Commit f9c4bb0b authored by Roopa Prabhu's avatar Roopa Prabhu Committed by David S. Miller

vxlan: vni filtering support on collect metadata device

This patch adds vnifiltering support to collect metadata device.

Motivation:
You can only use a single vxlan collect metadata device for a given
vxlan udp port in the system today. The vxlan collect metadata device
terminates all received vxlan packets. As shown in the below diagram,
there are use-cases where you need to support multiple such vxlan devices in
independent bridge domains. Each vxlan device must terminate the vni's
it is configured for.
Example usecase: In a service provider network a service provider
typically supports multiple bridge domains with overlapping vlans.
One bridge domain per customer. Vlans in each bridge domain are
mapped to globally unique vxlan ranges assigned to each customer.

vnifiltering support in collect metadata devices terminates only configured
vnis. This is similar to vlan filtering in bridge driver. The vni filtering
capability is provided by a new flag on collect metadata device.

In the below pic:
	- customer1 is mapped to br1 bridge domain
	- customer2 is mapped to br2 bridge domain
	- customer1 vlan 10-11 is mapped to vni 1001-1002
	- customer2 vlan 10-11 is mapped to vni 2001-2002
	- br1 and br2 are vlan filtering bridges
	- vxlan1 and vxlan2 are collect metadata devices with
	  vnifiltering enabled

┌──────────────────────────────────────────────────────────────────┐
│  switch                                                          │
│                                                                  │
│         ┌───────────┐                 ┌───────────┐              │
│         │           │                 │           │              │
│         │   br1     │                 │   br2     │              │
│         └┬─────────┬┘                 └──┬───────┬┘              │
│     vlans│         │               vlans │       │               │
│     10,11│         │                10,11│       │               │
│          │     vlanvnimap:               │    vlanvnimap:        │
│          │       10-1001,11-1002         │      10-2001,11-2002  │
│          │         │                     │       │               │
│   ┌──────┴┐     ┌──┴─────────┐       ┌───┴────┐  │               │
│   │ swp1  │     │vxlan1      │       │ swp2   │ ┌┴─────────────┐ │
│   │       │     │  vnifilter:│       │        │ │vxlan2        │ │
│   └───┬───┘     │   1001,1002│       └───┬────┘ │ vnifilter:   │ │
│       │         └────────────┘           │      │  2001,2002   │ │
│       │                                  │      └──────────────┘ │
│       │                                  │                       │
└───────┼──────────────────────────────────┼───────────────────────┘
        │                                  │
        │                                  │
  ┌─────┴───────┐                          │
  │  customer1  │                    ┌─────┴──────┐
  │ host/VM     │                    │customer2   │
  └─────────────┘                    │ host/VM    │
                                     └────────────┘

With this implementation, vxlan dst metadata device can
be associated with range of vnis.
struct vxlan_vni_node is introduced to represent
a configured vni. We start with vni and its
associated remote_ip in this structure. This
structure can be extended to bring in other
per vni attributes if there are usecases for it.
A vni inherits an attribute from the base vxlan device
if there is no per vni attributes defined.

struct vxlan_dev gets a new rhashtable for
vnis called vxlan_vni_group. vxlan_vnifilter.c
implements the necessary netlink api, notifications
and helper functions to process and manage lifecycle
of vxlan_vni_node.

This patch also adds new helper functions in vxlan_multicast.c
to handle per vni remote_ip multicast groups which are part
of vxlan_vni_group.

Fix build problems:
Reported-by: default avatarkernel test robot <lkp@intel.com>
Signed-off-by: default avatarRoopa Prabhu <roopa@nvidia.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
parent a498c595
......@@ -4,4 +4,4 @@
obj-$(CONFIG_VXLAN) += vxlan.o
vxlan-objs := vxlan_core.o vxlan_multicast.o
vxlan-objs := vxlan_core.o vxlan_multicast.o vxlan_vnifilter.o
......@@ -145,12 +145,19 @@ static struct vxlan_dev *vxlan_vs_find_vni(struct vxlan_sock *vs, int ifindex,
struct vxlan_dev_node *node;
/* For flow based devices, map all packets to VNI 0 */
if (vs->flags & VXLAN_F_COLLECT_METADATA)
if (vs->flags & VXLAN_F_COLLECT_METADATA &&
!(vs->flags & VXLAN_F_VNIFILTER))
vni = 0;
hlist_for_each_entry_rcu(node, vni_head(vs, vni), hlist) {
if (node->vxlan->default_dst.remote_vni != vni)
if (!node->vxlan)
continue;
if (node->vxlan->cfg.flags & VXLAN_F_VNIFILTER) {
if (!vxlan_vnifilter_lookup(node->vxlan, vni))
continue;
} else if (node->vxlan->default_dst.remote_vni != vni) {
continue;
}
if (IS_ENABLED(CONFIG_IPV6)) {
const struct vxlan_config *cfg = &node->vxlan->cfg;
......@@ -1478,6 +1485,9 @@ static void vxlan_sock_release(struct vxlan_dev *vxlan)
RCU_INIT_POINTER(vxlan->vn4_sock, NULL);
synchronize_net();
if (vxlan->cfg.flags & VXLAN_F_VNIFILTER)
vxlan_vs_del_vnigrp(vxlan);
else
vxlan_vs_del_dev(vxlan);
if (__vxlan_sock_release_prep(sock4)) {
......@@ -2857,6 +2867,9 @@ static int vxlan_init(struct net_device *dev)
struct vxlan_dev *vxlan = netdev_priv(dev);
int err;
if (vxlan->cfg.flags & VXLAN_F_VNIFILTER)
vxlan_vnigroup_init(vxlan);
dev->tstats = netdev_alloc_pcpu_stats(struct pcpu_sw_netstats);
if (!dev->tstats)
return -ENOMEM;
......@@ -2886,6 +2899,9 @@ static void vxlan_uninit(struct net_device *dev)
{
struct vxlan_dev *vxlan = netdev_priv(dev);
if (vxlan->cfg.flags & VXLAN_F_VNIFILTER)
vxlan_vnigroup_uninit(vxlan);
gro_cells_destroy(&vxlan->gro_cells);
vxlan_fdb_delete_default(vxlan, vxlan->cfg.vni);
......@@ -2903,16 +2919,11 @@ static int vxlan_open(struct net_device *dev)
if (ret < 0)
return ret;
if (vxlan_addr_multicast(&vxlan->default_dst.remote_ip)) {
ret = vxlan_igmp_join(vxlan, &vxlan->default_dst.remote_ip,
vxlan->default_dst.remote_ifindex);
if (ret == -EADDRINUSE)
ret = 0;
ret = vxlan_multicast_join(vxlan);
if (ret) {
vxlan_sock_release(vxlan);
return ret;
}
}
if (vxlan->cfg.age_interval)
mod_timer(&vxlan->age_timer, jiffies + FDB_AGE_INTERVAL);
......@@ -2948,13 +2959,9 @@ static void vxlan_flush(struct vxlan_dev *vxlan, bool do_all)
static int vxlan_stop(struct net_device *dev)
{
struct vxlan_dev *vxlan = netdev_priv(dev);
struct vxlan_net *vn = net_generic(vxlan->net, vxlan_net_id);
int ret = 0;
if (vxlan_addr_multicast(&vxlan->default_dst.remote_ip) &&
!vxlan_group_used(vn, vxlan, NULL, 0))
ret = vxlan_igmp_leave(vxlan, &vxlan->default_dst.remote_ip,
vxlan->default_dst.remote_ifindex);
vxlan_multicast_leave(vxlan);
del_timer_sync(&vxlan->age_timer);
......@@ -3184,6 +3191,7 @@ static const struct nla_policy vxlan_policy[IFLA_VXLAN_MAX + 1] = {
[IFLA_VXLAN_REMCSUM_NOPARTIAL] = { .type = NLA_FLAG },
[IFLA_VXLAN_TTL_INHERIT] = { .type = NLA_FLAG },
[IFLA_VXLAN_DF] = { .type = NLA_U8 },
[IFLA_VXLAN_VNIFILTER] = { .type = NLA_U8 },
};
static int vxlan_validate(struct nlattr *tb[], struct nlattr *data[],
......@@ -3369,6 +3377,7 @@ static struct vxlan_sock *vxlan_socket_create(struct net *net, bool ipv6,
static int __vxlan_sock_add(struct vxlan_dev *vxlan, bool ipv6)
{
struct vxlan_net *vn = net_generic(vxlan->net, vxlan_net_id);
bool metadata = vxlan->cfg.flags & VXLAN_F_COLLECT_METADATA;
struct vxlan_sock *vs = NULL;
struct vxlan_dev_node *node;
int l3mdev_index = 0;
......@@ -3404,7 +3413,12 @@ static int __vxlan_sock_add(struct vxlan_dev *vxlan, bool ipv6)
rcu_assign_pointer(vxlan->vn4_sock, vs);
node = &vxlan->hlist4;
}
if (metadata && (vxlan->cfg.flags & VXLAN_F_VNIFILTER))
vxlan_vs_add_vnigrp(vxlan, vs, ipv6);
else
vxlan_vs_add_dev(vs, vxlan, node);
return 0;
}
......@@ -3431,7 +3445,7 @@ static int vxlan_sock_add(struct vxlan_dev *vxlan)
return ret;
}
static int vxlan_vni_in_use(struct net *src_net, struct vxlan_dev *vxlan,
int vxlan_vni_in_use(struct net *src_net, struct vxlan_dev *vxlan,
struct vxlan_config *conf, __be32 vni)
{
struct vxlan_net *vn = net_generic(src_net, vxlan_net_id);
......@@ -3440,8 +3454,12 @@ static int vxlan_vni_in_use(struct net *src_net, struct vxlan_dev *vxlan,
list_for_each_entry(tmp, &vn->vxlan_list, next) {
if (tmp == vxlan)
continue;
if (tmp->cfg.vni != vni)
if (tmp->cfg.flags & VXLAN_F_VNIFILTER) {
if (!vxlan_vnifilter_lookup(tmp, vni))
continue;
} else if (tmp->cfg.vni != vni) {
continue;
}
if (tmp->cfg.dst_port != conf->dst_port)
continue;
if ((tmp->cfg.flags & (VXLAN_F_RCV_FLAGS | VXLAN_F_IPV6)) !=
......@@ -4051,6 +4069,21 @@ static int vxlan_nl2conf(struct nlattr *tb[], struct nlattr *data[],
if (data[IFLA_VXLAN_DF])
conf->df = nla_get_u8(data[IFLA_VXLAN_DF]);
if (data[IFLA_VXLAN_VNIFILTER]) {
err = vxlan_nl2flag(conf, data, IFLA_VXLAN_VNIFILTER,
VXLAN_F_VNIFILTER, changelink, false,
extack);
if (err)
return err;
if ((conf->flags & VXLAN_F_VNIFILTER) &&
!(conf->flags & VXLAN_F_COLLECT_METADATA)) {
NL_SET_ERR_MSG_ATTR(extack, data[IFLA_VXLAN_VNIFILTER],
"vxlan vnifilter only valid in collect metadata mode");
return -EINVAL;
}
}
return 0;
}
......@@ -4126,6 +4159,19 @@ static int vxlan_changelink(struct net_device *dev, struct nlattr *tb[],
dst->remote_ifindex,
true);
spin_unlock_bh(&vxlan->hash_lock[hash_index]);
/* If vni filtering device, also update fdb entries of
* all vnis that were using default remote ip
*/
if (vxlan->cfg.flags & VXLAN_F_VNIFILTER) {
err = vxlan_vnilist_update_group(vxlan, &dst->remote_ip,
&conf.remote_ip, extack);
if (err) {
netdev_adjacent_change_abort(dst->remote_dev,
lowerdev, dev);
return err;
}
}
}
if (conf.age_interval != vxlan->cfg.age_interval)
......@@ -4271,6 +4317,11 @@ static int vxlan_fill_info(struct sk_buff *skb, const struct net_device *dev)
nla_put_flag(skb, IFLA_VXLAN_REMCSUM_NOPARTIAL))
goto nla_put_failure;
if (vxlan->cfg.flags & VXLAN_F_VNIFILTER &&
nla_put_u8(skb, IFLA_VXLAN_VNIFILTER,
!!(vxlan->cfg.flags & VXLAN_F_VNIFILTER)))
goto nla_put_failure;
return 0;
nla_put_failure:
......@@ -4630,6 +4681,8 @@ static int __init vxlan_init_module(void)
if (rc)
goto out4;
vxlan_vnifilter_init();
return 0;
out4:
unregister_switchdev_notifier(&vxlan_switchdev_notifier_block);
......@@ -4644,6 +4697,7 @@ late_initcall(vxlan_init_module);
static void __exit vxlan_cleanup_module(void)
{
vxlan_vnifilter_uninit();
rtnl_link_unregister(&vxlan_link_ops);
unregister_switchdev_notifier(&vxlan_switchdev_notifier_block);
unregister_netdevice_notifier(&vxlan_notifier_block);
......
......@@ -82,9 +82,48 @@ int vxlan_igmp_leave(struct vxlan_dev *vxlan, union vxlan_addr *rip,
return ret;
}
static bool vxlan_group_used_match(union vxlan_addr *ip, int ifindex,
union vxlan_addr *rip, int rifindex)
{
if (!vxlan_addr_multicast(rip))
return false;
if (!vxlan_addr_equal(rip, ip))
return false;
if (rifindex != ifindex)
return false;
return true;
}
static bool vxlan_group_used_by_vnifilter(struct vxlan_dev *vxlan,
union vxlan_addr *ip, int ifindex)
{
struct vxlan_vni_group *vg = rtnl_dereference(vxlan->vnigrp);
struct vxlan_vni_node *v, *tmp;
if (vxlan_group_used_match(ip, ifindex,
&vxlan->default_dst.remote_ip,
vxlan->default_dst.remote_ifindex))
return true;
list_for_each_entry_safe(v, tmp, &vg->vni_list, vlist) {
if (!vxlan_addr_multicast(&v->remote_ip))
continue;
if (vxlan_group_used_match(ip, ifindex,
&v->remote_ip,
vxlan->default_dst.remote_ifindex))
return true;
}
return false;
}
/* See if multicast group is already in use by other ID */
bool vxlan_group_used(struct vxlan_net *vn, struct vxlan_dev *dev,
union vxlan_addr *rip, int rifindex)
__be32 vni, union vxlan_addr *rip, int rifindex)
{
union vxlan_addr *ip = (rip ? : &dev->default_dst.remote_ip);
int ifindex = (rifindex ? : dev->default_dst.remote_ifindex);
......@@ -121,14 +160,113 @@ bool vxlan_group_used(struct vxlan_net *vn, struct vxlan_dev *dev,
rtnl_dereference(vxlan->vn6_sock) != sock6)
continue;
#endif
if (!vxlan_addr_equal(&vxlan->default_dst.remote_ip, ip))
if (vxlan->cfg.flags & VXLAN_F_VNIFILTER) {
if (!vxlan_group_used_by_vnifilter(vxlan, ip, ifindex))
continue;
if (vxlan->default_dst.remote_ifindex != ifindex)
} else {
if (!vxlan_group_used_match(ip, ifindex,
&vxlan->default_dst.remote_ip,
vxlan->default_dst.remote_ifindex))
continue;
}
return true;
}
return false;
}
static int vxlan_multicast_join_vnigrp(struct vxlan_dev *vxlan)
{
struct vxlan_vni_group *vg = rtnl_dereference(vxlan->vnigrp);
struct vxlan_vni_node *v, *tmp, *vgood = NULL;
int ret = 0;
list_for_each_entry_safe(v, tmp, &vg->vni_list, vlist) {
if (!vxlan_addr_multicast(&v->remote_ip))
continue;
/* skip if address is same as default address */
if (vxlan_addr_equal(&v->remote_ip,
&vxlan->default_dst.remote_ip))
continue;
ret = vxlan_igmp_join(vxlan, &v->remote_ip, 0);
if (ret == -EADDRINUSE)
ret = 0;
if (ret)
goto out;
vgood = v;
}
out:
if (ret) {
list_for_each_entry_safe(v, tmp, &vg->vni_list, vlist) {
if (!vxlan_addr_multicast(&v->remote_ip))
continue;
if (vxlan_addr_equal(&v->remote_ip,
&vxlan->default_dst.remote_ip))
continue;
vxlan_igmp_leave(vxlan, &v->remote_ip, 0);
if (v == vgood)
break;
}
}
return ret;
}
static int vxlan_multicast_leave_vnigrp(struct vxlan_dev *vxlan)
{
struct vxlan_net *vn = net_generic(vxlan->net, vxlan_net_id);
struct vxlan_vni_group *vg = rtnl_dereference(vxlan->vnigrp);
struct vxlan_vni_node *v, *tmp;
int last_err = 0, ret;
list_for_each_entry_safe(v, tmp, &vg->vni_list, vlist) {
if (vxlan_addr_multicast(&v->remote_ip) &&
!vxlan_group_used(vn, vxlan, v->vni, &v->remote_ip,
0)) {
ret = vxlan_igmp_leave(vxlan, &v->remote_ip, 0);
if (ret)
last_err = ret;
}
}
return last_err;
}
int vxlan_multicast_join(struct vxlan_dev *vxlan)
{
int ret = 0;
if (vxlan_addr_multicast(&vxlan->default_dst.remote_ip)) {
ret = vxlan_igmp_join(vxlan, &vxlan->default_dst.remote_ip,
vxlan->default_dst.remote_ifindex);
if (ret == -EADDRINUSE)
ret = 0;
if (ret)
return ret;
}
if (vxlan->cfg.flags & VXLAN_F_VNIFILTER)
return vxlan_multicast_join_vnigrp(vxlan);
return 0;
}
int vxlan_multicast_leave(struct vxlan_dev *vxlan)
{
struct vxlan_net *vn = net_generic(vxlan->net, vxlan_net_id);
int ret = 0;
if (vxlan_addr_multicast(&vxlan->default_dst.remote_ip) &&
!vxlan_group_used(vn, vxlan, 0, NULL, 0)) {
ret = vxlan_igmp_leave(vxlan, &vxlan->default_dst.remote_ip,
vxlan->default_dst.remote_ifindex);
if (ret)
return ret;
}
if (vxlan->cfg.flags & VXLAN_F_VNIFILTER)
return vxlan_multicast_leave_vnigrp(vxlan);
return 0;
}
......@@ -7,8 +7,11 @@
#ifndef _VXLAN_PRIVATE_H
#define _VXLAN_PRIVATE_H
#include <linux/rhashtable.h>
extern unsigned int vxlan_net_id;
extern const u8 all_zeros_mac[ETH_ALEN + 2];
extern const struct rhashtable_params vxlan_vni_rht_params;
#define PORT_HASH_BITS 8
#define PORT_HASH_SIZE (1 << PORT_HASH_BITS)
......@@ -92,6 +95,19 @@ bool vxlan_addr_equal(const union vxlan_addr *a, const union vxlan_addr *b)
#endif
static inline struct vxlan_vni_node *
vxlan_vnifilter_lookup(struct vxlan_dev *vxlan, __be32 vni)
{
struct vxlan_vni_group *vg;
vg = rcu_dereference_rtnl(vxlan->vnigrp);
if (!vg)
return NULL;
return rhashtable_lookup_fast(&vg->vni_hash, &vni,
vxlan_vni_rht_params);
}
/* vxlan_core.c */
int vxlan_fdb_create(struct vxlan_dev *vxlan,
const u8 *mac, union vxlan_addr *ip,
......@@ -111,12 +127,33 @@ int vxlan_fdb_update(struct vxlan_dev *vxlan,
__be16 port, __be32 src_vni, __be32 vni,
__u32 ifindex, __u16 ndm_flags, u32 nhid,
bool swdev_notify, struct netlink_ext_ack *extack);
int vxlan_vni_in_use(struct net *src_net, struct vxlan_dev *vxlan,
struct vxlan_config *conf, __be32 vni);
/* vxlan_vnifilter.c */
int vxlan_vnigroup_init(struct vxlan_dev *vxlan);
void vxlan_vnigroup_uninit(struct vxlan_dev *vxlan);
void vxlan_vnifilter_init(void);
void vxlan_vnifilter_uninit(void);
void vxlan_vs_add_vnigrp(struct vxlan_dev *vxlan,
struct vxlan_sock *vs,
bool ipv6);
void vxlan_vs_del_vnigrp(struct vxlan_dev *vxlan);
int vxlan_vnilist_update_group(struct vxlan_dev *vxlan,
union vxlan_addr *old_remote_ip,
union vxlan_addr *new_remote_ip,
struct netlink_ext_ack *extack);
/* vxlan_multicast.c */
int vxlan_multicast_join(struct vxlan_dev *vxlan);
int vxlan_multicast_leave(struct vxlan_dev *vxlan);
bool vxlan_group_used(struct vxlan_net *vn, struct vxlan_dev *dev,
__be32 vni, union vxlan_addr *rip, int rifindex);
int vxlan_igmp_join(struct vxlan_dev *vxlan, union vxlan_addr *rip,
int rifindex);
int vxlan_igmp_leave(struct vxlan_dev *vxlan, union vxlan_addr *rip,
int rifindex);
bool vxlan_group_used(struct vxlan_net *vn, struct vxlan_dev *dev,
union vxlan_addr *rip, int rifindex);
#endif
This diff is collapsed.
......@@ -232,6 +232,25 @@ struct vxlan_dev_node {
struct vxlan_dev *vxlan;
};
struct vxlan_vni_node {
struct rhash_head vnode;
struct vxlan_dev_node hlist4; /* vni hash table for IPv4 socket */
#if IS_ENABLED(CONFIG_IPV6)
struct vxlan_dev_node hlist6; /* vni hash table for IPv6 socket */
#endif
struct list_head vlist;
__be32 vni;
union vxlan_addr remote_ip; /* default remote ip for this vni */
struct rcu_head rcu;
};
struct vxlan_vni_group {
struct rhashtable vni_hash;
struct list_head vni_list;
u32 num_vnis;
};
/* Pseudo network device */
struct vxlan_dev {
struct vxlan_dev_node hlist4; /* vni hash table for IPv4 socket */
......@@ -254,6 +273,8 @@ struct vxlan_dev {
struct vxlan_config cfg;
struct vxlan_vni_group __rcu *vnigrp;
struct hlist_head fdb_head[FDB_HASH_SIZE];
};
......@@ -274,6 +295,7 @@ struct vxlan_dev {
#define VXLAN_F_GPE 0x4000
#define VXLAN_F_IPV6_LINKLOCAL 0x8000
#define VXLAN_F_TTL_INHERIT 0x10000
#define VXLAN_F_VNIFILTER 0x20000
/* Flags that are used in the receive path. These flags must match in
* order for a socket to be shareable
......@@ -283,7 +305,8 @@ struct vxlan_dev {
VXLAN_F_UDP_ZERO_CSUM6_RX | \
VXLAN_F_REMCSUM_RX | \
VXLAN_F_REMCSUM_NOPARTIAL | \
VXLAN_F_COLLECT_METADATA)
VXLAN_F_COLLECT_METADATA | \
VXLAN_F_VNIFILTER)
/* Flags that can be set together with VXLAN_F_GPE. */
#define VXLAN_F_ALLOWED_GPE (VXLAN_F_GPE | \
......@@ -292,7 +315,8 @@ struct vxlan_dev {
VXLAN_F_UDP_ZERO_CSUM_TX | \
VXLAN_F_UDP_ZERO_CSUM6_TX | \
VXLAN_F_UDP_ZERO_CSUM6_RX | \
VXLAN_F_COLLECT_METADATA)
VXLAN_F_COLLECT_METADATA | \
VXLAN_F_VNIFILTER)
struct net_device *vxlan_dev_create(struct net *net, const char *name,
u8 name_assign_type, struct vxlan_config *conf);
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment