diff options
Diffstat (limited to 'drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c')
| -rw-r--r-- | drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c | 1857 |
1 files changed, 1487 insertions, 370 deletions
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c index 09e32778b012..a2033837182e 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c @@ -19,6 +19,7 @@ #include <linux/net_namespace.h> #include <linux/mutex.h> #include <linux/genalloc.h> +#include <linux/xarray.h> #include <net/netevent.h> #include <net/neighbour.h> #include <net/arp.h> @@ -51,10 +52,28 @@ struct mlxsw_sp_vr; struct mlxsw_sp_lpm_tree; struct mlxsw_sp_rif_ops; -struct mlxsw_sp_rif { +struct mlxsw_sp_crif_key { + struct net_device *dev; +}; + +struct mlxsw_sp_crif { + struct mlxsw_sp_crif_key key; + struct rhash_head ht_node; + bool can_destroy; struct list_head nexthop_list; + struct mlxsw_sp_rif *rif; +}; + +static const struct rhashtable_params mlxsw_sp_crif_ht_params = { + .key_offset = offsetof(struct mlxsw_sp_crif, key), + .key_len = sizeof_field(struct mlxsw_sp_crif, key), + .head_offset = offsetof(struct mlxsw_sp_crif, ht_node), +}; + +struct mlxsw_sp_rif { + struct mlxsw_sp_crif *crif; /* NULL for underlay RIF */ + netdevice_tracker dev_tracker; struct list_head neigh_list; - struct net_device *dev; /* NULL for underlay RIF */ struct mlxsw_sp_fid *fid; unsigned char addr[ETH_ALEN]; int mtu; @@ -71,6 +90,13 @@ struct mlxsw_sp_rif { bool counter_egress_valid; }; +static struct net_device *mlxsw_sp_rif_dev(const struct mlxsw_sp_rif *rif) +{ + if (!rif->crif) + return NULL; + return rif->crif->key.dev; +} + struct mlxsw_sp_rif_params { struct net_device *dev; union { @@ -96,8 +122,8 @@ struct mlxsw_sp_rif_subport { struct mlxsw_sp_rif_ipip_lb { struct mlxsw_sp_rif common; struct mlxsw_sp_rif_ipip_lb_config lb_config; - u16 ul_vr_id; /* Reserved for Spectrum-2. */ - u16 ul_rif_id; /* Reserved for Spectrum. */ + u16 ul_vr_id; /* Spectrum-1. */ + u16 ul_rif_id; /* Spectrum-2+. */ }; struct mlxsw_sp_rif_params_ipip_lb { @@ -115,6 +141,7 @@ struct mlxsw_sp_rif_ops { struct netlink_ext_ack *extack); void (*deconfigure)(struct mlxsw_sp_rif *rif); struct mlxsw_sp_fid * (*fid_get)(struct mlxsw_sp_rif *rif, + const struct mlxsw_sp_rif_params *params, struct netlink_ext_ack *extack); void (*fdb_del)(struct mlxsw_sp_rif *rif, const char *mac); }; @@ -475,7 +502,7 @@ struct mlxsw_sp_rt6 { struct mlxsw_sp_lpm_tree { u8 id; /* tree ID */ - unsigned int ref_count; + refcount_t ref_count; enum mlxsw_sp_l3proto proto; unsigned long prefix_ref_count[MLXSW_SP_PREFIX_COUNT]; struct mlxsw_sp_prefix_usage prefix_usage; @@ -552,7 +579,7 @@ mlxsw_sp_lpm_tree_find_unused(struct mlxsw_sp *mlxsw_sp) for (i = 0; i < mlxsw_sp->router->lpm.tree_count; i++) { lpm_tree = &mlxsw_sp->router->lpm.trees[i]; - if (lpm_tree->ref_count == 0) + if (refcount_read(&lpm_tree->ref_count) == 0) return lpm_tree; } return NULL; @@ -628,7 +655,7 @@ mlxsw_sp_lpm_tree_create(struct mlxsw_sp *mlxsw_sp, sizeof(lpm_tree->prefix_usage)); memset(&lpm_tree->prefix_ref_count, 0, sizeof(lpm_tree->prefix_ref_count)); - lpm_tree->ref_count = 1; + refcount_set(&lpm_tree->ref_count, 1); return lpm_tree; err_left_struct_set: @@ -652,7 +679,7 @@ mlxsw_sp_lpm_tree_get(struct mlxsw_sp *mlxsw_sp, for (i = 0; i < mlxsw_sp->router->lpm.tree_count; i++) { lpm_tree = &mlxsw_sp->router->lpm.trees[i]; - if (lpm_tree->ref_count != 0 && + if (refcount_read(&lpm_tree->ref_count) && lpm_tree->proto == proto && mlxsw_sp_prefix_usage_eq(&lpm_tree->prefix_usage, prefix_usage)) { @@ -665,14 +692,15 @@ mlxsw_sp_lpm_tree_get(struct mlxsw_sp *mlxsw_sp, static void mlxsw_sp_lpm_tree_hold(struct mlxsw_sp_lpm_tree *lpm_tree) { - lpm_tree->ref_count++; + refcount_inc(&lpm_tree->ref_count); } static void mlxsw_sp_lpm_tree_put(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_lpm_tree *lpm_tree) { - if (--lpm_tree->ref_count == 0) - mlxsw_sp_lpm_tree_destroy(mlxsw_sp, lpm_tree); + if (!refcount_dec_and_test(&lpm_tree->ref_count)) + return; + mlxsw_sp_lpm_tree_destroy(mlxsw_sp, lpm_tree); } #define MLXSW_SP_LPM_TREE_MIN 1 /* tree 0 is reserved */ @@ -748,10 +776,11 @@ static bool mlxsw_sp_vr_is_used(const struct mlxsw_sp_vr *vr) static struct mlxsw_sp_vr *mlxsw_sp_vr_find_unused(struct mlxsw_sp *mlxsw_sp) { + int max_vrs = MLXSW_CORE_RES_GET(mlxsw_sp->core, MAX_VRS); struct mlxsw_sp_vr *vr; int i; - for (i = 0; i < MLXSW_CORE_RES_GET(mlxsw_sp->core, MAX_VRS); i++) { + for (i = 0; i < max_vrs; i++) { vr = &mlxsw_sp->router->vrs[i]; if (!mlxsw_sp_vr_is_used(vr)) return vr; @@ -792,12 +821,13 @@ static u32 mlxsw_sp_fix_tb_id(u32 tb_id) static struct mlxsw_sp_vr *mlxsw_sp_vr_find(struct mlxsw_sp *mlxsw_sp, u32 tb_id) { + int max_vrs = MLXSW_CORE_RES_GET(mlxsw_sp->core, MAX_VRS); struct mlxsw_sp_vr *vr; int i; tb_id = mlxsw_sp_fix_tb_id(tb_id); - for (i = 0; i < MLXSW_CORE_RES_GET(mlxsw_sp->core, MAX_VRS); i++) { + for (i = 0; i < max_vrs; i++) { vr = &mlxsw_sp->router->vrs[i]; if (mlxsw_sp_vr_is_used(vr) && vr->tb_id == tb_id) return vr; @@ -959,6 +989,7 @@ static int mlxsw_sp_vrs_lpm_tree_replace(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_fib *fib, struct mlxsw_sp_lpm_tree *new_tree) { + int max_vrs = MLXSW_CORE_RES_GET(mlxsw_sp->core, MAX_VRS); enum mlxsw_sp_l3proto proto = fib->proto; struct mlxsw_sp_lpm_tree *old_tree; u8 old_id, new_id = new_tree->id; @@ -968,7 +999,7 @@ static int mlxsw_sp_vrs_lpm_tree_replace(struct mlxsw_sp *mlxsw_sp, old_tree = mlxsw_sp->router->lpm.proto_trees[proto]; old_id = old_tree->id; - for (i = 0; i < MLXSW_CORE_RES_GET(mlxsw_sp->core, MAX_VRS); i++) { + for (i = 0; i < max_vrs; i++) { vr = &mlxsw_sp->router->vrs[i]; if (!mlxsw_sp_vr_lpm_tree_should_replace(vr, proto, old_id)) continue; @@ -1052,6 +1083,61 @@ u32 mlxsw_sp_ipip_dev_ul_tb_id(const struct net_device *ol_dev) return tb_id; } +static void +mlxsw_sp_crif_init(struct mlxsw_sp_crif *crif, struct net_device *dev) +{ + crif->key.dev = dev; + INIT_LIST_HEAD(&crif->nexthop_list); +} + +static struct mlxsw_sp_crif * +mlxsw_sp_crif_alloc(struct net_device *dev) +{ + struct mlxsw_sp_crif *crif; + + crif = kzalloc(sizeof(*crif), GFP_KERNEL); + if (!crif) + return NULL; + + mlxsw_sp_crif_init(crif, dev); + return crif; +} + +static void mlxsw_sp_crif_free(struct mlxsw_sp_crif *crif) +{ + if (WARN_ON(crif->rif)) + return; + + WARN_ON(!list_empty(&crif->nexthop_list)); + kfree(crif); +} + +static int mlxsw_sp_crif_insert(struct mlxsw_sp_router *router, + struct mlxsw_sp_crif *crif) +{ + return rhashtable_insert_fast(&router->crif_ht, &crif->ht_node, + mlxsw_sp_crif_ht_params); +} + +static void mlxsw_sp_crif_remove(struct mlxsw_sp_router *router, + struct mlxsw_sp_crif *crif) +{ + rhashtable_remove_fast(&router->crif_ht, &crif->ht_node, + mlxsw_sp_crif_ht_params); +} + +static struct mlxsw_sp_crif * +mlxsw_sp_crif_lookup(struct mlxsw_sp_router *router, + const struct net_device *dev) +{ + struct mlxsw_sp_crif_key key = { + .dev = (struct net_device *)dev, + }; + + return rhashtable_lookup_fast(&router->crif_ht, &key, + mlxsw_sp_crif_ht_params); +} + static struct mlxsw_sp_rif * mlxsw_sp_rif_create(struct mlxsw_sp *mlxsw_sp, const struct mlxsw_sp_rif_params *params, @@ -1557,6 +1643,7 @@ mlxsw_sp_rif_ipip_lb_op(struct mlxsw_sp_rif_ipip_lb *lb_rif, u16 ul_vr_id, u16 ul_rif_id, bool enable) { struct mlxsw_sp_rif_ipip_lb_config lb_cf = lb_rif->lb_config; + struct net_device *dev = mlxsw_sp_rif_dev(&lb_rif->common); enum mlxsw_reg_ritr_loopback_ipip_options ipip_options; struct mlxsw_sp_rif *rif = &lb_rif->common; struct mlxsw_sp *mlxsw_sp = rif->mlxsw_sp; @@ -1569,7 +1656,7 @@ mlxsw_sp_rif_ipip_lb_op(struct mlxsw_sp_rif_ipip_lb *lb_rif, u16 ul_vr_id, case MLXSW_SP_L3_PROTO_IPV4: saddr4 = be32_to_cpu(lb_cf.saddr.addr4); mlxsw_reg_ritr_pack(ritr_pl, enable, MLXSW_REG_RITR_LOOPBACK_IF, - rif->rif_index, rif->vr_id, rif->dev->mtu); + rif->rif_index, rif->vr_id, dev->mtu); mlxsw_reg_ritr_loopback_ipip4_pack(ritr_pl, lb_cf.lb_ipipt, ipip_options, ul_vr_id, ul_rif_id, saddr4, @@ -1579,7 +1666,7 @@ mlxsw_sp_rif_ipip_lb_op(struct mlxsw_sp_rif_ipip_lb *lb_rif, u16 ul_vr_id, case MLXSW_SP_L3_PROTO_IPV6: saddr6 = &lb_cf.saddr.addr6; mlxsw_reg_ritr_pack(ritr_pl, enable, MLXSW_REG_RITR_LOOPBACK_IF, - rif->rif_index, rif->vr_id, rif->dev->mtu); + rif->rif_index, rif->vr_id, dev->mtu); mlxsw_reg_ritr_loopback_ipip6_pack(ritr_pl, lb_cf.lb_ipipt, ipip_options, ul_vr_id, ul_rif_id, saddr6, @@ -1639,9 +1726,29 @@ static void mlxsw_sp_netdevice_ipip_ol_down_event(struct mlxsw_sp *mlxsw_sp, mlxsw_sp_ipip_entry_ol_down_event(mlxsw_sp, ipip_entry); } -static void mlxsw_sp_nexthop_rif_migrate(struct mlxsw_sp *mlxsw_sp, +static void mlxsw_sp_nexthop_rif_update(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_rif *rif); + +static void mlxsw_sp_rif_migrate_destroy(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_rif *old_rif, - struct mlxsw_sp_rif *new_rif); + struct mlxsw_sp_rif *new_rif, + bool migrate_nhs) +{ + struct mlxsw_sp_crif *crif = old_rif->crif; + struct mlxsw_sp_crif mock_crif = {}; + + if (migrate_nhs) + mlxsw_sp_nexthop_rif_update(mlxsw_sp, new_rif); + + /* Plant a mock CRIF so that destroying the old RIF doesn't unoffload + * our nexthops and IPIP tunnels, and doesn't sever the crif->rif link. + */ + mlxsw_sp_crif_init(&mock_crif, crif->key.dev); + old_rif->crif = &mock_crif; + mock_crif.rif = old_rif; + mlxsw_sp_rif_destroy(old_rif); +} + static int mlxsw_sp_ipip_entry_ol_lb_update(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_ipip_entry *ipip_entry, @@ -1659,18 +1766,11 @@ mlxsw_sp_ipip_entry_ol_lb_update(struct mlxsw_sp *mlxsw_sp, return PTR_ERR(new_lb_rif); ipip_entry->ol_lb = new_lb_rif; - if (keep_encap) - mlxsw_sp_nexthop_rif_migrate(mlxsw_sp, &old_lb_rif->common, - &new_lb_rif->common); - - mlxsw_sp_rif_destroy(&old_lb_rif->common); - + mlxsw_sp_rif_migrate_destroy(mlxsw_sp, &old_lb_rif->common, + &new_lb_rif->common, keep_encap); return 0; } -static void mlxsw_sp_nexthop_rif_update(struct mlxsw_sp *mlxsw_sp, - struct mlxsw_sp_rif *rif); - /** * __mlxsw_sp_ipip_entry_update_tunnel - Update offload related to IPIP entry. * @mlxsw_sp: mlxsw_sp. @@ -2152,7 +2252,7 @@ int mlxsw_sp_neigh_counter_get(struct mlxsw_sp *mlxsw_sp, return -EINVAL; return mlxsw_sp_flow_counter_get(mlxsw_sp, neigh_entry->counter_index, - p_counter, NULL); + false, p_counter, NULL); } static struct mlxsw_sp_neigh_entry * @@ -2329,7 +2429,7 @@ static void mlxsw_sp_router_neigh_ent_ipv4_process(struct mlxsw_sp *mlxsw_sp, } dipn = htonl(dip); - dev = mlxsw_sp->router->rifs[rif]->dev; + dev = mlxsw_sp_rif_dev(mlxsw_sp->router->rifs[rif]); n = neigh_lookup(&arp_tbl, &dipn, dev); if (!n) return; @@ -2357,7 +2457,7 @@ static void mlxsw_sp_router_neigh_ent_ipv6_process(struct mlxsw_sp *mlxsw_sp, return; } - dev = mlxsw_sp->router->rifs[rif]->dev; + dev = mlxsw_sp_rif_dev(mlxsw_sp->router->rifs[rif]); n = neigh_lookup(&nd_tbl, &dip, dev); if (!n) return; @@ -2745,13 +2845,12 @@ static void mlxsw_sp_router_update_priority_work(struct work_struct *work) } static int mlxsw_sp_router_schedule_work(struct net *net, - struct notifier_block *nb, + struct mlxsw_sp_router *router, + struct neighbour *n, void (*cb)(struct work_struct *)) { struct mlxsw_sp_netevent_work *net_work; - struct mlxsw_sp_router *router; - router = container_of(nb, struct mlxsw_sp_router, netevent_nb); if (!net_eq(net, mlxsw_sp_net(router->mlxsw_sp))) return NOTIFY_DONE; @@ -2761,20 +2860,46 @@ static int mlxsw_sp_router_schedule_work(struct net *net, INIT_WORK(&net_work->work, cb); net_work->mlxsw_sp = router->mlxsw_sp; + net_work->n = n; mlxsw_core_schedule_work(&net_work->work); return NOTIFY_DONE; } +static bool mlxsw_sp_dev_lower_is_port(struct net_device *dev) +{ + struct mlxsw_sp_port *mlxsw_sp_port; + + rcu_read_lock(); + mlxsw_sp_port = mlxsw_sp_port_dev_lower_find_rcu(dev); + rcu_read_unlock(); + return !!mlxsw_sp_port; +} + +static int mlxsw_sp_router_schedule_neigh_work(struct mlxsw_sp_router *router, + struct neighbour *n) +{ + struct net *net; + + net = neigh_parms_net(n->parms); + + /* Take a reference to ensure the neighbour won't be destructed until we + * drop the reference in delayed work. + */ + neigh_clone(n); + return mlxsw_sp_router_schedule_work(net, router, n, + mlxsw_sp_router_neigh_event_work); +} + static int mlxsw_sp_router_netevent_event(struct notifier_block *nb, unsigned long event, void *ptr) { - struct mlxsw_sp_netevent_work *net_work; - struct mlxsw_sp_port *mlxsw_sp_port; - struct mlxsw_sp *mlxsw_sp; + struct mlxsw_sp_router *router; unsigned long interval; struct neigh_parms *p; struct neighbour *n; + router = container_of(nb, struct mlxsw_sp_router, netevent_nb); + switch (event) { case NETEVENT_DELAY_PROBE_TIME_UPDATE: p = ptr; @@ -2787,15 +2912,11 @@ static int mlxsw_sp_router_netevent_event(struct notifier_block *nb, /* We are in atomic context and can't take RTNL mutex, * so use RCU variant to walk the device chain. */ - mlxsw_sp_port = mlxsw_sp_port_lower_dev_hold(p->dev); - if (!mlxsw_sp_port) + if (!mlxsw_sp_dev_lower_is_port(p->dev)) return NOTIFY_DONE; - mlxsw_sp = mlxsw_sp_port->mlxsw_sp; interval = jiffies_to_msecs(NEIGH_VAR(p, DELAY_PROBE_TIME)); - mlxsw_sp->router->neighs_update.interval = interval; - - mlxsw_sp_port_dev_put(mlxsw_sp_port); + router->neighs_update.interval = interval; break; case NETEVENT_NEIGH_UPDATE: n = ptr; @@ -2803,35 +2924,18 @@ static int mlxsw_sp_router_netevent_event(struct notifier_block *nb, if (n->tbl->family != AF_INET && n->tbl->family != AF_INET6) return NOTIFY_DONE; - mlxsw_sp_port = mlxsw_sp_port_lower_dev_hold(n->dev); - if (!mlxsw_sp_port) + if (!mlxsw_sp_dev_lower_is_port(n->dev)) return NOTIFY_DONE; - net_work = kzalloc(sizeof(*net_work), GFP_ATOMIC); - if (!net_work) { - mlxsw_sp_port_dev_put(mlxsw_sp_port); - return NOTIFY_BAD; - } - - INIT_WORK(&net_work->work, mlxsw_sp_router_neigh_event_work); - net_work->mlxsw_sp = mlxsw_sp_port->mlxsw_sp; - net_work->n = n; + return mlxsw_sp_router_schedule_neigh_work(router, n); - /* Take a reference to ensure the neighbour won't be - * destructed until we drop the reference in delayed - * work. - */ - neigh_clone(n); - mlxsw_core_schedule_work(&net_work->work); - mlxsw_sp_port_dev_put(mlxsw_sp_port); - break; case NETEVENT_IPV4_MPATH_HASH_UPDATE: case NETEVENT_IPV6_MPATH_HASH_UPDATE: - return mlxsw_sp_router_schedule_work(ptr, nb, + return mlxsw_sp_router_schedule_work(ptr, router, NULL, mlxsw_sp_router_mp_hash_event_work); case NETEVENT_IPV4_FWD_UPDATE_PRIORITY_UPDATE: - return mlxsw_sp_router_schedule_work(ptr, nb, + return mlxsw_sp_router_schedule_work(ptr, router, NULL, mlxsw_sp_router_update_priority_work); } @@ -2882,6 +2986,55 @@ static void mlxsw_sp_neigh_rif_gone_sync(struct mlxsw_sp *mlxsw_sp, } } +struct mlxsw_sp_neigh_rif_made_sync { + struct mlxsw_sp *mlxsw_sp; + struct mlxsw_sp_rif *rif; + int err; +}; + +static void mlxsw_sp_neigh_rif_made_sync_each(struct neighbour *n, void *data) +{ + struct mlxsw_sp_neigh_rif_made_sync *rms = data; + int rc; + + if (rms->err) + return; + if (n->dev != mlxsw_sp_rif_dev(rms->rif)) + return; + rc = mlxsw_sp_router_schedule_neigh_work(rms->mlxsw_sp->router, n); + if (rc != NOTIFY_DONE) + rms->err = -ENOMEM; +} + +static int mlxsw_sp_neigh_rif_made_sync(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_rif *rif) +{ + struct mlxsw_sp_neigh_rif_made_sync rms = { + .mlxsw_sp = mlxsw_sp, + .rif = rif, + }; + + if (!mlxsw_sp_dev_lower_is_port(mlxsw_sp_rif_dev(rif))) + return 0; + + neigh_for_each(&arp_tbl, mlxsw_sp_neigh_rif_made_sync_each, &rms); + if (rms.err) + goto err_arp; + +#if IS_ENABLED(CONFIG_IPV6) + neigh_for_each(&nd_tbl, mlxsw_sp_neigh_rif_made_sync_each, &rms); +#endif + if (rms.err) + goto err_nd; + + return 0; + +err_nd: +err_arp: + mlxsw_sp_neigh_rif_gone_sync(mlxsw_sp, rif); + return rms.err; +} + enum mlxsw_sp_nexthop_type { MLXSW_SP_NEXTHOP_TYPE_ETH, MLXSW_SP_NEXTHOP_TYPE_IPIP, @@ -2900,9 +3053,11 @@ struct mlxsw_sp_nexthop_key { struct fib_nh *fib_nh; }; +struct mlxsw_sp_nexthop_counter; + struct mlxsw_sp_nexthop { struct list_head neigh_list_node; /* member of neigh entry list */ - struct list_head rif_list_node; + struct list_head crif_list_node; struct list_head router_list_node; struct mlxsw_sp_nexthop_group_info *nhgi; /* pointer back to the group * this nexthop belongs to @@ -2915,7 +3070,7 @@ struct mlxsw_sp_nexthop { int nh_weight; int norm_nh_weight; int num_adj_entries; - struct mlxsw_sp_rif *rif; + struct mlxsw_sp_crif *crif; u8 should_offload:1, /* set indicates this nexthop should be written * to the adjacency table. */ @@ -2931,10 +3086,18 @@ struct mlxsw_sp_nexthop { struct mlxsw_sp_neigh_entry *neigh_entry; struct mlxsw_sp_ipip_entry *ipip_entry; }; - unsigned int counter_index; - bool counter_valid; + struct mlxsw_sp_nexthop_counter *counter; + u32 id; /* NH ID for members of a NH object group. */ }; +static struct net_device * +mlxsw_sp_nexthop_dev(const struct mlxsw_sp_nexthop *nh) +{ + if (!nh->crif) + return NULL; + return nh->crif->key.dev; +} + enum mlxsw_sp_nexthop_group_type { MLXSW_SP_NEXTHOP_GROUP_TYPE_IPV4, MLXSW_SP_NEXTHOP_GROUP_TYPE_IPV6, @@ -2949,12 +3112,23 @@ struct mlxsw_sp_nexthop_group_info { int sum_norm_weight; u8 adj_index_valid:1, gateway:1, /* routes using the group use a gateway */ - is_resilient:1; + is_resilient:1, + hw_stats:1; struct list_head list; /* member in nh_res_grp_list */ - struct mlxsw_sp_nexthop nexthops[]; -#define nh_rif nexthops[0].rif + struct xarray nexthop_counters; + struct mlxsw_sp_nexthop nexthops[] __counted_by(count); }; +static struct mlxsw_sp_rif * +mlxsw_sp_nhgi_rif(const struct mlxsw_sp_nexthop_group_info *nhgi) +{ + struct mlxsw_sp_crif *crif = nhgi->nexthops[0].crif; + + if (!crif) + return NULL; + return crif->rif; +} + struct mlxsw_sp_nexthop_group_vr_key { u16 vr_id; enum mlxsw_sp_l3proto proto; @@ -2985,39 +3159,145 @@ struct mlxsw_sp_nexthop_group { bool can_destroy; }; -void mlxsw_sp_nexthop_counter_alloc(struct mlxsw_sp *mlxsw_sp, +struct mlxsw_sp_nexthop_counter { + unsigned int counter_index; + refcount_t ref_count; +}; + +static struct mlxsw_sp_nexthop_counter * +mlxsw_sp_nexthop_counter_alloc(struct mlxsw_sp *mlxsw_sp) +{ + struct mlxsw_sp_nexthop_counter *nhct; + int err; + + nhct = kzalloc(sizeof(*nhct), GFP_KERNEL); + if (!nhct) + return ERR_PTR(-ENOMEM); + + err = mlxsw_sp_flow_counter_alloc(mlxsw_sp, &nhct->counter_index); + if (err) + goto err_counter_alloc; + + refcount_set(&nhct->ref_count, 1); + return nhct; + +err_counter_alloc: + kfree(nhct); + return ERR_PTR(err); +} + +static void +mlxsw_sp_nexthop_counter_free(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_nexthop_counter *nhct) +{ + mlxsw_sp_flow_counter_free(mlxsw_sp, nhct->counter_index); + kfree(nhct); +} + +static struct mlxsw_sp_nexthop_counter * +mlxsw_sp_nexthop_sh_counter_get(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_nexthop *nh) +{ + struct mlxsw_sp_nexthop_group *nh_grp = nh->nhgi->nh_grp; + struct mlxsw_sp_nexthop_counter *nhct; + int err; + + nhct = xa_load(&nh_grp->nhgi->nexthop_counters, nh->id); + if (nhct) { + refcount_inc(&nhct->ref_count); + return nhct; + } + + nhct = mlxsw_sp_nexthop_counter_alloc(mlxsw_sp); + if (IS_ERR(nhct)) + return nhct; + + err = xa_err(xa_store(&nh_grp->nhgi->nexthop_counters, nh->id, nhct, + GFP_KERNEL)); + if (err) + goto err_store; + + return nhct; + +err_store: + mlxsw_sp_nexthop_counter_free(mlxsw_sp, nhct); + return ERR_PTR(err); +} + +static void mlxsw_sp_nexthop_sh_counter_put(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_nexthop *nh) +{ + struct mlxsw_sp_nexthop_group *nh_grp = nh->nhgi->nh_grp; + struct mlxsw_sp_nexthop_counter *nhct; + + nhct = xa_load(&nh_grp->nhgi->nexthop_counters, nh->id); + if (WARN_ON(!nhct)) + return; + + if (!refcount_dec_and_test(&nhct->ref_count)) + return; + + xa_erase(&nh_grp->nhgi->nexthop_counters, nh->id); + mlxsw_sp_nexthop_counter_free(mlxsw_sp, nhct); +} + +int mlxsw_sp_nexthop_counter_enable(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_nexthop *nh) { + const char *table_adj = MLXSW_SP_DPIPE_TABLE_NAME_ADJ; + struct mlxsw_sp_nexthop_counter *nhct; struct devlink *devlink; + bool dpipe_stats; + + if (nh->counter) + return 0; devlink = priv_to_devlink(mlxsw_sp->core); - if (!devlink_dpipe_table_counter_enabled(devlink, - MLXSW_SP_DPIPE_TABLE_NAME_ADJ)) - return; + dpipe_stats = devlink_dpipe_table_counter_enabled(devlink, table_adj); + if (!(nh->nhgi->hw_stats || dpipe_stats)) + return 0; - if (mlxsw_sp_flow_counter_alloc(mlxsw_sp, &nh->counter_index)) - return; + if (nh->id) + nhct = mlxsw_sp_nexthop_sh_counter_get(mlxsw_sp, nh); + else + nhct = mlxsw_sp_nexthop_counter_alloc(mlxsw_sp); + if (IS_ERR(nhct)) + return PTR_ERR(nhct); - nh->counter_valid = true; + nh->counter = nhct; + return 0; } -void mlxsw_sp_nexthop_counter_free(struct mlxsw_sp *mlxsw_sp, - struct mlxsw_sp_nexthop *nh) +void mlxsw_sp_nexthop_counter_disable(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_nexthop *nh) { - if (!nh->counter_valid) + if (!nh->counter) return; - mlxsw_sp_flow_counter_free(mlxsw_sp, nh->counter_index); - nh->counter_valid = false; + + if (nh->id) + mlxsw_sp_nexthop_sh_counter_put(mlxsw_sp, nh); + else + mlxsw_sp_nexthop_counter_free(mlxsw_sp, nh->counter); + nh->counter = NULL; +} + +static int mlxsw_sp_nexthop_counter_update(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_nexthop *nh) +{ + if (nh->nhgi->hw_stats) + return mlxsw_sp_nexthop_counter_enable(mlxsw_sp, nh); + mlxsw_sp_nexthop_counter_disable(mlxsw_sp, nh); + return 0; } int mlxsw_sp_nexthop_counter_get(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_nexthop *nh, u64 *p_counter) { - if (!nh->counter_valid) + if (!nh->counter) return -EINVAL; - return mlxsw_sp_flow_counter_get(mlxsw_sp, nh->counter_index, - p_counter, NULL); + return mlxsw_sp_flow_counter_get(mlxsw_sp, nh->counter->counter_index, + true, p_counter, NULL); } struct mlxsw_sp_nexthop *mlxsw_sp_nexthop_next(struct mlxsw_sp_router *router, @@ -3076,7 +3356,9 @@ int mlxsw_sp_nexthop_indexes(struct mlxsw_sp_nexthop *nh, u32 *p_adj_index, struct mlxsw_sp_rif *mlxsw_sp_nexthop_rif(struct mlxsw_sp_nexthop *nh) { - return nh->rif; + if (WARN_ON(!nh->crif)) + return NULL; + return nh->crif->rif; } bool mlxsw_sp_nexthop_group_has_ipip(struct mlxsw_sp_nexthop *nh) @@ -3461,11 +3743,12 @@ static int __mlxsw_sp_nexthop_eth_update(struct mlxsw_sp *mlxsw_sp, bool force, char *ratr_pl) { struct mlxsw_sp_neigh_entry *neigh_entry = nh->neigh_entry; + struct mlxsw_sp_rif *rif = mlxsw_sp_nexthop_rif(nh); enum mlxsw_reg_ratr_op op; u16 rif_index; - rif_index = nh->rif ? nh->rif->rif_index : - mlxsw_sp->router->lb_rif_index; + rif_index = rif ? rif->rif_index : + mlxsw_sp->router->lb_crif->rif->rif_index; op = force ? MLXSW_REG_RATR_OP_WRITE_WRITE_ENTRY : MLXSW_REG_RATR_OP_WRITE_WRITE_ENTRY_ON_ACTIVITY; mlxsw_reg_ratr_pack(ratr_pl, op, true, MLXSW_REG_RATR_TYPE_ETHERNET, @@ -3487,8 +3770,9 @@ static int __mlxsw_sp_nexthop_eth_update(struct mlxsw_sp *mlxsw_sp, WARN_ON_ONCE(1); return -EINVAL; } - if (nh->counter_valid) - mlxsw_reg_ratr_counter_pack(ratr_pl, nh->counter_index, true); + if (nh->counter) + mlxsw_reg_ratr_counter_pack(ratr_pl, nh->counter->counter_index, + true); else mlxsw_reg_ratr_counter_pack(ratr_pl, 0, false); @@ -3575,6 +3859,7 @@ mlxsw_sp_nexthop_group_update(struct mlxsw_sp *mlxsw_sp, nh = &nhgi->nexthops[i]; if (!nh->should_offload) { + mlxsw_sp_nexthop_counter_disable(mlxsw_sp, nh); nh->offloaded = 0; continue; } @@ -3582,6 +3867,10 @@ mlxsw_sp_nexthop_group_update(struct mlxsw_sp *mlxsw_sp, if (nh->update || reallocate) { int err = 0; + err = mlxsw_sp_nexthop_counter_update(mlxsw_sp, nh); + if (err) + return err; + err = mlxsw_sp_nexthop_update(mlxsw_sp, adj_index, nh, true, ratr_pl); if (err) @@ -4008,16 +4297,18 @@ mlxsw_sp_nexthop_dead_neigh_replace(struct mlxsw_sp *mlxsw_sp, { struct neighbour *n, *old_n = neigh_entry->key.n; struct mlxsw_sp_nexthop *nh; + struct net_device *dev; bool entry_connected; u8 nud_state, dead; int err; nh = list_first_entry(&neigh_entry->nexthop_list, struct mlxsw_sp_nexthop, neigh_list_node); + dev = mlxsw_sp_nexthop_dev(nh); - n = neigh_lookup(nh->neigh_tbl, &nh->gw_addr, nh->rif->dev); + n = neigh_lookup(nh->neigh_tbl, &nh->gw_addr, dev); if (!n) { - n = neigh_create(nh->neigh_tbl, &nh->gw_addr, nh->rif->dev); + n = neigh_create(nh->neigh_tbl, &nh->gw_addr, dev); if (IS_ERR(n)) return PTR_ERR(n); neigh_event_send(n, NULL); @@ -4081,44 +4372,49 @@ mlxsw_sp_nexthop_neigh_update(struct mlxsw_sp *mlxsw_sp, } } -static void mlxsw_sp_nexthop_rif_init(struct mlxsw_sp_nexthop *nh, - struct mlxsw_sp_rif *rif) +static void mlxsw_sp_nexthop_crif_init(struct mlxsw_sp_nexthop *nh, + struct mlxsw_sp_crif *crif) { - if (nh->rif) + if (nh->crif) return; - nh->rif = rif; - list_add(&nh->rif_list_node, &rif->nexthop_list); + nh->crif = crif; + list_add(&nh->crif_list_node, &crif->nexthop_list); } -static void mlxsw_sp_nexthop_rif_fini(struct mlxsw_sp_nexthop *nh) +static void mlxsw_sp_nexthop_crif_fini(struct mlxsw_sp_nexthop *nh) { - if (!nh->rif) + if (!nh->crif) return; - list_del(&nh->rif_list_node); - nh->rif = NULL; + list_del(&nh->crif_list_node); + nh->crif = NULL; } static int mlxsw_sp_nexthop_neigh_init(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_nexthop *nh) { struct mlxsw_sp_neigh_entry *neigh_entry; + struct net_device *dev; struct neighbour *n; u8 nud_state, dead; int err; + if (WARN_ON(!nh->crif->rif)) + return 0; + if (!nh->nhgi->gateway || nh->neigh_entry) return 0; + dev = mlxsw_sp_nexthop_dev(nh); /* Take a reference of neigh here ensuring that neigh would * not be destructed before the nexthop entry is finished. * The reference is taken either in neigh_lookup() or * in neigh_create() in case n is not found. */ - n = neigh_lookup(nh->neigh_tbl, &nh->gw_addr, nh->rif->dev); + n = neigh_lookup(nh->neigh_tbl, &nh->gw_addr, dev); if (!n) { - n = neigh_create(nh->neigh_tbl, &nh->gw_addr, nh->rif->dev); + n = neigh_create(nh->neigh_tbl, &nh->gw_addr, dev); if (IS_ERR(n)) return PTR_ERR(n); neigh_event_send(n, NULL); @@ -4197,15 +4493,20 @@ static void mlxsw_sp_nexthop_ipip_init(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_nexthop *nh, struct mlxsw_sp_ipip_entry *ipip_entry) { + struct mlxsw_sp_crif *crif; bool removing; if (!nh->nhgi->gateway || nh->ipip_entry) return; + crif = mlxsw_sp_crif_lookup(mlxsw_sp->router, ipip_entry->ol_dev); + if (WARN_ON(!crif)) + return; + nh->ipip_entry = ipip_entry; removing = !mlxsw_sp_ipip_netdev_ul_up(ipip_entry->ol_dev); __mlxsw_sp_nexthop_neigh_update(nh, removing); - mlxsw_sp_nexthop_rif_init(nh, &ipip_entry->ol_lb->common); + mlxsw_sp_nexthop_crif_init(nh, crif); } static void mlxsw_sp_nexthop_ipip_fini(struct mlxsw_sp *mlxsw_sp, @@ -4237,7 +4538,7 @@ static int mlxsw_sp_nexthop_type_init(struct mlxsw_sp *mlxsw_sp, { const struct mlxsw_sp_ipip_ops *ipip_ops; struct mlxsw_sp_ipip_entry *ipip_entry; - struct mlxsw_sp_rif *rif; + struct mlxsw_sp_crif *crif; int err; ipip_entry = mlxsw_sp_ipip_entry_find_by_ol_dev(mlxsw_sp, dev); @@ -4251,11 +4552,15 @@ static int mlxsw_sp_nexthop_type_init(struct mlxsw_sp *mlxsw_sp, } nh->type = MLXSW_SP_NEXTHOP_TYPE_ETH; - rif = mlxsw_sp_rif_find_by_dev(mlxsw_sp, dev); - if (!rif) + crif = mlxsw_sp_crif_lookup(mlxsw_sp->router, dev); + if (!crif) + return 0; + + mlxsw_sp_nexthop_crif_init(nh, crif); + + if (!crif->rif) return 0; - mlxsw_sp_nexthop_rif_init(nh, rif); err = mlxsw_sp_nexthop_neigh_init(mlxsw_sp, nh); if (err) goto err_neigh_init; @@ -4263,25 +4568,43 @@ static int mlxsw_sp_nexthop_type_init(struct mlxsw_sp *mlxsw_sp, return 0; err_neigh_init: - mlxsw_sp_nexthop_rif_fini(nh); + mlxsw_sp_nexthop_crif_fini(nh); return err; } -static void mlxsw_sp_nexthop_type_fini(struct mlxsw_sp *mlxsw_sp, - struct mlxsw_sp_nexthop *nh) +static int mlxsw_sp_nexthop_type_rif_made(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_nexthop *nh) +{ + switch (nh->type) { + case MLXSW_SP_NEXTHOP_TYPE_ETH: + return mlxsw_sp_nexthop_neigh_init(mlxsw_sp, nh); + case MLXSW_SP_NEXTHOP_TYPE_IPIP: + break; + } + + return 0; +} + +static void mlxsw_sp_nexthop_type_rif_gone(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_nexthop *nh) { switch (nh->type) { case MLXSW_SP_NEXTHOP_TYPE_ETH: mlxsw_sp_nexthop_neigh_fini(mlxsw_sp, nh); - mlxsw_sp_nexthop_rif_fini(nh); break; case MLXSW_SP_NEXTHOP_TYPE_IPIP: - mlxsw_sp_nexthop_rif_fini(nh); mlxsw_sp_nexthop_ipip_fini(mlxsw_sp, nh); break; } } +static void mlxsw_sp_nexthop_type_fini(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_nexthop *nh) +{ + mlxsw_sp_nexthop_type_rif_gone(mlxsw_sp, nh); + mlxsw_sp_nexthop_crif_fini(nh); +} + static int mlxsw_sp_nexthop4_init(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_nexthop_group *nh_grp, struct mlxsw_sp_nexthop *nh, @@ -4304,7 +4627,10 @@ static int mlxsw_sp_nexthop4_init(struct mlxsw_sp *mlxsw_sp, if (err) return err; - mlxsw_sp_nexthop_counter_alloc(mlxsw_sp, nh); + err = mlxsw_sp_nexthop_counter_enable(mlxsw_sp, nh); + if (err) + goto err_counter_enable; + list_add_tail(&nh->router_list_node, &mlxsw_sp->router->nexthop_list); if (!dev) @@ -4328,7 +4654,8 @@ static int mlxsw_sp_nexthop4_init(struct mlxsw_sp *mlxsw_sp, err_nexthop_neigh_init: list_del(&nh->router_list_node); - mlxsw_sp_nexthop_counter_free(mlxsw_sp, nh); + mlxsw_sp_nexthop_counter_disable(mlxsw_sp, nh); +err_counter_enable: mlxsw_sp_nexthop_remove(mlxsw_sp, nh); return err; } @@ -4338,7 +4665,7 @@ static void mlxsw_sp_nexthop4_fini(struct mlxsw_sp *mlxsw_sp, { mlxsw_sp_nexthop_type_fini(mlxsw_sp, nh); list_del(&nh->router_list_node); - mlxsw_sp_nexthop_counter_free(mlxsw_sp, nh); + mlxsw_sp_nexthop_counter_disable(mlxsw_sp, nh); mlxsw_sp_nexthop_remove(mlxsw_sp, nh); } @@ -4368,16 +4695,17 @@ static void mlxsw_sp_nexthop4_event(struct mlxsw_sp *mlxsw_sp, static void mlxsw_sp_nexthop_rif_update(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_rif *rif) { + struct net_device *dev = mlxsw_sp_rif_dev(rif); struct mlxsw_sp_nexthop *nh; bool removing; - list_for_each_entry(nh, &rif->nexthop_list, rif_list_node) { + list_for_each_entry(nh, &rif->crif->nexthop_list, crif_list_node) { switch (nh->type) { case MLXSW_SP_NEXTHOP_TYPE_ETH: removing = false; break; case MLXSW_SP_NEXTHOP_TYPE_IPIP: - removing = !mlxsw_sp_ipip_netdev_ul_up(rif->dev); + removing = !mlxsw_sp_ipip_netdev_ul_up(dev); break; default: WARN_ON(1); @@ -4389,16 +4717,33 @@ static void mlxsw_sp_nexthop_rif_update(struct mlxsw_sp *mlxsw_sp, } } -static void mlxsw_sp_nexthop_rif_migrate(struct mlxsw_sp *mlxsw_sp, - struct mlxsw_sp_rif *old_rif, - struct mlxsw_sp_rif *new_rif) +static int mlxsw_sp_nexthop_rif_made_sync(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_rif *rif) { - struct mlxsw_sp_nexthop *nh; + struct mlxsw_sp_nexthop *nh, *tmp; + unsigned int n = 0; + int err; - list_splice_init(&old_rif->nexthop_list, &new_rif->nexthop_list); - list_for_each_entry(nh, &new_rif->nexthop_list, rif_list_node) - nh->rif = new_rif; - mlxsw_sp_nexthop_rif_update(mlxsw_sp, new_rif); + list_for_each_entry_safe(nh, tmp, &rif->crif->nexthop_list, + crif_list_node) { + err = mlxsw_sp_nexthop_type_rif_made(mlxsw_sp, nh); + if (err) + goto err_nexthop_type_rif; + mlxsw_sp_nexthop_group_refresh(mlxsw_sp, nh->nhgi->nh_grp); + n++; + } + + return 0; + +err_nexthop_type_rif: + list_for_each_entry_safe(nh, tmp, &rif->crif->nexthop_list, + crif_list_node) { + if (!n--) + break; + mlxsw_sp_nexthop_type_rif_gone(mlxsw_sp, nh); + mlxsw_sp_nexthop_group_refresh(mlxsw_sp, nh->nhgi->nh_grp); + } + return err; } static void mlxsw_sp_nexthop_rif_gone_sync(struct mlxsw_sp *mlxsw_sp, @@ -4406,8 +4751,9 @@ static void mlxsw_sp_nexthop_rif_gone_sync(struct mlxsw_sp *mlxsw_sp, { struct mlxsw_sp_nexthop *nh, *tmp; - list_for_each_entry_safe(nh, tmp, &rif->nexthop_list, rif_list_node) { - mlxsw_sp_nexthop_type_fini(mlxsw_sp, nh); + list_for_each_entry_safe(nh, tmp, &rif->crif->nexthop_list, + crif_list_node) { + mlxsw_sp_nexthop_type_rif_gone(mlxsw_sp, nh); mlxsw_sp_nexthop_group_refresh(mlxsw_sp, nh->nhgi->nh_grp); } } @@ -4427,7 +4773,7 @@ static int mlxsw_sp_adj_trap_entry_init(struct mlxsw_sp *mlxsw_sp) mlxsw_reg_ratr_pack(ratr_pl, MLXSW_REG_RATR_OP_WRITE_WRITE_ENTRY, true, MLXSW_REG_RATR_TYPE_ETHERNET, mlxsw_sp->router->adj_trap_index, - mlxsw_sp->router->lb_rif_index); + mlxsw_sp->router->lb_crif->rif->rif_index); mlxsw_reg_ratr_trap_action_set(ratr_pl, trap_action); mlxsw_reg_ratr_trap_id_set(ratr_pl, MLXSW_TRAP_ID_RTR_EGRESS0); err = mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(ratr), ratr_pl); @@ -4743,21 +5089,19 @@ static bool mlxsw_sp_nexthop_obj_is_gateway(struct mlxsw_sp *mlxsw_sp, static void mlxsw_sp_nexthop_obj_blackhole_init(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_nexthop *nh) { - u16 lb_rif_index = mlxsw_sp->router->lb_rif_index; - nh->action = MLXSW_SP_NEXTHOP_ACTION_DISCARD; nh->should_offload = 1; /* While nexthops that discard packets do not forward packets * via an egress RIF, they still need to be programmed using a * valid RIF, so use the loopback RIF created during init. */ - nh->rif = mlxsw_sp->router->rifs[lb_rif_index]; + nh->crif = mlxsw_sp->router->lb_crif; } static void mlxsw_sp_nexthop_obj_blackhole_fini(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_nexthop *nh) { - nh->rif = NULL; + nh->crif = NULL; nh->should_offload = 0; } @@ -4786,9 +5130,9 @@ mlxsw_sp_nexthop_obj_init(struct mlxsw_sp *mlxsw_sp, break; } - mlxsw_sp_nexthop_counter_alloc(mlxsw_sp, nh); list_add_tail(&nh->router_list_node, &mlxsw_sp->router->nexthop_list); nh->ifindex = dev->ifindex; + nh->id = nh_obj->id; err = mlxsw_sp_nexthop_type_init(mlxsw_sp, nh, dev); if (err) @@ -4810,7 +5154,6 @@ mlxsw_sp_nexthop_obj_init(struct mlxsw_sp *mlxsw_sp, err_type_init: list_del(&nh->router_list_node); - mlxsw_sp_nexthop_counter_free(mlxsw_sp, nh); return err; } @@ -4821,7 +5164,7 @@ static void mlxsw_sp_nexthop_obj_fini(struct mlxsw_sp *mlxsw_sp, mlxsw_sp_nexthop_obj_blackhole_fini(mlxsw_sp, nh); mlxsw_sp_nexthop_type_fini(mlxsw_sp, nh); list_del(&nh->router_list_node); - mlxsw_sp_nexthop_counter_free(mlxsw_sp, nh); + mlxsw_sp_nexthop_counter_disable(mlxsw_sp, nh); nh->should_offload = 0; } @@ -4833,6 +5176,7 @@ mlxsw_sp_nexthop_obj_group_info_init(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_nexthop_group_info *nhgi; struct mlxsw_sp_nexthop *nh; bool is_resilient = false; + bool hw_stats = false; unsigned int nhs; int err, i; @@ -4842,9 +5186,11 @@ mlxsw_sp_nexthop_obj_group_info_init(struct mlxsw_sp *mlxsw_sp, break; case NH_NOTIFIER_INFO_TYPE_GRP: nhs = info->nh_grp->num_nh; + hw_stats = info->nh_grp->hw_stats; break; case NH_NOTIFIER_INFO_TYPE_RES_TABLE: nhs = info->nh_res_table->num_nh_buckets; + hw_stats = info->nh_res_table->hw_stats; is_resilient = true; break; default: @@ -4859,6 +5205,10 @@ mlxsw_sp_nexthop_obj_group_info_init(struct mlxsw_sp *mlxsw_sp, nhgi->gateway = mlxsw_sp_nexthop_obj_is_gateway(mlxsw_sp, info); nhgi->is_resilient = is_resilient; nhgi->count = nhs; + nhgi->hw_stats = hw_stats; + + xa_init_flags(&nhgi->nexthop_counters, XA_FLAGS_ALLOC1); + for (i = 0; i < nhgi->count; i++) { struct nh_notifier_single_info *nh_obj; int weight; @@ -4941,6 +5291,8 @@ mlxsw_sp_nexthop_obj_group_info_fini(struct mlxsw_sp *mlxsw_sp, } mlxsw_sp_nexthop_group_refresh(mlxsw_sp, nh_grp); WARN_ON_ONCE(nhgi->adj_index_valid); + WARN_ON(!xa_empty(&nhgi->nexthop_counters)); + xa_destroy(&nhgi->nexthop_counters); kfree(nhgi); } @@ -5080,6 +5432,43 @@ err_out: return err; } +static int mlxsw_sp_nexthop_obj_res_group_pre(struct mlxsw_sp *mlxsw_sp, + struct nh_notifier_info *info) +{ + struct nh_notifier_grp_info *grp_info = info->nh_grp; + struct mlxsw_sp_nexthop_group_info *nhgi; + struct mlxsw_sp_nexthop_group *nh_grp; + int err; + int i; + + nh_grp = mlxsw_sp_nexthop_obj_group_lookup(mlxsw_sp, info->id); + if (!nh_grp) + return 0; + nhgi = nh_grp->nhgi; + + if (nhgi->hw_stats == grp_info->hw_stats) + return 0; + + nhgi->hw_stats = grp_info->hw_stats; + + for (i = 0; i < nhgi->count; i++) { + struct mlxsw_sp_nexthop *nh = &nhgi->nexthops[i]; + + if (nh->offloaded) + nh->update = 1; + } + + err = mlxsw_sp_nexthop_group_refresh(mlxsw_sp, nh_grp); + if (err) + goto err_group_refresh; + + return 0; + +err_group_refresh: + nhgi->hw_stats = !grp_info->hw_stats; + return err; +} + static int mlxsw_sp_nexthop_obj_new(struct mlxsw_sp *mlxsw_sp, struct nh_notifier_info *info) { @@ -5256,6 +5645,79 @@ err_nexthop_obj_init: return err; } +static void +mlxsw_sp_nexthop_obj_mp_hw_stats_get(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_nexthop_group_info *nhgi, + struct nh_notifier_grp_hw_stats_info *info) +{ + int nhi; + + for (nhi = 0; nhi < info->num_nh; nhi++) { + struct mlxsw_sp_nexthop *nh = &nhgi->nexthops[nhi]; + u64 packets; + int err; + + err = mlxsw_sp_nexthop_counter_get(mlxsw_sp, nh, &packets); + if (err) + continue; + + nh_grp_hw_stats_report_delta(info, nhi, packets); + } +} + +static void +mlxsw_sp_nexthop_obj_res_hw_stats_get(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_nexthop_group_info *nhgi, + struct nh_notifier_grp_hw_stats_info *info) +{ + int nhi = -1; + int bucket; + + for (bucket = 0; bucket < nhgi->count; bucket++) { + struct mlxsw_sp_nexthop *nh = &nhgi->nexthops[bucket]; + u64 packets; + int err; + + if (nhi == -1 || info->stats[nhi].id != nh->id) { + for (nhi = 0; nhi < info->num_nh; nhi++) + if (info->stats[nhi].id == nh->id) + break; + if (WARN_ON_ONCE(nhi == info->num_nh)) { + nhi = -1; + continue; + } + } + + err = mlxsw_sp_nexthop_counter_get(mlxsw_sp, nh, &packets); + if (err) + continue; + + nh_grp_hw_stats_report_delta(info, nhi, packets); + } +} + +static void mlxsw_sp_nexthop_obj_hw_stats_get(struct mlxsw_sp *mlxsw_sp, + struct nh_notifier_info *info) +{ + struct mlxsw_sp_nexthop_group_info *nhgi; + struct mlxsw_sp_nexthop_group *nh_grp; + + if (info->type != NH_NOTIFIER_INFO_TYPE_GRP_HW_STATS) + return; + + nh_grp = mlxsw_sp_nexthop_obj_group_lookup(mlxsw_sp, info->id); + if (!nh_grp) + return; + nhgi = nh_grp->nhgi; + + if (nhgi->is_resilient) + mlxsw_sp_nexthop_obj_res_hw_stats_get(mlxsw_sp, nhgi, + info->nh_grp_hw_stats); + else + mlxsw_sp_nexthop_obj_mp_hw_stats_get(mlxsw_sp, nhgi, + info->nh_grp_hw_stats); +} + static int mlxsw_sp_nexthop_obj_event(struct notifier_block *nb, unsigned long event, void *ptr) { @@ -5271,6 +5733,10 @@ static int mlxsw_sp_nexthop_obj_event(struct notifier_block *nb, mutex_lock(&router->lock); switch (event) { + case NEXTHOP_EVENT_RES_TABLE_PRE_REPLACE: + err = mlxsw_sp_nexthop_obj_res_group_pre(router->mlxsw_sp, + info); + break; case NEXTHOP_EVENT_REPLACE: err = mlxsw_sp_nexthop_obj_new(router->mlxsw_sp, info); break; @@ -5281,6 +5747,9 @@ static int mlxsw_sp_nexthop_obj_event(struct notifier_block *nb, err = mlxsw_sp_nexthop_obj_bucket_replace(router->mlxsw_sp, info); break; + case NEXTHOP_EVENT_HW_STATS_REPORT_DELTA: + mlxsw_sp_nexthop_obj_hw_stats_get(router->mlxsw_sp, info); + break; default: break; } @@ -5491,7 +5960,7 @@ mlxsw_sp_fib_entry_should_offload(const struct mlxsw_sp_fib_entry *fib_entry) case MLXSW_SP_FIB_ENTRY_TYPE_REMOTE: return !!nh_group->nhgi->adj_index_valid; case MLXSW_SP_FIB_ENTRY_TYPE_LOCAL: - return !!nh_group->nhgi->nh_rif; + return !!mlxsw_sp_nhgi_rif(nh_group->nhgi); case MLXSW_SP_FIB_ENTRY_TYPE_BLACKHOLE: case MLXSW_SP_FIB_ENTRY_TYPE_IPIP_DECAP: case MLXSW_SP_FIB_ENTRY_TYPE_NVE_DECAP: @@ -5509,9 +5978,10 @@ mlxsw_sp_rt6_nexthop(struct mlxsw_sp_nexthop_group *nh_grp, for (i = 0; i < nh_grp->nhgi->count; i++) { struct mlxsw_sp_nexthop *nh = &nh_grp->nhgi->nexthops[i]; + struct net_device *dev = mlxsw_sp_nexthop_dev(nh); struct fib6_info *rt = mlxsw_sp_rt6->rt; - if (nh->rif && nh->rif->dev == rt->fib6_nh->fib_nh_dev && + if (dev && dev == rt->fib6_nh->fib_nh_dev && ipv6_addr_equal((const struct in6_addr *) &nh->gw_addr, &rt->fib6_nh->fib_nh_gw6)) return nh; @@ -5752,7 +6222,8 @@ static int mlxsw_sp_fib_entry_op_remote(struct mlxsw_sp *mlxsw_sp, trap_action = MLXSW_REG_RALUE_TRAP_ACTION_NOP; adjacency_index = nhgi->adj_index; ecmp_size = nhgi->ecmp_size; - } else if (!nhgi->adj_index_valid && nhgi->count && nhgi->nh_rif) { + } else if (!nhgi->adj_index_valid && nhgi->count && + mlxsw_sp_nhgi_rif(nhgi)) { trap_action = MLXSW_REG_RALUE_TRAP_ACTION_NOP; adjacency_index = mlxsw_sp->router->adj_trap_index; ecmp_size = 1; @@ -5771,7 +6242,7 @@ static int mlxsw_sp_fib_entry_op_local(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_fib_entry *fib_entry, enum mlxsw_reg_ralue_op op) { - struct mlxsw_sp_rif *rif = fib_entry->nh_group->nhgi->nh_rif; + struct mlxsw_sp_rif *rif = mlxsw_sp_nhgi_rif(fib_entry->nh_group->nhgi); enum mlxsw_reg_ralue_trap_action trap_action; char ralue_pl[MLXSW_REG_RALUE_LEN]; u16 trap_id = 0; @@ -6512,7 +6983,10 @@ static int mlxsw_sp_nexthop6_init(struct mlxsw_sp *mlxsw_sp, #if IS_ENABLED(CONFIG_IPV6) nh->neigh_tbl = &nd_tbl; #endif - mlxsw_sp_nexthop_counter_alloc(mlxsw_sp, nh); + + err = mlxsw_sp_nexthop_counter_enable(mlxsw_sp, nh); + if (err) + return err; list_add_tail(&nh->router_list_node, &mlxsw_sp->router->nexthop_list); @@ -6528,7 +7002,7 @@ static int mlxsw_sp_nexthop6_init(struct mlxsw_sp *mlxsw_sp, err_nexthop_type_init: list_del(&nh->router_list_node); - mlxsw_sp_nexthop_counter_free(mlxsw_sp, nh); + mlxsw_sp_nexthop_counter_disable(mlxsw_sp, nh); return err; } @@ -6537,7 +7011,7 @@ static void mlxsw_sp_nexthop6_fini(struct mlxsw_sp *mlxsw_sp, { mlxsw_sp_nexthop_type_fini(mlxsw_sp, nh); list_del(&nh->router_list_node); - mlxsw_sp_nexthop_counter_free(mlxsw_sp, nh); + mlxsw_sp_nexthop_counter_disable(mlxsw_sp, nh); } static bool mlxsw_sp_rt6_is_gateway(const struct mlxsw_sp *mlxsw_sp, @@ -7298,9 +7772,10 @@ static void mlxsw_sp_vr_fib_flush(struct mlxsw_sp *mlxsw_sp, static void mlxsw_sp_router_fib_flush(struct mlxsw_sp *mlxsw_sp) { + int max_vrs = MLXSW_CORE_RES_GET(mlxsw_sp->core, MAX_VRS); int i, j; - for (i = 0; i < MLXSW_CORE_RES_GET(mlxsw_sp->core, MAX_VRS); i++) { + for (i = 0; i < max_vrs; i++) { struct mlxsw_sp_vr *vr = &mlxsw_sp->router->vrs[i]; if (!mlxsw_sp_vr_is_used(vr)) @@ -7326,6 +7801,7 @@ struct mlxsw_sp_fib6_event_work { struct mlxsw_sp_fib_event_work { struct work_struct work; + netdevice_tracker dev_tracker; union { struct mlxsw_sp_fib6_event_work fib6_work; struct fib_entry_notifier_info fen_info; @@ -7499,12 +7975,12 @@ static void mlxsw_sp_router_fibmr_event_work(struct work_struct *work) &fib_work->ven_info); if (err) dev_warn(mlxsw_sp->bus_info->dev, "MR VIF add failed.\n"); - dev_put(fib_work->ven_info.dev); + netdev_put(fib_work->ven_info.dev, &fib_work->dev_tracker); break; case FIB_EVENT_VIF_DEL: mlxsw_sp_router_fibmr_vif_del(mlxsw_sp, &fib_work->ven_info); - dev_put(fib_work->ven_info.dev); + netdev_put(fib_work->ven_info.dev, &fib_work->dev_tracker); break; } mutex_unlock(&mlxsw_sp->router->lock); @@ -7575,7 +8051,8 @@ mlxsw_sp_router_fibmr_event(struct mlxsw_sp_fib_event_work *fib_work, case FIB_EVENT_VIF_ADD: case FIB_EVENT_VIF_DEL: memcpy(&fib_work->ven_info, info, sizeof(fib_work->ven_info)); - dev_hold(fib_work->ven_info.dev); + netdev_hold(fib_work->ven_info.dev, &fib_work->dev_tracker, + GFP_ATOMIC); break; } } @@ -7699,51 +8176,17 @@ static struct mlxsw_sp_rif * mlxsw_sp_rif_find_by_dev(const struct mlxsw_sp *mlxsw_sp, const struct net_device *dev) { + int max_rifs = MLXSW_CORE_RES_GET(mlxsw_sp->core, MAX_RIFS); int i; - for (i = 0; i < MLXSW_CORE_RES_GET(mlxsw_sp->core, MAX_RIFS); i++) + for (i = 0; i < max_rifs; i++) if (mlxsw_sp->router->rifs[i] && - mlxsw_sp->router->rifs[i]->dev == dev) + mlxsw_sp_rif_dev_is(mlxsw_sp->router->rifs[i], dev)) return mlxsw_sp->router->rifs[i]; return NULL; } -bool mlxsw_sp_rif_exists(struct mlxsw_sp *mlxsw_sp, - const struct net_device *dev) -{ - struct mlxsw_sp_rif *rif; - - mutex_lock(&mlxsw_sp->router->lock); - rif = mlxsw_sp_rif_find_by_dev(mlxsw_sp, dev); - mutex_unlock(&mlxsw_sp->router->lock); - - return rif; -} - -u16 mlxsw_sp_rif_vid(struct mlxsw_sp *mlxsw_sp, const struct net_device *dev) -{ - struct mlxsw_sp_rif *rif; - u16 vid = 0; - - mutex_lock(&mlxsw_sp->router->lock); - rif = mlxsw_sp_rif_find_by_dev(mlxsw_sp, dev); - if (!rif) - goto out; - - /* We only return the VID for VLAN RIFs. Otherwise we return an - * invalid value (0). - */ - if (rif->ops->type != MLXSW_SP_RIF_TYPE_VLAN) - goto out; - - vid = mlxsw_sp_fid_8021q_vid(rif->fid); - -out: - mutex_unlock(&mlxsw_sp->router->lock); - return vid; -} - static int mlxsw_sp_router_rif_disable(struct mlxsw_sp *mlxsw_sp, u16 rif) { char ritr_pl[MLXSW_REG_RITR_LEN]; @@ -7758,36 +8201,75 @@ static int mlxsw_sp_router_rif_disable(struct mlxsw_sp *mlxsw_sp, u16 rif) return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(ritr), ritr_pl); } +static int mlxsw_sp_router_rif_made_sync(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_rif *rif) +{ + int err; + + err = mlxsw_sp_neigh_rif_made_sync(mlxsw_sp, rif); + if (err) + return err; + + err = mlxsw_sp_nexthop_rif_made_sync(mlxsw_sp, rif); + if (err) + goto err_nexthop; + + return 0; + +err_nexthop: + mlxsw_sp_neigh_rif_gone_sync(mlxsw_sp, rif); + return err; +} + static void mlxsw_sp_router_rif_gone_sync(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_rif *rif) { + /* Signal to nexthop cleanup that the RIF is going away. */ + rif->crif->rif = NULL; + mlxsw_sp_router_rif_disable(mlxsw_sp, rif->rif_index); mlxsw_sp_nexthop_rif_gone_sync(mlxsw_sp, rif); mlxsw_sp_neigh_rif_gone_sync(mlxsw_sp, rif); } +static bool __mlxsw_sp_dev_addr_list_empty(const struct net_device *dev) +{ + struct inet6_dev *inet6_dev; + struct in_device *idev; + + idev = __in_dev_get_rcu(dev); + if (idev && idev->ifa_list) + return false; + + inet6_dev = __in6_dev_get(dev); + if (inet6_dev && !list_empty(&inet6_dev->addr_list)) + return false; + + return true; +} + +static bool mlxsw_sp_dev_addr_list_empty(const struct net_device *dev) +{ + bool addr_list_empty; + + rcu_read_lock(); + addr_list_empty = __mlxsw_sp_dev_addr_list_empty(dev); + rcu_read_unlock(); + + return addr_list_empty; +} + static bool mlxsw_sp_rif_should_config(struct mlxsw_sp_rif *rif, struct net_device *dev, unsigned long event) { - struct inet6_dev *inet6_dev; - bool addr_list_empty = true; - struct in_device *idev; + bool addr_list_empty; switch (event) { case NETDEV_UP: return rif == NULL; case NETDEV_DOWN: - rcu_read_lock(); - idev = __in_dev_get_rcu(dev); - if (idev && idev->ifa_list) - addr_list_empty = false; - - inet6_dev = __in6_dev_get(dev); - if (addr_list_empty && inet6_dev && - !list_empty(&inet6_dev->addr_list)) - addr_list_empty = false; - rcu_read_unlock(); + addr_list_empty = mlxsw_sp_dev_addr_list_empty(dev); /* macvlans do not have a RIF, but rather piggy back on the * RIF of their lower device. @@ -7796,7 +8278,7 @@ mlxsw_sp_rif_should_config(struct mlxsw_sp_rif *rif, struct net_device *dev, return true; if (rif && addr_list_empty && - !netif_is_l3_slave(rif->dev)) + !netif_is_l3_slave(mlxsw_sp_rif_dev(rif))) return true; /* It is possible we already removed the RIF ourselves * if it was assigned to a netdev that is now a bridge @@ -7854,27 +8336,39 @@ static void mlxsw_sp_rif_index_free(struct mlxsw_sp *mlxsw_sp, u16 rif_index, static struct mlxsw_sp_rif *mlxsw_sp_rif_alloc(size_t rif_size, u16 rif_index, u16 vr_id, - struct net_device *l3_dev) + struct mlxsw_sp_crif *crif) { + struct net_device *l3_dev = crif ? crif->key.dev : NULL; struct mlxsw_sp_rif *rif; rif = kzalloc(rif_size, GFP_KERNEL); if (!rif) return NULL; - INIT_LIST_HEAD(&rif->nexthop_list); INIT_LIST_HEAD(&rif->neigh_list); if (l3_dev) { ether_addr_copy(rif->addr, l3_dev->dev_addr); rif->mtu = l3_dev->mtu; - rif->dev = l3_dev; } rif->vr_id = vr_id; rif->rif_index = rif_index; + if (crif) { + rif->crif = crif; + crif->rif = rif; + } return rif; } +static void mlxsw_sp_rif_free(struct mlxsw_sp_rif *rif) +{ + WARN_ON(!list_empty(&rif->neigh_list)); + + if (rif->crif) + rif->crif->rif = NULL; + kfree(rif); +} + struct mlxsw_sp_rif *mlxsw_sp_rif_by_index(const struct mlxsw_sp *mlxsw_sp, u16 rif_index) { @@ -7891,18 +8385,6 @@ u16 mlxsw_sp_ipip_lb_rif_index(const struct mlxsw_sp_rif_ipip_lb *lb_rif) return lb_rif->common.rif_index; } -u16 mlxsw_sp_ipip_lb_ul_vr_id(const struct mlxsw_sp_rif_ipip_lb *lb_rif) -{ - u32 ul_tb_id = mlxsw_sp_ipip_dev_ul_tb_id(lb_rif->common.dev); - struct mlxsw_sp_vr *ul_vr; - - ul_vr = mlxsw_sp_vr_get(lb_rif->common.mlxsw_sp, ul_tb_id, NULL); - if (WARN_ON(IS_ERR(ul_vr))) - return 0; - - return ul_vr->id; -} - u16 mlxsw_sp_ipip_lb_ul_rif_id(const struct mlxsw_sp_rif_ipip_lb *lb_rif) { return lb_rif->ul_rif_id; @@ -8032,6 +8514,7 @@ mlxsw_sp_router_port_l3_stats_report_delta(struct mlxsw_sp_rif *rif, struct mlxsw_sp_router_hwstats_notify_work { struct work_struct work; struct net_device *dev; + netdevice_tracker dev_tracker; }; static void mlxsw_sp_router_hwstats_notify_work(struct work_struct *work) @@ -8043,7 +8526,7 @@ static void mlxsw_sp_router_hwstats_notify_work(struct work_struct *work) rtnl_lock(); rtnl_offload_xstats_notify(hws_work->dev); rtnl_unlock(); - dev_put(hws_work->dev); + netdev_put(hws_work->dev, &hws_work->dev_tracker); kfree(hws_work); } @@ -8063,19 +8546,25 @@ mlxsw_sp_router_hwstats_notify_schedule(struct net_device *dev) return; INIT_WORK(&hws_work->work, mlxsw_sp_router_hwstats_notify_work); - dev_hold(dev); + netdev_hold(dev, &hws_work->dev_tracker, GFP_KERNEL); hws_work->dev = dev; mlxsw_core_schedule_work(&hws_work->work); } int mlxsw_sp_rif_dev_ifindex(const struct mlxsw_sp_rif *rif) { - return rif->dev->ifindex; + return mlxsw_sp_rif_dev(rif)->ifindex; +} + +bool mlxsw_sp_rif_has_dev(const struct mlxsw_sp_rif *rif) +{ + return !!mlxsw_sp_rif_dev(rif); } -const struct net_device *mlxsw_sp_rif_dev(const struct mlxsw_sp_rif *rif) +bool mlxsw_sp_rif_dev_is(const struct mlxsw_sp_rif *rif, + const struct net_device *dev) { - return rif->dev; + return mlxsw_sp_rif_dev(rif) == dev; } static void mlxsw_sp_rif_push_l3_stats(struct mlxsw_sp_rif *rif) @@ -8083,7 +8572,7 @@ static void mlxsw_sp_rif_push_l3_stats(struct mlxsw_sp_rif *rif) struct rtnl_hw_stats64 stats = {}; if (!mlxsw_sp_router_port_l3_stats_fetch(rif, &stats)) - netdev_offload_xstats_push_delta(rif->dev, + netdev_offload_xstats_push_delta(mlxsw_sp_rif_dev(rif), NETDEV_OFFLOAD_XSTATS_TYPE_L3, &stats); } @@ -8098,6 +8587,7 @@ mlxsw_sp_rif_create(struct mlxsw_sp *mlxsw_sp, const struct mlxsw_sp_rif_ops *ops; struct mlxsw_sp_fid *fid = NULL; enum mlxsw_sp_rif_type type; + struct mlxsw_sp_crif *crif; struct mlxsw_sp_rif *rif; struct mlxsw_sp_vr *vr; u16 rif_index; @@ -8117,19 +8607,28 @@ mlxsw_sp_rif_create(struct mlxsw_sp *mlxsw_sp, goto err_rif_index_alloc; } - rif = mlxsw_sp_rif_alloc(ops->rif_size, rif_index, vr->id, params->dev); + crif = mlxsw_sp_crif_lookup(mlxsw_sp->router, params->dev); + if (WARN_ON(!crif)) { + err = -ENOENT; + goto err_crif_lookup; + } + + rif = mlxsw_sp_rif_alloc(ops->rif_size, rif_index, vr->id, crif); if (!rif) { err = -ENOMEM; goto err_rif_alloc; } - dev_hold(rif->dev); + netdev_hold(params->dev, &rif->dev_tracker, GFP_KERNEL); mlxsw_sp->router->rifs[rif_index] = rif; rif->mlxsw_sp = mlxsw_sp; rif->ops = ops; rif->rif_entries = rif_entries; + if (ops->setup) + ops->setup(rif, params); + if (ops->fid_get) { - fid = ops->fid_get(rif, extack); + fid = ops->fid_get(rif, params, extack); if (IS_ERR(fid)) { err = PTR_ERR(fid); goto err_fid_get; @@ -8137,9 +8636,6 @@ mlxsw_sp_rif_create(struct mlxsw_sp *mlxsw_sp, rif->fid = fid; } - if (ops->setup) - ops->setup(rif, params); - err = ops->configure(rif, extack); if (err) goto err_configure; @@ -8150,12 +8646,16 @@ mlxsw_sp_rif_create(struct mlxsw_sp *mlxsw_sp, goto err_mr_rif_add; } - if (netdev_offload_xstats_enabled(rif->dev, + err = mlxsw_sp_router_rif_made_sync(mlxsw_sp, rif); + if (err) + goto err_rif_made_sync; + + if (netdev_offload_xstats_enabled(params->dev, NETDEV_OFFLOAD_XSTATS_TYPE_L3)) { err = mlxsw_sp_router_port_l3_stats_enable(rif); if (err) goto err_stats_enable; - mlxsw_sp_router_hwstats_notify_schedule(rif->dev); + mlxsw_sp_router_hwstats_notify_schedule(params->dev); } else { mlxsw_sp_rif_counters_alloc(rif); } @@ -8164,6 +8664,8 @@ mlxsw_sp_rif_create(struct mlxsw_sp *mlxsw_sp, return rif; err_stats_enable: + mlxsw_sp_router_rif_gone_sync(mlxsw_sp, rif); +err_rif_made_sync: err_mr_rif_add: for (i--; i >= 0; i--) mlxsw_sp_mr_rif_del(vr->mr_table[i], rif); @@ -8173,9 +8675,10 @@ err_configure: mlxsw_sp_fid_put(fid); err_fid_get: mlxsw_sp->router->rifs[rif_index] = NULL; - dev_put(rif->dev); - kfree(rif); + netdev_put(params->dev, &rif->dev_tracker); + mlxsw_sp_rif_free(rif); err_rif_alloc: +err_crif_lookup: mlxsw_sp_rif_index_free(mlxsw_sp, rif_index, rif_entries); err_rif_index_alloc: vr->rif_count--; @@ -8185,8 +8688,10 @@ err_rif_index_alloc: static void mlxsw_sp_rif_destroy(struct mlxsw_sp_rif *rif) { + struct net_device *dev = mlxsw_sp_rif_dev(rif); const struct mlxsw_sp_rif_ops *ops = rif->ops; struct mlxsw_sp *mlxsw_sp = rif->mlxsw_sp; + struct mlxsw_sp_crif *crif = rif->crif; struct mlxsw_sp_fid *fid = rif->fid; u8 rif_entries = rif->rif_entries; u16 rif_index = rif->rif_index; @@ -8197,11 +8702,10 @@ static void mlxsw_sp_rif_destroy(struct mlxsw_sp_rif *rif) mlxsw_sp_router_rif_gone_sync(mlxsw_sp, rif); vr = &mlxsw_sp->router->vrs[rif->vr_id]; - if (netdev_offload_xstats_enabled(rif->dev, - NETDEV_OFFLOAD_XSTATS_TYPE_L3)) { + if (netdev_offload_xstats_enabled(dev, NETDEV_OFFLOAD_XSTATS_TYPE_L3)) { mlxsw_sp_rif_push_l3_stats(rif); mlxsw_sp_router_port_l3_stats_disable(rif); - mlxsw_sp_router_hwstats_notify_schedule(rif->dev); + mlxsw_sp_router_hwstats_notify_schedule(dev); } else { mlxsw_sp_rif_counters_free(rif); } @@ -8213,11 +8717,14 @@ static void mlxsw_sp_rif_destroy(struct mlxsw_sp_rif *rif) /* Loopback RIFs are not associated with a FID. */ mlxsw_sp_fid_put(fid); mlxsw_sp->router->rifs[rif->rif_index] = NULL; - dev_put(rif->dev); - kfree(rif); + netdev_put(dev, &rif->dev_tracker); + mlxsw_sp_rif_free(rif); mlxsw_sp_rif_index_free(mlxsw_sp, rif_index, rif_entries); vr->rif_count--; mlxsw_sp_vr_put(mlxsw_sp, vr); + + if (crif->can_destroy) + mlxsw_sp_crif_free(crif); } void mlxsw_sp_rif_destroy_by_dev(struct mlxsw_sp *mlxsw_sp, @@ -8234,6 +8741,110 @@ out: mutex_unlock(&mlxsw_sp->router->lock); } +static void mlxsw_sp_rif_destroy_vlan_upper(struct mlxsw_sp *mlxsw_sp, + struct net_device *br_dev, + u16 vid) +{ + struct net_device *upper_dev; + struct mlxsw_sp_crif *crif; + + rcu_read_lock(); + upper_dev = __vlan_find_dev_deep_rcu(br_dev, htons(ETH_P_8021Q), vid); + rcu_read_unlock(); + + if (!upper_dev) + return; + + crif = mlxsw_sp_crif_lookup(mlxsw_sp->router, upper_dev); + if (!crif || !crif->rif) + return; + + mlxsw_sp_rif_destroy(crif->rif); +} + +static int mlxsw_sp_inetaddr_bridge_event(struct mlxsw_sp *mlxsw_sp, + struct net_device *l3_dev, + int lower_pvid, + unsigned long event, + struct netlink_ext_ack *extack); + +int mlxsw_sp_router_bridge_vlan_add(struct mlxsw_sp *mlxsw_sp, + struct net_device *br_dev, + u16 new_vid, bool is_pvid, + struct netlink_ext_ack *extack) +{ + struct mlxsw_sp_rif *old_rif; + struct mlxsw_sp_rif *new_rif; + struct net_device *upper_dev; + u16 old_pvid = 0; + u16 new_pvid; + int err = 0; + + mutex_lock(&mlxsw_sp->router->lock); + old_rif = mlxsw_sp_rif_find_by_dev(mlxsw_sp, br_dev); + if (old_rif) { + /* If the RIF on the bridge is not a VLAN RIF, we shouldn't have + * gotten a PVID notification. + */ + if (WARN_ON(old_rif->ops->type != MLXSW_SP_RIF_TYPE_VLAN)) + old_rif = NULL; + else + old_pvid = mlxsw_sp_fid_8021q_vid(old_rif->fid); + } + + if (is_pvid) + new_pvid = new_vid; + else if (old_pvid == new_vid) + new_pvid = 0; + else + goto out; + + if (old_pvid == new_pvid) + goto out; + + if (new_pvid) { + struct mlxsw_sp_rif_params params = { + .dev = br_dev, + .vid = new_pvid, + }; + + /* If there is a VLAN upper with the same VID as the new PVID, + * kill its RIF, if there is one. + */ + mlxsw_sp_rif_destroy_vlan_upper(mlxsw_sp, br_dev, new_pvid); + + if (mlxsw_sp_dev_addr_list_empty(br_dev)) + goto out; + new_rif = mlxsw_sp_rif_create(mlxsw_sp, ¶ms, extack); + if (IS_ERR(new_rif)) { + err = PTR_ERR(new_rif); + goto out; + } + + if (old_pvid) + mlxsw_sp_rif_migrate_destroy(mlxsw_sp, old_rif, new_rif, + true); + } else { + mlxsw_sp_rif_destroy(old_rif); + } + + if (old_pvid) { + rcu_read_lock(); + upper_dev = __vlan_find_dev_deep_rcu(br_dev, htons(ETH_P_8021Q), + old_pvid); + rcu_read_unlock(); + if (upper_dev) + err = mlxsw_sp_inetaddr_bridge_event(mlxsw_sp, + upper_dev, + new_pvid, + NETDEV_UP, extack); + } + +out: + mutex_unlock(&mlxsw_sp->router->lock); + return err; +} + static void mlxsw_sp_rif_subport_params_init(struct mlxsw_sp_rif_params *params, struct mlxsw_sp_port_vlan *mlxsw_sp_port_vlan) @@ -8254,6 +8865,20 @@ mlxsw_sp_rif_subport_rif(const struct mlxsw_sp_rif *rif) return container_of(rif, struct mlxsw_sp_rif_subport, common); } +int mlxsw_sp_rif_subport_port(const struct mlxsw_sp_rif *rif, + u16 *port, bool *is_lag) +{ + struct mlxsw_sp_rif_subport *rif_subport; + + if (WARN_ON(rif->ops->type != MLXSW_SP_RIF_TYPE_SUBPORT)) + return -EINVAL; + + rif_subport = mlxsw_sp_rif_subport_rif(rif); + *is_lag = rif_subport->lag; + *port = *is_lag ? rif_subport->lag_id : rif_subport->system_port; + return 0; +} + static struct mlxsw_sp_rif * mlxsw_sp_rif_subport_get(struct mlxsw_sp *mlxsw_sp, const struct mlxsw_sp_rif_params *params, @@ -8488,21 +9113,24 @@ __mlxsw_sp_port_vlan_router_join(struct mlxsw_sp_port_vlan *mlxsw_sp_port_vlan, { struct mlxsw_sp_port *mlxsw_sp_port = mlxsw_sp_port_vlan->mlxsw_sp_port; struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; - struct mlxsw_sp_rif_params params = { - .dev = l3_dev, - }; + struct mlxsw_sp_rif_params params; u16 vid = mlxsw_sp_port_vlan->vid; struct mlxsw_sp_rif *rif; struct mlxsw_sp_fid *fid; int err; + params = (struct mlxsw_sp_rif_params) { + .dev = l3_dev, + .vid = vid, + }; + mlxsw_sp_rif_subport_params_init(¶ms, mlxsw_sp_port_vlan); rif = mlxsw_sp_rif_subport_get(mlxsw_sp, ¶ms, extack); if (IS_ERR(rif)) return PTR_ERR(rif); /* FID was already created, just take a reference */ - fid = rif->ops->fid_get(rif, extack); + fid = rif->ops->fid_get(rif, ¶ms, extack); err = mlxsw_sp_fid_port_vid_map(fid, mlxsw_sp_port, vid); if (err) goto err_fid_port_vid_map; @@ -8549,25 +9177,20 @@ __mlxsw_sp_port_vlan_router_leave(struct mlxsw_sp_port_vlan *mlxsw_sp_port_vlan) mlxsw_sp_rif_subport_put(rif); } -int -mlxsw_sp_port_vlan_router_join(struct mlxsw_sp_port_vlan *mlxsw_sp_port_vlan, - struct net_device *l3_dev, - struct netlink_ext_ack *extack) +static int +mlxsw_sp_port_vlan_router_join_existing(struct mlxsw_sp_port_vlan *mlxsw_sp_port_vlan, + struct net_device *l3_dev, + struct netlink_ext_ack *extack) { struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port_vlan->mlxsw_sp_port->mlxsw_sp; - struct mlxsw_sp_rif *rif; - int err = 0; - mutex_lock(&mlxsw_sp->router->lock); - rif = mlxsw_sp_rif_find_by_dev(mlxsw_sp, l3_dev); - if (!rif) - goto out; + lockdep_assert_held(&mlxsw_sp->router->lock); - err = __mlxsw_sp_port_vlan_router_join(mlxsw_sp_port_vlan, l3_dev, - extack); -out: - mutex_unlock(&mlxsw_sp->router->lock); - return err; + if (!mlxsw_sp_rif_find_by_dev(mlxsw_sp, l3_dev)) + return 0; + + return __mlxsw_sp_port_vlan_router_join(mlxsw_sp_port_vlan, l3_dev, + extack); } void @@ -8605,10 +9228,11 @@ static int mlxsw_sp_inetaddr_port_vlan_event(struct net_device *l3_dev, } static int mlxsw_sp_inetaddr_port_event(struct net_device *port_dev, - unsigned long event, + unsigned long event, bool nomaster, struct netlink_ext_ack *extack) { - if (netif_is_any_bridge_port(port_dev) || netif_is_lag_port(port_dev)) + if (!nomaster && (netif_is_any_bridge_port(port_dev) || + netif_is_lag_port(port_dev))) return 0; return mlxsw_sp_inetaddr_port_vlan_event(port_dev, port_dev, event, @@ -8639,10 +9263,10 @@ static int __mlxsw_sp_inetaddr_lag_event(struct net_device *l3_dev, } static int mlxsw_sp_inetaddr_lag_event(struct net_device *lag_dev, - unsigned long event, + unsigned long event, bool nomaster, struct netlink_ext_ack *extack) { - if (netif_is_bridge_port(lag_dev)) + if (!nomaster && netif_is_bridge_port(lag_dev)) return 0; return __mlxsw_sp_inetaddr_lag_event(lag_dev, lag_dev, event, @@ -8651,6 +9275,7 @@ static int mlxsw_sp_inetaddr_lag_event(struct net_device *lag_dev, static int mlxsw_sp_inetaddr_bridge_event(struct mlxsw_sp *mlxsw_sp, struct net_device *l3_dev, + int lower_pvid, unsigned long event, struct netlink_ext_ack *extack) { @@ -8658,6 +9283,7 @@ static int mlxsw_sp_inetaddr_bridge_event(struct mlxsw_sp *mlxsw_sp, .dev = l3_dev, }; struct mlxsw_sp_rif *rif; + int err; switch (event) { case NETDEV_UP: @@ -8669,7 +9295,21 @@ static int mlxsw_sp_inetaddr_bridge_event(struct mlxsw_sp *mlxsw_sp, NL_SET_ERR_MSG_MOD(extack, "Adding an IP address to 802.1ad bridge is not supported"); return -EOPNOTSUPP; } + err = br_vlan_get_pvid(l3_dev, ¶ms.vid); + if (err) + return err; + if (!params.vid) + return 0; + } else if (is_vlan_dev(l3_dev)) { + params.vid = vlan_dev_vlan_id(l3_dev); + + /* If the VID matches PVID of the bridge below, the + * bridge owns the RIF for this VLAN. Don't do anything. + */ + if ((int)params.vid == lower_pvid) + return 0; } + rif = mlxsw_sp_rif_create(mlxsw_sp, ¶ms, extack); if (IS_ERR(rif)) return PTR_ERR(rif); @@ -8685,24 +9325,32 @@ static int mlxsw_sp_inetaddr_bridge_event(struct mlxsw_sp *mlxsw_sp, static int mlxsw_sp_inetaddr_vlan_event(struct mlxsw_sp *mlxsw_sp, struct net_device *vlan_dev, - unsigned long event, + unsigned long event, bool nomaster, struct netlink_ext_ack *extack) { struct net_device *real_dev = vlan_dev_real_dev(vlan_dev); u16 vid = vlan_dev_vlan_id(vlan_dev); + u16 lower_pvid; + int err; - if (netif_is_bridge_port(vlan_dev)) + if (!nomaster && netif_is_bridge_port(vlan_dev)) return 0; - if (mlxsw_sp_port_dev_check(real_dev)) + if (mlxsw_sp_port_dev_check(real_dev)) { return mlxsw_sp_inetaddr_port_vlan_event(vlan_dev, real_dev, event, vid, extack); - else if (netif_is_lag_master(real_dev)) + } else if (netif_is_lag_master(real_dev)) { return __mlxsw_sp_inetaddr_lag_event(vlan_dev, real_dev, event, vid, extack); - else if (netif_is_bridge_master(real_dev) && br_vlan_enabled(real_dev)) - return mlxsw_sp_inetaddr_bridge_event(mlxsw_sp, vlan_dev, event, + } else if (netif_is_bridge_master(real_dev) && + br_vlan_enabled(real_dev)) { + err = br_vlan_get_pvid(real_dev, &lower_pvid); + if (err) + return err; + return mlxsw_sp_inetaddr_bridge_event(mlxsw_sp, vlan_dev, + lower_pvid, event, extack); + } return 0; } @@ -8756,10 +9404,8 @@ static int mlxsw_sp_rif_macvlan_add(struct mlxsw_sp *mlxsw_sp, int err; rif = mlxsw_sp_rif_find_by_dev(mlxsw_sp, vlan->lowerdev); - if (!rif) { - NL_SET_ERR_MSG_MOD(extack, "macvlan is only supported on top of router interfaces"); - return -EOPNOTSUPP; - } + if (!rif) + return 0; err = mlxsw_sp_rif_fdb_op(mlxsw_sp, macvlan_dev->dev_addr, mlxsw_sp_fid_index(rif->fid), true); @@ -8829,19 +9475,21 @@ static int mlxsw_sp_inetaddr_macvlan_event(struct mlxsw_sp *mlxsw_sp, static int __mlxsw_sp_inetaddr_event(struct mlxsw_sp *mlxsw_sp, struct net_device *dev, - unsigned long event, + unsigned long event, bool nomaster, struct netlink_ext_ack *extack) { if (mlxsw_sp_port_dev_check(dev)) - return mlxsw_sp_inetaddr_port_event(dev, event, extack); + return mlxsw_sp_inetaddr_port_event(dev, event, nomaster, + extack); else if (netif_is_lag_master(dev)) - return mlxsw_sp_inetaddr_lag_event(dev, event, extack); + return mlxsw_sp_inetaddr_lag_event(dev, event, nomaster, + extack); else if (netif_is_bridge_master(dev)) - return mlxsw_sp_inetaddr_bridge_event(mlxsw_sp, dev, event, + return mlxsw_sp_inetaddr_bridge_event(mlxsw_sp, dev, -1, event, extack); else if (is_vlan_dev(dev)) return mlxsw_sp_inetaddr_vlan_event(mlxsw_sp, dev, event, - extack); + nomaster, extack); else if (netif_is_macvlan(dev)) return mlxsw_sp_inetaddr_macvlan_event(mlxsw_sp, dev, event, extack); @@ -8868,14 +9516,15 @@ static int mlxsw_sp_inetaddr_event(struct notifier_block *nb, if (!mlxsw_sp_rif_should_config(rif, dev, event)) goto out; - err = __mlxsw_sp_inetaddr_event(router->mlxsw_sp, dev, event, NULL); + err = __mlxsw_sp_inetaddr_event(router->mlxsw_sp, dev, event, false, + NULL); out: mutex_unlock(&router->lock); return notifier_from_errno(err); } -int mlxsw_sp_inetaddr_valid_event(struct notifier_block *unused, - unsigned long event, void *ptr) +static int mlxsw_sp_inetaddr_valid_event(struct notifier_block *unused, + unsigned long event, void *ptr) { struct in_validator_info *ivi = (struct in_validator_info *) ptr; struct net_device *dev = ivi->ivi_dev->dev; @@ -8892,7 +9541,8 @@ int mlxsw_sp_inetaddr_valid_event(struct notifier_block *unused, if (!mlxsw_sp_rif_should_config(rif, dev, event)) goto out; - err = __mlxsw_sp_inetaddr_event(mlxsw_sp, dev, event, ivi->extack); + err = __mlxsw_sp_inetaddr_event(mlxsw_sp, dev, event, false, + ivi->extack); out: mutex_unlock(&mlxsw_sp->router->lock); return notifier_from_errno(err); @@ -8902,6 +9552,7 @@ struct mlxsw_sp_inet6addr_event_work { struct work_struct work; struct mlxsw_sp *mlxsw_sp; struct net_device *dev; + netdevice_tracker dev_tracker; unsigned long event; }; @@ -8921,11 +9572,11 @@ static void mlxsw_sp_inet6addr_event_work(struct work_struct *work) if (!mlxsw_sp_rif_should_config(rif, dev, event)) goto out; - __mlxsw_sp_inetaddr_event(mlxsw_sp, dev, event, NULL); + __mlxsw_sp_inetaddr_event(mlxsw_sp, dev, event, false, NULL); out: mutex_unlock(&mlxsw_sp->router->lock); rtnl_unlock(); - dev_put(dev); + netdev_put(dev, &inet6addr_work->dev_tracker); kfree(inet6addr_work); } @@ -8951,14 +9602,14 @@ static int mlxsw_sp_inet6addr_event(struct notifier_block *nb, inet6addr_work->mlxsw_sp = router->mlxsw_sp; inet6addr_work->dev = dev; inet6addr_work->event = event; - dev_hold(dev); + netdev_hold(dev, &inet6addr_work->dev_tracker, GFP_ATOMIC); mlxsw_core_schedule_work(&inet6addr_work->work); return NOTIFY_DONE; } -int mlxsw_sp_inet6addr_valid_event(struct notifier_block *unused, - unsigned long event, void *ptr) +static int mlxsw_sp_inet6addr_valid_event(struct notifier_block *unused, + unsigned long event, void *ptr) { struct in6_validator_info *i6vi = (struct in6_validator_info *) ptr; struct net_device *dev = i6vi->i6vi_dev->dev; @@ -8975,7 +9626,8 @@ int mlxsw_sp_inet6addr_valid_event(struct notifier_block *unused, if (!mlxsw_sp_rif_should_config(rif, dev, event)) goto out; - err = __mlxsw_sp_inetaddr_event(mlxsw_sp, dev, event, i6vi->extack); + err = __mlxsw_sp_inetaddr_event(mlxsw_sp, dev, event, false, + i6vi->extack); out: mutex_unlock(&mlxsw_sp->router->lock); return notifier_from_errno(err); @@ -9004,7 +9656,7 @@ mlxsw_sp_router_port_change_event(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_rif *rif, struct netlink_ext_ack *extack) { - struct net_device *dev = rif->dev; + struct net_device *dev = mlxsw_sp_rif_dev(rif); u8 old_mac_profile; u16 fid_index; int err; @@ -9088,6 +9740,104 @@ static int mlxsw_sp_router_port_pre_changeaddr_event(struct mlxsw_sp_rif *rif, return -ENOBUFS; } +static bool mlxsw_sp_router_netdevice_interesting(struct mlxsw_sp *mlxsw_sp, + struct net_device *dev) +{ + struct vlan_dev_priv *vlan; + + if (netif_is_lag_master(dev) || + netif_is_bridge_master(dev) || + mlxsw_sp_port_dev_check(dev) || + mlxsw_sp_netdev_is_ipip_ol(mlxsw_sp, dev) || + netif_is_l3_master(dev)) + return true; + + if (!is_vlan_dev(dev)) + return false; + + vlan = vlan_dev_priv(dev); + return netif_is_lag_master(vlan->real_dev) || + netif_is_bridge_master(vlan->real_dev) || + mlxsw_sp_port_dev_check(vlan->real_dev); +} + +static struct mlxsw_sp_crif * +mlxsw_sp_crif_register(struct mlxsw_sp_router *router, struct net_device *dev) +{ + struct mlxsw_sp_crif *crif; + int err; + + if (WARN_ON(mlxsw_sp_crif_lookup(router, dev))) + return NULL; + + crif = mlxsw_sp_crif_alloc(dev); + if (!crif) + return ERR_PTR(-ENOMEM); + + err = mlxsw_sp_crif_insert(router, crif); + if (err) + goto err_netdev_insert; + + return crif; + +err_netdev_insert: + mlxsw_sp_crif_free(crif); + return ERR_PTR(err); +} + +static void mlxsw_sp_crif_unregister(struct mlxsw_sp_router *router, + struct mlxsw_sp_crif *crif) +{ + struct mlxsw_sp_nexthop *nh, *tmp; + + mlxsw_sp_crif_remove(router, crif); + + list_for_each_entry_safe(nh, tmp, &crif->nexthop_list, crif_list_node) + mlxsw_sp_nexthop_type_fini(router->mlxsw_sp, nh); + + if (crif->rif) + crif->can_destroy = true; + else + mlxsw_sp_crif_free(crif); +} + +static int mlxsw_sp_netdevice_register(struct mlxsw_sp_router *router, + struct net_device *dev) +{ + struct mlxsw_sp_crif *crif; + + if (!mlxsw_sp_router_netdevice_interesting(router->mlxsw_sp, dev)) + return 0; + + crif = mlxsw_sp_crif_register(router, dev); + return PTR_ERR_OR_ZERO(crif); +} + +static void mlxsw_sp_netdevice_unregister(struct mlxsw_sp_router *router, + struct net_device *dev) +{ + struct mlxsw_sp_crif *crif; + + if (!mlxsw_sp_router_netdevice_interesting(router->mlxsw_sp, dev)) + return; + + /* netdev_run_todo(), by way of netdev_wait_allrefs_any(), rebroadcasts + * the NETDEV_UNREGISTER message, so we can get here twice. If that's + * what happened, the netdevice state is NETREG_UNREGISTERED. In that + * case, we expect to have collected the CRIF already, and warn if it + * still exists. Otherwise we expect the CRIF to exist. + */ + crif = mlxsw_sp_crif_lookup(router, dev); + if (dev->reg_state == NETREG_UNREGISTERED) { + if (!WARN_ON(crif)) + return; + } + if (WARN_ON(!crif)) + return; + + mlxsw_sp_crif_unregister(router, crif); +} + static bool mlxsw_sp_is_offload_xstats_event(unsigned long event) { switch (event) { @@ -9197,10 +9947,11 @@ static int mlxsw_sp_port_vrf_join(struct mlxsw_sp *mlxsw_sp, */ rif = mlxsw_sp_rif_find_by_dev(mlxsw_sp, l3_dev); if (rif) - __mlxsw_sp_inetaddr_event(mlxsw_sp, l3_dev, NETDEV_DOWN, + __mlxsw_sp_inetaddr_event(mlxsw_sp, l3_dev, NETDEV_DOWN, false, extack); - return __mlxsw_sp_inetaddr_event(mlxsw_sp, l3_dev, NETDEV_UP, extack); + return __mlxsw_sp_inetaddr_event(mlxsw_sp, l3_dev, NETDEV_UP, false, + extack); } static void mlxsw_sp_port_vrf_leave(struct mlxsw_sp *mlxsw_sp, @@ -9211,7 +9962,7 @@ static void mlxsw_sp_port_vrf_leave(struct mlxsw_sp *mlxsw_sp, rif = mlxsw_sp_rif_find_by_dev(mlxsw_sp, l3_dev); if (!rif) return; - __mlxsw_sp_inetaddr_event(mlxsw_sp, l3_dev, NETDEV_DOWN, NULL); + __mlxsw_sp_inetaddr_event(mlxsw_sp, l3_dev, NETDEV_DOWN, false, NULL); } static bool mlxsw_sp_is_vrf_event(unsigned long event, void *ptr) @@ -9254,6 +10005,233 @@ mlxsw_sp_netdevice_vrf_event(struct net_device *l3_dev, unsigned long event, return err; } +struct mlxsw_sp_router_replay_inetaddr_up { + struct mlxsw_sp *mlxsw_sp; + struct netlink_ext_ack *extack; + unsigned int done; + bool deslavement; +}; + +static int mlxsw_sp_router_replay_inetaddr_up(struct net_device *dev, + struct netdev_nested_priv *priv) +{ + struct mlxsw_sp_router_replay_inetaddr_up *ctx = priv->data; + bool nomaster = ctx->deslavement; + struct mlxsw_sp_crif *crif; + int err; + + if (mlxsw_sp_dev_addr_list_empty(dev)) + return 0; + + crif = mlxsw_sp_crif_lookup(ctx->mlxsw_sp->router, dev); + if (!crif || crif->rif) + return 0; + + if (!mlxsw_sp_rif_should_config(crif->rif, dev, NETDEV_UP)) + return 0; + + err = __mlxsw_sp_inetaddr_event(ctx->mlxsw_sp, dev, NETDEV_UP, + nomaster, ctx->extack); + if (err) + return err; + + ctx->done++; + return 0; +} + +static int mlxsw_sp_router_unreplay_inetaddr_up(struct net_device *dev, + struct netdev_nested_priv *priv) +{ + struct mlxsw_sp_router_replay_inetaddr_up *ctx = priv->data; + bool nomaster = ctx->deslavement; + struct mlxsw_sp_crif *crif; + + if (!ctx->done) + return 0; + + if (mlxsw_sp_dev_addr_list_empty(dev)) + return 0; + + crif = mlxsw_sp_crif_lookup(ctx->mlxsw_sp->router, dev); + if (!crif || !crif->rif) + return 0; + + /* We are rolling back NETDEV_UP, so ask for that. */ + if (!mlxsw_sp_rif_should_config(crif->rif, dev, NETDEV_UP)) + return 0; + + __mlxsw_sp_inetaddr_event(ctx->mlxsw_sp, dev, NETDEV_DOWN, nomaster, + NULL); + + ctx->done--; + return 0; +} + +int mlxsw_sp_netdevice_enslavement_replay(struct mlxsw_sp *mlxsw_sp, + struct net_device *upper_dev, + struct netlink_ext_ack *extack) +{ + struct mlxsw_sp_router_replay_inetaddr_up ctx = { + .mlxsw_sp = mlxsw_sp, + .extack = extack, + .deslavement = false, + }; + struct netdev_nested_priv priv = { + .data = &ctx, + }; + int err; + + err = mlxsw_sp_router_replay_inetaddr_up(upper_dev, &priv); + if (err) + return err; + + err = netdev_walk_all_upper_dev_rcu(upper_dev, + mlxsw_sp_router_replay_inetaddr_up, + &priv); + if (err) + goto err_replay_up; + + return 0; + +err_replay_up: + netdev_walk_all_upper_dev_rcu(upper_dev, + mlxsw_sp_router_unreplay_inetaddr_up, + &priv); + mlxsw_sp_router_unreplay_inetaddr_up(upper_dev, &priv); + return err; +} + +void mlxsw_sp_netdevice_deslavement_replay(struct mlxsw_sp *mlxsw_sp, + struct net_device *dev) +{ + struct mlxsw_sp_router_replay_inetaddr_up ctx = { + .mlxsw_sp = mlxsw_sp, + .deslavement = true, + }; + struct netdev_nested_priv priv = { + .data = &ctx, + }; + + mlxsw_sp_router_replay_inetaddr_up(dev, &priv); +} + +static int +mlxsw_sp_port_vid_router_join_existing(struct mlxsw_sp_port *mlxsw_sp_port, + u16 vid, struct net_device *dev, + struct netlink_ext_ack *extack) +{ + struct mlxsw_sp_port_vlan *mlxsw_sp_port_vlan; + + mlxsw_sp_port_vlan = mlxsw_sp_port_vlan_find_by_vid(mlxsw_sp_port, + vid); + if (WARN_ON(!mlxsw_sp_port_vlan)) + return -EINVAL; + + return mlxsw_sp_port_vlan_router_join_existing(mlxsw_sp_port_vlan, + dev, extack); +} + +static void +mlxsw_sp_port_vid_router_leave(struct mlxsw_sp_port *mlxsw_sp_port, u16 vid, + struct net_device *dev) +{ + struct mlxsw_sp_port_vlan *mlxsw_sp_port_vlan; + + mlxsw_sp_port_vlan = mlxsw_sp_port_vlan_find_by_vid(mlxsw_sp_port, + vid); + if (WARN_ON(!mlxsw_sp_port_vlan)) + return; + + __mlxsw_sp_port_vlan_router_leave(mlxsw_sp_port_vlan); +} + +static int __mlxsw_sp_router_port_join_lag(struct mlxsw_sp_port *mlxsw_sp_port, + struct net_device *lag_dev, + struct netlink_ext_ack *extack) +{ + u16 default_vid = MLXSW_SP_DEFAULT_VID; + struct net_device *upper_dev; + struct list_head *iter; + int done = 0; + u16 vid; + int err; + + err = mlxsw_sp_port_vid_router_join_existing(mlxsw_sp_port, default_vid, + lag_dev, extack); + if (err) + return err; + + netdev_for_each_upper_dev_rcu(lag_dev, upper_dev, iter) { + if (!is_vlan_dev(upper_dev)) + continue; + + vid = vlan_dev_vlan_id(upper_dev); + err = mlxsw_sp_port_vid_router_join_existing(mlxsw_sp_port, vid, + upper_dev, extack); + if (err) + goto err_router_join_dev; + + ++done; + } + + return 0; + +err_router_join_dev: + netdev_for_each_upper_dev_rcu(lag_dev, upper_dev, iter) { + if (!is_vlan_dev(upper_dev)) + continue; + if (!done--) + break; + + vid = vlan_dev_vlan_id(upper_dev); + mlxsw_sp_port_vid_router_leave(mlxsw_sp_port, vid, upper_dev); + } + + mlxsw_sp_port_vid_router_leave(mlxsw_sp_port, default_vid, lag_dev); + return err; +} + +static void +__mlxsw_sp_router_port_leave_lag(struct mlxsw_sp_port *mlxsw_sp_port, + struct net_device *lag_dev) +{ + u16 default_vid = MLXSW_SP_DEFAULT_VID; + struct net_device *upper_dev; + struct list_head *iter; + u16 vid; + + netdev_for_each_upper_dev_rcu(lag_dev, upper_dev, iter) { + if (!is_vlan_dev(upper_dev)) + continue; + + vid = vlan_dev_vlan_id(upper_dev); + mlxsw_sp_port_vid_router_leave(mlxsw_sp_port, vid, upper_dev); + } + + mlxsw_sp_port_vid_router_leave(mlxsw_sp_port, default_vid, lag_dev); +} + +int mlxsw_sp_router_port_join_lag(struct mlxsw_sp_port *mlxsw_sp_port, + struct net_device *lag_dev, + struct netlink_ext_ack *extack) +{ + int err; + + mutex_lock(&mlxsw_sp_port->mlxsw_sp->router->lock); + err = __mlxsw_sp_router_port_join_lag(mlxsw_sp_port, lag_dev, extack); + mutex_unlock(&mlxsw_sp_port->mlxsw_sp->router->lock); + + return err; +} + +void mlxsw_sp_router_port_leave_lag(struct mlxsw_sp_port *mlxsw_sp_port, + struct net_device *lag_dev) +{ + mutex_lock(&mlxsw_sp_port->mlxsw_sp->router->lock); + __mlxsw_sp_router_port_leave_lag(mlxsw_sp_port, lag_dev); + mutex_unlock(&mlxsw_sp_port->mlxsw_sp->router->lock); +} + static int mlxsw_sp_router_netdevice_event(struct notifier_block *nb, unsigned long event, void *ptr) { @@ -9267,6 +10245,15 @@ static int mlxsw_sp_router_netdevice_event(struct notifier_block *nb, mutex_lock(&mlxsw_sp->router->lock); + if (event == NETDEV_REGISTER) { + err = mlxsw_sp_netdevice_register(router, dev); + if (err) + /* No need to roll this back, UNREGISTER will collect it + * anyhow. + */ + goto out; + } + if (mlxsw_sp_is_offload_xstats_event(event)) err = mlxsw_sp_netdevice_offload_xstats_cmd(mlxsw_sp, dev, event, ptr); @@ -9281,11 +10268,49 @@ static int mlxsw_sp_router_netdevice_event(struct notifier_block *nb, else if (mlxsw_sp_is_vrf_event(event, ptr)) err = mlxsw_sp_netdevice_vrf_event(dev, event, ptr); + if (event == NETDEV_UNREGISTER) + mlxsw_sp_netdevice_unregister(router, dev); + +out: mutex_unlock(&mlxsw_sp->router->lock); return notifier_from_errno(err); } +struct mlxsw_sp_macvlan_replay { + struct mlxsw_sp *mlxsw_sp; + struct netlink_ext_ack *extack; +}; + +static int mlxsw_sp_macvlan_replay_upper(struct net_device *dev, + struct netdev_nested_priv *priv) +{ + const struct mlxsw_sp_macvlan_replay *rms = priv->data; + struct netlink_ext_ack *extack = rms->extack; + struct mlxsw_sp *mlxsw_sp = rms->mlxsw_sp; + + if (!netif_is_macvlan(dev)) + return 0; + + return mlxsw_sp_rif_macvlan_add(mlxsw_sp, dev, extack); +} + +static int mlxsw_sp_macvlan_replay(struct mlxsw_sp_rif *rif, + struct netlink_ext_ack *extack) +{ + struct mlxsw_sp_macvlan_replay rms = { + .mlxsw_sp = rif->mlxsw_sp, + .extack = extack, + }; + struct netdev_nested_priv priv = { + .data = &rms, + }; + + return netdev_walk_all_upper_dev_rcu(mlxsw_sp_rif_dev(rif), + mlxsw_sp_macvlan_replay_upper, + &priv); +} + static int __mlxsw_sp_rif_macvlan_flush(struct net_device *dev, struct netdev_nested_priv *priv) { @@ -9300,15 +10325,15 @@ static int __mlxsw_sp_rif_macvlan_flush(struct net_device *dev, static int mlxsw_sp_rif_macvlan_flush(struct mlxsw_sp_rif *rif) { + struct net_device *dev = mlxsw_sp_rif_dev(rif); struct netdev_nested_priv priv = { .data = (void *)rif, }; - if (!netif_is_macvlan_port(rif->dev)) + if (!netif_is_macvlan_port(dev)) return 0; - netdev_warn(rif->dev, "Router interface is deleted. Upper macvlans will not work\n"); - return netdev_walk_all_upper_dev_rcu(rif->dev, + return netdev_walk_all_upper_dev_rcu(dev, __mlxsw_sp_rif_macvlan_flush, &priv); } @@ -9329,6 +10354,7 @@ static void mlxsw_sp_rif_subport_setup(struct mlxsw_sp_rif *rif, static int mlxsw_sp_rif_subport_op(struct mlxsw_sp_rif *rif, bool enable) { + struct net_device *dev = mlxsw_sp_rif_dev(rif); struct mlxsw_sp *mlxsw_sp = rif->mlxsw_sp; struct mlxsw_sp_rif_subport *rif_subport; char ritr_pl[MLXSW_REG_RITR_LEN]; @@ -9336,8 +10362,8 @@ static int mlxsw_sp_rif_subport_op(struct mlxsw_sp_rif *rif, bool enable) rif_subport = mlxsw_sp_rif_subport_rif(rif); mlxsw_reg_ritr_pack(ritr_pl, enable, MLXSW_REG_RITR_SP_IF, - rif->rif_index, rif->vr_id, rif->dev->mtu); - mlxsw_reg_ritr_mac_pack(ritr_pl, rif->dev->dev_addr); + rif->rif_index, rif->vr_id, dev->mtu); + mlxsw_reg_ritr_mac_pack(ritr_pl, dev->dev_addr); mlxsw_reg_ritr_if_mac_profile_id_set(ritr_pl, rif->mac_profile_id); efid = mlxsw_sp_fid_index(rif->fid); mlxsw_reg_ritr_sp_if_pack(ritr_pl, rif_subport->lag, @@ -9350,6 +10376,7 @@ static int mlxsw_sp_rif_subport_op(struct mlxsw_sp_rif *rif, bool enable) static int mlxsw_sp_rif_subport_configure(struct mlxsw_sp_rif *rif, struct netlink_ext_ack *extack) { + struct net_device *dev = mlxsw_sp_rif_dev(rif); u8 mac_profile; int err; @@ -9363,7 +10390,11 @@ static int mlxsw_sp_rif_subport_configure(struct mlxsw_sp_rif *rif, if (err) goto err_rif_subport_op; - err = mlxsw_sp_rif_fdb_op(rif->mlxsw_sp, rif->dev->dev_addr, + err = mlxsw_sp_macvlan_replay(rif, extack); + if (err) + goto err_macvlan_replay; + + err = mlxsw_sp_rif_fdb_op(rif->mlxsw_sp, dev->dev_addr, mlxsw_sp_fid_index(rif->fid), true); if (err) goto err_rif_fdb_op; @@ -9375,9 +10406,11 @@ static int mlxsw_sp_rif_subport_configure(struct mlxsw_sp_rif *rif, return 0; err_fid_rif_set: - mlxsw_sp_rif_fdb_op(rif->mlxsw_sp, rif->dev->dev_addr, + mlxsw_sp_rif_fdb_op(rif->mlxsw_sp, dev->dev_addr, mlxsw_sp_fid_index(rif->fid), false); err_rif_fdb_op: + mlxsw_sp_rif_macvlan_flush(rif); +err_macvlan_replay: mlxsw_sp_rif_subport_op(rif, false); err_rif_subport_op: mlxsw_sp_rif_mac_profile_put(rif->mlxsw_sp, mac_profile); @@ -9386,10 +10419,11 @@ err_rif_subport_op: static void mlxsw_sp_rif_subport_deconfigure(struct mlxsw_sp_rif *rif) { + struct net_device *dev = mlxsw_sp_rif_dev(rif); struct mlxsw_sp_fid *fid = rif->fid; mlxsw_sp_fid_rif_unset(fid); - mlxsw_sp_rif_fdb_op(rif->mlxsw_sp, rif->dev->dev_addr, + mlxsw_sp_rif_fdb_op(rif->mlxsw_sp, dev->dev_addr, mlxsw_sp_fid_index(fid), false); mlxsw_sp_rif_macvlan_flush(rif); mlxsw_sp_rif_subport_op(rif, false); @@ -9398,6 +10432,7 @@ static void mlxsw_sp_rif_subport_deconfigure(struct mlxsw_sp_rif *rif) static struct mlxsw_sp_fid * mlxsw_sp_rif_subport_fid_get(struct mlxsw_sp_rif *rif, + const struct mlxsw_sp_rif_params *params, struct netlink_ext_ack *extack) { return mlxsw_sp_fid_rfid_get(rif->mlxsw_sp, rif->rif_index); @@ -9415,12 +10450,13 @@ static const struct mlxsw_sp_rif_ops mlxsw_sp_rif_subport_ops = { static int mlxsw_sp_rif_fid_op(struct mlxsw_sp_rif *rif, u16 fid, bool enable) { enum mlxsw_reg_ritr_if_type type = MLXSW_REG_RITR_FID_IF; + struct net_device *dev = mlxsw_sp_rif_dev(rif); struct mlxsw_sp *mlxsw_sp = rif->mlxsw_sp; char ritr_pl[MLXSW_REG_RITR_LEN]; mlxsw_reg_ritr_pack(ritr_pl, enable, type, rif->rif_index, rif->vr_id, - rif->dev->mtu); - mlxsw_reg_ritr_mac_pack(ritr_pl, rif->dev->dev_addr); + dev->mtu); + mlxsw_reg_ritr_mac_pack(ritr_pl, dev->dev_addr); mlxsw_reg_ritr_if_mac_profile_id_set(ritr_pl, rif->mac_profile_id); mlxsw_reg_ritr_fid_if_fid_set(ritr_pl, fid); @@ -9435,6 +10471,7 @@ u16 mlxsw_sp_router_port(const struct mlxsw_sp *mlxsw_sp) static int mlxsw_sp_rif_fid_configure(struct mlxsw_sp_rif *rif, struct netlink_ext_ack *extack) { + struct net_device *dev = mlxsw_sp_rif_dev(rif); struct mlxsw_sp *mlxsw_sp = rif->mlxsw_sp; u16 fid_index = mlxsw_sp_fid_index(rif->fid); u8 mac_profile; @@ -9460,7 +10497,11 @@ static int mlxsw_sp_rif_fid_configure(struct mlxsw_sp_rif *rif, if (err) goto err_fid_bc_flood_set; - err = mlxsw_sp_rif_fdb_op(rif->mlxsw_sp, rif->dev->dev_addr, + err = mlxsw_sp_macvlan_replay(rif, extack); + if (err) + goto err_macvlan_replay; + + err = mlxsw_sp_rif_fdb_op(rif->mlxsw_sp, dev->dev_addr, mlxsw_sp_fid_index(rif->fid), true); if (err) goto err_rif_fdb_op; @@ -9472,9 +10513,11 @@ static int mlxsw_sp_rif_fid_configure(struct mlxsw_sp_rif *rif, return 0; err_fid_rif_set: - mlxsw_sp_rif_fdb_op(rif->mlxsw_sp, rif->dev->dev_addr, + mlxsw_sp_rif_fdb_op(rif->mlxsw_sp, dev->dev_addr, mlxsw_sp_fid_index(rif->fid), false); err_rif_fdb_op: + mlxsw_sp_rif_macvlan_flush(rif); +err_macvlan_replay: mlxsw_sp_fid_flood_set(rif->fid, MLXSW_SP_FLOOD_TYPE_BC, mlxsw_sp_router_port(mlxsw_sp), false); err_fid_bc_flood_set: @@ -9489,12 +10532,13 @@ err_rif_fid_op: static void mlxsw_sp_rif_fid_deconfigure(struct mlxsw_sp_rif *rif) { + struct net_device *dev = mlxsw_sp_rif_dev(rif); u16 fid_index = mlxsw_sp_fid_index(rif->fid); struct mlxsw_sp *mlxsw_sp = rif->mlxsw_sp; struct mlxsw_sp_fid *fid = rif->fid; mlxsw_sp_fid_rif_unset(fid); - mlxsw_sp_rif_fdb_op(rif->mlxsw_sp, rif->dev->dev_addr, + mlxsw_sp_rif_fdb_op(rif->mlxsw_sp, dev->dev_addr, mlxsw_sp_fid_index(fid), false); mlxsw_sp_rif_macvlan_flush(rif); mlxsw_sp_fid_flood_set(rif->fid, MLXSW_SP_FLOOD_TYPE_BC, @@ -9507,9 +10551,12 @@ static void mlxsw_sp_rif_fid_deconfigure(struct mlxsw_sp_rif *rif) static struct mlxsw_sp_fid * mlxsw_sp_rif_fid_fid_get(struct mlxsw_sp_rif *rif, + const struct mlxsw_sp_rif_params *params, struct netlink_ext_ack *extack) { - return mlxsw_sp_fid_8021d_get(rif->mlxsw_sp, rif->dev->ifindex); + int rif_ifindex = mlxsw_sp_rif_dev_ifindex(rif); + + return mlxsw_sp_fid_8021d_get(rif->mlxsw_sp, rif_ifindex); } static void mlxsw_sp_rif_fid_fdb_del(struct mlxsw_sp_rif *rif, const char *mac) @@ -9517,7 +10564,7 @@ static void mlxsw_sp_rif_fid_fdb_del(struct mlxsw_sp_rif *rif, const char *mac) struct switchdev_notifier_fdb_info info = {}; struct net_device *dev; - dev = br_fdb_find_port(rif->dev, mac, 0); + dev = br_fdb_find_port(mlxsw_sp_rif_dev(rif), mac, 0); if (!dev) return; @@ -9538,36 +10585,33 @@ static const struct mlxsw_sp_rif_ops mlxsw_sp_rif_fid_ops = { static struct mlxsw_sp_fid * mlxsw_sp_rif_vlan_fid_get(struct mlxsw_sp_rif *rif, + const struct mlxsw_sp_rif_params *params, struct netlink_ext_ack *extack) { + struct net_device *dev = mlxsw_sp_rif_dev(rif); struct net_device *br_dev; - u16 vid; - int err; - if (is_vlan_dev(rif->dev)) { - vid = vlan_dev_vlan_id(rif->dev); - br_dev = vlan_dev_real_dev(rif->dev); + if (WARN_ON(!params->vid)) + return ERR_PTR(-EINVAL); + + if (is_vlan_dev(dev)) { + br_dev = vlan_dev_real_dev(dev); if (WARN_ON(!netif_is_bridge_master(br_dev))) return ERR_PTR(-EINVAL); - } else { - err = br_vlan_get_pvid(rif->dev, &vid); - if (err < 0 || !vid) { - NL_SET_ERR_MSG_MOD(extack, "Couldn't determine bridge PVID"); - return ERR_PTR(-EINVAL); - } } - return mlxsw_sp_fid_8021q_get(rif->mlxsw_sp, vid); + return mlxsw_sp_fid_8021q_get(rif->mlxsw_sp, params->vid); } static void mlxsw_sp_rif_vlan_fdb_del(struct mlxsw_sp_rif *rif, const char *mac) { + struct net_device *rif_dev = mlxsw_sp_rif_dev(rif); struct switchdev_notifier_fdb_info info = {}; u16 vid = mlxsw_sp_fid_8021q_vid(rif->fid); struct net_device *br_dev; struct net_device *dev; - br_dev = is_vlan_dev(rif->dev) ? vlan_dev_real_dev(rif->dev) : rif->dev; + br_dev = is_vlan_dev(rif_dev) ? vlan_dev_real_dev(rif_dev) : rif_dev; dev = br_fdb_find_port(br_dev, mac, vid); if (!dev) return; @@ -9581,11 +10625,12 @@ static void mlxsw_sp_rif_vlan_fdb_del(struct mlxsw_sp_rif *rif, const char *mac) static int mlxsw_sp_rif_vlan_op(struct mlxsw_sp_rif *rif, u16 vid, u16 efid, bool enable) { + struct net_device *dev = mlxsw_sp_rif_dev(rif); struct mlxsw_sp *mlxsw_sp = rif->mlxsw_sp; char ritr_pl[MLXSW_REG_RITR_LEN]; mlxsw_reg_ritr_vlan_if_pack(ritr_pl, enable, rif->rif_index, rif->vr_id, - rif->dev->mtu, rif->dev->dev_addr, + dev->mtu, dev->dev_addr, rif->mac_profile_id, vid, efid); return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(ritr), ritr_pl); @@ -9594,6 +10639,7 @@ static int mlxsw_sp_rif_vlan_op(struct mlxsw_sp_rif *rif, u16 vid, u16 efid, static int mlxsw_sp_rif_vlan_configure(struct mlxsw_sp_rif *rif, u16 efid, struct netlink_ext_ack *extack) { + struct net_device *dev = mlxsw_sp_rif_dev(rif); u16 vid = mlxsw_sp_fid_8021q_vid(rif->fid); struct mlxsw_sp *mlxsw_sp = rif->mlxsw_sp; u8 mac_profile; @@ -9619,7 +10665,11 @@ static int mlxsw_sp_rif_vlan_configure(struct mlxsw_sp_rif *rif, u16 efid, if (err) goto err_fid_bc_flood_set; - err = mlxsw_sp_rif_fdb_op(rif->mlxsw_sp, rif->dev->dev_addr, + err = mlxsw_sp_macvlan_replay(rif, extack); + if (err) + goto err_macvlan_replay; + + err = mlxsw_sp_rif_fdb_op(rif->mlxsw_sp, dev->dev_addr, mlxsw_sp_fid_index(rif->fid), true); if (err) goto err_rif_fdb_op; @@ -9631,9 +10681,11 @@ static int mlxsw_sp_rif_vlan_configure(struct mlxsw_sp_rif *rif, u16 efid, return 0; err_fid_rif_set: - mlxsw_sp_rif_fdb_op(rif->mlxsw_sp, rif->dev->dev_addr, + mlxsw_sp_rif_fdb_op(rif->mlxsw_sp, dev->dev_addr, mlxsw_sp_fid_index(rif->fid), false); err_rif_fdb_op: + mlxsw_sp_rif_macvlan_flush(rif); +err_macvlan_replay: mlxsw_sp_fid_flood_set(rif->fid, MLXSW_SP_FLOOD_TYPE_BC, mlxsw_sp_router_port(mlxsw_sp), false); err_fid_bc_flood_set: @@ -9648,11 +10700,12 @@ err_rif_vlan_fid_op: static void mlxsw_sp_rif_vlan_deconfigure(struct mlxsw_sp_rif *rif) { + struct net_device *dev = mlxsw_sp_rif_dev(rif); u16 vid = mlxsw_sp_fid_8021q_vid(rif->fid); struct mlxsw_sp *mlxsw_sp = rif->mlxsw_sp; mlxsw_sp_fid_rif_unset(rif->fid); - mlxsw_sp_rif_fdb_op(rif->mlxsw_sp, rif->dev->dev_addr, + mlxsw_sp_rif_fdb_op(rif->mlxsw_sp, dev->dev_addr, mlxsw_sp_fid_index(rif->fid), false); mlxsw_sp_rif_macvlan_flush(rif); mlxsw_sp_fid_flood_set(rif->fid, MLXSW_SP_FLOOD_TYPE_BC, @@ -9719,12 +10772,13 @@ mlxsw_sp1_rif_ipip_lb_configure(struct mlxsw_sp_rif *rif, struct netlink_ext_ack *extack) { struct mlxsw_sp_rif_ipip_lb *lb_rif = mlxsw_sp_rif_ipip_lb_rif(rif); - u32 ul_tb_id = mlxsw_sp_ipip_dev_ul_tb_id(rif->dev); + struct net_device *dev = mlxsw_sp_rif_dev(rif); + u32 ul_tb_id = mlxsw_sp_ipip_dev_ul_tb_id(dev); struct mlxsw_sp *mlxsw_sp = rif->mlxsw_sp; struct mlxsw_sp_vr *ul_vr; int err; - ul_vr = mlxsw_sp_vr_get(mlxsw_sp, ul_tb_id, NULL); + ul_vr = mlxsw_sp_vr_get(mlxsw_sp, ul_tb_id, extack); if (IS_ERR(ul_vr)) return PTR_ERR(ul_vr); @@ -9786,6 +10840,7 @@ mlxsw_sp_rif_ipip_lb_ul_rif_op(struct mlxsw_sp_rif *ul_rif, bool enable) static struct mlxsw_sp_rif * mlxsw_sp_ul_rif_create(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_vr *vr, + struct mlxsw_sp_crif *ul_crif, struct netlink_ext_ack *extack) { struct mlxsw_sp_rif *ul_rif; @@ -9799,7 +10854,8 @@ mlxsw_sp_ul_rif_create(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_vr *vr, return ERR_PTR(err); } - ul_rif = mlxsw_sp_rif_alloc(sizeof(*ul_rif), rif_index, vr->id, NULL); + ul_rif = mlxsw_sp_rif_alloc(sizeof(*ul_rif), rif_index, vr->id, + ul_crif); if (!ul_rif) { err = -ENOMEM; goto err_rif_alloc; @@ -9817,7 +10873,7 @@ mlxsw_sp_ul_rif_create(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_vr *vr, ul_rif_op_err: mlxsw_sp->router->rifs[rif_index] = NULL; - kfree(ul_rif); + mlxsw_sp_rif_free(ul_rif); err_rif_alloc: mlxsw_sp_rif_index_free(mlxsw_sp, rif_index, rif_entries); return ERR_PTR(err); @@ -9832,12 +10888,13 @@ static void mlxsw_sp_ul_rif_destroy(struct mlxsw_sp_rif *ul_rif) atomic_sub(rif_entries, &mlxsw_sp->router->rifs_count); mlxsw_sp_rif_ipip_lb_ul_rif_op(ul_rif, false); mlxsw_sp->router->rifs[ul_rif->rif_index] = NULL; - kfree(ul_rif); + mlxsw_sp_rif_free(ul_rif); mlxsw_sp_rif_index_free(mlxsw_sp, rif_index, rif_entries); } static struct mlxsw_sp_rif * mlxsw_sp_ul_rif_get(struct mlxsw_sp *mlxsw_sp, u32 tb_id, + struct mlxsw_sp_crif *ul_crif, struct netlink_ext_ack *extack) { struct mlxsw_sp_vr *vr; @@ -9850,7 +10907,7 @@ mlxsw_sp_ul_rif_get(struct mlxsw_sp *mlxsw_sp, u32 tb_id, if (refcount_inc_not_zero(&vr->ul_rif_refcnt)) return vr->ul_rif; - vr->ul_rif = mlxsw_sp_ul_rif_create(mlxsw_sp, vr, extack); + vr->ul_rif = mlxsw_sp_ul_rif_create(mlxsw_sp, vr, ul_crif, extack); if (IS_ERR(vr->ul_rif)) { err = PTR_ERR(vr->ul_rif); goto err_ul_rif_create; @@ -9888,7 +10945,7 @@ int mlxsw_sp_router_ul_rif_get(struct mlxsw_sp *mlxsw_sp, u32 ul_tb_id, int err = 0; mutex_lock(&mlxsw_sp->router->lock); - ul_rif = mlxsw_sp_ul_rif_get(mlxsw_sp, ul_tb_id, NULL); + ul_rif = mlxsw_sp_ul_rif_get(mlxsw_sp, ul_tb_id, NULL, NULL); if (IS_ERR(ul_rif)) { err = PTR_ERR(ul_rif); goto out; @@ -9918,12 +10975,13 @@ mlxsw_sp2_rif_ipip_lb_configure(struct mlxsw_sp_rif *rif, struct netlink_ext_ack *extack) { struct mlxsw_sp_rif_ipip_lb *lb_rif = mlxsw_sp_rif_ipip_lb_rif(rif); - u32 ul_tb_id = mlxsw_sp_ipip_dev_ul_tb_id(rif->dev); + struct net_device *dev = mlxsw_sp_rif_dev(rif); + u32 ul_tb_id = mlxsw_sp_ipip_dev_ul_tb_id(dev); struct mlxsw_sp *mlxsw_sp = rif->mlxsw_sp; struct mlxsw_sp_rif *ul_rif; int err; - ul_rif = mlxsw_sp_ul_rif_get(mlxsw_sp, ul_tb_id, NULL); + ul_rif = mlxsw_sp_ul_rif_get(mlxsw_sp, ul_tb_id, NULL, extack); if (IS_ERR(ul_rif)) return PTR_ERR(ul_rif); @@ -10041,11 +11099,12 @@ err_rifs_table_init: static void mlxsw_sp_rifs_fini(struct mlxsw_sp *mlxsw_sp) { + int max_rifs = MLXSW_CORE_RES_GET(mlxsw_sp->core, MAX_RIFS); struct devlink *devlink = priv_to_devlink(mlxsw_sp->core); int i; WARN_ON_ONCE(atomic_read(&mlxsw_sp->router->rifs_count)); - for (i = 0; i < MLXSW_CORE_RES_GET(mlxsw_sp->core, MAX_RIFS); i++) + for (i = 0; i < max_rifs; i++) WARN_ON_ONCE(mlxsw_sp->router->rifs[i]); devl_resource_occ_get_unregister(devlink, MLXSW_SP_RESOURCE_RIFS); @@ -10343,12 +11402,16 @@ static int mlxsw_sp_mp_hash_init(struct mlxsw_sp *mlxsw_sp) { bool old_inc_parsing_depth, new_inc_parsing_depth; struct mlxsw_sp_mp_hash_config config = {}; + struct net *net = mlxsw_sp_net(mlxsw_sp); char recr2_pl[MLXSW_REG_RECR2_LEN]; unsigned long bit; u32 seed; int err; - seed = jhash(mlxsw_sp->base_mac, sizeof(mlxsw_sp->base_mac), 0); + seed = READ_ONCE(net->ipv4.sysctl_fib_multipath_hash_seed).user_seed; + if (!seed) + seed = jhash(mlxsw_sp->base_mac, sizeof(mlxsw_sp->base_mac), 0); + mlxsw_reg_recr2_pack(recr2_pl, seed); mlxsw_sp_mp4_hash_init(mlxsw_sp, &config); mlxsw_sp_mp6_hash_init(mlxsw_sp, &config); @@ -10381,11 +11444,23 @@ err_reg_write: old_inc_parsing_depth); return err; } + +static void mlxsw_sp_mp_hash_fini(struct mlxsw_sp *mlxsw_sp) +{ + bool old_inc_parsing_depth = mlxsw_sp->router->inc_parsing_depth; + + mlxsw_sp_mp_hash_parsing_depth_adjust(mlxsw_sp, old_inc_parsing_depth, + false); +} #else static int mlxsw_sp_mp_hash_init(struct mlxsw_sp *mlxsw_sp) { return 0; } + +static void mlxsw_sp_mp_hash_fini(struct mlxsw_sp *mlxsw_sp) +{ +} #endif static int mlxsw_sp_dscp_init(struct mlxsw_sp *mlxsw_sp) @@ -10432,28 +11507,41 @@ static void __mlxsw_sp_router_fini(struct mlxsw_sp *mlxsw_sp) mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(rgcr), rgcr_pl); } -static int mlxsw_sp_lb_rif_init(struct mlxsw_sp *mlxsw_sp) +static int mlxsw_sp_lb_rif_init(struct mlxsw_sp *mlxsw_sp, + struct netlink_ext_ack *extack) { - u16 lb_rif_index; + struct mlxsw_sp_router *router = mlxsw_sp->router; + struct mlxsw_sp_rif *lb_rif; int err; + router->lb_crif = mlxsw_sp_crif_alloc(NULL); + if (!router->lb_crif) + return -ENOMEM; + /* Create a generic loopback RIF associated with the main table * (default VRF). Any table can be used, but the main table exists - * anyway, so we do not waste resources. + * anyway, so we do not waste resources. Loopback RIFs are usually + * created with a NULL CRIF, but this RIF is used as a fallback RIF + * for blackhole nexthops, and nexthops expect to have a valid CRIF. */ - err = mlxsw_sp_router_ul_rif_get(mlxsw_sp, RT_TABLE_MAIN, - &lb_rif_index); - if (err) - return err; - - mlxsw_sp->router->lb_rif_index = lb_rif_index; + lb_rif = mlxsw_sp_ul_rif_get(mlxsw_sp, RT_TABLE_MAIN, router->lb_crif, + extack); + if (IS_ERR(lb_rif)) { + err = PTR_ERR(lb_rif); + goto err_ul_rif_get; + } return 0; + +err_ul_rif_get: + mlxsw_sp_crif_free(router->lb_crif); + return err; } static void mlxsw_sp_lb_rif_fini(struct mlxsw_sp *mlxsw_sp) { - mlxsw_sp_router_ul_rif_put(mlxsw_sp, mlxsw_sp->router->lb_rif_index); + mlxsw_sp_ul_rif_put(mlxsw_sp->router->lb_crif->rif); + mlxsw_sp_crif_free(mlxsw_sp->router->lb_crif); } static int mlxsw_sp1_router_init(struct mlxsw_sp *mlxsw_sp) @@ -10492,6 +11580,7 @@ int mlxsw_sp_router_init(struct mlxsw_sp *mlxsw_sp, struct netlink_ext_ack *extack) { struct mlxsw_sp_router *router; + struct notifier_block *nb; int err; router = kzalloc(sizeof(*mlxsw_sp->router), GFP_KERNEL); @@ -10513,14 +11602,19 @@ int mlxsw_sp_router_init(struct mlxsw_sp *mlxsw_sp, if (err) goto err_router_init; - err = mlxsw_sp_rifs_init(mlxsw_sp); - if (err) - goto err_rifs_init; - err = mlxsw_sp->router_ops->ipips_init(mlxsw_sp); if (err) goto err_ipips_init; + err = rhashtable_init(&mlxsw_sp->router->crif_ht, + &mlxsw_sp_crif_ht_params); + if (err) + goto err_crif_ht_init; + + err = mlxsw_sp_rifs_init(mlxsw_sp); + if (err) + goto err_rifs_init; + err = rhashtable_init(&mlxsw_sp->router->nexthop_ht, &mlxsw_sp_nexthop_ht_params); if (err) @@ -10544,7 +11638,7 @@ int mlxsw_sp_router_init(struct mlxsw_sp *mlxsw_sp, if (err) goto err_vrs_init; - err = mlxsw_sp_lb_rif_init(mlxsw_sp); + err = mlxsw_sp_lb_rif_init(mlxsw_sp, extack); if (err) goto err_lb_rif_init; @@ -10570,12 +11664,30 @@ int mlxsw_sp_router_init(struct mlxsw_sp *mlxsw_sp, if (err) goto err_register_inet6addr_notifier; + router->inetaddr_valid_nb.notifier_call = mlxsw_sp_inetaddr_valid_event; + err = register_inetaddr_validator_notifier(&router->inetaddr_valid_nb); + if (err) + goto err_register_inetaddr_valid_notifier; + + nb = &router->inet6addr_valid_nb; + nb->notifier_call = mlxsw_sp_inet6addr_valid_event; + err = register_inet6addr_validator_notifier(nb); + if (err) + goto err_register_inet6addr_valid_notifier; + mlxsw_sp->router->netevent_nb.notifier_call = mlxsw_sp_router_netevent_event; err = register_netevent_notifier(&mlxsw_sp->router->netevent_nb); if (err) goto err_register_netevent_notifier; + mlxsw_sp->router->netdevice_nb.notifier_call = + mlxsw_sp_router_netdevice_event; + err = register_netdevice_notifier_net(mlxsw_sp_net(mlxsw_sp), + &mlxsw_sp->router->netdevice_nb); + if (err) + goto err_register_netdev_notifier; + mlxsw_sp->router->nexthop_nb.notifier_call = mlxsw_sp_nexthop_obj_event; err = register_nexthop_notifier(mlxsw_sp_net(mlxsw_sp), @@ -10591,30 +11703,28 @@ int mlxsw_sp_router_init(struct mlxsw_sp *mlxsw_sp, if (err) goto err_register_fib_notifier; - mlxsw_sp->router->netdevice_nb.notifier_call = - mlxsw_sp_router_netdevice_event; - err = register_netdevice_notifier_net(mlxsw_sp_net(mlxsw_sp), - &mlxsw_sp->router->netdevice_nb); - if (err) - goto err_register_netdev_notifier; - return 0; -err_register_netdev_notifier: - unregister_fib_notifier(mlxsw_sp_net(mlxsw_sp), - &mlxsw_sp->router->fib_nb); err_register_fib_notifier: unregister_nexthop_notifier(mlxsw_sp_net(mlxsw_sp), &mlxsw_sp->router->nexthop_nb); err_register_nexthop_notifier: + unregister_netdevice_notifier_net(mlxsw_sp_net(mlxsw_sp), + &router->netdevice_nb); +err_register_netdev_notifier: unregister_netevent_notifier(&mlxsw_sp->router->netevent_nb); err_register_netevent_notifier: + unregister_inet6addr_validator_notifier(&router->inet6addr_valid_nb); +err_register_inet6addr_valid_notifier: + unregister_inetaddr_validator_notifier(&router->inetaddr_valid_nb); +err_register_inetaddr_valid_notifier: unregister_inet6addr_notifier(&router->inet6addr_nb); err_register_inet6addr_notifier: unregister_inetaddr_notifier(&router->inetaddr_nb); err_register_inetaddr_notifier: mlxsw_core_flush_owq(); err_dscp_init: + mlxsw_sp_mp_hash_fini(mlxsw_sp); err_mp_hash_init: mlxsw_sp_neigh_fini(mlxsw_sp); err_neigh_init: @@ -10630,10 +11740,12 @@ err_lpm_init: err_nexthop_group_ht_init: rhashtable_destroy(&mlxsw_sp->router->nexthop_ht); err_nexthop_ht_init: - mlxsw_sp_ipips_fini(mlxsw_sp); -err_ipips_init: mlxsw_sp_rifs_fini(mlxsw_sp); err_rifs_init: + rhashtable_destroy(&mlxsw_sp->router->crif_ht); +err_crif_ht_init: + mlxsw_sp_ipips_fini(mlxsw_sp); +err_ipips_init: __mlxsw_sp_router_fini(mlxsw_sp); err_router_init: cancel_delayed_work_sync(&mlxsw_sp->router->nh_grp_activity_dw); @@ -10645,27 +11757,32 @@ err_router_ops_init: void mlxsw_sp_router_fini(struct mlxsw_sp *mlxsw_sp) { - unregister_netdevice_notifier_net(mlxsw_sp_net(mlxsw_sp), - &mlxsw_sp->router->netdevice_nb); - unregister_fib_notifier(mlxsw_sp_net(mlxsw_sp), - &mlxsw_sp->router->fib_nb); + struct mlxsw_sp_router *router = mlxsw_sp->router; + + unregister_fib_notifier(mlxsw_sp_net(mlxsw_sp), &router->fib_nb); unregister_nexthop_notifier(mlxsw_sp_net(mlxsw_sp), - &mlxsw_sp->router->nexthop_nb); - unregister_netevent_notifier(&mlxsw_sp->router->netevent_nb); - unregister_inet6addr_notifier(&mlxsw_sp->router->inet6addr_nb); - unregister_inetaddr_notifier(&mlxsw_sp->router->inetaddr_nb); + &router->nexthop_nb); + unregister_netdevice_notifier_net(mlxsw_sp_net(mlxsw_sp), + &router->netdevice_nb); + unregister_netevent_notifier(&router->netevent_nb); + unregister_inet6addr_validator_notifier(&router->inet6addr_valid_nb); + unregister_inetaddr_validator_notifier(&router->inetaddr_valid_nb); + unregister_inet6addr_notifier(&router->inet6addr_nb); + unregister_inetaddr_notifier(&router->inetaddr_nb); mlxsw_core_flush_owq(); + mlxsw_sp_mp_hash_fini(mlxsw_sp); mlxsw_sp_neigh_fini(mlxsw_sp); mlxsw_sp_lb_rif_fini(mlxsw_sp); mlxsw_sp_vrs_fini(mlxsw_sp); mlxsw_sp_mr_fini(mlxsw_sp); mlxsw_sp_lpm_fini(mlxsw_sp); - rhashtable_destroy(&mlxsw_sp->router->nexthop_group_ht); - rhashtable_destroy(&mlxsw_sp->router->nexthop_ht); - mlxsw_sp_ipips_fini(mlxsw_sp); + rhashtable_destroy(&router->nexthop_group_ht); + rhashtable_destroy(&router->nexthop_ht); mlxsw_sp_rifs_fini(mlxsw_sp); + rhashtable_destroy(&mlxsw_sp->router->crif_ht); + mlxsw_sp_ipips_fini(mlxsw_sp); __mlxsw_sp_router_fini(mlxsw_sp); - cancel_delayed_work_sync(&mlxsw_sp->router->nh_grp_activity_dw); - mutex_destroy(&mlxsw_sp->router->lock); - kfree(mlxsw_sp->router); + cancel_delayed_work_sync(&router->nh_grp_activity_dw); + mutex_destroy(&router->lock); + kfree(router); } |