From 5a6d13c86f2fe6304450e907fc1d8d72da82efef Mon Sep 17 00:00:00 2001
From: Daniel Baumann <daniel.baumann@progress-linux.org>
Date: Mon, 8 Apr 2024 18:58:07 +0200
Subject: Adding upstream version 6.1.82.

Signed-off-by: Daniel Baumann <daniel.baumann@progress-linux.org>
---
 net/netfilter/core.c                    |  16 +++++
 net/netfilter/ipset/ip_set_bitmap_gen.h |  14 ++++-
 net/netfilter/ipset/ip_set_core.c       |  39 +++++++++---
 net/netfilter/ipset/ip_set_hash_gen.h   |  19 ++++--
 net/netfilter/ipset/ip_set_list_set.c   |  13 +++-
 net/netfilter/nf_conntrack_core.c       |  13 ++++
 net/netfilter/nf_conntrack_h323_asn1.c  |   4 ++
 net/netfilter/nf_conntrack_proto_sctp.c |   2 +-
 net/netfilter/nf_conntrack_proto_tcp.c  |  45 +++++++++++--
 net/netfilter/nf_flow_table_core.c      |  41 +++++-------
 net/netfilter/nf_log.c                  |   7 ++-
 net/netfilter/nf_tables_api.c           | 108 ++++++++++++++++++--------------
 net/netfilter/nft_compat.c              |  37 +++++++++--
 net/netfilter/nft_ct.c                  |  26 ++++++++
 net/netfilter/nft_flow_offload.c        |  12 ++--
 net/netfilter/nft_set_pipapo.c          | 108 ++++++++++++++++----------------
 net/netfilter/nft_set_pipapo.h          |  18 ++++--
 net/netfilter/nft_set_pipapo_avx2.c     |  19 +++---
 net/netfilter/nft_set_rbtree.c          |   6 +-
 net/netfilter/nft_tunnel.c              |   1 +
 20 files changed, 368 insertions(+), 180 deletions(-)

(limited to 'net/netfilter')

diff --git a/net/netfilter/core.c b/net/netfilter/core.c
index 55a7f72d5..edf920742 100644
--- a/net/netfilter/core.c
+++ b/net/netfilter/core.c
@@ -707,6 +707,22 @@ void nf_conntrack_destroy(struct nf_conntrack *nfct)
 }
 EXPORT_SYMBOL(nf_conntrack_destroy);
 
+void nf_ct_set_closing(struct nf_conntrack *nfct)
+{
+	const struct nf_ct_hook *ct_hook;
+
+	if (!nfct)
+		return;
+
+	rcu_read_lock();
+	ct_hook = rcu_dereference(nf_ct_hook);
+	if (ct_hook)
+		ct_hook->set_closing(nfct);
+
+	rcu_read_unlock();
+}
+EXPORT_SYMBOL_GPL(nf_ct_set_closing);
+
 bool nf_ct_get_tuple_skb(struct nf_conntrack_tuple *dst_tuple,
 			 const struct sk_buff *skb)
 {
diff --git a/net/netfilter/ipset/ip_set_bitmap_gen.h b/net/netfilter/ipset/ip_set_bitmap_gen.h
index 26ab0e961..9523104a9 100644
--- a/net/netfilter/ipset/ip_set_bitmap_gen.h
+++ b/net/netfilter/ipset/ip_set_bitmap_gen.h
@@ -28,6 +28,7 @@
 #define mtype_del		IPSET_TOKEN(MTYPE, _del)
 #define mtype_list		IPSET_TOKEN(MTYPE, _list)
 #define mtype_gc		IPSET_TOKEN(MTYPE, _gc)
+#define mtype_cancel_gc		IPSET_TOKEN(MTYPE, _cancel_gc)
 #define mtype			MTYPE
 
 #define get_ext(set, map, id)	((map)->extensions + ((set)->dsize * (id)))
@@ -57,9 +58,6 @@ mtype_destroy(struct ip_set *set)
 {
 	struct mtype *map = set->data;
 
-	if (SET_WITH_TIMEOUT(set))
-		del_timer_sync(&map->gc);
-
 	if (set->dsize && set->extensions & IPSET_EXT_DESTROY)
 		mtype_ext_cleanup(set);
 	ip_set_free(map->members);
@@ -288,6 +286,15 @@ mtype_gc(struct timer_list *t)
 	add_timer(&map->gc);
 }
 
+static void
+mtype_cancel_gc(struct ip_set *set)
+{
+	struct mtype *map = set->data;
+
+	if (SET_WITH_TIMEOUT(set))
+		del_timer_sync(&map->gc);
+}
+
 static const struct ip_set_type_variant mtype = {
 	.kadt	= mtype_kadt,
 	.uadt	= mtype_uadt,
@@ -301,6 +308,7 @@ static const struct ip_set_type_variant mtype = {
 	.head	= mtype_head,
 	.list	= mtype_list,
 	.same_set = mtype_same_set,
+	.cancel_gc = mtype_cancel_gc,
 };
 
 #endif /* __IP_SET_BITMAP_IP_GEN_H */
diff --git a/net/netfilter/ipset/ip_set_core.c b/net/netfilter/ipset/ip_set_core.c
index d47dfdcb8..f645da82d 100644
--- a/net/netfilter/ipset/ip_set_core.c
+++ b/net/netfilter/ipset/ip_set_core.c
@@ -1156,6 +1156,7 @@ static int ip_set_create(struct sk_buff *skb, const struct nfnl_info *info,
 	return ret;
 
 cleanup:
+	set->variant->cancel_gc(set);
 	set->variant->destroy(set);
 put_out:
 	module_put(set->type->me);
@@ -1184,6 +1185,14 @@ ip_set_destroy_set(struct ip_set *set)
 	kfree(set);
 }
 
+static void
+ip_set_destroy_set_rcu(struct rcu_head *head)
+{
+	struct ip_set *set = container_of(head, struct ip_set, rcu);
+
+	ip_set_destroy_set(set);
+}
+
 static int ip_set_destroy(struct sk_buff *skb, const struct nfnl_info *info,
 			  const struct nlattr * const attr[])
 {
@@ -1195,8 +1204,6 @@ static int ip_set_destroy(struct sk_buff *skb, const struct nfnl_info *info,
 	if (unlikely(protocol_min_failed(attr)))
 		return -IPSET_ERR_PROTOCOL;
 
-	/* Must wait for flush to be really finished in list:set */
-	rcu_barrier();
 
 	/* Commands are serialized and references are
 	 * protected by the ip_set_ref_lock.
@@ -1208,8 +1215,10 @@ static int ip_set_destroy(struct sk_buff *skb, const struct nfnl_info *info,
 	 * counter, so if it's already zero, we can proceed
 	 * without holding the lock.
 	 */
-	read_lock_bh(&ip_set_ref_lock);
 	if (!attr[IPSET_ATTR_SETNAME]) {
+		/* Must wait for flush to be really finished in list:set */
+		rcu_barrier();
+		read_lock_bh(&ip_set_ref_lock);
 		for (i = 0; i < inst->ip_set_max; i++) {
 			s = ip_set(inst, i);
 			if (s && (s->ref || s->ref_netlink)) {
@@ -1223,6 +1232,8 @@ static int ip_set_destroy(struct sk_buff *skb, const struct nfnl_info *info,
 			s = ip_set(inst, i);
 			if (s) {
 				ip_set(inst, i) = NULL;
+				/* Must cancel garbage collectors */
+				s->variant->cancel_gc(s);
 				ip_set_destroy_set(s);
 			}
 		}
@@ -1230,6 +1241,9 @@ static int ip_set_destroy(struct sk_buff *skb, const struct nfnl_info *info,
 		inst->is_destroyed = false;
 	} else {
 		u32 flags = flag_exist(info->nlh);
+		u16 features = 0;
+
+		read_lock_bh(&ip_set_ref_lock);
 		s = find_set_and_id(inst, nla_data(attr[IPSET_ATTR_SETNAME]),
 				    &i);
 		if (!s) {
@@ -1240,10 +1254,16 @@ static int ip_set_destroy(struct sk_buff *skb, const struct nfnl_info *info,
 			ret = -IPSET_ERR_BUSY;
 			goto out;
 		}
+		features = s->type->features;
 		ip_set(inst, i) = NULL;
 		read_unlock_bh(&ip_set_ref_lock);
-
-		ip_set_destroy_set(s);
+		if (features & IPSET_TYPE_NAME) {
+			/* Must wait for flush to be really finished  */
+			rcu_barrier();
+		}
+		/* Must cancel garbage collectors */
+		s->variant->cancel_gc(s);
+		call_rcu(&s->rcu, ip_set_destroy_set_rcu);
 	}
 	return 0;
 out:
@@ -1396,9 +1416,6 @@ static int ip_set_swap(struct sk_buff *skb, const struct nfnl_info *info,
 	ip_set(inst, to_id) = from;
 	write_unlock_bh(&ip_set_ref_lock);
 
-	/* Make sure all readers of the old set pointers are completed. */
-	synchronize_rcu();
-
 	return 0;
 }
 
@@ -2364,6 +2381,7 @@ ip_set_net_exit(struct net *net)
 		set = ip_set(inst, i);
 		if (set) {
 			ip_set(inst, i) = NULL;
+			set->variant->cancel_gc(set);
 			ip_set_destroy_set(set);
 		}
 	}
@@ -2411,8 +2429,11 @@ ip_set_fini(void)
 {
 	nf_unregister_sockopt(&so_set);
 	nfnetlink_subsys_unregister(&ip_set_netlink_subsys);
-
 	unregister_pernet_subsys(&ip_set_net_ops);
+
+	/* Wait for call_rcu() in destroy */
+	rcu_barrier();
+
 	pr_debug("these are the famous last words\n");
 }
 
diff --git a/net/netfilter/ipset/ip_set_hash_gen.h b/net/netfilter/ipset/ip_set_hash_gen.h
index 7499192af..ef04e556a 100644
--- a/net/netfilter/ipset/ip_set_hash_gen.h
+++ b/net/netfilter/ipset/ip_set_hash_gen.h
@@ -210,6 +210,7 @@ htable_size(u8 hbits)
 #undef mtype_gc_do
 #undef mtype_gc
 #undef mtype_gc_init
+#undef mtype_cancel_gc
 #undef mtype_variant
 #undef mtype_data_match
 
@@ -254,6 +255,7 @@ htable_size(u8 hbits)
 #define mtype_gc_do		IPSET_TOKEN(MTYPE, _gc_do)
 #define mtype_gc		IPSET_TOKEN(MTYPE, _gc)
 #define mtype_gc_init		IPSET_TOKEN(MTYPE, _gc_init)
+#define mtype_cancel_gc		IPSET_TOKEN(MTYPE, _cancel_gc)
 #define mtype_variant		IPSET_TOKEN(MTYPE, _variant)
 #define mtype_data_match	IPSET_TOKEN(MTYPE, _data_match)
 
@@ -417,7 +419,7 @@ mtype_ahash_destroy(struct ip_set *set, struct htable *t, bool ext_destroy)
 	u32 i;
 
 	for (i = 0; i < jhash_size(t->htable_bits); i++) {
-		n = __ipset_dereference(hbucket(t, i));
+		n = (__force struct hbucket *)hbucket(t, i);
 		if (!n)
 			continue;
 		if (set->extensions & IPSET_EXT_DESTROY && ext_destroy)
@@ -437,10 +439,7 @@ mtype_destroy(struct ip_set *set)
 	struct htype *h = set->data;
 	struct list_head *l, *lt;
 
-	if (SET_WITH_TIMEOUT(set))
-		cancel_delayed_work_sync(&h->gc.dwork);
-
-	mtype_ahash_destroy(set, ipset_dereference_nfnl(h->table), true);
+	mtype_ahash_destroy(set, (__force struct htable *)h->table, true);
 	list_for_each_safe(l, lt, &h->ad) {
 		list_del(l);
 		kfree(l);
@@ -586,6 +585,15 @@ mtype_gc_init(struct htable_gc *gc)
 	queue_delayed_work(system_power_efficient_wq, &gc->dwork, HZ);
 }
 
+static void
+mtype_cancel_gc(struct ip_set *set)
+{
+	struct htype *h = set->data;
+
+	if (SET_WITH_TIMEOUT(set))
+		cancel_delayed_work_sync(&h->gc.dwork);
+}
+
 static int
 mtype_add(struct ip_set *set, void *value, const struct ip_set_ext *ext,
 	  struct ip_set_ext *mext, u32 flags);
@@ -1416,6 +1424,7 @@ static const struct ip_set_type_variant mtype_variant = {
 	.uref	= mtype_uref,
 	.resize	= mtype_resize,
 	.same_set = mtype_same_set,
+	.cancel_gc = mtype_cancel_gc,
 	.region_lock = true,
 };
 
diff --git a/net/netfilter/ipset/ip_set_list_set.c b/net/netfilter/ipset/ip_set_list_set.c
index 5a67f7966..6bc701998 100644
--- a/net/netfilter/ipset/ip_set_list_set.c
+++ b/net/netfilter/ipset/ip_set_list_set.c
@@ -426,9 +426,6 @@ list_set_destroy(struct ip_set *set)
 	struct list_set *map = set->data;
 	struct set_elem *e, *n;
 
-	if (SET_WITH_TIMEOUT(set))
-		del_timer_sync(&map->gc);
-
 	list_for_each_entry_safe(e, n, &map->members, list) {
 		list_del(&e->list);
 		ip_set_put_byindex(map->net, e->id);
@@ -545,6 +542,15 @@ list_set_same_set(const struct ip_set *a, const struct ip_set *b)
 	       a->extensions == b->extensions;
 }
 
+static void
+list_set_cancel_gc(struct ip_set *set)
+{
+	struct list_set *map = set->data;
+
+	if (SET_WITH_TIMEOUT(set))
+		del_timer_sync(&map->gc);
+}
+
 static const struct ip_set_type_variant set_variant = {
 	.kadt	= list_set_kadt,
 	.uadt	= list_set_uadt,
@@ -558,6 +564,7 @@ static const struct ip_set_type_variant set_variant = {
 	.head	= list_set_head,
 	.list	= list_set_list,
 	.same_set = list_set_same_set,
+	.cancel_gc = list_set_cancel_gc,
 };
 
 static void
diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c
index 796026296..024f93fc8 100644
--- a/net/netfilter/nf_conntrack_core.c
+++ b/net/netfilter/nf_conntrack_core.c
@@ -2772,11 +2772,24 @@ err_cachep:
 	return ret;
 }
 
+static void nf_conntrack_set_closing(struct nf_conntrack *nfct)
+{
+	struct nf_conn *ct = nf_ct_to_nf_conn(nfct);
+
+	switch (nf_ct_protonum(ct)) {
+	case IPPROTO_TCP:
+		nf_conntrack_tcp_set_closing(ct);
+		break;
+	}
+}
+
 static const struct nf_ct_hook nf_conntrack_hook = {
 	.update		= nf_conntrack_update,
 	.destroy	= nf_ct_destroy,
 	.get_tuple_skb  = nf_conntrack_get_tuple_skb,
 	.attach		= nf_conntrack_attach,
+	.set_closing	= nf_conntrack_set_closing,
+	.confirm	= __nf_conntrack_confirm,
 };
 
 void nf_conntrack_init_end(void)
diff --git a/net/netfilter/nf_conntrack_h323_asn1.c b/net/netfilter/nf_conntrack_h323_asn1.c
index e697a824b..540d97715 100644
--- a/net/netfilter/nf_conntrack_h323_asn1.c
+++ b/net/netfilter/nf_conntrack_h323_asn1.c
@@ -533,6 +533,8 @@ static int decode_seq(struct bitstr *bs, const struct field_t *f,
 	/* Get fields bitmap */
 	if (nf_h323_error_boundary(bs, 0, f->sz))
 		return H323_ERROR_BOUND;
+	if (f->sz > 32)
+		return H323_ERROR_RANGE;
 	bmp = get_bitmap(bs, f->sz);
 	if (base)
 		*(unsigned int *)base = bmp;
@@ -589,6 +591,8 @@ static int decode_seq(struct bitstr *bs, const struct field_t *f,
 	bmp2_len = get_bits(bs, 7) + 1;
 	if (nf_h323_error_boundary(bs, 0, bmp2_len))
 		return H323_ERROR_BOUND;
+	if (bmp2_len > 32)
+		return H323_ERROR_RANGE;
 	bmp2 = get_bitmap(bs, bmp2_len);
 	bmp |= bmp2 >> f->sz;
 	if (base)
diff --git a/net/netfilter/nf_conntrack_proto_sctp.c b/net/netfilter/nf_conntrack_proto_sctp.c
index c94a9971d..7ffd69849 100644
--- a/net/netfilter/nf_conntrack_proto_sctp.c
+++ b/net/netfilter/nf_conntrack_proto_sctp.c
@@ -299,7 +299,7 @@ sctp_new(struct nf_conn *ct, const struct sk_buff *skb,
 			pr_debug("Setting vtag %x for secondary conntrack\n",
 				 sh->vtag);
 			ct->proto.sctp.vtag[IP_CT_DIR_ORIGINAL] = sh->vtag;
-		} else {
+		} else if (sch->type == SCTP_CID_SHUTDOWN_ACK) {
 		/* If it is a shutdown ack OOTB packet, we expect a return
 		   shutdown complete, otherwise an ABORT Sec 8.4 (5) and (8) */
 			pr_debug("Setting vtag %x for new conn OOTB\n",
diff --git a/net/netfilter/nf_conntrack_proto_tcp.c b/net/netfilter/nf_conntrack_proto_tcp.c
index 3ac1af6f5..9480e638e 100644
--- a/net/netfilter/nf_conntrack_proto_tcp.c
+++ b/net/netfilter/nf_conntrack_proto_tcp.c
@@ -457,7 +457,8 @@ static void tcp_init_sender(struct ip_ct_tcp_state *sender,
 			    const struct sk_buff *skb,
 			    unsigned int dataoff,
 			    const struct tcphdr *tcph,
-			    u32 end, u32 win)
+			    u32 end, u32 win,
+			    enum ip_conntrack_dir dir)
 {
 	/* SYN-ACK in reply to a SYN
 	 * or SYN from reply direction in simultaneous open.
@@ -471,7 +472,8 @@ static void tcp_init_sender(struct ip_ct_tcp_state *sender,
 	 * Both sides must send the Window Scale option
 	 * to enable window scaling in either direction.
 	 */
-	if (!(sender->flags & IP_CT_TCP_FLAG_WINDOW_SCALE &&
+	if (dir == IP_CT_DIR_REPLY &&
+	    !(sender->flags & IP_CT_TCP_FLAG_WINDOW_SCALE &&
 	      receiver->flags & IP_CT_TCP_FLAG_WINDOW_SCALE)) {
 		sender->td_scale = 0;
 		receiver->td_scale = 0;
@@ -542,7 +544,7 @@ tcp_in_window(struct nf_conn *ct, enum ip_conntrack_dir dir,
 		if (tcph->syn) {
 			tcp_init_sender(sender, receiver,
 					skb, dataoff, tcph,
-					end, win);
+					end, win, dir);
 			if (!tcph->ack)
 				/* Simultaneous open */
 				return NFCT_TCP_ACCEPT;
@@ -585,7 +587,7 @@ tcp_in_window(struct nf_conn *ct, enum ip_conntrack_dir dir,
 		 */
 		tcp_init_sender(sender, receiver,
 				skb, dataoff, tcph,
-				end, win);
+				end, win, dir);
 
 		if (dir == IP_CT_DIR_REPLY && !tcph->ack)
 			return NFCT_TCP_ACCEPT;
@@ -911,6 +913,41 @@ static bool tcp_can_early_drop(const struct nf_conn *ct)
 	return false;
 }
 
+void nf_conntrack_tcp_set_closing(struct nf_conn *ct)
+{
+	enum tcp_conntrack old_state;
+	const unsigned int *timeouts;
+	u32 timeout;
+
+	if (!nf_ct_is_confirmed(ct))
+		return;
+
+	spin_lock_bh(&ct->lock);
+	old_state = ct->proto.tcp.state;
+	ct->proto.tcp.state = TCP_CONNTRACK_CLOSE;
+
+	if (old_state == TCP_CONNTRACK_CLOSE ||
+	    test_bit(IPS_FIXED_TIMEOUT_BIT, &ct->status)) {
+		spin_unlock_bh(&ct->lock);
+		return;
+	}
+
+	timeouts = nf_ct_timeout_lookup(ct);
+	if (!timeouts) {
+		const struct nf_tcp_net *tn;
+
+		tn = nf_tcp_pernet(nf_ct_net(ct));
+		timeouts = tn->timeouts;
+	}
+
+	timeout = timeouts[TCP_CONNTRACK_CLOSE];
+	WRITE_ONCE(ct->timeout, timeout + nfct_time_stamp);
+
+	spin_unlock_bh(&ct->lock);
+
+	nf_conntrack_event_cache(IPCT_PROTOINFO, ct);
+}
+
 static void nf_ct_tcp_state_reset(struct ip_ct_tcp_state *state)
 {
 	state->td_end		= 0;
diff --git a/net/netfilter/nf_flow_table_core.c b/net/netfilter/nf_flow_table_core.c
index c1d99cb37..99195cf6b 100644
--- a/net/netfilter/nf_flow_table_core.c
+++ b/net/netfilter/nf_flow_table_core.c
@@ -87,12 +87,22 @@ static u32 flow_offload_dst_cookie(struct flow_offload_tuple *flow_tuple)
 	return 0;
 }
 
+static struct dst_entry *nft_route_dst_fetch(struct nf_flow_route *route,
+					     enum flow_offload_tuple_dir dir)
+{
+	struct dst_entry *dst = route->tuple[dir].dst;
+
+	route->tuple[dir].dst = NULL;
+
+	return dst;
+}
+
 static int flow_offload_fill_route(struct flow_offload *flow,
-				   const struct nf_flow_route *route,
+				   struct nf_flow_route *route,
 				   enum flow_offload_tuple_dir dir)
 {
 	struct flow_offload_tuple *flow_tuple = &flow->tuplehash[dir].tuple;
-	struct dst_entry *dst = route->tuple[dir].dst;
+	struct dst_entry *dst = nft_route_dst_fetch(route, dir);
 	int i, j = 0;
 
 	switch (flow_tuple->l3proto) {
@@ -122,12 +132,10 @@ static int flow_offload_fill_route(struct flow_offload *flow,
 		       ETH_ALEN);
 		flow_tuple->out.ifidx = route->tuple[dir].out.ifindex;
 		flow_tuple->out.hw_ifidx = route->tuple[dir].out.hw_ifindex;
+		dst_release(dst);
 		break;
 	case FLOW_OFFLOAD_XMIT_XFRM:
 	case FLOW_OFFLOAD_XMIT_NEIGH:
-		if (!dst_hold_safe(route->tuple[dir].dst))
-			return -1;
-
 		flow_tuple->dst_cache = dst;
 		flow_tuple->dst_cookie = flow_offload_dst_cookie(flow_tuple);
 		break;
@@ -148,27 +156,12 @@ static void nft_flow_dst_release(struct flow_offload *flow,
 		dst_release(flow->tuplehash[dir].tuple.dst_cache);
 }
 
-int flow_offload_route_init(struct flow_offload *flow,
-			    const struct nf_flow_route *route)
+void flow_offload_route_init(struct flow_offload *flow,
+			     struct nf_flow_route *route)
 {
-	int err;
-
-	err = flow_offload_fill_route(flow, route, FLOW_OFFLOAD_DIR_ORIGINAL);
-	if (err < 0)
-		return err;
-
-	err = flow_offload_fill_route(flow, route, FLOW_OFFLOAD_DIR_REPLY);
-	if (err < 0)
-		goto err_route_reply;
-
+	flow_offload_fill_route(flow, route, FLOW_OFFLOAD_DIR_ORIGINAL);
+	flow_offload_fill_route(flow, route, FLOW_OFFLOAD_DIR_REPLY);
 	flow->type = NF_FLOW_OFFLOAD_ROUTE;
-
-	return 0;
-
-err_route_reply:
-	nft_flow_dst_release(flow, FLOW_OFFLOAD_DIR_ORIGINAL);
-
-	return err;
 }
 EXPORT_SYMBOL_GPL(flow_offload_route_init);
 
diff --git a/net/netfilter/nf_log.c b/net/netfilter/nf_log.c
index 8a2929014..be93a0249 100644
--- a/net/netfilter/nf_log.c
+++ b/net/netfilter/nf_log.c
@@ -193,11 +193,12 @@ void nf_logger_put(int pf, enum nf_log_type type)
 		return;
 	}
 
-	BUG_ON(loggers[pf][type] == NULL);
-
 	rcu_read_lock();
 	logger = rcu_dereference(loggers[pf][type]);
-	module_put(logger->me);
+	if (!logger)
+		WARN_ON_ONCE(1);
+	else
+		module_put(logger->me);
 	rcu_read_unlock();
 }
 EXPORT_SYMBOL_GPL(nf_logger_put);
diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c
index 1edb21382..d3ba947f4 100644
--- a/net/netfilter/nf_tables_api.c
+++ b/net/netfilter/nf_tables_api.c
@@ -686,15 +686,16 @@ static int nft_delobj(struct nft_ctx *ctx, struct nft_object *obj)
 	return err;
 }
 
-static int nft_trans_flowtable_add(struct nft_ctx *ctx, int msg_type,
-				   struct nft_flowtable *flowtable)
+static struct nft_trans *
+nft_trans_flowtable_add(struct nft_ctx *ctx, int msg_type,
+		        struct nft_flowtable *flowtable)
 {
 	struct nft_trans *trans;
 
 	trans = nft_trans_alloc(ctx, msg_type,
 				sizeof(struct nft_trans_flowtable));
 	if (trans == NULL)
-		return -ENOMEM;
+		return ERR_PTR(-ENOMEM);
 
 	if (msg_type == NFT_MSG_NEWFLOWTABLE)
 		nft_activate_next(ctx->net, flowtable);
@@ -703,22 +704,22 @@ static int nft_trans_flowtable_add(struct nft_ctx *ctx, int msg_type,
 	nft_trans_flowtable(trans) = flowtable;
 	nft_trans_commit_list_add_tail(ctx->net, trans);
 
-	return 0;
+	return trans;
 }
 
 static int nft_delflowtable(struct nft_ctx *ctx,
 			    struct nft_flowtable *flowtable)
 {
-	int err;
+	struct nft_trans *trans;
 
-	err = nft_trans_flowtable_add(ctx, NFT_MSG_DELFLOWTABLE, flowtable);
-	if (err < 0)
-		return err;
+	trans = nft_trans_flowtable_add(ctx, NFT_MSG_DELFLOWTABLE, flowtable);
+	if (IS_ERR(trans))
+		return PTR_ERR(trans);
 
 	nft_deactivate_next(ctx->net, flowtable);
 	nft_use_dec(&ctx->table->use);
 
-	return err;
+	return 0;
 }
 
 static void __nft_reg_track_clobber(struct nft_regs_track *track, u8 dreg)
@@ -1245,6 +1246,7 @@ static int nf_tables_updtable(struct nft_ctx *ctx)
 	return 0;
 
 err_register_hooks:
+	ctx->table->flags |= NFT_TABLE_F_DORMANT;
 	nft_trans_destroy(trans);
 	return ret;
 }
@@ -2057,7 +2059,7 @@ static struct nft_hook *nft_netdev_hook_alloc(struct net *net,
 	struct nft_hook *hook;
 	int err;
 
-	hook = kmalloc(sizeof(struct nft_hook), GFP_KERNEL_ACCOUNT);
+	hook = kzalloc(sizeof(struct nft_hook), GFP_KERNEL_ACCOUNT);
 	if (!hook) {
 		err = -ENOMEM;
 		goto err_hook_alloc;
@@ -2458,19 +2460,15 @@ static int nf_tables_addchain(struct nft_ctx *ctx, u8 family, u8 genmask,
 	RCU_INIT_POINTER(chain->blob_gen_0, blob);
 	RCU_INIT_POINTER(chain->blob_gen_1, blob);
 
-	err = nf_tables_register_hook(net, table, chain);
-	if (err < 0)
-		goto err_destroy_chain;
-
 	if (!nft_use_inc(&table->use)) {
 		err = -EMFILE;
-		goto err_use;
+		goto err_destroy_chain;
 	}
 
 	trans = nft_trans_chain_add(ctx, NFT_MSG_NEWCHAIN);
 	if (IS_ERR(trans)) {
 		err = PTR_ERR(trans);
-		goto err_unregister_hook;
+		goto err_trans;
 	}
 
 	nft_trans_chain_policy(trans) = NFT_CHAIN_POLICY_UNSET;
@@ -2478,17 +2476,22 @@ static int nf_tables_addchain(struct nft_ctx *ctx, u8 family, u8 genmask,
 		nft_trans_chain_policy(trans) = policy;
 
 	err = nft_chain_add(table, chain);
-	if (err < 0) {
-		nft_trans_destroy(trans);
-		goto err_unregister_hook;
-	}
+	if (err < 0)
+		goto err_chain_add;
+
+	/* This must be LAST to ensure no packets are walking over this chain. */
+	err = nf_tables_register_hook(net, table, chain);
+	if (err < 0)
+		goto err_register_hook;
 
 	return 0;
 
-err_unregister_hook:
+err_register_hook:
+	nft_chain_del(chain);
+err_chain_add:
+	nft_trans_destroy(trans);
+err_trans:
 	nft_use_dec_restore(&table->use);
-err_use:
-	nf_tables_unregister_hook(net, table, chain);
 err_destroy_chain:
 	nf_tables_chain_destroy(ctx);
 
@@ -4749,6 +4752,9 @@ static int nf_tables_newset(struct sk_buff *skb, const struct nfnl_info *info,
 		if (!(flags & NFT_SET_TIMEOUT))
 			return -EINVAL;
 
+		if (flags & NFT_SET_ANONYMOUS)
+			return -EOPNOTSUPP;
+
 		err = nf_msecs_to_jiffies64(nla[NFTA_SET_TIMEOUT], &desc.timeout);
 		if (err)
 			return err;
@@ -4757,6 +4763,10 @@ static int nf_tables_newset(struct sk_buff *skb, const struct nfnl_info *info,
 	if (nla[NFTA_SET_GC_INTERVAL] != NULL) {
 		if (!(flags & NFT_SET_TIMEOUT))
 			return -EINVAL;
+
+		if (flags & NFT_SET_ANONYMOUS)
+			return -EOPNOTSUPP;
+
 		desc.gc_int = ntohl(nla_get_be32(nla[NFTA_SET_GC_INTERVAL]));
 	}
 
@@ -7129,11 +7139,15 @@ nla_put_failure:
 	return -1;
 }
 
-static const struct nft_object_type *__nft_obj_type_get(u32 objtype)
+static const struct nft_object_type *__nft_obj_type_get(u32 objtype, u8 family)
 {
 	const struct nft_object_type *type;
 
 	list_for_each_entry(type, &nf_tables_objects, list) {
+		if (type->family != NFPROTO_UNSPEC &&
+		    type->family != family)
+			continue;
+
 		if (objtype == type->type)
 			return type;
 	}
@@ -7141,11 +7155,11 @@ static const struct nft_object_type *__nft_obj_type_get(u32 objtype)
 }
 
 static const struct nft_object_type *
-nft_obj_type_get(struct net *net, u32 objtype)
+nft_obj_type_get(struct net *net, u32 objtype, u8 family)
 {
 	const struct nft_object_type *type;
 
-	type = __nft_obj_type_get(objtype);
+	type = __nft_obj_type_get(objtype, family);
 	if (type != NULL && try_module_get(type->owner))
 		return type;
 
@@ -7238,7 +7252,7 @@ static int nf_tables_newobj(struct sk_buff *skb, const struct nfnl_info *info,
 		if (info->nlh->nlmsg_flags & NLM_F_REPLACE)
 			return -EOPNOTSUPP;
 
-		type = __nft_obj_type_get(objtype);
+		type = __nft_obj_type_get(objtype, family);
 		if (WARN_ON_ONCE(!type))
 			return -ENOENT;
 
@@ -7252,7 +7266,7 @@ static int nf_tables_newobj(struct sk_buff *skb, const struct nfnl_info *info,
 	if (!nft_use_inc(&table->use))
 		return -EMFILE;
 
-	type = nft_obj_type_get(net, objtype);
+	type = nft_obj_type_get(net, objtype, family);
 	if (IS_ERR(type)) {
 		err = PTR_ERR(type);
 		goto err_type;
@@ -7933,7 +7947,7 @@ err_unregister_net_hooks:
 	return err;
 }
 
-static void nft_flowtable_hooks_destroy(struct list_head *hook_list)
+static void nft_hooks_destroy(struct list_head *hook_list)
 {
 	struct nft_hook *hook, *next;
 
@@ -8026,9 +8040,9 @@ static int nf_tables_newflowtable(struct sk_buff *skb,
 	u8 family = info->nfmsg->nfgen_family;
 	const struct nf_flowtable_type *type;
 	struct nft_flowtable *flowtable;
-	struct nft_hook *hook, *next;
 	struct net *net = info->net;
 	struct nft_table *table;
+	struct nft_trans *trans;
 	struct nft_ctx ctx;
 	int err;
 
@@ -8108,34 +8122,34 @@ static int nf_tables_newflowtable(struct sk_buff *skb,
 	err = nft_flowtable_parse_hook(&ctx, nla[NFTA_FLOWTABLE_HOOK],
 				       &flowtable_hook, flowtable, true);
 	if (err < 0)
-		goto err4;
+		goto err_flowtable_parse_hooks;
 
 	list_splice(&flowtable_hook.list, &flowtable->hook_list);
 	flowtable->data.priority = flowtable_hook.priority;
 	flowtable->hooknum = flowtable_hook.num;
 
+	trans = nft_trans_flowtable_add(&ctx, NFT_MSG_NEWFLOWTABLE, flowtable);
+	if (IS_ERR(trans)) {
+		err = PTR_ERR(trans);
+		goto err_flowtable_trans;
+	}
+
+	/* This must be LAST to ensure no packets are walking over this flowtable. */
 	err = nft_register_flowtable_net_hooks(ctx.net, table,
 					       &flowtable->hook_list,
 					       flowtable);
-	if (err < 0) {
-		nft_flowtable_hooks_destroy(&flowtable->hook_list);
-		goto err4;
-	}
-
-	err = nft_trans_flowtable_add(&ctx, NFT_MSG_NEWFLOWTABLE, flowtable);
 	if (err < 0)
-		goto err5;
+		goto err_flowtable_hooks;
 
 	list_add_tail_rcu(&flowtable->list, &table->flowtables);
 
 	return 0;
-err5:
-	list_for_each_entry_safe(hook, next, &flowtable->hook_list, list) {
-		nft_unregister_flowtable_hook(net, flowtable, hook);
-		list_del_rcu(&hook->list);
-		kfree_rcu(hook, rcu);
-	}
-err4:
+
+err_flowtable_hooks:
+	nft_trans_destroy(trans);
+err_flowtable_trans:
+	nft_hooks_destroy(&flowtable->hook_list);
+err_flowtable_parse_hooks:
 	flowtable->data.type->free(&flowtable->data);
 err3:
 	module_put(type->owner);
@@ -8888,7 +8902,7 @@ static void nft_commit_release(struct nft_trans *trans)
 		break;
 	case NFT_MSG_DELFLOWTABLE:
 		if (nft_trans_flowtable_update(trans))
-			nft_flowtable_hooks_destroy(&nft_trans_flowtable_hooks(trans));
+			nft_hooks_destroy(&nft_trans_flowtable_hooks(trans));
 		else
 			nf_tables_flowtable_destroy(nft_trans_flowtable(trans));
 		break;
@@ -9845,7 +9859,7 @@ static void nf_tables_abort_release(struct nft_trans *trans)
 		break;
 	case NFT_MSG_NEWFLOWTABLE:
 		if (nft_trans_flowtable_update(trans))
-			nft_flowtable_hooks_destroy(&nft_trans_flowtable_hooks(trans));
+			nft_hooks_destroy(&nft_trans_flowtable_hooks(trans));
 		else
 			nf_tables_flowtable_destroy(nft_trans_flowtable(trans));
 		break;
diff --git a/net/netfilter/nft_compat.c b/net/netfilter/nft_compat.c
index 6952da7df..e4b8c02c5 100644
--- a/net/netfilter/nft_compat.c
+++ b/net/netfilter/nft_compat.c
@@ -135,7 +135,7 @@ static void nft_target_eval_bridge(const struct nft_expr *expr,
 
 static const struct nla_policy nft_target_policy[NFTA_TARGET_MAX + 1] = {
 	[NFTA_TARGET_NAME]	= { .type = NLA_NUL_STRING },
-	[NFTA_TARGET_REV]	= { .type = NLA_U32 },
+	[NFTA_TARGET_REV]	= NLA_POLICY_MAX(NLA_BE32, 255),
 	[NFTA_TARGET_INFO]	= { .type = NLA_BINARY },
 };
 
@@ -200,6 +200,7 @@ static const struct nla_policy nft_rule_compat_policy[NFTA_RULE_COMPAT_MAX + 1]
 static int nft_parse_compat(const struct nlattr *attr, u16 *proto, bool *inv)
 {
 	struct nlattr *tb[NFTA_RULE_COMPAT_MAX+1];
+	u32 l4proto;
 	u32 flags;
 	int err;
 
@@ -212,12 +213,18 @@ static int nft_parse_compat(const struct nlattr *attr, u16 *proto, bool *inv)
 		return -EINVAL;
 
 	flags = ntohl(nla_get_be32(tb[NFTA_RULE_COMPAT_FLAGS]));
-	if (flags & ~NFT_RULE_COMPAT_F_MASK)
+	if (flags & NFT_RULE_COMPAT_F_UNUSED ||
+	    flags & ~NFT_RULE_COMPAT_F_MASK)
 		return -EINVAL;
 	if (flags & NFT_RULE_COMPAT_F_INV)
 		*inv = true;
 
-	*proto = ntohl(nla_get_be32(tb[NFTA_RULE_COMPAT_PROTO]));
+	l4proto = ntohl(nla_get_be32(tb[NFTA_RULE_COMPAT_PROTO]));
+	if (l4proto > U16_MAX)
+		return -EINVAL;
+
+	*proto = l4proto;
+
 	return 0;
 }
 
@@ -351,10 +358,20 @@ static int nft_target_validate(const struct nft_ctx *ctx,
 
 	if (ctx->family != NFPROTO_IPV4 &&
 	    ctx->family != NFPROTO_IPV6 &&
+	    ctx->family != NFPROTO_INET &&
 	    ctx->family != NFPROTO_BRIDGE &&
 	    ctx->family != NFPROTO_ARP)
 		return -EOPNOTSUPP;
 
+	ret = nft_chain_validate_hooks(ctx->chain,
+				       (1 << NF_INET_PRE_ROUTING) |
+				       (1 << NF_INET_LOCAL_IN) |
+				       (1 << NF_INET_FORWARD) |
+				       (1 << NF_INET_LOCAL_OUT) |
+				       (1 << NF_INET_POST_ROUTING));
+	if (ret)
+		return ret;
+
 	if (nft_is_base_chain(ctx->chain)) {
 		const struct nft_base_chain *basechain =
 						nft_base_chain(ctx->chain);
@@ -418,7 +435,7 @@ static void nft_match_eval(const struct nft_expr *expr,
 
 static const struct nla_policy nft_match_policy[NFTA_MATCH_MAX + 1] = {
 	[NFTA_MATCH_NAME]	= { .type = NLA_NUL_STRING },
-	[NFTA_MATCH_REV]	= { .type = NLA_U32 },
+	[NFTA_MATCH_REV]	= NLA_POLICY_MAX(NLA_BE32, 255),
 	[NFTA_MATCH_INFO]	= { .type = NLA_BINARY },
 };
 
@@ -600,10 +617,20 @@ static int nft_match_validate(const struct nft_ctx *ctx,
 
 	if (ctx->family != NFPROTO_IPV4 &&
 	    ctx->family != NFPROTO_IPV6 &&
+	    ctx->family != NFPROTO_INET &&
 	    ctx->family != NFPROTO_BRIDGE &&
 	    ctx->family != NFPROTO_ARP)
 		return -EOPNOTSUPP;
 
+	ret = nft_chain_validate_hooks(ctx->chain,
+				       (1 << NF_INET_PRE_ROUTING) |
+				       (1 << NF_INET_LOCAL_IN) |
+				       (1 << NF_INET_FORWARD) |
+				       (1 << NF_INET_LOCAL_OUT) |
+				       (1 << NF_INET_POST_ROUTING));
+	if (ret)
+		return ret;
+
 	if (nft_is_base_chain(ctx->chain)) {
 		const struct nft_base_chain *basechain =
 						nft_base_chain(ctx->chain);
@@ -721,7 +748,7 @@ out_put:
 static const struct nla_policy nfnl_compat_policy_get[NFTA_COMPAT_MAX+1] = {
 	[NFTA_COMPAT_NAME]	= { .type = NLA_NUL_STRING,
 				    .len = NFT_COMPAT_NAME_MAX-1 },
-	[NFTA_COMPAT_REV]	= { .type = NLA_U32 },
+	[NFTA_COMPAT_REV]	= NLA_POLICY_MAX(NLA_BE32, 255),
 	[NFTA_COMPAT_TYPE]	= { .type = NLA_U32 },
 };
 
diff --git a/net/netfilter/nft_ct.c b/net/netfilter/nft_ct.c
index 641dc21f9..2bfe3cdfb 100644
--- a/net/netfilter/nft_ct.c
+++ b/net/netfilter/nft_ct.c
@@ -484,6 +484,9 @@ static int nft_ct_get_init(const struct nft_ctx *ctx,
 		break;
 #endif
 	case NFT_CT_ID:
+		if (tb[NFTA_CT_DIRECTION])
+			return -EINVAL;
+
 		len = sizeof(u32);
 		break;
 	default:
@@ -1231,7 +1234,30 @@ static int nft_ct_expect_obj_init(const struct nft_ctx *ctx,
 	if (tb[NFTA_CT_EXPECT_L3PROTO])
 		priv->l3num = ntohs(nla_get_be16(tb[NFTA_CT_EXPECT_L3PROTO]));
 
+	switch (priv->l3num) {
+	case NFPROTO_IPV4:
+	case NFPROTO_IPV6:
+		if (priv->l3num == ctx->family || ctx->family == NFPROTO_INET)
+			break;
+
+		return -EINVAL;
+	case NFPROTO_INET: /* tuple.src.l3num supports NFPROTO_IPV4/6 only */
+	default:
+		return -EAFNOSUPPORT;
+	}
+
 	priv->l4proto = nla_get_u8(tb[NFTA_CT_EXPECT_L4PROTO]);
+	switch (priv->l4proto) {
+	case IPPROTO_TCP:
+	case IPPROTO_UDP:
+	case IPPROTO_UDPLITE:
+	case IPPROTO_DCCP:
+	case IPPROTO_SCTP:
+		break;
+	default:
+		return -EOPNOTSUPP;
+	}
+
 	priv->dport = nla_get_be16(tb[NFTA_CT_EXPECT_DPORT]);
 	priv->timeout = nla_get_u32(tb[NFTA_CT_EXPECT_TIMEOUT]);
 	priv->size = nla_get_u8(tb[NFTA_CT_EXPECT_SIZE]);
diff --git a/net/netfilter/nft_flow_offload.c b/net/netfilter/nft_flow_offload.c
index 3d9f6dda5..7a8707632 100644
--- a/net/netfilter/nft_flow_offload.c
+++ b/net/netfilter/nft_flow_offload.c
@@ -250,9 +250,14 @@ static int nft_flow_route(const struct nft_pktinfo *pkt,
 		break;
 	}
 
+	if (!dst_hold_safe(this_dst))
+		return -ENOENT;
+
 	nf_route(nft_net(pkt), &other_dst, &fl, false, nft_pf(pkt));
-	if (!other_dst)
+	if (!other_dst) {
+		dst_release(this_dst);
 		return -ENOENT;
+	}
 
 	nft_default_forward_path(route, this_dst, dir);
 	nft_default_forward_path(route, other_dst, !dir);
@@ -349,8 +354,7 @@ static void nft_flow_offload_eval(const struct nft_expr *expr,
 	if (!flow)
 		goto err_flow_alloc;
 
-	if (flow_offload_route_init(flow, &route) < 0)
-		goto err_flow_add;
+	flow_offload_route_init(flow, &route);
 
 	if (tcph) {
 		ct->proto.tcp.seen[0].flags |= IP_CT_TCP_FLAG_BE_LIBERAL;
@@ -361,12 +365,12 @@ static void nft_flow_offload_eval(const struct nft_expr *expr,
 	if (ret < 0)
 		goto err_flow_add;
 
-	dst_release(route.tuple[!dir].dst);
 	return;
 
 err_flow_add:
 	flow_offload_free(flow);
 err_flow_alloc:
+	dst_release(route.tuple[dir].dst);
 	dst_release(route.tuple[!dir].dst);
 err_flow_route:
 	clear_bit(IPS_OFFLOAD_BIT, &ct->status);
diff --git a/net/netfilter/nft_set_pipapo.c b/net/netfilter/nft_set_pipapo.c
index 4e1cc3172..e1969209b 100644
--- a/net/netfilter/nft_set_pipapo.c
+++ b/net/netfilter/nft_set_pipapo.c
@@ -342,9 +342,6 @@
 #include "nft_set_pipapo_avx2.h"
 #include "nft_set_pipapo.h"
 
-/* Current working bitmap index, toggled between field matches */
-static DEFINE_PER_CPU(bool, nft_pipapo_scratch_index);
-
 /**
  * pipapo_refill() - For each set bit, set bits from selected mapping table item
  * @map:	Bitmap to be scanned for set bits
@@ -412,6 +409,7 @@ bool nft_pipapo_lookup(const struct net *net, const struct nft_set *set,
 		       const u32 *key, const struct nft_set_ext **ext)
 {
 	struct nft_pipapo *priv = nft_set_priv(set);
+	struct nft_pipapo_scratch *scratch;
 	unsigned long *res_map, *fill_map;
 	u8 genmask = nft_genmask_cur(net);
 	const u8 *rp = (const u8 *)key;
@@ -422,15 +420,17 @@ bool nft_pipapo_lookup(const struct net *net, const struct nft_set *set,
 
 	local_bh_disable();
 
-	map_index = raw_cpu_read(nft_pipapo_scratch_index);
-
 	m = rcu_dereference(priv->match);
 
 	if (unlikely(!m || !*raw_cpu_ptr(m->scratch)))
 		goto out;
 
-	res_map  = *raw_cpu_ptr(m->scratch) + (map_index ? m->bsize_max : 0);
-	fill_map = *raw_cpu_ptr(m->scratch) + (map_index ? 0 : m->bsize_max);
+	scratch = *raw_cpu_ptr(m->scratch);
+
+	map_index = scratch->map_index;
+
+	res_map  = scratch->map + (map_index ? m->bsize_max : 0);
+	fill_map = scratch->map + (map_index ? 0 : m->bsize_max);
 
 	memset(res_map, 0xff, m->bsize_max * sizeof(*res_map));
 
@@ -460,7 +460,7 @@ next_match:
 		b = pipapo_refill(res_map, f->bsize, f->rules, fill_map, f->mt,
 				  last);
 		if (b < 0) {
-			raw_cpu_write(nft_pipapo_scratch_index, map_index);
+			scratch->map_index = map_index;
 			local_bh_enable();
 
 			return false;
@@ -477,7 +477,7 @@ next_match:
 			 * current inactive bitmap is clean and can be reused as
 			 * *next* bitmap (not initial) for the next packet.
 			 */
-			raw_cpu_write(nft_pipapo_scratch_index, map_index);
+			scratch->map_index = map_index;
 			local_bh_enable();
 
 			return true;
@@ -1101,6 +1101,25 @@ static void pipapo_map(struct nft_pipapo_match *m,
 		f->mt[map[i].to + j].e = e;
 }
 
+/**
+ * pipapo_free_scratch() - Free per-CPU map at original (not aligned) address
+ * @m:		Matching data
+ * @cpu:	CPU number
+ */
+static void pipapo_free_scratch(const struct nft_pipapo_match *m, unsigned int cpu)
+{
+	struct nft_pipapo_scratch *s;
+	void *mem;
+
+	s = *per_cpu_ptr(m->scratch, cpu);
+	if (!s)
+		return;
+
+	mem = s;
+	mem -= s->align_off;
+	kfree(mem);
+}
+
 /**
  * pipapo_realloc_scratch() - Reallocate scratch maps for partial match results
  * @clone:	Copy of matching data with pending insertions and deletions
@@ -1114,12 +1133,13 @@ static int pipapo_realloc_scratch(struct nft_pipapo_match *clone,
 	int i;
 
 	for_each_possible_cpu(i) {
-		unsigned long *scratch;
+		struct nft_pipapo_scratch *scratch;
 #ifdef NFT_PIPAPO_ALIGN
-		unsigned long *scratch_aligned;
+		void *scratch_aligned;
+		u32 align_off;
 #endif
-
-		scratch = kzalloc_node(bsize_max * sizeof(*scratch) * 2 +
+		scratch = kzalloc_node(struct_size(scratch, map,
+						   bsize_max * 2) +
 				       NFT_PIPAPO_ALIGN_HEADROOM,
 				       GFP_KERNEL, cpu_to_node(i));
 		if (!scratch) {
@@ -1133,14 +1153,25 @@ static int pipapo_realloc_scratch(struct nft_pipapo_match *clone,
 			return -ENOMEM;
 		}
 
-		kfree(*per_cpu_ptr(clone->scratch, i));
-
-		*per_cpu_ptr(clone->scratch, i) = scratch;
+		pipapo_free_scratch(clone, i);
 
 #ifdef NFT_PIPAPO_ALIGN
-		scratch_aligned = NFT_PIPAPO_LT_ALIGN(scratch);
-		*per_cpu_ptr(clone->scratch_aligned, i) = scratch_aligned;
+		/* Align &scratch->map (not the struct itself): the extra
+		 * %NFT_PIPAPO_ALIGN_HEADROOM bytes passed to kzalloc_node()
+		 * above guarantee we can waste up to those bytes in order
+		 * to align the map field regardless of its offset within
+		 * the struct.
+		 */
+		BUILD_BUG_ON(offsetof(struct nft_pipapo_scratch, map) > NFT_PIPAPO_ALIGN_HEADROOM);
+
+		scratch_aligned = NFT_PIPAPO_LT_ALIGN(&scratch->map);
+		scratch_aligned -= offsetof(struct nft_pipapo_scratch, map);
+		align_off = scratch_aligned - (void *)scratch;
+
+		scratch = scratch_aligned;
+		scratch->align_off = align_off;
 #endif
+		*per_cpu_ptr(clone->scratch, i) = scratch;
 	}
 
 	return 0;
@@ -1294,11 +1325,6 @@ static struct nft_pipapo_match *pipapo_clone(struct nft_pipapo_match *old)
 	if (!new->scratch)
 		goto out_scratch;
 
-#ifdef NFT_PIPAPO_ALIGN
-	new->scratch_aligned = alloc_percpu(*new->scratch_aligned);
-	if (!new->scratch_aligned)
-		goto out_scratch;
-#endif
 	for_each_possible_cpu(i)
 		*per_cpu_ptr(new->scratch, i) = NULL;
 
@@ -1350,10 +1376,7 @@ out_lt:
 	}
 out_scratch_realloc:
 	for_each_possible_cpu(i)
-		kfree(*per_cpu_ptr(new->scratch, i));
-#ifdef NFT_PIPAPO_ALIGN
-	free_percpu(new->scratch_aligned);
-#endif
+		pipapo_free_scratch(new, i);
 out_scratch:
 	free_percpu(new->scratch);
 	kfree(new);
@@ -1638,13 +1661,9 @@ static void pipapo_free_match(struct nft_pipapo_match *m)
 	int i;
 
 	for_each_possible_cpu(i)
-		kfree(*per_cpu_ptr(m->scratch, i));
+		pipapo_free_scratch(m, i);
 
-#ifdef NFT_PIPAPO_ALIGN
-	free_percpu(m->scratch_aligned);
-#endif
 	free_percpu(m->scratch);
-
 	pipapo_free_fields(m);
 
 	kfree(m);
@@ -2132,7 +2151,7 @@ static int nft_pipapo_init(const struct nft_set *set,
 	m->field_count = field_count;
 	m->bsize_max = 0;
 
-	m->scratch = alloc_percpu(unsigned long *);
+	m->scratch = alloc_percpu(struct nft_pipapo_scratch *);
 	if (!m->scratch) {
 		err = -ENOMEM;
 		goto out_scratch;
@@ -2140,16 +2159,6 @@ static int nft_pipapo_init(const struct nft_set *set,
 	for_each_possible_cpu(i)
 		*per_cpu_ptr(m->scratch, i) = NULL;
 
-#ifdef NFT_PIPAPO_ALIGN
-	m->scratch_aligned = alloc_percpu(unsigned long *);
-	if (!m->scratch_aligned) {
-		err = -ENOMEM;
-		goto out_free;
-	}
-	for_each_possible_cpu(i)
-		*per_cpu_ptr(m->scratch_aligned, i) = NULL;
-#endif
-
 	rcu_head_init(&m->rcu);
 
 	nft_pipapo_for_each_field(f, i, m) {
@@ -2180,9 +2189,6 @@ static int nft_pipapo_init(const struct nft_set *set,
 	return 0;
 
 out_free:
-#ifdef NFT_PIPAPO_ALIGN
-	free_percpu(m->scratch_aligned);
-#endif
 	free_percpu(m->scratch);
 out_scratch:
 	kfree(m);
@@ -2236,11 +2242,8 @@ static void nft_pipapo_destroy(const struct nft_ctx *ctx,
 
 		nft_set_pipapo_match_destroy(ctx, set, m);
 
-#ifdef NFT_PIPAPO_ALIGN
-		free_percpu(m->scratch_aligned);
-#endif
 		for_each_possible_cpu(cpu)
-			kfree(*per_cpu_ptr(m->scratch, cpu));
+			pipapo_free_scratch(m, cpu);
 		free_percpu(m->scratch);
 		pipapo_free_fields(m);
 		kfree(m);
@@ -2253,11 +2256,8 @@ static void nft_pipapo_destroy(const struct nft_ctx *ctx,
 		if (priv->dirty)
 			nft_set_pipapo_match_destroy(ctx, set, m);
 
-#ifdef NFT_PIPAPO_ALIGN
-		free_percpu(priv->clone->scratch_aligned);
-#endif
 		for_each_possible_cpu(cpu)
-			kfree(*per_cpu_ptr(priv->clone->scratch, cpu));
+			pipapo_free_scratch(priv->clone, cpu);
 		free_percpu(priv->clone->scratch);
 
 		pipapo_free_fields(priv->clone);
diff --git a/net/netfilter/nft_set_pipapo.h b/net/netfilter/nft_set_pipapo.h
index 25a755915..30a3d092c 100644
--- a/net/netfilter/nft_set_pipapo.h
+++ b/net/netfilter/nft_set_pipapo.h
@@ -130,21 +130,29 @@ struct nft_pipapo_field {
 	union nft_pipapo_map_bucket *mt;
 };
 
+/**
+ * struct nft_pipapo_scratch - percpu data used for lookup and matching
+ * @map_index:	Current working bitmap index, toggled between field matches
+ * @align_off:	Offset to get the originally allocated address
+ * @map:	store partial matching results during lookup
+ */
+struct nft_pipapo_scratch {
+	u8 map_index;
+	u32 align_off;
+	unsigned long map[];
+};
+
 /**
  * struct nft_pipapo_match - Data used for lookup and matching
  * @field_count		Amount of fields in set
  * @scratch:		Preallocated per-CPU maps for partial matching results
- * @scratch_aligned:	Version of @scratch aligned to NFT_PIPAPO_ALIGN bytes
  * @bsize_max:		Maximum lookup table bucket size of all fields, in longs
  * @rcu			Matching data is swapped on commits
  * @f:			Fields, with lookup and mapping tables
  */
 struct nft_pipapo_match {
 	int field_count;
-#ifdef NFT_PIPAPO_ALIGN
-	unsigned long * __percpu *scratch_aligned;
-#endif
-	unsigned long * __percpu *scratch;
+	struct nft_pipapo_scratch * __percpu *scratch;
 	size_t bsize_max;
 	struct rcu_head rcu;
 	struct nft_pipapo_field f[];
diff --git a/net/netfilter/nft_set_pipapo_avx2.c b/net/netfilter/nft_set_pipapo_avx2.c
index 52e0d026d..a3a8ddca9 100644
--- a/net/netfilter/nft_set_pipapo_avx2.c
+++ b/net/netfilter/nft_set_pipapo_avx2.c
@@ -57,7 +57,7 @@
 
 /* Jump to label if @reg is zero */
 #define NFT_PIPAPO_AVX2_NOMATCH_GOTO(reg, label)			\
-	asm_volatile_goto("vptest %%ymm" #reg ", %%ymm" #reg ";"	\
+	asm goto("vptest %%ymm" #reg ", %%ymm" #reg ";"	\
 			  "je %l[" #label "]" : : : : label)
 
 /* Store 256 bits from YMM register into memory. Contrary to bucket load
@@ -71,9 +71,6 @@
 #define NFT_PIPAPO_AVX2_ZERO(reg)					\
 	asm volatile("vpxor %ymm" #reg ", %ymm" #reg ", %ymm" #reg)
 
-/* Current working bitmap index, toggled between field matches */
-static DEFINE_PER_CPU(bool, nft_pipapo_avx2_scratch_index);
-
 /**
  * nft_pipapo_avx2_prepare() - Prepare before main algorithm body
  *
@@ -1120,11 +1117,12 @@ bool nft_pipapo_avx2_lookup(const struct net *net, const struct nft_set *set,
 			    const u32 *key, const struct nft_set_ext **ext)
 {
 	struct nft_pipapo *priv = nft_set_priv(set);
-	unsigned long *res, *fill, *scratch;
+	struct nft_pipapo_scratch *scratch;
 	u8 genmask = nft_genmask_cur(net);
 	const u8 *rp = (const u8 *)key;
 	struct nft_pipapo_match *m;
 	struct nft_pipapo_field *f;
+	unsigned long *res, *fill;
 	bool map_index;
 	int i, ret = 0;
 
@@ -1141,15 +1139,16 @@ bool nft_pipapo_avx2_lookup(const struct net *net, const struct nft_set *set,
 	 */
 	kernel_fpu_begin_mask(0);
 
-	scratch = *raw_cpu_ptr(m->scratch_aligned);
+	scratch = *raw_cpu_ptr(m->scratch);
 	if (unlikely(!scratch)) {
 		kernel_fpu_end();
 		return false;
 	}
-	map_index = raw_cpu_read(nft_pipapo_avx2_scratch_index);
 
-	res  = scratch + (map_index ? m->bsize_max : 0);
-	fill = scratch + (map_index ? 0 : m->bsize_max);
+	map_index = scratch->map_index;
+
+	res  = scratch->map + (map_index ? m->bsize_max : 0);
+	fill = scratch->map + (map_index ? 0 : m->bsize_max);
 
 	/* Starting map doesn't need to be set for this implementation */
 
@@ -1221,7 +1220,7 @@ next_match:
 
 out:
 	if (i % 2)
-		raw_cpu_write(nft_pipapo_avx2_scratch_index, !map_index);
+		scratch->map_index = !map_index;
 	kernel_fpu_end();
 
 	return ret >= 0;
diff --git a/net/netfilter/nft_set_rbtree.c b/net/netfilter/nft_set_rbtree.c
index e34662f4a..5bf5572e9 100644
--- a/net/netfilter/nft_set_rbtree.c
+++ b/net/netfilter/nft_set_rbtree.c
@@ -235,7 +235,7 @@ static void nft_rbtree_gc_remove(struct net *net, struct nft_set *set,
 
 static const struct nft_rbtree_elem *
 nft_rbtree_gc_elem(const struct nft_set *__set, struct nft_rbtree *priv,
-		   struct nft_rbtree_elem *rbe, u8 genmask)
+		   struct nft_rbtree_elem *rbe)
 {
 	struct nft_set *set = (struct nft_set *)__set;
 	struct rb_node *prev = rb_prev(&rbe->node);
@@ -254,7 +254,7 @@ nft_rbtree_gc_elem(const struct nft_set *__set, struct nft_rbtree *priv,
 	while (prev) {
 		rbe_prev = rb_entry(prev, struct nft_rbtree_elem, node);
 		if (nft_rbtree_interval_end(rbe_prev) &&
-		    nft_set_elem_active(&rbe_prev->ext, genmask))
+		    nft_set_elem_active(&rbe_prev->ext, NFT_GENMASK_ANY))
 			break;
 
 		prev = rb_prev(prev);
@@ -365,7 +365,7 @@ static int __nft_rbtree_insert(const struct net *net, const struct nft_set *set,
 		    nft_set_elem_active(&rbe->ext, cur_genmask)) {
 			const struct nft_rbtree_elem *removed_end;
 
-			removed_end = nft_rbtree_gc_elem(set, priv, rbe, genmask);
+			removed_end = nft_rbtree_gc_elem(set, priv, rbe);
 			if (IS_ERR(removed_end))
 				return PTR_ERR(removed_end);
 
diff --git a/net/netfilter/nft_tunnel.c b/net/netfilter/nft_tunnel.c
index 983ade4be..efb505445 100644
--- a/net/netfilter/nft_tunnel.c
+++ b/net/netfilter/nft_tunnel.c
@@ -713,6 +713,7 @@ static const struct nft_object_ops nft_tunnel_obj_ops = {
 
 static struct nft_object_type nft_tunnel_obj_type __read_mostly = {
 	.type		= NFT_OBJECT_TUNNEL,
+	.family		= NFPROTO_NETDEV,
 	.ops		= &nft_tunnel_obj_ops,
 	.maxattr	= NFTA_TUNNEL_KEY_MAX,
 	.policy		= nft_tunnel_key_policy,
-- 
cgit v1.2.3