diff options
Diffstat (limited to 'drivers/net/ethernet/mellanox')
44 files changed, 5928 insertions, 602 deletions
diff --git a/drivers/net/ethernet/mellanox/mlx4/cmd.c b/drivers/net/ethernet/mellanox/mlx4/cmd.c index 2177e56ed..d48d57934 100644 --- a/drivers/net/ethernet/mellanox/mlx4/cmd.c +++ b/drivers/net/ethernet/mellanox/mlx4/cmd.c @@ -1010,7 +1010,7 @@ static int mlx4_MAD_IFC_wrapper(struct mlx4_dev *dev, int slave, if (!(smp->mgmt_class == IB_MGMT_CLASS_SUBN_LID_ROUTED && smp->method == IB_MGMT_METHOD_GET) || network_view) { mlx4_err(dev, "Unprivileged slave %d is trying to execute a Subnet MGMT MAD, class 0x%x, method 0x%x, view=%s for attr 0x%x. Rejecting\n", - slave, smp->method, smp->mgmt_class, + slave, smp->mgmt_class, smp->method, network_view ? "Network" : "Host", be16_to_cpu(smp->attr_id)); return -EPERM; diff --git a/drivers/net/ethernet/mellanox/mlx4/en_clock.c b/drivers/net/ethernet/mellanox/mlx4/en_clock.c index 8a083d73e..038f9ce39 100644 --- a/drivers/net/ethernet/mellanox/mlx4/en_clock.c +++ b/drivers/net/ethernet/mellanox/mlx4/en_clock.c @@ -242,6 +242,13 @@ void mlx4_en_init_timestamp(struct mlx4_en_dev *mdev) unsigned long flags; u64 ns, zero = 0; + /* mlx4_en_init_timestamp is called for each netdev. + * mdev->ptp_clock is common for all ports, skip initialization if + * was done for other port. + */ + if (mdev->ptp_clock) + return; + rwlock_init(&mdev->clock_lock); memset(&mdev->cycles, 0, sizeof(mdev->cycles)); diff --git a/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c index f79d81243..ddb554188 100644 --- a/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c +++ b/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c @@ -95,9 +95,6 @@ mlx4_en_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *drvinfo) (u16) (mdev->dev->caps.fw_ver & 0xffff)); strlcpy(drvinfo->bus_info, pci_name(mdev->dev->persist->pdev), sizeof(drvinfo->bus_info)); - drvinfo->n_stats = 0; - drvinfo->regdump_len = 0; - drvinfo->eedump_len = 0; } static const char mlx4_en_priv_flags[][ETH_GSTRING_LEN] = { diff --git a/drivers/net/ethernet/mellanox/mlx4/en_main.c b/drivers/net/ethernet/mellanox/mlx4/en_main.c index a946e4bf7..e0ec280a7 100644 --- a/drivers/net/ethernet/mellanox/mlx4/en_main.c +++ b/drivers/net/ethernet/mellanox/mlx4/en_main.c @@ -123,6 +123,28 @@ void mlx4_en_update_loopback_state(struct net_device *dev, */ if (mlx4_is_mfunc(priv->mdev->dev) || priv->validate_loopback) priv->flags |= MLX4_EN_FLAG_ENABLE_HW_LOOPBACK; + + mutex_lock(&priv->mdev->state_lock); + if (priv->mdev->dev->caps.flags2 & + MLX4_DEV_CAP_FLAG2_UPDATE_QP_SRC_CHECK_LB && + priv->rss_map.indir_qp.qpn) { + int i; + int err = 0; + int loopback = !!(features & NETIF_F_LOOPBACK); + + for (i = 0; i < priv->rx_ring_num; i++) { + int ret; + + ret = mlx4_en_change_mcast_lb(priv, + &priv->rss_map.qps[i], + loopback); + if (!err) + err = ret; + } + if (err) + mlx4_warn(priv->mdev, "failed to change mcast loopback\n"); + } + mutex_unlock(&priv->mdev->state_lock); } static int mlx4_en_get_profile(struct mlx4_en_dev *mdev) @@ -210,9 +232,6 @@ static void mlx4_en_remove(struct mlx4_dev *dev, void *endev_ptr) if (mdev->pndev[i]) mlx4_en_destroy_netdev(mdev->pndev[i]); - if (mdev->dev->caps.flags2 & MLX4_DEV_CAP_FLAG2_TS) - mlx4_en_remove_timestamp(mdev); - flush_workqueue(mdev->workqueue); destroy_workqueue(mdev->workqueue); (void) mlx4_mr_free(dev, &mdev->mr); @@ -298,10 +317,6 @@ static void *mlx4_en_add(struct mlx4_dev *dev) mlx4_foreach_port(i, dev, MLX4_PORT_TYPE_ETH) mdev->port_cnt++; - /* Initialize time stamp mechanism */ - if (mdev->dev->caps.flags2 & MLX4_DEV_CAP_FLAG2_TS) - mlx4_en_init_timestamp(mdev); - /* Set default number of RX rings*/ mlx4_en_set_num_rx_rings(mdev); diff --git a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c index 4726122ea..7869f97de 100644 --- a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c +++ b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c @@ -573,10 +573,8 @@ static int mlx4_en_get_qp(struct mlx4_en_priv *priv) { struct mlx4_en_dev *mdev = priv->mdev; struct mlx4_dev *dev = mdev->dev; - struct mlx4_mac_entry *entry; int index = 0; int err = 0; - u64 reg_id = 0; int *qpn = &priv->base_qpn; u64 mac = mlx4_mac_to_u64(priv->dev->dev_addr); @@ -600,44 +598,11 @@ static int mlx4_en_get_qp(struct mlx4_en_priv *priv) en_dbg(DRV, priv, "Reserved qp %d\n", *qpn); if (err) { en_err(priv, "Failed to reserve qp for mac registration\n"); - goto qp_err; - } - - err = mlx4_en_uc_steer_add(priv, priv->dev->dev_addr, qpn, ®_id); - if (err) - goto steer_err; - - err = mlx4_en_tunnel_steer_add(priv, priv->dev->dev_addr, *qpn, - &priv->tunnel_reg_id); - if (err) - goto tunnel_err; - - entry = kmalloc(sizeof(*entry), GFP_KERNEL); - if (!entry) { - err = -ENOMEM; - goto alloc_err; + mlx4_unregister_mac(dev, priv->port, mac); + return err; } - memcpy(entry->mac, priv->dev->dev_addr, sizeof(entry->mac)); - memcpy(priv->current_mac, entry->mac, sizeof(priv->current_mac)); - entry->reg_id = reg_id; - - hlist_add_head_rcu(&entry->hlist, - &priv->mac_hash[entry->mac[MLX4_EN_MAC_HASH_IDX]]); return 0; - -alloc_err: - if (priv->tunnel_reg_id) - mlx4_flow_detach(priv->mdev->dev, priv->tunnel_reg_id); -tunnel_err: - mlx4_en_uc_steer_release(priv, priv->dev->dev_addr, *qpn, reg_id); - -steer_err: - mlx4_qp_release_range(dev, *qpn, 1); - -qp_err: - mlx4_unregister_mac(dev, priv->port, mac); - return err; } static void mlx4_en_put_qp(struct mlx4_en_priv *priv) @@ -645,39 +610,13 @@ static void mlx4_en_put_qp(struct mlx4_en_priv *priv) struct mlx4_en_dev *mdev = priv->mdev; struct mlx4_dev *dev = mdev->dev; int qpn = priv->base_qpn; - u64 mac; if (dev->caps.steering_mode == MLX4_STEERING_MODE_A0) { - mac = mlx4_mac_to_u64(priv->dev->dev_addr); + u64 mac = mlx4_mac_to_u64(priv->dev->dev_addr); en_dbg(DRV, priv, "Registering MAC: %pM for deleting\n", priv->dev->dev_addr); mlx4_unregister_mac(dev, priv->port, mac); } else { - struct mlx4_mac_entry *entry; - struct hlist_node *tmp; - struct hlist_head *bucket; - unsigned int i; - - for (i = 0; i < MLX4_EN_MAC_HASH_SIZE; ++i) { - bucket = &priv->mac_hash[i]; - hlist_for_each_entry_safe(entry, tmp, bucket, hlist) { - mac = mlx4_mac_to_u64(entry->mac); - en_dbg(DRV, priv, "Registering MAC: %pM for deleting\n", - entry->mac); - mlx4_en_uc_steer_release(priv, entry->mac, - qpn, entry->reg_id); - - mlx4_unregister_mac(dev, priv->port, mac); - hlist_del_rcu(&entry->hlist); - kfree_rcu(entry, rcu); - } - } - - if (priv->tunnel_reg_id) { - mlx4_flow_detach(priv->mdev->dev, priv->tunnel_reg_id); - priv->tunnel_reg_id = 0; - } - en_dbg(DRV, priv, "Releasing qp: port %d, qpn %d\n", priv->port, qpn); mlx4_qp_release_range(dev, qpn, 1); @@ -1283,6 +1222,75 @@ static void mlx4_en_netpoll(struct net_device *dev) } #endif +static int mlx4_en_set_rss_steer_rules(struct mlx4_en_priv *priv) +{ + u64 reg_id; + int err = 0; + int *qpn = &priv->base_qpn; + struct mlx4_mac_entry *entry; + + err = mlx4_en_uc_steer_add(priv, priv->dev->dev_addr, qpn, ®_id); + if (err) + return err; + + err = mlx4_en_tunnel_steer_add(priv, priv->dev->dev_addr, *qpn, + &priv->tunnel_reg_id); + if (err) + goto tunnel_err; + + entry = kmalloc(sizeof(*entry), GFP_KERNEL); + if (!entry) { + err = -ENOMEM; + goto alloc_err; + } + + memcpy(entry->mac, priv->dev->dev_addr, sizeof(entry->mac)); + memcpy(priv->current_mac, entry->mac, sizeof(priv->current_mac)); + entry->reg_id = reg_id; + hlist_add_head_rcu(&entry->hlist, + &priv->mac_hash[entry->mac[MLX4_EN_MAC_HASH_IDX]]); + + return 0; + +alloc_err: + if (priv->tunnel_reg_id) + mlx4_flow_detach(priv->mdev->dev, priv->tunnel_reg_id); + +tunnel_err: + mlx4_en_uc_steer_release(priv, priv->dev->dev_addr, *qpn, reg_id); + return err; +} + +static void mlx4_en_delete_rss_steer_rules(struct mlx4_en_priv *priv) +{ + u64 mac; + unsigned int i; + int qpn = priv->base_qpn; + struct hlist_head *bucket; + struct hlist_node *tmp; + struct mlx4_mac_entry *entry; + + for (i = 0; i < MLX4_EN_MAC_HASH_SIZE; ++i) { + bucket = &priv->mac_hash[i]; + hlist_for_each_entry_safe(entry, tmp, bucket, hlist) { + mac = mlx4_mac_to_u64(entry->mac); + en_dbg(DRV, priv, "Registering MAC:%pM for deleting\n", + entry->mac); + mlx4_en_uc_steer_release(priv, entry->mac, + qpn, entry->reg_id); + + mlx4_unregister_mac(priv->mdev->dev, priv->port, mac); + hlist_del_rcu(&entry->hlist); + kfree_rcu(entry, rcu); + } + } + + if (priv->tunnel_reg_id) { + mlx4_flow_detach(priv->mdev->dev, priv->tunnel_reg_id); + priv->tunnel_reg_id = 0; + } +} + static void mlx4_en_tx_timeout(struct net_device *dev) { struct mlx4_en_priv *priv = netdev_priv(dev); @@ -1684,6 +1692,11 @@ int mlx4_en_start_port(struct net_device *dev) goto tx_err; } + /* Set Unicast and VXLAN steering rules */ + if (mdev->dev->caps.steering_mode != MLX4_STEERING_MODE_A0 && + mlx4_en_set_rss_steer_rules(priv)) + mlx4_warn(mdev, "Failed setting steering rules\n"); + /* Attach rx QP to bradcast address */ eth_broadcast_addr(&mc_list[10]); mc_list[5] = priv->port; /* needed for B0 steering support */ @@ -1831,6 +1844,9 @@ void mlx4_en_stop_port(struct net_device *dev, int detach) for (i = 0; i < priv->tx_ring_num; i++) mlx4_en_free_tx_buf(dev, priv->tx_ring[i]); + if (mdev->dev->caps.steering_mode != MLX4_STEERING_MODE_A0) + mlx4_en_delete_rss_steer_rules(priv); + /* Free RSS qps */ mlx4_en_release_rss_steer(priv); @@ -2056,6 +2072,9 @@ void mlx4_en_destroy_netdev(struct net_device *dev) /* flush any pending task for this netdev */ flush_workqueue(mdev->workqueue); + if (mdev->dev->caps.flags2 & MLX4_DEV_CAP_FLAG2_TS) + mlx4_en_remove_timestamp(mdev); + /* Detach the netdev so tasks would not attempt to access it */ mutex_lock(&mdev->state_lock); mdev->pndev[priv->port] = NULL; @@ -2800,7 +2819,6 @@ int mlx4_en_init_netdev(struct mlx4_en_dev *mdev, int port, struct mlx4_en_priv *priv; int i; int err; - u64 mac_u64; dev = alloc_etherdev_mqs(sizeof(struct mlx4_en_priv), MAX_TX_RINGS, MAX_RX_RINGS); @@ -2892,17 +2910,17 @@ int mlx4_en_init_netdev(struct mlx4_en_dev *mdev, int port, dev->addr_len = ETH_ALEN; mlx4_en_u64_to_mac(dev->dev_addr, mdev->dev->caps.def_mac[priv->port]); if (!is_valid_ether_addr(dev->dev_addr)) { - if (mlx4_is_slave(priv->mdev->dev)) { - eth_hw_addr_random(dev); - en_warn(priv, "Assigned random MAC address %pM\n", dev->dev_addr); - mac_u64 = mlx4_mac_to_u64(dev->dev_addr); - mdev->dev->caps.def_mac[priv->port] = mac_u64; - } else { - en_err(priv, "Port: %d, invalid mac burned: %pM, quiting\n", - priv->port, dev->dev_addr); - err = -EINVAL; - goto out; - } + en_err(priv, "Port: %d, invalid mac burned: %pM, quiting\n", + priv->port, dev->dev_addr); + err = -EINVAL; + goto out; + } else if (mlx4_is_slave(priv->mdev->dev) && + (priv->mdev->dev->port_random_macs & 1 << priv->port)) { + /* Random MAC was assigned in mlx4_slave_cap + * in mlx4_core module + */ + dev->addr_assign_type |= NET_ADDR_RANDOM; + en_warn(priv, "Assigned random MAC address %pM\n", dev->dev_addr); } memcpy(priv->current_mac, dev->dev_addr, sizeof(priv->current_mac)); @@ -3043,9 +3061,12 @@ int mlx4_en_init_netdev(struct mlx4_en_dev *mdev, int port, } queue_delayed_work(mdev->workqueue, &priv->stats_task, STATS_DELAY); + /* Initialize time stamp mechanism */ if (mdev->dev->caps.flags2 & MLX4_DEV_CAP_FLAG2_TS) - queue_delayed_work(mdev->workqueue, &priv->service_task, - SERVICE_TASK_DELAY); + mlx4_en_init_timestamp(mdev); + + queue_delayed_work(mdev->workqueue, &priv->service_task, + SERVICE_TASK_DELAY); mlx4_en_set_stats_bitmap(mdev->dev, &priv->stats_bitmap, mdev->profile.prof[priv->port].rx_ppp, diff --git a/drivers/net/ethernet/mellanox/mlx4/en_resources.c b/drivers/net/ethernet/mellanox/mlx4/en_resources.c index e482fa1bb..12aab5a65 100644 --- a/drivers/net/ethernet/mellanox/mlx4/en_resources.c +++ b/drivers/net/ethernet/mellanox/mlx4/en_resources.c @@ -69,6 +69,15 @@ void mlx4_en_fill_qp_context(struct mlx4_en_priv *priv, int size, int stride, context->pri_path.counter_index = priv->counter_index; context->cqn_send = cpu_to_be32(cqn); context->cqn_recv = cpu_to_be32(cqn); + if (!rss && + (mdev->dev->caps.flags2 & MLX4_DEV_CAP_FLAG2_LB_SRC_CHK) && + context->pri_path.counter_index != + MLX4_SINK_COUNTER_INDEX(mdev->dev)) { + /* disable multicast loopback to qp with same counter */ + if (!(dev->features & NETIF_F_LOOPBACK)) + context->pri_path.fl |= MLX4_FL_ETH_SRC_CHECK_MC_LB; + context->pri_path.control |= MLX4_CTRL_ETH_SRC_CHECK_IF_COUNTER; + } context->db_rec_addr = cpu_to_be64(priv->res.db.dma << 2); if (!(dev->features & NETIF_F_HW_VLAN_CTAG_RX)) context->param3 |= cpu_to_be32(1 << 30); @@ -80,6 +89,22 @@ void mlx4_en_fill_qp_context(struct mlx4_en_priv *priv, int size, int stride, } } +int mlx4_en_change_mcast_lb(struct mlx4_en_priv *priv, struct mlx4_qp *qp, + int loopback) +{ + int ret; + struct mlx4_update_qp_params qp_params; + + memset(&qp_params, 0, sizeof(qp_params)); + if (!loopback) + qp_params.flags = MLX4_UPDATE_QP_PARAMS_FLAGS_ETH_CHECK_MC_LB; + + ret = mlx4_update_qp(priv->mdev->dev, qp->qpn, + MLX4_UPDATE_QP_ETH_SRC_CHECK_MC_LB, + &qp_params); + + return ret; +} int mlx4_en_map_buffer(struct mlx4_buf *buf) { diff --git a/drivers/net/ethernet/mellanox/mlx4/fw.c b/drivers/net/ethernet/mellanox/mlx4/fw.c index e8ec1dec5..90db94e83 100644 --- a/drivers/net/ethernet/mellanox/mlx4/fw.c +++ b/drivers/net/ethernet/mellanox/mlx4/fw.c @@ -155,6 +155,8 @@ static void dump_dev_cap_flags2(struct mlx4_dev *dev, u64 flags) [27] = "Port beacon support", [28] = "RX-ALL support", [29] = "802.1ad offload support", + [31] = "Modifying loopback source checks using UPDATE_QP support", + [32] = "Loopback source checks support", }; int i; @@ -964,6 +966,10 @@ int mlx4_QUERY_DEV_CAP(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap) MLX4_GET(field32, outbox, QUERY_DEV_CAP_EXT_2_FLAGS_OFFSET); if (field32 & (1 << 16)) dev_cap->flags2 |= MLX4_DEV_CAP_FLAG2_UPDATE_QP; + if (field32 & (1 << 18)) + dev_cap->flags2 |= MLX4_DEV_CAP_FLAG2_UPDATE_QP_SRC_CHECK_LB; + if (field32 & (1 << 19)) + dev_cap->flags2 |= MLX4_DEV_CAP_FLAG2_LB_SRC_CHK; if (field32 & (1 << 26)) dev_cap->flags2 |= MLX4_DEV_CAP_FLAG2_VLAN_CONTROL; if (field32 & (1 << 20)) @@ -2840,3 +2846,19 @@ int set_phv_bit(struct mlx4_dev *dev, u8 port, int new_val) return -EOPNOTSUPP; } EXPORT_SYMBOL(set_phv_bit); + +void mlx4_replace_zero_macs(struct mlx4_dev *dev) +{ + int i; + u8 mac_addr[ETH_ALEN]; + + dev->port_random_macs = 0; + for (i = 1; i <= dev->caps.num_ports; ++i) + if (!dev->caps.def_mac[i] && + dev->caps.port_type[i] == MLX4_PORT_TYPE_ETH) { + eth_random_addr(mac_addr); + dev->port_random_macs |= 1 << i; + dev->caps.def_mac[i] = mlx4_mac_to_u64(mac_addr); + } +} +EXPORT_SYMBOL_GPL(mlx4_replace_zero_macs); diff --git a/drivers/net/ethernet/mellanox/mlx4/main.c b/drivers/net/ethernet/mellanox/mlx4/main.c index cc3a98975..31c491e02 100644 --- a/drivers/net/ethernet/mellanox/mlx4/main.c +++ b/drivers/net/ethernet/mellanox/mlx4/main.c @@ -863,6 +863,8 @@ static int mlx4_slave_cap(struct mlx4_dev *dev) return -ENODEV; } + mlx4_replace_zero_macs(dev); + dev->caps.qp0_qkey = kcalloc(dev->caps.num_ports, sizeof(u32), GFP_KERNEL); dev->caps.qp0_tunnel = kcalloc(dev->caps.num_ports, sizeof (u32), GFP_KERNEL); dev->caps.qp0_proxy = kcalloc(dev->caps.num_ports, sizeof (u32), GFP_KERNEL); @@ -890,9 +892,10 @@ static int mlx4_slave_cap(struct mlx4_dev *dev) dev->caps.qp1_proxy[i - 1] = func_cap.qp1_proxy_qpn; dev->caps.port_mask[i] = dev->caps.port_type[i]; dev->caps.phys_port_id[i] = func_cap.phys_port_id; - if (mlx4_get_slave_pkey_gid_tbl_len(dev, i, - &dev->caps.gid_table_len[i], - &dev->caps.pkey_table_len[i])) + err = mlx4_get_slave_pkey_gid_tbl_len(dev, i, + &dev->caps.gid_table_len[i], + &dev->caps.pkey_table_len[i]); + if (err) goto err_mem; } @@ -904,6 +907,7 @@ static int mlx4_slave_cap(struct mlx4_dev *dev) dev->caps.uar_page_size * dev->caps.num_uars, (unsigned long long) pci_resource_len(dev->persist->pdev, 2)); + err = -ENOMEM; goto err_mem; } diff --git a/drivers/net/ethernet/mellanox/mlx4/mlx4.h b/drivers/net/ethernet/mellanox/mlx4/mlx4.h index 232b2b55f..e1cf9036a 100644 --- a/drivers/net/ethernet/mellanox/mlx4/mlx4.h +++ b/drivers/net/ethernet/mellanox/mlx4/mlx4.h @@ -1378,6 +1378,8 @@ void mlx4_vf_immed_vlan_work_handler(struct work_struct *_work); void mlx4_init_quotas(struct mlx4_dev *dev); +/* for VFs, replace zero MACs with randomly-generated MACs at driver start */ +void mlx4_replace_zero_macs(struct mlx4_dev *dev); int mlx4_get_slave_num_gids(struct mlx4_dev *dev, int slave, int port); /* Returns the VF index of slave */ int mlx4_get_vf_indx(struct mlx4_dev *dev, int slave); diff --git a/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h b/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h index defcf8c39..c41f15102 100644 --- a/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h +++ b/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h @@ -798,7 +798,8 @@ void mlx4_en_fill_qp_context(struct mlx4_en_priv *priv, int size, int stride, void mlx4_en_sqp_event(struct mlx4_qp *qp, enum mlx4_event event); int mlx4_en_map_buffer(struct mlx4_buf *buf); void mlx4_en_unmap_buffer(struct mlx4_buf *buf); - +int mlx4_en_change_mcast_lb(struct mlx4_en_priv *priv, struct mlx4_qp *qp, + int loopback); void mlx4_en_calc_rx_buf(struct net_device *dev); int mlx4_en_config_rss_steer(struct mlx4_en_priv *priv); void mlx4_en_release_rss_steer(struct mlx4_en_priv *priv); diff --git a/drivers/net/ethernet/mellanox/mlx4/mr.c b/drivers/net/ethernet/mellanox/mlx4/mr.c index 78f51e103..93195191f 100644 --- a/drivers/net/ethernet/mellanox/mlx4/mr.c +++ b/drivers/net/ethernet/mellanox/mlx4/mr.c @@ -318,7 +318,7 @@ int mlx4_mr_hw_get_mpt(struct mlx4_dev *dev, struct mlx4_mr *mmr, key, NULL); } else { mailbox = mlx4_alloc_cmd_mailbox(dev); - if (IS_ERR_OR_NULL(mailbox)) + if (IS_ERR(mailbox)) return PTR_ERR(mailbox); err = mlx4_cmd_box(dev, 0, mailbox->dma, key, diff --git a/drivers/net/ethernet/mellanox/mlx4/qp.c b/drivers/net/ethernet/mellanox/mlx4/qp.c index 20268634a..168823dde 100644 --- a/drivers/net/ethernet/mellanox/mlx4/qp.c +++ b/drivers/net/ethernet/mellanox/mlx4/qp.c @@ -422,20 +422,37 @@ int mlx4_update_qp(struct mlx4_dev *dev, u32 qpn, u64 qp_mask = 0; int err = 0; + if (!attr || (attr & ~MLX4_UPDATE_QP_SUPPORTED_ATTRS)) + return -EINVAL; + mailbox = mlx4_alloc_cmd_mailbox(dev); if (IS_ERR(mailbox)) return PTR_ERR(mailbox); cmd = (struct mlx4_update_qp_context *)mailbox->buf; - if (!attr || (attr & ~MLX4_UPDATE_QP_SUPPORTED_ATTRS)) - return -EINVAL; - if (attr & MLX4_UPDATE_QP_SMAC) { pri_addr_path_mask |= 1ULL << MLX4_UPD_QP_PATH_MASK_MAC_INDEX; cmd->qp_context.pri_path.grh_mylmc = params->smac_index; } + if (attr & MLX4_UPDATE_QP_ETH_SRC_CHECK_MC_LB) { + if (!(dev->caps.flags2 + & MLX4_DEV_CAP_FLAG2_UPDATE_QP_SRC_CHECK_LB)) { + mlx4_warn(dev, + "Trying to set src check LB, but it isn't supported\n"); + err = -ENOTSUPP; + goto out; + } + pri_addr_path_mask |= + 1ULL << MLX4_UPD_QP_PATH_MASK_ETH_SRC_CHECK_MC_LB; + if (params->flags & + MLX4_UPDATE_QP_PARAMS_FLAGS_ETH_CHECK_MC_LB) { + cmd->qp_context.pri_path.fl |= + MLX4_FL_ETH_SRC_CHECK_MC_LB; + } + } + if (attr & MLX4_UPDATE_QP_VSD) { qp_mask |= 1ULL << MLX4_UPD_QP_MASK_VSD; if (params->flags & MLX4_UPDATE_QP_PARAMS_FLAGS_VSD_ENABLE) @@ -458,7 +475,7 @@ int mlx4_update_qp(struct mlx4_dev *dev, u32 qpn, err = mlx4_cmd(dev, mailbox->dma, qpn & 0xffffff, 0, MLX4_CMD_UPDATE_QP, MLX4_CMD_TIME_CLASS_A, MLX4_CMD_NATIVE); - +out: mlx4_free_cmd_mailbox(dev, mailbox); return err; } diff --git a/drivers/net/ethernet/mellanox/mlx4/resource_tracker.c b/drivers/net/ethernet/mellanox/mlx4/resource_tracker.c index 8bead9737..cad6c44df 100644 --- a/drivers/net/ethernet/mellanox/mlx4/resource_tracker.c +++ b/drivers/net/ethernet/mellanox/mlx4/resource_tracker.c @@ -770,9 +770,12 @@ static int update_vport_qp_param(struct mlx4_dev *dev, } } + /* preserve IF_COUNTER flag */ + qpc->pri_path.vlan_control &= + MLX4_CTRL_ETH_SRC_CHECK_IF_COUNTER; if (vp_oper->state.link_state == IFLA_VF_LINK_STATE_DISABLE && dev->caps.flags2 & MLX4_DEV_CAP_FLAG2_UPDATE_QP) { - qpc->pri_path.vlan_control = + qpc->pri_path.vlan_control |= MLX4_VLAN_CTRL_ETH_TX_BLOCK_TAGGED | MLX4_VLAN_CTRL_ETH_TX_BLOCK_PRIO_TAGGED | MLX4_VLAN_CTRL_ETH_TX_BLOCK_UNTAGGED | @@ -780,12 +783,12 @@ static int update_vport_qp_param(struct mlx4_dev *dev, MLX4_VLAN_CTRL_ETH_RX_BLOCK_UNTAGGED | MLX4_VLAN_CTRL_ETH_RX_BLOCK_TAGGED; } else if (0 != vp_oper->state.default_vlan) { - qpc->pri_path.vlan_control = + qpc->pri_path.vlan_control |= MLX4_VLAN_CTRL_ETH_TX_BLOCK_TAGGED | MLX4_VLAN_CTRL_ETH_RX_BLOCK_PRIO_TAGGED | MLX4_VLAN_CTRL_ETH_RX_BLOCK_UNTAGGED; } else { /* priority tagged */ - qpc->pri_path.vlan_control = + qpc->pri_path.vlan_control |= MLX4_VLAN_CTRL_ETH_TX_BLOCK_TAGGED | MLX4_VLAN_CTRL_ETH_RX_BLOCK_TAGGED; } @@ -1238,8 +1241,10 @@ static int add_res_range(struct mlx4_dev *dev, int slave, u64 base, int count, return 0; undo: - for (--i; i >= base; --i) + for (--i; i >= 0; --i) { rb_erase(&res_arr[i]->node, root); + list_del_init(&res_arr[i]->list); + } spin_unlock_irq(mlx4_tlock(dev)); @@ -3762,9 +3767,6 @@ int mlx4_INIT2RTR_QP_wrapper(struct mlx4_dev *dev, int slave, update_gid(dev, inbox, (u8)slave); adjust_proxy_tun_qkey(dev, vhcr, qpc); orig_sched_queue = qpc->pri_path.sched_queue; - err = update_vport_qp_param(dev, inbox, slave, qpn); - if (err) - return err; err = get_res(dev, slave, qpn, RES_QP, &qp); if (err) @@ -3774,6 +3776,10 @@ int mlx4_INIT2RTR_QP_wrapper(struct mlx4_dev *dev, int slave, goto out; } + err = update_vport_qp_param(dev, inbox, slave, qpn); + if (err) + goto out; + err = mlx4_DMA_wrapper(dev, slave, vhcr, inbox, outbox, cmd); out: /* if no error, save sched queue value passed in by VF. This is @@ -4208,7 +4214,9 @@ static int add_eth_header(struct mlx4_dev *dev, int slave, } -#define MLX4_UPD_QP_PATH_MASK_SUPPORTED (1ULL << MLX4_UPD_QP_PATH_MASK_MAC_INDEX) +#define MLX4_UPD_QP_PATH_MASK_SUPPORTED ( \ + 1ULL << MLX4_UPD_QP_PATH_MASK_MAC_INDEX |\ + 1ULL << MLX4_UPD_QP_PATH_MASK_ETH_SRC_CHECK_MC_LB) int mlx4_UPDATE_QP_wrapper(struct mlx4_dev *dev, int slave, struct mlx4_vhcr *vhcr, struct mlx4_cmd_mailbox *inbox, @@ -4231,6 +4239,16 @@ int mlx4_UPDATE_QP_wrapper(struct mlx4_dev *dev, int slave, (pri_addr_path_mask & ~MLX4_UPD_QP_PATH_MASK_SUPPORTED)) return -EPERM; + if ((pri_addr_path_mask & + (1ULL << MLX4_UPD_QP_PATH_MASK_ETH_SRC_CHECK_MC_LB)) && + !(dev->caps.flags2 & + MLX4_DEV_CAP_FLAG2_UPDATE_QP_SRC_CHECK_LB)) { + mlx4_warn(dev, + "Src check LB for slave %d isn't supported\n", + slave); + return -ENOTSUPP; + } + /* Just change the smac for the QP */ err = get_res(dev, slave, qpn, RES_QP, &rqp); if (err) { @@ -4288,9 +4306,10 @@ int mlx4_QP_FLOW_STEERING_ATTACH_wrapper(struct mlx4_dev *dev, int slave, return -EOPNOTSUPP; ctrl = (struct mlx4_net_trans_rule_hw_ctrl *)inbox->buf; - ctrl->port = mlx4_slave_convert_port(dev, slave, ctrl->port); - if (ctrl->port <= 0) + err = mlx4_slave_convert_port(dev, slave, ctrl->port); + if (err <= 0) return -EINVAL; + ctrl->port = err; qpn = be32_to_cpu(ctrl->qpn) & 0xffffff; err = get_res(dev, slave, qpn, RES_QP, &rqp); if (err) { diff --git a/drivers/net/ethernet/mellanox/mlx5/core/cmd.c b/drivers/net/ethernet/mellanox/mlx5/core/cmd.c index 75ff58dc1..037fc4cdf 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/cmd.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/cmd.c @@ -30,7 +30,7 @@ * SOFTWARE. */ -#include <asm-generic/kmap_types.h> +#include <linux/highmem.h> #include <linux/module.h> #include <linux/errno.h> #include <linux/pci.h> @@ -254,6 +254,156 @@ static void dump_buf(void *buf, int size, int data_only, int offset) pr_debug("\n"); } +enum { + MLX5_DRIVER_STATUS_ABORTED = 0xfe, + MLX5_DRIVER_SYND = 0xbadd00de, +}; + +static int mlx5_internal_err_ret_value(struct mlx5_core_dev *dev, u16 op, + u32 *synd, u8 *status) +{ + *synd = 0; + *status = 0; + + switch (op) { + case MLX5_CMD_OP_TEARDOWN_HCA: + case MLX5_CMD_OP_DISABLE_HCA: + case MLX5_CMD_OP_MANAGE_PAGES: + case MLX5_CMD_OP_DESTROY_MKEY: + case MLX5_CMD_OP_DESTROY_EQ: + case MLX5_CMD_OP_DESTROY_CQ: + case MLX5_CMD_OP_DESTROY_QP: + case MLX5_CMD_OP_DESTROY_PSV: + case MLX5_CMD_OP_DESTROY_SRQ: + case MLX5_CMD_OP_DESTROY_XRC_SRQ: + case MLX5_CMD_OP_DESTROY_DCT: + case MLX5_CMD_OP_DEALLOC_Q_COUNTER: + case MLX5_CMD_OP_DEALLOC_PD: + case MLX5_CMD_OP_DEALLOC_UAR: + case MLX5_CMD_OP_DETTACH_FROM_MCG: + case MLX5_CMD_OP_DEALLOC_XRCD: + case MLX5_CMD_OP_DEALLOC_TRANSPORT_DOMAIN: + case MLX5_CMD_OP_DELETE_VXLAN_UDP_DPORT: + case MLX5_CMD_OP_DELETE_L2_TABLE_ENTRY: + case MLX5_CMD_OP_DESTROY_TIR: + case MLX5_CMD_OP_DESTROY_SQ: + case MLX5_CMD_OP_DESTROY_RQ: + case MLX5_CMD_OP_DESTROY_RMP: + case MLX5_CMD_OP_DESTROY_TIS: + case MLX5_CMD_OP_DESTROY_RQT: + case MLX5_CMD_OP_DESTROY_FLOW_TABLE: + case MLX5_CMD_OP_DESTROY_FLOW_GROUP: + case MLX5_CMD_OP_DELETE_FLOW_TABLE_ENTRY: + return MLX5_CMD_STAT_OK; + + case MLX5_CMD_OP_QUERY_HCA_CAP: + case MLX5_CMD_OP_QUERY_ADAPTER: + case MLX5_CMD_OP_INIT_HCA: + case MLX5_CMD_OP_ENABLE_HCA: + case MLX5_CMD_OP_QUERY_PAGES: + case MLX5_CMD_OP_SET_HCA_CAP: + case MLX5_CMD_OP_QUERY_ISSI: + case MLX5_CMD_OP_SET_ISSI: + case MLX5_CMD_OP_CREATE_MKEY: + case MLX5_CMD_OP_QUERY_MKEY: + case MLX5_CMD_OP_QUERY_SPECIAL_CONTEXTS: + case MLX5_CMD_OP_PAGE_FAULT_RESUME: + case MLX5_CMD_OP_CREATE_EQ: + case MLX5_CMD_OP_QUERY_EQ: + case MLX5_CMD_OP_GEN_EQE: + case MLX5_CMD_OP_CREATE_CQ: + case MLX5_CMD_OP_QUERY_CQ: + case MLX5_CMD_OP_MODIFY_CQ: + case MLX5_CMD_OP_CREATE_QP: + case MLX5_CMD_OP_RST2INIT_QP: + case MLX5_CMD_OP_INIT2RTR_QP: + case MLX5_CMD_OP_RTR2RTS_QP: + case MLX5_CMD_OP_RTS2RTS_QP: + case MLX5_CMD_OP_SQERR2RTS_QP: + case MLX5_CMD_OP_2ERR_QP: + case MLX5_CMD_OP_2RST_QP: + case MLX5_CMD_OP_QUERY_QP: + case MLX5_CMD_OP_SQD_RTS_QP: + case MLX5_CMD_OP_INIT2INIT_QP: + case MLX5_CMD_OP_CREATE_PSV: + case MLX5_CMD_OP_CREATE_SRQ: + case MLX5_CMD_OP_QUERY_SRQ: + case MLX5_CMD_OP_ARM_RQ: + case MLX5_CMD_OP_CREATE_XRC_SRQ: + case MLX5_CMD_OP_QUERY_XRC_SRQ: + case MLX5_CMD_OP_ARM_XRC_SRQ: + case MLX5_CMD_OP_CREATE_DCT: + case MLX5_CMD_OP_DRAIN_DCT: + case MLX5_CMD_OP_QUERY_DCT: + case MLX5_CMD_OP_ARM_DCT_FOR_KEY_VIOLATION: + case MLX5_CMD_OP_QUERY_VPORT_STATE: + case MLX5_CMD_OP_MODIFY_VPORT_STATE: + case MLX5_CMD_OP_QUERY_ESW_VPORT_CONTEXT: + case MLX5_CMD_OP_MODIFY_ESW_VPORT_CONTEXT: + case MLX5_CMD_OP_QUERY_NIC_VPORT_CONTEXT: + case MLX5_CMD_OP_MODIFY_NIC_VPORT_CONTEXT: + case MLX5_CMD_OP_QUERY_ROCE_ADDRESS: + case MLX5_CMD_OP_SET_ROCE_ADDRESS: + case MLX5_CMD_OP_QUERY_HCA_VPORT_CONTEXT: + case MLX5_CMD_OP_MODIFY_HCA_VPORT_CONTEXT: + case MLX5_CMD_OP_QUERY_HCA_VPORT_GID: + case MLX5_CMD_OP_QUERY_HCA_VPORT_PKEY: + case MLX5_CMD_OP_QUERY_VPORT_COUNTER: + case MLX5_CMD_OP_ALLOC_Q_COUNTER: + case MLX5_CMD_OP_QUERY_Q_COUNTER: + case MLX5_CMD_OP_ALLOC_PD: + case MLX5_CMD_OP_ALLOC_UAR: + case MLX5_CMD_OP_CONFIG_INT_MODERATION: + case MLX5_CMD_OP_ACCESS_REG: + case MLX5_CMD_OP_ATTACH_TO_MCG: + case MLX5_CMD_OP_GET_DROPPED_PACKET_LOG: + case MLX5_CMD_OP_MAD_IFC: + case MLX5_CMD_OP_QUERY_MAD_DEMUX: + case MLX5_CMD_OP_SET_MAD_DEMUX: + case MLX5_CMD_OP_NOP: + case MLX5_CMD_OP_ALLOC_XRCD: + case MLX5_CMD_OP_ALLOC_TRANSPORT_DOMAIN: + case MLX5_CMD_OP_QUERY_CONG_STATUS: + case MLX5_CMD_OP_MODIFY_CONG_STATUS: + case MLX5_CMD_OP_QUERY_CONG_PARAMS: + case MLX5_CMD_OP_MODIFY_CONG_PARAMS: + case MLX5_CMD_OP_QUERY_CONG_STATISTICS: + case MLX5_CMD_OP_ADD_VXLAN_UDP_DPORT: + case MLX5_CMD_OP_SET_L2_TABLE_ENTRY: + case MLX5_CMD_OP_QUERY_L2_TABLE_ENTRY: + case MLX5_CMD_OP_CREATE_TIR: + case MLX5_CMD_OP_MODIFY_TIR: + case MLX5_CMD_OP_QUERY_TIR: + case MLX5_CMD_OP_CREATE_SQ: + case MLX5_CMD_OP_MODIFY_SQ: + case MLX5_CMD_OP_QUERY_SQ: + case MLX5_CMD_OP_CREATE_RQ: + case MLX5_CMD_OP_MODIFY_RQ: + case MLX5_CMD_OP_QUERY_RQ: + case MLX5_CMD_OP_CREATE_RMP: + case MLX5_CMD_OP_MODIFY_RMP: + case MLX5_CMD_OP_QUERY_RMP: + case MLX5_CMD_OP_CREATE_TIS: + case MLX5_CMD_OP_MODIFY_TIS: + case MLX5_CMD_OP_QUERY_TIS: + case MLX5_CMD_OP_CREATE_RQT: + case MLX5_CMD_OP_MODIFY_RQT: + case MLX5_CMD_OP_QUERY_RQT: + case MLX5_CMD_OP_CREATE_FLOW_TABLE: + case MLX5_CMD_OP_QUERY_FLOW_TABLE: + case MLX5_CMD_OP_CREATE_FLOW_GROUP: + case MLX5_CMD_OP_QUERY_FLOW_GROUP: + case MLX5_CMD_OP_SET_FLOW_TABLE_ENTRY: + case MLX5_CMD_OP_QUERY_FLOW_TABLE_ENTRY: + *status = MLX5_DRIVER_STATUS_ABORTED; + *synd = MLX5_DRIVER_SYND; + return -EIO; + default: + mlx5_core_err(dev, "Unknown FW command (%d)\n", op); + return -EINVAL; + } +} + const char *mlx5_command_str(int command) { switch (command) { @@ -473,6 +623,7 @@ static void cmd_work_handler(struct work_struct *work) struct mlx5_core_dev *dev = container_of(cmd, struct mlx5_core_dev, cmd); struct mlx5_cmd_layout *lay; struct semaphore *sem; + unsigned long flags; sem = ent->page_queue ? &cmd->pages_sem : &cmd->sem; down(sem); @@ -485,6 +636,9 @@ static void cmd_work_handler(struct work_struct *work) } } else { ent->idx = cmd->max_reg_cmds; + spin_lock_irqsave(&cmd->alloc_lock, flags); + clear_bit(ent->idx, &cmd->bitmask); + spin_unlock_irqrestore(&cmd->alloc_lock, flags); } ent->token = alloc_token(cmd); @@ -584,6 +738,16 @@ static int wait_func(struct mlx5_core_dev *dev, struct mlx5_cmd_work_ent *ent) return err; } +static __be32 *get_synd_ptr(struct mlx5_outbox_hdr *out) +{ + return &out->syndrome; +} + +static u8 *get_status_ptr(struct mlx5_outbox_hdr *out) +{ + return &out->status; +} + /* Notes: * 1. Callback functions may not sleep * 2. page queue commands do not support asynchrous completion @@ -1081,7 +1245,7 @@ static void free_msg(struct mlx5_core_dev *dev, struct mlx5_cmd_msg *msg) } } -void mlx5_cmd_comp_handler(struct mlx5_core_dev *dev, unsigned long vector) +void mlx5_cmd_comp_handler(struct mlx5_core_dev *dev, u64 vec) { struct mlx5_cmd *cmd = &dev->cmd; struct mlx5_cmd_work_ent *ent; @@ -1092,7 +1256,10 @@ void mlx5_cmd_comp_handler(struct mlx5_core_dev *dev, unsigned long vector) s64 ds; struct mlx5_cmd_stats *stats; unsigned long flags; + unsigned long vector; + /* there can be at most 32 command queues */ + vector = vec & 0xffffffff; for (i = 0; i < (1 << cmd->log_sz); i++) { if (test_bit(i, &vector)) { struct semaphore *sem; @@ -1110,11 +1277,16 @@ void mlx5_cmd_comp_handler(struct mlx5_core_dev *dev, unsigned long vector) ent->ret = verify_signature(ent); else ent->ret = 0; - ent->status = ent->lay->status_own >> 1; + if (vec & MLX5_TRIGGERED_CMD_COMP) + ent->status = MLX5_DRIVER_STATUS_ABORTED; + else + ent->status = ent->lay->status_own >> 1; + mlx5_core_dbg(dev, "command completed. ret 0x%x, delivery status %s(0x%x)\n", ent->ret, deliv_status_to_str(ent->status), ent->status); } free_ent(cmd, ent->idx); + if (ent->callback) { ds = ent->ts2 - ent->ts1; if (ent->op < ARRAY_SIZE(cmd->stats)) { @@ -1136,6 +1308,7 @@ void mlx5_cmd_comp_handler(struct mlx5_core_dev *dev, unsigned long vector) mlx5_free_cmd_msg(dev, ent->out); free_msg(dev, ent->in); + err = err ? err : ent->status; free_cmd(ent); callback(err, context); } else { @@ -1183,6 +1356,11 @@ static struct mlx5_cmd_msg *alloc_msg(struct mlx5_core_dev *dev, int in_size, return msg; } +static u16 opcode_from_in(struct mlx5_inbox_hdr *in) +{ + return be16_to_cpu(in->opcode); +} + static int is_manage_pages(struct mlx5_inbox_hdr *in) { return be16_to_cpu(in->opcode) == MLX5_CMD_OP_MANAGE_PAGES; @@ -1197,6 +1375,15 @@ static int cmd_exec(struct mlx5_core_dev *dev, void *in, int in_size, void *out, gfp_t gfp; int err; u8 status = 0; + u32 drv_synd; + + if (pci_channel_offline(dev->pdev) || + dev->state == MLX5_DEVICE_STATE_INTERNAL_ERROR) { + err = mlx5_internal_err_ret_value(dev, opcode_from_in(in), &drv_synd, &status); + *get_synd_ptr(out) = cpu_to_be32(drv_synd); + *get_status_ptr(out) = status; + return err; + } pages_queue = is_manage_pages(in); gfp = callback ? GFP_ATOMIC : GFP_KERNEL; @@ -1363,6 +1550,7 @@ int mlx5_cmd_init(struct mlx5_core_dev *dev) int err; int i; + memset(cmd, 0, sizeof(*cmd)); cmd_if_rev = cmdif_rev(dev); if (cmd_if_rev != CMD_IF_REV) { dev_err(&dev->pdev->dev, diff --git a/drivers/net/ethernet/mellanox/mlx5/core/cq.c b/drivers/net/ethernet/mellanox/mlx5/core/cq.c index 04ab7e445..b51e42d6f 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/cq.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/cq.c @@ -242,6 +242,7 @@ int mlx5_init_cq_table(struct mlx5_core_dev *dev) struct mlx5_cq_table *table = &dev->priv.cq_table; int err; + memset(table, 0, sizeof(*table)); spin_lock_init(&table->lock); INIT_RADIX_TREE(&table->tree, GFP_ATOMIC); err = mlx5_cq_debugfs_init(dev); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h index 0983a208b..22e72bf1a 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h @@ -334,9 +334,15 @@ struct mlx5e_tx_skb_cb { #define MLX5E_TX_SKB_CB(__skb) ((struct mlx5e_tx_skb_cb *)__skb->cb) +enum mlx5e_dma_map_type { + MLX5E_DMA_MAP_SINGLE, + MLX5E_DMA_MAP_PAGE +}; + struct mlx5e_sq_dma { - dma_addr_t addr; - u32 size; + dma_addr_t addr; + u32 size; + enum mlx5e_dma_map_type type; }; enum { @@ -617,5 +623,11 @@ static inline void mlx5e_cq_arm(struct mlx5e_cq *cq) mlx5_cq_arm(mcq, MLX5_CQ_DB_REQ_NOT, mcq->uar->map, NULL, cq->wq.cc); } +static inline int mlx5e_get_max_num_channels(struct mlx5_core_dev *mdev) +{ + return min_t(int, mdev->priv.eq_table.num_comp_vectors, + MLX5E_MAX_NUM_CHANNELS); +} + extern const struct ethtool_ops mlx5e_ethtool_ops; u16 mlx5e_get_max_inline_cap(struct mlx5_core_dev *mdev); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c index bce912688..2e022e900 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c @@ -345,9 +345,8 @@ static void mlx5e_get_channels(struct net_device *dev, struct ethtool_channels *ch) { struct mlx5e_priv *priv = netdev_priv(dev); - int ncv = priv->mdev->priv.eq_table.num_comp_vectors; - ch->max_combined = ncv; + ch->max_combined = mlx5e_get_max_num_channels(priv->mdev); ch->combined_count = priv->params.num_channels; } @@ -355,7 +354,7 @@ static int mlx5e_set_channels(struct net_device *dev, struct ethtool_channels *ch) { struct mlx5e_priv *priv = netdev_priv(dev); - int ncv = priv->mdev->priv.eq_table.num_comp_vectors; + int ncv = mlx5e_get_max_num_channels(priv->mdev); unsigned int count = ch->combined_count; bool was_opened; int err = 0; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c index 443632df2..1e52db32c 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c @@ -442,12 +442,12 @@ static void mlx5e_disable_rq(struct mlx5e_rq *rq) static int mlx5e_wait_for_min_rx_wqes(struct mlx5e_rq *rq) { + unsigned long exp_time = jiffies + msecs_to_jiffies(20000); struct mlx5e_channel *c = rq->channel; struct mlx5e_priv *priv = c->priv; struct mlx5_wq_ll *wq = &rq->wq; - int i; - for (i = 0; i < 1000; i++) { + while (time_before(jiffies, exp_time)) { if (wq->cur_sz >= priv->params.min_rx_wqes) return 0; @@ -1449,6 +1449,12 @@ int mlx5e_close_locked(struct net_device *netdev) { struct mlx5e_priv *priv = netdev_priv(netdev); + /* May already be CLOSED in case a previous configuration operation + * (e.g RX/TX queue size change) that involves close&open failed. + */ + if (!test_bit(MLX5E_STATE_OPENED, &priv->state)) + return 0; + clear_bit(MLX5E_STATE_OPENED, &priv->state); mlx5e_redirect_rqts(priv); @@ -1882,7 +1888,7 @@ static int mlx5e_set_features(struct net_device *netdev, mlx5e_disable_vlan_filter(priv); } - return 0; + return err; } static int mlx5e_change_mtu(struct net_device *netdev, int new_mtu) @@ -1895,6 +1901,8 @@ static int mlx5e_change_mtu(struct net_device *netdev, int new_mtu) mlx5_query_port_max_mtu(mdev, &max_mtu, 1); + max_mtu = MLX5E_HW2SW_MTU(max_mtu); + if (new_mtu > max_mtu) { netdev_err(netdev, "%s: Bad MTU (%d) > (%d) Max\n", @@ -2046,6 +2054,7 @@ static void mlx5e_build_netdev(struct net_device *netdev) netdev->vlan_features |= NETIF_F_LRO; netdev->hw_features = netdev->vlan_features; + netdev->hw_features |= NETIF_F_HW_VLAN_CTAG_TX; netdev->hw_features |= NETIF_F_HW_VLAN_CTAG_RX; netdev->hw_features |= NETIF_F_HW_VLAN_CTAG_FILTER; @@ -2089,8 +2098,7 @@ static void *mlx5e_create_netdev(struct mlx5_core_dev *mdev) { struct net_device *netdev; struct mlx5e_priv *priv; - int nch = min_t(int, mdev->priv.eq_table.num_comp_vectors, - MLX5E_MAX_NUM_CHANNELS); + int nch = mlx5e_get_max_num_channels(mdev); int err; if (mlx5e_check_required_hca_cap(mdev)) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c index b73672f32..1341b1d3c 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c @@ -61,39 +61,47 @@ void mlx5e_send_nop(struct mlx5e_sq *sq, bool notify_hw) } } -static void mlx5e_dma_pop_last_pushed(struct mlx5e_sq *sq, dma_addr_t *addr, - u32 *size) +static inline void mlx5e_tx_dma_unmap(struct device *pdev, + struct mlx5e_sq_dma *dma) { - sq->dma_fifo_pc--; - *addr = sq->dma_fifo[sq->dma_fifo_pc & sq->dma_fifo_mask].addr; - *size = sq->dma_fifo[sq->dma_fifo_pc & sq->dma_fifo_mask].size; -} - -static void mlx5e_dma_unmap_wqe_err(struct mlx5e_sq *sq, struct sk_buff *skb) -{ - dma_addr_t addr; - u32 size; - int i; - - for (i = 0; i < MLX5E_TX_SKB_CB(skb)->num_dma; i++) { - mlx5e_dma_pop_last_pushed(sq, &addr, &size); - dma_unmap_single(sq->pdev, addr, size, DMA_TO_DEVICE); + switch (dma->type) { + case MLX5E_DMA_MAP_SINGLE: + dma_unmap_single(pdev, dma->addr, dma->size, DMA_TO_DEVICE); + break; + case MLX5E_DMA_MAP_PAGE: + dma_unmap_page(pdev, dma->addr, dma->size, DMA_TO_DEVICE); + break; + default: + WARN_ONCE(true, "mlx5e_tx_dma_unmap unknown DMA type!\n"); } } -static inline void mlx5e_dma_push(struct mlx5e_sq *sq, dma_addr_t addr, - u32 size) +static inline void mlx5e_dma_push(struct mlx5e_sq *sq, + dma_addr_t addr, + u32 size, + enum mlx5e_dma_map_type map_type) { sq->dma_fifo[sq->dma_fifo_pc & sq->dma_fifo_mask].addr = addr; sq->dma_fifo[sq->dma_fifo_pc & sq->dma_fifo_mask].size = size; + sq->dma_fifo[sq->dma_fifo_pc & sq->dma_fifo_mask].type = map_type; sq->dma_fifo_pc++; } -static inline void mlx5e_dma_get(struct mlx5e_sq *sq, u32 i, dma_addr_t *addr, - u32 *size) +static inline struct mlx5e_sq_dma *mlx5e_dma_get(struct mlx5e_sq *sq, u32 i) +{ + return &sq->dma_fifo[i & sq->dma_fifo_mask]; +} + +static void mlx5e_dma_unmap_wqe_err(struct mlx5e_sq *sq, struct sk_buff *skb) { - *addr = sq->dma_fifo[i & sq->dma_fifo_mask].addr; - *size = sq->dma_fifo[i & sq->dma_fifo_mask].size; + int i; + + for (i = 0; i < MLX5E_TX_SKB_CB(skb)->num_dma; i++) { + struct mlx5e_sq_dma *last_pushed_dma = + mlx5e_dma_get(sq, --sq->dma_fifo_pc); + + mlx5e_tx_dma_unmap(sq->pdev, last_pushed_dma); + } } u16 mlx5e_select_queue(struct net_device *dev, struct sk_buff *skb, @@ -116,14 +124,36 @@ static inline u16 mlx5e_get_inline_hdr_size(struct mlx5e_sq *sq, * headers and occur before the data gather. * Therefore these headers must be copied into the WQE */ -#define MLX5E_MIN_INLINE (ETH_HLEN + 2/*vlan tag*/) +#define MLX5E_MIN_INLINE ETH_HLEN + + if (bf) { + u16 ihs = skb_headlen(skb); + + if (skb_vlan_tag_present(skb)) + ihs += VLAN_HLEN; - if (bf && (skb_headlen(skb) <= sq->max_inline)) - return skb_headlen(skb); + if (ihs <= sq->max_inline) + return skb_headlen(skb); + } return MLX5E_MIN_INLINE; } +static inline void mlx5e_insert_vlan(void *start, struct sk_buff *skb, u16 ihs) +{ + struct vlan_ethhdr *vhdr = (struct vlan_ethhdr *)start; + int cpy1_sz = 2 * ETH_ALEN; + int cpy2_sz = ihs - cpy1_sz; + + skb_copy_from_linear_data(skb, vhdr, cpy1_sz); + skb_pull_inline(skb, cpy1_sz); + vhdr->h_vlan_proto = skb->vlan_proto; + vhdr->h_vlan_TCI = cpu_to_be16(skb_vlan_tag_get(skb)); + skb_copy_from_linear_data(skb, &vhdr->h_vlan_encapsulated_proto, + cpy2_sz); + skb_pull_inline(skb, cpy2_sz); +} + static netdev_tx_t mlx5e_sq_xmit(struct mlx5e_sq *sq, struct sk_buff *skb) { struct mlx5_wq_cyc *wq = &sq->wq; @@ -175,8 +205,13 @@ static netdev_tx_t mlx5e_sq_xmit(struct mlx5e_sq *sq, struct sk_buff *skb) ETH_ZLEN); } - skb_copy_from_linear_data(skb, eseg->inline_hdr_start, ihs); - skb_pull_inline(skb, ihs); + if (skb_vlan_tag_present(skb)) { + mlx5e_insert_vlan(eseg->inline_hdr_start, skb, ihs); + ihs += VLAN_HLEN; + } else { + skb_copy_from_linear_data(skb, eseg->inline_hdr_start, ihs); + skb_pull_inline(skb, ihs); + } eseg->inline_hdr_sz = cpu_to_be16(ihs); @@ -198,7 +233,7 @@ static netdev_tx_t mlx5e_sq_xmit(struct mlx5e_sq *sq, struct sk_buff *skb) dseg->lkey = sq->mkey_be; dseg->byte_count = cpu_to_be32(headlen); - mlx5e_dma_push(sq, dma_addr, headlen); + mlx5e_dma_push(sq, dma_addr, headlen, MLX5E_DMA_MAP_SINGLE); MLX5E_TX_SKB_CB(skb)->num_dma++; dseg++; @@ -217,7 +252,7 @@ static netdev_tx_t mlx5e_sq_xmit(struct mlx5e_sq *sq, struct sk_buff *skb) dseg->lkey = sq->mkey_be; dseg->byte_count = cpu_to_be32(fsz); - mlx5e_dma_push(sq, dma_addr, fsz); + mlx5e_dma_push(sq, dma_addr, fsz, MLX5E_DMA_MAP_PAGE); MLX5E_TX_SKB_CB(skb)->num_dma++; dseg++; @@ -333,13 +368,10 @@ bool mlx5e_poll_tx_cq(struct mlx5e_cq *cq) } for (j = 0; j < MLX5E_TX_SKB_CB(skb)->num_dma; j++) { - dma_addr_t addr; - u32 size; + struct mlx5e_sq_dma *dma = + mlx5e_dma_get(sq, dma_fifo_cc++); - mlx5e_dma_get(sq, dma_fifo_cc, &addr, &size); - dma_fifo_cc++; - dma_unmap_single(sq->pdev, addr, size, - DMA_TO_DEVICE); + mlx5e_tx_dma_unmap(sq->pdev, dma); } npkts++; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eq.c b/drivers/net/ethernet/mellanox/mlx5/core/eq.c index a40b96d4c..713ead583 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eq.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eq.c @@ -346,6 +346,7 @@ int mlx5_create_map_eq(struct mlx5_core_dev *dev, struct mlx5_eq *eq, u8 vecidx, int inlen; eq->nent = roundup_pow_of_two(nent + MLX5_NUM_SPARE_EQE); + eq->cons_index = 0; err = mlx5_buf_alloc(dev, eq->nent * MLX5_EQE_SIZE, &eq->buf); if (err) return err; @@ -381,10 +382,10 @@ int mlx5_create_map_eq(struct mlx5_core_dev *dev, struct mlx5_eq *eq, u8 vecidx, name, pci_name(dev->pdev)); eq->eqn = out.eq_number; - eq->irqn = vecidx; + eq->irqn = priv->msix_arr[vecidx].vector; eq->dev = dev; eq->doorbell = uar->map + MLX5_EQ_DOORBEL_OFFSET; - err = request_irq(priv->msix_arr[vecidx].vector, mlx5_msix_handler, 0, + err = request_irq(eq->irqn, mlx5_msix_handler, 0, priv->irq_info[vecidx].name, eq); if (err) goto err_eq; @@ -420,12 +421,12 @@ int mlx5_destroy_unmap_eq(struct mlx5_core_dev *dev, struct mlx5_eq *eq) int err; mlx5_debug_eq_remove(dev, eq); - free_irq(dev->priv.msix_arr[eq->irqn].vector, eq); + free_irq(eq->irqn, eq); err = mlx5_cmd_destroy_eq(dev, eq->eqn); if (err) mlx5_core_warn(dev, "failed to destroy a previously created eq: eqn %d\n", eq->eqn); - synchronize_irq(dev->priv.msix_arr[eq->irqn].vector); + synchronize_irq(eq->irqn); mlx5_buf_free(dev, &eq->buf); return err; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/health.c b/drivers/net/ethernet/mellanox/mlx5/core/health.c index 292d76f2a..f5deb642d 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/health.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/health.c @@ -34,6 +34,7 @@ #include <linux/module.h> #include <linux/random.h> #include <linux/vmalloc.h> +#include <linux/hardirq.h> #include <linux/mlx5/driver.h> #include <linux/mlx5/cmd.h> #include "mlx5_core.h" @@ -46,39 +47,113 @@ enum { enum { MLX5_HEALTH_SYNDR_FW_ERR = 0x1, MLX5_HEALTH_SYNDR_IRISC_ERR = 0x7, + MLX5_HEALTH_SYNDR_HW_UNRECOVERABLE_ERR = 0x8, MLX5_HEALTH_SYNDR_CRC_ERR = 0x9, MLX5_HEALTH_SYNDR_FETCH_PCI_ERR = 0xa, MLX5_HEALTH_SYNDR_HW_FTL_ERR = 0xb, MLX5_HEALTH_SYNDR_ASYNC_EQ_OVERRUN_ERR = 0xc, MLX5_HEALTH_SYNDR_EQ_ERR = 0xd, + MLX5_HEALTH_SYNDR_EQ_INV = 0xe, MLX5_HEALTH_SYNDR_FFSER_ERR = 0xf, + MLX5_HEALTH_SYNDR_HIGH_TEMP = 0x10 }; -static DEFINE_SPINLOCK(health_lock); -static LIST_HEAD(health_list); -static struct work_struct health_work; +enum { + MLX5_NIC_IFC_FULL = 0, + MLX5_NIC_IFC_DISABLED = 1, + MLX5_NIC_IFC_NO_DRAM_NIC = 2 +}; -static void health_care(struct work_struct *work) +static u8 get_nic_interface(struct mlx5_core_dev *dev) { - struct mlx5_core_health *health, *n; - struct mlx5_core_dev *dev; - struct mlx5_priv *priv; - LIST_HEAD(tlist); + return (ioread32be(&dev->iseg->cmdq_addr_l_sz) >> 8) & 3; +} + +static void trigger_cmd_completions(struct mlx5_core_dev *dev) +{ + unsigned long flags; + u64 vector; - spin_lock_irq(&health_lock); - list_splice_init(&health_list, &tlist); + /* wait for pending handlers to complete */ + synchronize_irq(dev->priv.msix_arr[MLX5_EQ_VEC_CMD].vector); + spin_lock_irqsave(&dev->cmd.alloc_lock, flags); + vector = ~dev->cmd.bitmask & ((1ul << (1 << dev->cmd.log_sz)) - 1); + if (!vector) + goto no_trig; + + vector |= MLX5_TRIGGERED_CMD_COMP; + spin_unlock_irqrestore(&dev->cmd.alloc_lock, flags); + + mlx5_core_dbg(dev, "vector 0x%llx\n", vector); + mlx5_cmd_comp_handler(dev, vector); + return; + +no_trig: + spin_unlock_irqrestore(&dev->cmd.alloc_lock, flags); +} + +static int in_fatal(struct mlx5_core_dev *dev) +{ + struct mlx5_core_health *health = &dev->priv.health; + struct health_buffer __iomem *h = health->health; - spin_unlock_irq(&health_lock); + if (get_nic_interface(dev) == MLX5_NIC_IFC_DISABLED) + return 1; - list_for_each_entry_safe(health, n, &tlist, list) { - priv = container_of(health, struct mlx5_priv, health); - dev = container_of(priv, struct mlx5_core_dev, priv); - mlx5_core_warn(dev, "handling bad device here\n"); - /* nothing yet */ - spin_lock_irq(&health_lock); - list_del_init(&health->list); - spin_unlock_irq(&health_lock); + if (ioread32be(&h->fw_ver) == 0xffffffff) + return 1; + + return 0; +} + +void mlx5_enter_error_state(struct mlx5_core_dev *dev) +{ + if (dev->state == MLX5_DEVICE_STATE_INTERNAL_ERROR) + return; + + mlx5_core_err(dev, "start\n"); + if (pci_channel_offline(dev->pdev) || in_fatal(dev)) + dev->state = MLX5_DEVICE_STATE_INTERNAL_ERROR; + + mlx5_core_event(dev, MLX5_DEV_EVENT_SYS_ERROR, 0); + mlx5_core_err(dev, "end\n"); +} + +static void mlx5_handle_bad_state(struct mlx5_core_dev *dev) +{ + u8 nic_interface = get_nic_interface(dev); + + switch (nic_interface) { + case MLX5_NIC_IFC_FULL: + mlx5_core_warn(dev, "Expected to see disabled NIC but it is full driver\n"); + break; + + case MLX5_NIC_IFC_DISABLED: + mlx5_core_warn(dev, "starting teardown\n"); + break; + + case MLX5_NIC_IFC_NO_DRAM_NIC: + mlx5_core_warn(dev, "Expected to see disabled NIC but it is no dram nic\n"); + break; + default: + mlx5_core_warn(dev, "Expected to see disabled NIC but it is has invalid value %d\n", + nic_interface); } + + mlx5_disable_device(dev); +} + +static void health_care(struct work_struct *work) +{ + struct mlx5_core_health *health; + struct mlx5_core_dev *dev; + struct mlx5_priv *priv; + + health = container_of(work, struct mlx5_core_health, work); + priv = container_of(health, struct mlx5_priv, health); + dev = container_of(priv, struct mlx5_core_dev, priv); + mlx5_core_warn(dev, "handling bad device here\n"); + mlx5_handle_bad_state(dev); } static const char *hsynd_str(u8 synd) @@ -88,6 +163,8 @@ static const char *hsynd_str(u8 synd) return "firmware internal error"; case MLX5_HEALTH_SYNDR_IRISC_ERR: return "irisc not responding"; + case MLX5_HEALTH_SYNDR_HW_UNRECOVERABLE_ERR: + return "unrecoverable hardware error"; case MLX5_HEALTH_SYNDR_CRC_ERR: return "firmware CRC error"; case MLX5_HEALTH_SYNDR_FETCH_PCI_ERR: @@ -98,48 +175,81 @@ static const char *hsynd_str(u8 synd) return "async EQ buffer overrun"; case MLX5_HEALTH_SYNDR_EQ_ERR: return "EQ error"; + case MLX5_HEALTH_SYNDR_EQ_INV: + return "Invalid EQ refrenced"; case MLX5_HEALTH_SYNDR_FFSER_ERR: return "FFSER error"; + case MLX5_HEALTH_SYNDR_HIGH_TEMP: + return "High temprature"; default: return "unrecognized error"; } } -static u16 read_be16(__be16 __iomem *p) +static u16 get_maj(u32 fw) { - return swab16(readl((__force u16 __iomem *) p)); + return fw >> 28; } -static u32 read_be32(__be32 __iomem *p) +static u16 get_min(u32 fw) { - return swab32(readl((__force u32 __iomem *) p)); + return fw >> 16 & 0xfff; +} + +static u16 get_sub(u32 fw) +{ + return fw & 0xffff; } static void print_health_info(struct mlx5_core_dev *dev) { struct mlx5_core_health *health = &dev->priv.health; struct health_buffer __iomem *h = health->health; + char fw_str[18]; + u32 fw; int i; + /* If the syndrom is 0, the device is OK and no need to print buffer */ + if (!ioread8(&h->synd)) + return; + for (i = 0; i < ARRAY_SIZE(h->assert_var); i++) - pr_info("assert_var[%d] 0x%08x\n", i, read_be32(h->assert_var + i)); + dev_err(&dev->pdev->dev, "assert_var[%d] 0x%08x\n", i, ioread32be(h->assert_var + i)); + + dev_err(&dev->pdev->dev, "assert_exit_ptr 0x%08x\n", ioread32be(&h->assert_exit_ptr)); + dev_err(&dev->pdev->dev, "assert_callra 0x%08x\n", ioread32be(&h->assert_callra)); + fw = ioread32be(&h->fw_ver); + sprintf(fw_str, "%d.%d.%d", get_maj(fw), get_min(fw), get_sub(fw)); + dev_err(&dev->pdev->dev, "fw_ver %s\n", fw_str); + dev_err(&dev->pdev->dev, "hw_id 0x%08x\n", ioread32be(&h->hw_id)); + dev_err(&dev->pdev->dev, "irisc_index %d\n", ioread8(&h->irisc_index)); + dev_err(&dev->pdev->dev, "synd 0x%x: %s\n", ioread8(&h->synd), hsynd_str(ioread8(&h->synd))); + dev_err(&dev->pdev->dev, "ext_synd 0x%04x\n", ioread16be(&h->ext_synd)); +} + +static unsigned long get_next_poll_jiffies(void) +{ + unsigned long next; - pr_info("assert_exit_ptr 0x%08x\n", read_be32(&h->assert_exit_ptr)); - pr_info("assert_callra 0x%08x\n", read_be32(&h->assert_callra)); - pr_info("fw_ver 0x%08x\n", read_be32(&h->fw_ver)); - pr_info("hw_id 0x%08x\n", read_be32(&h->hw_id)); - pr_info("irisc_index %d\n", readb(&h->irisc_index)); - pr_info("synd 0x%x: %s\n", readb(&h->synd), hsynd_str(readb(&h->synd))); - pr_info("ext_sync 0x%04x\n", read_be16(&h->ext_sync)); + get_random_bytes(&next, sizeof(next)); + next %= HZ; + next += jiffies + MLX5_HEALTH_POLL_INTERVAL; + + return next; } static void poll_health(unsigned long data) { struct mlx5_core_dev *dev = (struct mlx5_core_dev *)data; struct mlx5_core_health *health = &dev->priv.health; - unsigned long next; u32 count; + if (dev->state == MLX5_DEVICE_STATE_INTERNAL_ERROR) { + trigger_cmd_completions(dev); + mod_timer(&health->timer, get_next_poll_jiffies()); + return; + } + count = ioread32be(health->health_counter); if (count == health->prev) ++health->miss_counter; @@ -148,18 +258,16 @@ static void poll_health(unsigned long data) health->prev = count; if (health->miss_counter == MAX_MISSES) { - mlx5_core_err(dev, "device's health compromised\n"); + dev_err(&dev->pdev->dev, "device's health compromised - reached miss count\n"); print_health_info(dev); - spin_lock_irq(&health_lock); - list_add_tail(&health->list, &health_list); - spin_unlock_irq(&health_lock); - - queue_work(mlx5_core_wq, &health_work); } else { - get_random_bytes(&next, sizeof(next)); - next %= HZ; - next += jiffies + MLX5_HEALTH_POLL_INTERVAL; - mod_timer(&health->timer, next); + mod_timer(&health->timer, get_next_poll_jiffies()); + } + + if (in_fatal(dev) && !health->sick) { + health->sick = true; + print_health_info(dev); + queue_work(health->wq, &health->work); } } @@ -167,7 +275,6 @@ void mlx5_start_health_poll(struct mlx5_core_dev *dev) { struct mlx5_core_health *health = &dev->priv.health; - INIT_LIST_HEAD(&health->list); init_timer(&health->timer); health->health = &dev->iseg->health; health->health_counter = &dev->iseg->health_counter; @@ -183,18 +290,33 @@ void mlx5_stop_health_poll(struct mlx5_core_dev *dev) struct mlx5_core_health *health = &dev->priv.health; del_timer_sync(&health->timer); - - spin_lock_irq(&health_lock); - if (!list_empty(&health->list)) - list_del_init(&health->list); - spin_unlock_irq(&health_lock); } -void mlx5_health_cleanup(void) +void mlx5_health_cleanup(struct mlx5_core_dev *dev) { + struct mlx5_core_health *health = &dev->priv.health; + + destroy_workqueue(health->wq); } -void __init mlx5_health_init(void) +int mlx5_health_init(struct mlx5_core_dev *dev) { - INIT_WORK(&health_work, health_care); + struct mlx5_core_health *health; + char *name; + + health = &dev->priv.health; + name = kmalloc(64, GFP_KERNEL); + if (!name) + return -ENOMEM; + + strcpy(name, "mlx5_health"); + strcat(name, dev_name(&dev->pdev->dev)); + health->wq = create_singlethread_workqueue(name); + kfree(name); + if (!health->wq) + return -ENOMEM; + + INIT_WORK(&health->work, health_care); + + return 0; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/main.c b/drivers/net/ethernet/mellanox/mlx5/core/main.c index 03aabdd79..4ac8d4cc4 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/main.c @@ -30,7 +30,7 @@ * SOFTWARE. */ -#include <asm-generic/kmap_types.h> +#include <linux/highmem.h> #include <linux/module.h> #include <linux/init.h> #include <linux/errno.h> @@ -39,12 +39,14 @@ #include <linux/slab.h> #include <linux/io-mapping.h> #include <linux/interrupt.h> +#include <linux/delay.h> #include <linux/mlx5/driver.h> #include <linux/mlx5/cq.h> #include <linux/mlx5/qp.h> #include <linux/mlx5/srq.h> #include <linux/debugfs.h> #include <linux/kmod.h> +#include <linux/delay.h> #include <linux/mlx5/mlx5_ifc.h> #include "mlx5_core.h" @@ -62,7 +64,6 @@ static int prof_sel = MLX5_DEFAULT_PROF; module_param_named(prof_sel, prof_sel, int, 0444); MODULE_PARM_DESC(prof_sel, "profile selector. Valid range 0 - 2"); -struct workqueue_struct *mlx5_core_wq; static LIST_HEAD(intf_list); static LIST_HEAD(dev_list); static DEFINE_MUTEX(intf_mutex); @@ -152,6 +153,25 @@ static struct mlx5_profile profile[] = { }, }; +#define FW_INIT_TIMEOUT_MILI 2000 +#define FW_INIT_WAIT_MS 2 + +static int wait_fw_init(struct mlx5_core_dev *dev, u32 max_wait_mili) +{ + unsigned long end = jiffies + msecs_to_jiffies(max_wait_mili); + int err = 0; + + while (fw_initializing(dev)) { + if (time_after(jiffies, end)) { + err = -EBUSY; + break; + } + msleep(FW_INIT_WAIT_MS); + } + + return err; +} + static int set_dma_caps(struct pci_dev *pdev) { int err; @@ -182,6 +202,34 @@ static int set_dma_caps(struct pci_dev *pdev) return err; } +static int mlx5_pci_enable_device(struct mlx5_core_dev *dev) +{ + struct pci_dev *pdev = dev->pdev; + int err = 0; + + mutex_lock(&dev->pci_status_mutex); + if (dev->pci_status == MLX5_PCI_STATUS_DISABLED) { + err = pci_enable_device(pdev); + if (!err) + dev->pci_status = MLX5_PCI_STATUS_ENABLED; + } + mutex_unlock(&dev->pci_status_mutex); + + return err; +} + +static void mlx5_pci_disable_device(struct mlx5_core_dev *dev) +{ + struct pci_dev *pdev = dev->pdev; + + mutex_lock(&dev->pci_status_mutex); + if (dev->pci_status == MLX5_PCI_STATUS_ENABLED) { + pci_disable_device(pdev); + dev->pci_status = MLX5_PCI_STATUS_DISABLED; + } + mutex_unlock(&dev->pci_status_mutex); +} + static int request_bar(struct pci_dev *pdev) { int err = 0; @@ -672,12 +720,126 @@ static void unmap_bf_area(struct mlx5_core_dev *dev) io_mapping_free(dev->priv.bf_mapping); } -static int mlx5_dev_init(struct mlx5_core_dev *dev, struct pci_dev *pdev) +static void mlx5_add_device(struct mlx5_interface *intf, struct mlx5_priv *priv) +{ + struct mlx5_device_context *dev_ctx; + struct mlx5_core_dev *dev = container_of(priv, struct mlx5_core_dev, priv); + + dev_ctx = kmalloc(sizeof(*dev_ctx), GFP_KERNEL); + if (!dev_ctx) + return; + + dev_ctx->intf = intf; + dev_ctx->context = intf->add(dev); + + if (dev_ctx->context) { + spin_lock_irq(&priv->ctx_lock); + list_add_tail(&dev_ctx->list, &priv->ctx_list); + spin_unlock_irq(&priv->ctx_lock); + } else { + kfree(dev_ctx); + } +} + +static void mlx5_remove_device(struct mlx5_interface *intf, struct mlx5_priv *priv) +{ + struct mlx5_device_context *dev_ctx; + struct mlx5_core_dev *dev = container_of(priv, struct mlx5_core_dev, priv); + + list_for_each_entry(dev_ctx, &priv->ctx_list, list) + if (dev_ctx->intf == intf) { + spin_lock_irq(&priv->ctx_lock); + list_del(&dev_ctx->list); + spin_unlock_irq(&priv->ctx_lock); + + intf->remove(dev, dev_ctx->context); + kfree(dev_ctx); + return; + } +} + +static int mlx5_register_device(struct mlx5_core_dev *dev) { struct mlx5_priv *priv = &dev->priv; - int err; + struct mlx5_interface *intf; + + mutex_lock(&intf_mutex); + list_add_tail(&priv->dev_list, &dev_list); + list_for_each_entry(intf, &intf_list, list) + mlx5_add_device(intf, priv); + mutex_unlock(&intf_mutex); + + return 0; +} + +static void mlx5_unregister_device(struct mlx5_core_dev *dev) +{ + struct mlx5_priv *priv = &dev->priv; + struct mlx5_interface *intf; + + mutex_lock(&intf_mutex); + list_for_each_entry(intf, &intf_list, list) + mlx5_remove_device(intf, priv); + list_del(&priv->dev_list); + mutex_unlock(&intf_mutex); +} + +int mlx5_register_interface(struct mlx5_interface *intf) +{ + struct mlx5_priv *priv; + + if (!intf->add || !intf->remove) + return -EINVAL; + + mutex_lock(&intf_mutex); + list_add_tail(&intf->list, &intf_list); + list_for_each_entry(priv, &dev_list, dev_list) + mlx5_add_device(intf, priv); + mutex_unlock(&intf_mutex); + + return 0; +} +EXPORT_SYMBOL(mlx5_register_interface); + +void mlx5_unregister_interface(struct mlx5_interface *intf) +{ + struct mlx5_priv *priv; + + mutex_lock(&intf_mutex); + list_for_each_entry(priv, &dev_list, dev_list) + mlx5_remove_device(intf, priv); + list_del(&intf->list); + mutex_unlock(&intf_mutex); +} +EXPORT_SYMBOL(mlx5_unregister_interface); + +void *mlx5_get_protocol_dev(struct mlx5_core_dev *mdev, int protocol) +{ + struct mlx5_priv *priv = &mdev->priv; + struct mlx5_device_context *dev_ctx; + unsigned long flags; + void *result = NULL; + + spin_lock_irqsave(&priv->ctx_lock, flags); + + list_for_each_entry(dev_ctx, &mdev->priv.ctx_list, list) + if ((dev_ctx->intf->protocol == protocol) && + dev_ctx->intf->get_dev) { + result = dev_ctx->intf->get_dev(dev_ctx->context); + break; + } + + spin_unlock_irqrestore(&priv->ctx_lock, flags); + + return result; +} +EXPORT_SYMBOL(mlx5_get_protocol_dev); + +static int mlx5_pci_init(struct mlx5_core_dev *dev, struct mlx5_priv *priv) +{ + struct pci_dev *pdev = dev->pdev; + int err = 0; - dev->pdev = pdev; pci_set_drvdata(dev->pdev, dev); strncpy(priv->name, dev_name(&pdev->dev), MLX5_MAX_NAME_LEN); priv->name[MLX5_MAX_NAME_LEN - 1] = 0; @@ -694,7 +856,7 @@ static int mlx5_dev_init(struct mlx5_core_dev *dev, struct pci_dev *pdev) if (!priv->dbg_root) return -ENOMEM; - err = pci_enable_device(pdev); + err = mlx5_pci_enable_device(dev); if (err) { dev_err(&pdev->dev, "Cannot enable PCI device, aborting\n"); goto err_dbg; @@ -721,13 +883,61 @@ static int mlx5_dev_init(struct mlx5_core_dev *dev, struct pci_dev *pdev) dev_err(&pdev->dev, "Failed mapping initialization segment, aborting\n"); goto err_clr_master; } + + return 0; + +err_clr_master: + pci_clear_master(dev->pdev); + release_bar(dev->pdev); +err_disable: + mlx5_pci_disable_device(dev); + +err_dbg: + debugfs_remove(priv->dbg_root); + return err; +} + +static void mlx5_pci_close(struct mlx5_core_dev *dev, struct mlx5_priv *priv) +{ + iounmap(dev->iseg); + pci_clear_master(dev->pdev); + release_bar(dev->pdev); + mlx5_pci_disable_device(dev); + debugfs_remove(priv->dbg_root); +} + +#define MLX5_IB_MOD "mlx5_ib" +static int mlx5_load_one(struct mlx5_core_dev *dev, struct mlx5_priv *priv) +{ + struct pci_dev *pdev = dev->pdev; + int err; + + mutex_lock(&dev->intf_state_mutex); + if (dev->interface_state == MLX5_INTERFACE_STATE_UP) { + dev_warn(&dev->pdev->dev, "%s: interface is up, NOP\n", + __func__); + goto out; + } + dev_info(&pdev->dev, "firmware version: %d.%d.%d\n", fw_rev_maj(dev), fw_rev_min(dev), fw_rev_sub(dev)); + /* on load removing any previous indication of internal error, device is + * up + */ + dev->state = MLX5_DEVICE_STATE_UP; + err = mlx5_cmd_init(dev); if (err) { dev_err(&pdev->dev, "Failed initializing command interface, aborting\n"); - goto err_unmap; + goto out_err; + } + + err = wait_fw_init(dev, FW_INIT_TIMEOUT_MILI); + if (err) { + dev_err(&dev->pdev->dev, "Firmware over %d MS in initializing state, aborting\n", + FW_INIT_TIMEOUT_MILI); + goto out_err; } mlx5_pagealloc_init(dev); @@ -842,8 +1052,29 @@ static int mlx5_dev_init(struct mlx5_core_dev *dev, struct pci_dev *pdev) mlx5_init_srq_table(dev); mlx5_init_mr_table(dev); + err = mlx5_register_device(dev); + if (err) { + dev_err(&pdev->dev, "mlx5_register_device failed %d\n", err); + goto err_reg_dev; + } + + err = request_module_nowait(MLX5_IB_MOD); + if (err) + pr_info("failed request module on %s\n", MLX5_IB_MOD); + + dev->interface_state = MLX5_INTERFACE_STATE_UP; +out: + mutex_unlock(&dev->intf_state_mutex); + return 0; +err_reg_dev: + mlx5_cleanup_mr_table(dev); + mlx5_cleanup_srq_table(dev); + mlx5_cleanup_qp_table(dev); + mlx5_cleanup_cq_table(dev); + mlx5_irq_clear_affinity_hints(dev); + err_unmap_bf_area: unmap_bf_area(dev); @@ -865,7 +1096,7 @@ err_stop_poll: mlx5_stop_health_poll(dev); if (mlx5_cmd_teardown_hca(dev)) { dev_err(&dev->pdev->dev, "tear_down_hca failed, skip cleanup\n"); - return err; + goto out_err; } err_pagealloc_stop: @@ -881,25 +1112,25 @@ err_pagealloc_cleanup: mlx5_pagealloc_cleanup(dev); mlx5_cmd_cleanup(dev); -err_unmap: - iounmap(dev->iseg); - -err_clr_master: - pci_clear_master(dev->pdev); - release_bar(dev->pdev); - -err_disable: - pci_disable_device(dev->pdev); +out_err: + dev->state = MLX5_DEVICE_STATE_INTERNAL_ERROR; + mutex_unlock(&dev->intf_state_mutex); -err_dbg: - debugfs_remove(priv->dbg_root); return err; } -static void mlx5_dev_cleanup(struct mlx5_core_dev *dev) +static int mlx5_unload_one(struct mlx5_core_dev *dev, struct mlx5_priv *priv) { - struct mlx5_priv *priv = &dev->priv; + int err = 0; + mutex_lock(&dev->intf_state_mutex); + if (dev->interface_state == MLX5_INTERFACE_STATE_DOWN) { + dev_warn(&dev->pdev->dev, "%s: interface is down, NOP\n", + __func__); + goto out; + } + mlx5_unregister_device(dev); + mlx5_cleanup_mr_table(dev); mlx5_cleanup_srq_table(dev); mlx5_cleanup_qp_table(dev); mlx5_cleanup_cq_table(dev); @@ -911,139 +1142,25 @@ static void mlx5_dev_cleanup(struct mlx5_core_dev *dev) mlx5_eq_cleanup(dev); mlx5_disable_msix(dev); mlx5_stop_health_poll(dev); - if (mlx5_cmd_teardown_hca(dev)) { + err = mlx5_cmd_teardown_hca(dev); + if (err) { dev_err(&dev->pdev->dev, "tear_down_hca failed, skip cleanup\n"); - return; + goto out; } mlx5_pagealloc_stop(dev); mlx5_reclaim_startup_pages(dev); mlx5_core_disable_hca(dev); mlx5_pagealloc_cleanup(dev); mlx5_cmd_cleanup(dev); - iounmap(dev->iseg); - pci_clear_master(dev->pdev); - release_bar(dev->pdev); - pci_disable_device(dev->pdev); - debugfs_remove(priv->dbg_root); -} - -static void mlx5_add_device(struct mlx5_interface *intf, struct mlx5_priv *priv) -{ - struct mlx5_device_context *dev_ctx; - struct mlx5_core_dev *dev = container_of(priv, struct mlx5_core_dev, priv); - - dev_ctx = kmalloc(sizeof(*dev_ctx), GFP_KERNEL); - if (!dev_ctx) { - pr_warn("mlx5_add_device: alloc context failed\n"); - return; - } - - dev_ctx->intf = intf; - dev_ctx->context = intf->add(dev); - - if (dev_ctx->context) { - spin_lock_irq(&priv->ctx_lock); - list_add_tail(&dev_ctx->list, &priv->ctx_list); - spin_unlock_irq(&priv->ctx_lock); - } else { - kfree(dev_ctx); - } -} - -static void mlx5_remove_device(struct mlx5_interface *intf, struct mlx5_priv *priv) -{ - struct mlx5_device_context *dev_ctx; - struct mlx5_core_dev *dev = container_of(priv, struct mlx5_core_dev, priv); - - list_for_each_entry(dev_ctx, &priv->ctx_list, list) - if (dev_ctx->intf == intf) { - spin_lock_irq(&priv->ctx_lock); - list_del(&dev_ctx->list); - spin_unlock_irq(&priv->ctx_lock); - - intf->remove(dev, dev_ctx->context); - kfree(dev_ctx); - return; - } -} -static int mlx5_register_device(struct mlx5_core_dev *dev) -{ - struct mlx5_priv *priv = &dev->priv; - struct mlx5_interface *intf; - - mutex_lock(&intf_mutex); - list_add_tail(&priv->dev_list, &dev_list); - list_for_each_entry(intf, &intf_list, list) - mlx5_add_device(intf, priv); - mutex_unlock(&intf_mutex); - - return 0; -} -static void mlx5_unregister_device(struct mlx5_core_dev *dev) -{ - struct mlx5_priv *priv = &dev->priv; - struct mlx5_interface *intf; - - mutex_lock(&intf_mutex); - list_for_each_entry(intf, &intf_list, list) - mlx5_remove_device(intf, priv); - list_del(&priv->dev_list); - mutex_unlock(&intf_mutex); -} - -int mlx5_register_interface(struct mlx5_interface *intf) -{ - struct mlx5_priv *priv; - - if (!intf->add || !intf->remove) - return -EINVAL; - - mutex_lock(&intf_mutex); - list_add_tail(&intf->list, &intf_list); - list_for_each_entry(priv, &dev_list, dev_list) - mlx5_add_device(intf, priv); - mutex_unlock(&intf_mutex); - - return 0; -} -EXPORT_SYMBOL(mlx5_register_interface); - -void mlx5_unregister_interface(struct mlx5_interface *intf) -{ - struct mlx5_priv *priv; - - mutex_lock(&intf_mutex); - list_for_each_entry(priv, &dev_list, dev_list) - mlx5_remove_device(intf, priv); - list_del(&intf->list); - mutex_unlock(&intf_mutex); -} -EXPORT_SYMBOL(mlx5_unregister_interface); - -void *mlx5_get_protocol_dev(struct mlx5_core_dev *mdev, int protocol) -{ - struct mlx5_priv *priv = &mdev->priv; - struct mlx5_device_context *dev_ctx; - unsigned long flags; - void *result = NULL; - - spin_lock_irqsave(&priv->ctx_lock, flags); - - list_for_each_entry(dev_ctx, &mdev->priv.ctx_list, list) - if ((dev_ctx->intf->protocol == protocol) && - dev_ctx->intf->get_dev) { - result = dev_ctx->intf->get_dev(dev_ctx->context); - break; - } - spin_unlock_irqrestore(&priv->ctx_lock, flags); - - return result; +out: + dev->interface_state = MLX5_INTERFACE_STATE_DOWN; + mutex_unlock(&dev->intf_state_mutex); + return err; } -EXPORT_SYMBOL(mlx5_get_protocol_dev); -static void mlx5_core_event(struct mlx5_core_dev *dev, enum mlx5_dev_event event, - unsigned long param) +void mlx5_core_event(struct mlx5_core_dev *dev, enum mlx5_dev_event event, + unsigned long param) { struct mlx5_priv *priv = &dev->priv; struct mlx5_device_context *dev_ctx; @@ -1064,7 +1181,6 @@ struct mlx5_core_event_handler { void *data); }; -#define MLX5_IB_MOD "mlx5_ib" static int init_one(struct pci_dev *pdev, const struct pci_device_id *id) @@ -1088,43 +1204,166 @@ static int init_one(struct pci_dev *pdev, prof_sel = MLX5_DEFAULT_PROF; } dev->profile = &profile[prof_sel]; + dev->pdev = pdev; dev->event = mlx5_core_event; INIT_LIST_HEAD(&priv->ctx_list); spin_lock_init(&priv->ctx_lock); - err = mlx5_dev_init(dev, pdev); + mutex_init(&dev->pci_status_mutex); + mutex_init(&dev->intf_state_mutex); + err = mlx5_pci_init(dev, priv); if (err) { - dev_err(&pdev->dev, "mlx5_dev_init failed %d\n", err); - goto out; + dev_err(&pdev->dev, "mlx5_pci_init failed with error code %d\n", err); + goto clean_dev; } - err = mlx5_register_device(dev); + err = mlx5_health_init(dev); if (err) { - dev_err(&pdev->dev, "mlx5_register_device failed %d\n", err); - goto out_init; + dev_err(&pdev->dev, "mlx5_health_init failed with error code %d\n", err); + goto close_pci; } - err = request_module_nowait(MLX5_IB_MOD); - if (err) - pr_info("failed request module on %s\n", MLX5_IB_MOD); + err = mlx5_load_one(dev, priv); + if (err) { + dev_err(&pdev->dev, "mlx5_load_one failed with error code %d\n", err); + goto clean_health; + } return 0; -out_init: - mlx5_dev_cleanup(dev); -out: +clean_health: + mlx5_health_cleanup(dev); +close_pci: + mlx5_pci_close(dev, priv); +clean_dev: + pci_set_drvdata(pdev, NULL); kfree(dev); + return err; } + static void remove_one(struct pci_dev *pdev) { struct mlx5_core_dev *dev = pci_get_drvdata(pdev); + struct mlx5_priv *priv = &dev->priv; - mlx5_unregister_device(dev); - mlx5_dev_cleanup(dev); + if (mlx5_unload_one(dev, priv)) { + dev_err(&dev->pdev->dev, "mlx5_unload_one failed\n"); + mlx5_health_cleanup(dev); + return; + } + mlx5_health_cleanup(dev); + mlx5_pci_close(dev, priv); + pci_set_drvdata(pdev, NULL); kfree(dev); } +static pci_ers_result_t mlx5_pci_err_detected(struct pci_dev *pdev, + pci_channel_state_t state) +{ + struct mlx5_core_dev *dev = pci_get_drvdata(pdev); + struct mlx5_priv *priv = &dev->priv; + + dev_info(&pdev->dev, "%s was called\n", __func__); + mlx5_enter_error_state(dev); + mlx5_unload_one(dev, priv); + mlx5_pci_disable_device(dev); + return state == pci_channel_io_perm_failure ? + PCI_ERS_RESULT_DISCONNECT : PCI_ERS_RESULT_NEED_RESET; +} + +static pci_ers_result_t mlx5_pci_slot_reset(struct pci_dev *pdev) +{ + struct mlx5_core_dev *dev = pci_get_drvdata(pdev); + int err = 0; + + dev_info(&pdev->dev, "%s was called\n", __func__); + + err = mlx5_pci_enable_device(dev); + if (err) { + dev_err(&pdev->dev, "%s: mlx5_pci_enable_device failed with error code: %d\n" + , __func__, err); + return PCI_ERS_RESULT_DISCONNECT; + } + pci_set_master(pdev); + pci_set_power_state(pdev, PCI_D0); + pci_restore_state(pdev); + + return err ? PCI_ERS_RESULT_DISCONNECT : PCI_ERS_RESULT_RECOVERED; +} + +void mlx5_disable_device(struct mlx5_core_dev *dev) +{ + mlx5_pci_err_detected(dev->pdev, 0); +} + +/* wait for the device to show vital signs. For now we check + * that we can read the device ID and that the health buffer + * shows a non zero value which is different than 0xffffffff + */ +static void wait_vital(struct pci_dev *pdev) +{ + struct mlx5_core_dev *dev = pci_get_drvdata(pdev); + struct mlx5_core_health *health = &dev->priv.health; + const int niter = 100; + u32 count; + u16 did; + int i; + + /* Wait for firmware to be ready after reset */ + msleep(1000); + for (i = 0; i < niter; i++) { + if (pci_read_config_word(pdev, 2, &did)) { + dev_warn(&pdev->dev, "failed reading config word\n"); + break; + } + if (did == pdev->device) { + dev_info(&pdev->dev, "device ID correctly read after %d iterations\n", i); + break; + } + msleep(50); + } + if (i == niter) + dev_warn(&pdev->dev, "%s-%d: could not read device ID\n", __func__, __LINE__); + + for (i = 0; i < niter; i++) { + count = ioread32be(health->health_counter); + if (count && count != 0xffffffff) { + dev_info(&pdev->dev, "Counter value 0x%x after %d iterations\n", count, i); + break; + } + msleep(50); + } + + if (i == niter) + dev_warn(&pdev->dev, "%s-%d: could not read device ID\n", __func__, __LINE__); +} + +static void mlx5_pci_resume(struct pci_dev *pdev) +{ + struct mlx5_core_dev *dev = pci_get_drvdata(pdev); + struct mlx5_priv *priv = &dev->priv; + int err; + + dev_info(&pdev->dev, "%s was called\n", __func__); + + pci_save_state(pdev); + wait_vital(pdev); + + err = mlx5_load_one(dev, priv); + if (err) + dev_err(&pdev->dev, "%s: mlx5_load_one failed with error code: %d\n" + , __func__, err); + else + dev_info(&pdev->dev, "%s: device recovered\n", __func__); +} + +static const struct pci_error_handlers mlx5_err_handler = { + .error_detected = mlx5_pci_err_detected, + .slot_reset = mlx5_pci_slot_reset, + .resume = mlx5_pci_resume +}; + static const struct pci_device_id mlx5_core_pci_table[] = { { PCI_VDEVICE(MELLANOX, 0x1011) }, /* Connect-IB */ { PCI_VDEVICE(MELLANOX, 0x1012) }, /* Connect-IB VF */ @@ -1141,7 +1380,8 @@ static struct pci_driver mlx5_core_driver = { .name = DRIVER_NAME, .id_table = mlx5_core_pci_table, .probe = init_one, - .remove = remove_one + .remove = remove_one, + .err_handler = &mlx5_err_handler }; static int __init init(void) @@ -1149,16 +1389,10 @@ static int __init init(void) int err; mlx5_register_debugfs(); - mlx5_core_wq = create_singlethread_workqueue("mlx5_core_wq"); - if (!mlx5_core_wq) { - err = -ENOMEM; - goto err_debug; - } - mlx5_health_init(); err = pci_register_driver(&mlx5_core_driver); if (err) - goto err_health; + goto err_debug; #ifdef CONFIG_MLX5_CORE_EN mlx5e_init(); @@ -1166,9 +1400,6 @@ static int __init init(void) return 0; -err_health: - mlx5_health_cleanup(); - destroy_workqueue(mlx5_core_wq); err_debug: mlx5_unregister_debugfs(); return err; @@ -1180,8 +1411,6 @@ static void __exit cleanup(void) mlx5e_cleanup(); #endif pci_unregister_driver(&mlx5_core_driver); - mlx5_health_cleanup(); - destroy_workqueue(mlx5_core_wq); mlx5_unregister_debugfs(); } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h index 566a70488..cee5b7a83 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h @@ -43,25 +43,25 @@ extern int mlx5_core_debug_mask; -#define mlx5_core_dbg(dev, format, ...) \ - pr_debug("%s:%s:%d:(pid %d): " format, \ - (dev)->priv.name, __func__, __LINE__, current->pid, \ +#define mlx5_core_dbg(__dev, format, ...) \ + dev_dbg(&(__dev)->pdev->dev, "%s:%s:%d:(pid %d): " format, \ + (__dev)->priv.name, __func__, __LINE__, current->pid, \ ##__VA_ARGS__) -#define mlx5_core_dbg_mask(dev, mask, format, ...) \ +#define mlx5_core_dbg_mask(__dev, mask, format, ...) \ do { \ if ((mask) & mlx5_core_debug_mask) \ - mlx5_core_dbg(dev, format, ##__VA_ARGS__); \ + mlx5_core_dbg(__dev, format, ##__VA_ARGS__); \ } while (0) -#define mlx5_core_err(dev, format, ...) \ - pr_err("%s:%s:%d:(pid %d): " format, \ - (dev)->priv.name, __func__, __LINE__, current->pid, \ +#define mlx5_core_err(__dev, format, ...) \ + dev_err(&(__dev)->pdev->dev, "%s:%s:%d:(pid %d): " format, \ + (__dev)->priv.name, __func__, __LINE__, current->pid, \ ##__VA_ARGS__) -#define mlx5_core_warn(dev, format, ...) \ - pr_warn("%s:%s:%d:(pid %d): " format, \ - (dev)->priv.name, __func__, __LINE__, current->pid, \ +#define mlx5_core_warn(__dev, format, ...) \ + dev_warn(&(__dev)->pdev->dev, "%s:%s:%d:(pid %d): " format, \ + (__dev)->priv.name, __func__, __LINE__, current->pid, \ ##__VA_ARGS__) enum { @@ -86,6 +86,10 @@ int mlx5_query_hca_caps(struct mlx5_core_dev *dev); int mlx5_query_board_id(struct mlx5_core_dev *dev); int mlx5_cmd_init_hca(struct mlx5_core_dev *dev); int mlx5_cmd_teardown_hca(struct mlx5_core_dev *dev); +void mlx5_core_event(struct mlx5_core_dev *dev, enum mlx5_dev_event event, + unsigned long param); +void mlx5_enter_error_state(struct mlx5_core_dev *dev); +void mlx5_disable_device(struct mlx5_core_dev *dev); void mlx5e_init(void); void mlx5e_cleanup(void); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/mr.c b/drivers/net/ethernet/mellanox/mlx5/core/mr.c index 1adb300dd..6fa22b51e 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/mr.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/mr.c @@ -40,6 +40,7 @@ void mlx5_init_mr_table(struct mlx5_core_dev *dev) { struct mlx5_mr_table *table = &dev->priv.mr_table; + memset(table, 0, sizeof(*table)); rwlock_init(&table->lock); INIT_RADIX_TREE(&table->tree, GFP_ATOMIC); } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/pagealloc.c b/drivers/net/ethernet/mellanox/mlx5/core/pagealloc.c index 8a64542ab..4d3377b12 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/pagealloc.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/pagealloc.c @@ -30,7 +30,7 @@ * SOFTWARE. */ -#include <asm-generic/kmap_types.h> +#include <linux/highmem.h> #include <linux/kernel.h> #include <linux/module.h> #include <linux/mlx5/driver.h> @@ -275,12 +275,36 @@ out_alloc: return err; } + +static void page_notify_fail(struct mlx5_core_dev *dev, u16 func_id) +{ + struct mlx5_manage_pages_inbox *in; + struct mlx5_manage_pages_outbox out; + int err; + + in = kzalloc(sizeof(*in), GFP_KERNEL); + if (!in) + return; + + memset(&out, 0, sizeof(out)); + in->hdr.opcode = cpu_to_be16(MLX5_CMD_OP_MANAGE_PAGES); + in->hdr.opmod = cpu_to_be16(MLX5_PAGES_CANT_GIVE); + in->func_id = cpu_to_be16(func_id); + err = mlx5_cmd_exec(dev, in, sizeof(*in), &out, sizeof(out)); + if (!err) + err = mlx5_cmd_status_to_err(&out.hdr); + + if (err) + mlx5_core_warn(dev, "page notify failed\n"); + + kfree(in); +} + static int give_pages(struct mlx5_core_dev *dev, u16 func_id, int npages, int notify_fail) { struct mlx5_manage_pages_inbox *in; struct mlx5_manage_pages_outbox out; - struct mlx5_manage_pages_inbox *nin; int inlen; u64 addr; int err; @@ -289,8 +313,9 @@ static int give_pages(struct mlx5_core_dev *dev, u16 func_id, int npages, inlen = sizeof(*in) + npages * sizeof(in->pas[0]); in = mlx5_vzalloc(inlen); if (!in) { + err = -ENOMEM; mlx5_core_warn(dev, "vzalloc failed %d\n", inlen); - return -ENOMEM; + goto out_free; } memset(&out, 0, sizeof(out)); @@ -316,43 +341,29 @@ retry: if (err) { mlx5_core_warn(dev, "func_id 0x%x, npages %d, err %d\n", func_id, npages, err); - goto out_alloc; + goto out_4k; } dev->priv.fw_pages += npages; - if (out.hdr.status) { - err = mlx5_cmd_status_to_err(&out.hdr); - if (err) { - mlx5_core_warn(dev, "func_id 0x%x, npages %d, status %d\n", - func_id, npages, out.hdr.status); - goto out_alloc; - } + err = mlx5_cmd_status_to_err(&out.hdr); + if (err) { + mlx5_core_warn(dev, "func_id 0x%x, npages %d, status %d\n", + func_id, npages, out.hdr.status); + goto out_4k; } mlx5_core_dbg(dev, "err %d\n", err); - goto out_free; - -out_alloc: - if (notify_fail) { - nin = kzalloc(sizeof(*nin), GFP_KERNEL); - if (!nin) { - mlx5_core_warn(dev, "allocation failed\n"); - goto out_4k; - } - memset(&out, 0, sizeof(out)); - nin->hdr.opcode = cpu_to_be16(MLX5_CMD_OP_MANAGE_PAGES); - nin->hdr.opmod = cpu_to_be16(MLX5_PAGES_CANT_GIVE); - if (mlx5_cmd_exec(dev, nin, sizeof(*nin), &out, sizeof(out))) - mlx5_core_warn(dev, "page notify failed\n"); - kfree(nin); - } + kvfree(in); + return 0; out_4k: for (i--; i >= 0; i--) free_4k(dev, be64_to_cpu(in->pas[i])); out_free: kvfree(in); + if (notify_fail) + page_notify_fail(dev, func_id); return err; } @@ -482,15 +493,20 @@ int mlx5_reclaim_startup_pages(struct mlx5_core_dev *dev) struct fw_page *fwp; struct rb_node *p; int nclaimed = 0; - int err; + int err = 0; do { p = rb_first(&dev->priv.page_root); if (p) { fwp = rb_entry(p, struct fw_page, rb_node); - err = reclaim_pages(dev, fwp->func_id, - optimal_reclaimed_pages(), - &nclaimed); + if (dev->state == MLX5_DEVICE_STATE_INTERNAL_ERROR) { + free_4k(dev, fwp->addr); + nclaimed = 1; + } else { + err = reclaim_pages(dev, fwp->func_id, + optimal_reclaimed_pages(), + &nclaimed); + } if (err) { mlx5_core_warn(dev, "failed reclaiming pages (%d)\n", err); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/port.c b/drivers/net/ethernet/mellanox/mlx5/core/port.c index 3b9480fa3..a87e773e9 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/port.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/port.c @@ -90,16 +90,13 @@ int mlx5_set_port_caps(struct mlx5_core_dev *dev, u8 port_num, u32 caps) { struct mlx5_reg_pcap in; struct mlx5_reg_pcap out; - int err; memset(&in, 0, sizeof(in)); in.caps_127_96 = cpu_to_be32(caps); in.port_num = port_num; - err = mlx5_core_access_reg(dev, &in, sizeof(in), &out, - sizeof(out), MLX5_REG_PCAP, 0, 1); - - return err; + return mlx5_core_access_reg(dev, &in, sizeof(in), &out, + sizeof(out), MLX5_REG_PCAP, 0, 1); } EXPORT_SYMBOL_GPL(mlx5_set_port_caps); @@ -107,16 +104,13 @@ int mlx5_query_port_ptys(struct mlx5_core_dev *dev, u32 *ptys, int ptys_size, int proto_mask, u8 local_port) { u32 in[MLX5_ST_SZ_DW(ptys_reg)]; - int err; memset(in, 0, sizeof(in)); MLX5_SET(ptys_reg, in, local_port, local_port); MLX5_SET(ptys_reg, in, proto_mask, proto_mask); - err = mlx5_core_access_reg(dev, in, sizeof(in), ptys, - ptys_size, MLX5_REG_PTYS, 0, 0); - - return err; + return mlx5_core_access_reg(dev, in, sizeof(in), ptys, + ptys_size, MLX5_REG_PTYS, 0, 0); } EXPORT_SYMBOL_GPL(mlx5_query_port_ptys); @@ -199,7 +193,6 @@ int mlx5_set_port_proto(struct mlx5_core_dev *dev, u32 proto_admin, { u32 in[MLX5_ST_SZ_DW(ptys_reg)]; u32 out[MLX5_ST_SZ_DW(ptys_reg)]; - int err; memset(in, 0, sizeof(in)); @@ -210,9 +203,8 @@ int mlx5_set_port_proto(struct mlx5_core_dev *dev, u32 proto_admin, else MLX5_SET(ptys_reg, in, ib_proto_admin, proto_admin); - err = mlx5_core_access_reg(dev, in, sizeof(in), out, - sizeof(out), MLX5_REG_PTYS, 0, 1); - return err; + return mlx5_core_access_reg(dev, in, sizeof(in), out, + sizeof(out), MLX5_REG_PTYS, 0, 1); } EXPORT_SYMBOL_GPL(mlx5_set_port_proto); @@ -250,7 +242,7 @@ int mlx5_query_port_admin_status(struct mlx5_core_dev *dev, return err; *status = MLX5_GET(paos_reg, out, admin_status); - return err; + return 0; } EXPORT_SYMBOL_GPL(mlx5_query_port_admin_status); @@ -308,15 +300,12 @@ static int mlx5_query_port_pvlc(struct mlx5_core_dev *dev, u32 *pvlc, int pvlc_size, u8 local_port) { u32 in[MLX5_ST_SZ_DW(pvlc_reg)]; - int err; memset(in, 0, sizeof(in)); MLX5_SET(pvlc_reg, in, local_port, local_port); - err = mlx5_core_access_reg(dev, in, sizeof(in), pvlc, - pvlc_size, MLX5_REG_PVLC, 0, 0); - - return err; + return mlx5_core_access_reg(dev, in, sizeof(in), pvlc, + pvlc_size, MLX5_REG_PVLC, 0, 0); } int mlx5_query_port_vl_hw_cap(struct mlx5_core_dev *dev, @@ -339,16 +328,14 @@ int mlx5_set_port_pause(struct mlx5_core_dev *dev, u32 rx_pause, u32 tx_pause) { u32 in[MLX5_ST_SZ_DW(pfcc_reg)]; u32 out[MLX5_ST_SZ_DW(pfcc_reg)]; - int err; memset(in, 0, sizeof(in)); MLX5_SET(pfcc_reg, in, local_port, 1); MLX5_SET(pfcc_reg, in, pptx, tx_pause); MLX5_SET(pfcc_reg, in, pprx, rx_pause); - err = mlx5_core_access_reg(dev, in, sizeof(in), out, - sizeof(out), MLX5_REG_PFCC, 0, 1); - return err; + return mlx5_core_access_reg(dev, in, sizeof(in), out, + sizeof(out), MLX5_REG_PFCC, 0, 1); } EXPORT_SYMBOL_GPL(mlx5_set_port_pause); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/qp.c b/drivers/net/ethernet/mellanox/mlx5/core/qp.c index 8b494b562..30e2ba3f5 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/qp.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/qp.c @@ -345,6 +345,7 @@ void mlx5_init_qp_table(struct mlx5_core_dev *dev) { struct mlx5_qp_table *table = &dev->priv.qp_table; + memset(table, 0, sizeof(*table)); spin_lock_init(&table->lock); INIT_RADIX_TREE(&table->tree, GFP_ATOMIC); mlx5_qp_debugfs_init(dev); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/srq.c b/drivers/net/ethernet/mellanox/mlx5/core/srq.c index c48f504cc..ffada8019 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/srq.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/srq.c @@ -531,6 +531,7 @@ void mlx5_init_srq_table(struct mlx5_core_dev *dev) { struct mlx5_srq_table *table = &dev->priv.srq_table; + memset(table, 0, sizeof(*table)); spin_lock_init(&table->lock); INIT_RADIX_TREE(&table->tree, GFP_ATOMIC); } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/transobj.c b/drivers/net/ethernet/mellanox/mlx5/core/transobj.c index b4c87c7b0..d7068f54e 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/transobj.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/transobj.c @@ -177,7 +177,7 @@ int mlx5_core_modify_tir(struct mlx5_core_dev *dev, u32 tirn, u32 *in, void mlx5_core_destroy_tir(struct mlx5_core_dev *dev, u32 tirn) { - u32 in[MLX5_ST_SZ_DW(destroy_tir_out)]; + u32 in[MLX5_ST_SZ_DW(destroy_tir_in)]; u32 out[MLX5_ST_SZ_DW(destroy_tir_out)]; memset(in, 0, sizeof(in)); @@ -206,7 +206,7 @@ int mlx5_core_create_tis(struct mlx5_core_dev *dev, u32 *in, int inlen, void mlx5_core_destroy_tis(struct mlx5_core_dev *dev, u32 tisn) { - u32 in[MLX5_ST_SZ_DW(destroy_tis_out)]; + u32 in[MLX5_ST_SZ_DW(destroy_tis_in)]; u32 out[MLX5_ST_SZ_DW(destroy_tis_out)]; memset(in, 0, sizeof(in)); diff --git a/drivers/net/ethernet/mellanox/mlxsw/Kconfig b/drivers/net/ethernet/mellanox/mlxsw/Kconfig index 2941d9c5a..e36e12219 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/Kconfig +++ b/drivers/net/ethernet/mellanox/mlxsw/Kconfig @@ -30,3 +30,14 @@ config MLXSW_SWITCHX2 To compile this driver as a module, choose M here: the module will be called mlxsw_switchx2. + +config MLXSW_SPECTRUM + tristate "Mellanox Technologies Spectrum support" + depends on MLXSW_CORE && NET_SWITCHDEV + default m + ---help--- + This driver supports Mellanox Technologies Spectrum Ethernet + Switch ASICs. + + To compile this driver as a module, choose M here: the + module will be called mlxsw_spectrum. diff --git a/drivers/net/ethernet/mellanox/mlxsw/Makefile b/drivers/net/ethernet/mellanox/mlxsw/Makefile index 0a05f65ee..af015818f 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/Makefile +++ b/drivers/net/ethernet/mellanox/mlxsw/Makefile @@ -4,3 +4,6 @@ obj-$(CONFIG_MLXSW_PCI) += mlxsw_pci.o mlxsw_pci-objs := pci.o obj-$(CONFIG_MLXSW_SWITCHX2) += mlxsw_switchx2.o mlxsw_switchx2-objs := switchx2.o +obj-$(CONFIG_MLXSW_SPECTRUM) += mlxsw_spectrum.o +mlxsw_spectrum-objs := spectrum.o spectrum_buffers.o \ + spectrum_switchdev.o diff --git a/drivers/net/ethernet/mellanox/mlxsw/cmd.h b/drivers/net/ethernet/mellanox/mlxsw/cmd.h index 770db17eb..cd63b8263 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/cmd.h +++ b/drivers/net/ethernet/mellanox/mlxsw/cmd.h @@ -464,6 +464,8 @@ MLXSW_ITEM32(cmd_mbox, query_aq_cap, max_sg_rq, 0x10, 0, 8); * passed in this command must be pinned. */ +#define MLXSW_CMD_MAP_FA_VPM_ENTRIES_MAX 32 + static inline int mlxsw_cmd_map_fa(struct mlxsw_core *mlxsw_core, char *in_mbox, u32 vpm_entries_count) { @@ -568,7 +570,7 @@ MLXSW_ITEM32(cmd_mbox, config_profile, set_max_vlan_groups, 0x0C, 6, 1); */ MLXSW_ITEM32(cmd_mbox, config_profile, set_max_regions, 0x0C, 7, 1); -/* cmd_mbox_config_profile_set_fid_based +/* cmd_mbox_config_profile_set_flood_mode * Capability bit. Setting a bit to 1 configures the profile * according to the mailbox contents. */ @@ -649,12 +651,8 @@ MLXSW_ITEM32(cmd_mbox, config_profile, max_vlan_groups, 0x28, 0, 12); MLXSW_ITEM32(cmd_mbox, config_profile, max_regions, 0x2C, 0, 16); /* cmd_mbox_config_profile_max_flood_tables - * Maximum number of Flooding Tables. Flooding Tables are associated to - * the different packet types for the different switch partitions. - * Note that the table size depends on the fid_based mode. - * In SwitchX silicon, tables are split equally between the switch - * partitions. e.g. for 2 swids and 8 tables, the first 4 are associated - * with swid-1 and the last 4 are associated with swid-2. + * Maximum number of single-entry flooding tables. Different flooding tables + * can be associated with different packet types. */ MLXSW_ITEM32(cmd_mbox, config_profile, max_flood_tables, 0x30, 16, 4); @@ -665,15 +663,42 @@ MLXSW_ITEM32(cmd_mbox, config_profile, max_flood_tables, 0x30, 16, 4); */ MLXSW_ITEM32(cmd_mbox, config_profile, max_vid_flood_tables, 0x30, 8, 4); -/* cmd_mbox_config_profile_fid_based - * FID Based Flood Mode - * 00 Do not use FID to offset the index into the Port Group Table/Multicast ID - * 01 Use FID to offset the index to the Port Group Table (pgi) - * 10 Use FID to offset the index to the Port Group Table (pgi) and - * the Multicast ID +/* cmd_mbox_config_profile_flood_mode + * Flooding mode to use. + * 0-2 - Backward compatible modes for SwitchX devices. + * 3 - Mixed mode, where: + * max_flood_tables indicates the number of single-entry tables. + * max_vid_flood_tables indicates the number of per-VID tables. + * max_fid_offset_flood_tables indicates the number of FID-offset tables. + * max_fid_flood_tables indicates the number of per-FID tables. */ MLXSW_ITEM32(cmd_mbox, config_profile, flood_mode, 0x30, 0, 2); +/* cmd_mbox_config_profile_max_fid_offset_flood_tables + * Maximum number of FID-offset flooding tables. + */ +MLXSW_ITEM32(cmd_mbox, config_profile, + max_fid_offset_flood_tables, 0x34, 24, 4); + +/* cmd_mbox_config_profile_fid_offset_flood_table_size + * The size (number of entries) of each FID-offset flood table. + */ +MLXSW_ITEM32(cmd_mbox, config_profile, + fid_offset_flood_table_size, 0x34, 0, 16); + +/* cmd_mbox_config_profile_max_fid_flood_tables + * Maximum number of per-FID flooding tables. + * + * Note: This flooding tables cover special FIDs only (vFIDs), starting at + * FID value 4K and higher. + */ +MLXSW_ITEM32(cmd_mbox, config_profile, max_fid_flood_tables, 0x38, 24, 4); + +/* cmd_mbox_config_profile_fid_flood_table_size + * The size (number of entries) of each per-FID table. + */ +MLXSW_ITEM32(cmd_mbox, config_profile, fid_flood_table_size, 0x38, 0, 16); + /* cmd_mbox_config_profile_max_ib_mc * Maximum number of multicast FDB records for InfiniBand * FDB (in 512 chunks) per InfiniBand switch partition. diff --git a/drivers/net/ethernet/mellanox/mlxsw/core.c b/drivers/net/ethernet/mellanox/mlxsw/core.c index 28c19cc1a..97f0d93ca 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/core.c +++ b/drivers/net/ethernet/mellanox/mlxsw/core.c @@ -287,7 +287,7 @@ static void mlxsw_emad_pack_op_tlv(char *op_tlv, mlxsw_emad_op_tlv_status_set(op_tlv, 0); mlxsw_emad_op_tlv_register_id_set(op_tlv, reg->id); mlxsw_emad_op_tlv_r_set(op_tlv, MLXSW_EMAD_OP_TLV_REQUEST); - if (MLXSW_CORE_REG_ACCESS_TYPE_QUERY == type) + if (type == MLXSW_CORE_REG_ACCESS_TYPE_QUERY) mlxsw_emad_op_tlv_method_set(op_tlv, MLXSW_EMAD_OP_TLV_METHOD_QUERY); else @@ -362,7 +362,7 @@ static bool mlxsw_emad_is_resp(const struct sk_buff *skb) char *op_tlv; op_tlv = mlxsw_emad_op_tlv(skb); - return (MLXSW_EMAD_OP_TLV_RESPONSE == mlxsw_emad_op_tlv_r_get(op_tlv)); + return (mlxsw_emad_op_tlv_r_get(op_tlv) == MLXSW_EMAD_OP_TLV_RESPONSE); } #define MLXSW_EMAD_TIMEOUT_MS 200 @@ -511,7 +511,6 @@ static int mlxsw_emad_traps_set(struct mlxsw_core *mlxsw_core) return err; mlxsw_reg_hpkt_pack(hpkt_pl, MLXSW_REG_HPKT_ACTION_TRAP_TO_CPU, - MLXSW_REG_HTGT_TRAP_GROUP_EMAD, MLXSW_TRAP_ID_ETHEMAD); return mlxsw_reg_write(mlxsw_core, MLXSW_REG(hpkt), hpkt_pl); } @@ -556,8 +555,8 @@ static void mlxsw_emad_fini(struct mlxsw_core *mlxsw_core) { char hpkt_pl[MLXSW_REG_HPKT_LEN]; + mlxsw_core->emad.use_emad = false; mlxsw_reg_hpkt_pack(hpkt_pl, MLXSW_REG_HPKT_ACTION_DISCARD, - MLXSW_REG_HTGT_TRAP_GROUP_EMAD, MLXSW_TRAP_ID_ETHEMAD); mlxsw_reg_write(mlxsw_core, MLXSW_REG(hpkt), hpkt_pl); diff --git a/drivers/net/ethernet/mellanox/mlxsw/core.h b/drivers/net/ethernet/mellanox/mlxsw/core.h index 165808471..807827350 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/core.h +++ b/drivers/net/ethernet/mellanox/mlxsw/core.h @@ -54,6 +54,7 @@ MODULE_ALIAS(MLXSW_MODULE_ALIAS_PREFIX kind) #define MLXSW_DEVICE_KIND_SWITCHX2 "switchx2" +#define MLXSW_DEVICE_KIND_SPECTRUM "spectrum" struct mlxsw_core; struct mlxsw_driver; @@ -153,6 +154,10 @@ struct mlxsw_config_profile { u8 max_flood_tables; u8 max_vid_flood_tables; u8 flood_mode; + u8 max_fid_offset_flood_tables; + u16 fid_offset_flood_table_size; + u8 max_fid_flood_tables; + u16 fid_flood_table_size; u16 max_ib_mc; u16 max_pkey; u8 ar_sec; diff --git a/drivers/net/ethernet/mellanox/mlxsw/item.h b/drivers/net/ethernet/mellanox/mlxsw/item.h index 36fb1cec5..a94dbda65 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/item.h +++ b/drivers/net/ethernet/mellanox/mlxsw/item.h @@ -171,15 +171,21 @@ static inline void __mlxsw_item_set64(char *buf, struct mlxsw_item *item, } static inline void __mlxsw_item_memcpy_from(char *buf, char *dst, - struct mlxsw_item *item) + struct mlxsw_item *item, + unsigned short index) { - memcpy(dst, &buf[item->offset], item->size.bytes); + unsigned int offset = __mlxsw_item_offset(item, index, sizeof(char)); + + memcpy(dst, &buf[offset], item->size.bytes); } -static inline void __mlxsw_item_memcpy_to(char *buf, char *src, - struct mlxsw_item *item) +static inline void __mlxsw_item_memcpy_to(char *buf, const char *src, + struct mlxsw_item *item, + unsigned short index) { - memcpy(&buf[item->offset], src, item->size.bytes); + unsigned int offset = __mlxsw_item_offset(item, index, sizeof(char)); + + memcpy(&buf[offset], src, item->size.bytes); } static inline u16 @@ -373,12 +379,40 @@ static struct mlxsw_item __ITEM_NAME(_type, _cname, _iname) = { \ static inline void \ mlxsw_##_type##_##_cname##_##_iname##_memcpy_from(char *buf, char *dst) \ { \ - __mlxsw_item_memcpy_from(buf, dst, &__ITEM_NAME(_type, _cname, _iname));\ + __mlxsw_item_memcpy_from(buf, dst, \ + &__ITEM_NAME(_type, _cname, _iname), 0); \ +} \ +static inline void \ +mlxsw_##_type##_##_cname##_##_iname##_memcpy_to(char *buf, const char *src) \ +{ \ + __mlxsw_item_memcpy_to(buf, src, \ + &__ITEM_NAME(_type, _cname, _iname), 0); \ +} + +#define MLXSW_ITEM_BUF_INDEXED(_type, _cname, _iname, _offset, _sizebytes, \ + _step, _instepoffset) \ +static struct mlxsw_item __ITEM_NAME(_type, _cname, _iname) = { \ + .offset = _offset, \ + .step = _step, \ + .in_step_offset = _instepoffset, \ + .size = {.bytes = _sizebytes,}, \ + .name = #_type "_" #_cname "_" #_iname, \ +}; \ +static inline void \ +mlxsw_##_type##_##_cname##_##_iname##_memcpy_from(char *buf, \ + unsigned short index, \ + char *dst) \ +{ \ + __mlxsw_item_memcpy_from(buf, dst, \ + &__ITEM_NAME(_type, _cname, _iname), index); \ } \ static inline void \ -mlxsw_##_type##_##_cname##_##_iname##_memcpy_to(char *buf, char *src) \ +mlxsw_##_type##_##_cname##_##_iname##_memcpy_to(char *buf, \ + unsigned short index, \ + const char *src) \ { \ - __mlxsw_item_memcpy_to(buf, src, &__ITEM_NAME(_type, _cname, _iname)); \ + __mlxsw_item_memcpy_to(buf, src, \ + &__ITEM_NAME(_type, _cname, _iname), index); \ } #define MLXSW_ITEM_BIT_ARRAY(_type, _cname, _iname, _offset, _sizebytes, \ diff --git a/drivers/net/ethernet/mellanox/mlxsw/pci.c b/drivers/net/ethernet/mellanox/mlxsw/pci.c index cef866c37..de69e719d 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/pci.c +++ b/drivers/net/ethernet/mellanox/mlxsw/pci.c @@ -57,6 +57,7 @@ static const char mlxsw_pci_driver_name[] = "mlxsw_pci"; static const struct pci_device_id mlxsw_pci_id_table[] = { {PCI_VDEVICE(MELLANOX, PCI_DEVICE_ID_MELLANOX_SWITCHX2), 0}, + {PCI_VDEVICE(MELLANOX, PCI_DEVICE_ID_MELLANOX_SPECTRUM), 0}, {0, } }; @@ -67,6 +68,8 @@ static const char *mlxsw_pci_device_kind_get(const struct pci_device_id *id) switch (id->device) { case PCI_DEVICE_ID_MELLANOX_SWITCHX2: return MLXSW_DEVICE_KIND_SWITCHX2; + case PCI_DEVICE_ID_MELLANOX_SPECTRUM: + return MLXSW_DEVICE_KIND_SPECTRUM; default: BUG(); } @@ -171,8 +174,8 @@ struct mlxsw_pci { struct msix_entry msix_entry; struct mlxsw_core *core; struct { - u16 num_pages; struct mlxsw_pci_mem_item *items; + unsigned int count; } fw_area; struct { struct mlxsw_pci_mem_item out_mbox; @@ -431,8 +434,7 @@ static int mlxsw_pci_wqe_frag_map(struct mlxsw_pci *mlxsw_pci, char *wqe, mapaddr = pci_map_single(pdev, frag_data, frag_len, direction); if (unlikely(pci_dma_mapping_error(pdev, mapaddr))) { - if (net_ratelimit()) - dev_err(&pdev->dev, "failed to dma map tx frag\n"); + dev_err_ratelimited(&pdev->dev, "failed to dma map tx frag\n"); return -EIO; } mlxsw_pci_wqe_address_set(wqe, index, mapaddr); @@ -497,6 +499,7 @@ static int mlxsw_pci_rdq_init(struct mlxsw_pci *mlxsw_pci, char *mbox, struct mlxsw_pci_queue *q) { struct mlxsw_pci_queue_elem_info *elem_info; + u8 sdq_count = mlxsw_pci_sdq_count(mlxsw_pci); int i; int err; @@ -504,9 +507,9 @@ static int mlxsw_pci_rdq_init(struct mlxsw_pci *mlxsw_pci, char *mbox, q->consumer_counter = 0; /* Set CQ of same number of this RDQ with base - * above MLXSW_PCI_SDQS_MAX as the lower ones are assigned to SDQs. + * above SDQ count as the lower ones are assigned to SDQs. */ - mlxsw_cmd_mbox_sw2hw_dq_cq_set(mbox, q->num + MLXSW_PCI_SDQS_COUNT); + mlxsw_cmd_mbox_sw2hw_dq_cq_set(mbox, sdq_count + q->num); mlxsw_cmd_mbox_sw2hw_dq_log2_dq_sz_set(mbox, 3); /* 8 pages */ for (i = 0; i < MLXSW_PCI_AQ_PAGES; i++) { dma_addr_t mapaddr = __mlxsw_pci_queue_page_get(q, i); @@ -699,8 +702,8 @@ static void mlxsw_pci_cqe_rdq_handle(struct mlxsw_pci *mlxsw_pci, put_new_skb: memset(wqe, 0, q->elem_size); err = mlxsw_pci_rdq_skb_alloc(mlxsw_pci, elem_info); - if (err && net_ratelimit()) - dev_dbg(&pdev->dev, "Failed to alloc skb for RDQ\n"); + if (err) + dev_dbg_ratelimited(&pdev->dev, "Failed to alloc skb for RDQ\n"); /* Everything is set up, ring doorbell to pass elem to HW */ q->producer_counter++; mlxsw_pci_queue_doorbell_producer_ring(mlxsw_pci, q); @@ -830,7 +833,8 @@ static void mlxsw_pci_eq_tasklet(unsigned long data) { struct mlxsw_pci_queue *q = (struct mlxsw_pci_queue *) data; struct mlxsw_pci *mlxsw_pci = q->pci; - unsigned long active_cqns[BITS_TO_LONGS(MLXSW_PCI_CQS_COUNT)]; + u8 cq_count = mlxsw_pci_cq_count(mlxsw_pci); + unsigned long active_cqns[BITS_TO_LONGS(MLXSW_PCI_CQS_MAX)]; char *eqe; u8 cqn; bool cq_handle = false; @@ -866,7 +870,7 @@ static void mlxsw_pci_eq_tasklet(unsigned long data) if (!cq_handle) return; - for_each_set_bit(cqn, active_cqns, MLXSW_PCI_CQS_COUNT) { + for_each_set_bit(cqn, active_cqns, cq_count) { q = mlxsw_pci_cq_get(mlxsw_pci, cqn); mlxsw_pci_queue_tasklet_schedule(q); } @@ -1067,10 +1071,8 @@ static int mlxsw_pci_aqs_init(struct mlxsw_pci *mlxsw_pci, char *mbox) num_eqs = mlxsw_cmd_mbox_query_aq_cap_max_num_eqs_get(mbox); eq_log2sz = mlxsw_cmd_mbox_query_aq_cap_log_max_eq_sz_get(mbox); - if ((num_sdqs != MLXSW_PCI_SDQS_COUNT) || - (num_rdqs != MLXSW_PCI_RDQS_COUNT) || - (num_cqs != MLXSW_PCI_CQS_COUNT) || - (num_eqs != MLXSW_PCI_EQS_COUNT)) { + if (num_sdqs + num_rdqs > num_cqs || + num_cqs > MLXSW_PCI_CQS_MAX || num_eqs != MLXSW_PCI_EQS_COUNT) { dev_err(&pdev->dev, "Unsupported number of queues\n"); return -EINVAL; } @@ -1215,6 +1217,14 @@ static int mlxsw_pci_config_profile(struct mlxsw_pci *mlxsw_pci, char *mbox, mbox, profile->max_flood_tables); mlxsw_cmd_mbox_config_profile_max_vid_flood_tables_set( mbox, profile->max_vid_flood_tables); + mlxsw_cmd_mbox_config_profile_max_fid_offset_flood_tables_set( + mbox, profile->max_fid_offset_flood_tables); + mlxsw_cmd_mbox_config_profile_fid_offset_flood_table_size_set( + mbox, profile->fid_offset_flood_table_size); + mlxsw_cmd_mbox_config_profile_max_fid_flood_tables_set( + mbox, profile->max_fid_flood_tables); + mlxsw_cmd_mbox_config_profile_fid_flood_table_size_set( + mbox, profile->fid_flood_table_size); } if (profile->used_flood_mode) { mlxsw_cmd_mbox_config_profile_set_flood_mode_set( @@ -1272,6 +1282,7 @@ static int mlxsw_pci_fw_area_init(struct mlxsw_pci *mlxsw_pci, char *mbox, u16 num_pages) { struct mlxsw_pci_mem_item *mem_item; + int nent = 0; int i; int err; @@ -1279,7 +1290,7 @@ static int mlxsw_pci_fw_area_init(struct mlxsw_pci *mlxsw_pci, char *mbox, GFP_KERNEL); if (!mlxsw_pci->fw_area.items) return -ENOMEM; - mlxsw_pci->fw_area.num_pages = num_pages; + mlxsw_pci->fw_area.count = num_pages; mlxsw_cmd_mbox_zero(mbox); for (i = 0; i < num_pages; i++) { @@ -1293,13 +1304,22 @@ static int mlxsw_pci_fw_area_init(struct mlxsw_pci *mlxsw_pci, char *mbox, err = -ENOMEM; goto err_alloc; } - mlxsw_cmd_mbox_map_fa_pa_set(mbox, i, mem_item->mapaddr); - mlxsw_cmd_mbox_map_fa_log2size_set(mbox, i, 0); /* 1 page */ + mlxsw_cmd_mbox_map_fa_pa_set(mbox, nent, mem_item->mapaddr); + mlxsw_cmd_mbox_map_fa_log2size_set(mbox, nent, 0); /* 1 page */ + if (++nent == MLXSW_CMD_MAP_FA_VPM_ENTRIES_MAX) { + err = mlxsw_cmd_map_fa(mlxsw_pci->core, mbox, nent); + if (err) + goto err_cmd_map_fa; + nent = 0; + mlxsw_cmd_mbox_zero(mbox); + } } - err = mlxsw_cmd_map_fa(mlxsw_pci->core, mbox, num_pages); - if (err) - goto err_cmd_map_fa; + if (nent) { + err = mlxsw_cmd_map_fa(mlxsw_pci->core, mbox, nent); + if (err) + goto err_cmd_map_fa; + } return 0; @@ -1322,7 +1342,7 @@ static void mlxsw_pci_fw_area_fini(struct mlxsw_pci *mlxsw_pci) mlxsw_cmd_unmap_fa(mlxsw_pci->core); - for (i = 0; i < mlxsw_pci->fw_area.num_pages; i++) { + for (i = 0; i < mlxsw_pci->fw_area.count; i++) { mem_item = &mlxsw_pci->fw_area.items[i]; pci_free_consistent(mlxsw_pci->pdev, mem_item->size, @@ -1642,8 +1662,9 @@ static int mlxsw_pci_cmd_exec(void *bus_priv, u16 opcode, u8 opcode_mod, CIR_OUT_PARAM_LO)); memcpy(out_mbox + sizeof(tmp), &tmp, sizeof(tmp)); } - } else if (!err && out_mbox) + } else if (!err && out_mbox) { memcpy(out_mbox, mlxsw_pci->cmd.out_mbox.buf, out_mbox_size); + } mutex_unlock(&mlxsw_pci->cmd.lock); diff --git a/drivers/net/ethernet/mellanox/mlxsw/pci.h b/drivers/net/ethernet/mellanox/mlxsw/pci.h index 1ef9664b4..142f33d97 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/pci.h +++ b/drivers/net/ethernet/mellanox/mlxsw/pci.h @@ -40,6 +40,7 @@ #include "item.h" #define PCI_DEVICE_ID_MELLANOX_SWITCHX2 0xc738 +#define PCI_DEVICE_ID_MELLANOX_SPECTRUM 0xcb84 #define MLXSW_PCI_BAR0_SIZE (1024 * 1024) /* 1MB */ #define MLXSW_PCI_PAGE_SIZE 4096 @@ -71,9 +72,7 @@ #define MLXSW_PCI_DOORBELL(offset, type_offset, num) \ ((offset) + (type_offset) + (num) * 4) -#define MLXSW_PCI_RDQS_COUNT 24 -#define MLXSW_PCI_SDQS_COUNT 24 -#define MLXSW_PCI_CQS_COUNT (MLXSW_PCI_RDQS_COUNT + MLXSW_PCI_SDQS_COUNT) +#define MLXSW_PCI_CQS_MAX 96 #define MLXSW_PCI_EQS_COUNT 2 #define MLXSW_PCI_EQ_ASYNC_NUM 0 #define MLXSW_PCI_EQ_COMP_NUM 1 diff --git a/drivers/net/ethernet/mellanox/mlxsw/reg.h b/drivers/net/ethernet/mellanox/mlxsw/reg.h index 096e1c121..236fb5d2a 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/reg.h +++ b/drivers/net/ethernet/mellanox/mlxsw/reg.h @@ -99,57 +99,6 @@ static const struct mlxsw_reg_info mlxsw_reg_spad = { */ MLXSW_ITEM_BUF(reg, spad, base_mac, 0x02, 6); -/* SMID - Switch Multicast ID - * -------------------------- - * In multi-chip configuration, each device should maintain mapping between - * Multicast ID (MID) into a list of local ports. This mapping is used in all - * the devices other than the ingress device, and is implemented as part of the - * FDB. The MID record maps from a MID, which is a unique identi- fier of the - * multicast group within the stacking domain, into a list of local ports into - * which the packet is replicated. - */ -#define MLXSW_REG_SMID_ID 0x2007 -#define MLXSW_REG_SMID_LEN 0x420 - -static const struct mlxsw_reg_info mlxsw_reg_smid = { - .id = MLXSW_REG_SMID_ID, - .len = MLXSW_REG_SMID_LEN, -}; - -/* reg_smid_swid - * Switch partition ID. - * Access: Index - */ -MLXSW_ITEM32(reg, smid, swid, 0x00, 24, 8); - -/* reg_smid_mid - * Multicast identifier - global identifier that represents the multicast group - * across all devices - * Access: Index - */ -MLXSW_ITEM32(reg, smid, mid, 0x00, 0, 16); - -/* reg_smid_port - * Local port memebership (1 bit per port). - * Access: RW - */ -MLXSW_ITEM_BIT_ARRAY(reg, smid, port, 0x20, 0x20, 1); - -/* reg_smid_port_mask - * Local port mask (1 bit per port). - * Access: W - */ -MLXSW_ITEM_BIT_ARRAY(reg, smid, port_mask, 0x220, 0x20, 1); - -static inline void mlxsw_reg_smid_pack(char *payload, u16 mid) -{ - MLXSW_REG_ZERO(smid, payload); - mlxsw_reg_smid_swid_set(payload, 0); - mlxsw_reg_smid_mid_set(payload, mid); - mlxsw_reg_smid_port_set(payload, MLXSW_PORT_CPU_PORT, 1); - mlxsw_reg_smid_port_mask_set(payload, MLXSW_PORT_CPU_PORT, 1); -} - /* SSPR - Switch System Port Record Register * ----------------------------------------- * Configures the system port to local port mapping. @@ -208,11 +157,359 @@ static inline void mlxsw_reg_sspr_pack(char *payload, u8 local_port) mlxsw_reg_sspr_system_port_set(payload, local_port); } +/* SFDAT - Switch Filtering Database Aging Time + * -------------------------------------------- + * Controls the Switch aging time. Aging time is able to be set per Switch + * Partition. + */ +#define MLXSW_REG_SFDAT_ID 0x2009 +#define MLXSW_REG_SFDAT_LEN 0x8 + +static const struct mlxsw_reg_info mlxsw_reg_sfdat = { + .id = MLXSW_REG_SFDAT_ID, + .len = MLXSW_REG_SFDAT_LEN, +}; + +/* reg_sfdat_swid + * Switch partition ID. + * Access: Index + */ +MLXSW_ITEM32(reg, sfdat, swid, 0x00, 24, 8); + +/* reg_sfdat_age_time + * Aging time in seconds + * Min - 10 seconds + * Max - 1,000,000 seconds + * Default is 300 seconds. + * Access: RW + */ +MLXSW_ITEM32(reg, sfdat, age_time, 0x04, 0, 20); + +static inline void mlxsw_reg_sfdat_pack(char *payload, u32 age_time) +{ + MLXSW_REG_ZERO(sfdat, payload); + mlxsw_reg_sfdat_swid_set(payload, 0); + mlxsw_reg_sfdat_age_time_set(payload, age_time); +} + +/* SFD - Switch Filtering Database + * ------------------------------- + * The following register defines the access to the filtering database. + * The register supports querying, adding, removing and modifying the database. + * The access is optimized for bulk updates in which case more than one + * FDB record is present in the same command. + */ +#define MLXSW_REG_SFD_ID 0x200A +#define MLXSW_REG_SFD_BASE_LEN 0x10 /* base length, without records */ +#define MLXSW_REG_SFD_REC_LEN 0x10 /* record length */ +#define MLXSW_REG_SFD_REC_MAX_COUNT 64 +#define MLXSW_REG_SFD_LEN (MLXSW_REG_SFD_BASE_LEN + \ + MLXSW_REG_SFD_REC_LEN * MLXSW_REG_SFD_REC_MAX_COUNT) + +static const struct mlxsw_reg_info mlxsw_reg_sfd = { + .id = MLXSW_REG_SFD_ID, + .len = MLXSW_REG_SFD_LEN, +}; + +/* reg_sfd_swid + * Switch partition ID for queries. Reserved on Write. + * Access: Index + */ +MLXSW_ITEM32(reg, sfd, swid, 0x00, 24, 8); + +enum mlxsw_reg_sfd_op { + /* Dump entire FDB a (process according to record_locator) */ + MLXSW_REG_SFD_OP_QUERY_DUMP = 0, + /* Query records by {MAC, VID/FID} value */ + MLXSW_REG_SFD_OP_QUERY_QUERY = 1, + /* Query and clear activity. Query records by {MAC, VID/FID} value */ + MLXSW_REG_SFD_OP_QUERY_QUERY_AND_CLEAR_ACTIVITY = 2, + /* Test. Response indicates if each of the records could be + * added to the FDB. + */ + MLXSW_REG_SFD_OP_WRITE_TEST = 0, + /* Add/modify. Aged-out records cannot be added. This command removes + * the learning notification of the {MAC, VID/FID}. Response includes + * the entries that were added to the FDB. + */ + MLXSW_REG_SFD_OP_WRITE_EDIT = 1, + /* Remove record by {MAC, VID/FID}. This command also removes + * the learning notification and aged-out notifications + * of the {MAC, VID/FID}. The response provides current (pre-removal) + * entries as non-aged-out. + */ + MLXSW_REG_SFD_OP_WRITE_REMOVE = 2, + /* Remove learned notification by {MAC, VID/FID}. The response provides + * the removed learning notification. + */ + MLXSW_REG_SFD_OP_WRITE_REMOVE_NOTIFICATION = 2, +}; + +/* reg_sfd_op + * Operation. + * Access: OP + */ +MLXSW_ITEM32(reg, sfd, op, 0x04, 30, 2); + +/* reg_sfd_record_locator + * Used for querying the FDB. Use record_locator=0 to initiate the + * query. When a record is returned, a new record_locator is + * returned to be used in the subsequent query. + * Reserved for database update. + * Access: Index + */ +MLXSW_ITEM32(reg, sfd, record_locator, 0x04, 0, 30); + +/* reg_sfd_num_rec + * Request: Number of records to read/add/modify/remove + * Response: Number of records read/added/replaced/removed + * See above description for more details. + * Ranges 0..64 + * Access: RW + */ +MLXSW_ITEM32(reg, sfd, num_rec, 0x08, 0, 8); + +static inline void mlxsw_reg_sfd_pack(char *payload, enum mlxsw_reg_sfd_op op, + u32 record_locator) +{ + MLXSW_REG_ZERO(sfd, payload); + mlxsw_reg_sfd_op_set(payload, op); + mlxsw_reg_sfd_record_locator_set(payload, record_locator); +} + +/* reg_sfd_rec_swid + * Switch partition ID. + * Access: Index + */ +MLXSW_ITEM32_INDEXED(reg, sfd, rec_swid, MLXSW_REG_SFD_BASE_LEN, 24, 8, + MLXSW_REG_SFD_REC_LEN, 0x00, false); + +enum mlxsw_reg_sfd_rec_type { + MLXSW_REG_SFD_REC_TYPE_UNICAST = 0x0, +}; + +/* reg_sfd_rec_type + * FDB record type. + * Access: RW + */ +MLXSW_ITEM32_INDEXED(reg, sfd, rec_type, MLXSW_REG_SFD_BASE_LEN, 20, 4, + MLXSW_REG_SFD_REC_LEN, 0x00, false); + +enum mlxsw_reg_sfd_rec_policy { + /* Replacement disabled, aging disabled. */ + MLXSW_REG_SFD_REC_POLICY_STATIC_ENTRY = 0, + /* (mlag remote): Replacement enabled, aging disabled, + * learning notification enabled on this port. + */ + MLXSW_REG_SFD_REC_POLICY_DYNAMIC_ENTRY_MLAG = 1, + /* (ingress device): Replacement enabled, aging enabled. */ + MLXSW_REG_SFD_REC_POLICY_DYNAMIC_ENTRY_INGRESS = 3, +}; + +/* reg_sfd_rec_policy + * Policy. + * Access: RW + */ +MLXSW_ITEM32_INDEXED(reg, sfd, rec_policy, MLXSW_REG_SFD_BASE_LEN, 18, 2, + MLXSW_REG_SFD_REC_LEN, 0x00, false); + +/* reg_sfd_rec_a + * Activity. Set for new static entries. Set for static entries if a frame SMAC + * lookup hits on the entry. + * To clear the a bit, use "query and clear activity" op. + * Access: RO + */ +MLXSW_ITEM32_INDEXED(reg, sfd, rec_a, MLXSW_REG_SFD_BASE_LEN, 16, 1, + MLXSW_REG_SFD_REC_LEN, 0x00, false); + +/* reg_sfd_rec_mac + * MAC address. + * Access: Index + */ +MLXSW_ITEM_BUF_INDEXED(reg, sfd, rec_mac, MLXSW_REG_SFD_BASE_LEN, 6, + MLXSW_REG_SFD_REC_LEN, 0x02); + +enum mlxsw_reg_sfd_rec_action { + /* forward */ + MLXSW_REG_SFD_REC_ACTION_NOP = 0, + /* forward and trap, trap_id is FDB_TRAP */ + MLXSW_REG_SFD_REC_ACTION_MIRROR_TO_CPU = 1, + /* trap and do not forward, trap_id is FDB_TRAP */ + MLXSW_REG_SFD_REC_ACTION_TRAP = 3, + MLXSW_REG_SFD_REC_ACTION_DISCARD_ERROR = 15, +}; + +/* reg_sfd_rec_action + * Action to apply on the packet. + * Note: Dynamic entries can only be configured with NOP action. + * Access: RW + */ +MLXSW_ITEM32_INDEXED(reg, sfd, rec_action, MLXSW_REG_SFD_BASE_LEN, 28, 4, + MLXSW_REG_SFD_REC_LEN, 0x0C, false); + +/* reg_sfd_uc_sub_port + * VEPA channel on local port. + * Valid only if local port is a non-stacking port. Must be 0 if multichannel + * VEPA is not enabled. + * Access: RW + */ +MLXSW_ITEM32_INDEXED(reg, sfd, uc_sub_port, MLXSW_REG_SFD_BASE_LEN, 16, 8, + MLXSW_REG_SFD_REC_LEN, 0x08, false); + +/* reg_sfd_uc_fid_vid + * Filtering ID or VLAN ID + * For SwitchX and SwitchX-2: + * - Dynamic entries (policy 2,3) use FID + * - Static entries (policy 0) use VID + * - When independent learning is configured, VID=FID + * For Spectrum: use FID for both Dynamic and Static entries. + * VID should not be used. + * Access: Index + */ +MLXSW_ITEM32_INDEXED(reg, sfd, uc_fid_vid, MLXSW_REG_SFD_BASE_LEN, 0, 16, + MLXSW_REG_SFD_REC_LEN, 0x08, false); + +/* reg_sfd_uc_system_port + * Unique port identifier for the final destination of the packet. + * Access: RW + */ +MLXSW_ITEM32_INDEXED(reg, sfd, uc_system_port, MLXSW_REG_SFD_BASE_LEN, 0, 16, + MLXSW_REG_SFD_REC_LEN, 0x0C, false); + +static inline void mlxsw_reg_sfd_uc_pack(char *payload, int rec_index, + enum mlxsw_reg_sfd_rec_policy policy, + const char *mac, u16 vid, + enum mlxsw_reg_sfd_rec_action action, + u8 local_port) +{ + u8 num_rec = mlxsw_reg_sfd_num_rec_get(payload); + + if (rec_index >= num_rec) + mlxsw_reg_sfd_num_rec_set(payload, rec_index + 1); + mlxsw_reg_sfd_rec_swid_set(payload, rec_index, 0); + mlxsw_reg_sfd_rec_type_set(payload, rec_index, + MLXSW_REG_SFD_REC_TYPE_UNICAST); + mlxsw_reg_sfd_rec_policy_set(payload, rec_index, policy); + mlxsw_reg_sfd_rec_mac_memcpy_to(payload, rec_index, mac); + mlxsw_reg_sfd_uc_sub_port_set(payload, rec_index, 0); + mlxsw_reg_sfd_uc_fid_vid_set(payload, rec_index, vid); + mlxsw_reg_sfd_rec_action_set(payload, rec_index, action); + mlxsw_reg_sfd_uc_system_port_set(payload, rec_index, local_port); +} + +static inline void mlxsw_reg_sfd_uc_unpack(char *payload, int rec_index, + char *mac, u16 *p_vid, + u8 *p_local_port) +{ + mlxsw_reg_sfd_rec_mac_memcpy_from(payload, rec_index, mac); + *p_vid = mlxsw_reg_sfd_uc_fid_vid_get(payload, rec_index); + *p_local_port = mlxsw_reg_sfd_uc_system_port_get(payload, rec_index); +} + +/* SFN - Switch FDB Notification Register + * ------------------------------------------- + * The switch provides notifications on newly learned FDB entries and + * aged out entries. The notifications can be polled by software. + */ +#define MLXSW_REG_SFN_ID 0x200B +#define MLXSW_REG_SFN_BASE_LEN 0x10 /* base length, without records */ +#define MLXSW_REG_SFN_REC_LEN 0x10 /* record length */ +#define MLXSW_REG_SFN_REC_MAX_COUNT 64 +#define MLXSW_REG_SFN_LEN (MLXSW_REG_SFN_BASE_LEN + \ + MLXSW_REG_SFN_REC_LEN * MLXSW_REG_SFN_REC_MAX_COUNT) + +static const struct mlxsw_reg_info mlxsw_reg_sfn = { + .id = MLXSW_REG_SFN_ID, + .len = MLXSW_REG_SFN_LEN, +}; + +/* reg_sfn_swid + * Switch partition ID. + * Access: Index + */ +MLXSW_ITEM32(reg, sfn, swid, 0x00, 24, 8); + +/* reg_sfn_num_rec + * Request: Number of learned notifications and aged-out notification + * records requested. + * Response: Number of notification records returned (must be smaller + * than or equal to the value requested) + * Ranges 0..64 + * Access: OP + */ +MLXSW_ITEM32(reg, sfn, num_rec, 0x04, 0, 8); + +static inline void mlxsw_reg_sfn_pack(char *payload) +{ + MLXSW_REG_ZERO(sfn, payload); + mlxsw_reg_sfn_swid_set(payload, 0); + mlxsw_reg_sfn_num_rec_set(payload, MLXSW_REG_SFN_REC_MAX_COUNT); +} + +/* reg_sfn_rec_swid + * Switch partition ID. + * Access: RO + */ +MLXSW_ITEM32_INDEXED(reg, sfn, rec_swid, MLXSW_REG_SFN_BASE_LEN, 24, 8, + MLXSW_REG_SFN_REC_LEN, 0x00, false); + +enum mlxsw_reg_sfn_rec_type { + /* MAC addresses learned on a regular port. */ + MLXSW_REG_SFN_REC_TYPE_LEARNED_MAC = 0x5, + /* Aged-out MAC address on a regular port */ + MLXSW_REG_SFN_REC_TYPE_AGED_OUT_MAC = 0x7, +}; + +/* reg_sfn_rec_type + * Notification record type. + * Access: RO + */ +MLXSW_ITEM32_INDEXED(reg, sfn, rec_type, MLXSW_REG_SFN_BASE_LEN, 20, 4, + MLXSW_REG_SFN_REC_LEN, 0x00, false); + +/* reg_sfn_rec_mac + * MAC address. + * Access: RO + */ +MLXSW_ITEM_BUF_INDEXED(reg, sfn, rec_mac, MLXSW_REG_SFN_BASE_LEN, 6, + MLXSW_REG_SFN_REC_LEN, 0x02); + +/* reg_sfn_mac_sub_port + * VEPA channel on the local port. + * 0 if multichannel VEPA is not enabled. + * Access: RO + */ +MLXSW_ITEM32_INDEXED(reg, sfn, mac_sub_port, MLXSW_REG_SFN_BASE_LEN, 16, 8, + MLXSW_REG_SFN_REC_LEN, 0x08, false); + +/* reg_sfn_mac_fid + * Filtering identifier. + * Access: RO + */ +MLXSW_ITEM32_INDEXED(reg, sfn, mac_fid, MLXSW_REG_SFN_BASE_LEN, 0, 16, + MLXSW_REG_SFN_REC_LEN, 0x08, false); + +/* reg_sfn_mac_system_port + * Unique port identifier for the final destination of the packet. + * Access: RO + */ +MLXSW_ITEM32_INDEXED(reg, sfn, mac_system_port, MLXSW_REG_SFN_BASE_LEN, 0, 16, + MLXSW_REG_SFN_REC_LEN, 0x0C, false); + +static inline void mlxsw_reg_sfn_mac_unpack(char *payload, int rec_index, + char *mac, u16 *p_vid, + u8 *p_local_port) +{ + mlxsw_reg_sfn_rec_mac_memcpy_from(payload, rec_index, mac); + *p_vid = mlxsw_reg_sfn_mac_fid_get(payload, rec_index); + *p_local_port = mlxsw_reg_sfn_mac_system_port_get(payload, rec_index); +} + /* SPMS - Switch Port MSTP/RSTP State Register * ------------------------------------------- * Configures the spanning tree state of a physical port. */ -#define MLXSW_REG_SPMS_ID 0x200d +#define MLXSW_REG_SPMS_ID 0x200D #define MLXSW_REG_SPMS_LEN 0x404 static const struct mlxsw_reg_info mlxsw_reg_spms = { @@ -243,20 +540,166 @@ enum mlxsw_reg_spms_state { */ MLXSW_ITEM_BIT_ARRAY(reg, spms, state, 0x04, 0x400, 2); -static inline void mlxsw_reg_spms_pack(char *payload, u8 local_port, u16 vid, - enum mlxsw_reg_spms_state state) +static inline void mlxsw_reg_spms_pack(char *payload, u8 local_port) { MLXSW_REG_ZERO(spms, payload); mlxsw_reg_spms_local_port_set(payload, local_port); +} + +static inline void mlxsw_reg_spms_vid_pack(char *payload, u16 vid, + enum mlxsw_reg_spms_state state) +{ mlxsw_reg_spms_state_set(payload, vid, state); } +/* SPVID - Switch Port VID + * ----------------------- + * The switch port VID configures the default VID for a port. + */ +#define MLXSW_REG_SPVID_ID 0x200E +#define MLXSW_REG_SPVID_LEN 0x08 + +static const struct mlxsw_reg_info mlxsw_reg_spvid = { + .id = MLXSW_REG_SPVID_ID, + .len = MLXSW_REG_SPVID_LEN, +}; + +/* reg_spvid_local_port + * Local port number. + * Access: Index + */ +MLXSW_ITEM32(reg, spvid, local_port, 0x00, 16, 8); + +/* reg_spvid_sub_port + * Virtual port within the physical port. + * Should be set to 0 when virtual ports are not enabled on the port. + * Access: Index + */ +MLXSW_ITEM32(reg, spvid, sub_port, 0x00, 8, 8); + +/* reg_spvid_pvid + * Port default VID + * Access: RW + */ +MLXSW_ITEM32(reg, spvid, pvid, 0x04, 0, 12); + +static inline void mlxsw_reg_spvid_pack(char *payload, u8 local_port, u16 pvid) +{ + MLXSW_REG_ZERO(spvid, payload); + mlxsw_reg_spvid_local_port_set(payload, local_port); + mlxsw_reg_spvid_pvid_set(payload, pvid); +} + +/* SPVM - Switch Port VLAN Membership + * ---------------------------------- + * The Switch Port VLAN Membership register configures the VLAN membership + * of a port in a VLAN denoted by VID. VLAN membership is managed per + * virtual port. The register can be used to add and remove VID(s) from a port. + */ +#define MLXSW_REG_SPVM_ID 0x200F +#define MLXSW_REG_SPVM_BASE_LEN 0x04 /* base length, without records */ +#define MLXSW_REG_SPVM_REC_LEN 0x04 /* record length */ +#define MLXSW_REG_SPVM_REC_MAX_COUNT 256 +#define MLXSW_REG_SPVM_LEN (MLXSW_REG_SPVM_BASE_LEN + \ + MLXSW_REG_SPVM_REC_LEN * MLXSW_REG_SPVM_REC_MAX_COUNT) + +static const struct mlxsw_reg_info mlxsw_reg_spvm = { + .id = MLXSW_REG_SPVM_ID, + .len = MLXSW_REG_SPVM_LEN, +}; + +/* reg_spvm_pt + * Priority tagged. If this bit is set, packets forwarded to the port with + * untagged VLAN membership (u bit is set) will be tagged with priority tag + * (VID=0) + * Access: RW + */ +MLXSW_ITEM32(reg, spvm, pt, 0x00, 31, 1); + +/* reg_spvm_pte + * Priority Tagged Update Enable. On Write operations, if this bit is cleared, + * the pt bit will NOT be updated. To update the pt bit, pte must be set. + * Access: WO + */ +MLXSW_ITEM32(reg, spvm, pte, 0x00, 30, 1); + +/* reg_spvm_local_port + * Local port number. + * Access: Index + */ +MLXSW_ITEM32(reg, spvm, local_port, 0x00, 16, 8); + +/* reg_spvm_sub_port + * Virtual port within the physical port. + * Should be set to 0 when virtual ports are not enabled on the port. + * Access: Index + */ +MLXSW_ITEM32(reg, spvm, sub_port, 0x00, 8, 8); + +/* reg_spvm_num_rec + * Number of records to update. Each record contains: i, e, u, vid. + * Access: OP + */ +MLXSW_ITEM32(reg, spvm, num_rec, 0x00, 0, 8); + +/* reg_spvm_rec_i + * Ingress membership in VLAN ID. + * Access: Index + */ +MLXSW_ITEM32_INDEXED(reg, spvm, rec_i, + MLXSW_REG_SPVM_BASE_LEN, 14, 1, + MLXSW_REG_SPVM_REC_LEN, 0, false); + +/* reg_spvm_rec_e + * Egress membership in VLAN ID. + * Access: Index + */ +MLXSW_ITEM32_INDEXED(reg, spvm, rec_e, + MLXSW_REG_SPVM_BASE_LEN, 13, 1, + MLXSW_REG_SPVM_REC_LEN, 0, false); + +/* reg_spvm_rec_u + * Untagged - port is an untagged member - egress transmission uses untagged + * frames on VID<n> + * Access: Index + */ +MLXSW_ITEM32_INDEXED(reg, spvm, rec_u, + MLXSW_REG_SPVM_BASE_LEN, 12, 1, + MLXSW_REG_SPVM_REC_LEN, 0, false); + +/* reg_spvm_rec_vid + * Egress membership in VLAN ID. + * Access: Index + */ +MLXSW_ITEM32_INDEXED(reg, spvm, rec_vid, + MLXSW_REG_SPVM_BASE_LEN, 0, 12, + MLXSW_REG_SPVM_REC_LEN, 0, false); + +static inline void mlxsw_reg_spvm_pack(char *payload, u8 local_port, + u16 vid_begin, u16 vid_end, + bool is_member, bool untagged) +{ + int size = vid_end - vid_begin + 1; + int i; + + MLXSW_REG_ZERO(spvm, payload); + mlxsw_reg_spvm_local_port_set(payload, local_port); + mlxsw_reg_spvm_num_rec_set(payload, size); + + for (i = 0; i < size; i++) { + mlxsw_reg_spvm_rec_i_set(payload, i, is_member); + mlxsw_reg_spvm_rec_e_set(payload, i, is_member); + mlxsw_reg_spvm_rec_u_set(payload, i, untagged); + mlxsw_reg_spvm_rec_vid_set(payload, i, vid_begin + i); + } +} + /* SFGC - Switch Flooding Group Configuration * ------------------------------------------ * The following register controls the association of flooding tables and MIDs * to packet types used for flooding. */ -#define MLXSW_REG_SFGC_ID 0x2011 +#define MLXSW_REG_SFGC_ID 0x2011 #define MLXSW_REG_SFGC_LEN 0x10 static const struct mlxsw_reg_info mlxsw_reg_sfgc = { @@ -265,13 +708,15 @@ static const struct mlxsw_reg_info mlxsw_reg_sfgc = { }; enum mlxsw_reg_sfgc_type { - MLXSW_REG_SFGC_TYPE_BROADCAST = 0, - MLXSW_REG_SFGC_TYPE_UNKNOWN_UNICAST = 1, - MLXSW_REG_SFGC_TYPE_UNREGISTERED_MULTICAST_IPV4 = 2, - MLXSW_REG_SFGC_TYPE_UNREGISTERED_MULTICAST_IPV6 = 3, - MLXSW_REG_SFGC_TYPE_UNREGISTERED_MULTICAST_NON_IP = 5, - MLXSW_REG_SFGC_TYPE_IPV4_LINK_LOCAL = 6, - MLXSW_REG_SFGC_TYPE_IPV6_ALL_HOST = 7, + MLXSW_REG_SFGC_TYPE_BROADCAST, + MLXSW_REG_SFGC_TYPE_UNKNOWN_UNICAST, + MLXSW_REG_SFGC_TYPE_UNREGISTERED_MULTICAST_IPV4, + MLXSW_REG_SFGC_TYPE_UNREGISTERED_MULTICAST_IPV6, + MLXSW_REG_SFGC_TYPE_RESERVED, + MLXSW_REG_SFGC_TYPE_UNREGISTERED_MULTICAST_NON_IP, + MLXSW_REG_SFGC_TYPE_IPV4_LINK_LOCAL, + MLXSW_REG_SFGC_TYPE_IPV6_ALL_HOST, + MLXSW_REG_SFGC_TYPE_MAX, }; /* reg_sfgc_type @@ -408,7 +853,7 @@ static inline void mlxsw_reg_sftr_pack(char *payload, unsigned int flood_table, unsigned int index, enum mlxsw_flood_table_type table_type, - unsigned int range) + unsigned int range, u8 port, bool set) { MLXSW_REG_ZERO(sftr, payload); mlxsw_reg_sftr_swid_set(payload, 0); @@ -416,8 +861,8 @@ static inline void mlxsw_reg_sftr_pack(char *payload, mlxsw_reg_sftr_index_set(payload, index); mlxsw_reg_sftr_table_type_set(payload, table_type); mlxsw_reg_sftr_range_set(payload, range); - mlxsw_reg_sftr_port_set(payload, MLXSW_PORT_CPU_PORT, 1); - mlxsw_reg_sftr_port_mask_set(payload, MLXSW_PORT_CPU_PORT, 1); + mlxsw_reg_sftr_port_set(payload, port, set); + mlxsw_reg_sftr_port_mask_set(payload, port, 1); } /* SPMLR - Switch Port MAC Learning Register @@ -473,6 +918,285 @@ static inline void mlxsw_reg_spmlr_pack(char *payload, u8 local_port, mlxsw_reg_spmlr_learn_mode_set(payload, mode); } +/* SVFA - Switch VID to FID Allocation Register + * -------------------------------------------- + * Controls the VID to FID mapping and {Port, VID} to FID mapping for + * virtualized ports. + */ +#define MLXSW_REG_SVFA_ID 0x201C +#define MLXSW_REG_SVFA_LEN 0x10 + +static const struct mlxsw_reg_info mlxsw_reg_svfa = { + .id = MLXSW_REG_SVFA_ID, + .len = MLXSW_REG_SVFA_LEN, +}; + +/* reg_svfa_swid + * Switch partition ID. + * Access: Index + */ +MLXSW_ITEM32(reg, svfa, swid, 0x00, 24, 8); + +/* reg_svfa_local_port + * Local port number. + * Access: Index + * + * Note: Reserved for 802.1Q FIDs. + */ +MLXSW_ITEM32(reg, svfa, local_port, 0x00, 16, 8); + +enum mlxsw_reg_svfa_mt { + MLXSW_REG_SVFA_MT_VID_TO_FID, + MLXSW_REG_SVFA_MT_PORT_VID_TO_FID, +}; + +/* reg_svfa_mapping_table + * Mapping table: + * 0 - VID to FID + * 1 - {Port, VID} to FID + * Access: Index + * + * Note: Reserved for SwitchX-2. + */ +MLXSW_ITEM32(reg, svfa, mapping_table, 0x00, 8, 3); + +/* reg_svfa_v + * Valid. + * Valid if set. + * Access: RW + * + * Note: Reserved for SwitchX-2. + */ +MLXSW_ITEM32(reg, svfa, v, 0x00, 0, 1); + +/* reg_svfa_fid + * Filtering ID. + * Access: RW + */ +MLXSW_ITEM32(reg, svfa, fid, 0x04, 16, 16); + +/* reg_svfa_vid + * VLAN ID. + * Access: Index + */ +MLXSW_ITEM32(reg, svfa, vid, 0x04, 0, 12); + +/* reg_svfa_counter_set_type + * Counter set type for flow counters. + * Access: RW + * + * Note: Reserved for SwitchX-2. + */ +MLXSW_ITEM32(reg, svfa, counter_set_type, 0x08, 24, 8); + +/* reg_svfa_counter_index + * Counter index for flow counters. + * Access: RW + * + * Note: Reserved for SwitchX-2. + */ +MLXSW_ITEM32(reg, svfa, counter_index, 0x08, 0, 24); + +static inline void mlxsw_reg_svfa_pack(char *payload, u8 local_port, + enum mlxsw_reg_svfa_mt mt, bool valid, + u16 fid, u16 vid) +{ + MLXSW_REG_ZERO(svfa, payload); + local_port = mt == MLXSW_REG_SVFA_MT_VID_TO_FID ? 0 : local_port; + mlxsw_reg_svfa_swid_set(payload, 0); + mlxsw_reg_svfa_local_port_set(payload, local_port); + mlxsw_reg_svfa_mapping_table_set(payload, mt); + mlxsw_reg_svfa_v_set(payload, valid); + mlxsw_reg_svfa_fid_set(payload, fid); + mlxsw_reg_svfa_vid_set(payload, vid); +} + +/* SVPE - Switch Virtual-Port Enabling Register + * -------------------------------------------- + * Enables port virtualization. + */ +#define MLXSW_REG_SVPE_ID 0x201E +#define MLXSW_REG_SVPE_LEN 0x4 + +static const struct mlxsw_reg_info mlxsw_reg_svpe = { + .id = MLXSW_REG_SVPE_ID, + .len = MLXSW_REG_SVPE_LEN, +}; + +/* reg_svpe_local_port + * Local port number + * Access: Index + * + * Note: CPU port is not supported (uses VLAN mode only). + */ +MLXSW_ITEM32(reg, svpe, local_port, 0x00, 16, 8); + +/* reg_svpe_vp_en + * Virtual port enable. + * 0 - Disable, VLAN mode (VID to FID). + * 1 - Enable, Virtual port mode ({Port, VID} to FID). + * Access: RW + */ +MLXSW_ITEM32(reg, svpe, vp_en, 0x00, 8, 1); + +static inline void mlxsw_reg_svpe_pack(char *payload, u8 local_port, + bool enable) +{ + MLXSW_REG_ZERO(svpe, payload); + mlxsw_reg_svpe_local_port_set(payload, local_port); + mlxsw_reg_svpe_vp_en_set(payload, enable); +} + +/* SFMR - Switch FID Management Register + * ------------------------------------- + * Creates and configures FIDs. + */ +#define MLXSW_REG_SFMR_ID 0x201F +#define MLXSW_REG_SFMR_LEN 0x18 + +static const struct mlxsw_reg_info mlxsw_reg_sfmr = { + .id = MLXSW_REG_SFMR_ID, + .len = MLXSW_REG_SFMR_LEN, +}; + +enum mlxsw_reg_sfmr_op { + MLXSW_REG_SFMR_OP_CREATE_FID, + MLXSW_REG_SFMR_OP_DESTROY_FID, +}; + +/* reg_sfmr_op + * Operation. + * 0 - Create or edit FID. + * 1 - Destroy FID. + * Access: WO + */ +MLXSW_ITEM32(reg, sfmr, op, 0x00, 24, 4); + +/* reg_sfmr_fid + * Filtering ID. + * Access: Index + */ +MLXSW_ITEM32(reg, sfmr, fid, 0x00, 0, 16); + +/* reg_sfmr_fid_offset + * FID offset. + * Used to point into the flooding table selected by SFGC register if + * the table is of type FID-Offset. Otherwise, this field is reserved. + * Access: RW + */ +MLXSW_ITEM32(reg, sfmr, fid_offset, 0x08, 0, 16); + +/* reg_sfmr_vtfp + * Valid Tunnel Flood Pointer. + * If not set, then nve_tunnel_flood_ptr is reserved and considered NULL. + * Access: RW + * + * Note: Reserved for 802.1Q FIDs. + */ +MLXSW_ITEM32(reg, sfmr, vtfp, 0x0C, 31, 1); + +/* reg_sfmr_nve_tunnel_flood_ptr + * Underlay Flooding and BC Pointer. + * Used as a pointer to the first entry of the group based link lists of + * flooding or BC entries (for NVE tunnels). + * Access: RW + */ +MLXSW_ITEM32(reg, sfmr, nve_tunnel_flood_ptr, 0x0C, 0, 24); + +/* reg_sfmr_vv + * VNI Valid. + * If not set, then vni is reserved. + * Access: RW + * + * Note: Reserved for 802.1Q FIDs. + */ +MLXSW_ITEM32(reg, sfmr, vv, 0x10, 31, 1); + +/* reg_sfmr_vni + * Virtual Network Identifier. + * Access: RW + * + * Note: A given VNI can only be assigned to one FID. + */ +MLXSW_ITEM32(reg, sfmr, vni, 0x10, 0, 24); + +static inline void mlxsw_reg_sfmr_pack(char *payload, + enum mlxsw_reg_sfmr_op op, u16 fid, + u16 fid_offset) +{ + MLXSW_REG_ZERO(sfmr, payload); + mlxsw_reg_sfmr_op_set(payload, op); + mlxsw_reg_sfmr_fid_set(payload, fid); + mlxsw_reg_sfmr_fid_offset_set(payload, fid_offset); + mlxsw_reg_sfmr_vtfp_set(payload, false); + mlxsw_reg_sfmr_vv_set(payload, false); +} + +/* SPVMLR - Switch Port VLAN MAC Learning Register + * ----------------------------------------------- + * Controls the switch MAC learning policy per {Port, VID}. + */ +#define MLXSW_REG_SPVMLR_ID 0x2020 +#define MLXSW_REG_SPVMLR_BASE_LEN 0x04 /* base length, without records */ +#define MLXSW_REG_SPVMLR_REC_LEN 0x04 /* record length */ +#define MLXSW_REG_SPVMLR_REC_MAX_COUNT 256 +#define MLXSW_REG_SPVMLR_LEN (MLXSW_REG_SPVMLR_BASE_LEN + \ + MLXSW_REG_SPVMLR_REC_LEN * \ + MLXSW_REG_SPVMLR_REC_MAX_COUNT) + +static const struct mlxsw_reg_info mlxsw_reg_spvmlr = { + .id = MLXSW_REG_SPVMLR_ID, + .len = MLXSW_REG_SPVMLR_LEN, +}; + +/* reg_spvmlr_local_port + * Local ingress port. + * Access: Index + * + * Note: CPU port is not supported. + */ +MLXSW_ITEM32(reg, spvmlr, local_port, 0x00, 16, 8); + +/* reg_spvmlr_num_rec + * Number of records to update. + * Access: OP + */ +MLXSW_ITEM32(reg, spvmlr, num_rec, 0x00, 0, 8); + +/* reg_spvmlr_rec_learn_enable + * 0 - Disable learning for {Port, VID}. + * 1 - Enable learning for {Port, VID}. + * Access: RW + */ +MLXSW_ITEM32_INDEXED(reg, spvmlr, rec_learn_enable, MLXSW_REG_SPVMLR_BASE_LEN, + 31, 1, MLXSW_REG_SPVMLR_REC_LEN, 0x00, false); + +/* reg_spvmlr_rec_vid + * VLAN ID to be added/removed from port or for querying. + * Access: Index + */ +MLXSW_ITEM32_INDEXED(reg, spvmlr, rec_vid, MLXSW_REG_SPVMLR_BASE_LEN, 0, 12, + MLXSW_REG_SPVMLR_REC_LEN, 0x00, false); + +static inline void mlxsw_reg_spvmlr_pack(char *payload, u8 local_port, + u16 vid_begin, u16 vid_end, + bool learn_enable) +{ + int num_rec = vid_end - vid_begin + 1; + int i; + + WARN_ON(num_rec < 1 || num_rec > MLXSW_REG_SPVMLR_REC_MAX_COUNT); + + MLXSW_REG_ZERO(spvmlr, payload); + mlxsw_reg_spvmlr_local_port_set(payload, local_port); + mlxsw_reg_spvmlr_num_rec_set(payload, num_rec); + + for (i = 0; i < num_rec; i++) { + mlxsw_reg_spvmlr_rec_learn_enable_set(payload, i, learn_enable); + mlxsw_reg_spvmlr_rec_vid_set(payload, i, vid_begin + i); + } +} + /* PMLP - Ports Module to Local Port Register * ------------------------------------------ * Configures the assignment of modules to local ports. @@ -1008,12 +1732,88 @@ static inline void mlxsw_reg_ppcnt_pack(char *payload, u8 local_port) mlxsw_reg_ppcnt_prio_tc_set(payload, 0); } +/* PBMC - Port Buffer Management Control Register + * ---------------------------------------------- + * The PBMC register configures and retrieves the port packet buffer + * allocation for different Prios, and the Pause threshold management. + */ +#define MLXSW_REG_PBMC_ID 0x500C +#define MLXSW_REG_PBMC_LEN 0x68 + +static const struct mlxsw_reg_info mlxsw_reg_pbmc = { + .id = MLXSW_REG_PBMC_ID, + .len = MLXSW_REG_PBMC_LEN, +}; + +/* reg_pbmc_local_port + * Local port number. + * Access: Index + */ +MLXSW_ITEM32(reg, pbmc, local_port, 0x00, 16, 8); + +/* reg_pbmc_xoff_timer_value + * When device generates a pause frame, it uses this value as the pause + * timer (time for the peer port to pause in quota-512 bit time). + * Access: RW + */ +MLXSW_ITEM32(reg, pbmc, xoff_timer_value, 0x04, 16, 16); + +/* reg_pbmc_xoff_refresh + * The time before a new pause frame should be sent to refresh the pause RW + * state. Using the same units as xoff_timer_value above (in quota-512 bit + * time). + * Access: RW + */ +MLXSW_ITEM32(reg, pbmc, xoff_refresh, 0x04, 0, 16); + +/* reg_pbmc_buf_lossy + * The field indicates if the buffer is lossy. + * 0 - Lossless + * 1 - Lossy + * Access: RW + */ +MLXSW_ITEM32_INDEXED(reg, pbmc, buf_lossy, 0x0C, 25, 1, 0x08, 0x00, false); + +/* reg_pbmc_buf_epsb + * Eligible for Port Shared buffer. + * If epsb is set, packets assigned to buffer are allowed to insert the port + * shared buffer. + * When buf_lossy is MLXSW_REG_PBMC_LOSSY_LOSSY this field is reserved. + * Access: RW + */ +MLXSW_ITEM32_INDEXED(reg, pbmc, buf_epsb, 0x0C, 24, 1, 0x08, 0x00, false); + +/* reg_pbmc_buf_size + * The part of the packet buffer array is allocated for the specific buffer. + * Units are represented in cells. + * Access: RW + */ +MLXSW_ITEM32_INDEXED(reg, pbmc, buf_size, 0x0C, 0, 16, 0x08, 0x00, false); + +static inline void mlxsw_reg_pbmc_pack(char *payload, u8 local_port, + u16 xoff_timer_value, u16 xoff_refresh) +{ + MLXSW_REG_ZERO(pbmc, payload); + mlxsw_reg_pbmc_local_port_set(payload, local_port); + mlxsw_reg_pbmc_xoff_timer_value_set(payload, xoff_timer_value); + mlxsw_reg_pbmc_xoff_refresh_set(payload, xoff_refresh); +} + +static inline void mlxsw_reg_pbmc_lossy_buffer_pack(char *payload, + int buf_index, + u16 size) +{ + mlxsw_reg_pbmc_buf_lossy_set(payload, buf_index, 1); + mlxsw_reg_pbmc_buf_epsb_set(payload, buf_index, 0); + mlxsw_reg_pbmc_buf_size_set(payload, buf_index, size); +} + /* PSPA - Port Switch Partition Allocation * --------------------------------------- * Controls the association of a port with a switch partition and enables * configuring ports as stacking ports. */ -#define MLXSW_REG_PSPA_ID 0x500d +#define MLXSW_REG_PSPA_ID 0x500D #define MLXSW_REG_PSPA_LEN 0x8 static const struct mlxsw_reg_info mlxsw_reg_pspa = { @@ -1074,8 +1874,11 @@ MLXSW_ITEM32(reg, htgt, swid, 0x00, 24, 8); */ MLXSW_ITEM32(reg, htgt, type, 0x00, 8, 4); -#define MLXSW_REG_HTGT_TRAP_GROUP_EMAD 0x0 -#define MLXSW_REG_HTGT_TRAP_GROUP_RX 0x1 +enum mlxsw_reg_htgt_trap_group { + MLXSW_REG_HTGT_TRAP_GROUP_EMAD, + MLXSW_REG_HTGT_TRAP_GROUP_RX, + MLXSW_REG_HTGT_TRAP_GROUP_CTRL, +}; /* reg_htgt_trap_group * Trap group number. User defined number specifying which trap groups @@ -1142,6 +1945,7 @@ MLXSW_ITEM32(reg, htgt, local_path_cpu_tclass, 0x10, 16, 6); #define MLXSW_REG_HTGT_LOCAL_PATH_RDQ_EMAD 0x15 #define MLXSW_REG_HTGT_LOCAL_PATH_RDQ_RX 0x14 +#define MLXSW_REG_HTGT_LOCAL_PATH_RDQ_CTRL 0x13 /* reg_htgt_local_path_rdq * Receive descriptor queue (RDQ) to use for the trap group. @@ -1149,21 +1953,29 @@ MLXSW_ITEM32(reg, htgt, local_path_cpu_tclass, 0x10, 16, 6); */ MLXSW_ITEM32(reg, htgt, local_path_rdq, 0x10, 0, 6); -static inline void mlxsw_reg_htgt_pack(char *payload, u8 trap_group) +static inline void mlxsw_reg_htgt_pack(char *payload, + enum mlxsw_reg_htgt_trap_group group) { u8 swid, rdq; MLXSW_REG_ZERO(htgt, payload); - if (MLXSW_REG_HTGT_TRAP_GROUP_EMAD == trap_group) { + switch (group) { + case MLXSW_REG_HTGT_TRAP_GROUP_EMAD: swid = MLXSW_PORT_SWID_ALL_SWIDS; rdq = MLXSW_REG_HTGT_LOCAL_PATH_RDQ_EMAD; - } else { + break; + case MLXSW_REG_HTGT_TRAP_GROUP_RX: swid = 0; rdq = MLXSW_REG_HTGT_LOCAL_PATH_RDQ_RX; + break; + case MLXSW_REG_HTGT_TRAP_GROUP_CTRL: + swid = 0; + rdq = MLXSW_REG_HTGT_LOCAL_PATH_RDQ_CTRL; + break; } mlxsw_reg_htgt_swid_set(payload, swid); mlxsw_reg_htgt_type_set(payload, MLXSW_REG_HTGT_PATH_TYPE_LOCAL); - mlxsw_reg_htgt_trap_group_set(payload, trap_group); + mlxsw_reg_htgt_trap_group_set(payload, group); mlxsw_reg_htgt_pide_set(payload, MLXSW_REG_HTGT_POLICER_DISABLE); mlxsw_reg_htgt_pid_set(payload, 0); mlxsw_reg_htgt_mirror_action_set(payload, MLXSW_REG_HTGT_TRAP_TO_CPU); @@ -1254,17 +2066,290 @@ enum { */ MLXSW_ITEM32(reg, hpkt, ctrl, 0x04, 16, 2); -static inline void mlxsw_reg_hpkt_pack(char *payload, u8 action, - u8 trap_group, u16 trap_id) +static inline void mlxsw_reg_hpkt_pack(char *payload, u8 action, u16 trap_id) { + enum mlxsw_reg_htgt_trap_group trap_group; + MLXSW_REG_ZERO(hpkt, payload); mlxsw_reg_hpkt_ack_set(payload, MLXSW_REG_HPKT_ACK_NOT_REQUIRED); mlxsw_reg_hpkt_action_set(payload, action); + switch (trap_id) { + case MLXSW_TRAP_ID_ETHEMAD: + case MLXSW_TRAP_ID_PUDE: + trap_group = MLXSW_REG_HTGT_TRAP_GROUP_EMAD; + break; + default: + trap_group = MLXSW_REG_HTGT_TRAP_GROUP_RX; + break; + } mlxsw_reg_hpkt_trap_group_set(payload, trap_group); mlxsw_reg_hpkt_trap_id_set(payload, trap_id); mlxsw_reg_hpkt_ctrl_set(payload, MLXSW_REG_HPKT_CTRL_PACKET_DEFAULT); } +/* SBPR - Shared Buffer Pools Register + * ----------------------------------- + * The SBPR configures and retrieves the shared buffer pools and configuration. + */ +#define MLXSW_REG_SBPR_ID 0xB001 +#define MLXSW_REG_SBPR_LEN 0x14 + +static const struct mlxsw_reg_info mlxsw_reg_sbpr = { + .id = MLXSW_REG_SBPR_ID, + .len = MLXSW_REG_SBPR_LEN, +}; + +enum mlxsw_reg_sbpr_dir { + MLXSW_REG_SBPR_DIR_INGRESS, + MLXSW_REG_SBPR_DIR_EGRESS, +}; + +/* reg_sbpr_dir + * Direction. + * Access: Index + */ +MLXSW_ITEM32(reg, sbpr, dir, 0x00, 24, 2); + +/* reg_sbpr_pool + * Pool index. + * Access: Index + */ +MLXSW_ITEM32(reg, sbpr, pool, 0x00, 0, 4); + +/* reg_sbpr_size + * Pool size in buffer cells. + * Access: RW + */ +MLXSW_ITEM32(reg, sbpr, size, 0x04, 0, 24); + +enum mlxsw_reg_sbpr_mode { + MLXSW_REG_SBPR_MODE_STATIC, + MLXSW_REG_SBPR_MODE_DYNAMIC, +}; + +/* reg_sbpr_mode + * Pool quota calculation mode. + * Access: RW + */ +MLXSW_ITEM32(reg, sbpr, mode, 0x08, 0, 4); + +static inline void mlxsw_reg_sbpr_pack(char *payload, u8 pool, + enum mlxsw_reg_sbpr_dir dir, + enum mlxsw_reg_sbpr_mode mode, u32 size) +{ + MLXSW_REG_ZERO(sbpr, payload); + mlxsw_reg_sbpr_pool_set(payload, pool); + mlxsw_reg_sbpr_dir_set(payload, dir); + mlxsw_reg_sbpr_mode_set(payload, mode); + mlxsw_reg_sbpr_size_set(payload, size); +} + +/* SBCM - Shared Buffer Class Management Register + * ---------------------------------------------- + * The SBCM register configures and retrieves the shared buffer allocation + * and configuration according to Port-PG, including the binding to pool + * and definition of the associated quota. + */ +#define MLXSW_REG_SBCM_ID 0xB002 +#define MLXSW_REG_SBCM_LEN 0x28 + +static const struct mlxsw_reg_info mlxsw_reg_sbcm = { + .id = MLXSW_REG_SBCM_ID, + .len = MLXSW_REG_SBCM_LEN, +}; + +/* reg_sbcm_local_port + * Local port number. + * For Ingress: excludes CPU port and Router port + * For Egress: excludes IP Router + * Access: Index + */ +MLXSW_ITEM32(reg, sbcm, local_port, 0x00, 16, 8); + +/* reg_sbcm_pg_buff + * PG buffer - Port PG (dir=ingress) / traffic class (dir=egress) + * For PG buffer: range is 0..cap_max_pg_buffers - 1 + * For traffic class: range is 0..cap_max_tclass - 1 + * Note that when traffic class is in MC aware mode then the traffic + * classes which are MC aware cannot be configured. + * Access: Index + */ +MLXSW_ITEM32(reg, sbcm, pg_buff, 0x00, 8, 6); + +enum mlxsw_reg_sbcm_dir { + MLXSW_REG_SBCM_DIR_INGRESS, + MLXSW_REG_SBCM_DIR_EGRESS, +}; + +/* reg_sbcm_dir + * Direction. + * Access: Index + */ +MLXSW_ITEM32(reg, sbcm, dir, 0x00, 0, 2); + +/* reg_sbcm_min_buff + * Minimum buffer size for the limiter, in cells. + * Access: RW + */ +MLXSW_ITEM32(reg, sbcm, min_buff, 0x18, 0, 24); + +/* reg_sbcm_max_buff + * When the pool associated to the port-pg/tclass is configured to + * static, Maximum buffer size for the limiter configured in cells. + * When the pool associated to the port-pg/tclass is configured to + * dynamic, the max_buff holds the "alpha" parameter, supporting + * the following values: + * 0: 0 + * i: (1/128)*2^(i-1), for i=1..14 + * 0xFF: Infinity + * Access: RW + */ +MLXSW_ITEM32(reg, sbcm, max_buff, 0x1C, 0, 24); + +/* reg_sbcm_pool + * Association of the port-priority to a pool. + * Access: RW + */ +MLXSW_ITEM32(reg, sbcm, pool, 0x24, 0, 4); + +static inline void mlxsw_reg_sbcm_pack(char *payload, u8 local_port, u8 pg_buff, + enum mlxsw_reg_sbcm_dir dir, + u32 min_buff, u32 max_buff, u8 pool) +{ + MLXSW_REG_ZERO(sbcm, payload); + mlxsw_reg_sbcm_local_port_set(payload, local_port); + mlxsw_reg_sbcm_pg_buff_set(payload, pg_buff); + mlxsw_reg_sbcm_dir_set(payload, dir); + mlxsw_reg_sbcm_min_buff_set(payload, min_buff); + mlxsw_reg_sbcm_max_buff_set(payload, max_buff); + mlxsw_reg_sbcm_pool_set(payload, pool); +} + +/* SBPM - Shared Buffer Class Management Register + * ---------------------------------------------- + * The SBPM register configures and retrieves the shared buffer allocation + * and configuration according to Port-Pool, including the definition + * of the associated quota. + */ +#define MLXSW_REG_SBPM_ID 0xB003 +#define MLXSW_REG_SBPM_LEN 0x28 + +static const struct mlxsw_reg_info mlxsw_reg_sbpm = { + .id = MLXSW_REG_SBPM_ID, + .len = MLXSW_REG_SBPM_LEN, +}; + +/* reg_sbpm_local_port + * Local port number. + * For Ingress: excludes CPU port and Router port + * For Egress: excludes IP Router + * Access: Index + */ +MLXSW_ITEM32(reg, sbpm, local_port, 0x00, 16, 8); + +/* reg_sbpm_pool + * The pool associated to quota counting on the local_port. + * Access: Index + */ +MLXSW_ITEM32(reg, sbpm, pool, 0x00, 8, 4); + +enum mlxsw_reg_sbpm_dir { + MLXSW_REG_SBPM_DIR_INGRESS, + MLXSW_REG_SBPM_DIR_EGRESS, +}; + +/* reg_sbpm_dir + * Direction. + * Access: Index + */ +MLXSW_ITEM32(reg, sbpm, dir, 0x00, 0, 2); + +/* reg_sbpm_min_buff + * Minimum buffer size for the limiter, in cells. + * Access: RW + */ +MLXSW_ITEM32(reg, sbpm, min_buff, 0x18, 0, 24); + +/* reg_sbpm_max_buff + * When the pool associated to the port-pg/tclass is configured to + * static, Maximum buffer size for the limiter configured in cells. + * When the pool associated to the port-pg/tclass is configured to + * dynamic, the max_buff holds the "alpha" parameter, supporting + * the following values: + * 0: 0 + * i: (1/128)*2^(i-1), for i=1..14 + * 0xFF: Infinity + * Access: RW + */ +MLXSW_ITEM32(reg, sbpm, max_buff, 0x1C, 0, 24); + +static inline void mlxsw_reg_sbpm_pack(char *payload, u8 local_port, u8 pool, + enum mlxsw_reg_sbpm_dir dir, + u32 min_buff, u32 max_buff) +{ + MLXSW_REG_ZERO(sbpm, payload); + mlxsw_reg_sbpm_local_port_set(payload, local_port); + mlxsw_reg_sbpm_pool_set(payload, pool); + mlxsw_reg_sbpm_dir_set(payload, dir); + mlxsw_reg_sbpm_min_buff_set(payload, min_buff); + mlxsw_reg_sbpm_max_buff_set(payload, max_buff); +} + +/* SBMM - Shared Buffer Multicast Management Register + * -------------------------------------------------- + * The SBMM register configures and retrieves the shared buffer allocation + * and configuration for MC packets according to Switch-Priority, including + * the binding to pool and definition of the associated quota. + */ +#define MLXSW_REG_SBMM_ID 0xB004 +#define MLXSW_REG_SBMM_LEN 0x28 + +static const struct mlxsw_reg_info mlxsw_reg_sbmm = { + .id = MLXSW_REG_SBMM_ID, + .len = MLXSW_REG_SBMM_LEN, +}; + +/* reg_sbmm_prio + * Switch Priority. + * Access: Index + */ +MLXSW_ITEM32(reg, sbmm, prio, 0x00, 8, 4); + +/* reg_sbmm_min_buff + * Minimum buffer size for the limiter, in cells. + * Access: RW + */ +MLXSW_ITEM32(reg, sbmm, min_buff, 0x18, 0, 24); + +/* reg_sbmm_max_buff + * When the pool associated to the port-pg/tclass is configured to + * static, Maximum buffer size for the limiter configured in cells. + * When the pool associated to the port-pg/tclass is configured to + * dynamic, the max_buff holds the "alpha" parameter, supporting + * the following values: + * 0: 0 + * i: (1/128)*2^(i-1), for i=1..14 + * 0xFF: Infinity + * Access: RW + */ +MLXSW_ITEM32(reg, sbmm, max_buff, 0x1C, 0, 24); + +/* reg_sbmm_pool + * Association of the port-priority to a pool. + * Access: RW + */ +MLXSW_ITEM32(reg, sbmm, pool, 0x24, 0, 4); + +static inline void mlxsw_reg_sbmm_pack(char *payload, u8 prio, u32 min_buff, + u32 max_buff, u8 pool) +{ + MLXSW_REG_ZERO(sbmm, payload); + mlxsw_reg_sbmm_prio_set(payload, prio); + mlxsw_reg_sbmm_min_buff_set(payload, min_buff); + mlxsw_reg_sbmm_max_buff_set(payload, max_buff); + mlxsw_reg_sbmm_pool_set(payload, pool); +} + static inline const char *mlxsw_reg_id_str(u16 reg_id) { switch (reg_id) { @@ -1272,18 +2357,34 @@ static inline const char *mlxsw_reg_id_str(u16 reg_id) return "SGCR"; case MLXSW_REG_SPAD_ID: return "SPAD"; - case MLXSW_REG_SMID_ID: - return "SMID"; case MLXSW_REG_SSPR_ID: return "SSPR"; + case MLXSW_REG_SFDAT_ID: + return "SFDAT"; + case MLXSW_REG_SFD_ID: + return "SFD"; + case MLXSW_REG_SFN_ID: + return "SFN"; case MLXSW_REG_SPMS_ID: return "SPMS"; + case MLXSW_REG_SPVID_ID: + return "SPVID"; + case MLXSW_REG_SPVM_ID: + return "SPVM"; case MLXSW_REG_SFGC_ID: return "SFGC"; case MLXSW_REG_SFTR_ID: return "SFTR"; case MLXSW_REG_SPMLR_ID: return "SPMLR"; + case MLXSW_REG_SVFA_ID: + return "SVFA"; + case MLXSW_REG_SVPE_ID: + return "SVPE"; + case MLXSW_REG_SFMR_ID: + return "SFMR"; + case MLXSW_REG_SPVMLR_ID: + return "SPVMLR"; case MLXSW_REG_PMLP_ID: return "PMLP"; case MLXSW_REG_PMTU_ID: @@ -1296,12 +2397,22 @@ static inline const char *mlxsw_reg_id_str(u16 reg_id) return "PAOS"; case MLXSW_REG_PPCNT_ID: return "PPCNT"; + case MLXSW_REG_PBMC_ID: + return "PBMC"; case MLXSW_REG_PSPA_ID: return "PSPA"; case MLXSW_REG_HTGT_ID: return "HTGT"; case MLXSW_REG_HPKT_ID: return "HPKT"; + case MLXSW_REG_SBPR_ID: + return "SBPR"; + case MLXSW_REG_SBCM_ID: + return "SBCM"; + case MLXSW_REG_SBPM_ID: + return "SBPM"; + case MLXSW_REG_SBMM_ID: + return "SBMM"; default: return "*UNKNOWN*"; } diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c new file mode 100644 index 000000000..3be4a2355 --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c @@ -0,0 +1,1949 @@ +/* + * drivers/net/ethernet/mellanox/mlxsw/spectrum.c + * Copyright (c) 2015 Mellanox Technologies. All rights reserved. + * Copyright (c) 2015 Jiri Pirko <jiri@mellanox.com> + * Copyright (c) 2015 Ido Schimmel <idosch@mellanox.com> + * Copyright (c) 2015 Elad Raz <eladr@mellanox.com> + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. 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. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * 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. + */ + +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/types.h> +#include <linux/netdevice.h> +#include <linux/etherdevice.h> +#include <linux/ethtool.h> +#include <linux/slab.h> +#include <linux/device.h> +#include <linux/skbuff.h> +#include <linux/if_vlan.h> +#include <linux/if_bridge.h> +#include <linux/workqueue.h> +#include <linux/jiffies.h> +#include <linux/bitops.h> +#include <net/switchdev.h> +#include <generated/utsrelease.h> + +#include "spectrum.h" +#include "core.h" +#include "reg.h" +#include "port.h" +#include "trap.h" +#include "txheader.h" + +static const char mlxsw_sp_driver_name[] = "mlxsw_spectrum"; +static const char mlxsw_sp_driver_version[] = "1.0"; + +/* tx_hdr_version + * Tx header version. + * Must be set to 1. + */ +MLXSW_ITEM32(tx, hdr, version, 0x00, 28, 4); + +/* tx_hdr_ctl + * Packet control type. + * 0 - Ethernet control (e.g. EMADs, LACP) + * 1 - Ethernet data + */ +MLXSW_ITEM32(tx, hdr, ctl, 0x00, 26, 2); + +/* tx_hdr_proto + * Packet protocol type. Must be set to 1 (Ethernet). + */ +MLXSW_ITEM32(tx, hdr, proto, 0x00, 21, 3); + +/* tx_hdr_rx_is_router + * Packet is sent from the router. Valid for data packets only. + */ +MLXSW_ITEM32(tx, hdr, rx_is_router, 0x00, 19, 1); + +/* tx_hdr_fid_valid + * Indicates if the 'fid' field is valid and should be used for + * forwarding lookup. Valid for data packets only. + */ +MLXSW_ITEM32(tx, hdr, fid_valid, 0x00, 16, 1); + +/* tx_hdr_swid + * Switch partition ID. Must be set to 0. + */ +MLXSW_ITEM32(tx, hdr, swid, 0x00, 12, 3); + +/* tx_hdr_control_tclass + * Indicates if the packet should use the control TClass and not one + * of the data TClasses. + */ +MLXSW_ITEM32(tx, hdr, control_tclass, 0x00, 6, 1); + +/* tx_hdr_etclass + * Egress TClass to be used on the egress device on the egress port. + */ +MLXSW_ITEM32(tx, hdr, etclass, 0x00, 0, 4); + +/* tx_hdr_port_mid + * Destination local port for unicast packets. + * Destination multicast ID for multicast packets. + * + * Control packets are directed to a specific egress port, while data + * packets are transmitted through the CPU port (0) into the switch partition, + * where forwarding rules are applied. + */ +MLXSW_ITEM32(tx, hdr, port_mid, 0x04, 16, 16); + +/* tx_hdr_fid + * Forwarding ID used for L2 forwarding lookup. Valid only if 'fid_valid' is + * set, otherwise calculated based on the packet's VID using VID to FID mapping. + * Valid for data packets only. + */ +MLXSW_ITEM32(tx, hdr, fid, 0x08, 0, 16); + +/* tx_hdr_type + * 0 - Data packets + * 6 - Control packets + */ +MLXSW_ITEM32(tx, hdr, type, 0x0C, 0, 4); + +static void mlxsw_sp_txhdr_construct(struct sk_buff *skb, + const struct mlxsw_tx_info *tx_info) +{ + char *txhdr = skb_push(skb, MLXSW_TXHDR_LEN); + + memset(txhdr, 0, MLXSW_TXHDR_LEN); + + mlxsw_tx_hdr_version_set(txhdr, MLXSW_TXHDR_VERSION_1); + mlxsw_tx_hdr_ctl_set(txhdr, MLXSW_TXHDR_ETH_CTL); + mlxsw_tx_hdr_proto_set(txhdr, MLXSW_TXHDR_PROTO_ETH); + mlxsw_tx_hdr_swid_set(txhdr, 0); + mlxsw_tx_hdr_control_tclass_set(txhdr, 1); + mlxsw_tx_hdr_port_mid_set(txhdr, tx_info->local_port); + mlxsw_tx_hdr_type_set(txhdr, MLXSW_TXHDR_TYPE_CONTROL); +} + +static int mlxsw_sp_base_mac_get(struct mlxsw_sp *mlxsw_sp) +{ + char spad_pl[MLXSW_REG_SPAD_LEN]; + int err; + + err = mlxsw_reg_query(mlxsw_sp->core, MLXSW_REG(spad), spad_pl); + if (err) + return err; + mlxsw_reg_spad_base_mac_memcpy_from(spad_pl, mlxsw_sp->base_mac); + return 0; +} + +static int mlxsw_sp_port_admin_status_set(struct mlxsw_sp_port *mlxsw_sp_port, + bool is_up) +{ + struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; + char paos_pl[MLXSW_REG_PAOS_LEN]; + + mlxsw_reg_paos_pack(paos_pl, mlxsw_sp_port->local_port, + is_up ? MLXSW_PORT_ADMIN_STATUS_UP : + MLXSW_PORT_ADMIN_STATUS_DOWN); + return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(paos), paos_pl); +} + +static int mlxsw_sp_port_oper_status_get(struct mlxsw_sp_port *mlxsw_sp_port, + bool *p_is_up) +{ + struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; + char paos_pl[MLXSW_REG_PAOS_LEN]; + u8 oper_status; + int err; + + mlxsw_reg_paos_pack(paos_pl, mlxsw_sp_port->local_port, 0); + err = mlxsw_reg_query(mlxsw_sp->core, MLXSW_REG(paos), paos_pl); + if (err) + return err; + oper_status = mlxsw_reg_paos_oper_status_get(paos_pl); + *p_is_up = oper_status == MLXSW_PORT_ADMIN_STATUS_UP ? true : false; + return 0; +} + +static int mlxsw_sp_vfid_create(struct mlxsw_sp *mlxsw_sp, u16 vfid) +{ + char sfmr_pl[MLXSW_REG_SFMR_LEN]; + int err; + + mlxsw_reg_sfmr_pack(sfmr_pl, MLXSW_REG_SFMR_OP_CREATE_FID, + MLXSW_SP_VFID_BASE + vfid, 0); + err = mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sfmr), sfmr_pl); + + if (err) + return err; + + set_bit(vfid, mlxsw_sp->active_vfids); + return 0; +} + +static void mlxsw_sp_vfid_destroy(struct mlxsw_sp *mlxsw_sp, u16 vfid) +{ + char sfmr_pl[MLXSW_REG_SFMR_LEN]; + + clear_bit(vfid, mlxsw_sp->active_vfids); + + mlxsw_reg_sfmr_pack(sfmr_pl, MLXSW_REG_SFMR_OP_DESTROY_FID, + MLXSW_SP_VFID_BASE + vfid, 0); + mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sfmr), sfmr_pl); +} + +static int mlxsw_sp_port_dev_addr_set(struct mlxsw_sp_port *mlxsw_sp_port, + unsigned char *addr) +{ + struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; + char ppad_pl[MLXSW_REG_PPAD_LEN]; + + mlxsw_reg_ppad_pack(ppad_pl, true, mlxsw_sp_port->local_port); + mlxsw_reg_ppad_mac_memcpy_to(ppad_pl, addr); + return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(ppad), ppad_pl); +} + +static int mlxsw_sp_port_dev_addr_init(struct mlxsw_sp_port *mlxsw_sp_port) +{ + struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; + unsigned char *addr = mlxsw_sp_port->dev->dev_addr; + + ether_addr_copy(addr, mlxsw_sp->base_mac); + addr[ETH_ALEN - 1] += mlxsw_sp_port->local_port; + return mlxsw_sp_port_dev_addr_set(mlxsw_sp_port, addr); +} + +static int mlxsw_sp_port_stp_state_set(struct mlxsw_sp_port *mlxsw_sp_port, + u16 vid, enum mlxsw_reg_spms_state state) +{ + struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; + char *spms_pl; + int err; + + spms_pl = kmalloc(MLXSW_REG_SPMS_LEN, GFP_KERNEL); + if (!spms_pl) + return -ENOMEM; + mlxsw_reg_spms_pack(spms_pl, mlxsw_sp_port->local_port); + mlxsw_reg_spms_vid_pack(spms_pl, vid, state); + err = mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(spms), spms_pl); + kfree(spms_pl); + return err; +} + +static int mlxsw_sp_port_mtu_set(struct mlxsw_sp_port *mlxsw_sp_port, u16 mtu) +{ + struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; + char pmtu_pl[MLXSW_REG_PMTU_LEN]; + int max_mtu; + int err; + + mtu += MLXSW_TXHDR_LEN + ETH_HLEN; + mlxsw_reg_pmtu_pack(pmtu_pl, mlxsw_sp_port->local_port, 0); + err = mlxsw_reg_query(mlxsw_sp->core, MLXSW_REG(pmtu), pmtu_pl); + if (err) + return err; + max_mtu = mlxsw_reg_pmtu_max_mtu_get(pmtu_pl); + + if (mtu > max_mtu) + return -EINVAL; + + mlxsw_reg_pmtu_pack(pmtu_pl, mlxsw_sp_port->local_port, mtu); + return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(pmtu), pmtu_pl); +} + +static int mlxsw_sp_port_swid_set(struct mlxsw_sp_port *mlxsw_sp_port, u8 swid) +{ + struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; + char pspa_pl[MLXSW_REG_PSPA_LEN]; + + mlxsw_reg_pspa_pack(pspa_pl, swid, mlxsw_sp_port->local_port); + return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(pspa), pspa_pl); +} + +static int mlxsw_sp_port_vp_mode_set(struct mlxsw_sp_port *mlxsw_sp_port, + bool enable) +{ + struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; + char svpe_pl[MLXSW_REG_SVPE_LEN]; + + mlxsw_reg_svpe_pack(svpe_pl, mlxsw_sp_port->local_port, enable); + return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(svpe), svpe_pl); +} + +int mlxsw_sp_port_vid_to_fid_set(struct mlxsw_sp_port *mlxsw_sp_port, + enum mlxsw_reg_svfa_mt mt, bool valid, u16 fid, + u16 vid) +{ + struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; + char svfa_pl[MLXSW_REG_SVFA_LEN]; + + mlxsw_reg_svfa_pack(svfa_pl, mlxsw_sp_port->local_port, mt, valid, + fid, vid); + return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(svfa), svfa_pl); +} + +static int mlxsw_sp_port_vid_learning_set(struct mlxsw_sp_port *mlxsw_sp_port, + u16 vid, bool learn_enable) +{ + struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; + char *spvmlr_pl; + int err; + + spvmlr_pl = kmalloc(MLXSW_REG_SPVMLR_LEN, GFP_KERNEL); + if (!spvmlr_pl) + return -ENOMEM; + mlxsw_reg_spvmlr_pack(spvmlr_pl, mlxsw_sp_port->local_port, vid, vid, + learn_enable); + err = mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(spvmlr), spvmlr_pl); + kfree(spvmlr_pl); + return err; +} + +static int +mlxsw_sp_port_system_port_mapping_set(struct mlxsw_sp_port *mlxsw_sp_port) +{ + struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; + char sspr_pl[MLXSW_REG_SSPR_LEN]; + + mlxsw_reg_sspr_pack(sspr_pl, mlxsw_sp_port->local_port); + return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sspr), sspr_pl); +} + +static int mlxsw_sp_port_module_check(struct mlxsw_sp_port *mlxsw_sp_port, + bool *p_usable) +{ + struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; + char pmlp_pl[MLXSW_REG_PMLP_LEN]; + int err; + + mlxsw_reg_pmlp_pack(pmlp_pl, mlxsw_sp_port->local_port); + err = mlxsw_reg_query(mlxsw_sp->core, MLXSW_REG(pmlp), pmlp_pl); + if (err) + return err; + *p_usable = mlxsw_reg_pmlp_width_get(pmlp_pl) ? true : false; + return 0; +} + +static int mlxsw_sp_port_open(struct net_device *dev) +{ + struct mlxsw_sp_port *mlxsw_sp_port = netdev_priv(dev); + int err; + + err = mlxsw_sp_port_admin_status_set(mlxsw_sp_port, true); + if (err) + return err; + netif_start_queue(dev); + return 0; +} + +static int mlxsw_sp_port_stop(struct net_device *dev) +{ + struct mlxsw_sp_port *mlxsw_sp_port = netdev_priv(dev); + + netif_stop_queue(dev); + return mlxsw_sp_port_admin_status_set(mlxsw_sp_port, false); +} + +static netdev_tx_t mlxsw_sp_port_xmit(struct sk_buff *skb, + struct net_device *dev) +{ + struct mlxsw_sp_port *mlxsw_sp_port = netdev_priv(dev); + struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; + struct mlxsw_sp_port_pcpu_stats *pcpu_stats; + const struct mlxsw_tx_info tx_info = { + .local_port = mlxsw_sp_port->local_port, + .is_emad = false, + }; + u64 len; + int err; + + if (mlxsw_core_skb_transmit_busy(mlxsw_sp, &tx_info)) + return NETDEV_TX_BUSY; + + if (unlikely(skb_headroom(skb) < MLXSW_TXHDR_LEN)) { + struct sk_buff *skb_orig = skb; + + skb = skb_realloc_headroom(skb, MLXSW_TXHDR_LEN); + if (!skb) { + this_cpu_inc(mlxsw_sp_port->pcpu_stats->tx_dropped); + dev_kfree_skb_any(skb_orig); + return NETDEV_TX_OK; + } + } + + if (eth_skb_pad(skb)) { + this_cpu_inc(mlxsw_sp_port->pcpu_stats->tx_dropped); + return NETDEV_TX_OK; + } + + mlxsw_sp_txhdr_construct(skb, &tx_info); + len = skb->len; + /* Due to a race we might fail here because of a full queue. In that + * unlikely case we simply drop the packet. + */ + err = mlxsw_core_skb_transmit(mlxsw_sp, skb, &tx_info); + + if (!err) { + pcpu_stats = this_cpu_ptr(mlxsw_sp_port->pcpu_stats); + u64_stats_update_begin(&pcpu_stats->syncp); + pcpu_stats->tx_packets++; + pcpu_stats->tx_bytes += len; + u64_stats_update_end(&pcpu_stats->syncp); + } else { + this_cpu_inc(mlxsw_sp_port->pcpu_stats->tx_dropped); + dev_kfree_skb_any(skb); + } + return NETDEV_TX_OK; +} + +static int mlxsw_sp_port_set_mac_address(struct net_device *dev, void *p) +{ + struct mlxsw_sp_port *mlxsw_sp_port = netdev_priv(dev); + struct sockaddr *addr = p; + int err; + + if (!is_valid_ether_addr(addr->sa_data)) + return -EADDRNOTAVAIL; + + err = mlxsw_sp_port_dev_addr_set(mlxsw_sp_port, addr->sa_data); + if (err) + return err; + memcpy(dev->dev_addr, addr->sa_data, dev->addr_len); + return 0; +} + +static int mlxsw_sp_port_change_mtu(struct net_device *dev, int mtu) +{ + struct mlxsw_sp_port *mlxsw_sp_port = netdev_priv(dev); + int err; + + err = mlxsw_sp_port_mtu_set(mlxsw_sp_port, mtu); + if (err) + return err; + dev->mtu = mtu; + return 0; +} + +static struct rtnl_link_stats64 * +mlxsw_sp_port_get_stats64(struct net_device *dev, + struct rtnl_link_stats64 *stats) +{ + struct mlxsw_sp_port *mlxsw_sp_port = netdev_priv(dev); + struct mlxsw_sp_port_pcpu_stats *p; + u64 rx_packets, rx_bytes, tx_packets, tx_bytes; + u32 tx_dropped = 0; + unsigned int start; + int i; + + for_each_possible_cpu(i) { + p = per_cpu_ptr(mlxsw_sp_port->pcpu_stats, i); + do { + start = u64_stats_fetch_begin_irq(&p->syncp); + rx_packets = p->rx_packets; + rx_bytes = p->rx_bytes; + tx_packets = p->tx_packets; + tx_bytes = p->tx_bytes; + } while (u64_stats_fetch_retry_irq(&p->syncp, start)); + + stats->rx_packets += rx_packets; + stats->rx_bytes += rx_bytes; + stats->tx_packets += tx_packets; + stats->tx_bytes += tx_bytes; + /* tx_dropped is u32, updated without syncp protection. */ + tx_dropped += p->tx_dropped; + } + stats->tx_dropped = tx_dropped; + return stats; +} + +int mlxsw_sp_port_vlan_set(struct mlxsw_sp_port *mlxsw_sp_port, u16 vid_begin, + u16 vid_end, bool is_member, bool untagged) +{ + struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; + char *spvm_pl; + int err; + + spvm_pl = kmalloc(MLXSW_REG_SPVM_LEN, GFP_KERNEL); + if (!spvm_pl) + return -ENOMEM; + + mlxsw_reg_spvm_pack(spvm_pl, mlxsw_sp_port->local_port, vid_begin, + vid_end, is_member, untagged); + err = mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(spvm), spvm_pl); + kfree(spvm_pl); + return err; +} + +static int mlxsw_sp_port_vp_mode_trans(struct mlxsw_sp_port *mlxsw_sp_port) +{ + enum mlxsw_reg_svfa_mt mt = MLXSW_REG_SVFA_MT_PORT_VID_TO_FID; + u16 vid, last_visited_vid; + int err; + + for_each_set_bit(vid, mlxsw_sp_port->active_vlans, VLAN_N_VID) { + err = mlxsw_sp_port_vid_to_fid_set(mlxsw_sp_port, mt, true, vid, + vid); + if (err) { + last_visited_vid = vid; + goto err_port_vid_to_fid_set; + } + } + + err = mlxsw_sp_port_vp_mode_set(mlxsw_sp_port, true); + if (err) { + last_visited_vid = VLAN_N_VID; + goto err_port_vid_to_fid_set; + } + + return 0; + +err_port_vid_to_fid_set: + for_each_set_bit(vid, mlxsw_sp_port->active_vlans, last_visited_vid) + mlxsw_sp_port_vid_to_fid_set(mlxsw_sp_port, mt, false, vid, + vid); + return err; +} + +static int mlxsw_sp_port_vlan_mode_trans(struct mlxsw_sp_port *mlxsw_sp_port) +{ + enum mlxsw_reg_svfa_mt mt = MLXSW_REG_SVFA_MT_PORT_VID_TO_FID; + u16 vid; + int err; + + err = mlxsw_sp_port_vp_mode_set(mlxsw_sp_port, false); + if (err) + return err; + + for_each_set_bit(vid, mlxsw_sp_port->active_vlans, VLAN_N_VID) { + err = mlxsw_sp_port_vid_to_fid_set(mlxsw_sp_port, mt, false, + vid, vid); + if (err) + return err; + } + + return 0; +} + +int mlxsw_sp_port_add_vid(struct net_device *dev, __be16 __always_unused proto, + u16 vid) +{ + struct mlxsw_sp_port *mlxsw_sp_port = netdev_priv(dev); + struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; + char *sftr_pl; + int err; + + /* VLAN 0 is added to HW filter when device goes up, but it is + * reserved in our case, so simply return. + */ + if (!vid) + return 0; + + if (test_bit(vid, mlxsw_sp_port->active_vfids)) { + netdev_warn(dev, "VID=%d already configured\n", vid); + return 0; + } + + if (!test_bit(vid, mlxsw_sp->active_vfids)) { + err = mlxsw_sp_vfid_create(mlxsw_sp, vid); + if (err) { + netdev_err(dev, "Failed to create vFID=%d\n", + MLXSW_SP_VFID_BASE + vid); + return err; + } + + sftr_pl = kmalloc(MLXSW_REG_SFTR_LEN, GFP_KERNEL); + if (!sftr_pl) { + err = -ENOMEM; + goto err_flood_table_alloc; + } + mlxsw_reg_sftr_pack(sftr_pl, 0, vid, + MLXSW_REG_SFGC_TABLE_TYPE_FID, 0, + MLXSW_PORT_CPU_PORT, true); + err = mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sftr), sftr_pl); + kfree(sftr_pl); + if (err) { + netdev_err(dev, "Failed to configure flood table\n"); + goto err_flood_table_config; + } + } + + /* In case we fail in the following steps, we intentionally do not + * destroy the associated vFID. + */ + + /* When adding the first VLAN interface on a bridged port we need to + * transition all the active 802.1Q bridge VLANs to use explicit + * {Port, VID} to FID mappings and set the port's mode to Virtual mode. + */ + if (!mlxsw_sp_port->nr_vfids) { + err = mlxsw_sp_port_vp_mode_trans(mlxsw_sp_port); + if (err) { + netdev_err(dev, "Failed to set to Virtual mode\n"); + return err; + } + } + + err = mlxsw_sp_port_vid_to_fid_set(mlxsw_sp_port, + MLXSW_REG_SVFA_MT_PORT_VID_TO_FID, + true, MLXSW_SP_VFID_BASE + vid, vid); + if (err) { + netdev_err(dev, "Failed to map {Port, VID=%d} to vFID=%d\n", + vid, MLXSW_SP_VFID_BASE + vid); + goto err_port_vid_to_fid_set; + } + + err = mlxsw_sp_port_vid_learning_set(mlxsw_sp_port, vid, false); + if (err) { + netdev_err(dev, "Failed to disable learning for VID=%d\n", vid); + goto err_port_vid_learning_set; + } + + err = mlxsw_sp_port_vlan_set(mlxsw_sp_port, vid, vid, true, false); + if (err) { + netdev_err(dev, "Failed to set VLAN membership for VID=%d\n", + vid); + goto err_port_add_vid; + } + + err = mlxsw_sp_port_stp_state_set(mlxsw_sp_port, vid, + MLXSW_REG_SPMS_STATE_FORWARDING); + if (err) { + netdev_err(dev, "Failed to set STP state for VID=%d\n", vid); + goto err_port_stp_state_set; + } + + mlxsw_sp_port->nr_vfids++; + set_bit(vid, mlxsw_sp_port->active_vfids); + + return 0; + +err_flood_table_config: +err_flood_table_alloc: + mlxsw_sp_vfid_destroy(mlxsw_sp, vid); + return err; + +err_port_stp_state_set: + mlxsw_sp_port_vlan_set(mlxsw_sp_port, vid, vid, false, false); +err_port_add_vid: + mlxsw_sp_port_vid_learning_set(mlxsw_sp_port, vid, true); +err_port_vid_learning_set: + mlxsw_sp_port_vid_to_fid_set(mlxsw_sp_port, + MLXSW_REG_SVFA_MT_PORT_VID_TO_FID, false, + MLXSW_SP_VFID_BASE + vid, vid); +err_port_vid_to_fid_set: + mlxsw_sp_port_vlan_mode_trans(mlxsw_sp_port); + return err; +} + +int mlxsw_sp_port_kill_vid(struct net_device *dev, + __be16 __always_unused proto, u16 vid) +{ + struct mlxsw_sp_port *mlxsw_sp_port = netdev_priv(dev); + int err; + + /* VLAN 0 is removed from HW filter when device goes down, but + * it is reserved in our case, so simply return. + */ + if (!vid) + return 0; + + if (!test_bit(vid, mlxsw_sp_port->active_vfids)) { + netdev_warn(dev, "VID=%d does not exist\n", vid); + return 0; + } + + err = mlxsw_sp_port_stp_state_set(mlxsw_sp_port, vid, + MLXSW_REG_SPMS_STATE_DISCARDING); + if (err) { + netdev_err(dev, "Failed to set STP state for VID=%d\n", vid); + return err; + } + + err = mlxsw_sp_port_vlan_set(mlxsw_sp_port, vid, vid, false, false); + if (err) { + netdev_err(dev, "Failed to set VLAN membership for VID=%d\n", + vid); + return err; + } + + err = mlxsw_sp_port_vid_learning_set(mlxsw_sp_port, vid, true); + if (err) { + netdev_err(dev, "Failed to enable learning for VID=%d\n", vid); + return err; + } + + err = mlxsw_sp_port_vid_to_fid_set(mlxsw_sp_port, + MLXSW_REG_SVFA_MT_PORT_VID_TO_FID, + false, MLXSW_SP_VFID_BASE + vid, + vid); + if (err) { + netdev_err(dev, "Failed to invalidate {Port, VID=%d} to vFID=%d mapping\n", + vid, MLXSW_SP_VFID_BASE + vid); + return err; + } + + /* When removing the last VLAN interface on a bridged port we need to + * transition all active 802.1Q bridge VLANs to use VID to FID + * mappings and set port's mode to VLAN mode. + */ + if (mlxsw_sp_port->nr_vfids == 1) { + err = mlxsw_sp_port_vlan_mode_trans(mlxsw_sp_port); + if (err) { + netdev_err(dev, "Failed to set to VLAN mode\n"); + return err; + } + } + + mlxsw_sp_port->nr_vfids--; + clear_bit(vid, mlxsw_sp_port->active_vfids); + + return 0; +} + +static const struct net_device_ops mlxsw_sp_port_netdev_ops = { + .ndo_open = mlxsw_sp_port_open, + .ndo_stop = mlxsw_sp_port_stop, + .ndo_start_xmit = mlxsw_sp_port_xmit, + .ndo_set_mac_address = mlxsw_sp_port_set_mac_address, + .ndo_change_mtu = mlxsw_sp_port_change_mtu, + .ndo_get_stats64 = mlxsw_sp_port_get_stats64, + .ndo_vlan_rx_add_vid = mlxsw_sp_port_add_vid, + .ndo_vlan_rx_kill_vid = mlxsw_sp_port_kill_vid, + .ndo_fdb_add = switchdev_port_fdb_add, + .ndo_fdb_del = switchdev_port_fdb_del, + .ndo_fdb_dump = switchdev_port_fdb_dump, + .ndo_bridge_setlink = switchdev_port_bridge_setlink, + .ndo_bridge_getlink = switchdev_port_bridge_getlink, + .ndo_bridge_dellink = switchdev_port_bridge_dellink, +}; + +static void mlxsw_sp_port_get_drvinfo(struct net_device *dev, + struct ethtool_drvinfo *drvinfo) +{ + struct mlxsw_sp_port *mlxsw_sp_port = netdev_priv(dev); + struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; + + strlcpy(drvinfo->driver, mlxsw_sp_driver_name, sizeof(drvinfo->driver)); + strlcpy(drvinfo->version, mlxsw_sp_driver_version, + sizeof(drvinfo->version)); + snprintf(drvinfo->fw_version, sizeof(drvinfo->fw_version), + "%d.%d.%d", + mlxsw_sp->bus_info->fw_rev.major, + mlxsw_sp->bus_info->fw_rev.minor, + mlxsw_sp->bus_info->fw_rev.subminor); + strlcpy(drvinfo->bus_info, mlxsw_sp->bus_info->device_name, + sizeof(drvinfo->bus_info)); +} + +struct mlxsw_sp_port_hw_stats { + char str[ETH_GSTRING_LEN]; + u64 (*getter)(char *payload); +}; + +static const struct mlxsw_sp_port_hw_stats mlxsw_sp_port_hw_stats[] = { + { + .str = "a_frames_transmitted_ok", + .getter = mlxsw_reg_ppcnt_a_frames_transmitted_ok_get, + }, + { + .str = "a_frames_received_ok", + .getter = mlxsw_reg_ppcnt_a_frames_received_ok_get, + }, + { + .str = "a_frame_check_sequence_errors", + .getter = mlxsw_reg_ppcnt_a_frame_check_sequence_errors_get, + }, + { + .str = "a_alignment_errors", + .getter = mlxsw_reg_ppcnt_a_alignment_errors_get, + }, + { + .str = "a_octets_transmitted_ok", + .getter = mlxsw_reg_ppcnt_a_octets_transmitted_ok_get, + }, + { + .str = "a_octets_received_ok", + .getter = mlxsw_reg_ppcnt_a_octets_received_ok_get, + }, + { + .str = "a_multicast_frames_xmitted_ok", + .getter = mlxsw_reg_ppcnt_a_multicast_frames_xmitted_ok_get, + }, + { + .str = "a_broadcast_frames_xmitted_ok", + .getter = mlxsw_reg_ppcnt_a_broadcast_frames_xmitted_ok_get, + }, + { + .str = "a_multicast_frames_received_ok", + .getter = mlxsw_reg_ppcnt_a_multicast_frames_received_ok_get, + }, + { + .str = "a_broadcast_frames_received_ok", + .getter = mlxsw_reg_ppcnt_a_broadcast_frames_received_ok_get, + }, + { + .str = "a_in_range_length_errors", + .getter = mlxsw_reg_ppcnt_a_in_range_length_errors_get, + }, + { + .str = "a_out_of_range_length_field", + .getter = mlxsw_reg_ppcnt_a_out_of_range_length_field_get, + }, + { + .str = "a_frame_too_long_errors", + .getter = mlxsw_reg_ppcnt_a_frame_too_long_errors_get, + }, + { + .str = "a_symbol_error_during_carrier", + .getter = mlxsw_reg_ppcnt_a_symbol_error_during_carrier_get, + }, + { + .str = "a_mac_control_frames_transmitted", + .getter = mlxsw_reg_ppcnt_a_mac_control_frames_transmitted_get, + }, + { + .str = "a_mac_control_frames_received", + .getter = mlxsw_reg_ppcnt_a_mac_control_frames_received_get, + }, + { + .str = "a_unsupported_opcodes_received", + .getter = mlxsw_reg_ppcnt_a_unsupported_opcodes_received_get, + }, + { + .str = "a_pause_mac_ctrl_frames_received", + .getter = mlxsw_reg_ppcnt_a_pause_mac_ctrl_frames_received_get, + }, + { + .str = "a_pause_mac_ctrl_frames_xmitted", + .getter = mlxsw_reg_ppcnt_a_pause_mac_ctrl_frames_transmitted_get, + }, +}; + +#define MLXSW_SP_PORT_HW_STATS_LEN ARRAY_SIZE(mlxsw_sp_port_hw_stats) + +static void mlxsw_sp_port_get_strings(struct net_device *dev, + u32 stringset, u8 *data) +{ + u8 *p = data; + int i; + + switch (stringset) { + case ETH_SS_STATS: + for (i = 0; i < MLXSW_SP_PORT_HW_STATS_LEN; i++) { + memcpy(p, mlxsw_sp_port_hw_stats[i].str, + ETH_GSTRING_LEN); + p += ETH_GSTRING_LEN; + } + break; + } +} + +static void mlxsw_sp_port_get_stats(struct net_device *dev, + struct ethtool_stats *stats, u64 *data) +{ + struct mlxsw_sp_port *mlxsw_sp_port = netdev_priv(dev); + struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; + char ppcnt_pl[MLXSW_REG_PPCNT_LEN]; + int i; + int err; + + mlxsw_reg_ppcnt_pack(ppcnt_pl, mlxsw_sp_port->local_port); + err = mlxsw_reg_query(mlxsw_sp->core, MLXSW_REG(ppcnt), ppcnt_pl); + for (i = 0; i < MLXSW_SP_PORT_HW_STATS_LEN; i++) + data[i] = !err ? mlxsw_sp_port_hw_stats[i].getter(ppcnt_pl) : 0; +} + +static int mlxsw_sp_port_get_sset_count(struct net_device *dev, int sset) +{ + switch (sset) { + case ETH_SS_STATS: + return MLXSW_SP_PORT_HW_STATS_LEN; + default: + return -EOPNOTSUPP; + } +} + +struct mlxsw_sp_port_link_mode { + u32 mask; + u32 supported; + u32 advertised; + u32 speed; +}; + +static const struct mlxsw_sp_port_link_mode mlxsw_sp_port_link_mode[] = { + { + .mask = MLXSW_REG_PTYS_ETH_SPEED_100BASE_T, + .supported = SUPPORTED_100baseT_Full, + .advertised = ADVERTISED_100baseT_Full, + .speed = 100, + }, + { + .mask = MLXSW_REG_PTYS_ETH_SPEED_100BASE_TX, + .speed = 100, + }, + { + .mask = MLXSW_REG_PTYS_ETH_SPEED_SGMII | + MLXSW_REG_PTYS_ETH_SPEED_1000BASE_KX, + .supported = SUPPORTED_1000baseKX_Full, + .advertised = ADVERTISED_1000baseKX_Full, + .speed = 1000, + }, + { + .mask = MLXSW_REG_PTYS_ETH_SPEED_10GBASE_T, + .supported = SUPPORTED_10000baseT_Full, + .advertised = ADVERTISED_10000baseT_Full, + .speed = 10000, + }, + { + .mask = MLXSW_REG_PTYS_ETH_SPEED_10GBASE_CX4 | + MLXSW_REG_PTYS_ETH_SPEED_10GBASE_KX4, + .supported = SUPPORTED_10000baseKX4_Full, + .advertised = ADVERTISED_10000baseKX4_Full, + .speed = 10000, + }, + { + .mask = MLXSW_REG_PTYS_ETH_SPEED_10GBASE_KR | + MLXSW_REG_PTYS_ETH_SPEED_10GBASE_CR | + MLXSW_REG_PTYS_ETH_SPEED_10GBASE_SR | + MLXSW_REG_PTYS_ETH_SPEED_10GBASE_ER_LR, + .supported = SUPPORTED_10000baseKR_Full, + .advertised = ADVERTISED_10000baseKR_Full, + .speed = 10000, + }, + { + .mask = MLXSW_REG_PTYS_ETH_SPEED_20GBASE_KR2, + .supported = SUPPORTED_20000baseKR2_Full, + .advertised = ADVERTISED_20000baseKR2_Full, + .speed = 20000, + }, + { + .mask = MLXSW_REG_PTYS_ETH_SPEED_40GBASE_CR4, + .supported = SUPPORTED_40000baseCR4_Full, + .advertised = ADVERTISED_40000baseCR4_Full, + .speed = 40000, + }, + { + .mask = MLXSW_REG_PTYS_ETH_SPEED_40GBASE_KR4, + .supported = SUPPORTED_40000baseKR4_Full, + .advertised = ADVERTISED_40000baseKR4_Full, + .speed = 40000, + }, + { + .mask = MLXSW_REG_PTYS_ETH_SPEED_40GBASE_SR4, + .supported = SUPPORTED_40000baseSR4_Full, + .advertised = ADVERTISED_40000baseSR4_Full, + .speed = 40000, + }, + { + .mask = MLXSW_REG_PTYS_ETH_SPEED_40GBASE_LR4_ER4, + .supported = SUPPORTED_40000baseLR4_Full, + .advertised = ADVERTISED_40000baseLR4_Full, + .speed = 40000, + }, + { + .mask = MLXSW_REG_PTYS_ETH_SPEED_25GBASE_CR | + MLXSW_REG_PTYS_ETH_SPEED_25GBASE_KR | + MLXSW_REG_PTYS_ETH_SPEED_25GBASE_SR, + .speed = 25000, + }, + { + .mask = MLXSW_REG_PTYS_ETH_SPEED_50GBASE_KR4 | + MLXSW_REG_PTYS_ETH_SPEED_50GBASE_CR2 | + MLXSW_REG_PTYS_ETH_SPEED_50GBASE_KR2, + .speed = 50000, + }, + { + .mask = MLXSW_REG_PTYS_ETH_SPEED_56GBASE_R4, + .supported = SUPPORTED_56000baseKR4_Full, + .advertised = ADVERTISED_56000baseKR4_Full, + .speed = 56000, + }, + { + .mask = MLXSW_REG_PTYS_ETH_SPEED_100GBASE_CR4 | + MLXSW_REG_PTYS_ETH_SPEED_100GBASE_SR4 | + MLXSW_REG_PTYS_ETH_SPEED_100GBASE_KR4 | + MLXSW_REG_PTYS_ETH_SPEED_100GBASE_LR4_ER4, + .speed = 100000, + }, +}; + +#define MLXSW_SP_PORT_LINK_MODE_LEN ARRAY_SIZE(mlxsw_sp_port_link_mode) + +static u32 mlxsw_sp_from_ptys_supported_port(u32 ptys_eth_proto) +{ + if (ptys_eth_proto & (MLXSW_REG_PTYS_ETH_SPEED_10GBASE_CR | + MLXSW_REG_PTYS_ETH_SPEED_10GBASE_SR | + MLXSW_REG_PTYS_ETH_SPEED_40GBASE_CR4 | + MLXSW_REG_PTYS_ETH_SPEED_40GBASE_SR4 | + MLXSW_REG_PTYS_ETH_SPEED_100GBASE_SR4 | + MLXSW_REG_PTYS_ETH_SPEED_SGMII)) + return SUPPORTED_FIBRE; + + if (ptys_eth_proto & (MLXSW_REG_PTYS_ETH_SPEED_10GBASE_KR | + MLXSW_REG_PTYS_ETH_SPEED_10GBASE_KX4 | + MLXSW_REG_PTYS_ETH_SPEED_40GBASE_KR4 | + MLXSW_REG_PTYS_ETH_SPEED_100GBASE_KR4 | + MLXSW_REG_PTYS_ETH_SPEED_1000BASE_KX)) + return SUPPORTED_Backplane; + return 0; +} + +static u32 mlxsw_sp_from_ptys_supported_link(u32 ptys_eth_proto) +{ + u32 modes = 0; + int i; + + for (i = 0; i < MLXSW_SP_PORT_LINK_MODE_LEN; i++) { + if (ptys_eth_proto & mlxsw_sp_port_link_mode[i].mask) + modes |= mlxsw_sp_port_link_mode[i].supported; + } + return modes; +} + +static u32 mlxsw_sp_from_ptys_advert_link(u32 ptys_eth_proto) +{ + u32 modes = 0; + int i; + + for (i = 0; i < MLXSW_SP_PORT_LINK_MODE_LEN; i++) { + if (ptys_eth_proto & mlxsw_sp_port_link_mode[i].mask) + modes |= mlxsw_sp_port_link_mode[i].advertised; + } + return modes; +} + +static void mlxsw_sp_from_ptys_speed_duplex(bool carrier_ok, u32 ptys_eth_proto, + struct ethtool_cmd *cmd) +{ + u32 speed = SPEED_UNKNOWN; + u8 duplex = DUPLEX_UNKNOWN; + int i; + + if (!carrier_ok) + goto out; + + for (i = 0; i < MLXSW_SP_PORT_LINK_MODE_LEN; i++) { + if (ptys_eth_proto & mlxsw_sp_port_link_mode[i].mask) { + speed = mlxsw_sp_port_link_mode[i].speed; + duplex = DUPLEX_FULL; + break; + } + } +out: + ethtool_cmd_speed_set(cmd, speed); + cmd->duplex = duplex; +} + +static u8 mlxsw_sp_port_connector_port(u32 ptys_eth_proto) +{ + if (ptys_eth_proto & (MLXSW_REG_PTYS_ETH_SPEED_10GBASE_SR | + MLXSW_REG_PTYS_ETH_SPEED_40GBASE_SR4 | + MLXSW_REG_PTYS_ETH_SPEED_100GBASE_SR4 | + MLXSW_REG_PTYS_ETH_SPEED_SGMII)) + return PORT_FIBRE; + + if (ptys_eth_proto & (MLXSW_REG_PTYS_ETH_SPEED_10GBASE_CR | + MLXSW_REG_PTYS_ETH_SPEED_40GBASE_CR4 | + MLXSW_REG_PTYS_ETH_SPEED_100GBASE_CR4)) + return PORT_DA; + + if (ptys_eth_proto & (MLXSW_REG_PTYS_ETH_SPEED_10GBASE_KR | + MLXSW_REG_PTYS_ETH_SPEED_10GBASE_KX4 | + MLXSW_REG_PTYS_ETH_SPEED_40GBASE_KR4 | + MLXSW_REG_PTYS_ETH_SPEED_100GBASE_KR4)) + return PORT_NONE; + + return PORT_OTHER; +} + +static int mlxsw_sp_port_get_settings(struct net_device *dev, + struct ethtool_cmd *cmd) +{ + struct mlxsw_sp_port *mlxsw_sp_port = netdev_priv(dev); + struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; + char ptys_pl[MLXSW_REG_PTYS_LEN]; + u32 eth_proto_cap; + u32 eth_proto_admin; + u32 eth_proto_oper; + int err; + + mlxsw_reg_ptys_pack(ptys_pl, mlxsw_sp_port->local_port, 0); + err = mlxsw_reg_query(mlxsw_sp->core, MLXSW_REG(ptys), ptys_pl); + if (err) { + netdev_err(dev, "Failed to get proto"); + return err; + } + mlxsw_reg_ptys_unpack(ptys_pl, ð_proto_cap, + ð_proto_admin, ð_proto_oper); + + cmd->supported = mlxsw_sp_from_ptys_supported_port(eth_proto_cap) | + mlxsw_sp_from_ptys_supported_link(eth_proto_cap) | + SUPPORTED_Pause | SUPPORTED_Asym_Pause; + cmd->advertising = mlxsw_sp_from_ptys_advert_link(eth_proto_admin); + mlxsw_sp_from_ptys_speed_duplex(netif_carrier_ok(dev), + eth_proto_oper, cmd); + + eth_proto_oper = eth_proto_oper ? eth_proto_oper : eth_proto_cap; + cmd->port = mlxsw_sp_port_connector_port(eth_proto_oper); + cmd->lp_advertising = mlxsw_sp_from_ptys_advert_link(eth_proto_oper); + + cmd->transceiver = XCVR_INTERNAL; + return 0; +} + +static u32 mlxsw_sp_to_ptys_advert_link(u32 advertising) +{ + u32 ptys_proto = 0; + int i; + + for (i = 0; i < MLXSW_SP_PORT_LINK_MODE_LEN; i++) { + if (advertising & mlxsw_sp_port_link_mode[i].advertised) + ptys_proto |= mlxsw_sp_port_link_mode[i].mask; + } + return ptys_proto; +} + +static u32 mlxsw_sp_to_ptys_speed(u32 speed) +{ + u32 ptys_proto = 0; + int i; + + for (i = 0; i < MLXSW_SP_PORT_LINK_MODE_LEN; i++) { + if (speed == mlxsw_sp_port_link_mode[i].speed) + ptys_proto |= mlxsw_sp_port_link_mode[i].mask; + } + return ptys_proto; +} + +static int mlxsw_sp_port_set_settings(struct net_device *dev, + struct ethtool_cmd *cmd) +{ + struct mlxsw_sp_port *mlxsw_sp_port = netdev_priv(dev); + struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; + char ptys_pl[MLXSW_REG_PTYS_LEN]; + u32 speed; + u32 eth_proto_new; + u32 eth_proto_cap; + u32 eth_proto_admin; + bool is_up; + int err; + + speed = ethtool_cmd_speed(cmd); + + eth_proto_new = cmd->autoneg == AUTONEG_ENABLE ? + mlxsw_sp_to_ptys_advert_link(cmd->advertising) : + mlxsw_sp_to_ptys_speed(speed); + + mlxsw_reg_ptys_pack(ptys_pl, mlxsw_sp_port->local_port, 0); + err = mlxsw_reg_query(mlxsw_sp->core, MLXSW_REG(ptys), ptys_pl); + if (err) { + netdev_err(dev, "Failed to get proto"); + return err; + } + mlxsw_reg_ptys_unpack(ptys_pl, ð_proto_cap, ð_proto_admin, NULL); + + eth_proto_new = eth_proto_new & eth_proto_cap; + if (!eth_proto_new) { + netdev_err(dev, "Not supported proto admin requested"); + return -EINVAL; + } + if (eth_proto_new == eth_proto_admin) + return 0; + + mlxsw_reg_ptys_pack(ptys_pl, mlxsw_sp_port->local_port, eth_proto_new); + err = mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(ptys), ptys_pl); + if (err) { + netdev_err(dev, "Failed to set proto admin"); + return err; + } + + err = mlxsw_sp_port_oper_status_get(mlxsw_sp_port, &is_up); + if (err) { + netdev_err(dev, "Failed to get oper status"); + return err; + } + if (!is_up) + return 0; + + err = mlxsw_sp_port_admin_status_set(mlxsw_sp_port, false); + if (err) { + netdev_err(dev, "Failed to set admin status"); + return err; + } + + err = mlxsw_sp_port_admin_status_set(mlxsw_sp_port, true); + if (err) { + netdev_err(dev, "Failed to set admin status"); + return err; + } + + return 0; +} + +static const struct ethtool_ops mlxsw_sp_port_ethtool_ops = { + .get_drvinfo = mlxsw_sp_port_get_drvinfo, + .get_link = ethtool_op_get_link, + .get_strings = mlxsw_sp_port_get_strings, + .get_ethtool_stats = mlxsw_sp_port_get_stats, + .get_sset_count = mlxsw_sp_port_get_sset_count, + .get_settings = mlxsw_sp_port_get_settings, + .set_settings = mlxsw_sp_port_set_settings, +}; + +static int mlxsw_sp_port_create(struct mlxsw_sp *mlxsw_sp, u8 local_port) +{ + struct mlxsw_sp_port *mlxsw_sp_port; + struct net_device *dev; + bool usable; + int err; + + dev = alloc_etherdev(sizeof(struct mlxsw_sp_port)); + if (!dev) + return -ENOMEM; + mlxsw_sp_port = netdev_priv(dev); + mlxsw_sp_port->dev = dev; + mlxsw_sp_port->mlxsw_sp = mlxsw_sp; + mlxsw_sp_port->local_port = local_port; + mlxsw_sp_port->learning = 1; + mlxsw_sp_port->learning_sync = 1; + mlxsw_sp_port->uc_flood = 1; + mlxsw_sp_port->pvid = 1; + + mlxsw_sp_port->pcpu_stats = + netdev_alloc_pcpu_stats(struct mlxsw_sp_port_pcpu_stats); + if (!mlxsw_sp_port->pcpu_stats) { + err = -ENOMEM; + goto err_alloc_stats; + } + + dev->netdev_ops = &mlxsw_sp_port_netdev_ops; + dev->ethtool_ops = &mlxsw_sp_port_ethtool_ops; + + err = mlxsw_sp_port_dev_addr_init(mlxsw_sp_port); + if (err) { + dev_err(mlxsw_sp->bus_info->dev, "Port %d: Unable to init port mac address\n", + mlxsw_sp_port->local_port); + goto err_dev_addr_init; + } + + netif_carrier_off(dev); + + dev->features |= NETIF_F_NETNS_LOCAL | NETIF_F_LLTX | NETIF_F_SG | + NETIF_F_HW_VLAN_CTAG_FILTER; + + /* Each packet needs to have a Tx header (metadata) on top all other + * headers. + */ + dev->hard_header_len += MLXSW_TXHDR_LEN; + + err = mlxsw_sp_port_module_check(mlxsw_sp_port, &usable); + if (err) { + dev_err(mlxsw_sp->bus_info->dev, "Port %d: Failed to check module\n", + mlxsw_sp_port->local_port); + goto err_port_module_check; + } + + if (!usable) { + dev_dbg(mlxsw_sp->bus_info->dev, "Port %d: Not usable, skipping initialization\n", + mlxsw_sp_port->local_port); + goto port_not_usable; + } + + err = mlxsw_sp_port_system_port_mapping_set(mlxsw_sp_port); + if (err) { + dev_err(mlxsw_sp->bus_info->dev, "Port %d: Failed to set system port mapping\n", + mlxsw_sp_port->local_port); + goto err_port_system_port_mapping_set; + } + + err = mlxsw_sp_port_swid_set(mlxsw_sp_port, 0); + if (err) { + dev_err(mlxsw_sp->bus_info->dev, "Port %d: Failed to set SWID\n", + mlxsw_sp_port->local_port); + goto err_port_swid_set; + } + + err = mlxsw_sp_port_mtu_set(mlxsw_sp_port, ETH_DATA_LEN); + if (err) { + dev_err(mlxsw_sp->bus_info->dev, "Port %d: Failed to set MTU\n", + mlxsw_sp_port->local_port); + goto err_port_mtu_set; + } + + err = mlxsw_sp_port_admin_status_set(mlxsw_sp_port, false); + if (err) + goto err_port_admin_status_set; + + err = mlxsw_sp_port_buffers_init(mlxsw_sp_port); + if (err) { + dev_err(mlxsw_sp->bus_info->dev, "Port %d: Failed to initialize buffers\n", + mlxsw_sp_port->local_port); + goto err_port_buffers_init; + } + + mlxsw_sp_port_switchdev_init(mlxsw_sp_port); + err = register_netdev(dev); + if (err) { + dev_err(mlxsw_sp->bus_info->dev, "Port %d: Failed to register netdev\n", + mlxsw_sp_port->local_port); + goto err_register_netdev; + } + + err = mlxsw_sp_port_vlan_init(mlxsw_sp_port); + if (err) + goto err_port_vlan_init; + + mlxsw_sp->ports[local_port] = mlxsw_sp_port; + return 0; + +err_port_vlan_init: + unregister_netdev(dev); +err_register_netdev: +err_port_buffers_init: +err_port_admin_status_set: +err_port_mtu_set: +err_port_swid_set: +err_port_system_port_mapping_set: +port_not_usable: +err_port_module_check: +err_dev_addr_init: + free_percpu(mlxsw_sp_port->pcpu_stats); +err_alloc_stats: + free_netdev(dev); + return err; +} + +static void mlxsw_sp_vfids_fini(struct mlxsw_sp *mlxsw_sp) +{ + u16 vfid; + + for_each_set_bit(vfid, mlxsw_sp->active_vfids, VLAN_N_VID) + mlxsw_sp_vfid_destroy(mlxsw_sp, vfid); +} + +static void mlxsw_sp_port_remove(struct mlxsw_sp *mlxsw_sp, u8 local_port) +{ + struct mlxsw_sp_port *mlxsw_sp_port = mlxsw_sp->ports[local_port]; + + if (!mlxsw_sp_port) + return; + mlxsw_sp_port_kill_vid(mlxsw_sp_port->dev, 0, 1); + unregister_netdev(mlxsw_sp_port->dev); /* This calls ndo_stop */ + mlxsw_sp_port_switchdev_fini(mlxsw_sp_port); + free_percpu(mlxsw_sp_port->pcpu_stats); + free_netdev(mlxsw_sp_port->dev); +} + +static void mlxsw_sp_ports_remove(struct mlxsw_sp *mlxsw_sp) +{ + int i; + + for (i = 1; i < MLXSW_PORT_MAX_PORTS; i++) + mlxsw_sp_port_remove(mlxsw_sp, i); + kfree(mlxsw_sp->ports); +} + +static int mlxsw_sp_ports_create(struct mlxsw_sp *mlxsw_sp) +{ + size_t alloc_size; + int i; + int err; + + alloc_size = sizeof(struct mlxsw_sp_port *) * MLXSW_PORT_MAX_PORTS; + mlxsw_sp->ports = kzalloc(alloc_size, GFP_KERNEL); + if (!mlxsw_sp->ports) + return -ENOMEM; + + for (i = 1; i < MLXSW_PORT_MAX_PORTS; i++) { + err = mlxsw_sp_port_create(mlxsw_sp, i); + if (err) + goto err_port_create; + } + return 0; + +err_port_create: + for (i--; i >= 1; i--) + mlxsw_sp_port_remove(mlxsw_sp, i); + kfree(mlxsw_sp->ports); + return err; +} + +static void mlxsw_sp_pude_event_func(const struct mlxsw_reg_info *reg, + char *pude_pl, void *priv) +{ + struct mlxsw_sp *mlxsw_sp = priv; + struct mlxsw_sp_port *mlxsw_sp_port; + enum mlxsw_reg_pude_oper_status status; + u8 local_port; + + local_port = mlxsw_reg_pude_local_port_get(pude_pl); + mlxsw_sp_port = mlxsw_sp->ports[local_port]; + if (!mlxsw_sp_port) { + dev_warn(mlxsw_sp->bus_info->dev, "Port %d: Link event received for non-existent port\n", + local_port); + return; + } + + status = mlxsw_reg_pude_oper_status_get(pude_pl); + if (status == MLXSW_PORT_OPER_STATUS_UP) { + netdev_info(mlxsw_sp_port->dev, "link up\n"); + netif_carrier_on(mlxsw_sp_port->dev); + } else { + netdev_info(mlxsw_sp_port->dev, "link down\n"); + netif_carrier_off(mlxsw_sp_port->dev); + } +} + +static struct mlxsw_event_listener mlxsw_sp_pude_event = { + .func = mlxsw_sp_pude_event_func, + .trap_id = MLXSW_TRAP_ID_PUDE, +}; + +static int mlxsw_sp_event_register(struct mlxsw_sp *mlxsw_sp, + enum mlxsw_event_trap_id trap_id) +{ + struct mlxsw_event_listener *el; + char hpkt_pl[MLXSW_REG_HPKT_LEN]; + int err; + + switch (trap_id) { + case MLXSW_TRAP_ID_PUDE: + el = &mlxsw_sp_pude_event; + break; + } + err = mlxsw_core_event_listener_register(mlxsw_sp->core, el, mlxsw_sp); + if (err) + return err; + + mlxsw_reg_hpkt_pack(hpkt_pl, MLXSW_REG_HPKT_ACTION_FORWARD, trap_id); + err = mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(hpkt), hpkt_pl); + if (err) + goto err_event_trap_set; + + return 0; + +err_event_trap_set: + mlxsw_core_event_listener_unregister(mlxsw_sp->core, el, mlxsw_sp); + return err; +} + +static void mlxsw_sp_event_unregister(struct mlxsw_sp *mlxsw_sp, + enum mlxsw_event_trap_id trap_id) +{ + struct mlxsw_event_listener *el; + + switch (trap_id) { + case MLXSW_TRAP_ID_PUDE: + el = &mlxsw_sp_pude_event; + break; + } + mlxsw_core_event_listener_unregister(mlxsw_sp->core, el, mlxsw_sp); +} + +static void mlxsw_sp_rx_listener_func(struct sk_buff *skb, u8 local_port, + void *priv) +{ + struct mlxsw_sp *mlxsw_sp = priv; + struct mlxsw_sp_port *mlxsw_sp_port = mlxsw_sp->ports[local_port]; + struct mlxsw_sp_port_pcpu_stats *pcpu_stats; + + if (unlikely(!mlxsw_sp_port)) { + dev_warn_ratelimited(mlxsw_sp->bus_info->dev, "Port %d: skb received for non-existent port\n", + local_port); + return; + } + + skb->dev = mlxsw_sp_port->dev; + + pcpu_stats = this_cpu_ptr(mlxsw_sp_port->pcpu_stats); + u64_stats_update_begin(&pcpu_stats->syncp); + pcpu_stats->rx_packets++; + pcpu_stats->rx_bytes += skb->len; + u64_stats_update_end(&pcpu_stats->syncp); + + skb->protocol = eth_type_trans(skb, skb->dev); + netif_receive_skb(skb); +} + +static const struct mlxsw_rx_listener mlxsw_sp_rx_listener[] = { + { + .func = mlxsw_sp_rx_listener_func, + .local_port = MLXSW_PORT_DONT_CARE, + .trap_id = MLXSW_TRAP_ID_FDB_MC, + }, + /* Traps for specific L2 packet types, not trapped as FDB MC */ + { + .func = mlxsw_sp_rx_listener_func, + .local_port = MLXSW_PORT_DONT_CARE, + .trap_id = MLXSW_TRAP_ID_STP, + }, + { + .func = mlxsw_sp_rx_listener_func, + .local_port = MLXSW_PORT_DONT_CARE, + .trap_id = MLXSW_TRAP_ID_LACP, + }, + { + .func = mlxsw_sp_rx_listener_func, + .local_port = MLXSW_PORT_DONT_CARE, + .trap_id = MLXSW_TRAP_ID_EAPOL, + }, + { + .func = mlxsw_sp_rx_listener_func, + .local_port = MLXSW_PORT_DONT_CARE, + .trap_id = MLXSW_TRAP_ID_LLDP, + }, + { + .func = mlxsw_sp_rx_listener_func, + .local_port = MLXSW_PORT_DONT_CARE, + .trap_id = MLXSW_TRAP_ID_MMRP, + }, + { + .func = mlxsw_sp_rx_listener_func, + .local_port = MLXSW_PORT_DONT_CARE, + .trap_id = MLXSW_TRAP_ID_MVRP, + }, + { + .func = mlxsw_sp_rx_listener_func, + .local_port = MLXSW_PORT_DONT_CARE, + .trap_id = MLXSW_TRAP_ID_RPVST, + }, + { + .func = mlxsw_sp_rx_listener_func, + .local_port = MLXSW_PORT_DONT_CARE, + .trap_id = MLXSW_TRAP_ID_DHCP, + }, + { + .func = mlxsw_sp_rx_listener_func, + .local_port = MLXSW_PORT_DONT_CARE, + .trap_id = MLXSW_TRAP_ID_IGMP_QUERY, + }, + { + .func = mlxsw_sp_rx_listener_func, + .local_port = MLXSW_PORT_DONT_CARE, + .trap_id = MLXSW_TRAP_ID_IGMP_V1_REPORT, + }, + { + .func = mlxsw_sp_rx_listener_func, + .local_port = MLXSW_PORT_DONT_CARE, + .trap_id = MLXSW_TRAP_ID_IGMP_V2_REPORT, + }, + { + .func = mlxsw_sp_rx_listener_func, + .local_port = MLXSW_PORT_DONT_CARE, + .trap_id = MLXSW_TRAP_ID_IGMP_V2_LEAVE, + }, + { + .func = mlxsw_sp_rx_listener_func, + .local_port = MLXSW_PORT_DONT_CARE, + .trap_id = MLXSW_TRAP_ID_IGMP_V3_REPORT, + }, +}; + +static int mlxsw_sp_traps_init(struct mlxsw_sp *mlxsw_sp) +{ + char htgt_pl[MLXSW_REG_HTGT_LEN]; + char hpkt_pl[MLXSW_REG_HPKT_LEN]; + int i; + int err; + + mlxsw_reg_htgt_pack(htgt_pl, MLXSW_REG_HTGT_TRAP_GROUP_RX); + err = mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(htgt), htgt_pl); + if (err) + return err; + + mlxsw_reg_htgt_pack(htgt_pl, MLXSW_REG_HTGT_TRAP_GROUP_CTRL); + err = mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(htgt), htgt_pl); + if (err) + return err; + + for (i = 0; i < ARRAY_SIZE(mlxsw_sp_rx_listener); i++) { + err = mlxsw_core_rx_listener_register(mlxsw_sp->core, + &mlxsw_sp_rx_listener[i], + mlxsw_sp); + if (err) + goto err_rx_listener_register; + + mlxsw_reg_hpkt_pack(hpkt_pl, MLXSW_REG_HPKT_ACTION_TRAP_TO_CPU, + mlxsw_sp_rx_listener[i].trap_id); + err = mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(hpkt), hpkt_pl); + if (err) + goto err_rx_trap_set; + } + return 0; + +err_rx_trap_set: + mlxsw_core_rx_listener_unregister(mlxsw_sp->core, + &mlxsw_sp_rx_listener[i], + mlxsw_sp); +err_rx_listener_register: + for (i--; i >= 0; i--) { + mlxsw_reg_hpkt_pack(hpkt_pl, MLXSW_REG_HPKT_ACTION_FORWARD, + mlxsw_sp_rx_listener[i].trap_id); + mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(hpkt), hpkt_pl); + + mlxsw_core_rx_listener_unregister(mlxsw_sp->core, + &mlxsw_sp_rx_listener[i], + mlxsw_sp); + } + return err; +} + +static void mlxsw_sp_traps_fini(struct mlxsw_sp *mlxsw_sp) +{ + char hpkt_pl[MLXSW_REG_HPKT_LEN]; + int i; + + for (i = 0; i < ARRAY_SIZE(mlxsw_sp_rx_listener); i++) { + mlxsw_reg_hpkt_pack(hpkt_pl, MLXSW_REG_HPKT_ACTION_FORWARD, + mlxsw_sp_rx_listener[i].trap_id); + mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(hpkt), hpkt_pl); + + mlxsw_core_rx_listener_unregister(mlxsw_sp->core, + &mlxsw_sp_rx_listener[i], + mlxsw_sp); + } +} + +static int __mlxsw_sp_flood_init(struct mlxsw_core *mlxsw_core, + enum mlxsw_reg_sfgc_type type, + enum mlxsw_reg_sfgc_bridge_type bridge_type) +{ + enum mlxsw_flood_table_type table_type; + enum mlxsw_sp_flood_table flood_table; + char sfgc_pl[MLXSW_REG_SFGC_LEN]; + + if (bridge_type == MLXSW_REG_SFGC_BRIDGE_TYPE_VFID) { + table_type = MLXSW_REG_SFGC_TABLE_TYPE_FID; + flood_table = 0; + } else { + table_type = MLXSW_REG_SFGC_TABLE_TYPE_FID_OFFEST; + if (type == MLXSW_REG_SFGC_TYPE_UNKNOWN_UNICAST) + flood_table = MLXSW_SP_FLOOD_TABLE_UC; + else + flood_table = MLXSW_SP_FLOOD_TABLE_BM; + } + + mlxsw_reg_sfgc_pack(sfgc_pl, type, bridge_type, table_type, + flood_table); + return mlxsw_reg_write(mlxsw_core, MLXSW_REG(sfgc), sfgc_pl); +} + +static int mlxsw_sp_flood_init(struct mlxsw_sp *mlxsw_sp) +{ + int type, err; + + /* For non-offloaded netdevs, flood all traffic types to CPU + * port. + */ + for (type = 0; type < MLXSW_REG_SFGC_TYPE_MAX; type++) { + if (type == MLXSW_REG_SFGC_TYPE_RESERVED) + continue; + + err = __mlxsw_sp_flood_init(mlxsw_sp->core, type, + MLXSW_REG_SFGC_BRIDGE_TYPE_VFID); + if (err) + return err; + } + + /* For bridged ports, use one flooding table for unknown unicast + * traffic and a second table for unregistered multicast and + * broadcast. + */ + for (type = 0; type < MLXSW_REG_SFGC_TYPE_MAX; type++) { + if (type == MLXSW_REG_SFGC_TYPE_RESERVED) + continue; + + err = __mlxsw_sp_flood_init(mlxsw_sp->core, type, + MLXSW_REG_SFGC_BRIDGE_TYPE_1Q_FID); + if (err) + return err; + } + + return 0; +} + +static int mlxsw_sp_init(void *priv, struct mlxsw_core *mlxsw_core, + const struct mlxsw_bus_info *mlxsw_bus_info) +{ + struct mlxsw_sp *mlxsw_sp = priv; + int err; + + mlxsw_sp->core = mlxsw_core; + mlxsw_sp->bus_info = mlxsw_bus_info; + + err = mlxsw_sp_base_mac_get(mlxsw_sp); + if (err) { + dev_err(mlxsw_sp->bus_info->dev, "Failed to get base mac\n"); + return err; + } + + err = mlxsw_sp_ports_create(mlxsw_sp); + if (err) { + dev_err(mlxsw_sp->bus_info->dev, "Failed to create ports\n"); + goto err_ports_create; + } + + err = mlxsw_sp_event_register(mlxsw_sp, MLXSW_TRAP_ID_PUDE); + if (err) { + dev_err(mlxsw_sp->bus_info->dev, "Failed to register for PUDE events\n"); + goto err_event_register; + } + + err = mlxsw_sp_traps_init(mlxsw_sp); + if (err) { + dev_err(mlxsw_sp->bus_info->dev, "Failed to set traps for RX\n"); + goto err_rx_listener_register; + } + + err = mlxsw_sp_flood_init(mlxsw_sp); + if (err) { + dev_err(mlxsw_sp->bus_info->dev, "Failed to initialize flood tables\n"); + goto err_flood_init; + } + + err = mlxsw_sp_buffers_init(mlxsw_sp); + if (err) { + dev_err(mlxsw_sp->bus_info->dev, "Failed to initialize buffers\n"); + goto err_buffers_init; + } + + err = mlxsw_sp_switchdev_init(mlxsw_sp); + if (err) { + dev_err(mlxsw_sp->bus_info->dev, "Failed to initialize switchdev\n"); + goto err_switchdev_init; + } + + return 0; + +err_switchdev_init: +err_buffers_init: +err_flood_init: + mlxsw_sp_traps_fini(mlxsw_sp); +err_rx_listener_register: + mlxsw_sp_event_unregister(mlxsw_sp, MLXSW_TRAP_ID_PUDE); +err_event_register: + mlxsw_sp_ports_remove(mlxsw_sp); +err_ports_create: + mlxsw_sp_vfids_fini(mlxsw_sp); + return err; +} + +static void mlxsw_sp_fini(void *priv) +{ + struct mlxsw_sp *mlxsw_sp = priv; + + mlxsw_sp_switchdev_fini(mlxsw_sp); + mlxsw_sp_traps_fini(mlxsw_sp); + mlxsw_sp_event_unregister(mlxsw_sp, MLXSW_TRAP_ID_PUDE); + mlxsw_sp_ports_remove(mlxsw_sp); + mlxsw_sp_vfids_fini(mlxsw_sp); +} + +static struct mlxsw_config_profile mlxsw_sp_config_profile = { + .used_max_vepa_channels = 1, + .max_vepa_channels = 0, + .used_max_lag = 1, + .max_lag = 64, + .used_max_port_per_lag = 1, + .max_port_per_lag = 16, + .used_max_mid = 1, + .max_mid = 7000, + .used_max_pgt = 1, + .max_pgt = 0, + .used_max_system_port = 1, + .max_system_port = 64, + .used_max_vlan_groups = 1, + .max_vlan_groups = 127, + .used_max_regions = 1, + .max_regions = 400, + .used_flood_tables = 1, + .used_flood_mode = 1, + .flood_mode = 3, + .max_fid_offset_flood_tables = 2, + .fid_offset_flood_table_size = VLAN_N_VID - 1, + .max_fid_flood_tables = 1, + .fid_flood_table_size = VLAN_N_VID, + .used_max_ib_mc = 1, + .max_ib_mc = 0, + .used_max_pkey = 1, + .max_pkey = 0, + .swid_config = { + { + .used_type = 1, + .type = MLXSW_PORT_SWID_TYPE_ETH, + } + }, +}; + +static struct mlxsw_driver mlxsw_sp_driver = { + .kind = MLXSW_DEVICE_KIND_SPECTRUM, + .owner = THIS_MODULE, + .priv_size = sizeof(struct mlxsw_sp), + .init = mlxsw_sp_init, + .fini = mlxsw_sp_fini, + .txhdr_construct = mlxsw_sp_txhdr_construct, + .txhdr_len = MLXSW_TXHDR_LEN, + .profile = &mlxsw_sp_config_profile, +}; + +static bool mlxsw_sp_port_dev_check(const struct net_device *dev) +{ + return dev->netdev_ops == &mlxsw_sp_port_netdev_ops; +} + +static int mlxsw_sp_port_bridge_join(struct mlxsw_sp_port *mlxsw_sp_port) +{ + struct net_device *dev = mlxsw_sp_port->dev; + int err; + + /* When port is not bridged untagged packets are tagged with + * PVID=VID=1, thereby creating an implicit VLAN interface in + * the device. Remove it and let bridge code take care of its + * own VLANs. + */ + err = mlxsw_sp_port_kill_vid(dev, 0, 1); + if (err) + netdev_err(dev, "Failed to remove VID 1\n"); + + return err; +} + +static int mlxsw_sp_port_bridge_leave(struct mlxsw_sp_port *mlxsw_sp_port) +{ + struct net_device *dev = mlxsw_sp_port->dev; + int err; + + /* Add implicit VLAN interface in the device, so that untagged + * packets will be classified to the default vFID. + */ + err = mlxsw_sp_port_add_vid(dev, 0, 1); + if (err) + netdev_err(dev, "Failed to add VID 1\n"); + + return err; +} + +static bool mlxsw_sp_master_bridge_check(struct mlxsw_sp *mlxsw_sp, + struct net_device *br_dev) +{ + return !mlxsw_sp->master_bridge.dev || + mlxsw_sp->master_bridge.dev == br_dev; +} + +static void mlxsw_sp_master_bridge_inc(struct mlxsw_sp *mlxsw_sp, + struct net_device *br_dev) +{ + mlxsw_sp->master_bridge.dev = br_dev; + mlxsw_sp->master_bridge.ref_count++; +} + +static void mlxsw_sp_master_bridge_dec(struct mlxsw_sp *mlxsw_sp, + struct net_device *br_dev) +{ + if (--mlxsw_sp->master_bridge.ref_count == 0) + mlxsw_sp->master_bridge.dev = NULL; +} + +static int mlxsw_sp_netdevice_event(struct notifier_block *unused, + unsigned long event, void *ptr) +{ + struct net_device *dev = netdev_notifier_info_to_dev(ptr); + struct netdev_notifier_changeupper_info *info; + struct mlxsw_sp_port *mlxsw_sp_port; + struct net_device *upper_dev; + struct mlxsw_sp *mlxsw_sp; + int err; + + if (!mlxsw_sp_port_dev_check(dev)) + return NOTIFY_DONE; + + mlxsw_sp_port = netdev_priv(dev); + mlxsw_sp = mlxsw_sp_port->mlxsw_sp; + info = ptr; + + switch (event) { + case NETDEV_PRECHANGEUPPER: + upper_dev = info->upper_dev; + /* HW limitation forbids to put ports to multiple bridges. */ + if (info->master && info->linking && + netif_is_bridge_master(upper_dev) && + !mlxsw_sp_master_bridge_check(mlxsw_sp, upper_dev)) + return NOTIFY_BAD; + break; + case NETDEV_CHANGEUPPER: + upper_dev = info->upper_dev; + if (info->master && + netif_is_bridge_master(upper_dev)) { + if (info->linking) { + err = mlxsw_sp_port_bridge_join(mlxsw_sp_port); + if (err) + netdev_err(dev, "Failed to join bridge\n"); + mlxsw_sp_master_bridge_inc(mlxsw_sp, upper_dev); + mlxsw_sp_port->bridged = 1; + } else { + err = mlxsw_sp_port_bridge_leave(mlxsw_sp_port); + if (err) + netdev_err(dev, "Failed to leave bridge\n"); + mlxsw_sp_port->bridged = 0; + mlxsw_sp_master_bridge_dec(mlxsw_sp, upper_dev); + } + } + break; + } + + return NOTIFY_DONE; +} + +static struct notifier_block mlxsw_sp_netdevice_nb __read_mostly = { + .notifier_call = mlxsw_sp_netdevice_event, +}; + +static int __init mlxsw_sp_module_init(void) +{ + int err; + + register_netdevice_notifier(&mlxsw_sp_netdevice_nb); + err = mlxsw_core_driver_register(&mlxsw_sp_driver); + if (err) + goto err_core_driver_register; + return 0; + +err_core_driver_register: + unregister_netdevice_notifier(&mlxsw_sp_netdevice_nb); + return err; +} + +static void __exit mlxsw_sp_module_exit(void) +{ + mlxsw_core_driver_unregister(&mlxsw_sp_driver); + unregister_netdevice_notifier(&mlxsw_sp_netdevice_nb); +} + +module_init(mlxsw_sp_module_init); +module_exit(mlxsw_sp_module_exit); + +MODULE_LICENSE("Dual BSD/GPL"); +MODULE_AUTHOR("Jiri Pirko <jiri@mellanox.com>"); +MODULE_DESCRIPTION("Mellanox Spectrum driver"); +MODULE_MLXSW_DRIVER_ALIAS(MLXSW_DEVICE_KIND_SPECTRUM); diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h new file mode 100644 index 000000000..4365c8bcc --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h @@ -0,0 +1,122 @@ +/* + * drivers/net/ethernet/mellanox/mlxsw/spectrum.h + * Copyright (c) 2015 Mellanox Technologies. All rights reserved. + * Copyright (c) 2015 Jiri Pirko <jiri@mellanox.com> + * Copyright (c) 2015 Ido Schimmel <idosch@mellanox.com> + * Copyright (c) 2015 Elad Raz <eladr@mellanox.com> + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. 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. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * 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. + */ + +#ifndef _MLXSW_SPECTRUM_H +#define _MLXSW_SPECTRUM_H + +#include <linux/types.h> +#include <linux/netdevice.h> +#include <linux/bitops.h> +#include <linux/if_vlan.h> +#include <net/switchdev.h> + +#include "core.h" + +#define MLXSW_SP_VFID_BASE VLAN_N_VID + +struct mlxsw_sp_port; + +struct mlxsw_sp { + unsigned long active_vfids[BITS_TO_LONGS(VLAN_N_VID)]; + unsigned long active_fids[BITS_TO_LONGS(VLAN_N_VID)]; + struct mlxsw_sp_port **ports; + struct mlxsw_core *core; + const struct mlxsw_bus_info *bus_info; + unsigned char base_mac[ETH_ALEN]; + struct { + struct delayed_work dw; +#define MLXSW_SP_DEFAULT_LEARNING_INTERVAL 100 + unsigned int interval; /* ms */ + } fdb_notify; +#define MLXSW_SP_DEFAULT_AGEING_TIME 300 + u32 ageing_time; + struct { + struct net_device *dev; + unsigned int ref_count; + } master_bridge; +}; + +struct mlxsw_sp_port_pcpu_stats { + u64 rx_packets; + u64 rx_bytes; + u64 tx_packets; + u64 tx_bytes; + struct u64_stats_sync syncp; + u32 tx_dropped; +}; + +struct mlxsw_sp_port { + struct net_device *dev; + struct mlxsw_sp_port_pcpu_stats __percpu *pcpu_stats; + struct mlxsw_sp *mlxsw_sp; + u8 local_port; + u8 stp_state; + u8 learning:1, + learning_sync:1, + uc_flood:1, + bridged:1; + u16 pvid; + /* 802.1Q bridge VLANs */ + unsigned long active_vlans[BITS_TO_LONGS(VLAN_N_VID)]; + /* VLAN interfaces */ + unsigned long active_vfids[BITS_TO_LONGS(VLAN_N_VID)]; + u16 nr_vfids; +}; + +enum mlxsw_sp_flood_table { + MLXSW_SP_FLOOD_TABLE_UC, + MLXSW_SP_FLOOD_TABLE_BM, +}; + +int mlxsw_sp_buffers_init(struct mlxsw_sp *mlxsw_sp); +int mlxsw_sp_port_buffers_init(struct mlxsw_sp_port *mlxsw_sp_port); + +int mlxsw_sp_switchdev_init(struct mlxsw_sp *mlxsw_sp); +void mlxsw_sp_switchdev_fini(struct mlxsw_sp *mlxsw_sp); +int mlxsw_sp_port_vlan_init(struct mlxsw_sp_port *mlxsw_sp_port); +void mlxsw_sp_port_switchdev_init(struct mlxsw_sp_port *mlxsw_sp_port); +void mlxsw_sp_port_switchdev_fini(struct mlxsw_sp_port *mlxsw_sp_port); +int mlxsw_sp_port_vid_to_fid_set(struct mlxsw_sp_port *mlxsw_sp_port, + enum mlxsw_reg_svfa_mt mt, bool valid, u16 fid, + u16 vid); +int mlxsw_sp_port_vlan_set(struct mlxsw_sp_port *mlxsw_sp_port, u16 vid_begin, + u16 vid_end, bool is_member, bool untagged); +int mlxsw_sp_port_add_vid(struct net_device *dev, __be16 __always_unused proto, + u16 vid); +int mlxsw_sp_port_kill_vid(struct net_device *dev, + __be16 __always_unused proto, u16 vid); + +#endif diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_buffers.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_buffers.c new file mode 100644 index 000000000..d59195e3f --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_buffers.c @@ -0,0 +1,422 @@ +/* + * drivers/net/ethernet/mellanox/mlxsw/spectrum_buffers.c + * Copyright (c) 2015 Mellanox Technologies. All rights reserved. + * Copyright (c) 2015 Jiri Pirko <jiri@mellanox.com> + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. 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. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * 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. + */ + +#include <linux/kernel.h> +#include <linux/types.h> + +#include "spectrum.h" +#include "core.h" +#include "port.h" +#include "reg.h" + +struct mlxsw_sp_pb { + u8 index; + u16 size; +}; + +#define MLXSW_SP_PB(_index, _size) \ + { \ + .index = _index, \ + .size = _size, \ + } + +static const struct mlxsw_sp_pb mlxsw_sp_pbs[] = { + MLXSW_SP_PB(0, 208), + MLXSW_SP_PB(1, 208), + MLXSW_SP_PB(2, 208), + MLXSW_SP_PB(3, 208), + MLXSW_SP_PB(4, 208), + MLXSW_SP_PB(5, 208), + MLXSW_SP_PB(6, 208), + MLXSW_SP_PB(7, 208), + MLXSW_SP_PB(9, 208), +}; + +#define MLXSW_SP_PBS_LEN ARRAY_SIZE(mlxsw_sp_pbs) + +static int mlxsw_sp_port_pb_init(struct mlxsw_sp_port *mlxsw_sp_port) +{ + char pbmc_pl[MLXSW_REG_PBMC_LEN]; + int i; + + mlxsw_reg_pbmc_pack(pbmc_pl, mlxsw_sp_port->local_port, + 0xffff, 0xffff / 2); + for (i = 0; i < MLXSW_SP_PBS_LEN; i++) { + const struct mlxsw_sp_pb *pb; + + pb = &mlxsw_sp_pbs[i]; + mlxsw_reg_pbmc_lossy_buffer_pack(pbmc_pl, pb->index, pb->size); + } + return mlxsw_reg_write(mlxsw_sp_port->mlxsw_sp->core, + MLXSW_REG(pbmc), pbmc_pl); +} + +#define MLXSW_SP_SB_BYTES_PER_CELL 96 + +struct mlxsw_sp_sb_pool { + u8 pool; + enum mlxsw_reg_sbpr_dir dir; + enum mlxsw_reg_sbpr_mode mode; + u32 size; +}; + +#define MLXSW_SP_SB_POOL_INGRESS_SIZE \ + ((15000000 - (2 * 20000 * MLXSW_PORT_MAX_PORTS)) / \ + MLXSW_SP_SB_BYTES_PER_CELL) +#define MLXSW_SP_SB_POOL_EGRESS_SIZE \ + ((14000000 - (8 * 1500 * MLXSW_PORT_MAX_PORTS)) / \ + MLXSW_SP_SB_BYTES_PER_CELL) + +#define MLXSW_SP_SB_POOL(_pool, _dir, _mode, _size) \ + { \ + .pool = _pool, \ + .dir = _dir, \ + .mode = _mode, \ + .size = _size, \ + } + +#define MLXSW_SP_SB_POOL_INGRESS(_pool, _size) \ + MLXSW_SP_SB_POOL(_pool, MLXSW_REG_SBPR_DIR_INGRESS, \ + MLXSW_REG_SBPR_MODE_DYNAMIC, _size) + +#define MLXSW_SP_SB_POOL_EGRESS(_pool, _size) \ + MLXSW_SP_SB_POOL(_pool, MLXSW_REG_SBPR_DIR_EGRESS, \ + MLXSW_REG_SBPR_MODE_DYNAMIC, _size) + +static const struct mlxsw_sp_sb_pool mlxsw_sp_sb_pools[] = { + MLXSW_SP_SB_POOL_INGRESS(0, MLXSW_SP_SB_POOL_INGRESS_SIZE), + MLXSW_SP_SB_POOL_INGRESS(1, 0), + MLXSW_SP_SB_POOL_INGRESS(2, 0), + MLXSW_SP_SB_POOL_INGRESS(3, 0), + MLXSW_SP_SB_POOL_EGRESS(0, MLXSW_SP_SB_POOL_EGRESS_SIZE), + MLXSW_SP_SB_POOL_EGRESS(1, 0), + MLXSW_SP_SB_POOL_EGRESS(2, 0), + MLXSW_SP_SB_POOL_EGRESS(2, MLXSW_SP_SB_POOL_EGRESS_SIZE), +}; + +#define MLXSW_SP_SB_POOLS_LEN ARRAY_SIZE(mlxsw_sp_sb_pools) + +static int mlxsw_sp_sb_pools_init(struct mlxsw_sp *mlxsw_sp) +{ + char sbpr_pl[MLXSW_REG_SBPR_LEN]; + int i; + int err; + + for (i = 0; i < MLXSW_SP_SB_POOLS_LEN; i++) { + const struct mlxsw_sp_sb_pool *pool; + + pool = &mlxsw_sp_sb_pools[i]; + mlxsw_reg_sbpr_pack(sbpr_pl, pool->pool, pool->dir, + pool->mode, pool->size); + err = mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sbpr), sbpr_pl); + if (err) + return err; + } + return 0; +} + +struct mlxsw_sp_sb_cm { + union { + u8 pg; + u8 tc; + } u; + enum mlxsw_reg_sbcm_dir dir; + u32 min_buff; + u32 max_buff; + u8 pool; +}; + +#define MLXSW_SP_SB_CM(_pg_tc, _dir, _min_buff, _max_buff, _pool) \ + { \ + .u.pg = _pg_tc, \ + .dir = _dir, \ + .min_buff = _min_buff, \ + .max_buff = _max_buff, \ + .pool = _pool, \ + } + +#define MLXSW_SP_SB_CM_INGRESS(_pg, _min_buff, _max_buff) \ + MLXSW_SP_SB_CM(_pg, MLXSW_REG_SBCM_DIR_INGRESS, \ + _min_buff, _max_buff, 0) + +#define MLXSW_SP_SB_CM_EGRESS(_tc, _min_buff, _max_buff) \ + MLXSW_SP_SB_CM(_tc, MLXSW_REG_SBCM_DIR_EGRESS, \ + _min_buff, _max_buff, 0) + +#define MLXSW_SP_CPU_PORT_SB_CM_EGRESS(_tc) \ + MLXSW_SP_SB_CM(_tc, MLXSW_REG_SBCM_DIR_EGRESS, 104, 2, 3) + +static const struct mlxsw_sp_sb_cm mlxsw_sp_sb_cms[] = { + MLXSW_SP_SB_CM_INGRESS(0, 10000 / MLXSW_SP_SB_BYTES_PER_CELL, 8), + MLXSW_SP_SB_CM_INGRESS(1, 0, 0), + MLXSW_SP_SB_CM_INGRESS(2, 0, 0), + MLXSW_SP_SB_CM_INGRESS(3, 0, 0), + MLXSW_SP_SB_CM_INGRESS(4, 0, 0), + MLXSW_SP_SB_CM_INGRESS(5, 0, 0), + MLXSW_SP_SB_CM_INGRESS(6, 0, 0), + MLXSW_SP_SB_CM_INGRESS(7, 0, 0), + MLXSW_SP_SB_CM_INGRESS(9, 20000 / MLXSW_SP_SB_BYTES_PER_CELL, 0xff), + MLXSW_SP_SB_CM_EGRESS(0, 1500 / MLXSW_SP_SB_BYTES_PER_CELL, 9), + MLXSW_SP_SB_CM_EGRESS(1, 1500 / MLXSW_SP_SB_BYTES_PER_CELL, 9), + MLXSW_SP_SB_CM_EGRESS(2, 1500 / MLXSW_SP_SB_BYTES_PER_CELL, 9), + MLXSW_SP_SB_CM_EGRESS(3, 1500 / MLXSW_SP_SB_BYTES_PER_CELL, 9), + MLXSW_SP_SB_CM_EGRESS(4, 1500 / MLXSW_SP_SB_BYTES_PER_CELL, 9), + MLXSW_SP_SB_CM_EGRESS(5, 1500 / MLXSW_SP_SB_BYTES_PER_CELL, 9), + MLXSW_SP_SB_CM_EGRESS(6, 1500 / MLXSW_SP_SB_BYTES_PER_CELL, 9), + MLXSW_SP_SB_CM_EGRESS(7, 1500 / MLXSW_SP_SB_BYTES_PER_CELL, 9), + MLXSW_SP_SB_CM_EGRESS(8, 0, 0), + MLXSW_SP_SB_CM_EGRESS(9, 0, 0), + MLXSW_SP_SB_CM_EGRESS(10, 0, 0), + MLXSW_SP_SB_CM_EGRESS(11, 0, 0), + MLXSW_SP_SB_CM_EGRESS(12, 0, 0), + MLXSW_SP_SB_CM_EGRESS(13, 0, 0), + MLXSW_SP_SB_CM_EGRESS(14, 0, 0), + MLXSW_SP_SB_CM_EGRESS(15, 0, 0), + MLXSW_SP_SB_CM_EGRESS(16, 1, 0xff), +}; + +#define MLXSW_SP_SB_CMS_LEN ARRAY_SIZE(mlxsw_sp_sb_cms) + +static const struct mlxsw_sp_sb_cm mlxsw_sp_cpu_port_sb_cms[] = { + MLXSW_SP_CPU_PORT_SB_CM_EGRESS(0), + MLXSW_SP_CPU_PORT_SB_CM_EGRESS(1), + MLXSW_SP_CPU_PORT_SB_CM_EGRESS(2), + MLXSW_SP_CPU_PORT_SB_CM_EGRESS(3), + MLXSW_SP_CPU_PORT_SB_CM_EGRESS(4), + MLXSW_SP_CPU_PORT_SB_CM_EGRESS(5), + MLXSW_SP_CPU_PORT_SB_CM_EGRESS(6), + MLXSW_SP_CPU_PORT_SB_CM_EGRESS(7), + MLXSW_SP_CPU_PORT_SB_CM_EGRESS(8), + MLXSW_SP_CPU_PORT_SB_CM_EGRESS(9), + MLXSW_SP_CPU_PORT_SB_CM_EGRESS(10), + MLXSW_SP_CPU_PORT_SB_CM_EGRESS(11), + MLXSW_SP_CPU_PORT_SB_CM_EGRESS(12), + MLXSW_SP_CPU_PORT_SB_CM_EGRESS(13), + MLXSW_SP_CPU_PORT_SB_CM_EGRESS(14), + MLXSW_SP_CPU_PORT_SB_CM_EGRESS(15), + MLXSW_SP_CPU_PORT_SB_CM_EGRESS(16), + MLXSW_SP_CPU_PORT_SB_CM_EGRESS(17), + MLXSW_SP_CPU_PORT_SB_CM_EGRESS(18), + MLXSW_SP_CPU_PORT_SB_CM_EGRESS(19), + MLXSW_SP_CPU_PORT_SB_CM_EGRESS(20), + MLXSW_SP_CPU_PORT_SB_CM_EGRESS(21), + MLXSW_SP_CPU_PORT_SB_CM_EGRESS(22), + MLXSW_SP_CPU_PORT_SB_CM_EGRESS(23), + MLXSW_SP_CPU_PORT_SB_CM_EGRESS(24), + MLXSW_SP_CPU_PORT_SB_CM_EGRESS(25), + MLXSW_SP_CPU_PORT_SB_CM_EGRESS(26), + MLXSW_SP_CPU_PORT_SB_CM_EGRESS(27), + MLXSW_SP_CPU_PORT_SB_CM_EGRESS(28), + MLXSW_SP_CPU_PORT_SB_CM_EGRESS(29), + MLXSW_SP_CPU_PORT_SB_CM_EGRESS(30), + MLXSW_SP_CPU_PORT_SB_CM_EGRESS(31), +}; + +#define MLXSW_SP_CPU_PORT_SB_MCS_LEN \ + ARRAY_SIZE(mlxsw_sp_cpu_port_sb_cms) + +static int mlxsw_sp_sb_cms_init(struct mlxsw_sp *mlxsw_sp, u8 local_port, + const struct mlxsw_sp_sb_cm *cms, + size_t cms_len) +{ + char sbcm_pl[MLXSW_REG_SBCM_LEN]; + int i; + int err; + + for (i = 0; i < cms_len; i++) { + const struct mlxsw_sp_sb_cm *cm; + + cm = &cms[i]; + mlxsw_reg_sbcm_pack(sbcm_pl, local_port, cm->u.pg, cm->dir, + cm->min_buff, cm->max_buff, cm->pool); + err = mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sbcm), sbcm_pl); + if (err) + return err; + } + return 0; +} + +static int mlxsw_sp_port_sb_cms_init(struct mlxsw_sp_port *mlxsw_sp_port) +{ + return mlxsw_sp_sb_cms_init(mlxsw_sp_port->mlxsw_sp, + mlxsw_sp_port->local_port, mlxsw_sp_sb_cms, + MLXSW_SP_SB_CMS_LEN); +} + +static int mlxsw_sp_cpu_port_sb_cms_init(struct mlxsw_sp *mlxsw_sp) +{ + return mlxsw_sp_sb_cms_init(mlxsw_sp, 0, mlxsw_sp_cpu_port_sb_cms, + MLXSW_SP_CPU_PORT_SB_MCS_LEN); +} + +struct mlxsw_sp_sb_pm { + u8 pool; + enum mlxsw_reg_sbpm_dir dir; + u32 min_buff; + u32 max_buff; +}; + +#define MLXSW_SP_SB_PM(_pool, _dir, _min_buff, _max_buff) \ + { \ + .pool = _pool, \ + .dir = _dir, \ + .min_buff = _min_buff, \ + .max_buff = _max_buff, \ + } + +#define MLXSW_SP_SB_PM_INGRESS(_pool, _min_buff, _max_buff) \ + MLXSW_SP_SB_PM(_pool, MLXSW_REG_SBPM_DIR_INGRESS, \ + _min_buff, _max_buff) + +#define MLXSW_SP_SB_PM_EGRESS(_pool, _min_buff, _max_buff) \ + MLXSW_SP_SB_PM(_pool, MLXSW_REG_SBPM_DIR_EGRESS, \ + _min_buff, _max_buff) + +static const struct mlxsw_sp_sb_pm mlxsw_sp_sb_pms[] = { + MLXSW_SP_SB_PM_INGRESS(0, 0, 0xff), + MLXSW_SP_SB_PM_INGRESS(1, 0, 0), + MLXSW_SP_SB_PM_INGRESS(2, 0, 0), + MLXSW_SP_SB_PM_INGRESS(3, 0, 0), + MLXSW_SP_SB_PM_EGRESS(0, 0, 7), + MLXSW_SP_SB_PM_EGRESS(1, 0, 0), + MLXSW_SP_SB_PM_EGRESS(2, 0, 0), + MLXSW_SP_SB_PM_EGRESS(3, 0, 0), +}; + +#define MLXSW_SP_SB_PMS_LEN ARRAY_SIZE(mlxsw_sp_sb_pms) + +static int mlxsw_sp_port_sb_pms_init(struct mlxsw_sp_port *mlxsw_sp_port) +{ + char sbpm_pl[MLXSW_REG_SBPM_LEN]; + int i; + int err; + + for (i = 0; i < MLXSW_SP_SB_PMS_LEN; i++) { + const struct mlxsw_sp_sb_pm *pm; + + pm = &mlxsw_sp_sb_pms[i]; + mlxsw_reg_sbpm_pack(sbpm_pl, mlxsw_sp_port->local_port, + pm->pool, pm->dir, + pm->min_buff, pm->max_buff); + err = mlxsw_reg_write(mlxsw_sp_port->mlxsw_sp->core, + MLXSW_REG(sbpm), sbpm_pl); + if (err) + return err; + } + return 0; +} + +struct mlxsw_sp_sb_mm { + u8 prio; + u32 min_buff; + u32 max_buff; + u8 pool; +}; + +#define MLXSW_SP_SB_MM(_prio, _min_buff, _max_buff, _pool) \ + { \ + .prio = _prio, \ + .min_buff = _min_buff, \ + .max_buff = _max_buff, \ + .pool = _pool, \ + } + +static const struct mlxsw_sp_sb_mm mlxsw_sp_sb_mms[] = { + MLXSW_SP_SB_MM(0, 20000 / MLXSW_SP_SB_BYTES_PER_CELL, 0xff, 0), + MLXSW_SP_SB_MM(1, 20000 / MLXSW_SP_SB_BYTES_PER_CELL, 0xff, 0), + MLXSW_SP_SB_MM(2, 20000 / MLXSW_SP_SB_BYTES_PER_CELL, 0xff, 0), + MLXSW_SP_SB_MM(3, 20000 / MLXSW_SP_SB_BYTES_PER_CELL, 0xff, 0), + MLXSW_SP_SB_MM(4, 20000 / MLXSW_SP_SB_BYTES_PER_CELL, 0xff, 0), + MLXSW_SP_SB_MM(5, 20000 / MLXSW_SP_SB_BYTES_PER_CELL, 0xff, 0), + MLXSW_SP_SB_MM(6, 20000 / MLXSW_SP_SB_BYTES_PER_CELL, 0xff, 0), + MLXSW_SP_SB_MM(7, 20000 / MLXSW_SP_SB_BYTES_PER_CELL, 0xff, 0), + MLXSW_SP_SB_MM(8, 20000 / MLXSW_SP_SB_BYTES_PER_CELL, 0xff, 0), + MLXSW_SP_SB_MM(9, 20000 / MLXSW_SP_SB_BYTES_PER_CELL, 0xff, 0), + MLXSW_SP_SB_MM(10, 20000 / MLXSW_SP_SB_BYTES_PER_CELL, 0xff, 0), + MLXSW_SP_SB_MM(11, 20000 / MLXSW_SP_SB_BYTES_PER_CELL, 0xff, 0), + MLXSW_SP_SB_MM(12, 20000 / MLXSW_SP_SB_BYTES_PER_CELL, 0xff, 0), + MLXSW_SP_SB_MM(13, 20000 / MLXSW_SP_SB_BYTES_PER_CELL, 0xff, 0), + MLXSW_SP_SB_MM(14, 20000 / MLXSW_SP_SB_BYTES_PER_CELL, 0xff, 0), +}; + +#define MLXSW_SP_SB_MMS_LEN ARRAY_SIZE(mlxsw_sp_sb_mms) + +static int mlxsw_sp_sb_mms_init(struct mlxsw_sp *mlxsw_sp) +{ + char sbmm_pl[MLXSW_REG_SBMM_LEN]; + int i; + int err; + + for (i = 0; i < MLXSW_SP_SB_MMS_LEN; i++) { + const struct mlxsw_sp_sb_mm *mc; + + mc = &mlxsw_sp_sb_mms[i]; + mlxsw_reg_sbmm_pack(sbmm_pl, mc->prio, mc->min_buff, + mc->max_buff, mc->pool); + err = mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sbmm), sbmm_pl); + if (err) + return err; + } + return 0; +} + +int mlxsw_sp_buffers_init(struct mlxsw_sp *mlxsw_sp) +{ + int err; + + err = mlxsw_sp_sb_pools_init(mlxsw_sp); + if (err) + return err; + err = mlxsw_sp_cpu_port_sb_cms_init(mlxsw_sp); + if (err) + return err; + err = mlxsw_sp_sb_mms_init(mlxsw_sp); + + return err; +} + +int mlxsw_sp_port_buffers_init(struct mlxsw_sp_port *mlxsw_sp_port) +{ + int err; + + err = mlxsw_sp_port_pb_init(mlxsw_sp_port); + if (err) + return err; + err = mlxsw_sp_port_sb_cms_init(mlxsw_sp_port); + if (err) + return err; + err = mlxsw_sp_port_sb_pms_init(mlxsw_sp_port); + + return err; +} diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c new file mode 100644 index 000000000..617fb22b5 --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c @@ -0,0 +1,903 @@ +/* + * drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c + * Copyright (c) 2015 Mellanox Technologies. All rights reserved. + * Copyright (c) 2015 Jiri Pirko <jiri@mellanox.com> + * Copyright (c) 2015 Ido Schimmel <idosch@mellanox.com> + * Copyright (c) 2015 Elad Raz <eladr@mellanox.com> + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. 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. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * 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. + */ + +#include <linux/kernel.h> +#include <linux/types.h> +#include <linux/netdevice.h> +#include <linux/etherdevice.h> +#include <linux/slab.h> +#include <linux/device.h> +#include <linux/skbuff.h> +#include <linux/if_vlan.h> +#include <linux/if_bridge.h> +#include <linux/workqueue.h> +#include <linux/jiffies.h> +#include <net/switchdev.h> + +#include "spectrum.h" +#include "core.h" +#include "reg.h" + +static int mlxsw_sp_port_attr_get(struct net_device *dev, + struct switchdev_attr *attr) +{ + struct mlxsw_sp_port *mlxsw_sp_port = netdev_priv(dev); + struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; + + switch (attr->id) { + case SWITCHDEV_ATTR_ID_PORT_PARENT_ID: + attr->u.ppid.id_len = sizeof(mlxsw_sp->base_mac); + memcpy(&attr->u.ppid.id, &mlxsw_sp->base_mac, + attr->u.ppid.id_len); + break; + case SWITCHDEV_ATTR_ID_PORT_BRIDGE_FLAGS: + attr->u.brport_flags = + (mlxsw_sp_port->learning ? BR_LEARNING : 0) | + (mlxsw_sp_port->learning_sync ? BR_LEARNING_SYNC : 0) | + (mlxsw_sp_port->uc_flood ? BR_FLOOD : 0); + break; + default: + return -EOPNOTSUPP; + } + + return 0; +} + +static int mlxsw_sp_port_stp_state_set(struct mlxsw_sp_port *mlxsw_sp_port, + u8 state) +{ + struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; + enum mlxsw_reg_spms_state spms_state; + char *spms_pl; + u16 vid; + int err; + + switch (state) { + case BR_STATE_DISABLED: /* fall-through */ + case BR_STATE_FORWARDING: + spms_state = MLXSW_REG_SPMS_STATE_FORWARDING; + break; + case BR_STATE_LISTENING: /* fall-through */ + case BR_STATE_LEARNING: + spms_state = MLXSW_REG_SPMS_STATE_LEARNING; + break; + case BR_STATE_BLOCKING: + spms_state = MLXSW_REG_SPMS_STATE_DISCARDING; + break; + default: + BUG(); + } + + spms_pl = kmalloc(MLXSW_REG_SPMS_LEN, GFP_KERNEL); + if (!spms_pl) + return -ENOMEM; + mlxsw_reg_spms_pack(spms_pl, mlxsw_sp_port->local_port); + for_each_set_bit(vid, mlxsw_sp_port->active_vlans, VLAN_N_VID) + mlxsw_reg_spms_vid_pack(spms_pl, vid, spms_state); + + err = mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(spms), spms_pl); + kfree(spms_pl); + return err; +} + +static int mlxsw_sp_port_attr_stp_state_set(struct mlxsw_sp_port *mlxsw_sp_port, + struct switchdev_trans *trans, + u8 state) +{ + if (switchdev_trans_ph_prepare(trans)) + return 0; + + mlxsw_sp_port->stp_state = state; + return mlxsw_sp_port_stp_state_set(mlxsw_sp_port, state); +} + +static int __mlxsw_sp_port_flood_set(struct mlxsw_sp_port *mlxsw_sp_port, + u16 fid_begin, u16 fid_end, bool set, + bool only_uc) +{ + struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; + u16 range = fid_end - fid_begin + 1; + char *sftr_pl; + int err; + + sftr_pl = kmalloc(MLXSW_REG_SFTR_LEN, GFP_KERNEL); + if (!sftr_pl) + return -ENOMEM; + + mlxsw_reg_sftr_pack(sftr_pl, MLXSW_SP_FLOOD_TABLE_UC, fid_begin, + MLXSW_REG_SFGC_TABLE_TYPE_FID_OFFEST, range, + mlxsw_sp_port->local_port, set); + err = mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sftr), sftr_pl); + if (err) + goto buffer_out; + + /* Flooding control allows one to decide whether a given port will + * flood unicast traffic for which there is no FDB entry. + */ + if (only_uc) + goto buffer_out; + + mlxsw_reg_sftr_pack(sftr_pl, MLXSW_SP_FLOOD_TABLE_BM, fid_begin, + MLXSW_REG_SFGC_TABLE_TYPE_FID_OFFEST, range, + mlxsw_sp_port->local_port, set); + err = mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sftr), sftr_pl); + +buffer_out: + kfree(sftr_pl); + return err; +} + +static int mlxsw_sp_port_uc_flood_set(struct mlxsw_sp_port *mlxsw_sp_port, + bool set) +{ + struct net_device *dev = mlxsw_sp_port->dev; + u16 vid, last_visited_vid; + int err; + + for_each_set_bit(vid, mlxsw_sp_port->active_vlans, VLAN_N_VID) { + err = __mlxsw_sp_port_flood_set(mlxsw_sp_port, vid, vid, set, + true); + if (err) { + last_visited_vid = vid; + goto err_port_flood_set; + } + } + + return 0; + +err_port_flood_set: + for_each_set_bit(vid, mlxsw_sp_port->active_vlans, last_visited_vid) + __mlxsw_sp_port_flood_set(mlxsw_sp_port, vid, vid, !set, true); + netdev_err(dev, "Failed to configure unicast flooding\n"); + return err; +} + +static int mlxsw_sp_port_attr_br_flags_set(struct mlxsw_sp_port *mlxsw_sp_port, + struct switchdev_trans *trans, + unsigned long brport_flags) +{ + unsigned long uc_flood = mlxsw_sp_port->uc_flood ? BR_FLOOD : 0; + bool set; + int err; + + if (switchdev_trans_ph_prepare(trans)) + return 0; + + if ((uc_flood ^ brport_flags) & BR_FLOOD) { + set = mlxsw_sp_port->uc_flood ? false : true; + err = mlxsw_sp_port_uc_flood_set(mlxsw_sp_port, set); + if (err) + return err; + } + + mlxsw_sp_port->uc_flood = brport_flags & BR_FLOOD ? 1 : 0; + mlxsw_sp_port->learning = brport_flags & BR_LEARNING ? 1 : 0; + mlxsw_sp_port->learning_sync = brport_flags & BR_LEARNING_SYNC ? 1 : 0; + + return 0; +} + +static int mlxsw_sp_ageing_set(struct mlxsw_sp *mlxsw_sp, u32 ageing_time) +{ + char sfdat_pl[MLXSW_REG_SFDAT_LEN]; + int err; + + mlxsw_reg_sfdat_pack(sfdat_pl, ageing_time); + err = mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sfdat), sfdat_pl); + if (err) + return err; + mlxsw_sp->ageing_time = ageing_time; + return 0; +} + +static int mlxsw_sp_port_attr_br_ageing_set(struct mlxsw_sp_port *mlxsw_sp_port, + struct switchdev_trans *trans, + unsigned long ageing_clock_t) +{ + struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; + unsigned long ageing_jiffies = clock_t_to_jiffies(ageing_clock_t); + u32 ageing_time = jiffies_to_msecs(ageing_jiffies) / 1000; + + if (switchdev_trans_ph_prepare(trans)) + return 0; + + return mlxsw_sp_ageing_set(mlxsw_sp, ageing_time); +} + +static int mlxsw_sp_port_attr_set(struct net_device *dev, + const struct switchdev_attr *attr, + struct switchdev_trans *trans) +{ + struct mlxsw_sp_port *mlxsw_sp_port = netdev_priv(dev); + int err = 0; + + switch (attr->id) { + case SWITCHDEV_ATTR_ID_PORT_STP_STATE: + err = mlxsw_sp_port_attr_stp_state_set(mlxsw_sp_port, trans, + attr->u.stp_state); + break; + case SWITCHDEV_ATTR_ID_PORT_BRIDGE_FLAGS: + err = mlxsw_sp_port_attr_br_flags_set(mlxsw_sp_port, trans, + attr->u.brport_flags); + break; + case SWITCHDEV_ATTR_ID_BRIDGE_AGEING_TIME: + err = mlxsw_sp_port_attr_br_ageing_set(mlxsw_sp_port, trans, + attr->u.ageing_time); + break; + default: + err = -EOPNOTSUPP; + break; + } + + return err; +} + +static int mlxsw_sp_port_pvid_set(struct mlxsw_sp_port *mlxsw_sp_port, u16 vid) +{ + struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; + char spvid_pl[MLXSW_REG_SPVID_LEN]; + + mlxsw_reg_spvid_pack(spvid_pl, mlxsw_sp_port->local_port, vid); + return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(spvid), spvid_pl); +} + +static int mlxsw_sp_fid_create(struct mlxsw_sp *mlxsw_sp, u16 fid) +{ + char sfmr_pl[MLXSW_REG_SFMR_LEN]; + int err; + + mlxsw_reg_sfmr_pack(sfmr_pl, MLXSW_REG_SFMR_OP_CREATE_FID, fid, fid); + err = mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sfmr), sfmr_pl); + + if (err) + return err; + + set_bit(fid, mlxsw_sp->active_fids); + return 0; +} + +static void mlxsw_sp_fid_destroy(struct mlxsw_sp *mlxsw_sp, u16 fid) +{ + char sfmr_pl[MLXSW_REG_SFMR_LEN]; + + clear_bit(fid, mlxsw_sp->active_fids); + + mlxsw_reg_sfmr_pack(sfmr_pl, MLXSW_REG_SFMR_OP_DESTROY_FID, + fid, fid); + mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sfmr), sfmr_pl); +} + +static int mlxsw_sp_port_fid_map(struct mlxsw_sp_port *mlxsw_sp_port, u16 fid) +{ + enum mlxsw_reg_svfa_mt mt; + + if (mlxsw_sp_port->nr_vfids) + mt = MLXSW_REG_SVFA_MT_PORT_VID_TO_FID; + else + mt = MLXSW_REG_SVFA_MT_VID_TO_FID; + + return mlxsw_sp_port_vid_to_fid_set(mlxsw_sp_port, mt, true, fid, fid); +} + +static int mlxsw_sp_port_fid_unmap(struct mlxsw_sp_port *mlxsw_sp_port, u16 fid) +{ + enum mlxsw_reg_svfa_mt mt; + + if (!mlxsw_sp_port->nr_vfids) + return 0; + + mt = MLXSW_REG_SVFA_MT_PORT_VID_TO_FID; + return mlxsw_sp_port_vid_to_fid_set(mlxsw_sp_port, mt, false, fid, fid); +} + +static int mlxsw_sp_port_add_vids(struct net_device *dev, u16 vid_begin, + u16 vid_end) +{ + u16 vid; + int err; + + for (vid = vid_begin; vid <= vid_end; vid++) { + err = mlxsw_sp_port_add_vid(dev, 0, vid); + if (err) + goto err_port_add_vid; + } + return 0; + +err_port_add_vid: + for (vid--; vid >= vid_begin; vid--) + mlxsw_sp_port_kill_vid(dev, 0, vid); + return err; +} + +static int __mlxsw_sp_port_vlans_add(struct mlxsw_sp_port *mlxsw_sp_port, + u16 vid_begin, u16 vid_end, + bool flag_untagged, bool flag_pvid) +{ + struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; + struct net_device *dev = mlxsw_sp_port->dev; + enum mlxsw_reg_svfa_mt mt; + u16 vid, vid_e; + int err; + + /* In case this is invoked with BRIDGE_FLAGS_SELF and port is + * not bridged, then packets ingressing through the port with + * the specified VIDs will be directed to CPU. + */ + if (!mlxsw_sp_port->bridged) + return mlxsw_sp_port_add_vids(dev, vid_begin, vid_end); + + for (vid = vid_begin; vid <= vid_end; vid++) { + if (!test_bit(vid, mlxsw_sp->active_fids)) { + err = mlxsw_sp_fid_create(mlxsw_sp, vid); + if (err) { + netdev_err(dev, "Failed to create FID=%d\n", + vid); + return err; + } + + /* When creating a FID, we set a VID to FID mapping + * regardless of the port's mode. + */ + mt = MLXSW_REG_SVFA_MT_VID_TO_FID; + err = mlxsw_sp_port_vid_to_fid_set(mlxsw_sp_port, mt, + true, vid, vid); + if (err) { + netdev_err(dev, "Failed to create FID=VID=%d mapping\n", + vid); + return err; + } + } + + /* Set FID mapping according to port's mode */ + err = mlxsw_sp_port_fid_map(mlxsw_sp_port, vid); + if (err) { + netdev_err(dev, "Failed to map FID=%d", vid); + return err; + } + } + + err = __mlxsw_sp_port_flood_set(mlxsw_sp_port, vid_begin, vid_end, + true, false); + if (err) { + netdev_err(dev, "Failed to configure flooding\n"); + return err; + } + + for (vid = vid_begin; vid <= vid_end; + vid += MLXSW_REG_SPVM_REC_MAX_COUNT) { + vid_e = min((u16) (vid + MLXSW_REG_SPVM_REC_MAX_COUNT - 1), + vid_end); + + err = mlxsw_sp_port_vlan_set(mlxsw_sp_port, vid, vid_e, true, + flag_untagged); + if (err) { + netdev_err(mlxsw_sp_port->dev, "Unable to add VIDs %d-%d\n", + vid, vid_e); + return err; + } + } + + vid = vid_begin; + if (flag_pvid && mlxsw_sp_port->pvid != vid) { + err = mlxsw_sp_port_pvid_set(mlxsw_sp_port, vid); + if (err) { + netdev_err(mlxsw_sp_port->dev, "Unable to add PVID %d\n", + vid); + return err; + } + mlxsw_sp_port->pvid = vid; + } + + /* Changing activity bits only if HW operation succeded */ + for (vid = vid_begin; vid <= vid_end; vid++) + set_bit(vid, mlxsw_sp_port->active_vlans); + + return mlxsw_sp_port_stp_state_set(mlxsw_sp_port, + mlxsw_sp_port->stp_state); +} + +static int mlxsw_sp_port_vlans_add(struct mlxsw_sp_port *mlxsw_sp_port, + const struct switchdev_obj_port_vlan *vlan, + struct switchdev_trans *trans) +{ + bool untagged_flag = vlan->flags & BRIDGE_VLAN_INFO_UNTAGGED; + bool pvid_flag = vlan->flags & BRIDGE_VLAN_INFO_PVID; + + if (switchdev_trans_ph_prepare(trans)) + return 0; + + return __mlxsw_sp_port_vlans_add(mlxsw_sp_port, + vlan->vid_begin, vlan->vid_end, + untagged_flag, pvid_flag); +} + +static int mlxsw_sp_port_fdb_op(struct mlxsw_sp_port *mlxsw_sp_port, + const char *mac, u16 vid, bool adding, + bool dynamic) +{ + enum mlxsw_reg_sfd_rec_policy policy; + enum mlxsw_reg_sfd_op op; + char *sfd_pl; + int err; + + if (!vid) + vid = mlxsw_sp_port->pvid; + + sfd_pl = kmalloc(MLXSW_REG_SFD_LEN, GFP_KERNEL); + if (!sfd_pl) + return -ENOMEM; + + policy = dynamic ? MLXSW_REG_SFD_REC_POLICY_DYNAMIC_ENTRY_INGRESS : + MLXSW_REG_SFD_REC_POLICY_STATIC_ENTRY; + op = adding ? MLXSW_REG_SFD_OP_WRITE_EDIT : + MLXSW_REG_SFD_OP_WRITE_REMOVE; + mlxsw_reg_sfd_pack(sfd_pl, op, 0); + mlxsw_reg_sfd_uc_pack(sfd_pl, 0, policy, + mac, vid, MLXSW_REG_SFD_REC_ACTION_NOP, + mlxsw_sp_port->local_port); + err = mlxsw_reg_write(mlxsw_sp_port->mlxsw_sp->core, MLXSW_REG(sfd), + sfd_pl); + kfree(sfd_pl); + + return err; +} + +static int +mlxsw_sp_port_fdb_static_add(struct mlxsw_sp_port *mlxsw_sp_port, + const struct switchdev_obj_port_fdb *fdb, + struct switchdev_trans *trans) +{ + if (switchdev_trans_ph_prepare(trans)) + return 0; + + return mlxsw_sp_port_fdb_op(mlxsw_sp_port, fdb->addr, fdb->vid, + true, false); +} + +static int mlxsw_sp_port_obj_add(struct net_device *dev, + const struct switchdev_obj *obj, + struct switchdev_trans *trans) +{ + struct mlxsw_sp_port *mlxsw_sp_port = netdev_priv(dev); + int err = 0; + + switch (obj->id) { + case SWITCHDEV_OBJ_ID_PORT_VLAN: + err = mlxsw_sp_port_vlans_add(mlxsw_sp_port, + SWITCHDEV_OBJ_PORT_VLAN(obj), + trans); + break; + case SWITCHDEV_OBJ_ID_PORT_FDB: + err = mlxsw_sp_port_fdb_static_add(mlxsw_sp_port, + SWITCHDEV_OBJ_PORT_FDB(obj), + trans); + break; + default: + err = -EOPNOTSUPP; + break; + } + + return err; +} + +static int mlxsw_sp_port_kill_vids(struct net_device *dev, u16 vid_begin, + u16 vid_end) +{ + u16 vid; + int err; + + for (vid = vid_begin; vid <= vid_end; vid++) { + err = mlxsw_sp_port_kill_vid(dev, 0, vid); + if (err) + return err; + } + + return 0; +} + +static int __mlxsw_sp_port_vlans_del(struct mlxsw_sp_port *mlxsw_sp_port, + u16 vid_begin, u16 vid_end, bool init) +{ + struct net_device *dev = mlxsw_sp_port->dev; + u16 vid, vid_e; + int err; + + /* In case this is invoked with BRIDGE_FLAGS_SELF and port is + * not bridged, then prevent packets ingressing through the + * port with the specified VIDs from being trapped to CPU. + */ + if (!init && !mlxsw_sp_port->bridged) + return mlxsw_sp_port_kill_vids(dev, vid_begin, vid_end); + + for (vid = vid_begin; vid <= vid_end; + vid += MLXSW_REG_SPVM_REC_MAX_COUNT) { + vid_e = min((u16) (vid + MLXSW_REG_SPVM_REC_MAX_COUNT - 1), + vid_end); + err = mlxsw_sp_port_vlan_set(mlxsw_sp_port, vid, vid_e, false, + false); + if (err) { + netdev_err(mlxsw_sp_port->dev, "Unable to del VIDs %d-%d\n", + vid, vid_e); + return err; + } + } + + if ((mlxsw_sp_port->pvid >= vid_begin) && + (mlxsw_sp_port->pvid <= vid_end)) { + /* Default VLAN is always 1 */ + mlxsw_sp_port->pvid = 1; + err = mlxsw_sp_port_pvid_set(mlxsw_sp_port, + mlxsw_sp_port->pvid); + if (err) { + netdev_err(mlxsw_sp_port->dev, "Unable to del PVID %d\n", + vid); + return err; + } + } + + if (init) + goto out; + + err = __mlxsw_sp_port_flood_set(mlxsw_sp_port, vid_begin, vid_end, + false, false); + if (err) { + netdev_err(dev, "Failed to clear flooding\n"); + return err; + } + + for (vid = vid_begin; vid <= vid_end; vid++) { + /* Remove FID mapping in case of Virtual mode */ + err = mlxsw_sp_port_fid_unmap(mlxsw_sp_port, vid); + if (err) { + netdev_err(dev, "Failed to unmap FID=%d", vid); + return err; + } + } + +out: + /* Changing activity bits only if HW operation succeded */ + for (vid = vid_begin; vid <= vid_end; vid++) + clear_bit(vid, mlxsw_sp_port->active_vlans); + + return 0; +} + +static int mlxsw_sp_port_vlans_del(struct mlxsw_sp_port *mlxsw_sp_port, + const struct switchdev_obj_port_vlan *vlan) +{ + return __mlxsw_sp_port_vlans_del(mlxsw_sp_port, + vlan->vid_begin, vlan->vid_end, false); +} + +static int +mlxsw_sp_port_fdb_static_del(struct mlxsw_sp_port *mlxsw_sp_port, + const struct switchdev_obj_port_fdb *fdb) +{ + return mlxsw_sp_port_fdb_op(mlxsw_sp_port, fdb->addr, fdb->vid, + false, false); +} + +static int mlxsw_sp_port_obj_del(struct net_device *dev, + const struct switchdev_obj *obj) +{ + struct mlxsw_sp_port *mlxsw_sp_port = netdev_priv(dev); + int err = 0; + + switch (obj->id) { + case SWITCHDEV_OBJ_ID_PORT_VLAN: + err = mlxsw_sp_port_vlans_del(mlxsw_sp_port, + SWITCHDEV_OBJ_PORT_VLAN(obj)); + break; + case SWITCHDEV_OBJ_ID_PORT_FDB: + err = mlxsw_sp_port_fdb_static_del(mlxsw_sp_port, + SWITCHDEV_OBJ_PORT_FDB(obj)); + break; + default: + err = -EOPNOTSUPP; + break; + } + + return err; +} + +static int mlxsw_sp_port_fdb_dump(struct mlxsw_sp_port *mlxsw_sp_port, + struct switchdev_obj_port_fdb *fdb, + switchdev_obj_dump_cb_t *cb) +{ + char *sfd_pl; + char mac[ETH_ALEN]; + u16 vid; + u8 local_port; + u8 num_rec; + int stored_err = 0; + int i; + int err; + + sfd_pl = kmalloc(MLXSW_REG_SFD_LEN, GFP_KERNEL); + if (!sfd_pl) + return -ENOMEM; + + mlxsw_reg_sfd_pack(sfd_pl, MLXSW_REG_SFD_OP_QUERY_DUMP, 0); + do { + mlxsw_reg_sfd_num_rec_set(sfd_pl, MLXSW_REG_SFD_REC_MAX_COUNT); + err = mlxsw_reg_query(mlxsw_sp_port->mlxsw_sp->core, + MLXSW_REG(sfd), sfd_pl); + if (err) + goto out; + + num_rec = mlxsw_reg_sfd_num_rec_get(sfd_pl); + + /* Even in case of error, we have to run the dump to the end + * so the session in firmware is finished. + */ + if (stored_err) + continue; + + for (i = 0; i < num_rec; i++) { + switch (mlxsw_reg_sfd_rec_type_get(sfd_pl, i)) { + case MLXSW_REG_SFD_REC_TYPE_UNICAST: + mlxsw_reg_sfd_uc_unpack(sfd_pl, i, mac, &vid, + &local_port); + if (local_port == mlxsw_sp_port->local_port) { + ether_addr_copy(fdb->addr, mac); + fdb->ndm_state = NUD_REACHABLE; + fdb->vid = vid; + err = cb(&fdb->obj); + if (err) + stored_err = err; + } + } + } + } while (num_rec == MLXSW_REG_SFD_REC_MAX_COUNT); + +out: + kfree(sfd_pl); + return stored_err ? stored_err : err; +} + +static int mlxsw_sp_port_vlan_dump(struct mlxsw_sp_port *mlxsw_sp_port, + struct switchdev_obj_port_vlan *vlan, + switchdev_obj_dump_cb_t *cb) +{ + u16 vid; + int err = 0; + + for_each_set_bit(vid, mlxsw_sp_port->active_vlans, VLAN_N_VID) { + vlan->flags = 0; + if (vid == mlxsw_sp_port->pvid) + vlan->flags |= BRIDGE_VLAN_INFO_PVID; + vlan->vid_begin = vid; + vlan->vid_end = vid; + err = cb(&vlan->obj); + if (err) + break; + } + return err; +} + +static int mlxsw_sp_port_obj_dump(struct net_device *dev, + struct switchdev_obj *obj, + switchdev_obj_dump_cb_t *cb) +{ + struct mlxsw_sp_port *mlxsw_sp_port = netdev_priv(dev); + int err = 0; + + switch (obj->id) { + case SWITCHDEV_OBJ_ID_PORT_VLAN: + err = mlxsw_sp_port_vlan_dump(mlxsw_sp_port, + SWITCHDEV_OBJ_PORT_VLAN(obj), cb); + break; + case SWITCHDEV_OBJ_ID_PORT_FDB: + err = mlxsw_sp_port_fdb_dump(mlxsw_sp_port, + SWITCHDEV_OBJ_PORT_FDB(obj), cb); + break; + default: + err = -EOPNOTSUPP; + break; + } + + return err; +} + +static const struct switchdev_ops mlxsw_sp_port_switchdev_ops = { + .switchdev_port_attr_get = mlxsw_sp_port_attr_get, + .switchdev_port_attr_set = mlxsw_sp_port_attr_set, + .switchdev_port_obj_add = mlxsw_sp_port_obj_add, + .switchdev_port_obj_del = mlxsw_sp_port_obj_del, + .switchdev_port_obj_dump = mlxsw_sp_port_obj_dump, +}; + +static void mlxsw_sp_fdb_notify_mac_process(struct mlxsw_sp *mlxsw_sp, + char *sfn_pl, int rec_index, + bool adding) +{ + struct mlxsw_sp_port *mlxsw_sp_port; + char mac[ETH_ALEN]; + u8 local_port; + u16 vid; + int err; + + mlxsw_reg_sfn_mac_unpack(sfn_pl, rec_index, mac, &vid, &local_port); + mlxsw_sp_port = mlxsw_sp->ports[local_port]; + if (!mlxsw_sp_port) { + dev_err_ratelimited(mlxsw_sp->bus_info->dev, "Incorrect local port in FDB notification\n"); + return; + } + + err = mlxsw_sp_port_fdb_op(mlxsw_sp_port, mac, vid, + adding && mlxsw_sp_port->learning, true); + if (err) { + if (net_ratelimit()) + netdev_err(mlxsw_sp_port->dev, "Failed to set FDB entry\n"); + return; + } + + if (mlxsw_sp_port->learning && mlxsw_sp_port->learning_sync) { + struct switchdev_notifier_fdb_info info; + unsigned long notifier_type; + + info.addr = mac; + info.vid = vid; + notifier_type = adding ? SWITCHDEV_FDB_ADD : SWITCHDEV_FDB_DEL; + call_switchdev_notifiers(notifier_type, mlxsw_sp_port->dev, + &info.info); + } +} + +static void mlxsw_sp_fdb_notify_rec_process(struct mlxsw_sp *mlxsw_sp, + char *sfn_pl, int rec_index) +{ + switch (mlxsw_reg_sfn_rec_type_get(sfn_pl, rec_index)) { + case MLXSW_REG_SFN_REC_TYPE_LEARNED_MAC: + mlxsw_sp_fdb_notify_mac_process(mlxsw_sp, sfn_pl, + rec_index, true); + break; + case MLXSW_REG_SFN_REC_TYPE_AGED_OUT_MAC: + mlxsw_sp_fdb_notify_mac_process(mlxsw_sp, sfn_pl, + rec_index, false); + break; + } +} + +static void mlxsw_sp_fdb_notify_work_schedule(struct mlxsw_sp *mlxsw_sp) +{ + schedule_delayed_work(&mlxsw_sp->fdb_notify.dw, + msecs_to_jiffies(mlxsw_sp->fdb_notify.interval)); +} + +static void mlxsw_sp_fdb_notify_work(struct work_struct *work) +{ + struct mlxsw_sp *mlxsw_sp; + char *sfn_pl; + u8 num_rec; + int i; + int err; + + sfn_pl = kmalloc(MLXSW_REG_SFN_LEN, GFP_KERNEL); + if (!sfn_pl) + return; + + mlxsw_sp = container_of(work, struct mlxsw_sp, fdb_notify.dw.work); + + do { + mlxsw_reg_sfn_pack(sfn_pl); + err = mlxsw_reg_query(mlxsw_sp->core, MLXSW_REG(sfn), sfn_pl); + if (err) { + dev_err_ratelimited(mlxsw_sp->bus_info->dev, "Failed to get FDB notifications\n"); + break; + } + num_rec = mlxsw_reg_sfn_num_rec_get(sfn_pl); + for (i = 0; i < num_rec; i++) + mlxsw_sp_fdb_notify_rec_process(mlxsw_sp, sfn_pl, i); + + } while (num_rec); + + kfree(sfn_pl); + mlxsw_sp_fdb_notify_work_schedule(mlxsw_sp); +} + +static int mlxsw_sp_fdb_init(struct mlxsw_sp *mlxsw_sp) +{ + int err; + + err = mlxsw_sp_ageing_set(mlxsw_sp, MLXSW_SP_DEFAULT_AGEING_TIME); + if (err) { + dev_err(mlxsw_sp->bus_info->dev, "Failed to set default ageing time\n"); + return err; + } + INIT_DELAYED_WORK(&mlxsw_sp->fdb_notify.dw, mlxsw_sp_fdb_notify_work); + mlxsw_sp->fdb_notify.interval = MLXSW_SP_DEFAULT_LEARNING_INTERVAL; + mlxsw_sp_fdb_notify_work_schedule(mlxsw_sp); + return 0; +} + +static void mlxsw_sp_fdb_fini(struct mlxsw_sp *mlxsw_sp) +{ + cancel_delayed_work_sync(&mlxsw_sp->fdb_notify.dw); +} + +static void mlxsw_sp_fids_fini(struct mlxsw_sp *mlxsw_sp) +{ + u16 fid; + + for_each_set_bit(fid, mlxsw_sp->active_fids, VLAN_N_VID) + mlxsw_sp_fid_destroy(mlxsw_sp, fid); +} + +int mlxsw_sp_switchdev_init(struct mlxsw_sp *mlxsw_sp) +{ + return mlxsw_sp_fdb_init(mlxsw_sp); +} + +void mlxsw_sp_switchdev_fini(struct mlxsw_sp *mlxsw_sp) +{ + mlxsw_sp_fdb_fini(mlxsw_sp); + mlxsw_sp_fids_fini(mlxsw_sp); +} + +int mlxsw_sp_port_vlan_init(struct mlxsw_sp_port *mlxsw_sp_port) +{ + struct net_device *dev = mlxsw_sp_port->dev; + int err; + + /* Allow only untagged packets to ingress and tag them internally + * with VID 1. + */ + mlxsw_sp_port->pvid = 1; + err = __mlxsw_sp_port_vlans_del(mlxsw_sp_port, 0, VLAN_N_VID, true); + if (err) { + netdev_err(dev, "Unable to init VLANs\n"); + return err; + } + + /* Add implicit VLAN interface in the device, so that untagged + * packets will be classified to the default vFID. + */ + err = mlxsw_sp_port_add_vid(dev, 0, 1); + if (err) + netdev_err(dev, "Failed to configure default vFID\n"); + + return err; +} + +void mlxsw_sp_port_switchdev_init(struct mlxsw_sp_port *mlxsw_sp_port) +{ + mlxsw_sp_port->dev->switchdev_ops = &mlxsw_sp_port_switchdev_ops; +} + +void mlxsw_sp_port_switchdev_fini(struct mlxsw_sp_port *mlxsw_sp_port) +{ +} diff --git a/drivers/net/ethernet/mellanox/mlxsw/switchx2.c b/drivers/net/ethernet/mellanox/mlxsw/switchx2.c index 62cbbd1ad..d85960cfb 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/switchx2.c +++ b/drivers/net/ethernet/mellanox/mlxsw/switchx2.c @@ -57,13 +57,11 @@ static const char mlxsw_sx_driver_version[] = "1.0"; struct mlxsw_sx_port; -#define MLXSW_SW_HW_ID_LEN 6 - struct mlxsw_sx { struct mlxsw_sx_port **ports; struct mlxsw_core *core; const struct mlxsw_bus_info *bus_info; - u8 hw_id[MLXSW_SW_HW_ID_LEN]; + u8 hw_id[ETH_ALEN]; }; struct mlxsw_sx_port_pcpu_stats { @@ -868,7 +866,7 @@ static int mlxsw_sx_port_attr_get(struct net_device *dev, struct mlxsw_sx *mlxsw_sx = mlxsw_sx_port->mlxsw_sx; switch (attr->id) { - case SWITCHDEV_ATTR_PORT_PARENT_ID: + case SWITCHDEV_ATTR_ID_PORT_PARENT_ID: attr->u.ppid.id_len = sizeof(mlxsw_sx->hw_id); memcpy(&attr->u.ppid.id, &mlxsw_sx->hw_id, attr->u.ppid.id_len); break; @@ -925,7 +923,8 @@ static int mlxsw_sx_port_stp_state_set(struct mlxsw_sx_port *mlxsw_sx_port, spms_pl = kmalloc(MLXSW_REG_SPMS_LEN, GFP_KERNEL); if (!spms_pl) return -ENOMEM; - mlxsw_reg_spms_pack(spms_pl, mlxsw_sx_port->local_port, vid, state); + mlxsw_reg_spms_pack(spms_pl, mlxsw_sx_port->local_port); + mlxsw_reg_spms_vid_pack(spms_pl, vid, state); err = mlxsw_reg_write(mlxsw_sx->core, MLXSW_REG(spms), spms_pl); kfree(spms_pl); return err; @@ -1148,7 +1147,7 @@ static void mlxsw_sx_pude_event_func(const struct mlxsw_reg_info *reg, } status = mlxsw_reg_pude_oper_status_get(pude_pl); - if (MLXSW_PORT_OPER_STATUS_UP == status) { + if (status == MLXSW_PORT_OPER_STATUS_UP) { netdev_info(mlxsw_sx_port->dev, "link up\n"); netif_carrier_on(mlxsw_sx_port->dev); } else { @@ -1178,8 +1177,7 @@ static int mlxsw_sx_event_register(struct mlxsw_sx *mlxsw_sx, if (err) return err; - mlxsw_reg_hpkt_pack(hpkt_pl, MLXSW_REG_HPKT_ACTION_FORWARD, - MLXSW_REG_HTGT_TRAP_GROUP_EMAD, trap_id); + mlxsw_reg_hpkt_pack(hpkt_pl, MLXSW_REG_HPKT_ACTION_FORWARD, trap_id); err = mlxsw_reg_write(mlxsw_sx->core, MLXSW_REG(hpkt), hpkt_pl); if (err) goto err_event_trap_set; @@ -1212,9 +1210,8 @@ static void mlxsw_sx_rx_listener_func(struct sk_buff *skb, u8 local_port, struct mlxsw_sx_port_pcpu_stats *pcpu_stats; if (unlikely(!mlxsw_sx_port)) { - if (net_ratelimit()) - dev_warn(mlxsw_sx->bus_info->dev, "Port %d: skb received for non-existent port\n", - local_port); + dev_warn_ratelimited(mlxsw_sx->bus_info->dev, "Port %d: skb received for non-existent port\n", + local_port); return; } @@ -1316,6 +1313,11 @@ static int mlxsw_sx_traps_init(struct mlxsw_sx *mlxsw_sx) if (err) return err; + mlxsw_reg_htgt_pack(htgt_pl, MLXSW_REG_HTGT_TRAP_GROUP_CTRL); + err = mlxsw_reg_write(mlxsw_sx->core, MLXSW_REG(htgt), htgt_pl); + if (err) + return err; + for (i = 0; i < ARRAY_SIZE(mlxsw_sx_rx_listener); i++) { err = mlxsw_core_rx_listener_register(mlxsw_sx->core, &mlxsw_sx_rx_listener[i], @@ -1324,7 +1326,6 @@ static int mlxsw_sx_traps_init(struct mlxsw_sx *mlxsw_sx) goto err_rx_listener_register; mlxsw_reg_hpkt_pack(hpkt_pl, MLXSW_REG_HPKT_ACTION_TRAP_TO_CPU, - MLXSW_REG_HTGT_TRAP_GROUP_RX, mlxsw_sx_rx_listener[i].trap_id); err = mlxsw_reg_write(mlxsw_sx->core, MLXSW_REG(hpkt), hpkt_pl); if (err) @@ -1339,7 +1340,6 @@ err_rx_trap_set: err_rx_listener_register: for (i--; i >= 0; i--) { mlxsw_reg_hpkt_pack(hpkt_pl, MLXSW_REG_HPKT_ACTION_FORWARD, - MLXSW_REG_HTGT_TRAP_GROUP_RX, mlxsw_sx_rx_listener[i].trap_id); mlxsw_reg_write(mlxsw_sx->core, MLXSW_REG(hpkt), hpkt_pl); @@ -1357,7 +1357,6 @@ static void mlxsw_sx_traps_fini(struct mlxsw_sx *mlxsw_sx) for (i = 0; i < ARRAY_SIZE(mlxsw_sx_rx_listener); i++) { mlxsw_reg_hpkt_pack(hpkt_pl, MLXSW_REG_HPKT_ACTION_FORWARD, - MLXSW_REG_HTGT_TRAP_GROUP_RX, mlxsw_sx_rx_listener[i].trap_id); mlxsw_reg_write(mlxsw_sx->core, MLXSW_REG(hpkt), hpkt_pl); @@ -1371,25 +1370,15 @@ static int mlxsw_sx_flood_init(struct mlxsw_sx *mlxsw_sx) { char sfgc_pl[MLXSW_REG_SFGC_LEN]; char sgcr_pl[MLXSW_REG_SGCR_LEN]; - char *smid_pl; char *sftr_pl; int err; - /* Due to FW bug, we must configure SMID. */ - smid_pl = kmalloc(MLXSW_REG_SMID_LEN, GFP_KERNEL); - if (!smid_pl) - return -ENOMEM; - mlxsw_reg_smid_pack(smid_pl, MLXSW_PORT_MID); - err = mlxsw_reg_write(mlxsw_sx->core, MLXSW_REG(smid), smid_pl); - kfree(smid_pl); - if (err) - return err; - /* Configure a flooding table, which includes only CPU port. */ sftr_pl = kmalloc(MLXSW_REG_SFTR_LEN, GFP_KERNEL); if (!sftr_pl) return -ENOMEM; - mlxsw_reg_sftr_pack(sftr_pl, 0, 0, MLXSW_REG_SFGC_TABLE_TYPE_SINGLE, 0); + mlxsw_reg_sftr_pack(sftr_pl, 0, 0, MLXSW_REG_SFGC_TABLE_TYPE_SINGLE, 0, + MLXSW_PORT_CPU_PORT, true); err = mlxsw_reg_write(mlxsw_sx->core, MLXSW_REG(sftr), sftr_pl); kfree(sftr_pl); if (err) diff --git a/drivers/net/ethernet/mellanox/mlxsw/txheader.h b/drivers/net/ethernet/mellanox/mlxsw/txheader.h index 06fc46c78..fdf94720c 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/txheader.h +++ b/drivers/net/ethernet/mellanox/mlxsw/txheader.h @@ -38,6 +38,7 @@ #define MLXSW_TXHDR_LEN 0x10 #define MLXSW_TXHDR_VERSION_0 0 +#define MLXSW_TXHDR_VERSION_1 1 enum { MLXSW_TXHDR_ETH_CTL, |