS 10
+#define DEFAULT_POLLING_INTERVAL_10_MS (10)
#define HASH_L4_PORTS(h) ((h)->src_port ^ (h)->dst_port)
}
static inline uint8_t
-is_lacp_packets(uint16_t ethertype, uint8_t subtype, uint16_t vlan_tci)
+is_lacp_packets(uint16_t ethertype, uint8_t subtype, struct rte_mbuf *mbuf)
{
const uint16_t ether_type_slow_be = rte_be_to_cpu_16(ETHER_TYPE_SLOW);
- return !vlan_tci && (ethertype == ether_type_slow_be &&
+ return !((mbuf->ol_flags & PKT_RX_VLAN) ? mbuf->vlan_tci : 0) &&
+ (ethertype == ether_type_slow_be &&
(subtype == SLOW_SUBTYPE_MARKER || subtype == SLOW_SUBTYPE_LACP));
}
+/*****************************************************************************
+ * Flow director's setup for mode 4 optimization
+ */
+
+static struct rte_flow_item_eth flow_item_eth_type_8023ad = {
+ .dst.addr_bytes = { 0 },
+ .src.addr_bytes = { 0 },
+ .type = RTE_BE16(ETHER_TYPE_SLOW),
+};
+
+static struct rte_flow_item_eth flow_item_eth_mask_type_8023ad = {
+ .dst.addr_bytes = { 0 },
+ .src.addr_bytes = { 0 },
+ .type = 0xFFFF,
+};
+
+static struct rte_flow_item flow_item_8023ad[] = {
+ {
+ .type = RTE_FLOW_ITEM_TYPE_ETH,
+ .spec = &flow_item_eth_type_8023ad,
+ .last = NULL,
+ .mask = &flow_item_eth_mask_type_8023ad,
+ },
+ {
+ .type = RTE_FLOW_ITEM_TYPE_END,
+ .spec = NULL,
+ .last = NULL,
+ .mask = NULL,
+ }
+};
+
+const struct rte_flow_attr flow_attr_8023ad = {
+ .group = 0,
+ .priority = 0,
+ .ingress = 1,
+ .egress = 0,
+ .reserved = 0,
+};
+
+int
+bond_ethdev_8023ad_flow_verify(struct rte_eth_dev *bond_dev,
+ uint16_t slave_port) {
+ struct rte_eth_dev_info slave_info;
+ struct rte_flow_error error;
+ struct bond_dev_private *internals = (struct bond_dev_private *)
+ (bond_dev->data->dev_private);
+
+ const struct rte_flow_action_queue lacp_queue_conf = {
+ .index = 0,
+ };
+
+ const struct rte_flow_action actions[] = {
+ {
+ .type = RTE_FLOW_ACTION_TYPE_QUEUE,
+ .conf = &lacp_queue_conf
+ },
+ {
+ .type = RTE_FLOW_ACTION_TYPE_END,
+ }
+ };
+
+ int ret = rte_flow_validate(slave_port, &flow_attr_8023ad,
+ flow_item_8023ad, actions, &error);
+ if (ret < 0) {
+ RTE_BOND_LOG(ERR, "%s: %s (slave_port=%d queue_id=%d)",
+ __func__, error.message, slave_port,
+ internals->mode4.dedicated_queues.rx_qid);
+ return -1;
+ }
+
+ rte_eth_dev_info_get(slave_port, &slave_info);
+ if (slave_info.max_rx_queues < bond_dev->data->nb_rx_queues ||
+ slave_info.max_tx_queues < bond_dev->data->nb_tx_queues) {
+ RTE_BOND_LOG(ERR,
+ "%s: Slave %d capabilities doesn't allow to allocate additional queues",
+ __func__, slave_port);
+ return -1;
+ }
+
+ return 0;
+}
+
+int
+bond_8023ad_slow_pkt_hw_filter_supported(uint16_t port_id) {
+ struct rte_eth_dev *bond_dev = &rte_eth_devices[port_id];
+ struct bond_dev_private *internals = (struct bond_dev_private *)
+ (bond_dev->data->dev_private);
+ struct rte_eth_dev_info bond_info;
+ uint16_t idx;
+
+ /* Verify if all slaves in bonding supports flow director and */
+ if (internals->slave_count > 0) {
+ rte_eth_dev_info_get(bond_dev->data->port_id, &bond_info);
+
+ internals->mode4.dedicated_queues.rx_qid = bond_info.nb_rx_queues;
+ internals->mode4.dedicated_queues.tx_qid = bond_info.nb_tx_queues;
+
+ for (idx = 0; idx < internals->slave_count; idx++) {
+ if (bond_ethdev_8023ad_flow_verify(bond_dev,
+ internals->slaves[idx].port_id) != 0)
+ return -1;
+ }
+ }
+
+ return 0;
+}
+
+int
+bond_ethdev_8023ad_flow_set(struct rte_eth_dev *bond_dev, uint16_t slave_port) {
+
+ struct rte_flow_error error;
+ struct bond_dev_private *internals = (struct bond_dev_private *)
+ (bond_dev->data->dev_private);
+
+ struct rte_flow_action_queue lacp_queue_conf = {
+ .index = internals->mode4.dedicated_queues.rx_qid,
+ };
+
+ const struct rte_flow_action actions[] = {
+ {
+ .type = RTE_FLOW_ACTION_TYPE_QUEUE,
+ .conf = &lacp_queue_conf
+ },
+ {
+ .type = RTE_FLOW_ACTION_TYPE_END,
+ }
+ };
+
+ internals->mode4.dedicated_queues.flow[slave_port] = rte_flow_create(slave_port,
+ &flow_attr_8023ad, flow_item_8023ad, actions, &error);
+ if (internals->mode4.dedicated_queues.flow[slave_port] == NULL) {
+ RTE_BOND_LOG(ERR, "bond_ethdev_8023ad_flow_set: %s "
+ "(slave_port=%d queue_id=%d)",
+ error.message, slave_port,
+ internals->mode4.dedicated_queues.rx_qid);
+ return -1;
+ }
+
+ return 0;
+}
+
+static uint16_t
+bond_ethdev_rx_burst_8023ad_fast_queue(void *queue, struct rte_mbuf **bufs,
+ uint16_t nb_pkts)
+{
+ struct bond_rx_queue *bd_rx_q = (struct bond_rx_queue *)queue;
+ struct bond_dev_private *internals = bd_rx_q->dev_private;
+ 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;
+
+ /* Copy slave list to protect against slave up/down changes during tx
+ * bursting */
+ slave_count = internals->active_slave_count;
+ 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;
+
+ /* 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);
+ }
+
+ internals->active_slave = idx;
+
+ return num_rx_total;
+}
+
+static uint16_t
+bond_ethdev_tx_burst_8023ad_fast_queue(void *queue, struct rte_mbuf **bufs,
+ uint16_t nb_bufs)
+{
+ struct bond_tx_queue *bd_tx_q = (struct bond_tx_queue *)queue;
+ struct bond_dev_private *internals = bd_tx_q->dev_private;
+
+ uint16_t slave_port_ids[RTE_MAX_ETHPORTS];
+ uint16_t slave_count;
+
+ uint16_t dist_slave_port_ids[RTE_MAX_ETHPORTS];
+ uint16_t dist_slave_count;
+
+ /* 2-D array to sort mbufs for transmission on each slave into */
+ struct rte_mbuf *slave_bufs[RTE_MAX_ETHPORTS][nb_bufs];
+ /* Number of mbufs for transmission on each slave */
+ uint16_t slave_nb_bufs[RTE_MAX_ETHPORTS] = { 0 };
+ /* 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 total_tx_count = 0, total_tx_fail_count = 0;
+
+ uint16_t i, j;
+
+ if (unlikely(nb_bufs == 0))
+ return 0;
+
+ /* Copy slave list to protect against slave up/down changes during tx
+ * bursting */
+ slave_count = internals->active_slave_count;
+ if (unlikely(slave_count < 1))
+ return 0;
+
+ memcpy(slave_port_ids, internals->active_slaves,
+ sizeof(slave_port_ids[0]) * slave_count);
+
+
+ dist_slave_count = 0;
+ for (i = 0; i < slave_count; i++) {
+ struct port *port = &mode_8023ad_ports[slave_port_ids[i]];
+
+ if (ACTOR_STATE(port, DISTRIBUTING))
+ dist_slave_port_ids[dist_slave_count++] =
+ slave_port_ids[i];
+ }
+
+ if (unlikely(dist_slave_count < 1))
+ return 0;
+
+ /*
+ * Populate slaves mbuf with the packets which are to be sent on it
+ * selecting output slave using hash based on xmit policy
+ */
+ internals->burst_xmit_hash(bufs, nb_bufs, dist_slave_count,
+ bufs_slave_port_idxs);
+
+ for (i = 0; i < nb_bufs; i++) {
+ /* Populate slave mbuf arrays with mbufs for that slave. */
+ uint8_t slave_idx = bufs_slave_port_idxs[i];
+
+ slave_bufs[slave_idx][slave_nb_bufs[slave_idx]++] = bufs[i];
+ }
+
+
+ /* Send packet burst on each slave device */
+ for (i = 0; i < dist_slave_count; i++) {
+ if (slave_nb_bufs[i] == 0)
+ continue;
+
+ slave_tx_count = rte_eth_tx_burst(dist_slave_port_ids[i],
+ bd_tx_q->queue_id, slave_bufs[i],
+ slave_nb_bufs[i]);
+
+ total_tx_count += slave_tx_count;
+
+ /* 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] -
+ 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];
+ }
+ }
+ }
+
+ return total_tx_count;
+}
+
+
static uint16_t
bond_ethdev_rx_burst_8023ad(void *queue, struct rte_mbuf **bufs,
uint16_t nb_pkts)
const uint16_t ether_type_slow_be = rte_be_to_cpu_16(ETHER_TYPE_SLOW);
uint16_t num_rx_total = 0; /* Total number of received packets */
- uint8_t slaves[RTE_MAX_ETHPORTS];
- uint8_t slave_count, idx;
+ uint16_t slaves[RTE_MAX_ETHPORTS];
+ uint16_t slave_count, idx;
uint8_t collecting; /* current slave collecting status */
const uint8_t promisc = internals->promiscuous_en;
/* Handle slow protocol packets. */
while (j < num_rx_total) {
+
+ /* If packet is not pure L2 and is known, skip it */
+ if ((bufs[j]->packet_type & ~RTE_PTYPE_L2_ETHER) != 0) {
+ j++;
+ continue;
+ }
+
if (j + 3 < num_rx_total)
rte_prefetch0(rte_pktmbuf_mtod(bufs[j + 3], void *));
subtype = ((struct slow_protocol_frame *)hdr)->slow_protocol.subtype;
/* Remove packet from array if it is slow packet or slave is not
- * in collecting state or bondign interface is not in promiscus
+ * 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]->vlan_tci) ||
+ 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)))) {
#define MAX_CLIENTS_NUMBER 128
uint8_t active_clients;
struct client_stats_t {
- uint8_t port;
+ uint16_t port;
uint32_t ipv4_addr;
uint32_t ipv4_rx_packets;
uint32_t ipv4_tx_packets;
struct client_stats_t client_stats[MAX_CLIENTS_NUMBER];
static void
-update_client_stats(uint32_t addr, uint8_t port, uint32_t *TXorRXindicator)
+update_client_stats(uint32_t addr, uint16_t port, uint32_t *TXorRXindicator)
{
int i = 0;
static void
mode6_debug(const char __attribute__((unused)) *info, struct ether_hdr *eth_h,
- uint8_t port, uint32_t __attribute__((unused)) *burstnumber)
+ uint16_t port, uint32_t __attribute__((unused)) *burstnumber)
{
struct ipv4_hdr *ipv4_h;
#ifdef RTE_LIBRTE_BOND_DEBUG_ALB
uint16_t offset = get_vlan_offset(eth_h, ðer_type);
#ifdef RTE_LIBRTE_BOND_DEBUG_ALB
- snprintf(buf, 16, "%s", info);
+ strlcpy(buf, info, 16);
#endif
if (ether_type == rte_cpu_to_be_16(ETHER_TYPE_IPv4)) {
struct rte_mbuf *slave_bufs[RTE_MAX_ETHPORTS][nb_pkts];
uint16_t slave_nb_pkts[RTE_MAX_ETHPORTS] = { 0 };
- uint8_t num_of_slaves;
- uint8_t slaves[RTE_MAX_ETHPORTS];
+ uint16_t num_of_slaves;
+ uint16_t slaves[RTE_MAX_ETHPORTS];
uint16_t num_tx_total = 0, num_tx_slave;
(word_src_addr[3] ^ word_dst_addr[3]);
}
-uint16_t
-xmit_l2_hash(const struct rte_mbuf *buf, uint8_t slave_count)
+
+void
+burst_xmit_l2_hash(struct rte_mbuf **buf, uint16_t nb_pkts,
+ uint8_t slave_count, uint16_t *slaves)
{
- struct ether_hdr *eth_hdr = rte_pktmbuf_mtod(buf, struct ether_hdr *);
+ struct ether_hdr *eth_hdr;
+ uint32_t hash;
+ int i;
+
+ for (i = 0; i < nb_pkts; i++) {
+ eth_hdr = rte_pktmbuf_mtod(buf[i], struct ether_hdr *);
- uint32_t hash = ether_hash(eth_hdr);
+ hash = ether_hash(eth_hdr);
- return (hash ^= hash >> 8) % slave_count;
+ slaves[i] = (hash ^= hash >> 8) % slave_count;
+ }
}
-uint16_t
-xmit_l23_hash(const struct rte_mbuf *buf, uint8_t slave_count)
+void
+burst_xmit_l23_hash(struct rte_mbuf **buf, uint16_t nb_pkts,
+ uint8_t slave_count, uint16_t *slaves)
{
- struct ether_hdr *eth_hdr = rte_pktmbuf_mtod(buf, struct ether_hdr *);
- uint16_t proto = eth_hdr->ether_type;
- size_t vlan_offset = get_vlan_offset(eth_hdr, &proto);
- uint32_t hash, l3hash = 0;
+ uint16_t i;
+ struct ether_hdr *eth_hdr;
+ uint16_t proto;
+ size_t vlan_offset;
+ uint32_t hash, l3hash;
- hash = ether_hash(eth_hdr);
+ for (i = 0; i < nb_pkts; i++) {
+ eth_hdr = rte_pktmbuf_mtod(buf[i], struct ether_hdr *);
+ l3hash = 0;
- if (rte_cpu_to_be_16(ETHER_TYPE_IPv4) == proto) {
- struct ipv4_hdr *ipv4_hdr = (struct ipv4_hdr *)
- ((char *)(eth_hdr + 1) + vlan_offset);
- l3hash = ipv4_hash(ipv4_hdr);
+ proto = eth_hdr->ether_type;
+ hash = ether_hash(eth_hdr);
- } else if (rte_cpu_to_be_16(ETHER_TYPE_IPv6) == proto) {
- struct ipv6_hdr *ipv6_hdr = (struct ipv6_hdr *)
- ((char *)(eth_hdr + 1) + vlan_offset);
- l3hash = ipv6_hash(ipv6_hdr);
- }
+ vlan_offset = get_vlan_offset(eth_hdr, &proto);
- hash = hash ^ l3hash;
- hash ^= hash >> 16;
- hash ^= hash >> 8;
+ if (rte_cpu_to_be_16(ETHER_TYPE_IPv4) == proto) {
+ struct ipv4_hdr *ipv4_hdr = (struct ipv4_hdr *)
+ ((char *)(eth_hdr + 1) + vlan_offset);
+ l3hash = ipv4_hash(ipv4_hdr);
- return hash % slave_count;
-}
+ } else if (rte_cpu_to_be_16(ETHER_TYPE_IPv6) == proto) {
+ struct ipv6_hdr *ipv6_hdr = (struct ipv6_hdr *)
+ ((char *)(eth_hdr + 1) + vlan_offset);
+ l3hash = ipv6_hash(ipv6_hdr);
+ }
-uint16_t
-xmit_l34_hash(const struct rte_mbuf *buf, uint8_t slave_count)
-{
- struct ether_hdr *eth_hdr = rte_pktmbuf_mtod(buf, struct ether_hdr *);
- uint16_t proto = eth_hdr->ether_type;
- size_t vlan_offset = get_vlan_offset(eth_hdr, &proto);
+ hash = hash ^ l3hash;
+ hash ^= hash >> 16;
+ hash ^= hash >> 8;
- struct udp_hdr *udp_hdr = NULL;
- struct tcp_hdr *tcp_hdr = NULL;
- uint32_t hash, l3hash = 0, l4hash = 0;
+ slaves[i] = hash % slave_count;
+ }
+}
- if (rte_cpu_to_be_16(ETHER_TYPE_IPv4) == proto) {
- struct ipv4_hdr *ipv4_hdr = (struct ipv4_hdr *)
- ((char *)(eth_hdr + 1) + vlan_offset);
- size_t ip_hdr_offset;
+void
+burst_xmit_l34_hash(struct rte_mbuf **buf, uint16_t nb_pkts,
+ uint8_t slave_count, uint16_t *slaves)
+{
+ struct ether_hdr *eth_hdr;
+ uint16_t proto;
+ size_t vlan_offset;
+ int i;
- l3hash = ipv4_hash(ipv4_hdr);
+ struct udp_hdr *udp_hdr;
+ struct tcp_hdr *tcp_hdr;
+ uint32_t hash, l3hash, l4hash;
- /* there is no L4 header in fragmented packet */
- if (likely(rte_ipv4_frag_pkt_is_fragmented(ipv4_hdr) == 0)) {
- ip_hdr_offset = (ipv4_hdr->version_ihl & IPV4_HDR_IHL_MASK) *
+ for (i = 0; i < nb_pkts; i++) {
+ eth_hdr = rte_pktmbuf_mtod(buf[i], struct ether_hdr *);
+ proto = eth_hdr->ether_type;
+ vlan_offset = get_vlan_offset(eth_hdr, &proto);
+ l3hash = 0;
+ l4hash = 0;
+
+ if (rte_cpu_to_be_16(ETHER_TYPE_IPv4) == proto) {
+ struct ipv4_hdr *ipv4_hdr = (struct ipv4_hdr *)
+ ((char *)(eth_hdr + 1) + vlan_offset);
+ size_t ip_hdr_offset;
+
+ l3hash = ipv4_hash(ipv4_hdr);
+
+ /* there is no L4 header in fragmented packet */
+ if (likely(rte_ipv4_frag_pkt_is_fragmented(ipv4_hdr)
+ == 0)) {
+ ip_hdr_offset = (ipv4_hdr->version_ihl
+ & IPV4_HDR_IHL_MASK) *
IPV4_IHL_MULTIPLIER;
- if (ipv4_hdr->next_proto_id == IPPROTO_TCP) {
- tcp_hdr = (struct tcp_hdr *)((char *)ipv4_hdr +
- ip_hdr_offset);
+ if (ipv4_hdr->next_proto_id == IPPROTO_TCP) {
+ tcp_hdr = (struct tcp_hdr *)
+ ((char *)ipv4_hdr +
+ ip_hdr_offset);
+ l4hash = HASH_L4_PORTS(tcp_hdr);
+ } else if (ipv4_hdr->next_proto_id ==
+ IPPROTO_UDP) {
+ udp_hdr = (struct udp_hdr *)
+ ((char *)ipv4_hdr +
+ ip_hdr_offset);
+ l4hash = HASH_L4_PORTS(udp_hdr);
+ }
+ }
+ } else if (rte_cpu_to_be_16(ETHER_TYPE_IPv6) == proto) {
+ struct ipv6_hdr *ipv6_hdr = (struct ipv6_hdr *)
+ ((char *)(eth_hdr + 1) + vlan_offset);
+ l3hash = ipv6_hash(ipv6_hdr);
+
+ if (ipv6_hdr->proto == IPPROTO_TCP) {
+ tcp_hdr = (struct tcp_hdr *)(ipv6_hdr + 1);
l4hash = HASH_L4_PORTS(tcp_hdr);
- } else if (ipv4_hdr->next_proto_id == IPPROTO_UDP) {
- udp_hdr = (struct udp_hdr *)((char *)ipv4_hdr +
- ip_hdr_offset);
+ } else if (ipv6_hdr->proto == IPPROTO_UDP) {
+ udp_hdr = (struct udp_hdr *)(ipv6_hdr + 1);
l4hash = HASH_L4_PORTS(udp_hdr);
}
}
- } else if (rte_cpu_to_be_16(ETHER_TYPE_IPv6) == proto) {
- struct ipv6_hdr *ipv6_hdr = (struct ipv6_hdr *)
- ((char *)(eth_hdr + 1) + vlan_offset);
- l3hash = ipv6_hash(ipv6_hdr);
-
- if (ipv6_hdr->proto == IPPROTO_TCP) {
- tcp_hdr = (struct tcp_hdr *)(ipv6_hdr + 1);
- l4hash = HASH_L4_PORTS(tcp_hdr);
- } else if (ipv6_hdr->proto == IPPROTO_UDP) {
- udp_hdr = (struct udp_hdr *)(ipv6_hdr + 1);
- l4hash = HASH_L4_PORTS(udp_hdr);
- }
- }
- hash = l3hash ^ l4hash;
- hash ^= hash >> 16;
- hash ^= hash >> 8;
+ hash = l3hash ^ l4hash;
+ hash ^= hash >> 16;
+ hash ^= hash >> 8;
- return hash % slave_count;
+ slaves[i] = hash % slave_count;
+ }
}
struct bwg_slave {
}
static void
-bandwidth_left(uint8_t port_id, uint64_t load, uint8_t update_idx,
+bandwidth_left(uint16_t port_id, uint64_t load, uint8_t update_idx,
struct bwg_slave *bwg_slave)
{
struct rte_eth_link link_status;
- rte_eth_link_get(port_id, &link_status);
+ rte_eth_link_get_nowait(port_id, &link_status);
uint64_t link_bwg = link_status.link_speed * 1000000ULL / 8;
if (link_bwg == 0)
return;
struct rte_eth_dev *primary_port =
&rte_eth_devices[internals->primary_port];
uint16_t num_tx_total = 0;
- uint8_t i, j;
+ uint16_t i, j;
- uint8_t num_of_slaves = internals->active_slave_count;
- uint8_t slaves[RTE_MAX_ETHPORTS];
+ uint16_t num_of_slaves = internals->active_slave_count;
+ uint16_t slaves[RTE_MAX_ETHPORTS];
struct ether_hdr *ether_hdr;
struct ether_addr primary_slave_addr;
uint16_t slave_bufs_pkts[RTE_MAX_ETHPORTS + 1] = { 0 };
/*
- * We create separate transmit buffers for update packets as they wont be
- * counted in num_tx_total.
+ * We create separate transmit buffers for update packets as they won't
+ * be counted in num_tx_total.
*/
struct rte_mbuf *update_bufs[RTE_MAX_ETHPORTS][ALB_HASH_TABLE_SIZE];
uint16_t update_bufs_pkts[RTE_MAX_ETHPORTS] = { 0 };
uint16_t num_send, num_not_send = 0;
uint16_t num_tx_total = 0;
- uint8_t slave_idx;
+ uint16_t slave_idx;
int i, j;
static uint16_t
bond_ethdev_tx_burst_balance(void *queue, struct rte_mbuf **bufs,
- uint16_t nb_pkts)
+ uint16_t nb_bufs)
{
- struct bond_dev_private *internals;
- struct bond_tx_queue *bd_tx_q;
+ struct bond_tx_queue *bd_tx_q = (struct bond_tx_queue *)queue;
+ struct bond_dev_private *internals = bd_tx_q->dev_private;
- uint8_t num_of_slaves;
- uint8_t slaves[RTE_MAX_ETHPORTS];
+ uint16_t slave_port_ids[RTE_MAX_ETHPORTS];
+ uint16_t slave_count;
- uint16_t num_tx_total = 0, num_tx_slave = 0, tx_fail_total = 0;
+ /* Array to sort mbufs for transmission on each slave into */
+ struct rte_mbuf *slave_bufs[RTE_MAX_ETHPORTS][nb_bufs];
+ /* Number of mbufs for transmission on each slave */
+ uint16_t slave_nb_bufs[RTE_MAX_ETHPORTS] = { 0 };
+ /* Mapping array generated by hash function to map mbufs to slaves */
+ uint16_t bufs_slave_port_idxs[nb_bufs];
- int i, op_slave_id;
+ uint16_t slave_tx_count, slave_tx_fail_count[RTE_MAX_ETHPORTS] = { 0 };
+ uint16_t total_tx_count = 0, total_tx_fail_count = 0;
- struct rte_mbuf *slave_bufs[RTE_MAX_ETHPORTS][nb_pkts];
- uint16_t slave_nb_pkts[RTE_MAX_ETHPORTS] = { 0 };
+ uint16_t i, j;
- bd_tx_q = (struct bond_tx_queue *)queue;
- internals = bd_tx_q->dev_private;
+ if (unlikely(nb_bufs == 0))
+ return 0;
/* Copy slave list to protect against slave up/down changes during tx
* bursting */
- num_of_slaves = internals->active_slave_count;
- memcpy(slaves, internals->active_slaves,
- sizeof(internals->active_slaves[0]) * num_of_slaves);
+ slave_count = internals->active_slave_count;
+ if (unlikely(slave_count < 1))
+ return 0;
- if (num_of_slaves < 1)
- return num_tx_total;
+ memcpy(slave_port_ids, internals->active_slaves,
+ sizeof(slave_port_ids[0]) * slave_count);
- /* Populate slaves mbuf with the packets which are to be sent on it */
- for (i = 0; i < nb_pkts; i++) {
- /* Select output slave using hash based on xmit policy */
- op_slave_id = internals->xmit_hash(bufs[i], num_of_slaves);
+ /*
+ * Populate slaves mbuf with the packets which are to be sent on it
+ * selecting output slave using hash based on xmit policy
+ */
+ internals->burst_xmit_hash(bufs, nb_bufs, slave_count,
+ bufs_slave_port_idxs);
- /* Populate slave mbuf arrays with mbufs for that slave */
- slave_bufs[op_slave_id][slave_nb_pkts[op_slave_id]++] = bufs[i];
+ for (i = 0; i < nb_bufs; i++) {
+ /* Populate slave mbuf arrays with mbufs for that slave. */
+ uint8_t slave_idx = bufs_slave_port_idxs[i];
+
+ slave_bufs[slave_idx][slave_nb_bufs[slave_idx]++] = bufs[i];
}
/* Send packet burst on each slave device */
- for (i = 0; i < num_of_slaves; i++) {
- if (slave_nb_pkts[i] > 0) {
- num_tx_slave = rte_eth_tx_burst(slaves[i], bd_tx_q->queue_id,
- slave_bufs[i], slave_nb_pkts[i]);
+ for (i = 0; i < slave_count; i++) {
+ if (slave_nb_bufs[i] == 0)
+ continue;
- /* if tx burst fails move packets to end of bufs */
- if (unlikely(num_tx_slave < slave_nb_pkts[i])) {
- int slave_tx_fail_count = slave_nb_pkts[i] - num_tx_slave;
+ slave_tx_count = rte_eth_tx_burst(slave_port_ids[i],
+ bd_tx_q->queue_id, slave_bufs[i],
+ slave_nb_bufs[i]);
- tx_fail_total += slave_tx_fail_count;
- memcpy(&bufs[nb_pkts - tx_fail_total],
- &slave_bufs[i][num_tx_slave],
- slave_tx_fail_count * sizeof(bufs[0]));
+ total_tx_count += slave_tx_count;
+
+ /* 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] -
+ 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];
}
+ }
+ }
- num_tx_total += num_tx_slave;
+ /*
+ * 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];
+ }
}
}
- return num_tx_total;
+ return total_tx_count;
}
static uint16_t
bond_ethdev_tx_burst_8023ad(void *queue, struct rte_mbuf **bufs,
- uint16_t nb_pkts)
+ uint16_t nb_bufs)
{
- struct bond_dev_private *internals;
- struct bond_tx_queue *bd_tx_q;
+ struct bond_tx_queue *bd_tx_q = (struct bond_tx_queue *)queue;
+ struct bond_dev_private *internals = bd_tx_q->dev_private;
- uint8_t num_of_slaves;
- uint8_t slaves[RTE_MAX_ETHPORTS];
- /* positions in slaves, not ID */
- uint8_t distributing_offsets[RTE_MAX_ETHPORTS];
- uint8_t distributing_count;
+ uint16_t slave_port_ids[RTE_MAX_ETHPORTS];
+ uint16_t slave_count;
- uint16_t num_tx_slave, num_tx_total = 0, num_tx_fail_total = 0;
- uint16_t i, j, op_slave_idx;
- const uint16_t buffs_size = nb_pkts + BOND_MODE_8023AX_SLAVE_TX_PKTS + 1;
+ uint16_t dist_slave_port_ids[RTE_MAX_ETHPORTS];
+ uint16_t dist_slave_count;
- /* Allocate additional packets in case 8023AD mode. */
- struct rte_mbuf *slave_bufs[RTE_MAX_ETHPORTS][buffs_size];
- void *slow_pkts[BOND_MODE_8023AX_SLAVE_TX_PKTS] = { NULL };
+ /* 2-D array to sort mbufs for transmission on each slave into */
+ struct rte_mbuf *slave_bufs[RTE_MAX_ETHPORTS][nb_bufs];
+ /* Number of mbufs for transmission on each slave */
+ uint16_t slave_nb_bufs[RTE_MAX_ETHPORTS] = { 0 };
+ /* Mapping array generated by hash function to map mbufs to slaves */
+ uint16_t bufs_slave_port_idxs[RTE_MAX_ETHPORTS] = { 0 };
- /* Total amount of packets in slave_bufs */
- uint16_t slave_nb_pkts[RTE_MAX_ETHPORTS] = { 0 };
- /* Slow packets placed in each slave */
- uint8_t slave_slow_nb_pkts[RTE_MAX_ETHPORTS] = { 0 };
+ uint16_t slave_tx_count, slave_tx_fail_count[RTE_MAX_ETHPORTS] = { 0 };
+ uint16_t total_tx_count = 0, total_tx_fail_count = 0;
- bd_tx_q = (struct bond_tx_queue *)queue;
- internals = bd_tx_q->dev_private;
+ uint16_t i, j;
+
+ if (unlikely(nb_bufs == 0))
+ return 0;
/* Copy slave list to protect against slave up/down changes during tx
* bursting */
- num_of_slaves = internals->active_slave_count;
- if (num_of_slaves < 1)
- return num_tx_total;
-
- memcpy(slaves, internals->active_slaves, sizeof(slaves[0]) * num_of_slaves);
-
- distributing_count = 0;
- for (i = 0; i < num_of_slaves; i++) {
- struct port *port = &mode_8023ad_ports[slaves[i]];
+ slave_count = internals->active_slave_count;
+ if (unlikely(slave_count < 1))
+ return 0;
- slave_slow_nb_pkts[i] = rte_ring_dequeue_burst(port->tx_ring,
- slow_pkts, BOND_MODE_8023AX_SLAVE_TX_PKTS,
- NULL);
- slave_nb_pkts[i] = slave_slow_nb_pkts[i];
+ memcpy(slave_port_ids, internals->active_slaves,
+ sizeof(slave_port_ids[0]) * slave_count);
- for (j = 0; j < slave_slow_nb_pkts[i]; j++)
- slave_bufs[i][j] = slow_pkts[j];
+ dist_slave_count = 0;
+ for (i = 0; i < slave_count; i++) {
+ struct port *port = &mode_8023ad_ports[slave_port_ids[i]];
if (ACTOR_STATE(port, DISTRIBUTING))
- distributing_offsets[distributing_count++] = i;
+ dist_slave_port_ids[dist_slave_count++] =
+ slave_port_ids[i];
}
- if (likely(distributing_count > 0)) {
- /* Populate slaves mbuf with the packets which are to be sent on it */
- for (i = 0; i < nb_pkts; i++) {
- /* Select output slave using hash based on xmit policy */
- op_slave_idx = internals->xmit_hash(bufs[i], distributing_count);
+ if (likely(dist_slave_count > 1)) {
+
+ /*
+ * Populate slaves mbuf with the packets which are to be sent
+ * on it, selecting output slave using hash based on xmit policy
+ */
+ internals->burst_xmit_hash(bufs, nb_bufs, dist_slave_count,
+ bufs_slave_port_idxs);
- /* Populate slave mbuf arrays with mbufs for that slave. Use only
- * slaves that are currently distributing. */
- uint8_t slave_offset = distributing_offsets[op_slave_idx];
- slave_bufs[slave_offset][slave_nb_pkts[slave_offset]] = bufs[i];
- slave_nb_pkts[slave_offset]++;
+ for (i = 0; i < nb_bufs; i++) {
+ /*
+ * Populate slave mbuf arrays with mbufs for that
+ * slave
+ */
+ uint8_t slave_idx = bufs_slave_port_idxs[i];
+
+ slave_bufs[slave_idx][slave_nb_bufs[slave_idx]++] =
+ bufs[i];
}
- }
- /* Send packet burst on each slave device */
- for (i = 0; i < num_of_slaves; i++) {
- if (slave_nb_pkts[i] == 0)
- continue;
- num_tx_slave = rte_eth_tx_burst(slaves[i], bd_tx_q->queue_id,
- slave_bufs[i], slave_nb_pkts[i]);
+ /* Send packet burst on each slave device */
+ for (i = 0; i < dist_slave_count; i++) {
+ if (slave_nb_bufs[i] == 0)
+ continue;
- /* If tx burst fails drop slow packets */
- for ( ; num_tx_slave < slave_slow_nb_pkts[i]; num_tx_slave++)
- rte_pktmbuf_free(slave_bufs[i][num_tx_slave]);
+ slave_tx_count = rte_eth_tx_burst(
+ dist_slave_port_ids[i],
+ bd_tx_q->queue_id, slave_bufs[i],
+ slave_nb_bufs[i]);
+
+ total_tx_count += slave_tx_count;
+
+ /* 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] -
+ 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];
+ }
+ }
+ }
+ }
+ }
- num_tx_total += num_tx_slave - slave_slow_nb_pkts[i];
- num_tx_fail_total += slave_nb_pkts[i] - num_tx_slave;
+ /* 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 rte_mbuf *ctrl_pkt = NULL;
- /* If tx burst fails move packets to end of bufs */
- if (unlikely(num_tx_slave < slave_nb_pkts[i])) {
- uint16_t j = nb_pkts - num_tx_fail_total;
- for ( ; num_tx_slave < slave_nb_pkts[i]; j++, num_tx_slave++)
- bufs[j] = slave_bufs[i][num_tx_slave];
+ if (likely(rte_ring_empty(port->tx_ring)))
+ continue;
+
+ if (rte_ring_dequeue(port->tx_ring,
+ (void **)&ctrl_pkt) != -ENOENT) {
+ slave_tx_count = rte_eth_tx_burst(slave_port_ids[i],
+ bd_tx_q->queue_id, &ctrl_pkt, 1);
+ /*
+ * re-enqueue LAG control plane packets to buffering
+ * ring if transmission fails so the packet isn't lost.
+ */
+ if (slave_tx_count != 1)
+ rte_ring_enqueue(port->tx_ring, ctrl_pkt);
}
}
- return num_tx_total;
+ return total_tx_count;
}
static uint16_t
struct bond_tx_queue *bd_tx_q;
uint8_t tx_failed_flag = 0, num_of_slaves;
- uint8_t slaves[RTE_MAX_ETHPORTS];
+ uint16_t slaves[RTE_MAX_ETHPORTS];
uint16_t max_nb_of_tx_pkts = 0;
}
void
-link_properties_set(struct rte_eth_dev *bonded_eth_dev,
- struct rte_eth_link *slave_dev_link)
+link_properties_set(struct rte_eth_dev *ethdev, struct rte_eth_link *slave_link)
{
- struct rte_eth_link *bonded_dev_link = &bonded_eth_dev->data->dev_link;
- struct bond_dev_private *internals = bonded_eth_dev->data->dev_private;
+ struct bond_dev_private *bond_ctx = ethdev->data->dev_private;
- if (slave_dev_link->link_status &&
- bonded_eth_dev->data->dev_started) {
- bonded_dev_link->link_duplex = slave_dev_link->link_duplex;
- bonded_dev_link->link_speed = slave_dev_link->link_speed;
+ if (bond_ctx->mode == BONDING_MODE_8023AD) {
+ /**
+ * If in mode 4 then save the link properties of the first
+ * slave, all subsequent slaves must match these properties
+ */
+ struct rte_eth_link *bond_link = &bond_ctx->mode4.slave_link;
- internals->link_props_set = 1;
+ bond_link->link_autoneg = slave_link->link_autoneg;
+ bond_link->link_duplex = slave_link->link_duplex;
+ bond_link->link_speed = slave_link->link_speed;
+ } else {
+ /**
+ * In any other mode the link properties are set to default
+ * values of AUTONEG/DUPLEX
+ */
+ ethdev->data->dev_link.link_autoneg = ETH_LINK_AUTONEG;
+ ethdev->data->dev_link.link_duplex = ETH_LINK_FULL_DUPLEX;
}
}
-void
-link_properties_reset(struct rte_eth_dev *bonded_eth_dev)
+int
+link_properties_valid(struct rte_eth_dev *ethdev,
+ struct rte_eth_link *slave_link)
{
- struct bond_dev_private *internals = bonded_eth_dev->data->dev_private;
-
- memset(&(bonded_eth_dev->data->dev_link), 0,
- sizeof(bonded_eth_dev->data->dev_link));
+ struct bond_dev_private *bond_ctx = ethdev->data->dev_private;
- internals->link_props_set = 0;
-}
+ if (bond_ctx->mode == BONDING_MODE_8023AD) {
+ struct rte_eth_link *bond_link = &bond_ctx->mode4.slave_link;
-int
-link_properties_valid(struct rte_eth_link *bonded_dev_link,
- struct rte_eth_link *slave_dev_link)
-{
- if (bonded_dev_link->link_duplex != slave_dev_link->link_duplex ||
- bonded_dev_link->link_speed != slave_dev_link->link_speed)
- return -1;
+ if (bond_link->link_duplex != slave_link->link_duplex ||
+ bond_link->link_autoneg != slave_link->link_autoneg ||
+ bond_link->link_speed != slave_link->link_speed)
+ return -1;
+ }
return 0;
}
case BONDING_MODE_BALANCE:
case BONDING_MODE_BROADCAST:
for (i = 0; i < internals->slave_count; i++) {
- if (mac_address_set(&rte_eth_devices[internals->slaves[i].port_id],
+ if (rte_eth_dev_default_mac_addr_set(
+ internals->slaves[i].port_id,
bonded_eth_dev->data->mac_addrs)) {
RTE_BOND_LOG(ERR, "Failed to update port Id %d MAC address",
internals->slaves[i].port_id);
for (i = 0; i < internals->slave_count; i++) {
if (internals->slaves[i].port_id ==
internals->current_primary_port) {
- if (mac_address_set(&rte_eth_devices[internals->primary_port],
+ if (rte_eth_dev_default_mac_addr_set(
+ internals->primary_port,
bonded_eth_dev->data->mac_addrs)) {
RTE_BOND_LOG(ERR, "Failed to update port Id %d MAC address",
internals->current_primary_port);
return -1;
}
} else {
- if (mac_address_set(
- &rte_eth_devices[internals->slaves[i].port_id],
+ if (rte_eth_dev_default_mac_addr_set(
+ internals->slaves[i].port_id,
&internals->slaves[i].persisted_mac_addr)) {
RTE_BOND_LOG(ERR, "Failed to update port Id %d MAC address",
internals->slaves[i].port_id);
if (bond_mode_8023ad_enable(eth_dev) != 0)
return -1;
- eth_dev->rx_pkt_burst = bond_ethdev_rx_burst_8023ad;
- eth_dev->tx_pkt_burst = bond_ethdev_tx_burst_8023ad;
- RTE_LOG(WARNING, PMD,
- "Using mode 4, it is necessary to do TX burst and RX burst "
- "at least every 100ms.\n");
+ if (internals->mode4.dedicated_queues.enabled == 0) {
+ eth_dev->rx_pkt_burst = bond_ethdev_rx_burst_8023ad;
+ eth_dev->tx_pkt_burst = bond_ethdev_tx_burst_8023ad;
+ RTE_LOG(WARNING, PMD,
+ "Using mode 4, it is necessary to do TX burst "
+ "and RX burst at least every 100ms.\n");
+ } else {
+ /* Use flow director's optimization */
+ eth_dev->rx_pkt_burst =
+ bond_ethdev_rx_burst_8023ad_fast_queue;
+ eth_dev->tx_pkt_burst =
+ bond_ethdev_tx_burst_8023ad_fast_queue;
+ }
break;
case BONDING_MODE_TLB:
eth_dev->tx_pkt_burst = bond_ethdev_tx_burst_tlb;
return 0;
}
+
+static int
+slave_configure_slow_queue(struct rte_eth_dev *bonded_eth_dev,
+ struct rte_eth_dev *slave_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];
+
+ if (port->slow_pool == NULL) {
+ char mem_name[256];
+ int slave_id = slave_eth_dev->data->port_id;
+
+ snprintf(mem_name, RTE_DIM(mem_name), "slave_port%u_slow_pool",
+ slave_id);
+ port->slow_pool = rte_pktmbuf_pool_create(mem_name, 8191,
+ 250, 0, RTE_MBUF_DEFAULT_BUF_SIZE,
+ slave_eth_dev->data->numa_node);
+
+ /* Any memory allocation failure in initialization is critical because
+ * resources can't be free, so reinitialization is impossible. */
+ if (port->slow_pool == NULL) {
+ rte_panic("Slave %u: Failed to create memory pool '%s': %s\n",
+ slave_id, mem_name, rte_strerror(rte_errno));
+ }
+ }
+
+ if (internals->mode4.dedicated_queues.enabled == 1) {
+ /* Configure slow Rx queue */
+
+ errval = rte_eth_rx_queue_setup(slave_eth_dev->data->port_id,
+ internals->mode4.dedicated_queues.rx_qid, 128,
+ rte_eth_dev_socket_id(slave_eth_dev->data->port_id),
+ NULL, port->slow_pool);
+ if (errval != 0) {
+ RTE_BOND_LOG(ERR,
+ "rte_eth_rx_queue_setup: port=%d queue_id %d, err (%d)",
+ slave_eth_dev->data->port_id,
+ internals->mode4.dedicated_queues.rx_qid,
+ errval);
+ return errval;
+ }
+
+ errval = rte_eth_tx_queue_setup(slave_eth_dev->data->port_id,
+ internals->mode4.dedicated_queues.tx_qid, 512,
+ rte_eth_dev_socket_id(slave_eth_dev->data->port_id),
+ NULL);
+ if (errval != 0) {
+ RTE_BOND_LOG(ERR,
+ "rte_eth_tx_queue_setup: port=%d queue_id %d, err (%d)",
+ slave_eth_dev->data->port_id,
+ internals->mode4.dedicated_queues.tx_qid,
+ errval);
+ return errval;
+ }
+ }
+ return 0;
+}
+
int
slave_configure(struct rte_eth_dev *bonded_eth_dev,
struct rte_eth_dev *slave_eth_dev)
{
struct bond_rx_queue *bd_rx_q;
struct bond_tx_queue *bd_tx_q;
+ uint16_t nb_rx_queues;
+ uint16_t nb_tx_queues;
int errval;
uint16_t q_id;
+ struct rte_flow_error flow_error;
+
+ struct bond_dev_private *internals = (struct bond_dev_private *)
+ bonded_eth_dev->data->dev_private;
/* Stop slave */
rte_eth_dev_stop(slave_eth_dev->data->port_id);
bonded_eth_dev->data->dev_conf.rxmode.mq_mode;
}
- slave_eth_dev->data->dev_conf.rxmode.hw_vlan_filter =
- bonded_eth_dev->data->dev_conf.rxmode.hw_vlan_filter;
+ if (bonded_eth_dev->data->dev_conf.rxmode.offloads &
+ DEV_RX_OFFLOAD_VLAN_FILTER)
+ slave_eth_dev->data->dev_conf.rxmode.offloads |=
+ DEV_RX_OFFLOAD_VLAN_FILTER;
+ else
+ slave_eth_dev->data->dev_conf.rxmode.offloads &=
+ ~DEV_RX_OFFLOAD_VLAN_FILTER;
+
+ nb_rx_queues = bonded_eth_dev->data->nb_rx_queues;
+ nb_tx_queues = bonded_eth_dev->data->nb_tx_queues;
+
+ if (internals->mode == BONDING_MODE_8023AD) {
+ if (internals->mode4.dedicated_queues.enabled == 1) {
+ nb_rx_queues++;
+ nb_tx_queues++;
+ }
+ }
+
+ errval = rte_eth_dev_set_mtu(slave_eth_dev->data->port_id,
+ bonded_eth_dev->data->mtu);
+ if (errval != 0 && errval != -ENOTSUP) {
+ RTE_BOND_LOG(ERR, "rte_eth_dev_set_mtu: port %u, err (%d)",
+ slave_eth_dev->data->port_id, errval);
+ return errval;
+ }
/* Configure device */
errval = rte_eth_dev_configure(slave_eth_dev->data->port_id,
- bonded_eth_dev->data->nb_rx_queues,
- bonded_eth_dev->data->nb_tx_queues,
+ nb_rx_queues, nb_tx_queues,
&(slave_eth_dev->data->dev_conf));
if (errval != 0) {
RTE_BOND_LOG(ERR, "Cannot configure slave device: port %u , err (%d)",
&bd_tx_q->tx_conf);
if (errval != 0) {
RTE_BOND_LOG(ERR,
- "rte_eth_tx_queue_setup: port=%d queue_id %d, err (%d)",
- slave_eth_dev->data->port_id, q_id, errval);
+ "rte_eth_tx_queue_setup: port=%d queue_id %d, err (%d)",
+ slave_eth_dev->data->port_id, q_id, errval);
return errval;
}
}
+ if (internals->mode == BONDING_MODE_8023AD &&
+ internals->mode4.dedicated_queues.enabled == 1) {
+ if (slave_configure_slow_queue(bonded_eth_dev, slave_eth_dev)
+ != 0)
+ return errval;
+
+ if (bond_ethdev_8023ad_flow_verify(bonded_eth_dev,
+ slave_eth_dev->data->port_id) != 0) {
+ RTE_BOND_LOG(ERR,
+ "rte_eth_tx_queue_setup: port=%d queue_id %d, err (%d)",
+ slave_eth_dev->data->port_id, q_id, errval);
+ return -1;
+ }
+
+ if (internals->mode4.dedicated_queues.flow[slave_eth_dev->data->port_id] != NULL)
+ rte_flow_destroy(slave_eth_dev->data->port_id,
+ internals->mode4.dedicated_queues.flow[slave_eth_dev->data->port_id],
+ &flow_error);
+
+ bond_ethdev_8023ad_flow_set(bonded_eth_dev,
+ slave_eth_dev->data->port_id);
+ }
+
/* Start device */
errval = rte_eth_dev_start(slave_eth_dev->data->port_id);
if (errval != 0) {
}
/* If lsc interrupt is set, check initial slave's link status */
- if (slave_eth_dev->data->dev_flags & RTE_ETH_DEV_INTR_LSC)
+ if (slave_eth_dev->data->dev_flags & RTE_ETH_DEV_INTR_LSC) {
+ slave_eth_dev->dev_ops->link_update(slave_eth_dev, 0);
bond_ethdev_lsc_event_callback(slave_eth_dev->data->port_id,
- RTE_ETH_EVENT_INTR_LSC, &bonded_eth_dev->data->port_id);
+ RTE_ETH_EVENT_INTR_LSC, &bonded_eth_dev->data->port_id,
+ NULL);
+ }
return 0;
}
void
bond_ethdev_primary_set(struct bond_dev_private *internals,
- uint8_t slave_port_id)
+ uint16_t slave_port_id)
{
int i;
if (internals->slave_count == 0) {
RTE_BOND_LOG(ERR, "Cannot start port since there are no slave devices");
- return -1;
+ goto out_err;
}
if (internals->user_defined_mac == 0) {
new_mac_addr = &internals->slaves[i].persisted_mac_addr;
if (new_mac_addr == NULL)
- return -1;
+ goto out_err;
if (mac_address_set(eth_dev, new_mac_addr) != 0) {
RTE_BOND_LOG(ERR, "bonded port (%d) failed to update MAC address",
eth_dev->data->port_id);
- return -1;
+ goto out_err;
}
}
/* Update all slave devices MACs*/
if (mac_address_slaves_update(eth_dev) != 0)
- return -1;
+ goto out_err;
/* If bonded device is configure in promiscuous mode then re-apply config */
if (internals->promiscuous_en)
bond_ethdev_promiscuous_enable(eth_dev);
+ if (internals->mode == BONDING_MODE_8023AD) {
+ if (internals->mode4.dedicated_queues.enabled == 1) {
+ internals->mode4.dedicated_queues.rx_qid =
+ eth_dev->data->nb_rx_queues;
+ internals->mode4.dedicated_queues.tx_qid =
+ eth_dev->data->nb_tx_queues;
+ }
+ }
+
+
/* Reconfigure each slave device if starting bonded device */
for (i = 0; i < internals->slave_count; i++) {
- if (slave_configure(eth_dev,
- &(rte_eth_devices[internals->slaves[i].port_id])) != 0) {
+ struct rte_eth_dev *slave_ethdev =
+ &(rte_eth_devices[internals->slaves[i].port_id]);
+ if (slave_configure(eth_dev, slave_ethdev) != 0) {
RTE_BOND_LOG(ERR,
- "bonded port (%d) failed to reconfigure slave device (%d)",
- eth_dev->data->port_id, internals->slaves[i].port_id);
- return -1;
+ "bonded port (%d) failed to reconfigure slave device (%d)",
+ eth_dev->data->port_id,
+ internals->slaves[i].port_id);
+ goto out_err;
}
/* We will need to poll for link status if any slave doesn't
* support interrupts
if (internals->slaves[i].link_status_poll_enabled)
internals->link_status_polling_enabled = 1;
}
+
/* start polling if needed */
if (internals->link_status_polling_enabled) {
rte_eal_alarm_set(
bond_tlb_enable(internals);
return 0;
+
+out_err:
+ eth_dev->data->dev_started = 0;
+ return -1;
}
static void
bond_ethdev_close(struct rte_eth_dev *dev)
{
struct bond_dev_private *internals = dev->data->dev_private;
+ uint8_t bond_port_id = internals->port_id;
+ int skipped = 0;
+
+ RTE_LOG(INFO, EAL, "Closing bonded device %s\n", dev->device->name);
+ while (internals->slave_count != skipped) {
+ uint16_t port_id = internals->slaves[skipped].port_id;
+ rte_eth_dev_stop(port_id);
+
+ if (rte_eth_bond_slave_remove(bond_port_id, port_id) != 0) {
+ RTE_LOG(ERR, EAL,
+ "Failed to remove port %d from bonded device "
+ "%s\n", port_id, dev->device->name);
+ skipped++;
+ }
+ }
bond_ethdev_free_queues(dev);
rte_bitmap_reset(internals->vlan_filter_bmp);
}
{
struct bond_dev_private *internals = dev->data->dev_private;
+ uint16_t max_nb_rx_queues = UINT16_MAX;
+ uint16_t max_nb_tx_queues = UINT16_MAX;
+
dev_info->max_mac_addrs = 1;
dev_info->max_rx_pktlen = internals->candidate_max_rx_pktlen ?
- internals->candidate_max_rx_pktlen : 2048;
+ internals->candidate_max_rx_pktlen :
+ ETHER_MAX_JUMBO_FRAME_LEN;
+
+ /* Max number of tx/rx queues that the bonded device can support is the
+ * minimum values of the bonded slaves, as all slaves must be capable
+ * of supporting the same number of tx/rx queues.
+ */
+ if (internals->slave_count > 0) {
+ struct rte_eth_dev_info slave_info;
+ uint8_t idx;
+
+ for (idx = 0; idx < internals->slave_count; idx++) {
+ rte_eth_dev_info_get(internals->slaves[idx].port_id,
+ &slave_info);
- dev_info->max_rx_queues = (uint16_t)128;
- dev_info->max_tx_queues = (uint16_t)512;
+ if (slave_info.max_rx_queues < max_nb_rx_queues)
+ max_nb_rx_queues = slave_info.max_rx_queues;
+
+ if (slave_info.max_tx_queues < max_nb_tx_queues)
+ max_nb_tx_queues = slave_info.max_tx_queues;
+ }
+ }
+
+ dev_info->max_rx_queues = max_nb_rx_queues;
+ dev_info->max_tx_queues = max_nb_tx_queues;
+
+ /**
+ * 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.
+ */
+ if (internals->mode == BONDING_MODE_8023AD &&
+ internals->mode4.dedicated_queues.enabled == 1) {
+ dev_info->max_rx_queues--;
+ dev_info->max_tx_queues--;
+ }
dev_info->min_rx_bufsize = 0;
dev_info->rx_offload_capa = internals->rx_offload_capa;
dev_info->tx_offload_capa = internals->tx_offload_capa;
+ dev_info->rx_queue_offload_capa = internals->rx_queue_offload_capa;
+ dev_info->tx_queue_offload_capa = internals->tx_queue_offload_capa;
dev_info->flow_type_rss_offloads = internals->flow_type_rss_offloads;
dev_info->reta_size = internals->reta_size;
bond_ethdev_vlan_filter_set(struct rte_eth_dev *dev, uint16_t vlan_id, int on)
{
int res;
- uint8_t i;
+ uint16_t i;
struct bond_dev_private *internals = dev->data->dev_private;
/* don't do this while a slave is being added */
rte_bitmap_clear(internals->vlan_filter_bmp, vlan_id);
for (i = 0; i < internals->slave_count; i++) {
- uint8_t port_id = internals->slaves[i].port_id;
+ uint16_t port_id = internals->slaves[i].port_id;
res = rte_eth_dev_vlan_filter(port_id, vlan_id, on);
if (res == ENOTSUP)
bond_ethdev_lsc_event_callback(internals->slaves[i].port_id,
RTE_ETH_EVENT_INTR_LSC,
- &bonded_ethdev->data->port_id);
+ &bonded_ethdev->data->port_id,
+ NULL);
}
}
rte_spinlock_unlock(&internals->lock);
}
static int
-bond_ethdev_link_update(struct rte_eth_dev *bonded_eth_dev,
- int wait_to_complete)
+bond_ethdev_link_update(struct rte_eth_dev *ethdev, int wait_to_complete)
{
- struct bond_dev_private *internals = bonded_eth_dev->data->dev_private;
+ void (*link_update)(uint16_t port_id, struct rte_eth_link *eth_link);
+
+ struct bond_dev_private *bond_ctx;
+ struct rte_eth_link slave_link;
+
+ uint32_t idx;
- if (!bonded_eth_dev->data->dev_started ||
- internals->active_slave_count == 0) {
- bonded_eth_dev->data->dev_link.link_status = ETH_LINK_DOWN;
+ bond_ctx = ethdev->data->dev_private;
+
+ ethdev->data->dev_link.link_speed = ETH_SPEED_NUM_NONE;
+
+ if (ethdev->data->dev_started == 0 ||
+ bond_ctx->active_slave_count == 0) {
+ ethdev->data->dev_link.link_status = ETH_LINK_DOWN;
return 0;
- } else {
- struct rte_eth_dev *slave_eth_dev;
- int i, link_up = 0;
+ }
- for (i = 0; i < internals->active_slave_count; i++) {
- slave_eth_dev = &rte_eth_devices[internals->active_slaves[i]];
+ ethdev->data->dev_link.link_status = ETH_LINK_UP;
- (*slave_eth_dev->dev_ops->link_update)(slave_eth_dev,
- wait_to_complete);
- if (slave_eth_dev->data->dev_link.link_status == ETH_LINK_UP) {
- link_up = 1;
- break;
- }
+ if (wait_to_complete)
+ link_update = rte_eth_link_get;
+ else
+ link_update = rte_eth_link_get_nowait;
+
+ switch (bond_ctx->mode) {
+ case BONDING_MODE_BROADCAST:
+ /**
+ * Setting link speed to UINT32_MAX to ensure we pick up the
+ * value of the first active slave
+ */
+ ethdev->data->dev_link.link_speed = UINT32_MAX;
+
+ /**
+ * link speed is minimum value of all the slaves link speed as
+ * packet loss will occur on this slave if transmission at rates
+ * greater than this are attempted
+ */
+ for (idx = 1; idx < bond_ctx->active_slave_count; idx++) {
+ link_update(bond_ctx->active_slaves[0], &slave_link);
+
+ if (slave_link.link_speed <
+ ethdev->data->dev_link.link_speed)
+ ethdev->data->dev_link.link_speed =
+ slave_link.link_speed;
}
+ break;
+ case BONDING_MODE_ACTIVE_BACKUP:
+ /* Current primary slave */
+ link_update(bond_ctx->current_primary_port, &slave_link);
+
+ ethdev->data->dev_link.link_speed = slave_link.link_speed;
+ break;
+ case BONDING_MODE_8023AD:
+ ethdev->data->dev_link.link_autoneg =
+ bond_ctx->mode4.slave_link.link_autoneg;
+ ethdev->data->dev_link.link_duplex =
+ bond_ctx->mode4.slave_link.link_duplex;
+ /* fall through to update link speed */
+ case BONDING_MODE_ROUND_ROBIN:
+ case BONDING_MODE_BALANCE:
+ case BONDING_MODE_TLB:
+ case BONDING_MODE_ALB:
+ default:
+ /**
+ * In theses mode the maximum theoretical link speed is the sum
+ * of all the slaves
+ */
+ ethdev->data->dev_link.link_speed = ETH_SPEED_NUM_NONE;
- bonded_eth_dev->data->dev_link.link_status = link_up;
+ for (idx = 0; idx < bond_ctx->active_slave_count; idx++) {
+ link_update(bond_ctx->active_slaves[idx], &slave_link);
+
+ ethdev->data->dev_link.link_speed +=
+ slave_link.link_speed;
+ }
}
+
return 0;
}
-static void
+
+static int
bond_ethdev_stats_get(struct rte_eth_dev *dev, struct rte_eth_stats *stats)
{
struct bond_dev_private *internals = dev->data->dev_private;
}
}
+
+ return 0;
}
static void
RTE_ETH_EVENT_INTR_LSC, NULL);
}
-void
-bond_ethdev_lsc_event_callback(uint8_t port_id, enum rte_eth_event_type type,
- void *param)
+int
+bond_ethdev_lsc_event_callback(uint16_t port_id, enum rte_eth_event_type type,
+ void *param, void *ret_param __rte_unused)
{
- struct rte_eth_dev *bonded_eth_dev, *slave_eth_dev;
+ struct rte_eth_dev *bonded_eth_dev;
struct bond_dev_private *internals;
struct rte_eth_link link;
+ int rc = -1;
int i, valid_slave = 0;
uint8_t active_pos;
uint8_t lsc_flag = 0;
if (type != RTE_ETH_EVENT_INTR_LSC || param == NULL)
- return;
+ return rc;
bonded_eth_dev = &rte_eth_devices[*(uint8_t *)param];
- slave_eth_dev = &rte_eth_devices[port_id];
if (check_for_bonded_ethdev(bonded_eth_dev))
- return;
+ return rc;
internals = bonded_eth_dev->data->dev_private;
/* If the device isn't started don't handle interrupts */
if (!bonded_eth_dev->data->dev_started)
- return;
+ return rc;
/* verify that port_id is a valid slave of bonded port */
for (i = 0; i < internals->slave_count; i++) {
}
if (!valid_slave)
- return;
+ return rc;
/* Search for port in active port list */
active_pos = find_slave_by_id(internals->active_slaves,
rte_eth_link_get_nowait(port_id, &link);
if (link.link_status) {
if (active_pos < internals->active_slave_count)
- return;
+ return rc;
/* if no active slave ports then set this port to be primary port */
if (internals->active_slave_count < 1) {
lsc_flag = 1;
mac_address_slaves_update(bonded_eth_dev);
-
- /* Inherit eth dev link properties from first active slave */
- link_properties_set(bonded_eth_dev,
- &(slave_eth_dev->data->dev_link));
- } else {
- if (link_properties_valid(
- &bonded_eth_dev->data->dev_link, &link) != 0) {
- slave_eth_dev->data->dev_flags &=
- (~RTE_ETH_DEV_BONDED_SLAVE);
- RTE_LOG(ERR, PMD,
- "port %u invalid speed/duplex\n",
- port_id);
- return;
- }
}
activate_slave(bonded_eth_dev, port_id);
bond_ethdev_primary_set(internals, port_id);
} else {
if (active_pos == internals->active_slave_count)
- return;
+ return rc;
/* Remove from active slave list */
deactivate_slave(bonded_eth_dev, port_id);
- /* No active slaves, change link status to down and reset other
- * link properties */
- if (internals->active_slave_count < 1) {
+ if (internals->active_slave_count < 1)
lsc_flag = 1;
- bonded_eth_dev->data->dev_link.link_status = ETH_LINK_DOWN;
-
- link_properties_reset(bonded_eth_dev);
- }
/* Update primary id, take first active slave from list or if none
* available set to -1 */
}
}
+ /**
+ * Update bonded device link properties after any change to active
+ * slaves
+ */
+ bond_ethdev_link_update(bonded_eth_dev, 0);
+
if (lsc_flag) {
/* Cancel any possible outstanding interrupts if delays are enabled */
if (internals->link_up_delay_ms > 0 ||
(void *)bonded_eth_dev);
else
_rte_eth_dev_callback_process(bonded_eth_dev,
- RTE_ETH_EVENT_INTR_LSC, NULL);
+ RTE_ETH_EVENT_INTR_LSC,
+ NULL);
} else {
if (internals->link_down_delay_ms > 0)
(void *)bonded_eth_dev);
else
_rte_eth_dev_callback_process(bonded_eth_dev,
- RTE_ETH_EVENT_INTR_LSC, NULL);
+ RTE_ETH_EVENT_INTR_LSC,
+ NULL);
}
}
+ return 0;
}
static int
return 0;
}
+static int
+bond_ethdev_mtu_set(struct rte_eth_dev *dev, uint16_t mtu)
+{
+ struct rte_eth_dev *slave_eth_dev;
+ struct bond_dev_private *internals = dev->data->dev_private;
+ int ret, i;
+
+ rte_spinlock_lock(&internals->lock);
+
+ for (i = 0; i < internals->slave_count; i++) {
+ slave_eth_dev = &rte_eth_devices[internals->slaves[i].port_id];
+ if (*slave_eth_dev->dev_ops->mtu_set == NULL) {
+ rte_spinlock_unlock(&internals->lock);
+ return -ENOTSUP;
+ }
+ }
+ for (i = 0; i < internals->slave_count; i++) {
+ ret = rte_eth_dev_set_mtu(internals->slaves[i].port_id, mtu);
+ if (ret < 0) {
+ rte_spinlock_unlock(&internals->lock);
+ return ret;
+ }
+ }
+
+ rte_spinlock_unlock(&internals->lock);
+ return 0;
+}
+
+static int
+bond_ethdev_mac_address_set(struct rte_eth_dev *dev, struct ether_addr *addr)
+{
+ if (mac_address_set(dev, addr)) {
+ RTE_BOND_LOG(ERR, "Failed to update MAC address");
+ return -EINVAL;
+ }
+
+ return 0;
+}
+
const struct eth_dev_ops default_dev_ops = {
.dev_start = bond_ethdev_start,
.dev_stop = bond_ethdev_stop,
.reta_update = bond_ethdev_rss_reta_update,
.reta_query = bond_ethdev_rss_reta_query,
.rss_hash_update = bond_ethdev_rss_hash_update,
- .rss_hash_conf_get = bond_ethdev_rss_hash_conf_get
+ .rss_hash_conf_get = bond_ethdev_rss_hash_conf_get,
+ .mtu_set = bond_ethdev_mtu_set,
+ .mac_addr_set = bond_ethdev_mac_address_set
};
+static int
+bond_alloc(struct rte_vdev_device *dev, uint8_t mode)
+{
+ const char *name = rte_vdev_device_name(dev);
+ uint8_t socket_id = dev->device.numa_node;
+ 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
+ */
+
+ /* reserve an ethdev entry */
+ eth_dev = rte_eth_vdev_allocate(dev, sizeof(*internals));
+ if (eth_dev == NULL) {
+ RTE_BOND_LOG(ERR, "Unable to allocate rte_eth_dev");
+ goto err;
+ }
+
+ internals = eth_dev->data->dev_private;
+ 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_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->burst_xmit_hash = burst_xmit_l2_hash;
+ internals->user_defined_mac = 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->rx_queue_offload_capa = 0;
+ internals->tx_queue_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 to %d\n",
+ 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;
+}
+
static int
bond_probe(struct rte_vdev_device *dev)
{
const char *name;
struct bond_dev_private *internals;
struct rte_kvargs *kvlist;
- uint8_t bonding_mode, socket_id;
+ uint8_t bonding_mode, socket_id/*, agg_mode*/;
int arg_count, port_id;
+ uint8_t agg_mode;
+ struct rte_eth_dev *eth_dev;
+
+ if (!dev)
+ return -EINVAL;
name = rte_vdev_device_name(dev);
RTE_LOG(INFO, EAL, "Initializing pmd_bond for %s\n", name);
+ if (rte_eal_process_type() == RTE_PROC_SECONDARY &&
+ strlen(rte_vdev_device_args(dev)) == 0) {
+ eth_dev = rte_eth_dev_attach_secondary(name);
+ if (!eth_dev) {
+ RTE_LOG(ERR, PMD, "Failed to probe %s\n", name);
+ return -1;
+ }
+ /* TODO: request info from primary to set up Rx and Tx */
+ eth_dev->dev_ops = &default_dev_ops;
+ return 0;
+ }
+
kvlist = rte_kvargs_parse(rte_vdev_device_args(dev),
pmd_bond_init_valid_arguments);
if (kvlist == NULL)
socket_id = rte_socket_id();
}
+ dev->device.numa_node = socket_id;
+
/* Create link bonding eth device */
- port_id = rte_eth_bond_create(name, bonding_mode, socket_id);
+ port_id = bond_alloc(dev, bonding_mode);
if (port_id < 0) {
RTE_LOG(ERR, EAL, "Failed to create socket %s in mode %u on "
"socket %u.\n", name, bonding_mode, socket_id);
internals = rte_eth_devices[port_id].data->dev_private;
internals->kvlist = kvlist;
+
+ if (rte_kvargs_count(kvlist, PMD_BOND_AGG_MODE_KVARG) == 1) {
+ if (rte_kvargs_process(kvlist,
+ PMD_BOND_AGG_MODE_KVARG,
+ &bond_ethdev_parse_slave_agg_mode_kvarg,
+ &agg_mode) != 0) {
+ RTE_LOG(ERR, EAL,
+ "Failed to parse agg selection mode for bonded device %s\n",
+ name);
+ goto parse_error;
+ }
+
+ if (internals->mode == BONDING_MODE_8023AD)
+ rte_eth_bond_8023ad_agg_selection_set(port_id,
+ agg_mode);
+ } else {
+ rte_eth_bond_8023ad_agg_selection_set(port_id, AGG_STABLE);
+ }
+
RTE_LOG(INFO, EAL, "Create bonded device %s on port %d in mode %u on "
"socket %u.\n", name, port_id, bonding_mode, socket_id);
return 0;
static int
bond_remove(struct rte_vdev_device *dev)
{
+ struct rte_eth_dev *eth_dev;
+ struct bond_dev_private *internals;
const char *name;
- int ret;
if (!dev)
return -EINVAL;
name = rte_vdev_device_name(dev);
RTE_LOG(INFO, EAL, "Uninitializing pmd_bond for %s\n", name);
- /* free link bonding eth device */
- ret = rte_eth_bond_free(name);
- if (ret < 0)
- RTE_LOG(ERR, EAL, "Failed to free %s\n", name);
+ /* now free all data allocation - for eth_dev structure,
+ * dummy pci driver and internal (private) data
+ */
+
+ /* find an ethdev entry */
+ eth_dev = rte_eth_dev_allocated(name);
+ if (eth_dev == NULL)
+ return -ENODEV;
+
+ RTE_ASSERT(eth_dev->device == &dev->device);
+
+ internals = eth_dev->data->dev_private;
+ if (internals->slave_count != 0)
+ return -EBUSY;
- return ret;
+ if (eth_dev->data->dev_started == 1) {
+ bond_ethdev_stop(eth_dev);
+ bond_ethdev_close(eth_dev);
+ }
+
+ eth_dev->dev_ops = NULL;
+ eth_dev->rx_pkt_burst = NULL;
+ eth_dev->tx_pkt_burst = NULL;
+
+ internals = eth_dev->data->dev_private;
+ /* Try to release mempool used in mode6. If the bond
+ * device is not mode6, free the NULL is not problem.
+ */
+ 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);
+
+ return 0;
}
/* this part will resolve the slave portids after all the other pdev and vdev
static int
bond_ethdev_configure(struct rte_eth_dev *dev)
{
- char *name = dev->data->name;
+ const char *name = dev->device->name;
struct bond_dev_private *internals = dev->data->dev_private;
struct rte_kvargs *kvlist = internals->kvlist;
int arg_count;
- uint8_t port_id = dev - rte_eth_devices;
+ uint16_t port_id = dev - rte_eth_devices;
+ uint8_t agg_mode;
static const uint8_t default_rss_key[40] = {
0x6D, 0x5A, 0x56, 0xDA, 0x25, 0x5B, 0x0E, 0xC2, 0x41, 0x67, 0x25, 0x3D,
return -1;
}
+ if (rte_kvargs_count(kvlist, PMD_BOND_AGG_MODE_KVARG) == 1) {
+ if (rte_kvargs_process(kvlist,
+ PMD_BOND_AGG_MODE_KVARG,
+ &bond_ethdev_parse_slave_agg_mode_kvarg,
+ &agg_mode) != 0) {
+ RTE_LOG(ERR, EAL,
+ "Failed to parse agg selection mode for bonded device %s\n",
+ name);
+ }
+ if (internals->mode == BONDING_MODE_8023AD)
+ rte_eth_bond_8023ad_agg_selection_set(port_id,
+ agg_mode);
+ }
+
/* Parse/add slave ports to bonded device */
if (rte_kvargs_count(kvlist, PMD_BOND_SLAVE_PORT_KVARG) > 0) {
struct bond_ethdev_slave_ports slave_ports;
/* Parse/set primary slave port id*/
arg_count = rte_kvargs_count(kvlist, PMD_BOND_PRIMARY_SLAVE_KVARG);
if (arg_count == 1) {
- uint8_t primary_slave_port_id;
+ uint16_t primary_slave_port_id;
if (rte_kvargs_process(kvlist,
PMD_BOND_PRIMARY_SLAVE_KVARG,
}
/* Set balance mode transmit policy*/
- if (rte_eth_bond_primary_set(port_id, (uint8_t)primary_slave_port_id)
+ if (rte_eth_bond_primary_set(port_id, primary_slave_port_id)
!= 0) {
RTE_LOG(ERR, EAL,
"Failed to set primary slave port %d on bonded device %s\n",
"primary=<ifc> "
"mode=[0-6] "
"xmit_policy=[l2 | l23 | l34] "
+ "agg_mode=[count | stable | bandwidth] "
"socket_id=<int> "
"mac=<mac addr> "
"lsc_poll_period_ms=<int> "