diff options
Diffstat (limited to 'drivers/net/ipvlan/ipvlan_main.c')
| -rw-r--r-- | drivers/net/ipvlan/ipvlan_main.c | 258 |
1 files changed, 104 insertions, 154 deletions
diff --git a/drivers/net/ipvlan/ipvlan_main.c b/drivers/net/ipvlan/ipvlan_main.c index 19bdde60680c..660f3db11766 100644 --- a/drivers/net/ipvlan/ipvlan_main.c +++ b/drivers/net/ipvlan/ipvlan_main.c @@ -1,81 +1,16 @@ +// SPDX-License-Identifier: GPL-2.0-or-later /* Copyright (c) 2014 Mahesh Bandewar <maheshb@google.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 "ipvlan.h" - -static unsigned int ipvlan_netid __read_mostly; - -struct ipvlan_netns { - unsigned int ipvl_nf_hook_refcnt; -}; - -static const struct nf_hook_ops ipvl_nfops[] = { - { - .hook = ipvlan_nf_input, - .pf = NFPROTO_IPV4, - .hooknum = NF_INET_LOCAL_IN, - .priority = INT_MAX, - }, -#if IS_ENABLED(CONFIG_IPV6) - { - .hook = ipvlan_nf_input, - .pf = NFPROTO_IPV6, - .hooknum = NF_INET_LOCAL_IN, - .priority = INT_MAX, - }, -#endif -}; +#include <linux/ethtool.h> +#include <net/netdev_lock.h> -static const struct l3mdev_ops ipvl_l3mdev_ops = { - .l3mdev_l3_rcv = ipvlan_l3_rcv, -}; - -static void ipvlan_adjust_mtu(struct ipvl_dev *ipvlan, struct net_device *dev) -{ - ipvlan->dev->mtu = dev->mtu; -} - -static int ipvlan_register_nf_hook(struct net *net) -{ - struct ipvlan_netns *vnet = net_generic(net, ipvlan_netid); - int err = 0; - - if (!vnet->ipvl_nf_hook_refcnt) { - err = nf_register_net_hooks(net, ipvl_nfops, - ARRAY_SIZE(ipvl_nfops)); - if (!err) - vnet->ipvl_nf_hook_refcnt = 1; - } else { - vnet->ipvl_nf_hook_refcnt++; - } - - return err; -} - -static void ipvlan_unregister_nf_hook(struct net *net) -{ - struct ipvlan_netns *vnet = net_generic(net, ipvlan_netid); - - if (WARN_ON(!vnet->ipvl_nf_hook_refcnt)) - return; - - vnet->ipvl_nf_hook_refcnt--; - if (!vnet->ipvl_nf_hook_refcnt) - nf_unregister_net_hooks(net, ipvl_nfops, - ARRAY_SIZE(ipvl_nfops)); -} +#include "ipvlan.h" static int ipvlan_set_port_mode(struct ipvl_port *port, u16 nval, struct netlink_ext_ack *extack) { struct ipvl_dev *ipvlan; - struct net_device *mdev = port->dev; unsigned int flags; int err; @@ -97,17 +32,12 @@ static int ipvlan_set_port_mode(struct ipvl_port *port, u16 nval, } if (nval == IPVLAN_MODE_L3S) { /* New mode is L3S */ - err = ipvlan_register_nf_hook(read_pnet(&port->pnet)); - if (!err) { - mdev->l3mdev_ops = &ipvl_l3mdev_ops; - mdev->priv_flags |= IFF_L3MDEV_MASTER; - } else + err = ipvlan_l3s_register(port); + if (err) goto fail; } else if (port->mode == IPVLAN_MODE_L3S) { /* Old mode was L3S */ - mdev->priv_flags &= ~IFF_L3MDEV_MASTER; - ipvlan_unregister_nf_hook(read_pnet(&port->pnet)); - mdev->l3mdev_ops = NULL; + ipvlan_l3s_unregister(port); } port->mode = nval; } @@ -154,6 +84,7 @@ static int ipvlan_port_create(struct net_device *dev) if (err) goto err; + netdev_hold(dev, &port->dev_tracker, GFP_KERNEL); return 0; err: @@ -166,28 +97,34 @@ static void ipvlan_port_destroy(struct net_device *dev) struct ipvl_port *port = ipvlan_port_get_rtnl(dev); struct sk_buff *skb; - if (port->mode == IPVLAN_MODE_L3S) { - dev->priv_flags &= ~IFF_L3MDEV_MASTER; - ipvlan_unregister_nf_hook(dev_net(dev)); - dev->l3mdev_ops = NULL; - } + netdev_put(dev, &port->dev_tracker); + if (port->mode == IPVLAN_MODE_L3S) + ipvlan_l3s_unregister(port); netdev_rx_handler_unregister(dev); cancel_work_sync(&port->wq); while ((skb = __skb_dequeue(&port->backlog)) != NULL) { - if (skb->dev) - dev_put(skb->dev); + dev_put(skb->dev); kfree_skb(skb); } ida_destroy(&port->ida); kfree(port); } +#define IPVLAN_ALWAYS_ON_OFLOADS \ + (NETIF_F_SG | NETIF_F_HW_CSUM | \ + NETIF_F_GSO_ROBUST | NETIF_F_GSO_SOFTWARE | NETIF_F_GSO_ENCAP_ALL) + +#define IPVLAN_ALWAYS_ON \ + (IPVLAN_ALWAYS_ON_OFLOADS | NETIF_F_VLAN_CHALLENGED) + #define IPVLAN_FEATURES \ (NETIF_F_SG | NETIF_F_HW_CSUM | NETIF_F_HIGHDMA | NETIF_F_FRAGLIST | \ - NETIF_F_GSO | NETIF_F_TSO | NETIF_F_GSO_ROBUST | \ - NETIF_F_TSO_ECN | NETIF_F_TSO6 | NETIF_F_GRO | NETIF_F_RXCSUM | \ + NETIF_F_GSO | NETIF_F_ALL_TSO | NETIF_F_GSO_ROBUST | \ + NETIF_F_GRO | NETIF_F_RXCSUM | \ NETIF_F_HW_VLAN_CTAG_FILTER | NETIF_F_HW_VLAN_STAG_FILTER) + /* NETIF_F_GSO_ENCAP_ALL NETIF_F_GSO_SOFTWARE Newly added */ + #define IPVLAN_STATE_MASK \ ((1<<__LINK_STATE_NOCARRIER) | (1<<__LINK_STATE_DORMANT)) @@ -201,9 +138,12 @@ static int ipvlan_init(struct net_device *dev) dev->state = (dev->state & ~IPVLAN_STATE_MASK) | (phy_dev->state & IPVLAN_STATE_MASK); dev->features = phy_dev->features & IPVLAN_FEATURES; - dev->features |= NETIF_F_LLTX | NETIF_F_VLAN_CHALLENGED; - dev->gso_max_size = phy_dev->gso_max_size; - dev->gso_max_segs = phy_dev->gso_max_segs; + dev->features |= IPVLAN_ALWAYS_ON; + dev->vlan_features = phy_dev->vlan_features & IPVLAN_FEATURES; + dev->vlan_features |= IPVLAN_ALWAYS_ON_OFLOADS; + dev->hw_enc_features |= dev->features; + dev->lltx = true; + netif_inherit_tso_max(dev, phy_dev); dev->hard_header_len = phy_dev->hard_header_len; netdev_lockdep_set_classes(dev); @@ -241,7 +181,6 @@ static void ipvlan_uninit(struct net_device *dev) static int ipvlan_open(struct net_device *dev) { struct ipvl_dev *ipvlan = netdev_priv(dev); - struct net_device *phy_dev = ipvlan->phy_dev; struct ipvl_addr *addr; if (ipvlan->port->mode == IPVLAN_MODE_L3 || @@ -255,7 +194,7 @@ static int ipvlan_open(struct net_device *dev) ipvlan_ht_addr_add(ipvlan, addr); rcu_read_unlock(); - return dev_uc_add(phy_dev, phy_dev->dev_addr); + return 0; } static int ipvlan_stop(struct net_device *dev) @@ -267,8 +206,6 @@ static int ipvlan_stop(struct net_device *dev) dev_uc_unsync(phy_dev, dev); dev_mc_unsync(phy_dev, dev); - dev_uc_del(phy_dev, phy_dev->dev_addr); - rcu_read_lock(); list_for_each_entry_rcu(addr, &ipvlan->addrs, anode) ipvlan_ht_addr_del(addr); @@ -291,8 +228,8 @@ static netdev_tx_t ipvlan_start_xmit(struct sk_buff *skb, pcptr = this_cpu_ptr(ipvlan->pcpu_stats); u64_stats_update_begin(&pcptr->syncp); - pcptr->tx_pkts++; - pcptr->tx_bytes += skblen; + u64_stats_inc(&pcptr->tx_pkts); + u64_stats_add(&pcptr->tx_bytes, skblen); u64_stats_update_end(&pcptr->syncp); } else { this_cpu_inc(ipvlan->pcpu_stats->tx_drps); @@ -305,7 +242,14 @@ static netdev_features_t ipvlan_fix_features(struct net_device *dev, { struct ipvl_dev *ipvlan = netdev_priv(dev); - return features & (ipvlan->sfeatures | ~IPVLAN_FEATURES); + features |= NETIF_F_ALL_FOR_ALL; + features &= (ipvlan->sfeatures | ~IPVLAN_FEATURES); + features = netdev_increment_features(ipvlan->phy_dev->features, + features, features); + features |= IPVLAN_ALWAYS_ON; + features &= (IPVLAN_FEATURES | IPVLAN_ALWAYS_ON); + + return features; } static void ipvlan_change_rx_flags(struct net_device *dev, int change) @@ -359,13 +303,13 @@ static void ipvlan_get_stats64(struct net_device *dev, for_each_possible_cpu(idx) { pcptr = per_cpu_ptr(ipvlan->pcpu_stats, idx); do { - strt= u64_stats_fetch_begin_irq(&pcptr->syncp); - rx_pkts = pcptr->rx_pkts; - rx_bytes = pcptr->rx_bytes; - rx_mcast = pcptr->rx_mcast; - tx_pkts = pcptr->tx_pkts; - tx_bytes = pcptr->tx_bytes; - } while (u64_stats_fetch_retry_irq(&pcptr->syncp, + strt = u64_stats_fetch_begin(&pcptr->syncp); + rx_pkts = u64_stats_read(&pcptr->rx_pkts); + rx_bytes = u64_stats_read(&pcptr->rx_bytes); + rx_mcast = u64_stats_read(&pcptr->rx_mcast); + tx_pkts = u64_stats_read(&pcptr->tx_pkts); + tx_bytes = u64_stats_read(&pcptr->tx_bytes); + } while (u64_stats_fetch_retry(&pcptr->syncp, strt)); s->rx_packets += rx_pkts; @@ -375,13 +319,14 @@ static void ipvlan_get_stats64(struct net_device *dev, s->tx_bytes += tx_bytes; /* u32 values are updated without syncp protection. */ - rx_errs += pcptr->rx_errs; - tx_drps += pcptr->tx_drps; + rx_errs += READ_ONCE(pcptr->rx_errs); + tx_drps += READ_ONCE(pcptr->tx_drps); } s->rx_errors = rx_errs; s->rx_dropped = rx_errs; s->tx_dropped = tx_drps; } + s->tx_errors = DEV_STATS_READ(dev, tx_errors); } static int ipvlan_vlan_rx_add_vid(struct net_device *dev, __be16 proto, u16 vid) @@ -406,7 +351,7 @@ static int ipvlan_get_iflink(const struct net_device *dev) { struct ipvl_dev *ipvlan = netdev_priv(dev); - return ipvlan->phy_dev->ifindex; + return READ_ONCE(ipvlan->phy_dev->ifindex); } static const struct net_device_ops ipvlan_netdev_ops = { @@ -444,8 +389,14 @@ static const struct header_ops ipvlan_header_ops = { .parse = eth_header_parse, .cache = eth_header_cache, .cache_update = eth_header_cache_update, + .parse_protocol = eth_header_parse_protocol, }; +static void ipvlan_adjust_mtu(struct ipvl_dev *ipvlan, struct net_device *dev) +{ + ipvlan->dev->mtu = dev->mtu; +} + static bool netif_is_ipvlan(const struct net_device *dev) { /* both ipvlan and ipvtap devices use the same netdev_ops */ @@ -463,8 +414,8 @@ static int ipvlan_ethtool_get_link_ksettings(struct net_device *dev, static void ipvlan_ethtool_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *drvinfo) { - strlcpy(drvinfo->driver, IPVLAN_DRV, sizeof(drvinfo->driver)); - strlcpy(drvinfo->version, IPV_DRV_VER, sizeof(drvinfo->version)); + strscpy(drvinfo->driver, IPVLAN_DRV, sizeof(drvinfo->driver)); + strscpy(drvinfo->version, IPV_DRV_VER, sizeof(drvinfo->version)); } static u32 ipvlan_ethtool_get_msglevel(struct net_device *dev) @@ -499,6 +450,8 @@ static int ipvlan_nl_changelink(struct net_device *dev, if (!data) return 0; + if (!ns_capable(dev_net(ipvlan->phy_dev)->user_ns, CAP_NET_ADMIN)) + return -EPERM; if (data[IFLA_IPVLAN_MODE]) { u16 nmode = nla_get_u16(data[IFLA_IPVLAN_MODE]); @@ -580,11 +533,13 @@ err: return ret; } -int ipvlan_link_new(struct net *src_net, struct net_device *dev, - struct nlattr *tb[], struct nlattr *data[], +int ipvlan_link_new(struct net_device *dev, struct rtnl_newlink_params *params, struct netlink_ext_ack *extack) { + struct net *link_net = rtnl_newlink_link_net(params); struct ipvl_dev *ipvlan = netdev_priv(dev); + struct nlattr **data = params->data; + struct nlattr **tb = params->tb; struct ipvl_port *port; struct net_device *phy_dev; int err; @@ -593,7 +548,7 @@ int ipvlan_link_new(struct net *src_net, struct net_device *dev, if (!tb[IFLA_LINK]) return -EINVAL; - phy_dev = __dev_get_by_index(src_net, nla_get_u32(tb[IFLA_LINK])); + phy_dev = __dev_get_by_index(link_net, nla_get_u32(tb[IFLA_LINK])); if (!phy_dev) return -ENODEV; @@ -601,6 +556,8 @@ int ipvlan_link_new(struct net *src_net, struct net_device *dev, struct ipvl_dev *tmp = netdev_priv(phy_dev); phy_dev = tmp->phy_dev; + if (!ns_capable(dev_net(phy_dev)->user_ns, CAP_NET_ADMIN)) + return -EPERM; } else if (!netif_is_ipvlan_port(phy_dev)) { /* Exit early if the underlying link is invalid or busy */ if (phy_dev->type != ARPHRD_ETHER || @@ -628,7 +585,7 @@ int ipvlan_link_new(struct net *src_net, struct net_device *dev, * world but keep using the physical-dev address for the outgoing * packets. */ - memcpy(dev->dev_addr, phy_dev->dev_addr, ETH_ALEN); + eth_hw_addr_set(dev, phy_dev->dev_addr); dev->priv_flags |= IFF_NO_RX_HANDLER; @@ -648,15 +605,15 @@ int ipvlan_link_new(struct net *src_net, struct net_device *dev, port->dev_id_start = 0x1; /* Since L2 address is shared among all IPvlan slaves including - * master, use unique 16 bit dev-ids to diffentiate among them. + * master, use unique 16 bit dev-ids to differentiate among them. * Assign IDs between 0x1 and 0xFFFE (used by the master) to each * slave link [see addrconf_ifid_eui48()]. */ - err = ida_simple_get(&port->ida, port->dev_id_start, 0xFFFE, - GFP_KERNEL); + err = ida_alloc_range(&port->ida, port->dev_id_start, 0xFFFD, + GFP_KERNEL); if (err < 0) - err = ida_simple_get(&port->ida, 0x1, port->dev_id_start, - GFP_KERNEL); + err = ida_alloc_range(&port->ida, 0x1, port->dev_id_start - 1, + GFP_KERNEL); if (err < 0) goto unregister_netdev; dev->dev_id = err; @@ -688,7 +645,7 @@ int ipvlan_link_new(struct net *src_net, struct net_device *dev, unlink_netdev: netdev_upper_dev_unlink(phy_dev, dev); remove_ida: - ida_simple_remove(&port->ida, dev->dev_id); + ida_free(&port->ida, dev->dev_id); unregister_netdev: unregister_netdevice(dev); return err; @@ -708,7 +665,7 @@ void ipvlan_link_delete(struct net_device *dev, struct list_head *head) } spin_unlock_bh(&ipvlan->addrs_lock); - ida_simple_remove(&ipvlan->port->ida, dev->dev_id); + ida_free(&ipvlan->port->ida, dev->dev_id); list_del_rcu(&ipvlan->pnode); unregister_netdevice_queue(dev, head); netdev_upper_dev_unlink(ipvlan->phy_dev, dev); @@ -735,6 +692,13 @@ static const struct nla_policy ipvlan_nl_policy[IFLA_IPVLAN_MAX + 1] = [IFLA_IPVLAN_FLAGS] = { .type = NLA_U16 }, }; +static struct net *ipvlan_get_link_net(const struct net_device *dev) +{ + struct ipvl_dev *ipvlan = netdev_priv(dev); + + return dev_net(ipvlan->phy_dev); +} + static struct rtnl_link_ops ipvlan_link_ops = { .kind = "ipvlan", .priv_size = sizeof(struct ipvl_dev), @@ -742,6 +706,7 @@ static struct rtnl_link_ops ipvlan_link_ops = { .setup = ipvlan_link_setup, .newlink = ipvlan_link_new, .dellink = ipvlan_link_delete, + .get_link_net = ipvlan_get_link_net, }; int ipvlan_link_register(struct rtnl_link_ops *ops) @@ -773,6 +738,8 @@ static int ipvlan_device_event(struct notifier_block *unused, port = ipvlan_port_get_rtnl(dev); switch (event) { + case NETDEV_UP: + case NETDEV_DOWN: case NETDEV_CHANGE: list_for_each_entry(ipvlan, &port->ipvlans, pnode) netif_stacked_transfer_operstate(ipvlan->phy_dev, @@ -781,7 +748,6 @@ static int ipvlan_device_event(struct notifier_block *unused, case NETDEV_REGISTER: { struct net *oldnet, *newnet = dev_net(dev); - struct ipvlan_netns *old_vnet; oldnet = read_pnet(&port->pnet); if (net_eq(newnet, oldnet)) @@ -789,12 +755,8 @@ static int ipvlan_device_event(struct notifier_block *unused, write_pnet(&port->pnet, newnet); - old_vnet = net_generic(oldnet, ipvlan_netid); - if (!old_vnet->ipvl_nf_hook_refcnt) - break; - - ipvlan_register_nf_hook(newnet); - ipvlan_unregister_nf_hook(oldnet); + if (port->mode == IPVLAN_MODE_L3S) + ipvlan_migrate_l3s_hook(oldnet, newnet); break; } case NETDEV_UNREGISTER: @@ -809,10 +771,8 @@ static int ipvlan_device_event(struct notifier_block *unused, case NETDEV_FEAT_CHANGE: list_for_each_entry(ipvlan, &port->ipvlans, pnode) { - ipvlan->dev->features = dev->features & IPVLAN_FEATURES; - ipvlan->dev->gso_max_size = dev->gso_max_size; - ipvlan->dev->gso_max_segs = dev->gso_max_segs; - netdev_features_change(ipvlan->dev); + netif_inherit_tso_max(ipvlan->dev, dev); + netdev_update_features(ipvlan->dev); } break; @@ -824,9 +784,9 @@ static int ipvlan_device_event(struct notifier_block *unused, case NETDEV_PRE_CHANGEADDR: prechaddr_info = ptr; list_for_each_entry(ipvlan, &port->ipvlans, pnode) { - err = dev_pre_changeaddr_notify(ipvlan->dev, - prechaddr_info->dev_addr, - extack); + err = netif_pre_changeaddr_notify(ipvlan->dev, + prechaddr_info->dev_addr, + extack); if (err) return notifier_from_errno(err); } @@ -834,7 +794,7 @@ static int ipvlan_device_event(struct notifier_block *unused, case NETDEV_CHANGEADDR: list_for_each_entry(ipvlan, &port->ipvlans, pnode) { - ether_addr_copy(ipvlan->dev->dev_addr, dev->dev_addr); + eth_hw_addr_set(ipvlan->dev, dev->dev_addr); call_netdevice_notifiers(NETDEV_CHANGEADDR, ipvlan->dev); } break; @@ -842,6 +802,12 @@ static int ipvlan_device_event(struct notifier_block *unused, case NETDEV_PRE_TYPE_CHANGE: /* Forbid underlying device to change its type. */ return NOTIFY_BAD; + + case NETDEV_NOTIFY_PEERS: + case NETDEV_BONDING_FAILOVER: + case NETDEV_RESEND_IGMP: + list_for_each_entry(ipvlan, &port->ipvlans, pnode) + call_netdevice_notifiers(event, ipvlan->dev); } return NOTIFY_DONE; } @@ -1068,23 +1034,6 @@ static struct notifier_block ipvlan_addr6_vtor_notifier_block __read_mostly = { }; #endif -static void ipvlan_ns_exit(struct net *net) -{ - struct ipvlan_netns *vnet = net_generic(net, ipvlan_netid); - - if (WARN_ON_ONCE(vnet->ipvl_nf_hook_refcnt)) { - vnet->ipvl_nf_hook_refcnt = 0; - nf_unregister_net_hooks(net, ipvl_nfops, - ARRAY_SIZE(ipvl_nfops)); - } -} - -static struct pernet_operations ipvlan_net_ops = { - .id = &ipvlan_netid, - .size = sizeof(struct ipvlan_netns), - .exit = ipvlan_ns_exit, -}; - static int __init ipvlan_init_module(void) { int err; @@ -1099,13 +1048,13 @@ static int __init ipvlan_init_module(void) register_inetaddr_notifier(&ipvlan_addr4_notifier_block); register_inetaddr_validator_notifier(&ipvlan_addr4_vtor_notifier_block); - err = register_pernet_subsys(&ipvlan_net_ops); + err = ipvlan_l3s_init(); if (err < 0) goto error; err = ipvlan_link_register(&ipvlan_link_ops); if (err < 0) { - unregister_pernet_subsys(&ipvlan_net_ops); + ipvlan_l3s_cleanup(); goto error; } @@ -1126,7 +1075,7 @@ error: static void __exit ipvlan_cleanup_module(void) { rtnl_link_unregister(&ipvlan_link_ops); - unregister_pernet_subsys(&ipvlan_net_ops); + ipvlan_l3s_cleanup(); unregister_netdevice_notifier(&ipvlan_notifier_block); unregister_inetaddr_notifier(&ipvlan_addr4_notifier_block); unregister_inetaddr_validator_notifier( @@ -1145,3 +1094,4 @@ MODULE_LICENSE("GPL"); MODULE_AUTHOR("Mahesh Bandewar <maheshb@google.com>"); MODULE_DESCRIPTION("Driver for L3 (IPv6/IPv4) based VLANs"); MODULE_ALIAS_RTNL_LINK("ipvlan"); +MODULE_IMPORT_NS("NETDEV_INTERNAL"); |
