diff options
Diffstat (limited to 'net/switchdev')
| -rw-r--r-- | net/switchdev/Kconfig | 3 | ||||
| -rw-r--r-- | net/switchdev/Makefile | 3 | ||||
| -rw-r--r-- | net/switchdev/switchdev.c | 1392 |
3 files changed, 675 insertions, 723 deletions
diff --git a/net/switchdev/Kconfig b/net/switchdev/Kconfig index 651fa201a570..18a2d980e11d 100644 --- a/net/switchdev/Kconfig +++ b/net/switchdev/Kconfig @@ -1,3 +1,4 @@ +# SPDX-License-Identifier: GPL-2.0-only # # Configuration for Switch device support # @@ -5,7 +6,7 @@ config NET_SWITCHDEV bool "Switch (and switch-ish) device support" depends on INET - ---help--- + help This module provides glue between core networking code and device drivers in order to support hardware switch chips in very generic meaning of the word "switch". This include devices supporting L2/L3 but diff --git a/net/switchdev/Makefile b/net/switchdev/Makefile index 5ed63ed324d0..c5561d7f3a7c 100644 --- a/net/switchdev/Makefile +++ b/net/switchdev/Makefile @@ -1,5 +1,6 @@ +# SPDX-License-Identifier: GPL-2.0-only # # Makefile for the Switch device API # -obj-$(CONFIG_NET_SWITCHDEV) += switchdev.o +obj-y += switchdev.o diff --git a/net/switchdev/switchdev.c b/net/switchdev/switchdev.c index 25dc67ef9d37..4d5fbacef496 100644 --- a/net/switchdev/switchdev.c +++ b/net/switchdev/switchdev.c @@ -1,12 +1,8 @@ +// SPDX-License-Identifier: GPL-2.0-or-later /* * net/switchdev/switchdev.c - Switch device API * Copyright (c) 2014-2015 Jiri Pirko <jiri@resnulli.us> * Copyright (c) 2014-2015 Scott Feldman <sfeldma@gmail.com> - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License as published by - * the Free Software Foundation; either version 2 of the License, or - * (at your option) any later version. */ #include <linux/kernel.h> @@ -23,76 +19,33 @@ #include <linux/rtnetlink.h> #include <net/switchdev.h> -/** - * switchdev_trans_item_enqueue - Enqueue data item to transaction queue - * - * @trans: transaction - * @data: pointer to data being queued - * @destructor: data destructor - * @tritem: transaction item being queued - * - * Enqeueue data item to transaction queue. tritem is typically placed in - * cointainter pointed at by data pointer. Destructor is called on - * transaction abort and after successful commit phase in case - * the caller did not dequeue the item before. - */ -void switchdev_trans_item_enqueue(struct switchdev_trans *trans, - void *data, void (*destructor)(void const *), - struct switchdev_trans_item *tritem) -{ - tritem->data = data; - tritem->destructor = destructor; - list_add_tail(&tritem->list, &trans->item_list); -} -EXPORT_SYMBOL_GPL(switchdev_trans_item_enqueue); - -static struct switchdev_trans_item * -__switchdev_trans_item_dequeue(struct switchdev_trans *trans) +static bool switchdev_obj_eq(const struct switchdev_obj *a, + const struct switchdev_obj *b) { - struct switchdev_trans_item *tritem; - - if (list_empty(&trans->item_list)) - return NULL; - tritem = list_first_entry(&trans->item_list, - struct switchdev_trans_item, list); - list_del(&tritem->list); - return tritem; -} + const struct switchdev_obj_port_vlan *va, *vb; + const struct switchdev_obj_port_mdb *ma, *mb; -/** - * switchdev_trans_item_dequeue - Dequeue data item from transaction queue - * - * @trans: transaction - */ -void *switchdev_trans_item_dequeue(struct switchdev_trans *trans) -{ - struct switchdev_trans_item *tritem; - - tritem = __switchdev_trans_item_dequeue(trans); - BUG_ON(!tritem); - return tritem->data; -} -EXPORT_SYMBOL_GPL(switchdev_trans_item_dequeue); - -static void switchdev_trans_init(struct switchdev_trans *trans) -{ - INIT_LIST_HEAD(&trans->item_list); -} - -static void switchdev_trans_items_destroy(struct switchdev_trans *trans) -{ - struct switchdev_trans_item *tritem; + if (a->id != b->id || a->orig_dev != b->orig_dev) + return false; - while ((tritem = __switchdev_trans_item_dequeue(trans))) - tritem->destructor(tritem->data); -} + switch (a->id) { + case SWITCHDEV_OBJ_ID_PORT_VLAN: + va = SWITCHDEV_OBJ_PORT_VLAN(a); + vb = SWITCHDEV_OBJ_PORT_VLAN(b); + return va->flags == vb->flags && + va->vid == vb->vid && + va->changed == vb->changed; + case SWITCHDEV_OBJ_ID_PORT_MDB: + case SWITCHDEV_OBJ_ID_HOST_MDB: + ma = SWITCHDEV_OBJ_PORT_MDB(a); + mb = SWITCHDEV_OBJ_PORT_MDB(b); + return ma->vid == mb->vid && + ether_addr_equal(ma->addr, mb->addr); + default: + break; + } -static void switchdev_trans_items_warn_destroy(struct net_device *dev, - struct switchdev_trans *trans) -{ - WARN(!list_empty(&trans->item_list), "%s: transaction item queue is not empty.\n", - dev->name); - switchdev_trans_items_destroy(trans); + BUG(); } static LIST_HEAD(deferred); @@ -104,8 +57,9 @@ typedef void switchdev_deferred_func_t(struct net_device *dev, struct switchdev_deferred_item { struct list_head list; struct net_device *dev; + netdevice_tracker dev_tracker; switchdev_deferred_func_t *func; - unsigned long data[0]; + unsigned long data[]; }; static struct switchdev_deferred_item *switchdev_deferred_dequeue(void) @@ -139,7 +93,7 @@ void switchdev_deferred_process(void) while ((dfitem = switchdev_deferred_dequeue())) { dfitem->func(dfitem->dev, dfitem->data); - dev_put(dfitem->dev); + netdev_put(dfitem->dev, &dfitem->dev_tracker); kfree(dfitem); } } @@ -160,13 +114,13 @@ static int switchdev_deferred_enqueue(struct net_device *dev, { struct switchdev_deferred_item *dfitem; - dfitem = kmalloc(sizeof(*dfitem) + data_len, GFP_ATOMIC); + dfitem = kmalloc(struct_size(dfitem, data, data_len), GFP_ATOMIC); if (!dfitem) return -ENOMEM; dfitem->dev = dev; dfitem->func = func; memcpy(dfitem->data, data, data_len); - dev_hold(dev); + netdev_hold(dev, &dfitem->dev_tracker, GFP_ATOMIC); spin_lock_bh(&deferred_lock); list_add_tail(&dfitem->list, &deferred); spin_unlock_bh(&deferred_lock); @@ -174,124 +128,39 @@ static int switchdev_deferred_enqueue(struct net_device *dev, return 0; } -/** - * switchdev_port_attr_get - Get port attribute - * - * @dev: port device - * @attr: attribute to get - */ -int switchdev_port_attr_get(struct net_device *dev, struct switchdev_attr *attr) +static int switchdev_port_attr_notify(enum switchdev_notifier_type nt, + struct net_device *dev, + const struct switchdev_attr *attr, + struct netlink_ext_ack *extack) { - const struct switchdev_ops *ops = dev->switchdev_ops; - struct net_device *lower_dev; - struct list_head *iter; - struct switchdev_attr first = { - .id = SWITCHDEV_ATTR_ID_UNDEFINED - }; - int err = -EOPNOTSUPP; + int err; + int rc; - if (ops && ops->switchdev_port_attr_get) - return ops->switchdev_port_attr_get(dev, attr); + struct switchdev_notifier_port_attr_info attr_info = { + .attr = attr, + .handled = false, + }; - if (attr->flags & SWITCHDEV_F_NO_RECURSE) + rc = call_switchdev_blocking_notifiers(nt, dev, + &attr_info.info, extack); + err = notifier_to_errno(rc); + if (err) { + WARN_ON(!attr_info.handled); return err; - - /* Switch device port(s) may be stacked under - * bond/team/vlan dev, so recurse down to get attr on - * each port. Return -ENODATA if attr values don't - * compare across ports. - */ - - netdev_for_each_lower_dev(dev, lower_dev, iter) { - err = switchdev_port_attr_get(lower_dev, attr); - if (err) - break; - if (first.id == SWITCHDEV_ATTR_ID_UNDEFINED) - first = *attr; - else if (memcmp(&first, attr, sizeof(*attr))) - return -ENODATA; } - return err; -} -EXPORT_SYMBOL_GPL(switchdev_port_attr_get); - -static int __switchdev_port_attr_set(struct net_device *dev, - const struct switchdev_attr *attr, - struct switchdev_trans *trans) -{ - const struct switchdev_ops *ops = dev->switchdev_ops; - struct net_device *lower_dev; - struct list_head *iter; - int err = -EOPNOTSUPP; - - if (ops && ops->switchdev_port_attr_set) { - err = ops->switchdev_port_attr_set(dev, attr, trans); - goto done; - } - - if (attr->flags & SWITCHDEV_F_NO_RECURSE) - goto done; - - /* Switch device port(s) may be stacked under - * bond/team/vlan dev, so recurse down to set attr on - * each port. - */ - - netdev_for_each_lower_dev(dev, lower_dev, iter) { - err = __switchdev_port_attr_set(lower_dev, attr, trans); - if (err) - break; - } - -done: - if (err == -EOPNOTSUPP && attr->flags & SWITCHDEV_F_SKIP_EOPNOTSUPP) - err = 0; + if (!attr_info.handled) + return -EOPNOTSUPP; - return err; + return 0; } static int switchdev_port_attr_set_now(struct net_device *dev, - const struct switchdev_attr *attr) + const struct switchdev_attr *attr, + struct netlink_ext_ack *extack) { - struct switchdev_trans trans; - int err; - - switchdev_trans_init(&trans); - - /* Phase I: prepare for attr set. Driver/device should fail - * here if there are going to be issues in the commit phase, - * such as lack of resources or support. The driver/device - * should reserve resources needed for the commit phase here, - * but should not commit the attr. - */ - - trans.ph_prepare = true; - err = __switchdev_port_attr_set(dev, attr, &trans); - if (err) { - /* Prepare phase failed: abort the transaction. Any - * resources reserved in the prepare phase are - * released. - */ - - if (err != -EOPNOTSUPP) - switchdev_trans_items_destroy(&trans); - - return err; - } - - /* Phase II: commit attr set. This cannot fail as a fault - * of driver/device. If it does, it's a bug in the driver/device - * because the driver said everythings was OK in phase I. - */ - - trans.ph_prepare = false; - err = __switchdev_port_attr_set(dev, attr, &trans); - WARN(err, "%s: Commit of attribute (id=%d) failed.\n", - dev->name, attr->id); - switchdev_trans_items_warn_destroy(dev, &trans); - - return err; + return switchdev_port_attr_notify(SWITCHDEV_PORT_ATTR_SET, dev, attr, + extack); } static void switchdev_port_attr_set_deferred(struct net_device *dev, @@ -300,7 +169,7 @@ static void switchdev_port_attr_set_deferred(struct net_device *dev, const struct switchdev_attr *attr = data; int err; - err = switchdev_port_attr_set_now(dev, attr); + err = switchdev_port_attr_set_now(dev, attr, NULL); if (err && err != -EOPNOTSUPP) netdev_err(dev, "failed (err=%d) to set attribute (id=%d)\n", err, attr->id); @@ -320,21 +189,19 @@ static int switchdev_port_attr_set_defer(struct net_device *dev, * * @dev: port device * @attr: attribute to set - * - * Use a 2-phase prepare-commit transaction model to ensure - * system is not left in a partially updated state due to - * failure from driver/device. + * @extack: netlink extended ack, for error message propagation * * rtnl_lock must be held and must not be in atomic section, * in case SWITCHDEV_F_DEFER flag is not set. */ int switchdev_port_attr_set(struct net_device *dev, - const struct switchdev_attr *attr) + const struct switchdev_attr *attr, + struct netlink_ext_ack *extack) { if (attr->flags & SWITCHDEV_F_DEFER) return switchdev_port_attr_set_defer(dev, attr); ASSERT_RTNL(); - return switchdev_port_attr_set_now(dev, attr); + return switchdev_port_attr_set_now(dev, attr, extack); } EXPORT_SYMBOL_GPL(switchdev_port_attr_set); @@ -343,84 +210,131 @@ static size_t switchdev_obj_size(const struct switchdev_obj *obj) switch (obj->id) { case SWITCHDEV_OBJ_ID_PORT_VLAN: return sizeof(struct switchdev_obj_port_vlan); - case SWITCHDEV_OBJ_ID_PORT_FDB: - return sizeof(struct switchdev_obj_port_fdb); case SWITCHDEV_OBJ_ID_PORT_MDB: return sizeof(struct switchdev_obj_port_mdb); + case SWITCHDEV_OBJ_ID_HOST_MDB: + return sizeof(struct switchdev_obj_port_mdb); default: BUG(); } return 0; } -static int __switchdev_port_obj_add(struct net_device *dev, - const struct switchdev_obj *obj, - struct switchdev_trans *trans) -{ - const struct switchdev_ops *ops = dev->switchdev_ops; - struct net_device *lower_dev; - struct list_head *iter; - int err = -EOPNOTSUPP; - - if (ops && ops->switchdev_port_obj_add) - return ops->switchdev_port_obj_add(dev, obj, trans); - - /* Switch device port(s) may be stacked under - * bond/team/vlan dev, so recurse down to add object on - * each port. - */ - - netdev_for_each_lower_dev(dev, lower_dev, iter) { - err = __switchdev_port_obj_add(lower_dev, obj, trans); - if (err) - break; - } - - return err; -} - -static int switchdev_port_obj_add_now(struct net_device *dev, - const struct switchdev_obj *obj) +static int switchdev_port_obj_notify(enum switchdev_notifier_type nt, + struct net_device *dev, + const struct switchdev_obj *obj, + struct netlink_ext_ack *extack) { - struct switchdev_trans trans; + int rc; int err; - ASSERT_RTNL(); - - switchdev_trans_init(&trans); - - /* Phase I: prepare for obj add. Driver/device should fail - * here if there are going to be issues in the commit phase, - * such as lack of resources or support. The driver/device - * should reserve resources needed for the commit phase here, - * but should not commit the obj. - */ + struct switchdev_notifier_port_obj_info obj_info = { + .obj = obj, + .handled = false, + }; - trans.ph_prepare = true; - err = __switchdev_port_obj_add(dev, obj, &trans); + rc = call_switchdev_blocking_notifiers(nt, dev, &obj_info.info, extack); + err = notifier_to_errno(rc); if (err) { - /* Prepare phase failed: abort the transaction. Any - * resources reserved in the prepare phase are - * released. - */ - - if (err != -EOPNOTSUPP) - switchdev_trans_items_destroy(&trans); - + WARN_ON(!obj_info.handled); return err; } + if (!obj_info.handled) + return -EOPNOTSUPP; + return 0; +} - /* Phase II: commit obj add. This cannot fail as a fault - * of driver/device. If it does, it's a bug in the driver/device - * because the driver said everythings was OK in phase I. - */ +static void switchdev_obj_id_to_helpful_msg(struct net_device *dev, + enum switchdev_obj_id obj_id, + int err, bool add) +{ + const char *action = add ? "add" : "del"; + const char *reason = ""; + const char *problem; + const char *obj_str; + + switch (obj_id) { + case SWITCHDEV_OBJ_ID_UNDEFINED: + obj_str = "Undefined object"; + problem = "Attempted operation is undefined, indicating a possible programming\n" + "error.\n"; + break; + case SWITCHDEV_OBJ_ID_PORT_VLAN: + obj_str = "VLAN entry"; + problem = "Failure in VLAN settings on this port might disrupt network\n" + "segmentation or traffic isolation, affecting network partitioning.\n"; + break; + case SWITCHDEV_OBJ_ID_PORT_MDB: + obj_str = "Port Multicast Database entry"; + problem = "Failure in updating the port's Multicast Database could lead to\n" + "multicast forwarding issues.\n"; + break; + case SWITCHDEV_OBJ_ID_HOST_MDB: + obj_str = "Host Multicast Database entry"; + problem = "Failure in updating the host's Multicast Database may impact multicast\n" + "group memberships or traffic delivery, affecting multicast\n" + "communication.\n"; + break; + case SWITCHDEV_OBJ_ID_MRP: + obj_str = "Media Redundancy Protocol configuration for port"; + problem = "Failure to set MRP ring ID on this port prevents communication with\n" + "the specified redundancy ring, resulting in an inability to engage\n" + "in MRP-based network operations.\n"; + break; + case SWITCHDEV_OBJ_ID_RING_TEST_MRP: + obj_str = "MRP Test Frame Operations for port"; + problem = "Failure to generate/monitor MRP test frames may lead to inability to\n" + "assess the ring's operational integrity and fault response, hindering\n" + "proactive network management.\n"; + break; + case SWITCHDEV_OBJ_ID_RING_ROLE_MRP: + obj_str = "MRP Ring Role Configuration"; + problem = "Improper MRP ring role configuration may create conflicts in the ring,\n" + "disrupting communication for all participants, or isolate the local\n" + "system from the ring, hindering its ability to communicate with other\n" + "participants.\n"; + break; + case SWITCHDEV_OBJ_ID_RING_STATE_MRP: + obj_str = "MRP Ring State Configuration"; + problem = "Failure to correctly set the MRP ring state can result in network\n" + "loops or leave segments without communication. In a Closed state,\n" + "it maintains loop prevention by blocking one MRM port, while an Open\n" + "state activates in response to failures, changing port states to\n" + "preserve network connectivity.\n"; + break; + case SWITCHDEV_OBJ_ID_IN_TEST_MRP: + obj_str = "MRP_InTest Frame Generation Configuration"; + problem = "Failure in managing MRP_InTest frame generation can misjudge the\n" + "interconnection ring's state, leading to incorrect blocking or\n" + "unblocking of the I/C port. This misconfiguration might result\n" + "in unintended network loops or isolate critical network segments,\n" + "compromising network integrity and reliability.\n"; + break; + case SWITCHDEV_OBJ_ID_IN_ROLE_MRP: + obj_str = "Interconnection Ring Role Configuration"; + problem = "Failure in incorrect assignment of interconnection ring roles\n" + "(MIM/MIC) can impair the formation of the interconnection rings.\n"; + break; + case SWITCHDEV_OBJ_ID_IN_STATE_MRP: + obj_str = "Interconnection Ring State Configuration"; + problem = "Failure in updating the interconnection ring state can lead in\n" + "case of Open state to incorrect blocking or unblocking of the\n" + "I/C port, resulting in unintended network loops or isolation\n" + "of critical network\n"; + break; + default: + obj_str = "Unknown object"; + problem = "Indicating a possible programming error.\n"; + } - trans.ph_prepare = false; - err = __switchdev_port_obj_add(dev, obj, &trans); - WARN(err, "%s: Commit of object (id=%d) failed.\n", dev->name, obj->id); - switchdev_trans_items_warn_destroy(dev, &trans); + switch (err) { + case -ENOSPC: + reason = "Current HW/SW setup lacks sufficient resources.\n"; + break; + } - return err; + netdev_err(dev, "Failed to %s %s (object id=%d) with error: %pe (%d).\n%s%s\n", + action, obj_str, obj_id, ERR_PTR(err), err, problem, reason); } static void switchdev_port_obj_add_deferred(struct net_device *dev, @@ -429,10 +343,11 @@ static void switchdev_port_obj_add_deferred(struct net_device *dev, const struct switchdev_obj *obj = data; int err; - err = switchdev_port_obj_add_now(dev, obj); + ASSERT_RTNL(); + err = switchdev_port_obj_notify(SWITCHDEV_PORT_OBJ_ADD, + dev, obj, NULL); if (err && err != -EOPNOTSUPP) - netdev_err(dev, "failed (err=%d) to add object (id=%d)\n", - err, obj->id); + switchdev_obj_id_to_helpful_msg(dev, obj->id, err, true); if (obj->complete) obj->complete(dev, err, obj->complete_priv); } @@ -448,49 +363,29 @@ static int switchdev_port_obj_add_defer(struct net_device *dev, * switchdev_port_obj_add - Add port object * * @dev: port device - * @id: object ID * @obj: object to add - * - * Use a 2-phase prepare-commit transaction model to ensure - * system is not left in a partially updated state due to - * failure from driver/device. + * @extack: netlink extended ack * * rtnl_lock must be held and must not be in atomic section, * in case SWITCHDEV_F_DEFER flag is not set. */ int switchdev_port_obj_add(struct net_device *dev, - const struct switchdev_obj *obj) + const struct switchdev_obj *obj, + struct netlink_ext_ack *extack) { if (obj->flags & SWITCHDEV_F_DEFER) return switchdev_port_obj_add_defer(dev, obj); ASSERT_RTNL(); - return switchdev_port_obj_add_now(dev, obj); + return switchdev_port_obj_notify(SWITCHDEV_PORT_OBJ_ADD, + dev, obj, extack); } EXPORT_SYMBOL_GPL(switchdev_port_obj_add); static int switchdev_port_obj_del_now(struct net_device *dev, const struct switchdev_obj *obj) { - const struct switchdev_ops *ops = dev->switchdev_ops; - struct net_device *lower_dev; - struct list_head *iter; - int err = -EOPNOTSUPP; - - if (ops && ops->switchdev_port_obj_del) - return ops->switchdev_port_obj_del(dev, obj); - - /* Switch device port(s) may be stacked under - * bond/team/vlan dev, so recurse down to delete object on - * each port. - */ - - netdev_for_each_lower_dev(dev, lower_dev, iter) { - err = switchdev_port_obj_del_now(lower_dev, obj); - if (err) - break; - } - - return err; + return switchdev_port_obj_notify(SWITCHDEV_PORT_OBJ_DEL, + dev, obj, NULL); } static void switchdev_port_obj_del_deferred(struct net_device *dev, @@ -501,8 +396,7 @@ static void switchdev_port_obj_del_deferred(struct net_device *dev, err = switchdev_port_obj_del_now(dev, obj); if (err && err != -EOPNOTSUPP) - netdev_err(dev, "failed (err=%d) to del object (id=%d)\n", - err, obj->id); + switchdev_obj_id_to_helpful_msg(dev, obj->id, err, false); if (obj->complete) obj->complete(dev, err, obj->complete_priv); } @@ -518,7 +412,6 @@ static int switchdev_port_obj_del_defer(struct net_device *dev, * switchdev_port_obj_del - Delete port object * * @dev: port device - * @id: object ID * @obj: object to delete * * rtnl_lock must be held and must not be in atomic section, @@ -535,43 +428,51 @@ int switchdev_port_obj_del(struct net_device *dev, EXPORT_SYMBOL_GPL(switchdev_port_obj_del); /** - * switchdev_port_obj_dump - Dump port objects + * switchdev_port_obj_act_is_deferred - Is object action pending? * * @dev: port device - * @id: object ID - * @obj: object to dump - * @cb: function to call with a filled object + * @nt: type of action; add or delete + * @obj: object to test + * + * Returns true if a deferred item is pending, which is + * equivalent to the action @nt on an object @obj. * * rtnl_lock must be held. */ -int switchdev_port_obj_dump(struct net_device *dev, struct switchdev_obj *obj, - switchdev_obj_dump_cb_t *cb) +bool switchdev_port_obj_act_is_deferred(struct net_device *dev, + enum switchdev_notifier_type nt, + const struct switchdev_obj *obj) { - const struct switchdev_ops *ops = dev->switchdev_ops; - struct net_device *lower_dev; - struct list_head *iter; - int err = -EOPNOTSUPP; + struct switchdev_deferred_item *dfitem; + bool found = false; ASSERT_RTNL(); - if (ops && ops->switchdev_port_obj_dump) - return ops->switchdev_port_obj_dump(dev, obj, cb); + spin_lock_bh(&deferred_lock); - /* Switch device port(s) may be stacked under - * bond/team/vlan dev, so recurse down to dump objects on - * first port at bottom of stack. - */ + list_for_each_entry(dfitem, &deferred, list) { + if (dfitem->dev != dev) + continue; - netdev_for_each_lower_dev(dev, lower_dev, iter) { - err = switchdev_port_obj_dump(lower_dev, obj, cb); - break; + if ((dfitem->func == switchdev_port_obj_add_deferred && + nt == SWITCHDEV_PORT_OBJ_ADD) || + (dfitem->func == switchdev_port_obj_del_deferred && + nt == SWITCHDEV_PORT_OBJ_DEL)) { + if (switchdev_obj_eq((const void *)dfitem->data, obj)) { + found = true; + break; + } + } } - return err; + spin_unlock_bh(&deferred_lock); + + return found; } -EXPORT_SYMBOL_GPL(switchdev_port_obj_dump); +EXPORT_SYMBOL_GPL(switchdev_port_obj_act_is_deferred); static ATOMIC_NOTIFIER_HEAD(switchdev_notif_chain); +static RAW_NOTIFIER_HEAD(switchdev_blocking_notif_chain); /** * register_switchdev_notifier - Register notifier @@ -602,513 +503,562 @@ EXPORT_SYMBOL_GPL(unregister_switchdev_notifier); * @val: value passed unmodified to notifier function * @dev: port device * @info: notifier information data - * + * @extack: netlink extended ack * Call all network notifier blocks. */ int call_switchdev_notifiers(unsigned long val, struct net_device *dev, - struct switchdev_notifier_info *info) + struct switchdev_notifier_info *info, + struct netlink_ext_ack *extack) { info->dev = dev; + info->extack = extack; return atomic_notifier_call_chain(&switchdev_notif_chain, val, info); } EXPORT_SYMBOL_GPL(call_switchdev_notifiers); -struct switchdev_vlan_dump { - struct switchdev_obj_port_vlan vlan; - struct sk_buff *skb; - u32 filter_mask; - u16 flags; - u16 begin; - u16 end; -}; - -static int switchdev_port_vlan_dump_put(struct switchdev_vlan_dump *dump) +int register_switchdev_blocking_notifier(struct notifier_block *nb) { - struct bridge_vlan_info vinfo; - - vinfo.flags = dump->flags; + struct raw_notifier_head *chain = &switchdev_blocking_notif_chain; + int err; - if (dump->begin == 0 && dump->end == 0) { - return 0; - } else if (dump->begin == dump->end) { - vinfo.vid = dump->begin; - if (nla_put(dump->skb, IFLA_BRIDGE_VLAN_INFO, - sizeof(vinfo), &vinfo)) - return -EMSGSIZE; - } else { - vinfo.vid = dump->begin; - vinfo.flags |= BRIDGE_VLAN_INFO_RANGE_BEGIN; - if (nla_put(dump->skb, IFLA_BRIDGE_VLAN_INFO, - sizeof(vinfo), &vinfo)) - return -EMSGSIZE; - vinfo.vid = dump->end; - vinfo.flags &= ~BRIDGE_VLAN_INFO_RANGE_BEGIN; - vinfo.flags |= BRIDGE_VLAN_INFO_RANGE_END; - if (nla_put(dump->skb, IFLA_BRIDGE_VLAN_INFO, - sizeof(vinfo), &vinfo)) - return -EMSGSIZE; - } + rtnl_lock(); + err = raw_notifier_chain_register(chain, nb); + rtnl_unlock(); - return 0; + return err; } +EXPORT_SYMBOL_GPL(register_switchdev_blocking_notifier); -static int switchdev_port_vlan_dump_cb(struct switchdev_obj *obj) +int unregister_switchdev_blocking_notifier(struct notifier_block *nb) { - struct switchdev_obj_port_vlan *vlan = SWITCHDEV_OBJ_PORT_VLAN(obj); - struct switchdev_vlan_dump *dump = - container_of(vlan, struct switchdev_vlan_dump, vlan); - int err = 0; - - if (vlan->vid_begin > vlan->vid_end) - return -EINVAL; - - if (dump->filter_mask & RTEXT_FILTER_BRVLAN) { - dump->flags = vlan->flags; - for (dump->begin = dump->end = vlan->vid_begin; - dump->begin <= vlan->vid_end; - dump->begin++, dump->end++) { - err = switchdev_port_vlan_dump_put(dump); - if (err) - return err; - } - } else if (dump->filter_mask & RTEXT_FILTER_BRVLAN_COMPRESSED) { - if (dump->begin > vlan->vid_begin && - dump->begin >= vlan->vid_end) { - if ((dump->begin - 1) == vlan->vid_end && - dump->flags == vlan->flags) { - /* prepend */ - dump->begin = vlan->vid_begin; - } else { - err = switchdev_port_vlan_dump_put(dump); - dump->flags = vlan->flags; - dump->begin = vlan->vid_begin; - dump->end = vlan->vid_end; - } - } else if (dump->end <= vlan->vid_begin && - dump->end < vlan->vid_end) { - if ((dump->end + 1) == vlan->vid_begin && - dump->flags == vlan->flags) { - /* append */ - dump->end = vlan->vid_end; - } else { - err = switchdev_port_vlan_dump_put(dump); - dump->flags = vlan->flags; - dump->begin = vlan->vid_begin; - dump->end = vlan->vid_end; - } - } else { - err = -EINVAL; - } - } + struct raw_notifier_head *chain = &switchdev_blocking_notif_chain; + int err; + + rtnl_lock(); + err = raw_notifier_chain_unregister(chain, nb); + rtnl_unlock(); return err; } +EXPORT_SYMBOL_GPL(unregister_switchdev_blocking_notifier); -static int switchdev_port_vlan_fill(struct sk_buff *skb, struct net_device *dev, - u32 filter_mask) +int call_switchdev_blocking_notifiers(unsigned long val, struct net_device *dev, + struct switchdev_notifier_info *info, + struct netlink_ext_ack *extack) { - struct switchdev_vlan_dump dump = { - .vlan.obj.orig_dev = dev, - .vlan.obj.id = SWITCHDEV_OBJ_ID_PORT_VLAN, - .skb = skb, - .filter_mask = filter_mask, - }; - int err = 0; - - if ((filter_mask & RTEXT_FILTER_BRVLAN) || - (filter_mask & RTEXT_FILTER_BRVLAN_COMPRESSED)) { - err = switchdev_port_obj_dump(dev, &dump.vlan.obj, - switchdev_port_vlan_dump_cb); - if (err) - goto err_out; - if (filter_mask & RTEXT_FILTER_BRVLAN_COMPRESSED) - /* last one */ - err = switchdev_port_vlan_dump_put(&dump); + ASSERT_RTNL(); + info->dev = dev; + info->extack = extack; + return raw_notifier_call_chain(&switchdev_blocking_notif_chain, + val, info); +} +EXPORT_SYMBOL_GPL(call_switchdev_blocking_notifiers); + +struct switchdev_nested_priv { + bool (*check_cb)(const struct net_device *dev); + bool (*foreign_dev_check_cb)(const struct net_device *dev, + const struct net_device *foreign_dev); + const struct net_device *dev; + struct net_device *lower_dev; +}; + +static int switchdev_lower_dev_walk(struct net_device *lower_dev, + struct netdev_nested_priv *priv) +{ + struct switchdev_nested_priv *switchdev_priv = priv->data; + bool (*foreign_dev_check_cb)(const struct net_device *dev, + const struct net_device *foreign_dev); + bool (*check_cb)(const struct net_device *dev); + const struct net_device *dev; + + check_cb = switchdev_priv->check_cb; + foreign_dev_check_cb = switchdev_priv->foreign_dev_check_cb; + dev = switchdev_priv->dev; + + if (check_cb(lower_dev) && !foreign_dev_check_cb(lower_dev, dev)) { + switchdev_priv->lower_dev = lower_dev; + return 1; } -err_out: - return err == -EOPNOTSUPP ? 0 : err; + return 0; } -/** - * switchdev_port_bridge_getlink - Get bridge port attributes - * - * @dev: port device - * - * Called for SELF on rtnl_bridge_getlink to get bridge port - * attributes. - */ -int switchdev_port_bridge_getlink(struct sk_buff *skb, u32 pid, u32 seq, - struct net_device *dev, u32 filter_mask, - int nlflags) +static struct net_device * +switchdev_lower_dev_find_rcu(struct net_device *dev, + bool (*check_cb)(const struct net_device *dev), + bool (*foreign_dev_check_cb)(const struct net_device *dev, + const struct net_device *foreign_dev)) { - struct switchdev_attr attr = { - .orig_dev = dev, - .id = SWITCHDEV_ATTR_ID_PORT_BRIDGE_FLAGS, + struct switchdev_nested_priv switchdev_priv = { + .check_cb = check_cb, + .foreign_dev_check_cb = foreign_dev_check_cb, + .dev = dev, + .lower_dev = NULL, + }; + struct netdev_nested_priv priv = { + .data = &switchdev_priv, }; - u16 mode = BRIDGE_MODE_UNDEF; - u32 mask = BR_LEARNING | BR_LEARNING_SYNC | BR_FLOOD; - int err; - if (!netif_is_bridge_port(dev)) - return -EOPNOTSUPP; + netdev_walk_all_lower_dev_rcu(dev, switchdev_lower_dev_walk, &priv); - err = switchdev_port_attr_get(dev, &attr); - if (err && err != -EOPNOTSUPP) - return err; - - return ndo_dflt_bridge_getlink(skb, pid, seq, dev, mode, - attr.u.brport_flags, mask, nlflags, - filter_mask, switchdev_port_vlan_fill); + return switchdev_priv.lower_dev; } -EXPORT_SYMBOL_GPL(switchdev_port_bridge_getlink); -static int switchdev_port_br_setflag(struct net_device *dev, - struct nlattr *nlattr, - unsigned long brport_flag) +static struct net_device * +switchdev_lower_dev_find(struct net_device *dev, + bool (*check_cb)(const struct net_device *dev), + bool (*foreign_dev_check_cb)(const struct net_device *dev, + const struct net_device *foreign_dev)) { - struct switchdev_attr attr = { - .orig_dev = dev, - .id = SWITCHDEV_ATTR_ID_PORT_BRIDGE_FLAGS, + struct switchdev_nested_priv switchdev_priv = { + .check_cb = check_cb, + .foreign_dev_check_cb = foreign_dev_check_cb, + .dev = dev, + .lower_dev = NULL, + }; + struct netdev_nested_priv priv = { + .data = &switchdev_priv, }; - u8 flag = nla_get_u8(nlattr); - int err; - - err = switchdev_port_attr_get(dev, &attr); - if (err) - return err; - if (flag) - attr.u.brport_flags |= brport_flag; - else - attr.u.brport_flags &= ~brport_flag; + netdev_walk_all_lower_dev(dev, switchdev_lower_dev_walk, &priv); - return switchdev_port_attr_set(dev, &attr); + return switchdev_priv.lower_dev; } -static const struct nla_policy -switchdev_port_bridge_policy[IFLA_BRPORT_MAX + 1] = { - [IFLA_BRPORT_STATE] = { .type = NLA_U8 }, - [IFLA_BRPORT_COST] = { .type = NLA_U32 }, - [IFLA_BRPORT_PRIORITY] = { .type = NLA_U16 }, - [IFLA_BRPORT_MODE] = { .type = NLA_U8 }, - [IFLA_BRPORT_GUARD] = { .type = NLA_U8 }, - [IFLA_BRPORT_PROTECT] = { .type = NLA_U8 }, - [IFLA_BRPORT_FAST_LEAVE] = { .type = NLA_U8 }, - [IFLA_BRPORT_LEARNING] = { .type = NLA_U8 }, - [IFLA_BRPORT_LEARNING_SYNC] = { .type = NLA_U8 }, - [IFLA_BRPORT_UNICAST_FLOOD] = { .type = NLA_U8 }, -}; - -static int switchdev_port_br_setlink_protinfo(struct net_device *dev, - struct nlattr *protinfo) +static int __switchdev_handle_fdb_event_to_device(struct net_device *dev, + struct net_device *orig_dev, unsigned long event, + const struct switchdev_notifier_fdb_info *fdb_info, + bool (*check_cb)(const struct net_device *dev), + bool (*foreign_dev_check_cb)(const struct net_device *dev, + const struct net_device *foreign_dev), + int (*mod_cb)(struct net_device *dev, struct net_device *orig_dev, + unsigned long event, const void *ctx, + const struct switchdev_notifier_fdb_info *fdb_info)) { - struct nlattr *attr; - int rem; - int err; + const struct switchdev_notifier_info *info = &fdb_info->info; + struct net_device *br, *lower_dev, *switchdev; + struct list_head *iter; + int err = -EOPNOTSUPP; - err = nla_validate_nested(protinfo, IFLA_BRPORT_MAX, - switchdev_port_bridge_policy, NULL); - if (err) - return err; + if (check_cb(dev)) + return mod_cb(dev, orig_dev, event, info->ctx, fdb_info); - nla_for_each_nested(attr, protinfo, rem) { - switch (nla_type(attr)) { - case IFLA_BRPORT_LEARNING: - err = switchdev_port_br_setflag(dev, attr, - BR_LEARNING); - break; - case IFLA_BRPORT_LEARNING_SYNC: - err = switchdev_port_br_setflag(dev, attr, - BR_LEARNING_SYNC); - break; - case IFLA_BRPORT_UNICAST_FLOOD: - err = switchdev_port_br_setflag(dev, attr, BR_FLOOD); - break; - default: - err = -EOPNOTSUPP; - break; - } - if (err) + /* Recurse through lower interfaces in case the FDB entry is pointing + * towards a bridge or a LAG device. + */ + netdev_for_each_lower_dev(dev, lower_dev, iter) { + /* Do not propagate FDB entries across bridges */ + if (netif_is_bridge_master(lower_dev)) + continue; + + /* Bridge ports might be either us, or LAG interfaces + * that we offload. + */ + if (!check_cb(lower_dev) && + !switchdev_lower_dev_find_rcu(lower_dev, check_cb, + foreign_dev_check_cb)) + continue; + + err = __switchdev_handle_fdb_event_to_device(lower_dev, orig_dev, + event, fdb_info, check_cb, + foreign_dev_check_cb, + mod_cb); + if (err && err != -EOPNOTSUPP) return err; } - return 0; + /* Event is neither on a bridge nor a LAG. Check whether it is on an + * interface that is in a bridge with us. + */ + br = netdev_master_upper_dev_get_rcu(dev); + if (!br || !netif_is_bridge_master(br)) + return 0; + + switchdev = switchdev_lower_dev_find_rcu(br, check_cb, foreign_dev_check_cb); + if (!switchdev) + return 0; + + if (!foreign_dev_check_cb(switchdev, dev)) + return err; + + return __switchdev_handle_fdb_event_to_device(br, orig_dev, event, fdb_info, + check_cb, foreign_dev_check_cb, + mod_cb); } -static int switchdev_port_br_afspec(struct net_device *dev, - struct nlattr *afspec, - int (*f)(struct net_device *dev, - const struct switchdev_obj *obj)) +int switchdev_handle_fdb_event_to_device(struct net_device *dev, unsigned long event, + const struct switchdev_notifier_fdb_info *fdb_info, + bool (*check_cb)(const struct net_device *dev), + bool (*foreign_dev_check_cb)(const struct net_device *dev, + const struct net_device *foreign_dev), + int (*mod_cb)(struct net_device *dev, struct net_device *orig_dev, + unsigned long event, const void *ctx, + const struct switchdev_notifier_fdb_info *fdb_info)) { - struct nlattr *attr; - struct bridge_vlan_info *vinfo; - struct switchdev_obj_port_vlan vlan = { - .obj.orig_dev = dev, - .obj.id = SWITCHDEV_OBJ_ID_PORT_VLAN, - }; - int rem; int err; - nla_for_each_nested(attr, afspec, rem) { - if (nla_type(attr) != IFLA_BRIDGE_VLAN_INFO) - continue; - if (nla_len(attr) != sizeof(struct bridge_vlan_info)) - return -EINVAL; - vinfo = nla_data(attr); - if (!vinfo->vid || vinfo->vid >= VLAN_VID_MASK) - return -EINVAL; - vlan.flags = vinfo->flags; - if (vinfo->flags & BRIDGE_VLAN_INFO_RANGE_BEGIN) { - if (vlan.vid_begin) - return -EINVAL; - vlan.vid_begin = vinfo->vid; - /* don't allow range of pvids */ - if (vlan.flags & BRIDGE_VLAN_INFO_PVID) - return -EINVAL; - } else if (vinfo->flags & BRIDGE_VLAN_INFO_RANGE_END) { - if (!vlan.vid_begin) - return -EINVAL; - vlan.vid_end = vinfo->vid; - if (vlan.vid_end <= vlan.vid_begin) - return -EINVAL; - err = f(dev, &vlan.obj); - if (err) - return err; - vlan.vid_begin = 0; - } else { - if (vlan.vid_begin) - return -EINVAL; - vlan.vid_begin = vinfo->vid; - vlan.vid_end = vinfo->vid; - err = f(dev, &vlan.obj); - if (err) - return err; - vlan.vid_begin = 0; - } - } + err = __switchdev_handle_fdb_event_to_device(dev, dev, event, fdb_info, + check_cb, foreign_dev_check_cb, + mod_cb); + if (err == -EOPNOTSUPP) + err = 0; - return 0; + return err; } - -/** - * switchdev_port_bridge_setlink - Set bridge port attributes - * - * @dev: port device - * @nlh: netlink header - * @flags: netlink flags - * - * Called for SELF on rtnl_bridge_setlink to set bridge port - * attributes. - */ -int switchdev_port_bridge_setlink(struct net_device *dev, - struct nlmsghdr *nlh, u16 flags) +EXPORT_SYMBOL_GPL(switchdev_handle_fdb_event_to_device); + +static int __switchdev_handle_port_obj_add(struct net_device *dev, + struct switchdev_notifier_port_obj_info *port_obj_info, + bool (*check_cb)(const struct net_device *dev), + bool (*foreign_dev_check_cb)(const struct net_device *dev, + const struct net_device *foreign_dev), + int (*add_cb)(struct net_device *dev, const void *ctx, + const struct switchdev_obj *obj, + struct netlink_ext_ack *extack)) { - struct nlattr *protinfo; - struct nlattr *afspec; - int err = 0; + struct switchdev_notifier_info *info = &port_obj_info->info; + struct net_device *br, *lower_dev, *switchdev; + struct netlink_ext_ack *extack; + struct list_head *iter; + int err = -EOPNOTSUPP; - if (!netif_is_bridge_port(dev)) - return -EOPNOTSUPP; + extack = switchdev_notifier_info_to_extack(info); - protinfo = nlmsg_find_attr(nlh, sizeof(struct ifinfomsg), - IFLA_PROTINFO); - if (protinfo) { - err = switchdev_port_br_setlink_protinfo(dev, protinfo); - if (err) + if (check_cb(dev)) { + err = add_cb(dev, info->ctx, port_obj_info->obj, extack); + if (err != -EOPNOTSUPP) + port_obj_info->handled = true; + return err; + } + + /* Switch ports might be stacked under e.g. a LAG. Ignore the + * unsupported devices, another driver might be able to handle them. But + * propagate to the callers any hard errors. + * + * If the driver does its own bookkeeping of stacked ports, it's not + * necessary to go through this helper. + */ + netdev_for_each_lower_dev(dev, lower_dev, iter) { + if (netif_is_bridge_master(lower_dev)) + continue; + + /* When searching for switchdev interfaces that are neighbors + * of foreign ones, and @dev is a bridge, do not recurse on the + * foreign interface again, it was already visited. + */ + if (foreign_dev_check_cb && !check_cb(lower_dev) && + !switchdev_lower_dev_find(lower_dev, check_cb, foreign_dev_check_cb)) + continue; + + err = __switchdev_handle_port_obj_add(lower_dev, port_obj_info, + check_cb, foreign_dev_check_cb, + add_cb); + if (err && err != -EOPNOTSUPP) return err; } - afspec = nlmsg_find_attr(nlh, sizeof(struct ifinfomsg), - IFLA_AF_SPEC); - if (afspec) - err = switchdev_port_br_afspec(dev, afspec, - switchdev_port_obj_add); + /* Event is neither on a bridge nor a LAG. Check whether it is on an + * interface that is in a bridge with us. + */ + if (!foreign_dev_check_cb) + return err; + + br = netdev_master_upper_dev_get(dev); + if (!br || !netif_is_bridge_master(br)) + return err; + + switchdev = switchdev_lower_dev_find(br, check_cb, foreign_dev_check_cb); + if (!switchdev) + return err; + + if (!foreign_dev_check_cb(switchdev, dev)) + return err; + + return __switchdev_handle_port_obj_add(br, port_obj_info, check_cb, + foreign_dev_check_cb, add_cb); +} +/* Pass through a port object addition, if @dev passes @check_cb, or replicate + * it towards all lower interfaces of @dev that pass @check_cb, if @dev is a + * bridge or a LAG. + */ +int switchdev_handle_port_obj_add(struct net_device *dev, + struct switchdev_notifier_port_obj_info *port_obj_info, + bool (*check_cb)(const struct net_device *dev), + int (*add_cb)(struct net_device *dev, const void *ctx, + const struct switchdev_obj *obj, + struct netlink_ext_ack *extack)) +{ + int err; + + err = __switchdev_handle_port_obj_add(dev, port_obj_info, check_cb, + NULL, add_cb); + if (err == -EOPNOTSUPP) + err = 0; return err; } -EXPORT_SYMBOL_GPL(switchdev_port_bridge_setlink); +EXPORT_SYMBOL_GPL(switchdev_handle_port_obj_add); -/** - * switchdev_port_bridge_dellink - Set bridge port attributes - * - * @dev: port device - * @nlh: netlink header - * @flags: netlink flags - * - * Called for SELF on rtnl_bridge_dellink to set bridge port - * attributes. +/* Same as switchdev_handle_port_obj_add(), except if object is notified on a + * @dev that passes @foreign_dev_check_cb, it is replicated towards all devices + * that pass @check_cb and are in the same bridge as @dev. */ -int switchdev_port_bridge_dellink(struct net_device *dev, - struct nlmsghdr *nlh, u16 flags) +int switchdev_handle_port_obj_add_foreign(struct net_device *dev, + struct switchdev_notifier_port_obj_info *port_obj_info, + bool (*check_cb)(const struct net_device *dev), + bool (*foreign_dev_check_cb)(const struct net_device *dev, + const struct net_device *foreign_dev), + int (*add_cb)(struct net_device *dev, const void *ctx, + const struct switchdev_obj *obj, + struct netlink_ext_ack *extack)) { - struct nlattr *afspec; + int err; - if (!netif_is_bridge_port(dev)) - return -EOPNOTSUPP; + err = __switchdev_handle_port_obj_add(dev, port_obj_info, check_cb, + foreign_dev_check_cb, add_cb); + if (err == -EOPNOTSUPP) + err = 0; + return err; +} +EXPORT_SYMBOL_GPL(switchdev_handle_port_obj_add_foreign); + +static int __switchdev_handle_port_obj_del(struct net_device *dev, + struct switchdev_notifier_port_obj_info *port_obj_info, + bool (*check_cb)(const struct net_device *dev), + bool (*foreign_dev_check_cb)(const struct net_device *dev, + const struct net_device *foreign_dev), + int (*del_cb)(struct net_device *dev, const void *ctx, + const struct switchdev_obj *obj)) +{ + struct switchdev_notifier_info *info = &port_obj_info->info; + struct net_device *br, *lower_dev, *switchdev; + struct list_head *iter; + int err = -EOPNOTSUPP; - afspec = nlmsg_find_attr(nlh, sizeof(struct ifinfomsg), - IFLA_AF_SPEC); - if (afspec) - return switchdev_port_br_afspec(dev, afspec, - switchdev_port_obj_del); + if (check_cb(dev)) { + err = del_cb(dev, info->ctx, port_obj_info->obj); + if (err != -EOPNOTSUPP) + port_obj_info->handled = true; + return err; + } - return 0; + /* Switch ports might be stacked under e.g. a LAG. Ignore the + * unsupported devices, another driver might be able to handle them. But + * propagate to the callers any hard errors. + * + * If the driver does its own bookkeeping of stacked ports, it's not + * necessary to go through this helper. + */ + netdev_for_each_lower_dev(dev, lower_dev, iter) { + if (netif_is_bridge_master(lower_dev)) + continue; + + /* When searching for switchdev interfaces that are neighbors + * of foreign ones, and @dev is a bridge, do not recurse on the + * foreign interface again, it was already visited. + */ + if (foreign_dev_check_cb && !check_cb(lower_dev) && + !switchdev_lower_dev_find(lower_dev, check_cb, foreign_dev_check_cb)) + continue; + + err = __switchdev_handle_port_obj_del(lower_dev, port_obj_info, + check_cb, foreign_dev_check_cb, + del_cb); + if (err && err != -EOPNOTSUPP) + return err; + } + + /* Event is neither on a bridge nor a LAG. Check whether it is on an + * interface that is in a bridge with us. + */ + if (!foreign_dev_check_cb) + return err; + + br = netdev_master_upper_dev_get(dev); + if (!br || !netif_is_bridge_master(br)) + return err; + + switchdev = switchdev_lower_dev_find(br, check_cb, foreign_dev_check_cb); + if (!switchdev) + return err; + + if (!foreign_dev_check_cb(switchdev, dev)) + return err; + + return __switchdev_handle_port_obj_del(br, port_obj_info, check_cb, + foreign_dev_check_cb, del_cb); } -EXPORT_SYMBOL_GPL(switchdev_port_bridge_dellink); -/** - * switchdev_port_fdb_add - Add FDB (MAC/VLAN) entry to port - * - * @ndmsg: netlink hdr - * @nlattr: netlink attributes - * @dev: port device - * @addr: MAC address to add - * @vid: VLAN to add - * - * Add FDB entry to switch device. +/* Pass through a port object deletion, if @dev passes @check_cb, or replicate + * it towards all lower interfaces of @dev that pass @check_cb, if @dev is a + * bridge or a LAG. */ -int switchdev_port_fdb_add(struct ndmsg *ndm, struct nlattr *tb[], - struct net_device *dev, const unsigned char *addr, - u16 vid, u16 nlm_flags) +int switchdev_handle_port_obj_del(struct net_device *dev, + struct switchdev_notifier_port_obj_info *port_obj_info, + bool (*check_cb)(const struct net_device *dev), + int (*del_cb)(struct net_device *dev, const void *ctx, + const struct switchdev_obj *obj)) { - struct switchdev_obj_port_fdb fdb = { - .obj.orig_dev = dev, - .obj.id = SWITCHDEV_OBJ_ID_PORT_FDB, - .vid = vid, - }; + int err; - ether_addr_copy(fdb.addr, addr); - return switchdev_port_obj_add(dev, &fdb.obj); + err = __switchdev_handle_port_obj_del(dev, port_obj_info, check_cb, + NULL, del_cb); + if (err == -EOPNOTSUPP) + err = 0; + return err; } -EXPORT_SYMBOL_GPL(switchdev_port_fdb_add); +EXPORT_SYMBOL_GPL(switchdev_handle_port_obj_del); -/** - * switchdev_port_fdb_del - Delete FDB (MAC/VLAN) entry from port - * - * @ndmsg: netlink hdr - * @nlattr: netlink attributes - * @dev: port device - * @addr: MAC address to delete - * @vid: VLAN to delete - * - * Delete FDB entry from switch device. +/* Same as switchdev_handle_port_obj_del(), except if object is notified on a + * @dev that passes @foreign_dev_check_cb, it is replicated towards all devices + * that pass @check_cb and are in the same bridge as @dev. */ -int switchdev_port_fdb_del(struct ndmsg *ndm, struct nlattr *tb[], - struct net_device *dev, const unsigned char *addr, - u16 vid) +int switchdev_handle_port_obj_del_foreign(struct net_device *dev, + struct switchdev_notifier_port_obj_info *port_obj_info, + bool (*check_cb)(const struct net_device *dev), + bool (*foreign_dev_check_cb)(const struct net_device *dev, + const struct net_device *foreign_dev), + int (*del_cb)(struct net_device *dev, const void *ctx, + const struct switchdev_obj *obj)) { - struct switchdev_obj_port_fdb fdb = { - .obj.orig_dev = dev, - .obj.id = SWITCHDEV_OBJ_ID_PORT_FDB, - .vid = vid, - }; + int err; - ether_addr_copy(fdb.addr, addr); - return switchdev_port_obj_del(dev, &fdb.obj); + err = __switchdev_handle_port_obj_del(dev, port_obj_info, check_cb, + foreign_dev_check_cb, del_cb); + if (err == -EOPNOTSUPP) + err = 0; + return err; } -EXPORT_SYMBOL_GPL(switchdev_port_fdb_del); +EXPORT_SYMBOL_GPL(switchdev_handle_port_obj_del_foreign); + +static int __switchdev_handle_port_attr_set(struct net_device *dev, + struct switchdev_notifier_port_attr_info *port_attr_info, + bool (*check_cb)(const struct net_device *dev), + int (*set_cb)(struct net_device *dev, const void *ctx, + const struct switchdev_attr *attr, + struct netlink_ext_ack *extack)) +{ + struct switchdev_notifier_info *info = &port_attr_info->info; + struct netlink_ext_ack *extack; + struct net_device *lower_dev; + struct list_head *iter; + int err = -EOPNOTSUPP; -struct switchdev_fdb_dump { - struct switchdev_obj_port_fdb fdb; - struct net_device *dev; - struct sk_buff *skb; - struct netlink_callback *cb; - int idx; -}; + extack = switchdev_notifier_info_to_extack(info); -static int switchdev_port_fdb_dump_cb(struct switchdev_obj *obj) -{ - struct switchdev_obj_port_fdb *fdb = SWITCHDEV_OBJ_PORT_FDB(obj); - struct switchdev_fdb_dump *dump = - container_of(fdb, struct switchdev_fdb_dump, fdb); - u32 portid = NETLINK_CB(dump->cb->skb).portid; - u32 seq = dump->cb->nlh->nlmsg_seq; - struct nlmsghdr *nlh; - struct ndmsg *ndm; - - if (dump->idx < dump->cb->args[2]) - goto skip; - - nlh = nlmsg_put(dump->skb, portid, seq, RTM_NEWNEIGH, - sizeof(*ndm), NLM_F_MULTI); - if (!nlh) - return -EMSGSIZE; - - ndm = nlmsg_data(nlh); - ndm->ndm_family = AF_BRIDGE; - ndm->ndm_pad1 = 0; - ndm->ndm_pad2 = 0; - ndm->ndm_flags = NTF_SELF; - ndm->ndm_type = 0; - ndm->ndm_ifindex = dump->dev->ifindex; - ndm->ndm_state = fdb->ndm_state; - - if (nla_put(dump->skb, NDA_LLADDR, ETH_ALEN, fdb->addr)) - goto nla_put_failure; - - if (fdb->vid && nla_put_u16(dump->skb, NDA_VLAN, fdb->vid)) - goto nla_put_failure; - - nlmsg_end(dump->skb, nlh); - -skip: - dump->idx++; - return 0; + if (check_cb(dev)) { + err = set_cb(dev, info->ctx, port_attr_info->attr, extack); + if (err != -EOPNOTSUPP) + port_attr_info->handled = true; + return err; + } + + /* Switch ports might be stacked under e.g. a LAG. Ignore the + * unsupported devices, another driver might be able to handle them. But + * propagate to the callers any hard errors. + * + * If the driver does its own bookkeeping of stacked ports, it's not + * necessary to go through this helper. + */ + netdev_for_each_lower_dev(dev, lower_dev, iter) { + if (netif_is_bridge_master(lower_dev)) + continue; -nla_put_failure: - nlmsg_cancel(dump->skb, nlh); - return -EMSGSIZE; + err = __switchdev_handle_port_attr_set(lower_dev, port_attr_info, + check_cb, set_cb); + if (err && err != -EOPNOTSUPP) + return err; + } + + return err; } -/** - * switchdev_port_fdb_dump - Dump port FDB (MAC/VLAN) entries - * - * @skb: netlink skb - * @cb: netlink callback - * @dev: port device - * @filter_dev: filter device - * @idx: - * - * Dump FDB entries from switch device. - */ -int switchdev_port_fdb_dump(struct sk_buff *skb, struct netlink_callback *cb, - struct net_device *dev, - struct net_device *filter_dev, int *idx) +int switchdev_handle_port_attr_set(struct net_device *dev, + struct switchdev_notifier_port_attr_info *port_attr_info, + bool (*check_cb)(const struct net_device *dev), + int (*set_cb)(struct net_device *dev, const void *ctx, + const struct switchdev_attr *attr, + struct netlink_ext_ack *extack)) { - struct switchdev_fdb_dump dump = { - .fdb.obj.orig_dev = dev, - .fdb.obj.id = SWITCHDEV_OBJ_ID_PORT_FDB, - .dev = dev, - .skb = skb, - .cb = cb, - .idx = *idx, - }; int err; - err = switchdev_port_obj_dump(dev, &dump.fdb.obj, - switchdev_port_fdb_dump_cb); - *idx = dump.idx; + err = __switchdev_handle_port_attr_set(dev, port_attr_info, check_cb, + set_cb); + if (err == -EOPNOTSUPP) + err = 0; return err; } -EXPORT_SYMBOL_GPL(switchdev_port_fdb_dump); +EXPORT_SYMBOL_GPL(switchdev_handle_port_attr_set); + +int switchdev_bridge_port_offload(struct net_device *brport_dev, + struct net_device *dev, const void *ctx, + struct notifier_block *atomic_nb, + struct notifier_block *blocking_nb, + bool tx_fwd_offload, + struct netlink_ext_ack *extack) +{ + struct switchdev_notifier_brport_info brport_info = { + .brport = { + .dev = dev, + .ctx = ctx, + .atomic_nb = atomic_nb, + .blocking_nb = blocking_nb, + .tx_fwd_offload = tx_fwd_offload, + }, + }; + int err; + + ASSERT_RTNL(); + + err = call_switchdev_blocking_notifiers(SWITCHDEV_BRPORT_OFFLOADED, + brport_dev, &brport_info.info, + extack); + return notifier_to_errno(err); +} +EXPORT_SYMBOL_GPL(switchdev_bridge_port_offload); -bool switchdev_port_same_parent_id(struct net_device *a, - struct net_device *b) +void switchdev_bridge_port_unoffload(struct net_device *brport_dev, + const void *ctx, + struct notifier_block *atomic_nb, + struct notifier_block *blocking_nb) { - struct switchdev_attr a_attr = { - .orig_dev = a, - .id = SWITCHDEV_ATTR_ID_PORT_PARENT_ID, + struct switchdev_notifier_brport_info brport_info = { + .brport = { + .ctx = ctx, + .atomic_nb = atomic_nb, + .blocking_nb = blocking_nb, + }, }; - struct switchdev_attr b_attr = { - .orig_dev = b, - .id = SWITCHDEV_ATTR_ID_PORT_PARENT_ID, + + ASSERT_RTNL(); + + call_switchdev_blocking_notifiers(SWITCHDEV_BRPORT_UNOFFLOADED, + brport_dev, &brport_info.info, + NULL); +} +EXPORT_SYMBOL_GPL(switchdev_bridge_port_unoffload); + +int switchdev_bridge_port_replay(struct net_device *brport_dev, + struct net_device *dev, const void *ctx, + struct notifier_block *atomic_nb, + struct notifier_block *blocking_nb, + struct netlink_ext_ack *extack) +{ + struct switchdev_notifier_brport_info brport_info = { + .brport = { + .dev = dev, + .ctx = ctx, + .atomic_nb = atomic_nb, + .blocking_nb = blocking_nb, + }, }; + int err; - if (switchdev_port_attr_get(a, &a_attr) || - switchdev_port_attr_get(b, &b_attr)) - return false; + ASSERT_RTNL(); - return netdev_phys_item_id_same(&a_attr.u.ppid, &b_attr.u.ppid); + err = call_switchdev_blocking_notifiers(SWITCHDEV_BRPORT_REPLAY, + brport_dev, &brport_info.info, + extack); + return notifier_to_errno(err); } -EXPORT_SYMBOL_GPL(switchdev_port_same_parent_id); +EXPORT_SYMBOL_GPL(switchdev_bridge_port_replay); |