vxlan: add changelink support
Subsystems (2)
PATCH 1/1 vxlan: add changelink support
From: Roopa Prabhu <[email protected]>
This patch adds changelink rtnl op support for vxlan netdevs. code changes involve:
- refactor vxlan_newlink into vxlan_nl2conf to be used by vxlan_newlink and vxlan_changelink
- vxlan_nl2conf and vxlan_dev_configure take a changelink argument to isolate changelink checks and updates.
- Allow changing only a few attributes:
- return -EOPNOTSUPP for attributes that cannot be changed for now. Incremental patches can make the non-supported one available in the future if needed.
--- drivers/net/vxlan.c | 383 ++++++++++++++++++++++++++++++++++++--------------- 1 file changed, 270 insertions(+), 113 deletions(-)
drivers/net/vxlan.c +270 -113
@@ -2835,39 +2835,40 @@ static int vxlan_sock_add(struct vxlan_dev *vxlan) }static int vxlan_dev_configure (struct net *src_net,struct net_device *dev, -struct vxlan_config *conf) +struct vxlan_config *conf, +bool changelink) {struct vxlan_net *vn =net_generic (src_net, vxlan_net_id);struct vxlan_dev *vxlan =netdev_priv (dev), *tmp;struct vxlan_rdst *dst = &vxlan->default_dst;unsigned short needed_headroom =ETH_HLEN ; -int err;bool use_ipv6 = false;__be16 default_port = vxlan->cfg.dst_port;struct net_device *lowerdev =NULL ; -if (conf->flags &VXLAN_F_GPE ) { -/* For now, allow GPE only together with COLLECT_METADATA. -* This can be relaxed later; in such case, the other side -* of the PtP link will have to be provided. -*/ -if ((conf->flags & ~VXLAN_F_ALLOWED_GPE ) || - !(conf->flags &VXLAN_F_COLLECT_METADATA )) { -pr_info ("unsupported combination of extensions\n" ); -return -EINVAL ; +if (!changelink) { +if (conf->flags &VXLAN_F_GPE ) { +/* For now, allow GPE only together with +* COLLECT_METADATA. This can be relaxed later; in such +* case, the other side of the PtP link will have to be +* provided. +*/ +if ((conf->flags & ~VXLAN_F_ALLOWED_GPE ) || + !(conf->flags &VXLAN_F_COLLECT_METADATA )) { +pr_info ("unsupported combination of extensions\n" ); +return -EINVAL ; + } +vxlan_raw_setup (dev); + }else { +vxlan_ether_setup (dev); } -vxlan_raw_setup (dev); - }else { -vxlan_ether_setup (dev); +/* MTU range: 68 - 65535 */ + dev->min_mtu =ETH_MIN_MTU ; + dev->max_mtu =ETH_MAX_MTU ; + vxlan->net = src_net; } -/* MTU range: 68 - 65535 */ - dev->min_mtu =ETH_MIN_MTU ; - dev->max_mtu =ETH_MAX_MTU ; - - vxlan->net = src_net; - dst->remote_vni = conf->vni;memcpy (&dst->remote_ip, &conf->remote_ip,sizeof (conf->remote_ip));
@@ -2889,12 +2890,14 @@ static int vxlan_dev_configure(struct net *src_net, struct net_device *dev,return -EINVAL ; } -if (conf->remote_ifindex) { +if (conf->remote_ifindex && + conf->remote_ifindex != vxlan->cfg.remote_ifindex) { lowerdev =__dev_get_by_index (src_net, conf->remote_ifindex); dst->remote_ifindex = conf->remote_ifindex;if (!lowerdev) { -pr_info ("ifindex %d does not exist\n" , dst->remote_ifindex); +pr_info ("ifindex %d does not exist\n" , + dst->remote_ifindex);return -ENODEV ; }
@@ -2913,7 +2916,8 @@ static int vxlan_dev_configure(struct net *src_net, struct net_device *dev, (use_ipv6 ?VXLAN6_HEADROOM :VXLAN_HEADROOM ); needed_headroom = lowerdev->hard_header_len; - }else if (vxlan_addr_multicast (&dst->remote_ip)) { + }else if (!conf->remote_ifindex && +vxlan_addr_multicast (&dst->remote_ip)) {pr_info ("multicast destination requires interface to be specified\n" );return -EINVAL ; }
@@ -2953,6 +2957,9 @@ static int vxlan_dev_configure(struct net *src_net, struct net_device *dev,if (!vxlan->cfg.age_interval) vxlan->cfg.age_interval =FDB_AGE_DEFAULT ; +if (changelink) +return 0 ; +list_for_each_entry (tmp, &vn->vxlan_list, next) { if (tmp->cfg.vni == conf->vni && (tmp->default_dst.remote_ip.sa.sa_family == AF_INET6 ||
@@ -2965,147 +2972,296 @@ static int vxlan_dev_configure(struct net *src_net, struct net_device *dev, } } - dev->ethtool_ops = &vxlan_ethtool_ops; - -/* create an fdb entry for a valid default destination */ -if (!vxlan_addr_any (&vxlan->default_dst.remote_ip)) { - err =vxlan_fdb_create (vxlan, all_zeros_mac, - &vxlan->default_dst.remote_ip, -NUD_REACHABLE |NUD_PERMANENT , -NLM_F_EXCL |NLM_F_CREATE , - vxlan->cfg.dst_port, - vxlan->default_dst.remote_vni, - vxlan->default_dst.remote_vni, - vxlan->default_dst.remote_ifindex, -NTF_SELF ); -if (err) -return err; - } - - err =register_netdevice (dev); -if (err) { -vxlan_fdb_delete_default (vxlan, vxlan->cfg.vni); -return err; - } - -list_add (&vxlan->next, &vn->vxlan_list); -return 0 ; } -static int vxlan_newlink (struct net *src_net,struct net_device *dev, -struct nlattr *tb[],struct nlattr *data[]) +static int vxlan_nl2conf (struct nlattr *tb[],struct nlattr *data[], +struct net_device *dev,struct vxlan_config *conf, +bool changelink) { -struct vxlan_config conf; +struct vxlan_dev *vxlan =netdev_priv (dev); -memset (&conf,0 ,sizeof (conf)); +memset (conf,0 ,sizeof (*conf)); -if (data[IFLA_VXLAN_ID ]) - conf.vni =cpu_to_be32 (nla_get_u32 (data[IFLA_VXLAN_ID ])); +/* if changelink operation, start with old existing cfg */ +if (changelink) +memcpy (conf, &vxlan->cfg,sizeof (*conf)); + +if (data[IFLA_VXLAN_ID ]) { +__be32 vni =cpu_to_be32 (nla_get_u32 (data[IFLA_VXLAN_ID ])); + +if (changelink && (vni != conf->vni)) +return -EOPNOTSUPP ; + conf->vni =cpu_to_be32 (nla_get_u32 (data[IFLA_VXLAN_ID ])); + }if (data[IFLA_VXLAN_GROUP ]) { - conf.remote_ip.sin.sin_addr.s_addr =nla_get_in_addr (data[IFLA_VXLAN_GROUP ]); + conf->remote_ip.sin.sin_addr.s_addr =nla_get_in_addr (data[IFLA_VXLAN_GROUP ]); }else if (data[IFLA_VXLAN_GROUP6 ]) {if (!IS_ENABLED (CONFIG_IPV6 ))return -EPFNOSUPPORT ; - conf.remote_ip.sin6.sin6_addr =nla_get_in6_addr (data[IFLA_VXLAN_GROUP6 ]); - conf.remote_ip.sa.sa_family =AF_INET6 ; + conf->remote_ip.sin6.sin6_addr =nla_get_in6_addr (data[IFLA_VXLAN_GROUP6 ]); + conf->remote_ip.sa.sa_family =AF_INET6 ; }if (data[IFLA_VXLAN_LOCAL ]) { - conf.saddr.sin.sin_addr.s_addr =nla_get_in_addr (data[IFLA_VXLAN_LOCAL ]); - conf.saddr.sa.sa_family =AF_INET ; + conf->saddr.sin.sin_addr.s_addr =nla_get_in_addr (data[IFLA_VXLAN_LOCAL ]); + conf->saddr.sa.sa_family =AF_INET ; }else if (data[IFLA_VXLAN_LOCAL6 ]) {if (!IS_ENABLED (CONFIG_IPV6 ))return -EPFNOSUPPORT ;/* TODO: respect scope id */ - conf.saddr.sin6.sin6_addr =nla_get_in6_addr (data[IFLA_VXLAN_LOCAL6 ]); - conf.saddr.sa.sa_family =AF_INET6 ; + conf->saddr.sin6.sin6_addr =nla_get_in6_addr (data[IFLA_VXLAN_LOCAL6 ]); + conf->saddr.sa.sa_family =AF_INET6 ; }if (data[IFLA_VXLAN_LINK ]) - conf.remote_ifindex =nla_get_u32 (data[IFLA_VXLAN_LINK ]); + conf->remote_ifindex =nla_get_u32 (data[IFLA_VXLAN_LINK ]);if (data[IFLA_VXLAN_TOS ]) - conf.tos =nla_get_u8 (data[IFLA_VXLAN_TOS ]); + conf->tos =nla_get_u8 (data[IFLA_VXLAN_TOS ]);if (data[IFLA_VXLAN_TTL ]) - conf.ttl =nla_get_u8 (data[IFLA_VXLAN_TTL ]); + conf->ttl =nla_get_u8 (data[IFLA_VXLAN_TTL ]);if (data[IFLA_VXLAN_LABEL ]) - conf.label =nla_get_be32 (data[IFLA_VXLAN_LABEL ]) & + conf->label =nla_get_be32 (data[IFLA_VXLAN_LABEL ]) &IPV6_FLOWLABEL_MASK ; -if (!data[IFLA_VXLAN_LEARNING ] ||nla_get_u8 (data[IFLA_VXLAN_LEARNING ])) - conf.flags |=VXLAN_F_LEARN ; +if (data[IFLA_VXLAN_LEARNING ]) { +if (nla_get_u8 (data[IFLA_VXLAN_LEARNING ])) { + conf->flags |=VXLAN_F_LEARN ; + }else { + conf->flags &= ~VXLAN_F_LEARN ; + vxlan->flags &= ~VXLAN_F_LEARN ; + } + }else if (!changelink) { +/* default to learn on a new device */ + conf->flags |=VXLAN_F_LEARN ; + } -if (data[IFLA_VXLAN_AGEING ]) - conf.age_interval =nla_get_u32 (data[IFLA_VXLAN_AGEING ]); +if (data[IFLA_VXLAN_AGEING ]) { +if (changelink) +return -EOPNOTSUPP ; + conf->age_interval =nla_get_u32 (data[IFLA_VXLAN_AGEING ]); + } -if (data[IFLA_VXLAN_PROXY ] &&nla_get_u8 (data[IFLA_VXLAN_PROXY ])) - conf.flags |=VXLAN_F_PROXY ; +if (data[IFLA_VXLAN_PROXY ]) { +if (changelink) +return -EOPNOTSUPP ; +if (nla_get_u8 (data[IFLA_VXLAN_PROXY ])) + conf->flags |=VXLAN_F_PROXY ; + } -if (data[IFLA_VXLAN_RSC ] &&nla_get_u8 (data[IFLA_VXLAN_RSC ])) - conf.flags |=VXLAN_F_RSC ; +if (data[IFLA_VXLAN_RSC ]) { +if (changelink) +return -EOPNOTSUPP ; +if (nla_get_u8 (data[IFLA_VXLAN_RSC ])) + conf->flags |=VXLAN_F_RSC ; + } -if (data[IFLA_VXLAN_L2MISS ] &&nla_get_u8 (data[IFLA_VXLAN_L2MISS ])) - conf.flags |=VXLAN_F_L2MISS ; +if (data[IFLA_VXLAN_L2MISS ]) { +if (changelink) +return -EOPNOTSUPP ; +if (nla_get_u8 (data[IFLA_VXLAN_L2MISS ])) + conf->flags |=VXLAN_F_L2MISS ; + } -if (data[IFLA_VXLAN_L3MISS ] &&nla_get_u8 (data[IFLA_VXLAN_L3MISS ])) - conf.flags |=VXLAN_F_L3MISS ; +if (data[IFLA_VXLAN_L3MISS ]) { +if (changelink) +return -EOPNOTSUPP ; +if (nla_get_u8 (data[IFLA_VXLAN_L3MISS ])) + conf->flags |=VXLAN_F_L3MISS ; + } -if (data[IFLA_VXLAN_LIMIT ]) - conf.addrmax =nla_get_u32 (data[IFLA_VXLAN_LIMIT ]); +if (data[IFLA_VXLAN_LIMIT ]) { +if (changelink) +return -EOPNOTSUPP ; + conf->addrmax =nla_get_u32 (data[IFLA_VXLAN_LIMIT ]); + } -if (data[IFLA_VXLAN_COLLECT_METADATA ] && -nla_get_u8 (data[IFLA_VXLAN_COLLECT_METADATA ])) - conf.flags |=VXLAN_F_COLLECT_METADATA ; +if (data[IFLA_VXLAN_COLLECT_METADATA ]) { +if (changelink) +return -EOPNOTSUPP ; +if (nla_get_u8 (data[IFLA_VXLAN_COLLECT_METADATA ])) + conf->flags |=VXLAN_F_COLLECT_METADATA ; + }if (data[IFLA_VXLAN_PORT_RANGE ]) { -const struct ifla_vxlan_port_range *p - =nla_data (data[IFLA_VXLAN_PORT_RANGE ]); - conf.port_min =ntohs (p->low); - conf.port_max =ntohs (p->high); +if (!changelink) { +const struct ifla_vxlan_port_range *p + =nla_data (data[IFLA_VXLAN_PORT_RANGE ]); + conf->port_min =ntohs (p->low); + conf->port_max =ntohs (p->high); + }else { +return -EOPNOTSUPP ; + } } -if (data[IFLA_VXLAN_PORT ]) - conf.dst_port =nla_get_be16 (data[IFLA_VXLAN_PORT ]); +if (data[IFLA_VXLAN_PORT ]) { +if (changelink) +return -EOPNOTSUPP ; + conf->dst_port =nla_get_be16 (data[IFLA_VXLAN_PORT ]); + } -if (data[IFLA_VXLAN_UDP_CSUM ] && - !nla_get_u8 (data[IFLA_VXLAN_UDP_CSUM ])) - conf.flags |=VXLAN_F_UDP_ZERO_CSUM_TX ; +if (data[IFLA_VXLAN_UDP_CSUM ]) { +if (changelink) +return -EOPNOTSUPP ; +if (!nla_get_u8 (data[IFLA_VXLAN_UDP_CSUM ])) + conf->flags |=VXLAN_F_UDP_ZERO_CSUM_TX ; + } -if (data[IFLA_VXLAN_UDP_ZERO_CSUM6_TX ] && -nla_get_u8 (data[IFLA_VXLAN_UDP_ZERO_CSUM6_TX ])) - conf.flags |=VXLAN_F_UDP_ZERO_CSUM6_TX ; +if (data[IFLA_VXLAN_UDP_ZERO_CSUM6_TX ]) { +if (changelink) +return -EOPNOTSUPP ; +if (nla_get_u8 (data[IFLA_VXLAN_UDP_ZERO_CSUM6_TX ])) + conf->flags |=VXLAN_F_UDP_ZERO_CSUM6_TX ; + } -if (data[IFLA_VXLAN_UDP_ZERO_CSUM6_RX ] && -nla_get_u8 (data[IFLA_VXLAN_UDP_ZERO_CSUM6_RX ])) - conf.flags |=VXLAN_F_UDP_ZERO_CSUM6_RX ; +if (data[IFLA_VXLAN_UDP_ZERO_CSUM6_RX ]) { +if (changelink) +return -EOPNOTSUPP ; +if (nla_get_u8 (data[IFLA_VXLAN_UDP_ZERO_CSUM6_RX ])) + conf->flags |=VXLAN_F_UDP_ZERO_CSUM6_RX ; + } -if (data[IFLA_VXLAN_REMCSUM_TX ] && -nla_get_u8 (data[IFLA_VXLAN_REMCSUM_TX ])) - conf.flags |=VXLAN_F_REMCSUM_TX ; +if (data[IFLA_VXLAN_REMCSUM_TX ]) { +if (changelink) +return -EOPNOTSUPP ; +if (nla_get_u8 (data[IFLA_VXLAN_REMCSUM_TX ])) + conf->flags |=VXLAN_F_REMCSUM_TX ; + } -if (data[IFLA_VXLAN_REMCSUM_RX ] && -nla_get_u8 (data[IFLA_VXLAN_REMCSUM_RX ])) - conf.flags |=VXLAN_F_REMCSUM_RX ; +if (data[IFLA_VXLAN_REMCSUM_RX ]) { +if (changelink) +return -EOPNOTSUPP ; +if (nla_get_u8 (data[IFLA_VXLAN_REMCSUM_RX ])) + conf->flags |=VXLAN_F_REMCSUM_RX ; + } + +if (data[IFLA_VXLAN_GBP ]) { +if (changelink) +return -EOPNOTSUPP ; + conf->flags |=VXLAN_F_GBP ; + } + +if (data[IFLA_VXLAN_GPE ]) { +if (changelink) +return -EOPNOTSUPP ; + conf->flags |=VXLAN_F_GPE ; + } + +if (data[IFLA_VXLAN_REMCSUM_NOPARTIAL ]) { +if (changelink) +return -EOPNOTSUPP ; + conf->flags |=VXLAN_F_REMCSUM_NOPARTIAL ; + } + +if (tb[IFLA_MTU ]) { +if (changelink) +return -EOPNOTSUPP ; + conf->mtu =nla_get_u32 (tb[IFLA_MTU ]); + } + +return 0 ; +} + +static int vxlan_newlink (struct net *src_net,struct net_device *dev, +struct nlattr *tb[],struct nlattr *data[]) +{ +struct vxlan_net *vn =net_generic (src_net, vxlan_net_id); +struct vxlan_dev *vxlan =netdev_priv (dev); +struct vxlan_config conf; +int err; + + err =vxlan_nl2conf (tb, data, dev, &conf, false); +if (err) +return err; + + err =vxlan_dev_configure (src_net, dev, &conf, false); +if (err) +return err; + + dev->ethtool_ops = &vxlan_ethtool_ops; + +/* create an fdb entry for a valid default destination */ +if (!vxlan_addr_any (&vxlan->default_dst.remote_ip)) { + err =vxlan_fdb_create (vxlan, all_zeros_mac, + &vxlan->default_dst.remote_ip, +NUD_REACHABLE |NUD_PERMANENT , +NLM_F_EXCL |NLM_F_CREATE , + vxlan->cfg.dst_port, + vxlan->default_dst.remote_vni, + vxlan->default_dst.remote_vni, + vxlan->default_dst.remote_ifindex, +NTF_SELF ); +if (err) +return err; + } + + err =register_netdevice (dev); +if (err) { +vxlan_fdb_delete_default (vxlan, vxlan->default_dst.remote_vni); +return err; + } + +list_add (&vxlan->next, &vn->vxlan_list); + +return 0 ; +} -if (data[IFLA_VXLAN_GBP ]) - conf.flags |=VXLAN_F_GBP ; +static int vxlan_changelink (struct net_device *dev,struct nlattr *tb[], +struct nlattr *data[]) +{ +struct vxlan_dev *vxlan =netdev_priv (dev); +struct vxlan_rdst *dst = &vxlan->default_dst; +struct vxlan_rdst old_dst; +struct vxlan_config conf; +int err; + + err =vxlan_nl2conf (tb, data, + dev, &conf, true); +if (err) +return err; -if (data[IFLA_VXLAN_GPE ]) - conf.flags |=VXLAN_F_GPE ; +memcpy (&old_dst, dst,sizeof (struct vxlan_rdst )); -if (data[IFLA_VXLAN_REMCSUM_NOPARTIAL ]) - conf.flags |=VXLAN_F_REMCSUM_NOPARTIAL ; + err =vxlan_dev_configure (vxlan->net, dev, &conf, true); +if (err) +return err; -if (tb[IFLA_MTU ]) - conf.mtu =nla_get_u32 (tb[IFLA_MTU ]); +/* handle default dst entry */ +if (!vxlan_addr_equal (&dst->remote_ip, &old_dst.remote_ip)) { +spin_lock_bh (&vxlan->hash_lock); +if (!vxlan_addr_any (&old_dst.remote_ip)) +__vxlan_fdb_delete (vxlan, all_zeros_mac, + old_dst.remote_ip, + vxlan->cfg.dst_port, + old_dst.remote_vni, + old_dst.remote_vni, + old_dst.remote_ifindex,0 ); + +if (!vxlan_addr_any (&dst->remote_ip)) { + err =vxlan_fdb_create (vxlan, all_zeros_mac, + &dst->remote_ip, +NUD_REACHABLE |NUD_PERMANENT , +NLM_F_CREATE |NLM_F_APPEND , + vxlan->cfg.dst_port, + dst->remote_vni, + dst->remote_vni, + dst->remote_ifindex, +NTF_SELF ); +if (err) { +spin_unlock_bh (&vxlan->hash_lock); +return err; + } + } +spin_unlock_bh (&vxlan->hash_lock); + } -return vxlan_dev_configure (src_net, dev, &conf); +return 0 ; }static void vxlan_dellink (struct net_device *dev,struct list_head *head)
@@ -3261,6 +3417,7 @@ static struct net *vxlan_get_link_net(const struct net_device *dev) .setup = vxlan_setup, .validate = vxlan_validate, .newlink = vxlan_newlink, + .changelink = vxlan_changelink, .dellink = vxlan_dellink, .get_size = vxlan_get_size, .fill_info = vxlan_fill_info,
@@ -3282,7 +3439,7 @@ struct net_device *vxlan_dev_create(struct net *net, const char *name,if (IS_ERR (dev))return dev; - err =vxlan_dev_configure (net, dev, conf); + err =vxlan_dev_configure (net, dev, conf, false);if (err <0 ) {free_netdev (dev);return ERR_PTR (err);
Signature · 1 line
1.7.10.4
From: Roopa Prabhu <[email protected]>
Date: Mon, 20 Feb 2017 08:29:19 -0800
Quoted 15 lines
From: Roopa Prabhu <[email protected]>
This patch adds changelink rtnl op support for vxlan netdevs. code changes involve:
- refactorvxlan_newlinkintovxlan_nl2confto be used byvxlan_newlinkandvxlan_changelink
-vxlan_nl2confandvxlan_dev_configuretake a changelink argument to isolate changelink checks and updates.
- Allow changing only a few attributes:
- return-EOPNOTSUPPfor attributes that cannot be changed for now. Incremental patches can make the non-supported one available in the future if needed.
Quoted 1 line
Signed-off-by Roopa Prabhu <[email protected]>
Applied.