Linux Ethernet Bridging - Jan 2020 - [Bridge] [PATCH net-next 0/4] net: bridge: add per-vlan state option

Hi,
This set adds the first per-vlan option - state, which uses the new vlan
infrastructure that was recently added. It gives us forwarding control on
per-vlan basis. The first 3 patches prepare the vlan code to support option
dumping and modification. We still compress vlan ranges which have equal
options, each new option will have to add its own equality check to
br_vlan_opts_eq(). The vlans are created in forwarding state by default to
be backwards compatible and vlan state is considered only when the port
state is forwarding (more info in patch 4).
I'll send the selftest for the vlan state with the iproute2 patch-set.

Thanks,
 Nik

Nikolay Aleksandrov (4):
  net: bridge: check port state before br_allowed_egress
  net: bridge: vlan: add basic option dumping support
  net: bridge: vlan: add basic option setting support
  net: bridge: vlan: add per-vlan state

 include/uapi/linux/if_bridge.h |   2 +
 net/bridge/Makefile            |   2 +-
 net/bridge/br_device.c         |   3 +-
 net/bridge/br_forward.c        |   2 +-
 net/bridge/br_input.c          |   7 +-
 net/bridge/br_private.h        |  58 +++++++++++++-
 net/bridge/br_vlan.c           |  99 ++++++++++++++++++-----
 net/bridge/br_vlan_options.c   | 142 +++++++++++++++++++++++++++++++++
 8 files changed, 287 insertions(+), 28 deletions(-)
 create mode 100644 net/bridge/br_vlan_options.c

-- 
2.21.0

If we make sure that br_allowed_egress is called only when we have
BR_STATE_FORWARDING state then we can avoid a test later when we add
per-vlan state.

Signed-off-by: Nikolay Aleksandrov <nikolay at cumulusnetworks.com>
---
 net/bridge/br_forward.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/bridge/br_forward.c b/net/bridge/br_forward.c
index 86637000f275..7629b63f6f30 100644
--- a/net/bridge/br_forward.c
+++ b/net/bridge/br_forward.c
@@ -25,7 +25,7 @@ static inline int should_deliver(const struct net_bridge_port
*p,
 
 	vg = nbp_vlan_group_rcu(p);
 	return ((p->flags & BR_HAIRPIN_MODE) || skb->dev != p->dev)
&&
-		br_allowed_egress(vg, skb) && p->state == BR_STATE_FORWARDING
&&
+		p->state == BR_STATE_FORWARDING && br_allowed_egress(vg, skb)
&&
 		nbp_switchdev_allowed_egress(p, skb) &&
 		!br_skb_isolated(p, skb);
 }
-- 
2.21.0

We'll be dumping the options for the whole range if they're equal. The
first range vlan will be used to extract the options. The commit doesn't
change anything yet it just adds the skeleton for the support. The dump
will happen when the first option is added.

Signed-off-by: Nikolay Aleksandrov <nikolay at cumulusnetworks.com>
---
 net/bridge/Makefile          |  2 +-
 net/bridge/br_private.h      |  8 ++++++++
 net/bridge/br_vlan.c         | 20 ++++++++++++++------
 net/bridge/br_vlan_options.c | 25 +++++++++++++++++++++++++
 4 files changed, 48 insertions(+), 7 deletions(-)
 create mode 100644 net/bridge/br_vlan_options.c

diff --git a/net/bridge/Makefile b/net/bridge/Makefile
index ac9ef337f0fa..49da7ae6f077 100644
--- a/net/bridge/Makefile
+++ b/net/bridge/Makefile
@@ -20,7 +20,7 @@ obj-$(CONFIG_BRIDGE_NETFILTER) += br_netfilter.o
 
 bridge-$(CONFIG_BRIDGE_IGMP_SNOOPING) += br_multicast.o br_mdb.o
 
-bridge-$(CONFIG_BRIDGE_VLAN_FILTERING) += br_vlan.o br_vlan_tunnel.o
+bridge-$(CONFIG_BRIDGE_VLAN_FILTERING) += br_vlan.o br_vlan_tunnel.o
br_vlan_options.o
 
 bridge-$(CONFIG_NET_SWITCHDEV) += br_switchdev.o
 
diff --git a/net/bridge/br_private.h b/net/bridge/br_private.h
index a6226ff2f0cc..403df71d2cfa 100644
--- a/net/bridge/br_private.h
+++ b/net/bridge/br_private.h
@@ -1191,6 +1191,14 @@ static inline void br_vlan_notify(const struct net_bridge
*br,
 }
 #endif
 
+/* br_vlan_options.c */
+#ifdef CONFIG_BRIDGE_VLAN_FILTERING
+bool br_vlan_opts_eq(const struct net_bridge_vlan *v1,
+		     const struct net_bridge_vlan *v2);
+bool br_vlan_opts_fill(struct sk_buff *skb, const struct net_bridge_vlan *v);
+size_t br_vlan_opts_nl_size(void);
+#endif
+
 struct nf_br_ops {
 	int (*br_dev_xmit_hook)(struct sk_buff *skb);
 };
diff --git a/net/bridge/br_vlan.c b/net/bridge/br_vlan.c
index e4f7dd10c3f8..75ec3da92b0b 100644
--- a/net/bridge/br_vlan.c
+++ b/net/bridge/br_vlan.c
@@ -1547,7 +1547,9 @@ void br_vlan_port_event(struct net_bridge_port *p,
unsigned long event)
 	}
 }
 
+/* v_opts is used to dump the options which must be equal in the whole range */
 static bool br_vlan_fill_vids(struct sk_buff *skb, u16 vid, u16 vid_range,
+			      const struct net_bridge_vlan *v_opts,
 			      u16 flags)
 {
 	struct bridge_vlan_info info;
@@ -1572,6 +1574,9 @@ static bool br_vlan_fill_vids(struct sk_buff *skb, u16
vid, u16 vid_range,
 	    nla_put_u16(skb, BRIDGE_VLANDB_ENTRY_RANGE, vid_range))
 		goto out_err;
 
+	if (v_opts && !br_vlan_opts_fill(skb, v_opts))
+		goto out_err;
+
 	nla_nest_end(skb, nest);
 
 	return true;
@@ -1586,7 +1591,8 @@ static size_t rtnl_vlan_nlmsg_size(void)
 	return NLMSG_ALIGN(sizeof(struct br_vlan_msg))
 		+ nla_total_size(0) /* BRIDGE_VLANDB_ENTRY */
 		+ nla_total_size(sizeof(u16)) /* BRIDGE_VLANDB_ENTRY_RANGE */
-		+ nla_total_size(sizeof(struct bridge_vlan_info)); /*
BRIDGE_VLANDB_ENTRY_INFO */
+		+ nla_total_size(sizeof(struct bridge_vlan_info)) /* BRIDGE_VLANDB_ENTRY_INFO
*/
+		+ br_vlan_opts_nl_size(); /* bridge vlan options */
 }
 
 void br_vlan_notify(const struct net_bridge *br,
@@ -1595,7 +1601,7 @@ void br_vlan_notify(const struct net_bridge *br,
 		    int cmd)
 {
 	struct net_bridge_vlan_group *vg;
-	struct net_bridge_vlan *v;
+	struct net_bridge_vlan *v = NULL;
 	struct br_vlan_msg *bvm;
 	struct nlmsghdr *nlh;
 	struct sk_buff *skb;
@@ -1647,7 +1653,7 @@ void br_vlan_notify(const struct net_bridge *br,
 		goto out_kfree;
 	}
 
-	if (!br_vlan_fill_vids(skb, vid, vid_range, flags))
+	if (!br_vlan_fill_vids(skb, vid, vid_range, v, flags))
 		goto out_err;
 
 	nlmsg_end(skb, nlh);
@@ -1665,7 +1671,8 @@ static bool br_vlan_can_enter_range(const struct
net_bridge_vlan *v_curr,
 				    const struct net_bridge_vlan *range_end)
 {
 	return v_curr->vid - range_end->vid == 1 &&
-	       range_end->flags == v_curr->flags;
+	       range_end->flags == v_curr->flags &&
+	       br_vlan_opts_eq(v_curr, range_end);
 }
 
 static int br_vlan_dump_dev(const struct net_device *dev,
@@ -1729,7 +1736,8 @@ static int br_vlan_dump_dev(const struct net_device *dev,
 			u16 flags = br_vlan_flags(range_start, pvid);
 
 			if (!br_vlan_fill_vids(skb, range_start->vid,
-					       range_end->vid, flags)) {
+					       range_end->vid, range_start,
+					       flags)) {
 				err = -EMSGSIZE;
 				break;
 			}
@@ -1748,7 +1756,7 @@ static int br_vlan_dump_dev(const struct net_device *dev,
 	 */
 	if (!err && range_start &&
 	    !br_vlan_fill_vids(skb, range_start->vid, range_end->vid,
-			       br_vlan_flags(range_start, pvid)))
+			       range_start, br_vlan_flags(range_start, pvid)))
 		err = -EMSGSIZE;
 
 	cb->args[1] = err ? idx : 0;
diff --git a/net/bridge/br_vlan_options.c b/net/bridge/br_vlan_options.c
new file mode 100644
index 000000000000..55fcdc9c380c
--- /dev/null
+++ b/net/bridge/br_vlan_options.c
@@ -0,0 +1,25 @@
+// SPDX-License-Identifier: GPL-2.0-only
+// Copyright (c) 2020, Nikolay Aleksandrov <nikolay at
cumulusnetworks.com>
+#include <linux/kernel.h>
+#include <linux/netdevice.h>
+#include <linux/rtnetlink.h>
+#include <linux/slab.h>
+
+#include "br_private.h"
+
+/* check if the options between two vlans are equal */
+bool br_vlan_opts_eq(const struct net_bridge_vlan *v1,
+		     const struct net_bridge_vlan *v2)
+{
+	return true;
+}
+
+bool br_vlan_opts_fill(struct sk_buff *skb, const struct net_bridge_vlan *v)
+{
+	return true;
+}
+
+size_t br_vlan_opts_nl_size(void)
+{
+	return 0;
+}
-- 
2.21.0

This patch adds support for option modification of single vlans and
ranges. It allows to only modify options, i.e. skip create/delete by
using the BRIDGE_VLAN_INFO_ONLY_OPTS flag.

Signed-off-by: Nikolay Aleksandrov <nikolay at cumulusnetworks.com>
---
 include/uapi/linux/if_bridge.h |  1 +
 net/bridge/br_private.h        |  7 ++++
 net/bridge/br_vlan.c           | 32 ++++++++++++++--
 net/bridge/br_vlan_options.c   | 68 ++++++++++++++++++++++++++++++++++
 4 files changed, 105 insertions(+), 3 deletions(-)

diff --git a/include/uapi/linux/if_bridge.h b/include/uapi/linux/if_bridge.h
index ac38f0b674b8..06bbfefa2141 100644
--- a/include/uapi/linux/if_bridge.h
+++ b/include/uapi/linux/if_bridge.h
@@ -130,6 +130,7 @@ enum {
 #define BRIDGE_VLAN_INFO_RANGE_BEGIN	(1<<3) /* VLAN is start of vlan
range */
 #define BRIDGE_VLAN_INFO_RANGE_END	(1<<4) /* VLAN is end of vlan range */
 #define BRIDGE_VLAN_INFO_BRENTRY	(1<<5) /* Global bridge VLAN entry */
+#define BRIDGE_VLAN_INFO_ONLY_OPTS	(1<<6) /* skip create/delete */
 
 struct bridge_vlan_info {
 	__u16 flags;
diff --git a/net/bridge/br_private.h b/net/bridge/br_private.h
index 403df71d2cfa..8ca8a9258510 100644
--- a/net/bridge/br_private.h
+++ b/net/bridge/br_private.h
@@ -1197,6 +1197,13 @@ bool br_vlan_opts_eq(const struct net_bridge_vlan *v1,
 		     const struct net_bridge_vlan *v2);
 bool br_vlan_opts_fill(struct sk_buff *skb, const struct net_bridge_vlan *v);
 size_t br_vlan_opts_nl_size(void);
+int br_vlan_process_options(const struct net_bridge *br,
+			    const struct net_bridge_port *p,
+			    struct net_bridge_vlan *range_start,
+			    struct net_bridge_vlan *range_end,
+			    struct nlattr **tb,
+			    bool *changed,
+			    struct netlink_ext_ack *extack);
 #endif
 
 struct nf_br_ops {
diff --git a/net/bridge/br_vlan.c b/net/bridge/br_vlan.c
index 75ec3da92b0b..468d1a861c66 100644
--- a/net/bridge/br_vlan.c
+++ b/net/bridge/br_vlan.c
@@ -1824,11 +1824,11 @@ static int br_vlan_rtm_process_one(struct net_device
*dev,
 {
 	struct bridge_vlan_info *vinfo, vrange_end, *vinfo_last = NULL;
 	struct nlattr *tb[BRIDGE_VLANDB_ENTRY_MAX + 1];
+	bool changed = false, skip_processing = false;
 	struct net_bridge_vlan_group *vg;
 	struct net_bridge_port *p = NULL;
 	int err = 0, cmdmap = 0;
 	struct net_bridge *br;
-	bool changed = false;
 
 	if (netif_is_bridge_master(dev)) {
 		br = netdev_priv(dev);
@@ -1882,14 +1882,40 @@ static int br_vlan_rtm_process_one(struct net_device
*dev,
 	switch (cmd) {
 	case RTM_NEWVLAN:
 		cmdmap = RTM_SETLINK;
+		skip_processing = !!(vinfo->flags & BRIDGE_VLAN_INFO_ONLY_OPTS);
 		break;
 	case RTM_DELVLAN:
 		cmdmap = RTM_DELLINK;
 		break;
 	}
 
-	err = br_process_vlan_info(br, p, cmdmap, vinfo, &vinfo_last,
&changed,
-				   extack);
+	if (!skip_processing) {
+		struct bridge_vlan_info *tmp_last = vinfo_last;
+
+		/* br_process_vlan_info may overwrite vinfo_last */
+		err = br_process_vlan_info(br, p, cmdmap, vinfo, &tmp_last,
+					   &changed, extack);
+		if (err)
+			goto out;
+	}
+
+	/* deal with options */
+	if (cmd == RTM_NEWVLAN) {
+		struct net_bridge_vlan *range_start, *range_end;
+
+		if (vinfo_last) {
+			range_start = br_vlan_find(vg, vinfo_last->vid);
+			range_end = br_vlan_find(vg, vinfo->vid);
+		} else {
+			range_start = br_vlan_find(vg, vinfo->vid);
+			range_end = range_start;
+		}
+
+		err = br_vlan_process_options(br, p, range_start, range_end,
+					      tb, &changed, extack);
+	}
+
+out:
 	if (changed)
 		br_ifinfo_notify(cmdmap, br, p);
 
diff --git a/net/bridge/br_vlan_options.c b/net/bridge/br_vlan_options.c
index 55fcdc9c380c..1c76a1ba9a8c 100644
--- a/net/bridge/br_vlan_options.c
+++ b/net/bridge/br_vlan_options.c
@@ -23,3 +23,71 @@ size_t br_vlan_opts_nl_size(void)
 {
 	return 0;
 }
+
+static int br_vlan_process_one_opts(const struct net_bridge *br,
+				    const struct net_bridge_port *p,
+				    struct net_bridge_vlan *v,
+				    struct nlattr **tb,
+				    bool *changed,
+				    struct netlink_ext_ack *extack)
+{
+	return 0;
+}
+
+int br_vlan_process_options(const struct net_bridge *br,
+			    const struct net_bridge_port *p,
+			    struct net_bridge_vlan *range_start,
+			    struct net_bridge_vlan *range_end,
+			    struct nlattr **tb,
+			    bool *changed,
+			    struct netlink_ext_ack *extack)
+{
+	int vid, err = 0, v_change_start = 0;
+	struct net_bridge_vlan_group *vg;
+	struct net_bridge_vlan *v;
+
+	if (p)
+		vg = nbp_vlan_group(p);
+	else
+		vg = br_vlan_group(br);
+
+	/* we require at least range_start to process options */
+	if (!range_start || !range_end) {
+		NL_SET_ERR_MSG_MOD(extack, "Vlan doesn't exist, can't process
options");
+		return -ENOENT;
+	}
+
+	for (vid = range_start->vid; vid <= range_end->vid; vid++) {
+		bool curr_change = false;
+
+		v = br_vlan_find(vg, vid);
+		if (!v) {
+			NL_SET_ERR_MSG_MOD(extack, "Vlan doesn't exist, can't process
options");
+			err = -ENOENT;
+			break;
+		}
+
+		err = br_vlan_process_one_opts(br, p, v, tb, &curr_change,
+					       extack);
+		if (err)
+			break;
+
+		if (curr_change) {
+			*changed = curr_change;
+			if (!v_change_start)
+				v_change_start = vid;
+		} else {
+			/* nothing to notify yet */
+			if (!v_change_start)
+				continue;
+			br_vlan_notify(br, p, v_change_start, vid - 1,
+				       RTM_NEWVLAN);
+			v_change_start = 0;
+		}
+	}
+	/* v_change_start is set only if the last/whole range changed */
+	if (v_change_start)
+		br_vlan_notify(br, p, v_change_start, vid - 1, RTM_NEWVLAN);
+
+	return err;
+}
-- 
2.21.0

The first per-vlan option added is state, it is needed for EVPN and for
per-vlan STP. The state allows to control the forwarding on per-vlan
basis. The vlan state is considered only if the port state is forwarding
in order to avoid conflicts and be consistent. br_allowed_egress is
called only when the state is forwarding, but the ingress case is a bit
more complicated due to the fact that we may have the transition between
port:BR_STATE_FORWARDING -> vlan:BR_STATE_LEARNING which should still
allow the bridge to learn from the packet after vlan filtering and it will
be dropped after that. Also to optimize the pvid state check we keep a
copy in the vlan group to avoid one lookup. The state members are
modified with *_ONCE() to annotate the lockless access.

Signed-off-by: Nikolay Aleksandrov <nikolay at cumulusnetworks.com>
---
 include/uapi/linux/if_bridge.h |  1 +
 net/bridge/br_device.c         |  3 +-
 net/bridge/br_input.c          |  7 +++--
 net/bridge/br_private.h        | 43 ++++++++++++++++++++++++--
 net/bridge/br_vlan.c           | 47 +++++++++++++++++++++--------
 net/bridge/br_vlan_options.c   | 55 ++++++++++++++++++++++++++++++++--
 6 files changed, 136 insertions(+), 20 deletions(-)

diff --git a/include/uapi/linux/if_bridge.h b/include/uapi/linux/if_bridge.h
index 06bbfefa2141..e07d082d8a79 100644
--- a/include/uapi/linux/if_bridge.h
+++ b/include/uapi/linux/if_bridge.h
@@ -191,6 +191,7 @@ enum {
 	BRIDGE_VLANDB_ENTRY_UNSPEC,
 	BRIDGE_VLANDB_ENTRY_INFO,
 	BRIDGE_VLANDB_ENTRY_RANGE,
+	BRIDGE_VLANDB_ENTRY_STATE,
 	__BRIDGE_VLANDB_ENTRY_MAX,
 };
 #define BRIDGE_VLANDB_ENTRY_MAX (__BRIDGE_VLANDB_ENTRY_MAX - 1)
diff --git a/net/bridge/br_device.c b/net/bridge/br_device.c
index fb38add21b37..dc3d2c1dd9d5 100644
--- a/net/bridge/br_device.c
+++ b/net/bridge/br_device.c
@@ -32,6 +32,7 @@ netdev_tx_t br_dev_xmit(struct sk_buff *skb, struct net_device
*dev)
 	struct net_bridge_mdb_entry *mdst;
 	struct pcpu_sw_netstats *brstats = this_cpu_ptr(br->stats);
 	const struct nf_br_ops *nf_ops;
+	u8 state = BR_STATE_FORWARDING;
 	const unsigned char *dest;
 	struct ethhdr *eth;
 	u16 vid = 0;
@@ -56,7 +57,7 @@ netdev_tx_t br_dev_xmit(struct sk_buff *skb, struct net_device
*dev)
 	eth = eth_hdr(skb);
 	skb_pull(skb, ETH_HLEN);
 
-	if (!br_allowed_ingress(br, br_vlan_group_rcu(br), skb, &vid))
+	if (!br_allowed_ingress(br, br_vlan_group_rcu(br), skb, &vid, &state))
 		goto out;
 
 	if (IS_ENABLED(CONFIG_INET) &&
diff --git a/net/bridge/br_input.c b/net/bridge/br_input.c
index 8944ceb47fe9..fcc260840028 100644
--- a/net/bridge/br_input.c
+++ b/net/bridge/br_input.c
@@ -76,11 +76,14 @@ int br_handle_frame_finish(struct net *net, struct sock *sk,
struct sk_buff *skb
 	bool local_rcv, mcast_hit = false;
 	struct net_bridge *br;
 	u16 vid = 0;
+	u8 state;
 
 	if (!p || p->state == BR_STATE_DISABLED)
 		goto drop;
 
-	if (!br_allowed_ingress(p->br, nbp_vlan_group_rcu(p), skb, &vid))
+	state = p->state;
+	if (!br_allowed_ingress(p->br, nbp_vlan_group_rcu(p), skb, &vid,
+				&state))
 		goto out;
 
 	nbp_switchdev_frame_mark(p, skb);
@@ -103,7 +106,7 @@ int br_handle_frame_finish(struct net *net, struct sock *sk,
struct sk_buff *skb
 		}
 	}
 
-	if (p->state == BR_STATE_LEARNING)
+	if (state == BR_STATE_LEARNING)
 		goto drop;
 
 	BR_INPUT_SKB_CB(skb)->brdev = br->dev;
diff --git a/net/bridge/br_private.h b/net/bridge/br_private.h
index 8ca8a9258510..c1d6fe3a381f 100644
--- a/net/bridge/br_private.h
+++ b/net/bridge/br_private.h
@@ -113,6 +113,7 @@ enum {
  * @vid: VLAN id
  * @flags: bridge vlan flags
  * @priv_flags: private (in-kernel) bridge vlan flags
+ * @state: STP state (e.g. blocking, learning, forwarding)
  * @stats: per-cpu VLAN statistics
  * @br: if MASTER flag set, this points to a bridge struct
  * @port: if MASTER flag unset, this points to a port struct
@@ -133,6 +134,7 @@ struct net_bridge_vlan {
 	u16				vid;
 	u16				flags;
 	u16				priv_flags;
+	u8				state;
 	struct br_vlan_stats __percpu	*stats;
 	union {
 		struct net_bridge	*br;
@@ -157,6 +159,7 @@ struct net_bridge_vlan {
  * @vlan_list: sorted VLAN entry list
  * @num_vlans: number of total VLAN entries
  * @pvid: PVID VLAN id
+ * @pvid_state: PVID's STP state (e.g. forwarding, learning, blocking)
  *
  * IMPORTANT: Be careful when checking if there're VLAN entries using list
  *            primitives because the bridge can have entries in its list which
@@ -170,6 +173,7 @@ struct net_bridge_vlan_group {
 	struct list_head		vlan_list;
 	u16				num_vlans;
 	u16				pvid;
+	u8				pvid_state;
 };
 
 /* bridge fdb flags */
@@ -935,7 +939,7 @@ static inline int br_multicast_igmp_type(const struct
sk_buff *skb)
 #ifdef CONFIG_BRIDGE_VLAN_FILTERING
 bool br_allowed_ingress(const struct net_bridge *br,
 			struct net_bridge_vlan_group *vg, struct sk_buff *skb,
-			u16 *vid);
+			u16 *vid, u8 *state);
 bool br_allowed_egress(struct net_bridge_vlan_group *vg,
 		       const struct sk_buff *skb);
 bool br_should_learn(struct net_bridge_port *p, struct sk_buff *skb, u16 *vid);
@@ -1035,7 +1039,7 @@ static inline u16 br_vlan_flags(const struct
net_bridge_vlan *v, u16 pvid)
 static inline bool br_allowed_ingress(const struct net_bridge *br,
 				      struct net_bridge_vlan_group *vg,
 				      struct sk_buff *skb,
-				      u16 *vid)
+				      u16 *vid, u8 *state)
 {
 	return true;
 }
@@ -1204,6 +1208,41 @@ int br_vlan_process_options(const struct net_bridge *br,
 			    struct nlattr **tb,
 			    bool *changed,
 			    struct netlink_ext_ack *extack);
+
+/* vlan state manipulation helpers using *_ONCE to annotate lockless access */
+static inline u8 br_vlan_get_state(const struct net_bridge_vlan *v)
+{
+	return READ_ONCE(v->state);
+}
+
+static inline void br_vlan_set_state(struct net_bridge_vlan *v, u8 state)
+{
+	WRITE_ONCE(v->state, state);
+}
+
+static inline u8 br_vlan_get_pvid_state(const struct net_bridge_vlan_group *vg)
+{
+	return READ_ONCE(vg->pvid_state);
+}
+
+static inline void br_vlan_set_pvid_state(struct net_bridge_vlan_group *vg,
+					  u8 state)
+{
+	WRITE_ONCE(vg->pvid_state, state);
+}
+
+/* learn_allow is true at ingress and false at egress */
+static inline bool br_vlan_state_allowed(u8 state, bool learn_allow)
+{
+	switch (state) {
+	case BR_STATE_LEARNING:
+		return learn_allow;
+	case BR_STATE_FORWARDING:
+		return true;
+	default:
+		return false;
+	}
+}
 #endif
 
 struct nf_br_ops {
diff --git a/net/bridge/br_vlan.c b/net/bridge/br_vlan.c
index 468d1a861c66..3d97efcb4fd8 100644
--- a/net/bridge/br_vlan.c
+++ b/net/bridge/br_vlan.c
@@ -34,13 +34,15 @@ static struct net_bridge_vlan *br_vlan_lookup(struct
rhashtable *tbl, u16 vid)
 	return rhashtable_lookup_fast(tbl, &vid, br_vlan_rht_params);
 }
 
-static bool __vlan_add_pvid(struct net_bridge_vlan_group *vg, u16 vid)
+static bool __vlan_add_pvid(struct net_bridge_vlan_group *vg,
+			    const struct net_bridge_vlan *v)
 {
-	if (vg->pvid == vid)
+	if (vg->pvid == v->vid)
 		return false;
 
 	smp_wmb();
-	vg->pvid = vid;
+	br_vlan_set_pvid_state(vg, v->state);
+	vg->pvid = v->vid;
 
 	return true;
 }
@@ -69,7 +71,7 @@ static bool __vlan_add_flags(struct net_bridge_vlan *v, u16
flags)
 		vg = nbp_vlan_group(v->port);
 
 	if (flags & BRIDGE_VLAN_INFO_PVID)
-		ret = __vlan_add_pvid(vg, v->vid);
+		ret = __vlan_add_pvid(vg, v);
 	else
 		ret = __vlan_delete_pvid(vg, v->vid);
 
@@ -293,6 +295,9 @@ static int __vlan_add(struct net_bridge_vlan *v, u16 flags,
 		vg->num_vlans++;
 	}
 
+	/* set the state before publishing */
+	v->state = BR_STATE_FORWARDING;
+
 	err = rhashtable_lookup_insert_fast(&vg->vlan_hash, &v->vnode,
 					    br_vlan_rht_params);
 	if (err)
@@ -466,7 +471,8 @@ struct sk_buff *br_handle_vlan(struct net_bridge *br,
 /* Called under RCU */
 static bool __allowed_ingress(const struct net_bridge *br,
 			      struct net_bridge_vlan_group *vg,
-			      struct sk_buff *skb, u16 *vid)
+			      struct sk_buff *skb, u16 *vid,
+			      u8 *state)
 {
 	struct br_vlan_stats *stats;
 	struct net_bridge_vlan *v;
@@ -532,13 +538,25 @@ static bool __allowed_ingress(const struct net_bridge *br,
 			skb->vlan_tci |= pvid;
 
 		/* if stats are disabled we can avoid the lookup */
-		if (!br_opt_get(br, BROPT_VLAN_STATS_ENABLED))
-			return true;
+		if (!br_opt_get(br, BROPT_VLAN_STATS_ENABLED)) {
+			if (*state == BR_STATE_FORWARDING) {
+				*state = br_vlan_get_pvid_state(vg);
+				return br_vlan_state_allowed(*state, true);
+			} else {
+				return true;
+			}
+		}
 	}
 	v = br_vlan_find(vg, *vid);
 	if (!v || !br_vlan_should_use(v))
 		goto drop;
 
+	if (*state == BR_STATE_FORWARDING) {
+		*state = br_vlan_get_state(v);
+		if (!br_vlan_state_allowed(*state, true))
+			goto drop;
+	}
+
 	if (br_opt_get(br, BROPT_VLAN_STATS_ENABLED)) {
 		stats = this_cpu_ptr(v->stats);
 		u64_stats_update_begin(&stats->syncp);
@@ -556,7 +574,7 @@ static bool __allowed_ingress(const struct net_bridge *br,
 
 bool br_allowed_ingress(const struct net_bridge *br,
 			struct net_bridge_vlan_group *vg, struct sk_buff *skb,
-			u16 *vid)
+			u16 *vid, u8 *state)
 {
 	/* If VLAN filtering is disabled on the bridge, all packets are
 	 * permitted.
@@ -566,7 +584,7 @@ bool br_allowed_ingress(const struct net_bridge *br,
 		return true;
 	}
 
-	return __allowed_ingress(br, vg, skb, vid);
+	return __allowed_ingress(br, vg, skb, vid, state);
 }
 
 /* Called under RCU. */
@@ -582,7 +600,8 @@ bool br_allowed_egress(struct net_bridge_vlan_group *vg,
 
 	br_vlan_get_tag(skb, &vid);
 	v = br_vlan_find(vg, vid);
-	if (v && br_vlan_should_use(v))
+	if (v && br_vlan_should_use(v) &&
+	    br_vlan_state_allowed(br_vlan_get_state(v), false))
 		return true;
 
 	return false;
@@ -593,6 +612,7 @@ bool br_should_learn(struct net_bridge_port *p, struct
sk_buff *skb, u16 *vid)
 {
 	struct net_bridge_vlan_group *vg;
 	struct net_bridge *br = p->br;
+	struct net_bridge_vlan *v;
 
 	/* If filtering was disabled at input, let it pass. */
 	if (!br_opt_get(br, BROPT_VLAN_ENABLED))
@@ -607,13 +627,15 @@ bool br_should_learn(struct net_bridge_port *p, struct
sk_buff *skb, u16 *vid)
 
 	if (!*vid) {
 		*vid = br_get_pvid(vg);
-		if (!*vid)
+		if (!*vid ||
+		    !br_vlan_state_allowed(br_vlan_get_pvid_state(vg), true))
 			return false;
 
 		return true;
 	}
 
-	if (br_vlan_find(vg, *vid))
+	v = br_vlan_find(vg, *vid);
+	if (v && br_vlan_state_allowed(br_vlan_get_state(v), true))
 		return true;
 
 	return false;
@@ -1816,6 +1838,7 @@ static const struct nla_policy
br_vlan_db_policy[BRIDGE_VLANDB_ENTRY_MAX + 1]  	[BRIDGE_VLANDB_ENTRY_INFO]	= {
.type = NLA_EXACT_LEN,
 					    .len = sizeof(struct bridge_vlan_info) },
 	[BRIDGE_VLANDB_ENTRY_RANGE]	= { .type = NLA_U16 },
+	[BRIDGE_VLANDB_ENTRY_STATE]	= { .type = NLA_U8 },
 };
 
 static int br_vlan_rtm_process_one(struct net_device *dev,
diff --git a/net/bridge/br_vlan_options.c b/net/bridge/br_vlan_options.c
index 1c76a1ba9a8c..699dad08e44a 100644
--- a/net/bridge/br_vlan_options.c
+++ b/net/bridge/br_vlan_options.c
@@ -11,26 +11,75 @@
 bool br_vlan_opts_eq(const struct net_bridge_vlan *v1,
 		     const struct net_bridge_vlan *v2)
 {
-	return true;
+	return v1->state == v2->state;
 }
 
 bool br_vlan_opts_fill(struct sk_buff *skb, const struct net_bridge_vlan *v)
 {
-	return true;
+	return !nla_put_u8(skb, BRIDGE_VLANDB_ENTRY_STATE,
+			   br_vlan_get_state(v));
 }
 
 size_t br_vlan_opts_nl_size(void)
 {
+	return nla_total_size(sizeof(u8)); /* BRIDGE_VLANDB_ENTRY_STATE */
+}
+
+static int br_vlan_modify_state(struct net_bridge_vlan_group *vg,
+				struct net_bridge_vlan *v,
+				u8 state,
+				bool *changed,
+				struct netlink_ext_ack *extack)
+{
+	struct net_bridge *br;
+
+	ASSERT_RTNL();
+
+	if (state > BR_STATE_BLOCKING) {
+		NL_SET_ERR_MSG_MOD(extack, "Invalid vlan state");
+		return -EINVAL;
+	}
+
+	if (br_vlan_is_brentry(v))
+		br = v->br;
+	else
+		br = v->port->br;
+
+	if (br->stp_enabled == BR_KERNEL_STP) {
+		NL_SET_ERR_MSG_MOD(extack, "Can't modify vlan state when using
kernel STP");
+		return -EBUSY;
+	}
+
+	if (v->state == state)
+		return 0;
+
+	if (v->vid == br_get_pvid(vg))
+		br_vlan_set_pvid_state(vg, state);
+
+	br_vlan_set_state(v, state);
+	*changed = true;
+
 	return 0;
 }
 
 static int br_vlan_process_one_opts(const struct net_bridge *br,
 				    const struct net_bridge_port *p,
+				    struct net_bridge_vlan_group *vg,
 				    struct net_bridge_vlan *v,
 				    struct nlattr **tb,
 				    bool *changed,
 				    struct netlink_ext_ack *extack)
 {
+	int err;
+
+	if (tb[BRIDGE_VLANDB_ENTRY_STATE]) {
+		u8 state = nla_get_u8(tb[BRIDGE_VLANDB_ENTRY_STATE]);
+
+		err = br_vlan_modify_state(vg, v, state, changed, extack);
+		if (err)
+			return err;
+	}
+
 	return 0;
 }
 
@@ -67,7 +116,7 @@ int br_vlan_process_options(const struct net_bridge *br,
 			break;
 		}
 
-		err = br_vlan_process_one_opts(br, p, v, tb, &curr_change,
+		err = br_vlan_process_one_opts(br, p, vg, v, tb, &curr_change,
 					       extack);
 		if (err)
 			break;
-- 
2.21.0

On 1/23/20 3:28 PM, Nikolay Aleksandrov wrote:
> Hi,
> This set adds the first per-vlan option - state, which uses the new vlan
> infrastructure that was recently added. It gives us forwarding control on
> per-vlan basis. The first 3 patches prepare the vlan code to support option
> dumping and modification. We still compress vlan ranges which have equal
> options, each new option will have to add its own equality check to
> br_vlan_opts_eq(). The vlans are created in forwarding state by default to
> be backwards compatible and vlan state is considered only when the port
> state is forwarding (more info in patch 4).
> I'll send the selftest for the vlan state with the iproute2 patch-set.
> 
> Thanks,
>   Nik
> 
> Nikolay Aleksandrov (4):
>    net: bridge: check port state before br_allowed_egress
>    net: bridge: vlan: add basic option dumping support
>    net: bridge: vlan: add basic option setting support
>    net: bridge: vlan: add per-vlan state
> 
>   include/uapi/linux/if_bridge.h |   2 +
>   net/bridge/Makefile            |   2 +-
>   net/bridge/br_device.c         |   3 +-
>   net/bridge/br_forward.c        |   2 +-
>   net/bridge/br_input.c          |   7 +-
>   net/bridge/br_private.h        |  58 +++++++++++++-
>   net/bridge/br_vlan.c           |  99 ++++++++++++++++++-----
>   net/bridge/br_vlan_options.c   | 142 +++++++++++++++++++++++++++++++++
>   8 files changed, 287 insertions(+), 28 deletions(-)
>   create mode 100644 net/bridge/br_vlan_options.c
> 
Hm, actually I just noticed that we send port notifications even if only
changing options
which should be avoided since now we have per-vlan notifications. We should be
sending
only those when changing options without create/delete.
I'll make that change and send v2 after some tests tomorrow.

Cheers,
  Nik