static void print_ethaddr(const char *name, const struct ether_addr *eth_addr) { char buf[ETHER_ADDR_FMT_SIZE]; ether_format_addr(buf, ETHER_ADDR_FMT_SIZE, eth_addr); printf("%s%s", name, buf); }
int control_callback_setup(const char* cb, uint8_t nb_ports) { char cmd[CTRL_CBK_MAX_SIZE]; int len; char ether1[ETHER_ADDR_FMT_SIZE]; uint8_t port; const char* argv[4]; len = snprintf(cmd, CTRL_CBK_MAX_SIZE, "%s", cb); for (port = 0; port < nb_ports; port++) { ether_format_addr(ether1, ETHER_ADDR_FMT_SIZE, &ports_eth_addr[port]); len += snprintf(&cmd[len], CTRL_CBK_MAX_SIZE - len, " dpdk%d %s", port, ether1); if (len >= CTRL_CBK_MAX_SIZE) { rte_panic("control callback too long"); } } argv[0] = "/bin/sh"; argv[1] = "-c"; argv[2] = cmd; argv[3] = NULL; RTE_LOG(INFO, PKTJ_CTRL1, "executing command `%s`\n", cmd); return posix_spawn(NULL, "/bin/sh", NULL, NULL, __DECONST(char**, argv), environ); }
static void print_ethaddr(struct cmdline *cl, const char *name, const struct ether_addr *eth_addr) { char buf[ETHER_ADDR_FMT_SIZE]; ether_format_addr(buf, ETHER_ADDR_FMT_SIZE, eth_addr); cmdline_printf(cl, "%s%s\n", name, buf); }
static void ether_addr_dump(const char *what, const struct ether_addr *ea) { char buf[ETHER_ADDR_FMT_SIZE]; ether_format_addr(buf, ETHER_ADDR_FMT_SIZE, ea); if (what) printf("%s", what); printf("%s", buf); }
/* * @dev - real device kni attach to. * @kniname - optional, kni device name or auto generate. */ int kni_add_dev(struct netif_port *dev, const char *kniname) { struct rte_kni_conf conf; struct rte_kni *kni; int err; if (!dev) return EDPVS_INVAL; if (dev->type == PORT_TYPE_BOND_SLAVE) return EDPVS_NOTSUPP; if (kni_dev_exist(dev)) { RTE_LOG(ERR, Kni, "%s: dev %s has already attached with kni\n", __func__, dev->name); return EDPVS_EXIST; } kni_fill_conf(dev, kniname, &conf); kni = rte_kni_alloc(kni_mbuf_pool[dev->socket], &conf, NULL); if (!kni) return EDPVS_DPDKAPIFAIL; err = kni_rtnl_init(dev); if (err != EDPVS_OK) { rte_kni_release(kni); return err; } /* * kni device should use same mac as real device, * because it may config same IP of real device. * diff mac means kni cannot accept packets sent * to real-device. */ err = linux_set_if_mac(conf.name, (unsigned char *)&dev->addr); if (err != EDPVS_OK) { char mac[18]; ether_format_addr(mac, sizeof(mac), &dev->addr); RTE_LOG(WARNING, Kni, "%s: fail to set mac %s for %s: %s\n", __func__, mac, conf.name, strerror(errno)); } snprintf(dev->kni.name, sizeof(dev->kni.name), "%s", conf.name); dev->kni.addr = dev->addr; dev->kni.kni = kni; return EDPVS_OK; }
static void enicpmd_dev_info_get(struct rte_eth_dev *eth_dev, struct rte_eth_dev_info *device_info) { struct enic *enic = pmd_priv(eth_dev); ENICPMD_FUNC_TRACE(); /* Scattered Rx uses two receive queues per rx queue exposed to dpdk */ device_info->max_rx_queues = enic->conf_rq_count / 2; device_info->max_tx_queues = enic->conf_wq_count; device_info->min_rx_bufsize = ENIC_MIN_MTU; /* "Max" mtu is not a typo. HW receives packet sizes up to the * max mtu regardless of the current mtu (vNIC's mtu). vNIC mtu is * a hint to the driver to size receive buffers accordingly so that * larger-than-vnic-mtu packets get truncated.. For DPDK, we let * the user decide the buffer size via rxmode.max_rx_pkt_len, basically * ignoring vNIC mtu. */ device_info->max_rx_pktlen = enic_mtu_to_max_rx_pktlen(enic->max_mtu); device_info->max_mac_addrs = ENIC_UNICAST_PERFECT_FILTERS; device_info->rx_offload_capa = enic->rx_offload_capa; device_info->tx_offload_capa = enic->tx_offload_capa; device_info->tx_queue_offload_capa = enic->tx_queue_offload_capa; device_info->default_rxconf = (struct rte_eth_rxconf) { .rx_free_thresh = ENIC_DEFAULT_RX_FREE_THRESH }; device_info->reta_size = enic->reta_size; device_info->hash_key_size = enic->hash_key_size; device_info->flow_type_rss_offloads = enic->flow_type_rss_offloads; device_info->rx_desc_lim = (struct rte_eth_desc_lim) { .nb_max = enic->config.rq_desc_count, .nb_min = ENIC_MIN_RQ_DESCS, .nb_align = ENIC_ALIGN_DESCS, }; device_info->tx_desc_lim = (struct rte_eth_desc_lim) { .nb_max = enic->config.wq_desc_count, .nb_min = ENIC_MIN_WQ_DESCS, .nb_align = ENIC_ALIGN_DESCS, .nb_seg_max = ENIC_TX_XMIT_MAX, .nb_mtu_seg_max = ENIC_NON_TSO_MAX_DESC, }; device_info->default_rxportconf = (struct rte_eth_dev_portconf) { .burst_size = ENIC_DEFAULT_RX_BURST, .ring_size = RTE_MIN(device_info->rx_desc_lim.nb_max, ENIC_DEFAULT_RX_RING_SIZE), .nb_queues = ENIC_DEFAULT_RX_RINGS, }; device_info->default_txportconf = (struct rte_eth_dev_portconf) { .burst_size = ENIC_DEFAULT_TX_BURST, .ring_size = RTE_MIN(device_info->tx_desc_lim.nb_max, ENIC_DEFAULT_TX_RING_SIZE), .nb_queues = ENIC_DEFAULT_TX_RINGS, }; } static const uint32_t *enicpmd_dev_supported_ptypes_get(struct rte_eth_dev *dev) { static const uint32_t ptypes[] = { RTE_PTYPE_L2_ETHER, RTE_PTYPE_L2_ETHER_VLAN, RTE_PTYPE_L3_IPV4_EXT_UNKNOWN, RTE_PTYPE_L3_IPV6_EXT_UNKNOWN, RTE_PTYPE_L4_TCP, RTE_PTYPE_L4_UDP, RTE_PTYPE_L4_FRAG, RTE_PTYPE_L4_NONFRAG, RTE_PTYPE_UNKNOWN }; static const uint32_t ptypes_overlay[] = { RTE_PTYPE_L2_ETHER, RTE_PTYPE_L2_ETHER_VLAN, RTE_PTYPE_L3_IPV4_EXT_UNKNOWN, RTE_PTYPE_L3_IPV6_EXT_UNKNOWN, RTE_PTYPE_L4_TCP, RTE_PTYPE_L4_UDP, RTE_PTYPE_L4_FRAG, RTE_PTYPE_L4_NONFRAG, RTE_PTYPE_TUNNEL_GRENAT, RTE_PTYPE_INNER_L2_ETHER, RTE_PTYPE_INNER_L3_IPV4_EXT_UNKNOWN, RTE_PTYPE_INNER_L3_IPV6_EXT_UNKNOWN, RTE_PTYPE_INNER_L4_TCP, RTE_PTYPE_INNER_L4_UDP, RTE_PTYPE_INNER_L4_FRAG, RTE_PTYPE_INNER_L4_NONFRAG, RTE_PTYPE_UNKNOWN }; if (dev->rx_pkt_burst != enic_dummy_recv_pkts && dev->rx_pkt_burst != NULL) { struct enic *enic = pmd_priv(dev); if (enic->overlay_offload) return ptypes_overlay; else return ptypes; } return NULL; } static void enicpmd_dev_promiscuous_enable(struct rte_eth_dev *eth_dev) { struct enic *enic = pmd_priv(eth_dev); if (rte_eal_process_type() != RTE_PROC_PRIMARY) return; ENICPMD_FUNC_TRACE(); enic->promisc = 1; enic_add_packet_filter(enic); } static void enicpmd_dev_promiscuous_disable(struct rte_eth_dev *eth_dev) { struct enic *enic = pmd_priv(eth_dev); if (rte_eal_process_type() != RTE_PROC_PRIMARY) return; ENICPMD_FUNC_TRACE(); enic->promisc = 0; enic_add_packet_filter(enic); } static void enicpmd_dev_allmulticast_enable(struct rte_eth_dev *eth_dev) { struct enic *enic = pmd_priv(eth_dev); if (rte_eal_process_type() != RTE_PROC_PRIMARY) return; ENICPMD_FUNC_TRACE(); enic->allmulti = 1; enic_add_packet_filter(enic); } static void enicpmd_dev_allmulticast_disable(struct rte_eth_dev *eth_dev) { struct enic *enic = pmd_priv(eth_dev); if (rte_eal_process_type() != RTE_PROC_PRIMARY) return; ENICPMD_FUNC_TRACE(); enic->allmulti = 0; enic_add_packet_filter(enic); } static int enicpmd_add_mac_addr(struct rte_eth_dev *eth_dev, struct ether_addr *mac_addr, __rte_unused uint32_t index, __rte_unused uint32_t pool) { struct enic *enic = pmd_priv(eth_dev); if (rte_eal_process_type() != RTE_PROC_PRIMARY) return -E_RTE_SECONDARY; ENICPMD_FUNC_TRACE(); return enic_set_mac_address(enic, mac_addr->addr_bytes); } static void enicpmd_remove_mac_addr(struct rte_eth_dev *eth_dev, uint32_t index) { struct enic *enic = pmd_priv(eth_dev); if (rte_eal_process_type() != RTE_PROC_PRIMARY) return; ENICPMD_FUNC_TRACE(); if (enic_del_mac_address(enic, index)) dev_err(enic, "del mac addr failed\n"); } static int enicpmd_set_mac_addr(struct rte_eth_dev *eth_dev, struct ether_addr *addr) { struct enic *enic = pmd_priv(eth_dev); int ret; if (rte_eal_process_type() != RTE_PROC_PRIMARY) return -E_RTE_SECONDARY; ENICPMD_FUNC_TRACE(); ret = enic_del_mac_address(enic, 0); if (ret) return ret; return enic_set_mac_address(enic, addr->addr_bytes); } static void debug_log_add_del_addr(struct ether_addr *addr, bool add) { char mac_str[ETHER_ADDR_FMT_SIZE]; ether_format_addr(mac_str, ETHER_ADDR_FMT_SIZE, addr); PMD_INIT_LOG(DEBUG, " %s address %s\n", add ? "add" : "remove", mac_str); } static int enicpmd_set_mc_addr_list(struct rte_eth_dev *eth_dev, struct ether_addr *mc_addr_set, uint32_t nb_mc_addr) { struct enic *enic = pmd_priv(eth_dev); char mac_str[ETHER_ADDR_FMT_SIZE]; struct ether_addr *addr; uint32_t i, j; int ret; ENICPMD_FUNC_TRACE(); /* Validate the given addresses first */ for (i = 0; i < nb_mc_addr && mc_addr_set != NULL; i++) { addr = &mc_addr_set[i]; if (!is_multicast_ether_addr(addr) || is_broadcast_ether_addr(addr)) { ether_format_addr(mac_str, ETHER_ADDR_FMT_SIZE, addr); PMD_INIT_LOG(ERR, " invalid multicast address %s\n", mac_str); return -EINVAL; } } /* Flush all if requested */ if (nb_mc_addr == 0 || mc_addr_set == NULL) { PMD_INIT_LOG(DEBUG, " flush multicast addresses\n"); for (i = 0; i < enic->mc_count; i++) { addr = &enic->mc_addrs[i]; debug_log_add_del_addr(addr, false); ret = vnic_dev_del_addr(enic->vdev, addr->addr_bytes); if (ret) return ret; } enic->mc_count = 0; return 0; } if (nb_mc_addr > ENIC_MULTICAST_PERFECT_FILTERS) { PMD_INIT_LOG(ERR, " too many multicast addresses: max=%d\n", ENIC_MULTICAST_PERFECT_FILTERS); return -ENOSPC; } /* * devcmd is slow, so apply the difference instead of flushing and * adding everything. * 1. Delete addresses on the NIC but not on the host */ for (i = 0; i < enic->mc_count; i++) { addr = &enic->mc_addrs[i]; for (j = 0; j < nb_mc_addr; j++) { if (is_same_ether_addr(addr, &mc_addr_set[j])) break; } if (j < nb_mc_addr) continue; debug_log_add_del_addr(addr, false); ret = vnic_dev_del_addr(enic->vdev, addr->addr_bytes); if (ret) return ret; } /* 2. Add addresses on the host but not on the NIC */ for (i = 0; i < nb_mc_addr; i++) { addr = &mc_addr_set[i]; for (j = 0; j < enic->mc_count; j++) { if (is_same_ether_addr(addr, &enic->mc_addrs[j])) break; } if (j < enic->mc_count) continue; debug_log_add_del_addr(addr, true); ret = vnic_dev_add_addr(enic->vdev, addr->addr_bytes); if (ret) return ret; } /* Keep a copy so we can flush/apply later on.. */ memcpy(enic->mc_addrs, mc_addr_set, nb_mc_addr * sizeof(struct ether_addr)); enic->mc_count = nb_mc_addr; return 0; } static int enicpmd_mtu_set(struct rte_eth_dev *eth_dev, uint16_t mtu) { struct enic *enic = pmd_priv(eth_dev); ENICPMD_FUNC_TRACE(); return enic_set_mtu(enic, mtu); } static int enicpmd_dev_rss_reta_query(struct rte_eth_dev *dev, struct rte_eth_rss_reta_entry64 *reta_conf, uint16_t reta_size) { struct enic *enic = pmd_priv(dev); uint16_t i, idx, shift; ENICPMD_FUNC_TRACE(); if (reta_size != ENIC_RSS_RETA_SIZE) { dev_err(enic, "reta_query: wrong reta_size. given=%u expected=%u\n", reta_size, ENIC_RSS_RETA_SIZE); return -EINVAL; } for (i = 0; i < reta_size; i++) { idx = i / RTE_RETA_GROUP_SIZE; shift = i % RTE_RETA_GROUP_SIZE; if (reta_conf[idx].mask & (1ULL << shift)) reta_conf[idx].reta[shift] = enic_sop_rq_idx_to_rte_idx( enic->rss_cpu.cpu[i / 4].b[i % 4]); } return 0; } static int enicpmd_dev_rss_reta_update(struct rte_eth_dev *dev, struct rte_eth_rss_reta_entry64 *reta_conf, uint16_t reta_size) { struct enic *enic = pmd_priv(dev); union vnic_rss_cpu rss_cpu; uint16_t i, idx, shift; ENICPMD_FUNC_TRACE(); if (reta_size != ENIC_RSS_RETA_SIZE) { dev_err(enic, "reta_update: wrong reta_size. given=%u" " expected=%u\n", reta_size, ENIC_RSS_RETA_SIZE); return -EINVAL; } /* * Start with the current reta and modify it per reta_conf, as we * need to push the entire reta even if we only modify one entry. */ rss_cpu = enic->rss_cpu; for (i = 0; i < reta_size; i++) { idx = i / RTE_RETA_GROUP_SIZE; shift = i % RTE_RETA_GROUP_SIZE; if (reta_conf[idx].mask & (1ULL << shift)) rss_cpu.cpu[i / 4].b[i % 4] = enic_rte_rq_idx_to_sop_idx( reta_conf[idx].reta[shift]); } return enic_set_rss_reta(enic, &rss_cpu); } static int enicpmd_dev_rss_hash_update(struct rte_eth_dev *dev, struct rte_eth_rss_conf *rss_conf) { struct enic *enic = pmd_priv(dev); ENICPMD_FUNC_TRACE(); return enic_set_rss_conf(enic, rss_conf); } static int enicpmd_dev_rss_hash_conf_get(struct rte_eth_dev *dev, struct rte_eth_rss_conf *rss_conf) { struct enic *enic = pmd_priv(dev); ENICPMD_FUNC_TRACE(); if (rss_conf == NULL) return -EINVAL; if (rss_conf->rss_key != NULL && rss_conf->rss_key_len < ENIC_RSS_HASH_KEY_SIZE) { dev_err(enic, "rss_hash_conf_get: wrong rss_key_len. given=%u" " expected=%u+\n", rss_conf->rss_key_len, ENIC_RSS_HASH_KEY_SIZE); return -EINVAL; } rss_conf->rss_hf = enic->rss_hf; if (rss_conf->rss_key != NULL) { int i; for (i = 0; i < ENIC_RSS_HASH_KEY_SIZE; i++) { rss_conf->rss_key[i] = enic->rss_key.key[i / 10].b[i % 10]; } rss_conf->rss_key_len = ENIC_RSS_HASH_KEY_SIZE; } return 0; } static void enicpmd_dev_rxq_info_get(struct rte_eth_dev *dev, uint16_t rx_queue_id, struct rte_eth_rxq_info *qinfo) { struct enic *enic = pmd_priv(dev); struct vnic_rq *rq_sop; struct vnic_rq *rq_data; struct rte_eth_rxconf *conf; uint16_t sop_queue_idx; uint16_t data_queue_idx; ENICPMD_FUNC_TRACE(); sop_queue_idx = enic_rte_rq_idx_to_sop_idx(rx_queue_id); data_queue_idx = enic_rte_rq_idx_to_data_idx(rx_queue_id); rq_sop = &enic->rq[sop_queue_idx]; rq_data = &enic->rq[data_queue_idx]; /* valid if data_queue_enable */ qinfo->mp = rq_sop->mp; qinfo->scattered_rx = rq_sop->data_queue_enable; qinfo->nb_desc = rq_sop->ring.desc_count; if (qinfo->scattered_rx) qinfo->nb_desc += rq_data->ring.desc_count; conf = &qinfo->conf; memset(conf, 0, sizeof(*conf)); conf->rx_free_thresh = rq_sop->rx_free_thresh; conf->rx_drop_en = 1; /* * Except VLAN stripping (port setting), all the checksum offloads * are always enabled. */ conf->offloads = enic->rx_offload_capa; if (!enic->ig_vlan_strip_en) conf->offloads &= ~DEV_RX_OFFLOAD_VLAN_STRIP; /* rx_thresh and other fields are not applicable for enic */ } static void enicpmd_dev_txq_info_get(struct rte_eth_dev *dev, uint16_t tx_queue_id, struct rte_eth_txq_info *qinfo) { struct enic *enic = pmd_priv(dev); struct vnic_wq *wq = &enic->wq[tx_queue_id]; ENICPMD_FUNC_TRACE(); qinfo->nb_desc = wq->ring.desc_count; memset(&qinfo->conf, 0, sizeof(qinfo->conf)); qinfo->conf.offloads = wq->offloads; /* tx_thresh, and all the other fields are not applicable for enic */ } static int enicpmd_dev_rx_queue_intr_enable(struct rte_eth_dev *eth_dev, uint16_t rx_queue_id) { struct enic *enic = pmd_priv(eth_dev); ENICPMD_FUNC_TRACE(); vnic_intr_unmask(&enic->intr[rx_queue_id + ENICPMD_RXQ_INTR_OFFSET]); return 0; } static int enicpmd_dev_rx_queue_intr_disable(struct rte_eth_dev *eth_dev, uint16_t rx_queue_id) { struct enic *enic = pmd_priv(eth_dev); ENICPMD_FUNC_TRACE(); vnic_intr_mask(&enic->intr[rx_queue_id + ENICPMD_RXQ_INTR_OFFSET]); return 0; } static int udp_tunnel_common_check(struct enic *enic, struct rte_eth_udp_tunnel *tnl) { if (tnl->prot_type != RTE_TUNNEL_TYPE_VXLAN) return -ENOTSUP; if (!enic->overlay_offload) { PMD_INIT_LOG(DEBUG, " vxlan (overlay offload) is not " "supported\n"); return -ENOTSUP; } return 0; } static int update_vxlan_port(struct enic *enic, uint16_t port) { if (vnic_dev_overlay_offload_cfg(enic->vdev, OVERLAY_CFG_VXLAN_PORT_UPDATE, port)) { PMD_INIT_LOG(DEBUG, " failed to update vxlan port\n"); return -EINVAL; } PMD_INIT_LOG(DEBUG, " updated vxlan port to %u\n", port); enic->vxlan_port = port; return 0; } static int enicpmd_dev_udp_tunnel_port_add(struct rte_eth_dev *eth_dev, struct rte_eth_udp_tunnel *tnl) { struct enic *enic = pmd_priv(eth_dev); int ret; ENICPMD_FUNC_TRACE(); ret = udp_tunnel_common_check(enic, tnl); if (ret) return ret; /* * The NIC has 1 configurable VXLAN port number. "Adding" a new port * number replaces it. */ if (tnl->udp_port == enic->vxlan_port || tnl->udp_port == 0) { PMD_INIT_LOG(DEBUG, " %u is already configured or invalid\n", tnl->udp_port); return -EINVAL; } return update_vxlan_port(enic, tnl->udp_port); } static int enicpmd_dev_udp_tunnel_port_del(struct rte_eth_dev *eth_dev, struct rte_eth_udp_tunnel *tnl) { struct enic *enic = pmd_priv(eth_dev); int ret; ENICPMD_FUNC_TRACE(); ret = udp_tunnel_common_check(enic, tnl); if (ret) return ret; /* * Clear the previously set port number and restore the * hardware default port number. Some drivers disable VXLAN * offloads when there are no configured port numbers. But * enic does not do that as VXLAN is part of overlay offload, * which is tied to inner RSS and TSO. */ if (tnl->udp_port != enic->vxlan_port) { PMD_INIT_LOG(DEBUG, " %u is not a configured vxlan port\n", tnl->udp_port); return -EINVAL; } return update_vxlan_port(enic, ENIC_DEFAULT_VXLAN_PORT); } static int enicpmd_dev_fw_version_get(struct rte_eth_dev *eth_dev, char *fw_version, size_t fw_size) { struct vnic_devcmd_fw_info *info; struct enic *enic; int ret; ENICPMD_FUNC_TRACE(); if (fw_version == NULL || fw_size <= 0) return -EINVAL; enic = pmd_priv(eth_dev); ret = vnic_dev_fw_info(enic->vdev, &info); if (ret) return ret; snprintf(fw_version, fw_size, "%s %s", info->fw_version, info->fw_build); fw_version[fw_size - 1] = '\0'; return 0; } static const struct eth_dev_ops enicpmd_eth_dev_ops = { .dev_configure = enicpmd_dev_configure, .dev_start = enicpmd_dev_start, .dev_stop = enicpmd_dev_stop, .dev_set_link_up = NULL, .dev_set_link_down = NULL, .dev_close = enicpmd_dev_close, .promiscuous_enable = enicpmd_dev_promiscuous_enable, .promiscuous_disable = enicpmd_dev_promiscuous_disable, .allmulticast_enable = enicpmd_dev_allmulticast_enable, .allmulticast_disable = enicpmd_dev_allmulticast_disable, .link_update = enicpmd_dev_link_update, .stats_get = enicpmd_dev_stats_get, .stats_reset = enicpmd_dev_stats_reset, .queue_stats_mapping_set = NULL, .dev_infos_get = enicpmd_dev_info_get, .dev_supported_ptypes_get = enicpmd_dev_supported_ptypes_get, .mtu_set = enicpmd_mtu_set, .vlan_filter_set = NULL, .vlan_tpid_set = NULL, .vlan_offload_set = enicpmd_vlan_offload_set, .vlan_strip_queue_set = NULL, .rx_queue_start = enicpmd_dev_rx_queue_start, .rx_queue_stop = enicpmd_dev_rx_queue_stop, .tx_queue_start = enicpmd_dev_tx_queue_start, .tx_queue_stop = enicpmd_dev_tx_queue_stop, .rx_queue_setup = enicpmd_dev_rx_queue_setup, .rx_queue_release = enicpmd_dev_rx_queue_release, .rx_queue_count = enicpmd_dev_rx_queue_count, .rx_descriptor_done = NULL, .tx_queue_setup = enicpmd_dev_tx_queue_setup, .tx_queue_release = enicpmd_dev_tx_queue_release, .rx_queue_intr_enable = enicpmd_dev_rx_queue_intr_enable, .rx_queue_intr_disable = enicpmd_dev_rx_queue_intr_disable, .rxq_info_get = enicpmd_dev_rxq_info_get, .txq_info_get = enicpmd_dev_txq_info_get, .dev_led_on = NULL, .dev_led_off = NULL, .flow_ctrl_get = NULL, .flow_ctrl_set = NULL, .priority_flow_ctrl_set = NULL, .mac_addr_add = enicpmd_add_mac_addr, .mac_addr_remove = enicpmd_remove_mac_addr, .mac_addr_set = enicpmd_set_mac_addr, .set_mc_addr_list = enicpmd_set_mc_addr_list, .filter_ctrl = enicpmd_dev_filter_ctrl, .reta_query = enicpmd_dev_rss_reta_query, .reta_update = enicpmd_dev_rss_reta_update, .rss_hash_conf_get = enicpmd_dev_rss_hash_conf_get, .rss_hash_update = enicpmd_dev_rss_hash_update, .udp_tunnel_port_add = enicpmd_dev_udp_tunnel_port_add, .udp_tunnel_port_del = enicpmd_dev_udp_tunnel_port_del, .fw_version_get = enicpmd_dev_fw_version_get, }; static int enic_parse_zero_one(const char *key, const char *value, void *opaque) { struct enic *enic; bool b; enic = (struct enic *)opaque; if (strcmp(value, "0") == 0) { b = false; } else if (strcmp(value, "1") == 0) { b = true; } else { dev_err(enic, "Invalid value for %s" ": expected=0|1 given=%s\n", key, value); return -EINVAL; } if (strcmp(key, ENIC_DEVARG_DISABLE_OVERLAY) == 0) enic->disable_overlay = b; if (strcmp(key, ENIC_DEVARG_ENABLE_AVX2_RX) == 0) enic->enable_avx2_rx = b; return 0; } static int enic_parse_ig_vlan_rewrite(__rte_unused const char *key, const char *value, void *opaque) { struct enic *enic; enic = (struct enic *)opaque; if (strcmp(value, "trunk") == 0) { /* Trunk mode: always tag */ enic->ig_vlan_rewrite_mode = IG_VLAN_REWRITE_MODE_DEFAULT_TRUNK; } else if (strcmp(value, "untag") == 0) { /* Untag default VLAN mode: untag if VLAN = default VLAN */ enic->ig_vlan_rewrite_mode = IG_VLAN_REWRITE_MODE_UNTAG_DEFAULT_VLAN; } else if (strcmp(value, "priority") == 0) { /* * Priority-tag default VLAN mode: priority tag (VLAN header * with ID=0) if VLAN = default */ enic->ig_vlan_rewrite_mode = IG_VLAN_REWRITE_MODE_PRIORITY_TAG_DEFAULT_VLAN; } else if (strcmp(value, "pass") == 0) { /* Pass through mode: do not touch tags */ enic->ig_vlan_rewrite_mode = IG_VLAN_REWRITE_MODE_PASS_THRU; } else { dev_err(enic, "Invalid value for " ENIC_DEVARG_IG_VLAN_REWRITE ": expected=trunk|untag|priority|pass given=%s\n", value); return -EINVAL; } return 0; } static int enic_check_devargs(struct rte_eth_dev *dev) { static const char *const valid_keys[] = { ENIC_DEVARG_DISABLE_OVERLAY, ENIC_DEVARG_ENABLE_AVX2_RX, ENIC_DEVARG_IG_VLAN_REWRITE, NULL}; struct enic *enic = pmd_priv(dev); struct rte_kvargs *kvlist; ENICPMD_FUNC_TRACE(); enic->disable_overlay = false; enic->enable_avx2_rx = false; enic->ig_vlan_rewrite_mode = IG_VLAN_REWRITE_MODE_PASS_THRU; if (!dev->device->devargs) return 0; kvlist = rte_kvargs_parse(dev->device->devargs->args, valid_keys); if (!kvlist) return -EINVAL; if (rte_kvargs_process(kvlist, ENIC_DEVARG_DISABLE_OVERLAY, enic_parse_zero_one, enic) < 0 || rte_kvargs_process(kvlist, ENIC_DEVARG_ENABLE_AVX2_RX, enic_parse_zero_one, enic) < 0 || rte_kvargs_process(kvlist, ENIC_DEVARG_IG_VLAN_REWRITE, enic_parse_ig_vlan_rewrite, enic) < 0) { rte_kvargs_free(kvlist); return -EINVAL; } rte_kvargs_free(kvlist); return 0; } /* Initialize the driver * It returns 0 on success. */ static int eth_enicpmd_dev_init(struct rte_eth_dev *eth_dev) { struct rte_pci_device *pdev; struct rte_pci_addr *addr; struct enic *enic = pmd_priv(eth_dev); int err; ENICPMD_FUNC_TRACE(); enic->port_id = eth_dev->data->port_id; enic->rte_dev = eth_dev; eth_dev->dev_ops = &enicpmd_eth_dev_ops; eth_dev->rx_pkt_burst = &enic_recv_pkts; eth_dev->tx_pkt_burst = &enic_xmit_pkts; eth_dev->tx_pkt_prepare = &enic_prep_pkts; /* Let rte_eth_dev_close() release the port resources */ eth_dev->data->dev_flags |= RTE_ETH_DEV_CLOSE_REMOVE; pdev = RTE_ETH_DEV_TO_PCI(eth_dev); rte_eth_copy_pci_info(eth_dev, pdev); enic->pdev = pdev; addr = &pdev->addr; snprintf(enic->bdf_name, ENICPMD_BDF_LENGTH, "%04x:%02x:%02x.%x", addr->domain, addr->bus, addr->devid, addr->function); err = enic_check_devargs(eth_dev); if (err) return err; return enic_probe(enic); } static int eth_enic_pci_probe(struct rte_pci_driver *pci_drv __rte_unused, struct rte_pci_device *pci_dev) { return rte_eth_dev_pci_generic_probe(pci_dev, sizeof(struct enic), eth_enicpmd_dev_init); } static int eth_enic_pci_remove(struct rte_pci_device *pci_dev) { return rte_eth_dev_pci_generic_remove(pci_dev, NULL); } static struct rte_pci_driver rte_enic_pmd = { .id_table = pci_id_enic_map, .drv_flags = RTE_PCI_DRV_NEED_MAPPING | RTE_PCI_DRV_INTR_LSC | RTE_PCI_DRV_IOVA_AS_VA, .probe = eth_enic_pci_probe, .remove = eth_enic_pci_remove, }; RTE_PMD_REGISTER_PCI(net_enic, rte_enic_pmd); RTE_PMD_REGISTER_PCI_TABLE(net_enic, pci_id_enic_map); RTE_PMD_REGISTER_KMOD_DEP(net_enic, "* igb_uio | uio_pci_generic | vfio-pci"); RTE_PMD_REGISTER_PARAM_STRING(net_enic, ENIC_DEVARG_DISABLE_OVERLAY "=0|1 " ENIC_DEVARG_ENABLE_AVX2_RX "=0|1 " ENIC_DEVARG_IG_VLAN_REWRITE "=trunk|untag|priority|pass");
int main(int argc, char **argv) { struct pg_error *error = NULL; int ret; uint64_t args_flags; struct vtep_opts opt = {NULL, NULL, NULL, NULL}; int32_t ip; struct ether_addr eth_addr; struct ether_addr inner_addr; GList *neighbor_addrs = NULL; ret = pg_start(argc, argv, &error); g_assert(ret != -1); CHECK_ERROR(error); if (signal(SIGINT, sig_handler) == SIG_ERR) return -errno; /* accounting program name */ argc -= ret; argv += ret; args_flags = parse_args(argc, argv, &opt); if (args_flags & PRINT_USAGE) print_usage(); if (!!(args_flags & FAIL)) { dprintf(2, "Invalide arguments, use '-h'\n"); ret = -EINVAL; goto exit; } if (!pg_scan_ether_addr(ð_addr, opt.mac) || !is_valid_assigned_ether_addr(ð_addr)) { char buf[40]; ether_format_addr(buf, 40, ð_addr); dprintf(2, "%s is an invalide ethernet adress\n" "sould be an unicast addr and have format XX:XX:XX:XX:XX:XX\n", buf); ret = -EINVAL; goto exit; } if (!pg_scan_ether_addr(&inner_addr, opt.inner_mac) || !is_valid_assigned_ether_addr(&inner_addr)) { char buf[40]; ether_format_addr(buf, 40, &inner_addr); dprintf(2, "%s is an invalide ethernet adress\n" "sould be an unicast addr and have format XX:XX:XX:XX:XX:XX\n", buf); ret = -EINVAL; goto exit; } for (GList *lst = opt.neighbor_macs; lst != NULL; lst = lst->next) { const char *data = lst->data; struct ether_addr *tmp = g_new0(struct ether_addr, 1); if (!pg_scan_ether_addr(tmp, data) || !is_valid_assigned_ether_addr(tmp)) { char buf[40]; ether_format_addr(buf, 40, tmp); dprintf(2, "%s is an invalide ethernet adress\n" "sould be an unicast addr and have format XX:XX:XX:XX:XX:XX\n", buf); ret = -EINVAL; goto exit; } neighbor_addrs = g_list_append(neighbor_addrs, tmp); } ip = inet_addr(opt.ip); if (ip < 0) { dprintf(2, "invalide ip\n" "should have format: XXX.XXX.XXX.XXX\n"); return -EINVAL; } ret = start_loop(ip, ð_addr, &inner_addr, neighbor_addrs); exit: g_list_free(opt.neighbor_macs); g_list_free_full(neighbor_addrs, destroy_ether_addr); pg_stop(); return ret; }
static void log_packet(struct rte_mbuf *m) { char buf[4096]; int offset = 0; int n; uint16_t ether_type; uint8_t ipproto; char *l3_h = NULL; struct ether_hdr *eth_h; struct ipv4_hdr *ipv4_h = NULL; struct ipv6_hdr *ipv6_h = NULL; struct udp_hdr *udp_h = NULL; struct tcp_hdr *tcp_h = NULL; char ip_src_str[INET6_ADDRSTRLEN]; char ip_dst_str[INET6_ADDRSTRLEN]; eth_h = rte_pktmbuf_mtod(m, struct ether_hdr *); ether_format_addr(buf+offset, ETHER_ADDR_FMT_SIZE, ð_h->s_addr); offset += ETHER_ADDR_FMT_SIZE-1; strcpy(buf+offset, " -> "); offset += strlen(" -> "); ether_format_addr(buf+offset, ETHER_ADDR_FMT_SIZE, ð_h->d_addr); offset += ETHER_ADDR_FMT_SIZE-1; strcpy(buf+offset, "\n"); offset += strlen("\n"); ether_type = rte_be_to_cpu_16(eth_h->ether_type); l3_h = (char *)(eth_h + 1); switch (ether_type) { case ETHER_TYPE_ARP: return; case ETHER_TYPE_IPv4: ipv4_h = (struct ipv4_hdr *)l3_h; ipproto = ipv4_h->next_proto_id; inet_ntop(AF_INET, &(ipv4_h->src_addr), ip_src_str, INET6_ADDRSTRLEN); inet_ntop(AF_INET, &(ipv4_h->dst_addr), ip_dst_str, INET6_ADDRSTRLEN); n = snprintf(buf+offset, 4096-offset, " IPV4 %s -> %s (ttl %d, id %d, tlen: %d, offset %d, flags(%s%s))\n", ip_src_str, ip_dst_str, ipv4_h->time_to_live, rte_be_to_cpu_16(ipv4_h->packet_id), rte_be_to_cpu_16(ipv4_h->total_length), (rte_be_to_cpu_16(ipv4_h->fragment_offset) & \ IPV4_HDR_OFFSET_MASK) * IPV4_HDR_OFFSET_UNITS, (rte_be_to_cpu_16(ipv4_h->fragment_offset) & IPV4_HDR_DF_FLAG)? "DF":"", (rte_be_to_cpu_16(ipv4_h->fragment_offset) & IPV4_HDR_MF_FLAG)? "MF":""); offset += n; break; case ETHER_TYPE_IPv6: ipv6_h = (struct ipv6_hdr *)l3_h; ipproto = ipv6_h->proto; inet_ntop(AF_INET6, &(ipv6_h->src_addr), ip_src_str, INET6_ADDRSTRLEN); inet_ntop(AF_INET6, &(ipv6_h->dst_addr), ip_dst_str, INET6_ADDRSTRLEN); n = snprintf(buf+offset, 4096-offset, " IPV6 %s -> %s (ttl %d)\n", ip_src_str, ip_dst_str, ipv6_h->hop_limits); offset += n; break; default: return; } switch (ipproto) { case IPPROTO_UDP: udp_h = (struct udp_hdr *) (l3_h + m->l3_len); snprintf(buf+offset, 4096-offset, " UDP %d -> %d\n", rte_be_to_cpu_16(udp_h->src_port), rte_be_to_cpu_16(udp_h->dst_port)); break; case IPPROTO_TCP: tcp_h = (struct tcp_hdr *) (l3_h + m->l3_len); snprintf(buf+offset, 4096-offset, " TCP %d -> %d\n", rte_be_to_cpu_16(tcp_h->src_port), rte_be_to_cpu_16(tcp_h->dst_port)); default: return; } LOG_RAW(DEBUG, "%s", buf); }