#include <net/netns/generic.h>
#include <net/tun_proto.h>
#include <net/vxlan.h>
+#include <net/nexthop.h>
#if IS_ENABLED(CONFIG_IPV6)
#include <net/ip6_tunnel.h>
u16 state; /* see ndm_state */
__be32 vni;
u16 flags; /* see ndm_flags and below */
+ struct list_head nh_list;
+ struct nexthop __rcu *nh;
};
#define NTF_VXLAN_ADDED_BY_USER 0x100
*/
static inline struct vxlan_rdst *first_remote_rcu(struct vxlan_fdb *fdb)
{
+ if (rcu_access_pointer(fdb->nh))
+ return NULL;
return list_entry_rcu(fdb->remotes.next, struct vxlan_rdst, list);
}
static inline struct vxlan_rdst *first_remote_rtnl(struct vxlan_fdb *fdb)
{
+ if (rcu_access_pointer(fdb->nh))
+ return NULL;
return list_first_entry(&fdb->remotes, struct vxlan_rdst, list);
}
{
unsigned long now = jiffies;
struct nda_cacheinfo ci;
+ bool send_ip, send_eth;
struct nlmsghdr *nlh;
+ struct nexthop *nh;
struct ndmsg *ndm;
- bool send_ip, send_eth;
nlh = nlmsg_put(skb, portid, seq, type, sizeof(*ndm), flags);
if (nlh == NULL)
send_eth = send_ip = true;
+ nh = rcu_dereference_rtnl(fdb->nh);
if (type == RTM_GETNEIGH) {
- send_ip = !vxlan_addr_any(&rdst->remote_ip);
+ if (rdst) {
+ send_ip = !vxlan_addr_any(&rdst->remote_ip);
+ ndm->ndm_family = send_ip ? rdst->remote_ip.sa.sa_family : AF_INET;
+ } else if (nh) {
+ ndm->ndm_family = nexthop_get_family(nh);
+ }
send_eth = !is_zero_ether_addr(fdb->eth_addr);
- ndm->ndm_family = send_ip ? rdst->remote_ip.sa.sa_family : AF_INET;
} else
ndm->ndm_family = AF_BRIDGE;
ndm->ndm_state = fdb->state;
ndm->ndm_ifindex = vxlan->dev->ifindex;
ndm->ndm_flags = fdb->flags;
- if (rdst->offloaded)
+ if (rdst && rdst->offloaded)
ndm->ndm_flags |= NTF_OFFLOADED;
ndm->ndm_type = RTN_UNICAST;
if (send_eth && nla_put(skb, NDA_LLADDR, ETH_ALEN, &fdb->eth_addr))
goto nla_put_failure;
+ if (nh) {
+ if (nla_put_u32(skb, NDA_NH_ID, nh->id))
+ goto nla_put_failure;
+ } else if (rdst) {
+ if (send_ip && vxlan_nla_put_addr(skb, NDA_DST,
+ &rdst->remote_ip))
+ goto nla_put_failure;
+
+ if (rdst->remote_port &&
+ rdst->remote_port != vxlan->cfg.dst_port &&
+ nla_put_be16(skb, NDA_PORT, rdst->remote_port))
+ goto nla_put_failure;
+ if (rdst->remote_vni != vxlan->default_dst.remote_vni &&
+ nla_put_u32(skb, NDA_VNI, be32_to_cpu(rdst->remote_vni)))
+ goto nla_put_failure;
+ if (rdst->remote_ifindex &&
+ nla_put_u32(skb, NDA_IFINDEX, rdst->remote_ifindex))
+ goto nla_put_failure;
+ }
- if (send_ip && vxlan_nla_put_addr(skb, NDA_DST, &rdst->remote_ip))
- goto nla_put_failure;
-
- if (rdst->remote_port && rdst->remote_port != vxlan->cfg.dst_port &&
- nla_put_be16(skb, NDA_PORT, rdst->remote_port))
- goto nla_put_failure;
- if (rdst->remote_vni != vxlan->default_dst.remote_vni &&
- nla_put_u32(skb, NDA_VNI, be32_to_cpu(rdst->remote_vni)))
- goto nla_put_failure;
if ((vxlan->cfg.flags & VXLAN_F_COLLECT_METADATA) && fdb->vni &&
nla_put_u32(skb, NDA_SRC_VNI,
be32_to_cpu(fdb->vni)))
goto nla_put_failure;
- if (rdst->remote_ifindex &&
- nla_put_u32(skb, NDA_IFINDEX, rdst->remote_ifindex))
- goto nla_put_failure;
ci.ndm_used = jiffies_to_clock_t(now - fdb->used);
ci.ndm_confirmed = 0;
{
int err;
- if (swdev_notify) {
+ if (swdev_notify && rd) {
switch (type) {
case RTM_NEWNEIGH:
err = vxlan_fdb_switchdev_call_notifiers(vxlan, fdb, rd,
f->flags = ndm_flags;
f->updated = f->used = jiffies;
f->vni = src_vni;
+ f->nh = NULL;
+ INIT_LIST_HEAD(&f->nh_list);
INIT_LIST_HEAD(&f->remotes);
memcpy(f->eth_addr, mac, ETH_ALEN);
vxlan_fdb_head(vxlan, mac, src_vni));
}
+static int vxlan_fdb_nh_update(struct vxlan_dev *vxlan, struct vxlan_fdb *fdb,
+ u32 nhid, struct netlink_ext_ack *extack)
+{
+ struct nexthop *old_nh = rtnl_dereference(fdb->nh);
+ struct nh_group *nhg;
+ struct nexthop *nh;
+ int err = -EINVAL;
+
+ if (old_nh && old_nh->id == nhid)
+ return 0;
+
+ nh = nexthop_find_by_id(vxlan->net, nhid);
+ if (!nh) {
+ NL_SET_ERR_MSG(extack, "Nexthop id does not exist");
+ goto err_inval;
+ }
+
+ if (nh) {
+ if (!nexthop_get(nh)) {
+ NL_SET_ERR_MSG(extack, "Nexthop has been deleted");
+ nh = NULL;
+ goto err_inval;
+ }
+ if (!nh->is_fdb_nh) {
+ NL_SET_ERR_MSG(extack, "Nexthop is not a fdb nexthop");
+ goto err_inval;
+ }
+
+ if (!nh->is_group || !nh->nh_grp->mpath) {
+ NL_SET_ERR_MSG(extack, "Nexthop is not a multipath group");
+ goto err_inval;
+ }
+
+ /* check nexthop group family */
+ nhg = rtnl_dereference(nh->nh_grp);
+ switch (vxlan->default_dst.remote_ip.sa.sa_family) {
+ case AF_INET:
+ if (!nhg->has_v4) {
+ err = -EAFNOSUPPORT;
+ NL_SET_ERR_MSG(extack, "Nexthop group family not supported");
+ goto err_inval;
+ }
+ break;
+ case AF_INET6:
+ if (nhg->has_v4) {
+ err = -EAFNOSUPPORT;
+ NL_SET_ERR_MSG(extack, "Nexthop group family not supported");
+ goto err_inval;
+ }
+ }
+ }
+
+ if (old_nh) {
+ list_del_rcu(&fdb->nh_list);
+ nexthop_put(old_nh);
+ }
+ rcu_assign_pointer(fdb->nh, nh);
+ list_add_tail_rcu(&fdb->nh_list, &nh->fdb_list);
+ return 1;
+
+err_inval:
+ if (nh)
+ nexthop_put(nh);
+ return err;
+}
+
static int vxlan_fdb_create(struct vxlan_dev *vxlan,
const u8 *mac, union vxlan_addr *ip,
__u16 state, __be16 port, __be32 src_vni,
__be32 vni, __u32 ifindex, __u16 ndm_flags,
- struct vxlan_fdb **fdb)
+ u32 nhid, struct vxlan_fdb **fdb,
+ struct netlink_ext_ack *extack)
{
struct vxlan_rdst *rd = NULL;
struct vxlan_fdb *f;
if (!f)
return -ENOMEM;
- rc = vxlan_fdb_append(f, ip, port, vni, ifindex, &rd);
- if (rc < 0) {
- kfree(f);
- return rc;
- }
+ if (nhid)
+ rc = vxlan_fdb_nh_update(vxlan, f, nhid, extack);
+ else
+ rc = vxlan_fdb_append(f, ip, port, vni, ifindex, &rd);
+ if (rc < 0)
+ goto errout;
*fdb = f;
return 0;
+
+errout:
+ kfree(f);
+ return rc;
}
static void __vxlan_fdb_free(struct vxlan_fdb *f)
{
struct vxlan_rdst *rd, *nd;
+ struct nexthop *nh;
+
+ nh = rcu_dereference_raw(f->nh);
+ if (nh) {
+ rcu_assign_pointer(f->nh, NULL);
+ list_del_rcu(&f->nh_list);
+ nexthop_put(nh);
+ }
list_for_each_entry_safe(rd, nd, &f->remotes, list) {
dst_cache_destroy(&rd->dst_cache);
netdev_dbg(vxlan->dev, "delete %pM\n", f->eth_addr);
--vxlan->addrcnt;
- if (do_notify)
- list_for_each_entry(rd, &f->remotes, list)
- vxlan_fdb_notify(vxlan, f, rd, RTM_DELNEIGH,
+ if (do_notify) {
+ if (rcu_access_pointer(f->nh))
+ vxlan_fdb_notify(vxlan, f, NULL, RTM_DELNEIGH,
swdev_notify, NULL);
+ else
+ list_for_each_entry(rd, &f->remotes, list)
+ vxlan_fdb_notify(vxlan, f, rd, RTM_DELNEIGH,
+ swdev_notify, NULL);
+ }
hlist_del_rcu(&f->hlist);
call_rcu(&f->rcu, vxlan_fdb_free);
__u16 state, __u16 flags,
__be16 port, __be32 vni,
__u32 ifindex, __u16 ndm_flags,
- struct vxlan_fdb *f,
+ struct vxlan_fdb *f, u32 nhid,
bool swdev_notify,
struct netlink_ext_ack *extack)
{
int rc = 0;
int err;
+ if (nhid && !rcu_access_pointer(f->nh)) {
+ NL_SET_ERR_MSG(extack,
+ "Cannot replace an existing non nexthop fdb with a nexthop");
+ return -EOPNOTSUPP;
+ }
+
+ if (nhid && (flags & NLM_F_APPEND)) {
+ NL_SET_ERR_MSG(extack,
+ "Cannot append to a nexthop fdb");
+ return -EOPNOTSUPP;
+ }
+
/* Do not allow an externally learned entry to take over an entry added
* by the user.
*/
/* Only change unicasts */
if (!(is_multicast_ether_addr(f->eth_addr) ||
is_zero_ether_addr(f->eth_addr))) {
- rc = vxlan_fdb_replace(f, ip, port, vni,
- ifindex, &oldrd);
+ if (nhid) {
+ rc = vxlan_fdb_nh_update(vxlan, f, nhid, extack);
+ if (rc < 0)
+ return rc;
+ } else {
+ rc = vxlan_fdb_replace(f, ip, port, vni,
+ ifindex, &oldrd);
+ }
notify |= rc;
} else {
+ NL_SET_ERR_MSG(extack, "Cannot replace non-unicast fdb entries");
return -EOPNOTSUPP;
}
}
return 0;
err_notify:
+ if (nhid)
+ return err;
if ((flags & NLM_F_REPLACE) && rc)
*rd = oldrd;
else if ((flags & NLM_F_APPEND) && rc) {
const u8 *mac, union vxlan_addr *ip,
__u16 state, __u16 flags,
__be16 port, __be32 src_vni, __be32 vni,
- __u32 ifindex, __u16 ndm_flags,
+ __u32 ifindex, __u16 ndm_flags, u32 nhid,
bool swdev_notify,
struct netlink_ext_ack *extack)
{
netdev_dbg(vxlan->dev, "add %pM -> %pIS\n", mac, ip);
rc = vxlan_fdb_create(vxlan, mac, ip, state, port, src_vni,
- vni, ifindex, fdb_flags, &f);
+ vni, ifindex, fdb_flags, nhid, &f, extack);
if (rc < 0)
return rc;
const u8 *mac, union vxlan_addr *ip,
__u16 state, __u16 flags,
__be16 port, __be32 src_vni, __be32 vni,
- __u32 ifindex, __u16 ndm_flags,
+ __u32 ifindex, __u16 ndm_flags, u32 nhid,
bool swdev_notify,
struct netlink_ext_ack *extack)
{
return vxlan_fdb_update_existing(vxlan, ip, state, flags, port,
vni, ifindex, ndm_flags, f,
- swdev_notify, extack);
+ nhid, swdev_notify, extack);
} else {
if (!(flags & NLM_F_CREATE))
return -ENOENT;
return vxlan_fdb_update_create(vxlan, mac, ip, state, flags,
port, src_vni, vni, ifindex,
- ndm_flags, swdev_notify, extack);
+ ndm_flags, nhid, swdev_notify,
+ extack);
}
}
static int vxlan_fdb_parse(struct nlattr *tb[], struct vxlan_dev *vxlan,
union vxlan_addr *ip, __be16 *port, __be32 *src_vni,
- __be32 *vni, u32 *ifindex)
+ __be32 *vni, u32 *ifindex, u32 *nhid)
{
struct net *net = dev_net(vxlan->dev);
int err;
*ifindex = 0;
}
+ if (tb[NDA_NH_ID])
+ *nhid = nla_get_u32(tb[NDA_NH_ID]);
+ else
+ *nhid = 0;
+
return 0;
}
union vxlan_addr ip;
__be16 port;
__be32 src_vni, vni;
- u32 ifindex;
+ u32 ifindex, nhid;
u32 hash_index;
int err;
return -EINVAL;
}
- if (tb[NDA_DST] == NULL)
+ if (!tb || (!tb[NDA_DST] && !tb[NDA_NH_ID]))
return -EINVAL;
- err = vxlan_fdb_parse(tb, vxlan, &ip, &port, &src_vni, &vni, &ifindex);
+ err = vxlan_fdb_parse(tb, vxlan, &ip, &port, &src_vni, &vni, &ifindex,
+ &nhid);
if (err)
return err;
err = vxlan_fdb_update(vxlan, addr, &ip, ndm->ndm_state, flags,
port, src_vni, vni, ifindex,
ndm->ndm_flags | NTF_VXLAN_ADDED_BY_USER,
- true, extack);
+ nhid, true, extack);
spin_unlock_bh(&vxlan->hash_lock[hash_index]);
return err;
__be16 port, __be32 src_vni, __be32 vni,
u32 ifindex, bool swdev_notify)
{
- struct vxlan_fdb *f;
struct vxlan_rdst *rd = NULL;
+ struct vxlan_fdb *f;
int err = -ENOENT;
f = vxlan_find_mac(vxlan, addr, src_vni);
struct vxlan_dev *vxlan = netdev_priv(dev);
union vxlan_addr ip;
__be32 src_vni, vni;
- __be16 port;
- u32 ifindex;
+ u32 ifindex, nhid;
u32 hash_index;
+ __be16 port;
int err;
- err = vxlan_fdb_parse(tb, vxlan, &ip, &port, &src_vni, &vni, &ifindex);
+ err = vxlan_fdb_parse(tb, vxlan, &ip, &port, &src_vni, &vni, &ifindex,
+ &nhid);
if (err)
return err;
hlist_for_each_entry_rcu(f, &vxlan->fdb_head[h], hlist) {
struct vxlan_rdst *rd;
+ if (rcu_access_pointer(f->nh)) {
+ err = vxlan_fdb_info(skb, vxlan, f,
+ NETLINK_CB(cb->skb).portid,
+ cb->nlh->nlmsg_seq,
+ RTM_NEWNEIGH,
+ NLM_F_MULTI, NULL);
+ if (err < 0)
+ goto out;
+ continue;
+ }
+
list_for_each_entry_rcu(rd, &f->remotes, list) {
if (*idx < cb->args[2])
goto skip;
if (f->state & (NUD_PERMANENT | NUD_NOARP))
return true;
+ /* Don't override an fdb with nexthop with a learnt entry */
+ if (rcu_access_pointer(f->nh))
+ return true;
+
if (net_ratelimit())
netdev_info(dev,
"%pM migrated from %pIS to %pIS\n",
vxlan->cfg.dst_port,
vni,
vxlan->default_dst.remote_vni,
- ifindex, NTF_SELF, true, NULL);
+ ifindex, NTF_SELF, 0, true, NULL);
spin_unlock(&vxlan->hash_lock[hash_index]);
}
kfree_skb(skb);
}
+static void vxlan_xmit_nh(struct sk_buff *skb, struct net_device *dev,
+ struct vxlan_fdb *f, __be32 vni, bool did_rsc)
+{
+ struct vxlan_rdst nh_rdst;
+ struct nexthop *nh;
+ bool do_xmit;
+ u32 hash;
+
+ memset(&nh_rdst, 0, sizeof(struct vxlan_rdst));
+ hash = skb_get_hash(skb);
+
+ rcu_read_lock();
+ nh = rcu_dereference(f->nh);
+ if (!nh) {
+ rcu_read_unlock();
+ goto drop;
+ }
+ do_xmit = vxlan_fdb_nh_path_select(nh, hash, &nh_rdst);
+ rcu_read_unlock();
+
+ if (likely(do_xmit))
+ vxlan_xmit_one(skb, dev, vni, &nh_rdst, did_rsc);
+ else
+ goto drop;
+
+ return;
+
+drop:
+ dev->stats.tx_dropped++;
+ dev_kfree_skb(skb);
+}
+
/* Transmit local packets over Vxlan
*
* Outer IP header inherits ECN and DF from inner header.
}
}
- list_for_each_entry_rcu(rdst, &f->remotes, list) {
- struct sk_buff *skb1;
+ if (rcu_access_pointer(f->nh)) {
+ vxlan_xmit_nh(skb, dev, f,
+ (vni ? : vxlan->default_dst.remote_vni), did_rsc);
+ } else {
+ list_for_each_entry_rcu(rdst, &f->remotes, list) {
+ struct sk_buff *skb1;
- if (!fdst) {
- fdst = rdst;
- continue;
+ if (!fdst) {
+ fdst = rdst;
+ continue;
+ }
+ skb1 = skb_clone(skb, GFP_ATOMIC);
+ if (skb1)
+ vxlan_xmit_one(skb1, dev, vni, rdst, did_rsc);
}
- skb1 = skb_clone(skb, GFP_ATOMIC);
- if (skb1)
- vxlan_xmit_one(skb1, dev, vni, rdst, did_rsc);
+ if (fdst)
+ vxlan_xmit_one(skb, dev, vni, fdst, did_rsc);
+ else
+ kfree_skb(skb);
}
- if (fdst)
- vxlan_xmit_one(skb, dev, vni, fdst, did_rsc);
- else
- kfree_skb(skb);
return NETDEV_TX_OK;
}
dst->remote_vni,
dst->remote_vni,
dst->remote_ifindex,
- NTF_SELF, &f);
+ NTF_SELF, 0, &f, extack);
if (err)
return err;
}
vxlan->cfg.dst_port,
conf.vni, conf.vni,
conf.remote_ifindex,
- NTF_SELF, true, extack);
+ NTF_SELF, 0, true, extack);
if (err) {
spin_unlock_bh(&vxlan->hash_lock[hash_index]);
netdev_adjacent_change_abort(dst->remote_dev,
fdb_info->remote_vni,
fdb_info->remote_ifindex,
NTF_USE | NTF_SELF | NTF_EXT_LEARNED,
- false, extack);
+ 0, false, extack);
spin_unlock_bh(&vxlan->hash_lock[hash_index]);
return err;
projects / users / hch / configfs.git / commitdiff