* [PATCH net-next 0/8] mlxsw: Add support for nexthop objects
@ 2020-11-19 13:08 Ido Schimmel
2020-11-19 13:08 ` [PATCH net-next 1/8] mlxsw: spectrum_router: " Ido Schimmel
` (8 more replies)
0 siblings, 9 replies; 10+ messages in thread
From: Ido Schimmel @ 2020-11-19 13:08 UTC (permalink / raw)
To: netdev; +Cc: davem, kuba, dsahern, jiri, mlxsw, Ido Schimmel
From: Ido Schimmel <idosch@nvidia.com>
This patch set adds support for nexthop objects in mlxsw. Nexthop
objects are treated as another front-end for programming nexthops, in
addition to the existing IPv4 and IPv6 front-ends.
Patch #1 registers a listener to the nexthop notification chain and
parses the nexthop information into the existing mlxsw data structures
that are already used by the IPv4 and IPv6 front-ends. Blackhole
nexthops are currently rejected. Support will be added in a follow-up
patch set.
Patch #2 extends mlxsw to resolve its internal nexthop objects from the
nexthop identifier encoded in the FIB info of the notified routes.
Patch #3 finally removes the limitation of rejecting routes that use
nexthop objects.
Patch #4 adds a selftest.
Patches #5-#8 add generic forwarding selftests that can be used with
veth pairs or physical loopbacks.
Ido Schimmel (8):
mlxsw: spectrum_router: Add support for nexthop objects
mlxsw: spectrum_router: Enable resolution of nexthop groups from
nexthop objects
mlxsw: spectrum_router: Allow programming routes with nexthop objects
selftests: mlxsw: Add nexthop objects configuration tests
selftests: forwarding: Do not configure nexthop objects twice
selftests: forwarding: Test IPv4 routes with IPv6 link-local nexthops
selftests: forwarding: Add device-only nexthop test
selftests: forwarding: Add multipath tunneling nexthop test
.../ethernet/mellanox/mlxsw/spectrum_router.c | 498 +++++++++++++++++-
.../ethernet/mellanox/mlxsw/spectrum_router.h | 1 +
.../selftests/drivers/net/mlxsw/rtnetlink.sh | 189 +++++++
.../net/forwarding/gre_multipath_nh.sh | 356 +++++++++++++
.../net/forwarding/router_mpath_nh.sh | 12 +-
.../selftests/net/forwarding/router_nh.sh | 160 ++++++
6 files changed, 1201 insertions(+), 15 deletions(-)
create mode 100755 tools/testing/selftests/net/forwarding/gre_multipath_nh.sh
create mode 100755 tools/testing/selftests/net/forwarding/router_nh.sh
--
2.28.0
^ permalink raw reply [flat|nested] 10+ messages in thread* [PATCH net-next 1/8] mlxsw: spectrum_router: Add support for nexthop objects 2020-11-19 13:08 [PATCH net-next 0/8] mlxsw: Add support for nexthop objects Ido Schimmel @ 2020-11-19 13:08 ` Ido Schimmel 2020-11-19 13:08 ` [PATCH net-next 2/8] mlxsw: spectrum_router: Enable resolution of nexthop groups from " Ido Schimmel ` (7 subsequent siblings) 8 siblings, 0 replies; 10+ messages in thread From: Ido Schimmel @ 2020-11-19 13:08 UTC (permalink / raw) To: netdev; +Cc: davem, kuba, dsahern, jiri, mlxsw, Ido Schimmel From: Ido Schimmel <idosch@nvidia.com> Register a listener to the nexthop notification chain and parse notified nexthop objects into the existing mlxsw nexthop data structures. Signed-off-by: Ido Schimmel <idosch@nvidia.com> Reviewed-by: Jiri Pirko <jiri@nvidia.com> --- .../ethernet/mellanox/mlxsw/spectrum_router.c | 453 ++++++++++++++++++ .../ethernet/mellanox/mlxsw/spectrum_router.h | 1 + 2 files changed, 454 insertions(+) diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c index 147dd8fab2af..480d47ef9362 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c @@ -2873,6 +2873,7 @@ struct mlxsw_sp_nexthop { enum mlxsw_sp_nexthop_group_type { MLXSW_SP_NEXTHOP_GROUP_TYPE_IPV4, MLXSW_SP_NEXTHOP_GROUP_TYPE_IPV6, + MLXSW_SP_NEXTHOP_GROUP_TYPE_OBJ, }; struct mlxsw_sp_nexthop_group_info { @@ -2894,6 +2895,9 @@ struct mlxsw_sp_nexthop_group { struct { struct fib_info *fi; } ipv4; + struct { + u32 id; + } obj; }; struct mlxsw_sp_nexthop_group_info *nhgi; enum mlxsw_sp_nexthop_group_type type; @@ -3012,6 +3016,7 @@ struct mlxsw_sp_nexthop_group_cmp_arg { union { struct fib_info *fi; struct mlxsw_sp_fib6_entry *fib6_entry; + u32 id; }; }; @@ -3074,6 +3079,8 @@ mlxsw_sp_nexthop_group_cmp(struct rhashtable_compare_arg *arg, const void *ptr) case MLXSW_SP_NEXTHOP_GROUP_TYPE_IPV6: return !mlxsw_sp_nexthop6_group_cmp(nh_grp, cmp_arg->fib6_entry); + case MLXSW_SP_NEXTHOP_GROUP_TYPE_OBJ: + return cmp_arg->id != nh_grp->obj.id; default: WARN_ON(1); return 1; @@ -3100,6 +3107,8 @@ static u32 mlxsw_sp_nexthop_group_hash_obj(const void *data, u32 len, u32 seed) val ^= jhash(&nh->gw_addr, sizeof(nh->gw_addr), seed); } return jhash(&val, sizeof(val), seed); + case MLXSW_SP_NEXTHOP_GROUP_TYPE_OBJ: + return jhash(&nh_grp->obj.id, sizeof(nh_grp->obj.id), seed); default: WARN_ON(1); return 0; @@ -3134,6 +3143,8 @@ mlxsw_sp_nexthop_group_hash(const void *data, u32 len, u32 seed) return jhash(&cmp_arg->fi, sizeof(cmp_arg->fi), seed); case MLXSW_SP_NEXTHOP_GROUP_TYPE_IPV6: return mlxsw_sp_nexthop6_group_hash(cmp_arg->fib6_entry, seed); + case MLXSW_SP_NEXTHOP_GROUP_TYPE_OBJ: + return jhash(&cmp_arg->id, sizeof(cmp_arg->id), seed); default: WARN_ON(1); return 0; @@ -3538,6 +3549,25 @@ mlxsw_sp_nexthop6_group_offload_refresh(struct mlxsw_sp *mlxsw_sp, __mlxsw_sp_nexthop6_group_offload_refresh(nh_grp, fib6_entry); } +static void +mlxsw_sp_nexthop_obj_group_offload_refresh(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_nexthop_group *nh_grp) +{ + /* Do not update the flags if the nexthop group is being destroyed + * since: + * 1. The nexthop objects is being deleted, in which case the flags are + * irrelevant. + * 2. The nexthop group was replaced by a newer group, in which case + * the flags of the nexthop object were already updated based on the + * new group. + */ + if (nh_grp->can_destroy) + return; + + nexthop_set_hw_flags(mlxsw_sp_net(mlxsw_sp), nh_grp->obj.id, + nh_grp->nhgi->adj_index_valid, false); +} + static void mlxsw_sp_nexthop_group_offload_refresh(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_nexthop_group *nh_grp) @@ -3549,6 +3579,9 @@ mlxsw_sp_nexthop_group_offload_refresh(struct mlxsw_sp *mlxsw_sp, case MLXSW_SP_NEXTHOP_GROUP_TYPE_IPV6: mlxsw_sp_nexthop6_group_offload_refresh(mlxsw_sp, nh_grp); break; + case MLXSW_SP_NEXTHOP_GROUP_TYPE_OBJ: + mlxsw_sp_nexthop_obj_group_offload_refresh(mlxsw_sp, nh_grp); + break; } } @@ -4088,6 +4121,413 @@ static void mlxsw_sp_nexthop_rif_gone_sync(struct mlxsw_sp *mlxsw_sp, } } +static int +mlxsw_sp_nexthop_obj_single_validate(struct mlxsw_sp *mlxsw_sp, + const struct nh_notifier_single_info *nh, + struct netlink_ext_ack *extack) +{ + int err = -EINVAL; + + if (nh->is_reject) + NL_SET_ERR_MSG_MOD(extack, "Blackhole nexthops are not supported"); + else if (nh->is_fdb) + NL_SET_ERR_MSG_MOD(extack, "FDB nexthops are not supported"); + else if (nh->has_encap) + NL_SET_ERR_MSG_MOD(extack, "Encapsulating nexthops are not supported"); + else + err = 0; + + return err; +} + +static int +mlxsw_sp_nexthop_obj_group_validate(struct mlxsw_sp *mlxsw_sp, + const struct nh_notifier_grp_info *nh_grp, + struct netlink_ext_ack *extack) +{ + int i; + + if (nh_grp->is_fdb) { + NL_SET_ERR_MSG_MOD(extack, "FDB nexthop groups are not supported"); + return -EINVAL; + } + + for (i = 0; i < nh_grp->num_nh; i++) { + const struct nh_notifier_single_info *nh; + int err; + + nh = &nh_grp->nh_entries[i].nh; + err = mlxsw_sp_nexthop_obj_single_validate(mlxsw_sp, nh, + extack); + if (err) + return err; + + /* Device only nexthops with an IPIP device are programmed as + * encapsulating adjacency entries. + */ + if (!nh->gw_family && + !mlxsw_sp_netdev_ipip_type(mlxsw_sp, nh->dev, NULL)) { + NL_SET_ERR_MSG_MOD(extack, "Nexthop group entry does not have a gateway"); + return -EINVAL; + } + } + + return 0; +} + +static int mlxsw_sp_nexthop_obj_validate(struct mlxsw_sp *mlxsw_sp, + unsigned long event, + struct nh_notifier_info *info) +{ + if (event != NEXTHOP_EVENT_REPLACE) + return 0; + + if (!info->is_grp) + return mlxsw_sp_nexthop_obj_single_validate(mlxsw_sp, info->nh, + info->extack); + return mlxsw_sp_nexthop_obj_group_validate(mlxsw_sp, info->nh_grp, + info->extack); +} + +static bool mlxsw_sp_nexthop_obj_is_gateway(struct mlxsw_sp *mlxsw_sp, + const struct nh_notifier_info *info) +{ + const struct net_device *dev; + + if (info->is_grp) + /* Already validated earlier. */ + return true; + + dev = info->nh->dev; + return info->nh->gw_family || + mlxsw_sp_netdev_ipip_type(mlxsw_sp, dev, NULL); +} + +static int +mlxsw_sp_nexthop_obj_init(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_nexthop_group *nh_grp, + struct mlxsw_sp_nexthop *nh, + struct nh_notifier_single_info *nh_obj, int weight) +{ + struct net_device *dev = nh_obj->dev; + int err; + + nh->nhgi = nh_grp->nhgi; + nh->nh_weight = weight; + + switch (nh_obj->gw_family) { + case AF_INET: + memcpy(&nh->gw_addr, &nh_obj->ipv4, sizeof(nh_obj->ipv4)); + nh->neigh_tbl = &arp_tbl; + break; + case AF_INET6: + memcpy(&nh->gw_addr, &nh_obj->ipv6, sizeof(nh_obj->ipv6)); +#if IS_ENABLED(CONFIG_IPV6) + nh->neigh_tbl = &nd_tbl; +#endif + 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; + + err = mlxsw_sp_nexthop_type_init(mlxsw_sp, nh, dev); + if (err) + goto err_type_init; + + return 0; + +err_type_init: + list_del(&nh->router_list_node); + mlxsw_sp_nexthop_counter_free(mlxsw_sp, nh); + return err; +} + +static void mlxsw_sp_nexthop_obj_fini(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_nexthop *nh) +{ + mlxsw_sp_nexthop_type_fini(mlxsw_sp, nh); + list_del(&nh->router_list_node); + mlxsw_sp_nexthop_counter_free(mlxsw_sp, nh); +} + +static int +mlxsw_sp_nexthop_obj_group_info_init(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_nexthop_group *nh_grp, + struct nh_notifier_info *info) +{ + unsigned int nhs = info->is_grp ? info->nh_grp->num_nh : 1; + struct mlxsw_sp_nexthop_group_info *nhgi; + struct mlxsw_sp_nexthop *nh; + int err, i; + + nhgi = kzalloc(struct_size(nhgi, nexthops, nhs), GFP_KERNEL); + if (!nhgi) + return -ENOMEM; + nh_grp->nhgi = nhgi; + nhgi->nh_grp = nh_grp; + nhgi->gateway = mlxsw_sp_nexthop_obj_is_gateway(mlxsw_sp, info); + nhgi->count = nhs; + for (i = 0; i < nhgi->count; i++) { + struct nh_notifier_single_info *nh_obj; + int weight; + + nh = &nhgi->nexthops[i]; + if (info->is_grp) { + nh_obj = &info->nh_grp->nh_entries[i].nh; + weight = info->nh_grp->nh_entries[i].weight; + } else { + nh_obj = info->nh; + weight = 1; + } + err = mlxsw_sp_nexthop_obj_init(mlxsw_sp, nh_grp, nh, nh_obj, + weight); + if (err) + goto err_nexthop_obj_init; + } + err = mlxsw_sp_nexthop_group_refresh(mlxsw_sp, nh_grp); + if (err) { + NL_SET_ERR_MSG_MOD(info->extack, "Failed to write adjacency entries to the device"); + goto err_group_refresh; + } + + return 0; + +err_group_refresh: + i = nhgi->count; +err_nexthop_obj_init: + for (i--; i >= 0; i--) { + nh = &nhgi->nexthops[i]; + mlxsw_sp_nexthop_obj_fini(mlxsw_sp, nh); + } + kfree(nhgi); + return err; +} + +static void +mlxsw_sp_nexthop_obj_group_info_fini(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_nexthop_group *nh_grp) +{ + struct mlxsw_sp_nexthop_group_info *nhgi = nh_grp->nhgi; + int i; + + for (i = nhgi->count - 1; i >= 0; i--) { + struct mlxsw_sp_nexthop *nh = &nhgi->nexthops[i]; + + mlxsw_sp_nexthop_obj_fini(mlxsw_sp, nh); + } + mlxsw_sp_nexthop_group_refresh(mlxsw_sp, nh_grp); + WARN_ON_ONCE(nhgi->adj_index_valid); + kfree(nhgi); +} + +static struct mlxsw_sp_nexthop_group * +mlxsw_sp_nexthop_obj_group_create(struct mlxsw_sp *mlxsw_sp, + struct nh_notifier_info *info) +{ + struct mlxsw_sp_nexthop_group *nh_grp; + int err; + + nh_grp = kzalloc(sizeof(*nh_grp), GFP_KERNEL); + if (!nh_grp) + return ERR_PTR(-ENOMEM); + INIT_LIST_HEAD(&nh_grp->fib_list); + nh_grp->type = MLXSW_SP_NEXTHOP_GROUP_TYPE_OBJ; + nh_grp->obj.id = info->id; + + err = mlxsw_sp_nexthop_obj_group_info_init(mlxsw_sp, nh_grp, info); + if (err) + goto err_nexthop_group_info_init; + + nh_grp->can_destroy = false; + + return nh_grp; + +err_nexthop_group_info_init: + kfree(nh_grp); + return ERR_PTR(err); +} + +static void +mlxsw_sp_nexthop_obj_group_destroy(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_nexthop_group *nh_grp) +{ + if (!nh_grp->can_destroy) + return; + mlxsw_sp_nexthop_obj_group_info_fini(mlxsw_sp, nh_grp); + WARN_ON_ONCE(!list_empty(&nh_grp->fib_list)); + kfree(nh_grp); +} + +static struct mlxsw_sp_nexthop_group * +mlxsw_sp_nexthop_obj_group_lookup(struct mlxsw_sp *mlxsw_sp, u32 id) +{ + struct mlxsw_sp_nexthop_group_cmp_arg cmp_arg; + + cmp_arg.type = MLXSW_SP_NEXTHOP_GROUP_TYPE_OBJ; + cmp_arg.id = id; + return rhashtable_lookup_fast(&mlxsw_sp->router->nexthop_group_ht, + &cmp_arg, + mlxsw_sp_nexthop_group_ht_params); +} + +static int mlxsw_sp_nexthop_obj_group_add(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_nexthop_group *nh_grp) +{ + return mlxsw_sp_nexthop_group_insert(mlxsw_sp, nh_grp); +} + +static int +mlxsw_sp_nexthop_obj_group_replace(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_nexthop_group *nh_grp, + struct mlxsw_sp_nexthop_group *old_nh_grp, + struct netlink_ext_ack *extack) +{ + struct mlxsw_sp_nexthop_group_info *old_nhgi = old_nh_grp->nhgi; + struct mlxsw_sp_nexthop_group_info *new_nhgi = nh_grp->nhgi; + int err; + + old_nh_grp->nhgi = new_nhgi; + new_nhgi->nh_grp = old_nh_grp; + nh_grp->nhgi = old_nhgi; + old_nhgi->nh_grp = nh_grp; + + if (old_nhgi->adj_index_valid && new_nhgi->adj_index_valid) { + /* Both the old adjacency index and the new one are valid. + * Routes are currently using the old one. Tell the device to + * replace the old adjacency index with the new one. + */ + err = mlxsw_sp_adj_index_mass_update(mlxsw_sp, old_nh_grp, + old_nhgi->adj_index, + old_nhgi->ecmp_size); + if (err) { + NL_SET_ERR_MSG_MOD(extack, "Failed to replace old adjacency index with new one"); + goto err_out; + } + } else if (old_nhgi->adj_index_valid && !new_nhgi->adj_index_valid) { + /* The old adjacency index is valid, while the new one is not. + * Iterate over all the routes using the group and change them + * to trap packets to the CPU. + */ + err = mlxsw_sp_nexthop_fib_entries_update(mlxsw_sp, old_nh_grp); + if (err) { + NL_SET_ERR_MSG_MOD(extack, "Failed to update routes to trap packets"); + goto err_out; + } + } else if (!old_nhgi->adj_index_valid && new_nhgi->adj_index_valid) { + /* The old adjacency index is invalid, while the new one is. + * Iterate over all the routes using the group and change them + * to forward packets using the new valid index. + */ + err = mlxsw_sp_nexthop_fib_entries_update(mlxsw_sp, old_nh_grp); + if (err) { + NL_SET_ERR_MSG_MOD(extack, "Failed to update routes to forward packets"); + goto err_out; + } + } + + /* Make sure the flags are set / cleared based on the new nexthop group + * information. + */ + mlxsw_sp_nexthop_obj_group_offload_refresh(mlxsw_sp, old_nh_grp); + + /* At this point 'nh_grp' is just a shell that is not used by anyone + * and its nexthop group info is the old info that was just replaced + * with the new one. Remove it. + */ + nh_grp->can_destroy = true; + mlxsw_sp_nexthop_obj_group_destroy(mlxsw_sp, nh_grp); + + return 0; + +err_out: + old_nhgi->nh_grp = old_nh_grp; + nh_grp->nhgi = new_nhgi; + new_nhgi->nh_grp = nh_grp; + old_nh_grp->nhgi = old_nhgi; + return err; +} + +static int mlxsw_sp_nexthop_obj_new(struct mlxsw_sp *mlxsw_sp, + struct nh_notifier_info *info) +{ + struct mlxsw_sp_nexthop_group *nh_grp, *old_nh_grp; + struct netlink_ext_ack *extack = info->extack; + int err; + + nh_grp = mlxsw_sp_nexthop_obj_group_create(mlxsw_sp, info); + if (IS_ERR(nh_grp)) + return PTR_ERR(nh_grp); + + old_nh_grp = mlxsw_sp_nexthop_obj_group_lookup(mlxsw_sp, info->id); + if (!old_nh_grp) + err = mlxsw_sp_nexthop_obj_group_add(mlxsw_sp, nh_grp); + else + err = mlxsw_sp_nexthop_obj_group_replace(mlxsw_sp, nh_grp, + old_nh_grp, extack); + + if (err) { + nh_grp->can_destroy = true; + mlxsw_sp_nexthop_obj_group_destroy(mlxsw_sp, nh_grp); + } + + return err; +} + +static void mlxsw_sp_nexthop_obj_del(struct mlxsw_sp *mlxsw_sp, + struct nh_notifier_info *info) +{ + struct mlxsw_sp_nexthop_group *nh_grp; + + nh_grp = mlxsw_sp_nexthop_obj_group_lookup(mlxsw_sp, info->id); + if (!nh_grp) + return; + + nh_grp->can_destroy = true; + mlxsw_sp_nexthop_group_remove(mlxsw_sp, nh_grp); + + /* If the group still has routes using it, then defer the delete + * operation until the last route using it is deleted. + */ + if (!list_empty(&nh_grp->fib_list)) + return; + mlxsw_sp_nexthop_obj_group_destroy(mlxsw_sp, nh_grp); +} + +static int mlxsw_sp_nexthop_obj_event(struct notifier_block *nb, + unsigned long event, void *ptr) +{ + struct nh_notifier_info *info = ptr; + struct mlxsw_sp_router *router; + int err = 0; + + router = container_of(nb, struct mlxsw_sp_router, nexthop_nb); + err = mlxsw_sp_nexthop_obj_validate(router->mlxsw_sp, event, info); + if (err) + goto out; + + mutex_lock(&router->lock); + + ASSERT_RTNL(); + + switch (event) { + case NEXTHOP_EVENT_REPLACE: + err = mlxsw_sp_nexthop_obj_new(router->mlxsw_sp, info); + break; + case NEXTHOP_EVENT_DEL: + mlxsw_sp_nexthop_obj_del(router->mlxsw_sp, info); + break; + default: + break; + } + + mutex_unlock(&router->lock); + +out: + return notifier_from_errno(err); +} + static bool mlxsw_sp_fi_is_gateway(const struct mlxsw_sp *mlxsw_sp, struct fib_info *fi) { @@ -8549,6 +8989,14 @@ int mlxsw_sp_router_init(struct mlxsw_sp *mlxsw_sp, if (err) goto err_register_netevent_notifier; + mlxsw_sp->router->nexthop_nb.notifier_call = + mlxsw_sp_nexthop_obj_event; + err = register_nexthop_notifier(mlxsw_sp_net(mlxsw_sp), + &mlxsw_sp->router->nexthop_nb, + extack); + if (err) + goto err_register_nexthop_notifier; + mlxsw_sp->router->fib_nb.notifier_call = mlxsw_sp_router_fib_event; err = register_fib_notifier(mlxsw_sp_net(mlxsw_sp), &mlxsw_sp->router->fib_nb, @@ -8559,6 +9007,9 @@ int mlxsw_sp_router_init(struct mlxsw_sp *mlxsw_sp, return 0; err_register_fib_notifier: + unregister_nexthop_notifier(mlxsw_sp_net(mlxsw_sp), + &mlxsw_sp->router->nexthop_nb); +err_register_nexthop_notifier: unregister_netevent_notifier(&mlxsw_sp->router->netevent_nb); err_register_netevent_notifier: unregister_inet6addr_notifier(&router->inet6addr_nb); @@ -8598,6 +9049,8 @@ void mlxsw_sp_router_fini(struct mlxsw_sp *mlxsw_sp) { unregister_fib_notifier(mlxsw_sp_net(mlxsw_sp), &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); diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.h index 8230f6ff02ed..023f70827db0 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.h +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.h @@ -58,6 +58,7 @@ struct mlxsw_sp_router { struct list_head nexthop_neighs_list; struct list_head ipip_list; bool aborted; + struct notifier_block nexthop_nb; struct notifier_block fib_nb; struct notifier_block netevent_nb; struct notifier_block inetaddr_nb; -- 2.28.0 ^ permalink raw reply related [flat|nested] 10+ messages in thread
* [PATCH net-next 2/8] mlxsw: spectrum_router: Enable resolution of nexthop groups from nexthop objects 2020-11-19 13:08 [PATCH net-next 0/8] mlxsw: Add support for nexthop objects Ido Schimmel 2020-11-19 13:08 ` [PATCH net-next 1/8] mlxsw: spectrum_router: " Ido Schimmel @ 2020-11-19 13:08 ` Ido Schimmel 2020-11-19 13:08 ` [PATCH net-next 3/8] mlxsw: spectrum_router: Allow programming routes with " Ido Schimmel ` (6 subsequent siblings) 8 siblings, 0 replies; 10+ messages in thread From: Ido Schimmel @ 2020-11-19 13:08 UTC (permalink / raw) To: netdev; +Cc: davem, kuba, dsahern, jiri, mlxsw, Ido Schimmel From: Ido Schimmel <idosch@nvidia.com> If the FIB info (i.e, 'struct fib_info', 'struct fib6_info') uses a nexthop object, then use the object's identifier to resolve the nexthop group. Signed-off-by: Ido Schimmel <idosch@nvidia.com> Reviewed-by: Jiri Pirko <jiri@nvidia.com> --- .../ethernet/mellanox/mlxsw/spectrum_router.c | 31 +++++++++++++++++++ 1 file changed, 31 insertions(+) diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c index 480d47ef9362..ad335d5b8d66 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c @@ -4648,12 +4648,21 @@ static int mlxsw_sp_nexthop4_group_get(struct mlxsw_sp *mlxsw_sp, { struct mlxsw_sp_nexthop_group *nh_grp; + if (fi->nh) { + nh_grp = mlxsw_sp_nexthop_obj_group_lookup(mlxsw_sp, + fi->nh->id); + if (WARN_ON_ONCE(!nh_grp)) + return -EINVAL; + goto out; + } + nh_grp = mlxsw_sp_nexthop4_group_lookup(mlxsw_sp, fi); if (!nh_grp) { nh_grp = mlxsw_sp_nexthop4_group_create(mlxsw_sp, fi); if (IS_ERR(nh_grp)) return PTR_ERR(nh_grp); } +out: list_add_tail(&fib_entry->nexthop_group_node, &nh_grp->fib_list); fib_entry->nh_group = nh_grp; return 0; @@ -4667,6 +4676,12 @@ static void mlxsw_sp_nexthop4_group_put(struct mlxsw_sp *mlxsw_sp, list_del(&fib_entry->nexthop_group_node); if (!list_empty(&nh_grp->fib_list)) return; + + if (nh_grp->type == MLXSW_SP_NEXTHOP_GROUP_TYPE_OBJ) { + mlxsw_sp_nexthop_obj_group_destroy(mlxsw_sp, nh_grp); + return; + } + mlxsw_sp_nexthop4_group_destroy(mlxsw_sp, nh_grp); } @@ -5957,8 +5972,17 @@ mlxsw_sp_nexthop6_group_destroy(struct mlxsw_sp *mlxsw_sp, static int mlxsw_sp_nexthop6_group_get(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_fib6_entry *fib6_entry) { + struct fib6_info *rt = mlxsw_sp_fib6_entry_rt(fib6_entry); struct mlxsw_sp_nexthop_group *nh_grp; + if (rt->nh) { + nh_grp = mlxsw_sp_nexthop_obj_group_lookup(mlxsw_sp, + rt->nh->id); + if (WARN_ON_ONCE(!nh_grp)) + return -EINVAL; + goto out; + } + nh_grp = mlxsw_sp_nexthop6_group_lookup(mlxsw_sp, fib6_entry); if (!nh_grp) { nh_grp = mlxsw_sp_nexthop6_group_create(mlxsw_sp, fib6_entry); @@ -5971,6 +5995,7 @@ static int mlxsw_sp_nexthop6_group_get(struct mlxsw_sp *mlxsw_sp, */ __mlxsw_sp_nexthop6_group_offload_refresh(nh_grp, fib6_entry); +out: list_add_tail(&fib6_entry->common.nexthop_group_node, &nh_grp->fib_list); fib6_entry->common.nh_group = nh_grp; @@ -5986,6 +6011,12 @@ static void mlxsw_sp_nexthop6_group_put(struct mlxsw_sp *mlxsw_sp, list_del(&fib_entry->nexthop_group_node); if (!list_empty(&nh_grp->fib_list)) return; + + if (nh_grp->type == MLXSW_SP_NEXTHOP_GROUP_TYPE_OBJ) { + mlxsw_sp_nexthop_obj_group_destroy(mlxsw_sp, nh_grp); + return; + } + mlxsw_sp_nexthop6_group_destroy(mlxsw_sp, nh_grp); } -- 2.28.0 ^ permalink raw reply related [flat|nested] 10+ messages in thread
* [PATCH net-next 3/8] mlxsw: spectrum_router: Allow programming routes with nexthop objects 2020-11-19 13:08 [PATCH net-next 0/8] mlxsw: Add support for nexthop objects Ido Schimmel 2020-11-19 13:08 ` [PATCH net-next 1/8] mlxsw: spectrum_router: " Ido Schimmel 2020-11-19 13:08 ` [PATCH net-next 2/8] mlxsw: spectrum_router: Enable resolution of nexthop groups from " Ido Schimmel @ 2020-11-19 13:08 ` Ido Schimmel 2020-11-19 13:08 ` [PATCH net-next 4/8] selftests: mlxsw: Add nexthop objects configuration tests Ido Schimmel ` (5 subsequent siblings) 8 siblings, 0 replies; 10+ messages in thread From: Ido Schimmel @ 2020-11-19 13:08 UTC (permalink / raw) To: netdev; +Cc: davem, kuba, dsahern, jiri, mlxsw, Ido Schimmel From: Ido Schimmel <idosch@nvidia.com> Now that the driver supports nexthop objects, the check is no longer necessary. Remove it. Signed-off-by: Ido Schimmel <idosch@nvidia.com> Reviewed-by: Jiri Pirko <jiri@nvidia.com> --- .../net/ethernet/mellanox/mlxsw/spectrum_router.c | 14 -------------- 1 file changed, 14 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c index ad335d5b8d66..42a7bec3fd88 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c @@ -7068,20 +7068,6 @@ static int mlxsw_sp_router_fib_event(struct notifier_block *nb, NL_SET_ERR_MSG_MOD(info->extack, "IPv6 gateway with IPv4 route is not supported"); return notifier_from_errno(-EINVAL); } - if (fen_info->fi->nh) { - NL_SET_ERR_MSG_MOD(info->extack, "IPv4 route with nexthop objects is not supported"); - return notifier_from_errno(-EINVAL); - } - } else if (info->family == AF_INET6) { - struct fib6_entry_notifier_info *fen6_info; - - fen6_info = container_of(info, - struct fib6_entry_notifier_info, - info); - if (fen6_info->rt->nh) { - NL_SET_ERR_MSG_MOD(info->extack, "IPv6 route with nexthop objects is not supported"); - return notifier_from_errno(-EINVAL); - } } break; } -- 2.28.0 ^ permalink raw reply related [flat|nested] 10+ messages in thread
* [PATCH net-next 4/8] selftests: mlxsw: Add nexthop objects configuration tests 2020-11-19 13:08 [PATCH net-next 0/8] mlxsw: Add support for nexthop objects Ido Schimmel ` (2 preceding siblings ...) 2020-11-19 13:08 ` [PATCH net-next 3/8] mlxsw: spectrum_router: Allow programming routes with " Ido Schimmel @ 2020-11-19 13:08 ` Ido Schimmel 2020-11-19 13:08 ` [PATCH net-next 5/8] selftests: forwarding: Do not configure nexthop objects twice Ido Schimmel ` (4 subsequent siblings) 8 siblings, 0 replies; 10+ messages in thread From: Ido Schimmel @ 2020-11-19 13:08 UTC (permalink / raw) To: netdev; +Cc: davem, kuba, dsahern, jiri, mlxsw, Ido Schimmel From: Ido Schimmel <idosch@nvidia.com> Test that unsupported nexthop objects are rejected and that offload indication is correctly set on: nexthop objects, nexthop group objects and routes associated these objects. Signed-off-by: Ido Schimmel <idosch@nvidia.com> --- .../selftests/drivers/net/mlxsw/rtnetlink.sh | 189 ++++++++++++++++++ 1 file changed, 189 insertions(+) diff --git a/tools/testing/selftests/drivers/net/mlxsw/rtnetlink.sh b/tools/testing/selftests/drivers/net/mlxsw/rtnetlink.sh index f4031002d5e9..5de47d72f8c9 100755 --- a/tools/testing/selftests/drivers/net/mlxsw/rtnetlink.sh +++ b/tools/testing/selftests/drivers/net/mlxsw/rtnetlink.sh @@ -29,6 +29,10 @@ ALL_TESTS=" bridge_extern_learn_test neigh_offload_test nexthop_offload_test + nexthop_obj_invalid_test + nexthop_obj_offload_test + nexthop_obj_group_offload_test + nexthop_obj_route_offload_test devlink_reload_test " NUM_NETIFS=2 @@ -674,6 +678,191 @@ nexthop_offload_test() sysctl_restore net.ipv6.conf.$swp2.keep_addr_on_down } +nexthop_obj_invalid_test() +{ + # Test that invalid nexthop object configurations are rejected + RET=0 + + simple_if_init $swp1 192.0.2.1/24 2001:db8:1::1/64 + simple_if_init $swp2 192.0.2.2/24 2001:db8:1::2/64 + setup_wait + + ip nexthop add id 1 via 192.0.2.3 fdb + check_fail $? "managed to configure an FDB nexthop when should not" + + ip nexthop add id 1 encap mpls 200/300 via 192.0.2.3 dev $swp1 + check_fail $? "managed to configure a nexthop with MPLS encap when should not" + + ip nexthop add id 1 blackhole + check_fail $? "managed to configure a blackhole nexthop when should not" + + ip nexthop add id 1 dev $swp1 + ip nexthop add id 2 dev $swp1 + ip nexthop add id 10 group 1/2 + check_fail $? "managed to configure a nexthop group with device-only nexthops when should not" + + log_test "nexthop objects - invalid configurations" + + ip nexthop del id 2 + ip nexthop del id 1 + + simple_if_fini $swp2 192.0.2.2/24 2001:db8:1::2/64 + simple_if_fini $swp1 192.0.2.1/24 2001:db8:1::1/64 +} + +nexthop_obj_offload_test() +{ + # Test offload indication of nexthop objects + RET=0 + + simple_if_init $swp1 192.0.2.1/24 2001:db8:1::1/64 + simple_if_init $swp2 + setup_wait + + ip nexthop add id 1 via 192.0.2.2 dev $swp1 + ip neigh replace 192.0.2.2 lladdr 00:11:22:33:44:55 nud reachable \ + dev $swp1 + + busywait "$TIMEOUT" wait_for_offload \ + ip nexthop show id 1 + check_err $? "nexthop not marked as offloaded when should" + + ip neigh replace 192.0.2.2 nud failed dev $swp1 + busywait "$TIMEOUT" not wait_for_offload \ + ip nexthop show id 1 + check_err $? "nexthop marked as offloaded after setting neigh to failed state" + + ip neigh replace 192.0.2.2 lladdr 00:11:22:33:44:55 nud reachable \ + dev $swp1 + busywait "$TIMEOUT" wait_for_offload \ + ip nexthop show id 1 + check_err $? "nexthop not marked as offloaded after neigh replace" + + ip nexthop replace id 1 via 192.0.2.3 dev $swp1 + busywait "$TIMEOUT" not wait_for_offload \ + ip nexthop show id 1 + check_err $? "nexthop marked as offloaded after replacing to use an invalid address" + + ip nexthop replace id 1 via 192.0.2.2 dev $swp1 + busywait "$TIMEOUT" wait_for_offload \ + ip nexthop show id 1 + check_err $? "nexthop not marked as offloaded after replacing to use a valid address" + + log_test "nexthop objects offload indication" + + ip neigh del 192.0.2.2 dev $swp1 + ip nexthop del id 1 + + simple_if_fini $swp2 + simple_if_fini $swp1 192.0.2.1/24 2001:db8:1::1/64 +} + +nexthop_obj_group_offload_test() +{ + # Test offload indication of nexthop group objects + RET=0 + + simple_if_init $swp1 192.0.2.1/24 2001:db8:1::1/64 + simple_if_init $swp2 + setup_wait + + ip nexthop add id 1 via 192.0.2.2 dev $swp1 + ip nexthop add id 2 via 2001:db8:1::2 dev $swp1 + ip nexthop add id 10 group 1/2 + ip neigh replace 192.0.2.2 lladdr 00:11:22:33:44:55 nud reachable \ + dev $swp1 + ip neigh replace 192.0.2.3 lladdr 00:11:22:33:44:55 nud reachable \ + dev $swp1 + ip neigh replace 2001:db8:1::2 lladdr 00:11:22:33:44:55 nud reachable \ + dev $swp1 + + busywait "$TIMEOUT" wait_for_offload \ + ip nexthop show id 1 + check_err $? "IPv4 nexthop not marked as offloaded when should" + busywait "$TIMEOUT" wait_for_offload \ + ip nexthop show id 2 + check_err $? "IPv6 nexthop not marked as offloaded when should" + busywait "$TIMEOUT" wait_for_offload \ + ip nexthop show id 10 + check_err $? "nexthop group not marked as offloaded when should" + + # Invalidate nexthop id 1 + ip neigh replace 192.0.2.2 nud failed dev $swp1 + busywait "$TIMEOUT" not wait_for_offload \ + ip nexthop show id 10 + check_fail $? "nexthop group not marked as offloaded with one valid nexthop" + + # Invalidate nexthop id 2 + ip neigh replace 2001:db8:1::2 nud failed dev $swp1 + busywait "$TIMEOUT" not wait_for_offload \ + ip nexthop show id 10 + check_err $? "nexthop group marked as offloaded when should not" + + # Revalidate nexthop id 1 + ip nexthop replace id 1 via 192.0.2.3 dev $swp1 + busywait "$TIMEOUT" wait_for_offload \ + ip nexthop show id 10 + check_err $? "nexthop group not marked as offloaded after revalidating nexthop" + + log_test "nexthop group objects offload indication" + + ip neigh del 2001:db8:1::2 dev $swp1 + ip neigh del 192.0.2.3 dev $swp1 + ip neigh del 192.0.2.2 dev $swp1 + ip nexthop del id 10 + ip nexthop del id 2 + ip nexthop del id 1 + + simple_if_fini $swp2 + simple_if_fini $swp1 192.0.2.1/24 2001:db8:1::1/64 +} + +nexthop_obj_route_offload_test() +{ + # Test offload indication of routes using nexthop objects + RET=0 + + simple_if_init $swp1 192.0.2.1/24 2001:db8:1::1/64 + simple_if_init $swp2 + setup_wait + + ip nexthop add id 1 via 192.0.2.2 dev $swp1 + ip neigh replace 192.0.2.2 lladdr 00:11:22:33:44:55 nud reachable \ + dev $swp1 + ip neigh replace 192.0.2.3 lladdr 00:11:22:33:44:55 nud reachable \ + dev $swp1 + + ip route replace 198.51.100.0/24 nhid 1 + busywait "$TIMEOUT" wait_for_offload \ + ip route show 198.51.100.0/24 + check_err $? "route not marked as offloaded when using valid nexthop" + + ip nexthop replace id 1 via 192.0.2.3 dev $swp1 + busywait "$TIMEOUT" wait_for_offload \ + ip route show 198.51.100.0/24 + check_err $? "route not marked as offloaded after replacing valid nexthop with a valid one" + + ip nexthop replace id 1 via 192.0.2.4 dev $swp1 + busywait "$TIMEOUT" not wait_for_offload \ + ip route show 198.51.100.0/24 + check_err $? "route marked as offloaded after replacing valid nexthop with an invalid one" + + ip nexthop replace id 1 via 192.0.2.2 dev $swp1 + busywait "$TIMEOUT" wait_for_offload \ + ip route show 198.51.100.0/24 + check_err $? "route not marked as offloaded after replacing invalid nexthop with a valid one" + + log_test "routes using nexthop objects offload indication" + + ip route del 198.51.100.0/24 + ip neigh del 192.0.2.3 dev $swp1 + ip neigh del 192.0.2.2 dev $swp1 + ip nexthop del id 1 + + simple_if_fini $swp2 + simple_if_fini $swp1 192.0.2.1/24 2001:db8:1::1/64 +} + devlink_reload_test() { # Test that after executing all the above configuration tests, a -- 2.28.0 ^ permalink raw reply related [flat|nested] 10+ messages in thread
* [PATCH net-next 5/8] selftests: forwarding: Do not configure nexthop objects twice 2020-11-19 13:08 [PATCH net-next 0/8] mlxsw: Add support for nexthop objects Ido Schimmel ` (3 preceding siblings ...) 2020-11-19 13:08 ` [PATCH net-next 4/8] selftests: mlxsw: Add nexthop objects configuration tests Ido Schimmel @ 2020-11-19 13:08 ` Ido Schimmel 2020-11-19 13:08 ` [PATCH net-next 6/8] selftests: forwarding: Test IPv4 routes with IPv6 link-local nexthops Ido Schimmel ` (3 subsequent siblings) 8 siblings, 0 replies; 10+ messages in thread From: Ido Schimmel @ 2020-11-19 13:08 UTC (permalink / raw) To: netdev; +Cc: davem, kuba, dsahern, jiri, mlxsw, Ido Schimmel From: Ido Schimmel <idosch@nvidia.com> routing_nh_obj() is used to configure the nexthop objects employed by the test, but it is called twice resulting in "RTNETLINK answers: File exists" messages. Remove the first call, so that the function is only called after setup_wait(), when all the interfaces are up and ready. Signed-off-by: Ido Schimmel <idosch@nvidia.com> --- tools/testing/selftests/net/forwarding/router_mpath_nh.sh | 1 - 1 file changed, 1 deletion(-) diff --git a/tools/testing/selftests/net/forwarding/router_mpath_nh.sh b/tools/testing/selftests/net/forwarding/router_mpath_nh.sh index cf3d26c233e8..6067477ff326 100755 --- a/tools/testing/selftests/net/forwarding/router_mpath_nh.sh +++ b/tools/testing/selftests/net/forwarding/router_mpath_nh.sh @@ -312,7 +312,6 @@ setup_prepare() router1_create router2_create - routing_nh_obj forwarding_enable } -- 2.28.0 ^ permalink raw reply related [flat|nested] 10+ messages in thread
* [PATCH net-next 6/8] selftests: forwarding: Test IPv4 routes with IPv6 link-local nexthops 2020-11-19 13:08 [PATCH net-next 0/8] mlxsw: Add support for nexthop objects Ido Schimmel ` (4 preceding siblings ...) 2020-11-19 13:08 ` [PATCH net-next 5/8] selftests: forwarding: Do not configure nexthop objects twice Ido Schimmel @ 2020-11-19 13:08 ` Ido Schimmel 2020-11-19 13:08 ` [PATCH net-next 7/8] selftests: forwarding: Add device-only nexthop test Ido Schimmel ` (2 subsequent siblings) 8 siblings, 0 replies; 10+ messages in thread From: Ido Schimmel @ 2020-11-19 13:08 UTC (permalink / raw) To: netdev; +Cc: davem, kuba, dsahern, jiri, mlxsw, Ido Schimmel From: Ido Schimmel <idosch@nvidia.com> In addition to IPv4 multipath tests with IPv4 nexthops, also test IPv4 multipath with nexthops that use IPv6 link-local addresses. Signed-off-by: Ido Schimmel <idosch@nvidia.com> --- .../selftests/net/forwarding/router_mpath_nh.sh | 11 +++++++++++ 1 file changed, 11 insertions(+) diff --git a/tools/testing/selftests/net/forwarding/router_mpath_nh.sh b/tools/testing/selftests/net/forwarding/router_mpath_nh.sh index 6067477ff326..e8c2573d5232 100755 --- a/tools/testing/selftests/net/forwarding/router_mpath_nh.sh +++ b/tools/testing/selftests/net/forwarding/router_mpath_nh.sh @@ -280,6 +280,17 @@ multipath_test() multipath4_test "Weighted MP 2:1" 2 1 multipath4_test "Weighted MP 11:45" 11 45 + log_info "Running IPv4 multipath tests with IPv6 link-local nexthops" + ip nexthop replace id 101 via fe80:2::22 dev $rp12 + ip nexthop replace id 102 via fe80:3::23 dev $rp13 + + multipath4_test "ECMP" 1 1 + multipath4_test "Weighted MP 2:1" 2 1 + multipath4_test "Weighted MP 11:45" 11 45 + + ip nexthop replace id 102 via 169.254.3.23 dev $rp13 + ip nexthop replace id 101 via 169.254.2.22 dev $rp12 + log_info "Running IPv6 multipath tests" multipath6_test "ECMP" 1 1 multipath6_test "Weighted MP 2:1" 2 1 -- 2.28.0 ^ permalink raw reply related [flat|nested] 10+ messages in thread
* [PATCH net-next 7/8] selftests: forwarding: Add device-only nexthop test 2020-11-19 13:08 [PATCH net-next 0/8] mlxsw: Add support for nexthop objects Ido Schimmel ` (5 preceding siblings ...) 2020-11-19 13:08 ` [PATCH net-next 6/8] selftests: forwarding: Test IPv4 routes with IPv6 link-local nexthops Ido Schimmel @ 2020-11-19 13:08 ` Ido Schimmel 2020-11-19 13:08 ` [PATCH net-next 8/8] selftests: forwarding: Add multipath tunneling " Ido Schimmel 2020-11-20 23:30 ` [PATCH net-next 0/8] mlxsw: Add support for nexthop objects patchwork-bot+netdevbpf 8 siblings, 0 replies; 10+ messages in thread From: Ido Schimmel @ 2020-11-19 13:08 UTC (permalink / raw) To: netdev; +Cc: davem, kuba, dsahern, jiri, mlxsw, Ido Schimmel From: Ido Schimmel <idosch@nvidia.com> In a similar fashion to router_multipath.sh and its nexthop objects version router_mpath_nh.sh, create a nexthop objects version of router.sh. It reuses the same topology, but uses device-only nexthop objects instead of legacy ones. Signed-off-by: Ido Schimmel <idosch@nvidia.com> --- .../selftests/net/forwarding/router_nh.sh | 160 ++++++++++++++++++ 1 file changed, 160 insertions(+) create mode 100755 tools/testing/selftests/net/forwarding/router_nh.sh diff --git a/tools/testing/selftests/net/forwarding/router_nh.sh b/tools/testing/selftests/net/forwarding/router_nh.sh new file mode 100755 index 000000000000..f3a53738bdcc --- /dev/null +++ b/tools/testing/selftests/net/forwarding/router_nh.sh @@ -0,0 +1,160 @@ +#!/bin/bash +# SPDX-License-Identifier: GPL-2.0 + +ALL_TESTS=" + ping_ipv4 + ping_ipv6 +" + +NUM_NETIFS=4 +source lib.sh +source tc_common.sh + +h1_create() +{ + vrf_create "vrf-h1" + ip link set dev $h1 master vrf-h1 + + ip link set dev vrf-h1 up + ip link set dev $h1 up + + ip address add 192.0.2.2/24 dev $h1 + ip address add 2001:db8:1::2/64 dev $h1 + + ip route add 198.51.100.0/24 vrf vrf-h1 nexthop via 192.0.2.1 + ip route add 2001:db8:2::/64 vrf vrf-h1 nexthop via 2001:db8:1::1 +} + +h1_destroy() +{ + ip route del 2001:db8:2::/64 vrf vrf-h1 + ip route del 198.51.100.0/24 vrf vrf-h1 + + ip address del 2001:db8:1::2/64 dev $h1 + ip address del 192.0.2.2/24 dev $h1 + + ip link set dev $h1 down + vrf_destroy "vrf-h1" +} + +h2_create() +{ + vrf_create "vrf-h2" + ip link set dev $h2 master vrf-h2 + + ip link set dev vrf-h2 up + ip link set dev $h2 up + + ip address add 198.51.100.2/24 dev $h2 + ip address add 2001:db8:2::2/64 dev $h2 + + ip route add 192.0.2.0/24 vrf vrf-h2 nexthop via 198.51.100.1 + ip route add 2001:db8:1::/64 vrf vrf-h2 nexthop via 2001:db8:2::1 +} + +h2_destroy() +{ + ip route del 2001:db8:1::/64 vrf vrf-h2 + ip route del 192.0.2.0/24 vrf vrf-h2 + + ip address del 2001:db8:2::2/64 dev $h2 + ip address del 198.51.100.2/24 dev $h2 + + ip link set dev $h2 down + vrf_destroy "vrf-h2" +} + +router_create() +{ + ip link set dev $rp1 up + ip link set dev $rp2 up + + tc qdisc add dev $rp2 clsact + + ip address add 192.0.2.1/24 dev $rp1 + ip address add 2001:db8:1::1/64 dev $rp1 + + ip address add 198.51.100.1/24 dev $rp2 + ip address add 2001:db8:2::1/64 dev $rp2 +} + +router_destroy() +{ + ip address del 2001:db8:2::1/64 dev $rp2 + ip address del 198.51.100.1/24 dev $rp2 + + ip address del 2001:db8:1::1/64 dev $rp1 + ip address del 192.0.2.1/24 dev $rp1 + + tc qdisc del dev $rp2 clsact + + ip link set dev $rp2 down + ip link set dev $rp1 down +} + +routing_nh_obj() +{ + # Create the nexthops as AF_INET6, so that IPv4 and IPv6 routes could + # use them. + ip -6 nexthop add id 101 dev $rp1 + ip -6 nexthop add id 102 dev $rp2 + + ip route replace 192.0.2.0/24 nhid 101 + ip route replace 2001:db8:1::/64 nhid 101 + ip route replace 198.51.100.0/24 nhid 102 + ip route replace 2001:db8:2::/64 nhid 102 +} + +setup_prepare() +{ + h1=${NETIFS[p1]} + rp1=${NETIFS[p2]} + + rp2=${NETIFS[p3]} + h2=${NETIFS[p4]} + + rp1mac=$(mac_get $rp1) + + vrf_prepare + + h1_create + h2_create + + router_create + + forwarding_enable +} + +cleanup() +{ + pre_cleanup + + forwarding_restore + + router_destroy + + h2_destroy + h1_destroy + + vrf_cleanup +} + +ping_ipv4() +{ + ping_test $h1 198.51.100.2 +} + +ping_ipv6() +{ + ping6_test $h1 2001:db8:2::2 +} + +trap cleanup EXIT + +setup_prepare +setup_wait +routing_nh_obj + +tests_run + +exit $EXIT_STATUS -- 2.28.0 ^ permalink raw reply related [flat|nested] 10+ messages in thread
* [PATCH net-next 8/8] selftests: forwarding: Add multipath tunneling nexthop test 2020-11-19 13:08 [PATCH net-next 0/8] mlxsw: Add support for nexthop objects Ido Schimmel ` (6 preceding siblings ...) 2020-11-19 13:08 ` [PATCH net-next 7/8] selftests: forwarding: Add device-only nexthop test Ido Schimmel @ 2020-11-19 13:08 ` Ido Schimmel 2020-11-20 23:30 ` [PATCH net-next 0/8] mlxsw: Add support for nexthop objects patchwork-bot+netdevbpf 8 siblings, 0 replies; 10+ messages in thread From: Ido Schimmel @ 2020-11-19 13:08 UTC (permalink / raw) To: netdev; +Cc: davem, kuba, dsahern, jiri, mlxsw, Ido Schimmel From: Ido Schimmel <idosch@nvidia.com> Add a nexthop objects version of gre_multipath.sh. Unlike the original test, it also tests IPv6 overlay which is not possible with the legacy nexthop implementation. See commit 9a2ad3623868 ("selftests: forwarding: gre_multipath: Drop IPv6 tests") for more info. Signed-off-by: Ido Schimmel <idosch@nvidia.com> --- .../net/forwarding/gre_multipath_nh.sh | 356 ++++++++++++++++++ 1 file changed, 356 insertions(+) create mode 100755 tools/testing/selftests/net/forwarding/gre_multipath_nh.sh diff --git a/tools/testing/selftests/net/forwarding/gre_multipath_nh.sh b/tools/testing/selftests/net/forwarding/gre_multipath_nh.sh new file mode 100755 index 000000000000..d03aa2cab9fd --- /dev/null +++ b/tools/testing/selftests/net/forwarding/gre_multipath_nh.sh @@ -0,0 +1,356 @@ +#!/bin/bash +# SPDX-License-Identifier: GPL-2.0 + +# Test traffic distribution when a wECMP route forwards traffic to two GRE +# tunnels. +# +# +-------------------------+ +# | H1 | +# | $h1 + | +# | 192.0.2.1/28 | | +# | 2001:db8:1::1/64 | | +# +-------------------|-----+ +# | +# +-------------------|------------------------+ +# | SW1 | | +# | $ol1 + | +# | 192.0.2.2/28 | +# | 2001:db8:1::2/64 | +# | | +# | + g1a (gre) + g1b (gre) | +# | loc=192.0.2.65 loc=192.0.2.81 | +# | rem=192.0.2.66 --. rem=192.0.2.82 --. | +# | tos=inherit | tos=inherit | | +# | .------------------' | | +# | | .------------------' | +# | v v | +# | + $ul1.111 (vlan) + $ul1.222 (vlan) | +# | | 192.0.2.129/28 | 192.0.2.145/28 | +# | \ / | +# | \________________/ | +# | | | +# | + $ul1 | +# +------------|-------------------------------+ +# | +# +------------|-------------------------------+ +# | SW2 + $ul2 | +# | _______|________ | +# | / \ | +# | / \ | +# | + $ul2.111 (vlan) + $ul2.222 (vlan) | +# | ^ 192.0.2.130/28 ^ 192.0.2.146/28 | +# | | | | +# | | '------------------. | +# | '------------------. | | +# | + g2a (gre) | + g2b (gre) | | +# | loc=192.0.2.66 | loc=192.0.2.82 | | +# | rem=192.0.2.65 --' rem=192.0.2.81 --' | +# | tos=inherit tos=inherit | +# | | +# | $ol2 + | +# | 192.0.2.17/28 | | +# | 2001:db8:2::1/64 | | +# +-------------------|------------------------+ +# | +# +-------------------|-----+ +# | H2 | | +# | $h2 + | +# | 192.0.2.18/28 | +# | 2001:db8:2::2/64 | +# +-------------------------+ + +ALL_TESTS=" + ping_ipv4 + ping_ipv6 + multipath_ipv4 + multipath_ipv6 + multipath_ipv6_l4 +" + +NUM_NETIFS=6 +source lib.sh + +h1_create() +{ + simple_if_init $h1 192.0.2.1/28 2001:db8:1::1/64 + ip route add vrf v$h1 192.0.2.16/28 via 192.0.2.2 + ip route add vrf v$h1 2001:db8:2::/64 via 2001:db8:1::2 +} + +h1_destroy() +{ + ip route del vrf v$h1 2001:db8:2::/64 via 2001:db8:1::2 + ip route del vrf v$h1 192.0.2.16/28 via 192.0.2.2 + simple_if_fini $h1 192.0.2.1/28 +} + +sw1_create() +{ + simple_if_init $ol1 192.0.2.2/28 2001:db8:1::2/64 + __simple_if_init $ul1 v$ol1 + vlan_create $ul1 111 v$ol1 192.0.2.129/28 + vlan_create $ul1 222 v$ol1 192.0.2.145/28 + + tunnel_create g1a gre 192.0.2.65 192.0.2.66 tos inherit dev v$ol1 + __simple_if_init g1a v$ol1 192.0.2.65/32 + ip route add vrf v$ol1 192.0.2.66/32 via 192.0.2.130 + + tunnel_create g1b gre 192.0.2.81 192.0.2.82 tos inherit dev v$ol1 + __simple_if_init g1b v$ol1 192.0.2.81/32 + ip route add vrf v$ol1 192.0.2.82/32 via 192.0.2.146 + + ip -6 nexthop add id 101 dev g1a + ip -6 nexthop add id 102 dev g1b + ip nexthop add id 103 group 101/102 + + ip route add vrf v$ol1 192.0.2.16/28 nhid 103 + ip route add vrf v$ol1 2001:db8:2::/64 nhid 103 +} + +sw1_destroy() +{ + ip route del vrf v$ol1 2001:db8:2::/64 + ip route del vrf v$ol1 192.0.2.16/28 + + ip nexthop del id 103 + ip -6 nexthop del id 102 + ip -6 nexthop del id 101 + + ip route del vrf v$ol1 192.0.2.82/32 via 192.0.2.146 + __simple_if_fini g1b 192.0.2.81/32 + tunnel_destroy g1b + + ip route del vrf v$ol1 192.0.2.66/32 via 192.0.2.130 + __simple_if_fini g1a 192.0.2.65/32 + tunnel_destroy g1a + + vlan_destroy $ul1 222 + vlan_destroy $ul1 111 + __simple_if_fini $ul1 + simple_if_fini $ol1 192.0.2.2/28 2001:db8:1::2/64 +} + +sw2_create() +{ + simple_if_init $ol2 192.0.2.17/28 2001:db8:2::1/64 + __simple_if_init $ul2 v$ol2 + vlan_create $ul2 111 v$ol2 192.0.2.130/28 + vlan_create $ul2 222 v$ol2 192.0.2.146/28 + + tunnel_create g2a gre 192.0.2.66 192.0.2.65 tos inherit dev v$ol2 + __simple_if_init g2a v$ol2 192.0.2.66/32 + ip route add vrf v$ol2 192.0.2.65/32 via 192.0.2.129 + + tunnel_create g2b gre 192.0.2.82 192.0.2.81 tos inherit dev v$ol2 + __simple_if_init g2b v$ol2 192.0.2.82/32 + ip route add vrf v$ol2 192.0.2.81/32 via 192.0.2.145 + + ip -6 nexthop add id 201 dev g2a + ip -6 nexthop add id 202 dev g2b + ip nexthop add id 203 group 201/202 + + ip route add vrf v$ol2 192.0.2.0/28 nhid 203 + ip route add vrf v$ol2 2001:db8:1::/64 nhid 203 + + tc qdisc add dev $ul2 clsact + tc filter add dev $ul2 ingress pref 111 prot 802.1Q \ + flower vlan_id 111 action pass + tc filter add dev $ul2 ingress pref 222 prot 802.1Q \ + flower vlan_id 222 action pass +} + +sw2_destroy() +{ + tc qdisc del dev $ul2 clsact + + ip route del vrf v$ol2 2001:db8:1::/64 + ip route del vrf v$ol2 192.0.2.0/28 + + ip nexthop del id 203 + ip -6 nexthop del id 202 + ip -6 nexthop del id 201 + + ip route del vrf v$ol2 192.0.2.81/32 via 192.0.2.145 + __simple_if_fini g2b 192.0.2.82/32 + tunnel_destroy g2b + + ip route del vrf v$ol2 192.0.2.65/32 via 192.0.2.129 + __simple_if_fini g2a 192.0.2.66/32 + tunnel_destroy g2a + + vlan_destroy $ul2 222 + vlan_destroy $ul2 111 + __simple_if_fini $ul2 + simple_if_fini $ol2 192.0.2.17/28 2001:db8:2::1/64 +} + +h2_create() +{ + simple_if_init $h2 192.0.2.18/28 2001:db8:2::2/64 + ip route add vrf v$h2 192.0.2.0/28 via 192.0.2.17 + ip route add vrf v$h2 2001:db8:1::/64 via 2001:db8:2::1 +} + +h2_destroy() +{ + ip route del vrf v$h2 2001:db8:1::/64 via 2001:db8:2::1 + ip route del vrf v$h2 192.0.2.0/28 via 192.0.2.17 + simple_if_fini $h2 192.0.2.18/28 2001:db8:2::2/64 +} + +setup_prepare() +{ + h1=${NETIFS[p1]} + ol1=${NETIFS[p2]} + + ul1=${NETIFS[p3]} + ul2=${NETIFS[p4]} + + ol2=${NETIFS[p5]} + h2=${NETIFS[p6]} + + vrf_prepare + h1_create + sw1_create + sw2_create + h2_create + + forwarding_enable +} + +cleanup() +{ + pre_cleanup + + forwarding_restore + + h2_destroy + sw2_destroy + sw1_destroy + h1_destroy + vrf_cleanup +} + +multipath4_test() +{ + local what=$1; shift + local weight1=$1; shift + local weight2=$1; shift + + sysctl_set net.ipv4.fib_multipath_hash_policy 1 + ip nexthop replace id 103 group 101,$weight1/102,$weight2 + + local t0_111=$(tc_rule_stats_get $ul2 111 ingress) + local t0_222=$(tc_rule_stats_get $ul2 222 ingress) + + ip vrf exec v$h1 \ + $MZ $h1 -q -p 64 -A 192.0.2.1 -B 192.0.2.18 \ + -d 1msec -t udp "sp=1024,dp=0-32768" + + local t1_111=$(tc_rule_stats_get $ul2 111 ingress) + local t1_222=$(tc_rule_stats_get $ul2 222 ingress) + + local d111=$((t1_111 - t0_111)) + local d222=$((t1_222 - t0_222)) + multipath_eval "$what" $weight1 $weight2 $d111 $d222 + + ip nexthop replace id 103 group 101/102 + sysctl_restore net.ipv4.fib_multipath_hash_policy +} + +multipath6_test() +{ + local what=$1; shift + local weight1=$1; shift + local weight2=$1; shift + + sysctl_set net.ipv6.fib_multipath_hash_policy 0 + ip nexthop replace id 103 group 101,$weight1/102,$weight2 + + local t0_111=$(tc_rule_stats_get $ul2 111 ingress) + local t0_222=$(tc_rule_stats_get $ul2 222 ingress) + + # Generate 16384 echo requests, each with a random flow label. + for ((i=0; i < 16384; ++i)); do + ip vrf exec v$h1 $PING6 2001:db8:2::2 -F 0 -c 1 -q &> /dev/null + done + + local t1_111=$(tc_rule_stats_get $ul2 111 ingress) + local t1_222=$(tc_rule_stats_get $ul2 222 ingress) + + local d111=$((t1_111 - t0_111)) + local d222=$((t1_222 - t0_222)) + multipath_eval "$what" $weight1 $weight2 $d111 $d222 + + ip nexthop replace id 103 group 101/102 + sysctl_restore net.ipv6.fib_multipath_hash_policy +} + +multipath6_l4_test() +{ + local what=$1; shift + local weight1=$1; shift + local weight2=$1; shift + + sysctl_set net.ipv6.fib_multipath_hash_policy 1 + ip nexthop replace id 103 group 101,$weight1/102,$weight2 + + local t0_111=$(tc_rule_stats_get $ul2 111 ingress) + local t0_222=$(tc_rule_stats_get $ul2 222 ingress) + + ip vrf exec v$h1 \ + $MZ $h1 -6 -q -p 64 -A 2001:db8:1::1 -B 2001:db8:2::2 \ + -d 1msec -t udp "sp=1024,dp=0-32768" + + local t1_111=$(tc_rule_stats_get $ul2 111 ingress) + local t1_222=$(tc_rule_stats_get $ul2 222 ingress) + + local d111=$((t1_111 - t0_111)) + local d222=$((t1_222 - t0_222)) + multipath_eval "$what" $weight1 $weight2 $d111 $d222 + + ip nexthop replace id 103 group 101/102 + sysctl_restore net.ipv6.fib_multipath_hash_policy +} + +ping_ipv4() +{ + ping_test $h1 192.0.2.18 +} + +ping_ipv6() +{ + ping6_test $h1 2001:db8:2::2 +} + +multipath_ipv4() +{ + log_info "Running IPv4 multipath tests" + multipath4_test "ECMP" 1 1 + multipath4_test "Weighted MP 2:1" 2 1 + multipath4_test "Weighted MP 11:45" 11 45 +} + +multipath_ipv6() +{ + log_info "Running IPv6 multipath tests" + multipath6_test "ECMP" 1 1 + multipath6_test "Weighted MP 2:1" 2 1 + multipath6_test "Weighted MP 11:45" 11 45 +} + +multipath_ipv6_l4() +{ + log_info "Running IPv6 L4 hash multipath tests" + multipath6_l4_test "ECMP" 1 1 + multipath6_l4_test "Weighted MP 2:1" 2 1 + multipath6_l4_test "Weighted MP 11:45" 11 45 +} + +trap cleanup EXIT + +setup_prepare +setup_wait +tests_run + +exit $EXIT_STATUS -- 2.28.0 ^ permalink raw reply related [flat|nested] 10+ messages in thread
* Re: [PATCH net-next 0/8] mlxsw: Add support for nexthop objects 2020-11-19 13:08 [PATCH net-next 0/8] mlxsw: Add support for nexthop objects Ido Schimmel ` (7 preceding siblings ...) 2020-11-19 13:08 ` [PATCH net-next 8/8] selftests: forwarding: Add multipath tunneling " Ido Schimmel @ 2020-11-20 23:30 ` patchwork-bot+netdevbpf 8 siblings, 0 replies; 10+ messages in thread From: patchwork-bot+netdevbpf @ 2020-11-20 23:30 UTC (permalink / raw) To: Ido Schimmel; +Cc: netdev, davem, kuba, dsahern, jiri, mlxsw, idosch Hello: This series was applied to netdev/net-next.git (refs/heads/master): On Thu, 19 Nov 2020 15:08:40 +0200 you wrote: > From: Ido Schimmel <idosch@nvidia.com> > > This patch set adds support for nexthop objects in mlxsw. Nexthop > objects are treated as another front-end for programming nexthops, in > addition to the existing IPv4 and IPv6 front-ends. > > Patch #1 registers a listener to the nexthop notification chain and > parses the nexthop information into the existing mlxsw data structures > that are already used by the IPv4 and IPv6 front-ends. Blackhole > nexthops are currently rejected. Support will be added in a follow-up > patch set. > > [...] Here is the summary with links: - [net-next,1/8] mlxsw: spectrum_router: Add support for nexthop objects https://git.kernel.org/netdev/net-next/c/2a014b200bbd - [net-next,2/8] mlxsw: spectrum_router: Enable resolution of nexthop groups from nexthop objects https://git.kernel.org/netdev/net-next/c/c25db3a77f61 - [net-next,3/8] mlxsw: spectrum_router: Allow programming routes with nexthop objects https://git.kernel.org/netdev/net-next/c/cdd6cfc54c64 - [net-next,4/8] selftests: mlxsw: Add nexthop objects configuration tests https://git.kernel.org/netdev/net-next/c/20ac8f869053 - [net-next,5/8] selftests: forwarding: Do not configure nexthop objects twice https://git.kernel.org/netdev/net-next/c/ffb721515bf3 - [net-next,6/8] selftests: forwarding: Test IPv4 routes with IPv6 link-local nexthops https://git.kernel.org/netdev/net-next/c/3600f29ad139 - [net-next,7/8] selftests: forwarding: Add device-only nexthop test https://git.kernel.org/netdev/net-next/c/e96fa54bbd90 - [net-next,8/8] selftests: forwarding: Add multipath tunneling nexthop test https://git.kernel.org/netdev/net-next/c/e035146d6560 You are awesome, thank you! -- Deet-doot-dot, I am a bot. https://korg.docs.kernel.org/patchwork/pwbot.html ^ permalink raw reply [flat|nested] 10+ messages in thread
end of thread, other threads:[~2020-11-20 23:30 UTC | newest] Thread overview: 10+ messages (download: mbox.gz / follow: Atom feed) -- links below jump to the message on this page -- 2020-11-19 13:08 [PATCH net-next 0/8] mlxsw: Add support for nexthop objects Ido Schimmel 2020-11-19 13:08 ` [PATCH net-next 1/8] mlxsw: spectrum_router: " Ido Schimmel 2020-11-19 13:08 ` [PATCH net-next 2/8] mlxsw: spectrum_router: Enable resolution of nexthop groups from " Ido Schimmel 2020-11-19 13:08 ` [PATCH net-next 3/8] mlxsw: spectrum_router: Allow programming routes with " Ido Schimmel 2020-11-19 13:08 ` [PATCH net-next 4/8] selftests: mlxsw: Add nexthop objects configuration tests Ido Schimmel 2020-11-19 13:08 ` [PATCH net-next 5/8] selftests: forwarding: Do not configure nexthop objects twice Ido Schimmel 2020-11-19 13:08 ` [PATCH net-next 6/8] selftests: forwarding: Test IPv4 routes with IPv6 link-local nexthops Ido Schimmel 2020-11-19 13:08 ` [PATCH net-next 7/8] selftests: forwarding: Add device-only nexthop test Ido Schimmel 2020-11-19 13:08 ` [PATCH net-next 8/8] selftests: forwarding: Add multipath tunneling " Ido Schimmel 2020-11-20 23:30 ` [PATCH net-next 0/8] mlxsw: Add support for nexthop objects patchwork-bot+netdevbpf
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox; as well as URLs for NNTP newsgroup(s).