Merge git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net

Cross-merge networking fixes after downstream PR.

Conflicts:

include/linux/mlx5/driver.h
  617f5db1a626 ("RDMA/mlx5: Fix affinity assignment")
  dc13180824b7 ("net/mlx5: Enable devlink port for embedded cpu VF vports")
https://lore.kernel.org/all/20230613125939.595e50b8@canb.auug.org.au/

tools/testing/selftests/net/mptcp/mptcp_join.sh
  47867f0a7e83 ("selftests: mptcp: join: skip check if MIB counter not supported")
  425ba803124b ("selftests: mptcp: join: support RM_ADDR for used endpoints or not")
  45b1a1227a7a ("mptcp: introduces more address related mibs")
  0639fa230a21 ("selftests: mptcp: add explicit check for new mibs")
https://lore.kernel.org/netdev/20230609-upstream-net-20230610-mptcp-selftests-support-old-kernels-part-3-v1-0-2896fe2ee8a3@tessares.net/

No adjacent changes.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>

5 files changed, 112 insertions, 22 deletions

diff --git a/net/netfilter/nf_flow_table_core.c b/net/netfilter/nf_flow_table_core.c
index b46dd897f2c5..1d34d700bd09 100644
--- a/net/netfilter/nf_flow_table_core.c
+++ b/net/netfilter/nf_flow_table_core.c
@@ -299,12 +299,12 @@ int flow_offload_add(struct nf_flowtable *flow_table, struct flow_offload *flow)
 EXPORT_SYMBOL_GPL(flow_offload_add);
 
 void flow_offload_refresh(struct nf_flowtable *flow_table,
-			  struct flow_offload *flow)
+			  struct flow_offload *flow, bool force)
 {
 	u32 timeout;
 
 	timeout = nf_flowtable_time_stamp + flow_offload_get_timeout(flow);
-	if (timeout - READ_ONCE(flow->timeout) > HZ)
+	if (force || timeout - READ_ONCE(flow->timeout) > HZ)
 		WRITE_ONCE(flow->timeout, timeout);
 	else
 		return;
@@ -316,6 +316,12 @@ void flow_offload_refresh(struct nf_flowtable *flow_table,
 }
 EXPORT_SYMBOL_GPL(flow_offload_refresh);
 
+static bool nf_flow_is_outdated(const struct flow_offload *flow)
+{
+	return test_bit(IPS_SEEN_REPLY_BIT, &flow->ct->status) &&
+		!test_bit(NF_FLOW_HW_ESTABLISHED, &flow->flags);
+}
+
 static inline bool nf_flow_has_expired(const struct flow_offload *flow)
 {
 	return nf_flow_timeout_delta(flow->timeout) <= 0;
@@ -405,7 +411,8 @@ static void nf_flow_offload_gc_step(struct nf_flowtable *flow_table,
 				    struct flow_offload *flow, void *data)
 {
 	if (nf_flow_has_expired(flow) ||
-	    nf_ct_is_dying(flow->ct))
+	    nf_ct_is_dying(flow->ct) ||
+	    nf_flow_is_outdated(flow))
 		flow_offload_teardown(flow);
 
 	if (test_bit(NF_FLOW_TEARDOWN, &flow->flags)) {
diff --git a/net/netfilter/nf_flow_table_ip.c b/net/netfilter/nf_flow_table_ip.c
index d885d34edfe1..e45fade76409 100644
--- a/net/netfilter/nf_flow_table_ip.c
+++ b/net/netfilter/nf_flow_table_ip.c
@@ -388,7 +388,7 @@ static int nf_flow_offload_forward(struct nf_flowtable_ctx *ctx,
 	if (skb_try_make_writable(skb, thoff + ctx->hdrsize))
 		return -1;
 
-	flow_offload_refresh(flow_table, flow);
+	flow_offload_refresh(flow_table, flow, false);
 
 	nf_flow_encap_pop(skb, tuplehash);
 	thoff -= ctx->offset;
@@ -667,7 +667,7 @@ static int nf_flow_offload_ipv6_forward(struct nf_flowtable_ctx *ctx,
 	if (skb_try_make_writable(skb, thoff + ctx->hdrsize))
 		return -1;
 
-	flow_offload_refresh(flow_table, flow);
+	flow_offload_refresh(flow_table, flow, false);
 
 	nf_flow_encap_pop(skb, tuplehash);
 
diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c
index dc58991705d7..486d24b83fa9 100644
--- a/net/netfilter/nf_tables_api.c
+++ b/net/netfilter/nf_tables_api.c
@@ -3844,7 +3844,8 @@ err_destroy_flow_rule:
 	if (flow)
 		nft_flow_rule_destroy(flow);
 err_release_rule:
-	nf_tables_rule_release(&ctx, rule);
+	nft_rule_expr_deactivate(&ctx, rule, NFT_TRANS_PREPARE);
+	nf_tables_rule_destroy(&ctx, rule);
 err_release_expr:
 	for (i = 0; i < n; i++) {
 		if (expr_info[i].ops) {
@@ -4919,6 +4920,7 @@ static int nf_tables_newset(struct sk_buff *skb, const struct nfnl_info *info,
 
 	set->num_exprs = num_exprs;
 	set->handle = nf_tables_alloc_handle(table);
+	INIT_LIST_HEAD(&set->pending_update);
 
 	err = nft_trans_set_add(&ctx, NFT_MSG_NEWSET, set);
 	if (err < 0)
@@ -9278,10 +9280,25 @@ static void nf_tables_commit_audit_log(struct list_head *adl, u32 generation)
 	}
 }
 
+static void nft_set_commit_update(struct list_head *set_update_list)
+{
+	struct nft_set *set, *next;
+
+	list_for_each_entry_safe(set, next, set_update_list, pending_update) {
+		list_del_init(&set->pending_update);
+
+		if (!set->ops->commit)
+			continue;
+
+		set->ops->commit(set);
+	}
+}
+
 static int nf_tables_commit(struct net *net, struct sk_buff *skb)
 {
 	struct nftables_pernet *nft_net = nft_pernet(net);
 	struct nft_trans *trans, *next;
+	LIST_HEAD(set_update_list);
 	struct nft_trans_elem *te;
 	struct nft_chain *chain;
 	struct nft_table *table;
@@ -9456,6 +9473,11 @@ static int nf_tables_commit(struct net *net, struct sk_buff *skb)
 			nf_tables_setelem_notify(&trans->ctx, te->set,
 						 &te->elem,
 						 NFT_MSG_NEWSETELEM);
+			if (te->set->ops->commit &&
+			    list_empty(&te->set->pending_update)) {
+				list_add_tail(&te->set->pending_update,
+					      &set_update_list);
+			}
 			nft_trans_destroy(trans);
 			break;
 		case NFT_MSG_DELSETELEM:
@@ -9470,6 +9492,11 @@ static int nf_tables_commit(struct net *net, struct sk_buff *skb)
 				atomic_dec(&te->set->nelems);
 				te->set->ndeact--;
 			}
+			if (te->set->ops->commit &&
+			    list_empty(&te->set->pending_update)) {
+				list_add_tail(&te->set->pending_update,
+					      &set_update_list);
+			}
 			break;
 		case NFT_MSG_NEWOBJ:
 			if (nft_trans_obj_update(trans)) {
@@ -9532,6 +9559,8 @@ static int nf_tables_commit(struct net *net, struct sk_buff *skb)
 		}
 	}
 
+	nft_set_commit_update(&set_update_list);
+
 	nft_commit_notify(net, NETLINK_CB(skb).portid);
 	nf_tables_gen_notify(net, skb, NFT_MSG_NEWGEN);
 	nf_tables_commit_audit_log(&adl, nft_net->base_seq);
@@ -9591,10 +9620,25 @@ static void nf_tables_abort_release(struct nft_trans *trans)
 	kfree(trans);
 }
 
+static void nft_set_abort_update(struct list_head *set_update_list)
+{
+	struct nft_set *set, *next;
+
+	list_for_each_entry_safe(set, next, set_update_list, pending_update) {
+		list_del_init(&set->pending_update);
+
+		if (!set->ops->abort)
+			continue;
+
+		set->ops->abort(set);
+	}
+}
+
 static int __nf_tables_abort(struct net *net, enum nfnl_abort_action action)
 {
 	struct nftables_pernet *nft_net = nft_pernet(net);
 	struct nft_trans *trans, *next;
+	LIST_HEAD(set_update_list);
 	struct nft_trans_elem *te;
 
 	if (action == NFNL_ABORT_VALIDATE &&
@@ -9704,6 +9748,12 @@ static int __nf_tables_abort(struct net *net, enum nfnl_abort_action action)
 			nft_setelem_remove(net, te->set, &te->elem);
 			if (!nft_setelem_is_catchall(te->set, &te->elem))
 				atomic_dec(&te->set->nelems);
+
+			if (te->set->ops->abort &&
+			    list_empty(&te->set->pending_update)) {
+				list_add_tail(&te->set->pending_update,
+					      &set_update_list);
+			}
 			break;
 		case NFT_MSG_DELSETELEM:
 		case NFT_MSG_DESTROYSETELEM:
@@ -9714,6 +9764,11 @@ static int __nf_tables_abort(struct net *net, enum nfnl_abort_action action)
 			if (!nft_setelem_is_catchall(te->set, &te->elem))
 				te->set->ndeact--;
 
+			if (te->set->ops->abort &&
+			    list_empty(&te->set->pending_update)) {
+				list_add_tail(&te->set->pending_update,
+					      &set_update_list);
+			}
 			nft_trans_destroy(trans);
 			break;
 		case NFT_MSG_NEWOBJ:
@@ -9756,6 +9811,8 @@ static int __nf_tables_abort(struct net *net, enum nfnl_abort_action action)
 		}
 	}
 
+	nft_set_abort_update(&set_update_list);
+
 	synchronize_rcu();
 
 	list_for_each_entry_safe_reverse(trans, next,
diff --git a/net/netfilter/nfnetlink.c b/net/netfilter/nfnetlink.c
index ae7146475d17..c9fbe0f707b5 100644
--- a/net/netfilter/nfnetlink.c
+++ b/net/netfilter/nfnetlink.c
@@ -533,7 +533,8 @@ ack:
 			 * processed, this avoids that the same error is
 			 * reported several times when replaying the batch.
 			 */
-			if (nfnl_err_add(&err_list, nlh, err, &extack) < 0) {
+			if (err == -ENOMEM ||
+			    nfnl_err_add(&err_list, nlh, err, &extack) < 0) {
 				/* We failed to enqueue an error, reset the
 				 * list of errors and send OOM to userspace
 				 * pointing to the batch header.
diff --git a/net/netfilter/nft_set_pipapo.c b/net/netfilter/nft_set_pipapo.c
index 34c684e121d3..83131f23ad2d 100644
--- a/net/netfilter/nft_set_pipapo.c
+++ b/net/netfilter/nft_set_pipapo.c
@@ -1599,17 +1599,10 @@ static void pipapo_free_fields(struct nft_pipapo_match *m)
 	}
 }
 
-/**
- * pipapo_reclaim_match - RCU callback to free fields from old matching data
- * @rcu:	RCU head
- */
-static void pipapo_reclaim_match(struct rcu_head *rcu)
+static void pipapo_free_match(struct nft_pipapo_match *m)
 {
-	struct nft_pipapo_match *m;
 	int i;
 
-	m = container_of(rcu, struct nft_pipapo_match, rcu);
-
 	for_each_possible_cpu(i)
 		kfree(*per_cpu_ptr(m->scratch, i));
 
@@ -1624,7 +1617,19 @@ static void pipapo_reclaim_match(struct rcu_head *rcu)
 }
 
 /**
- * pipapo_commit() - Replace lookup data with current working copy
+ * pipapo_reclaim_match - RCU callback to free fields from old matching data
+ * @rcu:	RCU head
+ */
+static void pipapo_reclaim_match(struct rcu_head *rcu)
+{
+	struct nft_pipapo_match *m;
+
+	m = container_of(rcu, struct nft_pipapo_match, rcu);
+	pipapo_free_match(m);
+}
+
+/**
+ * nft_pipapo_commit() - Replace lookup data with current working copy
  * @set:	nftables API set representation
  *
  * While at it, check if we should perform garbage collection on the working
@@ -1634,7 +1639,7 @@ static void pipapo_reclaim_match(struct rcu_head *rcu)
  * We also need to create a new working copy for subsequent insertions and
  * deletions.
  */
-static void pipapo_commit(const struct nft_set *set)
+static void nft_pipapo_commit(const struct nft_set *set)
 {
 	struct nft_pipapo *priv = nft_set_priv(set);
 	struct nft_pipapo_match *new_clone, *old;
@@ -1659,6 +1664,26 @@ static void pipapo_commit(const struct nft_set *set)
 	priv->clone = new_clone;
 }
 
+static void nft_pipapo_abort(const struct nft_set *set)
+{
+	struct nft_pipapo *priv = nft_set_priv(set);
+	struct nft_pipapo_match *new_clone, *m;
+
+	if (!priv->dirty)
+		return;
+
+	m = rcu_dereference(priv->match);
+
+	new_clone = pipapo_clone(m);
+	if (IS_ERR(new_clone))
+		return;
+
+	priv->dirty = false;
+
+	pipapo_free_match(priv->clone);
+	priv->clone = new_clone;
+}
+
 /**
  * nft_pipapo_activate() - Mark element reference as active given key, commit
  * @net:	Network namespace
@@ -1666,8 +1691,7 @@ static void pipapo_commit(const struct nft_set *set)
  * @elem:	nftables API element representation containing key data
  *
  * On insertion, elements are added to a copy of the matching data currently
- * in use for lookups, and not directly inserted into current lookup data, so
- * we'll take care of that by calling pipapo_commit() here. Both
+ * in use for lookups, and not directly inserted into current lookup data. Both
  * nft_pipapo_insert() and nft_pipapo_activate() are called once for each
  * element, hence we can't purpose either one as a real commit operation.
  */
@@ -1683,8 +1707,6 @@ static void nft_pipapo_activate(const struct net *net,
 
 	nft_set_elem_change_active(net, set, &e->ext);
 	nft_set_elem_clear_busy(&e->ext);
-
-	pipapo_commit(set);
 }
 
 /**
@@ -1930,7 +1952,6 @@ static void nft_pipapo_remove(const struct net *net, const struct nft_set *set,
 		if (i == m->field_count) {
 			priv->dirty = true;
 			pipapo_drop(m, rulemap);
-			pipapo_commit(set);
 			return;
 		}
 
@@ -2228,6 +2249,8 @@ const struct nft_set_type nft_set_pipapo_type = {
 		.init		= nft_pipapo_init,
 		.destroy	= nft_pipapo_destroy,
 		.gc_init	= nft_pipapo_gc_init,
+		.commit		= nft_pipapo_commit,
+		.abort		= nft_pipapo_abort,
 		.elemsize	= offsetof(struct nft_pipapo_elem, ext),
 	},
 };
@@ -2250,6 +2273,8 @@ const struct nft_set_type nft_set_pipapo_avx2_type = {
 		.init		= nft_pipapo_init,
 		.destroy	= nft_pipapo_destroy,
 		.gc_init	= nft_pipapo_gc_init,
+		.commit		= nft_pipapo_commit,
+		.abort		= nft_pipapo_abort,
 		.elemsize	= offsetof(struct nft_pipapo_elem, ext),
 	},
 };