+ return 0;
+}
+
+/**
+ * Parse device parameters.
+ *
+ * @param config
+ * Pointer to device configuration structure.
+ * @param devargs
+ * Device arguments structure.
+ *
+ * @return
+ * 0 on success, a negative errno value otherwise and rte_errno is set.
+ */
+static int
+mlx5_args(struct mlx5_dev_config *config, struct rte_devargs *devargs)
+{
+ const char **params = (const char *[]){
+ MLX5_RXQ_CQE_COMP_EN,
+ MLX5_RXQ_CQE_PAD_EN,
+ MLX5_RXQ_PKT_PAD_EN,
+ MLX5_RX_MPRQ_EN,
+ MLX5_RX_MPRQ_LOG_STRIDE_NUM,
+ MLX5_RX_MPRQ_MAX_MEMCPY_LEN,
+ MLX5_RXQS_MIN_MPRQ,
+ MLX5_TXQ_INLINE,
+ MLX5_TXQ_INLINE_MIN,
+ MLX5_TXQ_INLINE_MAX,
+ MLX5_TXQ_INLINE_MPW,
+ MLX5_TXQS_MIN_INLINE,
+ MLX5_TXQS_MAX_VEC,
+ MLX5_TXQ_MPW_EN,
+ MLX5_TXQ_MPW_HDR_DSEG_EN,
+ MLX5_TXQ_MAX_INLINE_LEN,
+ MLX5_TX_DB_NC,
+ MLX5_TX_VEC_EN,
+ MLX5_RX_VEC_EN,
+ MLX5_L3_VXLAN_EN,
+ MLX5_VF_NL_EN,
+ MLX5_DV_ESW_EN,
+ MLX5_DV_FLOW_EN,
+ MLX5_DV_XMETA_EN,
+ MLX5_MR_EXT_MEMSEG_EN,
+ MLX5_REPRESENTOR,
+ MLX5_MAX_DUMP_FILES_NUM,
+ MLX5_LRO_TIMEOUT_USEC,
+ MLX5_CLASS_ARG_NAME,
+ MLX5_HP_BUF_SIZE,
+ NULL,
+ };
+ struct rte_kvargs *kvlist;
+ int ret = 0;
+ int i;
+
+ if (devargs == NULL)
+ return 0;
+ /* Following UGLY cast is done to pass checkpatch. */
+ kvlist = rte_kvargs_parse(devargs->args, params);
+ if (kvlist == NULL) {
+ rte_errno = EINVAL;
+ return -rte_errno;
+ }
+ /* Process parameters. */
+ for (i = 0; (params[i] != NULL); ++i) {
+ if (rte_kvargs_count(kvlist, params[i])) {
+ ret = rte_kvargs_process(kvlist, params[i],
+ mlx5_args_check, config);
+ if (ret) {
+ rte_errno = EINVAL;
+ rte_kvargs_free(kvlist);
+ return -rte_errno;
+ }
+ }
+ }
+ rte_kvargs_free(kvlist);
+ return 0;
+}
+
+static struct rte_pci_driver mlx5_driver;
+
+/**
+ * PMD global initialization.
+ *
+ * Independent from individual device, this function initializes global
+ * per-PMD data structures distinguishing primary and secondary processes.
+ * Hence, each initialization is called once per a process.
+ *
+ * @return
+ * 0 on success, a negative errno value otherwise and rte_errno is set.
+ */
+static int
+mlx5_init_once(void)
+{
+ struct mlx5_shared_data *sd;
+ struct mlx5_local_data *ld = &mlx5_local_data;
+ int ret = 0;
+
+ if (mlx5_init_shared_data())
+ return -rte_errno;
+ sd = mlx5_shared_data;
+ MLX5_ASSERT(sd);
+ rte_spinlock_lock(&sd->lock);
+ switch (rte_eal_process_type()) {
+ case RTE_PROC_PRIMARY:
+ if (sd->init_done)
+ break;
+ LIST_INIT(&sd->mem_event_cb_list);
+ rte_rwlock_init(&sd->mem_event_rwlock);
+ rte_mem_event_callback_register("MLX5_MEM_EVENT_CB",
+ mlx5_mr_mem_event_cb, NULL);
+ ret = mlx5_mp_init_primary();
+ if (ret)
+ goto out;
+ sd->init_done = true;
+ break;
+ case RTE_PROC_SECONDARY:
+ if (ld->init_done)
+ break;
+ ret = mlx5_mp_init_secondary();
+ if (ret)
+ goto out;
+ ++sd->secondary_cnt;
+ ld->init_done = true;
+ break;
+ default:
+ break;
+ }
+out:
+ rte_spinlock_unlock(&sd->lock);
+ return ret;
+}
+
+/**
+ * Configures the minimal amount of data to inline into WQE
+ * while sending packets.
+ *
+ * - the txq_inline_min has the maximal priority, if this
+ * key is specified in devargs
+ * - if DevX is enabled the inline mode is queried from the
+ * device (HCA attributes and NIC vport context if needed).
+ * - otherwise L2 mode (18 bytes) is assumed for ConnectX-4/4 Lx
+ * and none (0 bytes) for other NICs
+ *
+ * @param spawn
+ * Verbs device parameters (name, port, switch_info) to spawn.
+ * @param config
+ * Device configuration parameters.
+ */
+static void
+mlx5_set_min_inline(struct mlx5_dev_spawn_data *spawn,
+ struct mlx5_dev_config *config)
+{
+ if (config->txq_inline_min != MLX5_ARG_UNSET) {
+ /* Application defines size of inlined data explicitly. */
+ switch (spawn->pci_dev->id.device_id) {
+ case PCI_DEVICE_ID_MELLANOX_CONNECTX4:
+ case PCI_DEVICE_ID_MELLANOX_CONNECTX4VF:
+ if (config->txq_inline_min <
+ (int)MLX5_INLINE_HSIZE_L2) {
+ DRV_LOG(DEBUG,
+ "txq_inline_mix aligned to minimal"
+ " ConnectX-4 required value %d",
+ (int)MLX5_INLINE_HSIZE_L2);
+ config->txq_inline_min = MLX5_INLINE_HSIZE_L2;
+ }
+ break;
+ }
+ goto exit;
+ }
+ if (config->hca_attr.eth_net_offloads) {
+ /* We have DevX enabled, inline mode queried successfully. */
+ switch (config->hca_attr.wqe_inline_mode) {
+ case MLX5_CAP_INLINE_MODE_L2:
+ /* outer L2 header must be inlined. */
+ config->txq_inline_min = MLX5_INLINE_HSIZE_L2;
+ goto exit;
+ case MLX5_CAP_INLINE_MODE_NOT_REQUIRED:
+ /* No inline data are required by NIC. */
+ config->txq_inline_min = MLX5_INLINE_HSIZE_NONE;
+ config->hw_vlan_insert =
+ config->hca_attr.wqe_vlan_insert;
+ DRV_LOG(DEBUG, "Tx VLAN insertion is supported");
+ goto exit;
+ case MLX5_CAP_INLINE_MODE_VPORT_CONTEXT:
+ /* inline mode is defined by NIC vport context. */
+ if (!config->hca_attr.eth_virt)
+ break;
+ switch (config->hca_attr.vport_inline_mode) {
+ case MLX5_INLINE_MODE_NONE:
+ config->txq_inline_min =
+ MLX5_INLINE_HSIZE_NONE;
+ goto exit;
+ case MLX5_INLINE_MODE_L2:
+ config->txq_inline_min =
+ MLX5_INLINE_HSIZE_L2;
+ goto exit;
+ case MLX5_INLINE_MODE_IP:
+ config->txq_inline_min =
+ MLX5_INLINE_HSIZE_L3;
+ goto exit;
+ case MLX5_INLINE_MODE_TCP_UDP:
+ config->txq_inline_min =
+ MLX5_INLINE_HSIZE_L4;
+ goto exit;
+ case MLX5_INLINE_MODE_INNER_L2:
+ config->txq_inline_min =
+ MLX5_INLINE_HSIZE_INNER_L2;
+ goto exit;
+ case MLX5_INLINE_MODE_INNER_IP:
+ config->txq_inline_min =
+ MLX5_INLINE_HSIZE_INNER_L3;
+ goto exit;
+ case MLX5_INLINE_MODE_INNER_TCP_UDP:
+ config->txq_inline_min =
+ MLX5_INLINE_HSIZE_INNER_L4;
+ goto exit;
+ }
+ }
+ }
+ /*
+ * We get here if we are unable to deduce
+ * inline data size with DevX. Try PCI ID
+ * to determine old NICs.
+ */
+ switch (spawn->pci_dev->id.device_id) {
+ case PCI_DEVICE_ID_MELLANOX_CONNECTX4:
+ case PCI_DEVICE_ID_MELLANOX_CONNECTX4VF:
+ case PCI_DEVICE_ID_MELLANOX_CONNECTX4LX:
+ case PCI_DEVICE_ID_MELLANOX_CONNECTX4LXVF:
+ config->txq_inline_min = MLX5_INLINE_HSIZE_L2;
+ config->hw_vlan_insert = 0;
+ break;
+ case PCI_DEVICE_ID_MELLANOX_CONNECTX5:
+ case PCI_DEVICE_ID_MELLANOX_CONNECTX5VF:
+ case PCI_DEVICE_ID_MELLANOX_CONNECTX5EX:
+ case PCI_DEVICE_ID_MELLANOX_CONNECTX5EXVF:
+ /*
+ * These NICs support VLAN insertion from WQE and
+ * report the wqe_vlan_insert flag. But there is the bug
+ * and PFC control may be broken, so disable feature.
+ */
+ config->hw_vlan_insert = 0;
+ config->txq_inline_min = MLX5_INLINE_HSIZE_NONE;
+ break;
+ default:
+ config->txq_inline_min = MLX5_INLINE_HSIZE_NONE;
+ break;
+ }
+exit:
+ DRV_LOG(DEBUG, "min tx inline configured: %d", config->txq_inline_min);
+}
+
+/**
+ * Configures the metadata mask fields in the shared context.
+ *
+ * @param [in] dev
+ * Pointer to Ethernet device.
+ */
+static void
+mlx5_set_metadata_mask(struct rte_eth_dev *dev)
+{
+ struct mlx5_priv *priv = dev->data->dev_private;
+ struct mlx5_ibv_shared *sh = priv->sh;
+ uint32_t meta, mark, reg_c0;
+
+ reg_c0 = ~priv->vport_meta_mask;
+ switch (priv->config.dv_xmeta_en) {
+ case MLX5_XMETA_MODE_LEGACY:
+ meta = UINT32_MAX;
+ mark = MLX5_FLOW_MARK_MASK;
+ break;
+ case MLX5_XMETA_MODE_META16:
+ meta = reg_c0 >> rte_bsf32(reg_c0);
+ mark = MLX5_FLOW_MARK_MASK;
+ break;
+ case MLX5_XMETA_MODE_META32:
+ meta = UINT32_MAX;
+ mark = (reg_c0 >> rte_bsf32(reg_c0)) & MLX5_FLOW_MARK_MASK;
+ break;
+ default:
+ meta = 0;
+ mark = 0;
+ MLX5_ASSERT(false);
+ break;
+ }
+ if (sh->dv_mark_mask && sh->dv_mark_mask != mark)
+ DRV_LOG(WARNING, "metadata MARK mask mismatche %08X:%08X",
+ sh->dv_mark_mask, mark);
+ else
+ sh->dv_mark_mask = mark;
+ if (sh->dv_meta_mask && sh->dv_meta_mask != meta)
+ DRV_LOG(WARNING, "metadata META mask mismatche %08X:%08X",
+ sh->dv_meta_mask, meta);
+ else
+ sh->dv_meta_mask = meta;
+ if (sh->dv_regc0_mask && sh->dv_regc0_mask != reg_c0)
+ DRV_LOG(WARNING, "metadata reg_c0 mask mismatche %08X:%08X",
+ sh->dv_meta_mask, reg_c0);
+ else
+ sh->dv_regc0_mask = reg_c0;
+ DRV_LOG(DEBUG, "metadata mode %u", priv->config.dv_xmeta_en);
+ DRV_LOG(DEBUG, "metadata MARK mask %08X", sh->dv_mark_mask);
+ DRV_LOG(DEBUG, "metadata META mask %08X", sh->dv_meta_mask);
+ DRV_LOG(DEBUG, "metadata reg_c0 mask %08X", sh->dv_regc0_mask);
+}
+
+/**
+ * Allocate page of door-bells and register it using DevX API.
+ *
+ * @param [in] dev
+ * Pointer to Ethernet device.
+ *
+ * @return
+ * Pointer to new page on success, NULL otherwise.
+ */
+static struct mlx5_devx_dbr_page *
+mlx5_alloc_dbr_page(struct rte_eth_dev *dev)
+{
+ struct mlx5_priv *priv = dev->data->dev_private;
+ struct mlx5_devx_dbr_page *page;
+
+ /* Allocate space for door-bell page and management data. */
+ page = rte_calloc_socket(__func__, 1, sizeof(struct mlx5_devx_dbr_page),
+ RTE_CACHE_LINE_SIZE, dev->device->numa_node);
+ if (!page) {
+ DRV_LOG(ERR, "port %u cannot allocate dbr page",
+ dev->data->port_id);
+ return NULL;
+ }
+ /* Register allocated memory. */
+ page->umem = mlx5_glue->devx_umem_reg(priv->sh->ctx, page->dbrs,
+ MLX5_DBR_PAGE_SIZE, 0);
+ if (!page->umem) {
+ DRV_LOG(ERR, "port %u cannot umem reg dbr page",
+ dev->data->port_id);
+ rte_free(page);
+ return NULL;
+ }
+ return page;
+}
+
+/**
+ * Find the next available door-bell, allocate new page if needed.
+ *
+ * @param [in] dev
+ * Pointer to Ethernet device.
+ * @param [out] dbr_page
+ * Door-bell page containing the page data.
+ *
+ * @return
+ * Door-bell address offset on success, a negative error value otherwise.
+ */
+int64_t
+mlx5_get_dbr(struct rte_eth_dev *dev, struct mlx5_devx_dbr_page **dbr_page)
+{
+ struct mlx5_priv *priv = dev->data->dev_private;
+ struct mlx5_devx_dbr_page *page = NULL;
+ uint32_t i, j;
+
+ LIST_FOREACH(page, &priv->dbrpgs, next)
+ if (page->dbr_count < MLX5_DBR_PER_PAGE)
+ break;
+ if (!page) { /* No page with free door-bell exists. */
+ page = mlx5_alloc_dbr_page(dev);
+ if (!page) /* Failed to allocate new page. */
+ return (-1);
+ LIST_INSERT_HEAD(&priv->dbrpgs, page, next);
+ }
+ /* Loop to find bitmap part with clear bit. */
+ for (i = 0;
+ i < MLX5_DBR_BITMAP_SIZE && page->dbr_bitmap[i] == UINT64_MAX;
+ i++)
+ ; /* Empty. */
+ /* Find the first clear bit. */
+ j = rte_bsf64(~page->dbr_bitmap[i]);
+ MLX5_ASSERT(i < (MLX5_DBR_PER_PAGE / 64));
+ page->dbr_bitmap[i] |= (1 << j);
+ page->dbr_count++;
+ *dbr_page = page;
+ return (((i * 64) + j) * sizeof(uint64_t));
+}
+
+/**
+ * Release a door-bell record.
+ *
+ * @param [in] dev
+ * Pointer to Ethernet device.
+ * @param [in] umem_id
+ * UMEM ID of page containing the door-bell record to release.
+ * @param [in] offset
+ * Offset of door-bell record in page.
+ *
+ * @return
+ * 0 on success, a negative error value otherwise.
+ */
+int32_t
+mlx5_release_dbr(struct rte_eth_dev *dev, uint32_t umem_id, uint64_t offset)
+{
+ struct mlx5_priv *priv = dev->data->dev_private;
+ struct mlx5_devx_dbr_page *page = NULL;
+ int ret = 0;
+
+ LIST_FOREACH(page, &priv->dbrpgs, next)
+ /* Find the page this address belongs to. */
+ if (page->umem->umem_id == umem_id)
+ break;
+ if (!page)
+ return -EINVAL;
+ page->dbr_count--;
+ if (!page->dbr_count) {
+ /* Page not used, free it and remove from list. */
+ LIST_REMOVE(page, next);
+ if (page->umem)
+ ret = -mlx5_glue->devx_umem_dereg(page->umem);
+ rte_free(page);
+ } else {
+ /* Mark in bitmap that this door-bell is not in use. */
+ offset /= MLX5_DBR_SIZE;
+ int i = offset / 64;
+ int j = offset % 64;
+
+ page->dbr_bitmap[i] &= ~(1 << j);
+ }
+ return ret;
+}
+
+int
+rte_pmd_mlx5_get_dyn_flag_names(char *names[], unsigned int n)
+{
+ static const char *const dynf_names[] = {
+ RTE_PMD_MLX5_FINE_GRANULARITY_INLINE,
+ RTE_MBUF_DYNFLAG_METADATA_NAME
+ };
+ unsigned int i;
+
+ if (n < RTE_DIM(dynf_names))
+ return -ENOMEM;
+ for (i = 0; i < RTE_DIM(dynf_names); i++) {
+ if (names[i] == NULL)
+ return -EINVAL;
+ strcpy(names[i], dynf_names[i]);
+ }
+ return RTE_DIM(dynf_names);
+}
+
+/**
+ * Check sibling device configurations.
+ *
+ * Sibling devices sharing the Infiniband device context
+ * should have compatible configurations. This regards
+ * representors and bonding slaves.
+ *
+ * @param priv
+ * Private device descriptor.
+ * @param config
+ * Configuration of the device is going to be created.
+ *
+ * @return
+ * 0 on success, EINVAL otherwise
+ */
+static int
+mlx5_dev_check_sibling_config(struct mlx5_priv *priv,
+ struct mlx5_dev_config *config)
+{
+ struct mlx5_ibv_shared *sh = priv->sh;
+ struct mlx5_dev_config *sh_conf = NULL;
+ uint16_t port_id;
+
+ MLX5_ASSERT(sh);
+ /* Nothing to compare for the single/first device. */
+ if (sh->refcnt == 1)
+ return 0;
+ /* Find the device with shared context. */
+ MLX5_ETH_FOREACH_DEV(port_id, priv->pci_dev) {
+ struct mlx5_priv *opriv =
+ rte_eth_devices[port_id].data->dev_private;
+
+ if (opriv && opriv != priv && opriv->sh == sh) {
+ sh_conf = &opriv->config;
+ break;
+ }
+ }
+ if (!sh_conf)
+ return 0;
+ if (sh_conf->dv_flow_en ^ config->dv_flow_en) {
+ DRV_LOG(ERR, "\"dv_flow_en\" configuration mismatch"
+ " for shared %s context", sh->ibdev_name);
+ rte_errno = EINVAL;
+ return rte_errno;
+ }
+ if (sh_conf->dv_xmeta_en ^ config->dv_xmeta_en) {
+ DRV_LOG(ERR, "\"dv_xmeta_en\" configuration mismatch"
+ " for shared %s context", sh->ibdev_name);
+ rte_errno = EINVAL;
+ return rte_errno;
+ }
+ return 0;
+}
+/**
+ * Spawn an Ethernet device from Verbs information.
+ *
+ * @param dpdk_dev
+ * Backing DPDK device.
+ * @param spawn
+ * Verbs device parameters (name, port, switch_info) to spawn.
+ * @param config
+ * Device configuration parameters.
+ *
+ * @return
+ * A valid Ethernet device object on success, NULL otherwise and rte_errno
+ * is set. The following errors are defined:
+ *
+ * EBUSY: device is not supposed to be spawned.
+ * EEXIST: device is already spawned
+ */
+static struct rte_eth_dev *
+mlx5_dev_spawn(struct rte_device *dpdk_dev,
+ struct mlx5_dev_spawn_data *spawn,
+ struct mlx5_dev_config config)
+{
+ const struct mlx5_switch_info *switch_info = &spawn->info;
+ struct mlx5_ibv_shared *sh = NULL;
+ struct ibv_port_attr port_attr;
+ struct mlx5dv_context dv_attr = { .comp_mask = 0 };
+ struct rte_eth_dev *eth_dev = NULL;
+ struct mlx5_priv *priv = NULL;
+ int err = 0;
+ unsigned int hw_padding = 0;
+ unsigned int mps;
+ unsigned int cqe_comp;
+ unsigned int cqe_pad = 0;
+ unsigned int tunnel_en = 0;
+ unsigned int mpls_en = 0;
+ unsigned int swp = 0;
+ unsigned int mprq = 0;
+ unsigned int mprq_min_stride_size_n = 0;
+ unsigned int mprq_max_stride_size_n = 0;
+ unsigned int mprq_min_stride_num_n = 0;
+ unsigned int mprq_max_stride_num_n = 0;
+ struct rte_ether_addr mac;
+ char name[RTE_ETH_NAME_MAX_LEN];
+ int own_domain_id = 0;
+ uint16_t port_id;
+ unsigned int i;
+#ifdef HAVE_MLX5DV_DR_DEVX_PORT
+ struct mlx5dv_devx_port devx_port = { .comp_mask = 0 };
+#endif
+
+ /* Determine if this port representor is supposed to be spawned. */
+ if (switch_info->representor && dpdk_dev->devargs) {
+ struct rte_eth_devargs eth_da;
+
+ err = rte_eth_devargs_parse(dpdk_dev->devargs->args, ð_da);
+ if (err) {
+ rte_errno = -err;
+ DRV_LOG(ERR, "failed to process device arguments: %s",
+ strerror(rte_errno));
+ return NULL;
+ }
+ for (i = 0; i < eth_da.nb_representor_ports; ++i)
+ if (eth_da.representor_ports[i] ==
+ (uint16_t)switch_info->port_name)
+ break;
+ if (i == eth_da.nb_representor_ports) {
+ rte_errno = EBUSY;
+ return NULL;
+ }
+ }
+ /* Build device name. */
+ if (spawn->pf_bond < 0) {
+ /* Single device. */
+ if (!switch_info->representor)
+ strlcpy(name, dpdk_dev->name, sizeof(name));
+ else
+ snprintf(name, sizeof(name), "%s_representor_%u",
+ dpdk_dev->name, switch_info->port_name);
+ } else {
+ /* Bonding device. */
+ if (!switch_info->representor)
+ snprintf(name, sizeof(name), "%s_%s",
+ dpdk_dev->name, spawn->ibv_dev->name);
+ else
+ snprintf(name, sizeof(name), "%s_%s_representor_%u",
+ dpdk_dev->name, spawn->ibv_dev->name,
+ switch_info->port_name);
+ }
+ /* check if the device is already spawned */
+ if (rte_eth_dev_get_port_by_name(name, &port_id) == 0) {
+ rte_errno = EEXIST;
+ return NULL;
+ }
+ DRV_LOG(DEBUG, "naming Ethernet device \"%s\"", name);
+ if (rte_eal_process_type() == RTE_PROC_SECONDARY) {
+ eth_dev = rte_eth_dev_attach_secondary(name);
+ if (eth_dev == NULL) {
+ DRV_LOG(ERR, "can not attach rte ethdev");
+ rte_errno = ENOMEM;
+ return NULL;
+ }
+ eth_dev->device = dpdk_dev;
+ eth_dev->dev_ops = &mlx5_dev_sec_ops;
+ err = mlx5_proc_priv_init(eth_dev);
+ if (err)
+ return NULL;
+ /* Receive command fd from primary process */
+ err = mlx5_mp_req_verbs_cmd_fd(eth_dev);
+ if (err < 0)
+ return NULL;
+ /* Remap UAR for Tx queues. */
+ err = mlx5_tx_uar_init_secondary(eth_dev, err);
+ if (err)
+ return NULL;
+ /*
+ * Ethdev pointer is still required as input since
+ * the primary device is not accessible from the
+ * secondary process.
+ */
+ eth_dev->rx_pkt_burst = mlx5_select_rx_function(eth_dev);
+ eth_dev->tx_pkt_burst = mlx5_select_tx_function(eth_dev);
+ return eth_dev;
+ }
+ /*
+ * Some parameters ("tx_db_nc" in particularly) are needed in
+ * advance to create dv/verbs device context. We proceed the
+ * devargs here to get ones, and later proceed devargs again
+ * to override some hardware settings.
+ */
+ err = mlx5_args(&config, dpdk_dev->devargs);
+ if (err) {
+ err = rte_errno;
+ DRV_LOG(ERR, "failed to process device arguments: %s",
+ strerror(rte_errno));
+ goto error;
+ }
+ sh = mlx5_alloc_shared_ibctx(spawn, &config);
+ if (!sh)
+ return NULL;
+ config.devx = sh->devx;
+#ifdef HAVE_MLX5DV_DR_ACTION_DEST_DEVX_TIR
+ config.dest_tir = 1;
+#endif
+#ifdef HAVE_IBV_MLX5_MOD_SWP
+ dv_attr.comp_mask |= MLX5DV_CONTEXT_MASK_SWP;
+#endif
+ /*
+ * Multi-packet send is supported by ConnectX-4 Lx PF as well
+ * as all ConnectX-5 devices.
+ */
+#ifdef HAVE_IBV_DEVICE_TUNNEL_SUPPORT
+ dv_attr.comp_mask |= MLX5DV_CONTEXT_MASK_TUNNEL_OFFLOADS;
+#endif
+#ifdef HAVE_IBV_DEVICE_STRIDING_RQ_SUPPORT
+ dv_attr.comp_mask |= MLX5DV_CONTEXT_MASK_STRIDING_RQ;
+#endif
+ mlx5_glue->dv_query_device(sh->ctx, &dv_attr);
+ if (dv_attr.flags & MLX5DV_CONTEXT_FLAGS_MPW_ALLOWED) {
+ if (dv_attr.flags & MLX5DV_CONTEXT_FLAGS_ENHANCED_MPW) {
+ DRV_LOG(DEBUG, "enhanced MPW is supported");
+ mps = MLX5_MPW_ENHANCED;
+ } else {
+ DRV_LOG(DEBUG, "MPW is supported");
+ mps = MLX5_MPW;
+ }
+ } else {
+ DRV_LOG(DEBUG, "MPW isn't supported");
+ mps = MLX5_MPW_DISABLED;
+ }
+#ifdef HAVE_IBV_MLX5_MOD_SWP
+ if (dv_attr.comp_mask & MLX5DV_CONTEXT_MASK_SWP)
+ swp = dv_attr.sw_parsing_caps.sw_parsing_offloads;
+ DRV_LOG(DEBUG, "SWP support: %u", swp);
+#endif
+ config.swp = !!swp;
+#ifdef HAVE_IBV_DEVICE_STRIDING_RQ_SUPPORT
+ if (dv_attr.comp_mask & MLX5DV_CONTEXT_MASK_STRIDING_RQ) {
+ struct mlx5dv_striding_rq_caps mprq_caps =
+ dv_attr.striding_rq_caps;
+
+ DRV_LOG(DEBUG, "\tmin_single_stride_log_num_of_bytes: %d",
+ mprq_caps.min_single_stride_log_num_of_bytes);
+ DRV_LOG(DEBUG, "\tmax_single_stride_log_num_of_bytes: %d",
+ mprq_caps.max_single_stride_log_num_of_bytes);
+ DRV_LOG(DEBUG, "\tmin_single_wqe_log_num_of_strides: %d",
+ mprq_caps.min_single_wqe_log_num_of_strides);
+ DRV_LOG(DEBUG, "\tmax_single_wqe_log_num_of_strides: %d",
+ mprq_caps.max_single_wqe_log_num_of_strides);
+ DRV_LOG(DEBUG, "\tsupported_qpts: %d",
+ mprq_caps.supported_qpts);
+ DRV_LOG(DEBUG, "device supports Multi-Packet RQ");
+ mprq = 1;
+ mprq_min_stride_size_n =
+ mprq_caps.min_single_stride_log_num_of_bytes;
+ mprq_max_stride_size_n =
+ mprq_caps.max_single_stride_log_num_of_bytes;
+ mprq_min_stride_num_n =
+ mprq_caps.min_single_wqe_log_num_of_strides;
+ mprq_max_stride_num_n =
+ mprq_caps.max_single_wqe_log_num_of_strides;
+ config.mprq.stride_num_n = RTE_MAX(MLX5_MPRQ_STRIDE_NUM_N,
+ mprq_min_stride_num_n);
+ }
+#endif
+ if (RTE_CACHE_LINE_SIZE == 128 &&
+ !(dv_attr.flags & MLX5DV_CONTEXT_FLAGS_CQE_128B_COMP))
+ cqe_comp = 0;
+ else
+ cqe_comp = 1;
+ config.cqe_comp = cqe_comp;
+#ifdef HAVE_IBV_MLX5_MOD_CQE_128B_PAD
+ /* Whether device supports 128B Rx CQE padding. */
+ cqe_pad = RTE_CACHE_LINE_SIZE == 128 &&
+ (dv_attr.flags & MLX5DV_CONTEXT_FLAGS_CQE_128B_PAD);
+#endif
+#ifdef HAVE_IBV_DEVICE_TUNNEL_SUPPORT
+ if (dv_attr.comp_mask & MLX5DV_CONTEXT_MASK_TUNNEL_OFFLOADS) {
+ tunnel_en = ((dv_attr.tunnel_offloads_caps &
+ MLX5DV_RAW_PACKET_CAP_TUNNELED_OFFLOAD_VXLAN) &&
+ (dv_attr.tunnel_offloads_caps &
+ MLX5DV_RAW_PACKET_CAP_TUNNELED_OFFLOAD_GRE) &&
+ (dv_attr.tunnel_offloads_caps &
+ MLX5DV_RAW_PACKET_CAP_TUNNELED_OFFLOAD_GENEVE));
+ }
+ DRV_LOG(DEBUG, "tunnel offloading is %ssupported",
+ tunnel_en ? "" : "not ");
+#else
+ DRV_LOG(WARNING,
+ "tunnel offloading disabled due to old OFED/rdma-core version");
+#endif
+ config.tunnel_en = tunnel_en;
+#ifdef HAVE_IBV_DEVICE_MPLS_SUPPORT
+ mpls_en = ((dv_attr.tunnel_offloads_caps &
+ MLX5DV_RAW_PACKET_CAP_TUNNELED_OFFLOAD_CW_MPLS_OVER_GRE) &&
+ (dv_attr.tunnel_offloads_caps &
+ MLX5DV_RAW_PACKET_CAP_TUNNELED_OFFLOAD_CW_MPLS_OVER_UDP));
+ DRV_LOG(DEBUG, "MPLS over GRE/UDP tunnel offloading is %ssupported",
+ mpls_en ? "" : "not ");
+#else
+ DRV_LOG(WARNING, "MPLS over GRE/UDP tunnel offloading disabled due to"
+ " old OFED/rdma-core version or firmware configuration");
+#endif
+ config.mpls_en = mpls_en;
+ /* Check port status. */
+ err = mlx5_glue->query_port(sh->ctx, spawn->ibv_port, &port_attr);
+ if (err) {
+ DRV_LOG(ERR, "port query failed: %s", strerror(err));
+ goto error;
+ }
+ if (port_attr.link_layer != IBV_LINK_LAYER_ETHERNET) {
+ DRV_LOG(ERR, "port is not configured in Ethernet mode");
+ err = EINVAL;
+ goto error;
+ }
+ if (port_attr.state != IBV_PORT_ACTIVE)
+ DRV_LOG(DEBUG, "port is not active: \"%s\" (%d)",
+ mlx5_glue->port_state_str(port_attr.state),
+ port_attr.state);
+ /* Allocate private eth device data. */
+ priv = rte_zmalloc("ethdev private structure",
+ sizeof(*priv),
+ RTE_CACHE_LINE_SIZE);
+ if (priv == NULL) {
+ DRV_LOG(ERR, "priv allocation failure");
+ err = ENOMEM;
+ goto error;
+ }
+ priv->sh = sh;
+ priv->ibv_port = spawn->ibv_port;
+ priv->pci_dev = spawn->pci_dev;
+ priv->mtu = RTE_ETHER_MTU;
+#ifndef RTE_ARCH_64
+ /* Initialize UAR access locks for 32bit implementations. */
+ rte_spinlock_init(&priv->uar_lock_cq);
+ for (i = 0; i < MLX5_UAR_PAGE_NUM_MAX; i++)
+ rte_spinlock_init(&priv->uar_lock[i]);
+#endif
+ /* Some internal functions rely on Netlink sockets, open them now. */
+ priv->nl_socket_rdma = mlx5_nl_init(NETLINK_RDMA);
+ priv->nl_socket_route = mlx5_nl_init(NETLINK_ROUTE);
+ priv->representor = !!switch_info->representor;
+ priv->master = !!switch_info->master;
+ priv->domain_id = RTE_ETH_DEV_SWITCH_DOMAIN_ID_INVALID;
+ priv->vport_meta_tag = 0;
+ priv->vport_meta_mask = 0;
+ priv->pf_bond = spawn->pf_bond;
+#ifdef HAVE_MLX5DV_DR_DEVX_PORT
+ /*
+ * The DevX port query API is implemented. E-Switch may use
+ * either vport or reg_c[0] metadata register to match on
+ * vport index. The engaged part of metadata register is
+ * defined by mask.
+ */
+ if (switch_info->representor || switch_info->master) {
+ devx_port.comp_mask = MLX5DV_DEVX_PORT_VPORT |
+ MLX5DV_DEVX_PORT_MATCH_REG_C_0;
+ err = mlx5_glue->devx_port_query(sh->ctx, spawn->ibv_port,
+ &devx_port);
+ if (err) {
+ DRV_LOG(WARNING,
+ "can't query devx port %d on device %s",
+ spawn->ibv_port, spawn->ibv_dev->name);
+ devx_port.comp_mask = 0;
+ }
+ }
+ if (devx_port.comp_mask & MLX5DV_DEVX_PORT_MATCH_REG_C_0) {
+ priv->vport_meta_tag = devx_port.reg_c_0.value;
+ priv->vport_meta_mask = devx_port.reg_c_0.mask;
+ if (!priv->vport_meta_mask) {
+ DRV_LOG(ERR, "vport zero mask for port %d"
+ " on bonding device %s",
+ spawn->ibv_port, spawn->ibv_dev->name);
+ err = ENOTSUP;
+ goto error;
+ }
+ if (priv->vport_meta_tag & ~priv->vport_meta_mask) {
+ DRV_LOG(ERR, "invalid vport tag for port %d"
+ " on bonding device %s",
+ spawn->ibv_port, spawn->ibv_dev->name);
+ err = ENOTSUP;
+ goto error;
+ }
+ }
+ if (devx_port.comp_mask & MLX5DV_DEVX_PORT_VPORT) {
+ priv->vport_id = devx_port.vport_num;
+ } else if (spawn->pf_bond >= 0) {
+ DRV_LOG(ERR, "can't deduce vport index for port %d"
+ " on bonding device %s",
+ spawn->ibv_port, spawn->ibv_dev->name);
+ err = ENOTSUP;
+ goto error;
+ } else {
+ /* Suppose vport index in compatible way. */
+ priv->vport_id = switch_info->representor ?
+ switch_info->port_name + 1 : -1;
+ }
+#else
+ /*
+ * Kernel/rdma_core support single E-Switch per PF configurations
+ * only and vport_id field contains the vport index for
+ * associated VF, which is deduced from representor port name.
+ * For example, let's have the IB device port 10, it has
+ * attached network device eth0, which has port name attribute
+ * pf0vf2, we can deduce the VF number as 2, and set vport index
+ * as 3 (2+1). This assigning schema should be changed if the
+ * multiple E-Switch instances per PF configurations or/and PCI
+ * subfunctions are added.
+ */
+ priv->vport_id = switch_info->representor ?
+ switch_info->port_name + 1 : -1;
+#endif
+ /* representor_id field keeps the unmodified VF index. */
+ priv->representor_id = switch_info->representor ?
+ switch_info->port_name : -1;
+ /*
+ * Look for sibling devices in order to reuse their switch domain
+ * if any, otherwise allocate one.
+ */
+ MLX5_ETH_FOREACH_DEV(port_id, priv->pci_dev) {
+ const struct mlx5_priv *opriv =
+ rte_eth_devices[port_id].data->dev_private;
+
+ if (!opriv ||
+ opriv->sh != priv->sh ||
+ opriv->domain_id ==
+ RTE_ETH_DEV_SWITCH_DOMAIN_ID_INVALID)
+ continue;
+ priv->domain_id = opriv->domain_id;
+ break;
+ }
+ if (priv->domain_id == RTE_ETH_DEV_SWITCH_DOMAIN_ID_INVALID) {
+ err = rte_eth_switch_domain_alloc(&priv->domain_id);
+ if (err) {
+ err = rte_errno;
+ DRV_LOG(ERR, "unable to allocate switch domain: %s",
+ strerror(rte_errno));
+ goto error;
+ }
+ own_domain_id = 1;
+ }
+ /* Override some values set by hardware configuration. */
+ mlx5_args(&config, dpdk_dev->devargs);
+ err = mlx5_dev_check_sibling_config(priv, &config);
+ if (err)
+ goto error;
+ config.hw_csum = !!(sh->device_attr.device_cap_flags_ex &
+ IBV_DEVICE_RAW_IP_CSUM);
+ DRV_LOG(DEBUG, "checksum offloading is %ssupported",
+ (config.hw_csum ? "" : "not "));
+#if !defined(HAVE_IBV_DEVICE_COUNTERS_SET_V42) && \
+ !defined(HAVE_IBV_DEVICE_COUNTERS_SET_V45)
+ DRV_LOG(DEBUG, "counters are not supported");
+#endif
+#if !defined(HAVE_IBV_FLOW_DV_SUPPORT) || !defined(HAVE_MLX5DV_DR)
+ if (config.dv_flow_en) {
+ DRV_LOG(WARNING, "DV flow is not supported");
+ config.dv_flow_en = 0;
+ }
+#endif
+ config.ind_table_max_size =
+ sh->device_attr.rss_caps.max_rwq_indirection_table_size;
+ /*
+ * Remove this check once DPDK supports larger/variable
+ * indirection tables.
+ */
+ if (config.ind_table_max_size > (unsigned int)ETH_RSS_RETA_SIZE_512)
+ config.ind_table_max_size = ETH_RSS_RETA_SIZE_512;
+ DRV_LOG(DEBUG, "maximum Rx indirection table size is %u",
+ config.ind_table_max_size);
+ config.hw_vlan_strip = !!(sh->device_attr.raw_packet_caps &
+ IBV_RAW_PACKET_CAP_CVLAN_STRIPPING);
+ DRV_LOG(DEBUG, "VLAN stripping is %ssupported",
+ (config.hw_vlan_strip ? "" : "not "));
+ config.hw_fcs_strip = !!(sh->device_attr.raw_packet_caps &
+ IBV_RAW_PACKET_CAP_SCATTER_FCS);
+ DRV_LOG(DEBUG, "FCS stripping configuration is %ssupported",
+ (config.hw_fcs_strip ? "" : "not "));
+#if defined(HAVE_IBV_WQ_FLAG_RX_END_PADDING)
+ hw_padding = !!sh->device_attr.rx_pad_end_addr_align;
+#elif defined(HAVE_IBV_WQ_FLAGS_PCI_WRITE_END_PADDING)
+ hw_padding = !!(sh->device_attr.device_cap_flags_ex &
+ IBV_DEVICE_PCI_WRITE_END_PADDING);
+#endif
+ if (config.hw_padding && !hw_padding) {
+ DRV_LOG(DEBUG, "Rx end alignment padding isn't supported");
+ config.hw_padding = 0;
+ } else if (config.hw_padding) {
+ DRV_LOG(DEBUG, "Rx end alignment padding is enabled");
+ }
+ config.tso = (sh->device_attr.tso_caps.max_tso > 0 &&
+ (sh->device_attr.tso_caps.supported_qpts &
+ (1 << IBV_QPT_RAW_PACKET)));
+ if (config.tso)
+ config.tso_max_payload_sz = sh->device_attr.tso_caps.max_tso;
+ /*
+ * MPW is disabled by default, while the Enhanced MPW is enabled
+ * by default.
+ */
+ if (config.mps == MLX5_ARG_UNSET)
+ config.mps = (mps == MLX5_MPW_ENHANCED) ? MLX5_MPW_ENHANCED :
+ MLX5_MPW_DISABLED;
+ else
+ config.mps = config.mps ? mps : MLX5_MPW_DISABLED;
+ DRV_LOG(INFO, "%sMPS is %s",
+ config.mps == MLX5_MPW_ENHANCED ? "enhanced " :
+ config.mps == MLX5_MPW ? "legacy " : "",
+ config.mps != MLX5_MPW_DISABLED ? "enabled" : "disabled");
+ if (config.cqe_comp && !cqe_comp) {
+ DRV_LOG(WARNING, "Rx CQE compression isn't supported");
+ config.cqe_comp = 0;
+ }
+ if (config.cqe_pad && !cqe_pad) {
+ DRV_LOG(WARNING, "Rx CQE padding isn't supported");
+ config.cqe_pad = 0;
+ } else if (config.cqe_pad) {
+ DRV_LOG(INFO, "Rx CQE padding is enabled");
+ }
+ if (config.devx) {
+ priv->counter_fallback = 0;
+ err = mlx5_devx_cmd_query_hca_attr(sh->ctx, &config.hca_attr);
+ if (err) {
+ err = -err;
+ goto error;
+ }
+ if (!config.hca_attr.flow_counters_dump)
+ priv->counter_fallback = 1;
+#ifndef HAVE_IBV_DEVX_ASYNC
+ priv->counter_fallback = 1;
+#endif
+ if (priv->counter_fallback)
+ DRV_LOG(INFO, "Use fall-back DV counter management");
+ /* Check for LRO support. */
+ if (config.dest_tir && config.hca_attr.lro_cap &&
+ config.dv_flow_en) {
+ /* TBD check tunnel lro caps. */
+ config.lro.supported = config.hca_attr.lro_cap;
+ DRV_LOG(DEBUG, "Device supports LRO");
+ /*
+ * If LRO timeout is not configured by application,
+ * use the minimal supported value.
+ */
+ if (!config.lro.timeout)
+ config.lro.timeout =
+ config.hca_attr.lro_timer_supported_periods[0];
+ DRV_LOG(DEBUG, "LRO session timeout set to %d usec",
+ config.lro.timeout);
+ }
+#if defined(HAVE_MLX5DV_DR) && defined(HAVE_MLX5_DR_CREATE_ACTION_FLOW_METER)
+ if (config.hca_attr.qos.sup && config.hca_attr.qos.srtcm_sup &&
+ config.dv_flow_en) {
+ uint8_t reg_c_mask =
+ config.hca_attr.qos.flow_meter_reg_c_ids;
+ /*
+ * Meter needs two REG_C's for color match and pre-sfx
+ * flow match. Here get the REG_C for color match.
+ * REG_C_0 and REG_C_1 is reserved for metadata feature.
+ */
+ reg_c_mask &= 0xfc;
+ if (__builtin_popcount(reg_c_mask) < 1) {
+ priv->mtr_en = 0;
+ DRV_LOG(WARNING, "No available register for"
+ " meter.");
+ } else {
+ priv->mtr_color_reg = ffs(reg_c_mask) - 1 +
+ REG_C_0;
+ priv->mtr_en = 1;
+ priv->mtr_reg_share =
+ config.hca_attr.qos.flow_meter_reg_share;
+ DRV_LOG(DEBUG, "The REG_C meter uses is %d",
+ priv->mtr_color_reg);
+ }
+ }
+#endif
+ }
+ if (config.mprq.enabled && mprq) {
+ if (config.mprq.stride_num_n > mprq_max_stride_num_n ||
+ config.mprq.stride_num_n < mprq_min_stride_num_n) {
+ config.mprq.stride_num_n =
+ RTE_MAX(MLX5_MPRQ_STRIDE_NUM_N,
+ mprq_min_stride_num_n);
+ DRV_LOG(WARNING,
+ "the number of strides"
+ " for Multi-Packet RQ is out of range,"
+ " setting default value (%u)",
+ 1 << config.mprq.stride_num_n);
+ }
+ config.mprq.min_stride_size_n = mprq_min_stride_size_n;
+ config.mprq.max_stride_size_n = mprq_max_stride_size_n;
+ } else if (config.mprq.enabled && !mprq) {
+ DRV_LOG(WARNING, "Multi-Packet RQ isn't supported");
+ config.mprq.enabled = 0;
+ }
+ if (config.max_dump_files_num == 0)
+ config.max_dump_files_num = 128;
+ eth_dev = rte_eth_dev_allocate(name);
+ if (eth_dev == NULL) {
+ DRV_LOG(ERR, "can not allocate rte ethdev");
+ err = ENOMEM;
+ goto error;
+ }
+ /* Flag to call rte_eth_dev_release_port() in rte_eth_dev_close(). */
+ eth_dev->data->dev_flags |= RTE_ETH_DEV_CLOSE_REMOVE;
+ if (priv->representor) {
+ eth_dev->data->dev_flags |= RTE_ETH_DEV_REPRESENTOR;
+ eth_dev->data->representor_id = priv->representor_id;
+ }
+ /*
+ * Store associated network device interface index. This index
+ * is permanent throughout the lifetime of device. So, we may store
+ * the ifindex here and use the cached value further.
+ */
+ MLX5_ASSERT(spawn->ifindex);
+ priv->if_index = spawn->ifindex;
+ eth_dev->data->dev_private = priv;
+ priv->dev_data = eth_dev->data;
+ eth_dev->data->mac_addrs = priv->mac;
+ eth_dev->device = dpdk_dev;
+ /* Configure the first MAC address by default. */
+ if (mlx5_get_mac(eth_dev, &mac.addr_bytes)) {