1 files changed, 110 insertions, 255 deletions

diff --git a/net/sched/cls_api.c b/net/sched/cls_api.c
index 2ef8f5a6205a..50566db45949 100644
--- a/net/sched/cls_api.c
+++ b/net/sched/cls_api.c
@@ -49,6 +49,23 @@ static LIST_HEAD(tcf_proto_base);
 /* Protects list of registered TC modules. It is pure SMP lock. */
 static DEFINE_RWLOCK(cls_mod_lock);
 
+#ifdef CONFIG_NET_CLS_ACT
+DEFINE_STATIC_KEY_FALSE(tc_skb_ext_tc);
+EXPORT_SYMBOL(tc_skb_ext_tc);
+
+void tc_skb_ext_tc_enable(void)
+{
+	static_branch_inc(&tc_skb_ext_tc);
+}
+EXPORT_SYMBOL(tc_skb_ext_tc_enable);
+
+void tc_skb_ext_tc_disable(void)
+{
+	static_branch_dec(&tc_skb_ext_tc);
+}
+EXPORT_SYMBOL(tc_skb_ext_tc_disable);
+#endif
+
 static u32 destroy_obj_hashfn(const struct tcf_proto *tp)
 {
 	return jhash_3words(tp->chain->index, tp->prio,
@@ -177,7 +194,7 @@ EXPORT_SYMBOL(register_tcf_proto_ops);
 
 static struct workqueue_struct *tc_filter_wq;
 
-int unregister_tcf_proto_ops(struct tcf_proto_ops *ops)
+void unregister_tcf_proto_ops(struct tcf_proto_ops *ops)
 {
 	struct tcf_proto_ops *t;
 	int rc = -ENOENT;
@@ -197,7 +214,8 @@ int unregister_tcf_proto_ops(struct tcf_proto_ops *ops)
 		}
 	}
 	write_unlock(&cls_mod_lock);
-	return rc;
+
+	WARN(rc, "unregister tc filter kind(%s) failed %d\n", ops->kind, rc);
 }
 EXPORT_SYMBOL(unregister_tcf_proto_ops);
 
@@ -1044,7 +1062,7 @@ static int __tcf_qdisc_find(struct net *net, struct Qdisc **q,
 
 	/* Find qdisc */
 	if (!*parent) {
-		*q = dev->qdisc;
+		*q = rcu_dereference(dev->qdisc);
 		*parent = (*q)->handle;
 	} else {
 		*q = qdisc_lookup_rcu(dev, TC_H_MAJ(*parent));
@@ -1615,14 +1633,21 @@ int tcf_classify(struct sk_buff *skb,
 	ret = __tcf_classify(skb, tp, orig_tp, res, compat_mode,
 			     &last_executed_chain);
 
-	/* If we missed on some chain */
-	if (ret == TC_ACT_UNSPEC && last_executed_chain) {
-		ext = tc_skb_ext_alloc(skb);
-		if (WARN_ON_ONCE(!ext))
-			return TC_ACT_SHOT;
-		ext->chain = last_executed_chain;
-		ext->mru = qdisc_skb_cb(skb)->mru;
-		ext->post_ct = qdisc_skb_cb(skb)->post_ct;
+	if (tc_skb_ext_tc_enabled()) {
+		/* If we missed on some chain */
+		if (ret == TC_ACT_UNSPEC && last_executed_chain) {
+			struct tc_skb_cb *cb = tc_skb_cb(skb);
+
+			ext = tc_skb_ext_alloc(skb);
+			if (WARN_ON_ONCE(!ext))
+				return TC_ACT_SHOT;
+			ext->chain = last_executed_chain;
+			ext->mru = cb->mru;
+			ext->post_ct = cb->post_ct;
+			ext->post_ct_snat = cb->post_ct_snat;
+			ext->post_ct_dnat = cb->post_ct_dnat;
+			ext->zone = cb->zone;
+		}
 	}
 
 	return ret;
@@ -1648,10 +1673,10 @@ static int tcf_chain_tp_insert(struct tcf_chain *chain,
 	if (chain->flushing)
 		return -EAGAIN;
 
+	RCU_INIT_POINTER(tp->next, tcf_chain_tp_prev(chain, chain_info));
 	if (*chain_info->pprev == chain->filter_chain)
 		tcf_chain0_head_change(chain, tp);
 	tcf_proto_get(tp);
-	RCU_INIT_POINTER(tp->next, tcf_chain_tp_prev(chain, chain_info));
 	rcu_assign_pointer(*chain_info->pprev, tp);
 
 	return 0;
@@ -1940,9 +1965,9 @@ static int tc_new_tfilter(struct sk_buff *skb, struct nlmsghdr *n,
 	bool prio_allocate;
 	u32 parent;
 	u32 chain_index;
-	struct Qdisc *q = NULL;
+	struct Qdisc *q;
 	struct tcf_chain_info chain_info;
-	struct tcf_chain *chain = NULL;
+	struct tcf_chain *chain;
 	struct tcf_block *block;
 	struct tcf_proto *tp;
 	unsigned long cl;
@@ -1952,9 +1977,6 @@ static int tc_new_tfilter(struct sk_buff *skb, struct nlmsghdr *n,
 	bool rtnl_held = false;
 	u32 flags;
 
-	if (!netlink_ns_capable(skb, net->user_ns, CAP_NET_ADMIN))
-		return -EPERM;
-
 replay:
 	tp_created = 0;
 
@@ -1971,6 +1993,8 @@ replay:
 	tp = NULL;
 	cl = 0;
 	block = NULL;
+	q = NULL;
+	chain = NULL;
 	flags = 0;
 
 	if (prio == 0) {
@@ -2110,6 +2134,7 @@ replay:
 	}
 
 	if (chain->tmplt_ops && chain->tmplt_ops != tp->ops) {
+		tfilter_put(tp, fh);
 		NL_SET_ERR_MSG(extack, "Chain template is set to a different filter kind");
 		err = -EINVAL;
 		goto errout;
@@ -2181,9 +2206,6 @@ static int tc_del_tfilter(struct sk_buff *skb, struct nlmsghdr *n,
 	int err;
 	bool rtnl_held = false;
 
-	if (!netlink_ns_capable(skb, net->user_ns, CAP_NET_ADMIN))
-		return -EPERM;
-
 	err = nlmsg_parse_deprecated(n, sizeof(*t), tca, TCA_MAX,
 				     rtm_tca_policy, extack);
 	if (err < 0)
@@ -2580,7 +2602,7 @@ static int tc_dump_tfilter(struct sk_buff *skb, struct netlink_callback *cb)
 
 		parent = tcm->tcm_parent;
 		if (!parent)
-			q = dev->qdisc;
+			q = rtnl_dereference(dev->qdisc);
 		else
 			q = qdisc_lookup(dev, TC_H_MAJ(tcm->tcm_parent));
 		if (!q)
@@ -2793,17 +2815,14 @@ static int tc_ctl_chain(struct sk_buff *skb, struct nlmsghdr *n,
 	struct tcmsg *t;
 	u32 parent;
 	u32 chain_index;
-	struct Qdisc *q = NULL;
-	struct tcf_chain *chain = NULL;
+	struct Qdisc *q;
+	struct tcf_chain *chain;
 	struct tcf_block *block;
 	unsigned long cl;
 	int err;
 
-	if (n->nlmsg_type != RTM_GETCHAIN &&
-	    !netlink_ns_capable(skb, net->user_ns, CAP_NET_ADMIN))
-		return -EPERM;
-
 replay:
+	q = NULL;
 	err = nlmsg_parse_deprecated(n, sizeof(*t), tca, TCA_MAX,
 				     rtm_tca_policy, extack);
 	if (err < 0)
@@ -2954,7 +2973,7 @@ static int tc_dump_chain(struct sk_buff *skb, struct netlink_callback *cb)
 			return skb->len;
 
 		if (!tcm->tcm_parent)
-			q = dev->qdisc;
+			q = rtnl_dereference(dev->qdisc);
 		else
 			q = qdisc_lookup(dev, TC_H_MAJ(tcm->tcm_parent));
 
@@ -3025,9 +3044,9 @@ void tcf_exts_destroy(struct tcf_exts *exts)
 }
 EXPORT_SYMBOL(tcf_exts_destroy);
 
-int tcf_exts_validate(struct net *net, struct tcf_proto *tp, struct nlattr **tb,
-		      struct nlattr *rate_tlv, struct tcf_exts *exts,
-		      u32 flags, struct netlink_ext_ack *extack)
+int tcf_exts_validate_ex(struct net *net, struct tcf_proto *tp, struct nlattr **tb,
+			 struct nlattr *rate_tlv, struct tcf_exts *exts,
+			 u32 flags, u32 fl_flags, struct netlink_ext_ack *extack)
 {
 #ifdef CONFIG_NET_CLS_ACT
 	{
@@ -3061,7 +3080,8 @@ int tcf_exts_validate(struct net *net, struct tcf_proto *tp, struct nlattr **tb,
 			flags |= TCA_ACT_FLAGS_BIND;
 			err = tcf_action_init(net, tp, tb[exts->action],
 					      rate_tlv, exts->actions, init_res,
-					      &attr_size, flags, extack);
+					      &attr_size, flags, fl_flags,
+					      extack);
 			if (err < 0)
 				return err;
 			exts->nr_actions = err;
@@ -3077,6 +3097,15 @@ int tcf_exts_validate(struct net *net, struct tcf_proto *tp, struct nlattr **tb,
 
 	return 0;
 }
+EXPORT_SYMBOL(tcf_exts_validate_ex);
+
+int tcf_exts_validate(struct net *net, struct tcf_proto *tp, struct nlattr **tb,
+		      struct nlattr *rate_tlv, struct tcf_exts *exts,
+		      u32 flags, struct netlink_ext_ack *extack)
+{
+	return tcf_exts_validate_ex(net, tp, tb, rate_tlv, exts,
+				    flags, 0, extack);
+}
 EXPORT_SYMBOL(tcf_exts_validate);
 
 void tcf_exts_change(struct tcf_exts *dst, struct tcf_exts *src)
@@ -3320,7 +3349,7 @@ err_unlock:
 	up_read(&block->cb_lock);
 	if (take_rtnl)
 		rtnl_unlock();
-	return ok_count < 0 ? ok_count : 0;
+	return min(ok_count, 0);
 }
 EXPORT_SYMBOL(tc_setup_cb_add);
 
@@ -3376,7 +3405,7 @@ err_unlock:
 	up_read(&block->cb_lock);
 	if (take_rtnl)
 		rtnl_unlock();
-	return ok_count < 0 ? ok_count : 0;
+	return min(ok_count, 0);
 }
 EXPORT_SYMBOL(tc_setup_cb_replace);
 
@@ -3414,7 +3443,7 @@ retry:
 	up_read(&block->cb_lock);
 	if (take_rtnl)
 		rtnl_unlock();
-	return ok_count < 0 ? ok_count : 0;
+	return min(ok_count, 0);
 }
 EXPORT_SYMBOL(tc_setup_cb_destroy);
 
@@ -3461,7 +3490,7 @@ static void tcf_act_put_cookie(struct flow_action_entry *entry)
 	flow_action_cookie_destroy(entry->cookie);
 }
 
-void tc_cleanup_flow_action(struct flow_action *flow_action)
+void tc_cleanup_offload_action(struct flow_action *flow_action)
 {
 	struct flow_action_entry *entry;
 	int i;
@@ -3472,93 +3501,42 @@ void tc_cleanup_flow_action(struct flow_action *flow_action)
 			entry->destructor(entry->destructor_priv);
 	}
 }
-EXPORT_SYMBOL(tc_cleanup_flow_action);
+EXPORT_SYMBOL(tc_cleanup_offload_action);
 
-static void tcf_mirred_get_dev(struct flow_action_entry *entry,
-			       const struct tc_action *act)
+static int tc_setup_offload_act(struct tc_action *act,
+				struct flow_action_entry *entry,
+				u32 *index_inc,
+				struct netlink_ext_ack *extack)
 {
 #ifdef CONFIG_NET_CLS_ACT
-	entry->dev = act->ops->get_dev(act, &entry->destructor);
-	if (!entry->dev)
-		return;
-	entry->destructor_priv = entry->dev;
-#endif
-}
-
-static void tcf_tunnel_encap_put_tunnel(void *priv)
-{
-	struct ip_tunnel_info *tunnel = priv;
-
-	kfree(tunnel);
-}
-
-static int tcf_tunnel_encap_get_tunnel(struct flow_action_entry *entry,
-				       const struct tc_action *act)
-{
-	entry->tunnel = tcf_tunnel_info_copy(act);
-	if (!entry->tunnel)
-		return -ENOMEM;
-	entry->destructor = tcf_tunnel_encap_put_tunnel;
-	entry->destructor_priv = entry->tunnel;
+	if (act->ops->offload_act_setup) {
+		return act->ops->offload_act_setup(act, entry, index_inc, true,
+						   extack);
+	} else {
+		NL_SET_ERR_MSG(extack, "Action does not support offload");
+		return -EOPNOTSUPP;
+	}
+#else
 	return 0;
-}
-
-static void tcf_sample_get_group(struct flow_action_entry *entry,
-				 const struct tc_action *act)
-{
-#ifdef CONFIG_NET_CLS_ACT
-	entry->sample.psample_group =
-		act->ops->get_psample_group(act, &entry->destructor);
-	entry->destructor_priv = entry->sample.psample_group;
 #endif
 }
 
-static void tcf_gate_entry_destructor(void *priv)
-{
-	struct action_gate_entry *oe = priv;
-
-	kfree(oe);
-}
-
-static int tcf_gate_get_entries(struct flow_action_entry *entry,
-				const struct tc_action *act)
-{
-	entry->gate.entries = tcf_gate_get_list(act);
-
-	if (!entry->gate.entries)
-		return -EINVAL;
-
-	entry->destructor = tcf_gate_entry_destructor;
-	entry->destructor_priv = entry->gate.entries;
-
-	return 0;
-}
-
-static enum flow_action_hw_stats tc_act_hw_stats(u8 hw_stats)
-{
-	if (WARN_ON_ONCE(hw_stats > TCA_ACT_HW_STATS_ANY))
-		return FLOW_ACTION_HW_STATS_DONT_CARE;
-	else if (!hw_stats)
-		return FLOW_ACTION_HW_STATS_DISABLED;
-
-	return hw_stats;
-}
-
-int tc_setup_flow_action(struct flow_action *flow_action,
-			 const struct tcf_exts *exts)
+int tc_setup_action(struct flow_action *flow_action,
+		    struct tc_action *actions[],
+		    struct netlink_ext_ack *extack)
 {
+	int i, j, k, index, err = 0;
 	struct tc_action *act;
-	int i, j, k, err = 0;
 
 	BUILD_BUG_ON(TCA_ACT_HW_STATS_ANY != FLOW_ACTION_HW_STATS_ANY);
 	BUILD_BUG_ON(TCA_ACT_HW_STATS_IMMEDIATE != FLOW_ACTION_HW_STATS_IMMEDIATE);
 	BUILD_BUG_ON(TCA_ACT_HW_STATS_DELAYED != FLOW_ACTION_HW_STATS_DELAYED);
 
-	if (!exts)
+	if (!actions)
 		return 0;
 
 	j = 0;
-	tcf_exts_for_each_action(i, act, exts) {
+	tcf_act_for_each_action(i, act, actions) {
 		struct flow_action_entry *entry;
 
 		entry = &flow_action->entries[j];
@@ -3567,165 +3545,45 @@ int tc_setup_flow_action(struct flow_action *flow_action,
 		if (err)
 			goto err_out_locked;
 
-		entry->hw_stats = tc_act_hw_stats(act->hw_stats);
-
-		if (is_tcf_gact_ok(act)) {
-			entry->id = FLOW_ACTION_ACCEPT;
-		} else if (is_tcf_gact_shot(act)) {
-			entry->id = FLOW_ACTION_DROP;
-		} else if (is_tcf_gact_trap(act)) {
-			entry->id = FLOW_ACTION_TRAP;
-		} else if (is_tcf_gact_goto_chain(act)) {
-			entry->id = FLOW_ACTION_GOTO;
-			entry->chain_index = tcf_gact_goto_chain_index(act);
-		} else if (is_tcf_mirred_egress_redirect(act)) {
-			entry->id = FLOW_ACTION_REDIRECT;
-			tcf_mirred_get_dev(entry, act);
-		} else if (is_tcf_mirred_egress_mirror(act)) {
-			entry->id = FLOW_ACTION_MIRRED;
-			tcf_mirred_get_dev(entry, act);
-		} else if (is_tcf_mirred_ingress_redirect(act)) {
-			entry->id = FLOW_ACTION_REDIRECT_INGRESS;
-			tcf_mirred_get_dev(entry, act);
-		} else if (is_tcf_mirred_ingress_mirror(act)) {
-			entry->id = FLOW_ACTION_MIRRED_INGRESS;
-			tcf_mirred_get_dev(entry, act);
-		} else if (is_tcf_vlan(act)) {
-			switch (tcf_vlan_action(act)) {
-			case TCA_VLAN_ACT_PUSH:
-				entry->id = FLOW_ACTION_VLAN_PUSH;
-				entry->vlan.vid = tcf_vlan_push_vid(act);
-				entry->vlan.proto = tcf_vlan_push_proto(act);
-				entry->vlan.prio = tcf_vlan_push_prio(act);
-				break;
-			case TCA_VLAN_ACT_POP:
-				entry->id = FLOW_ACTION_VLAN_POP;
-				break;
-			case TCA_VLAN_ACT_MODIFY:
-				entry->id = FLOW_ACTION_VLAN_MANGLE;
-				entry->vlan.vid = tcf_vlan_push_vid(act);
-				entry->vlan.proto = tcf_vlan_push_proto(act);
-				entry->vlan.prio = tcf_vlan_push_prio(act);
-				break;
-			default:
-				err = -EOPNOTSUPP;
-				goto err_out_locked;
-			}
-		} else if (is_tcf_tunnel_set(act)) {
-			entry->id = FLOW_ACTION_TUNNEL_ENCAP;
-			err = tcf_tunnel_encap_get_tunnel(entry, act);
-			if (err)
-				goto err_out_locked;
-		} else if (is_tcf_tunnel_release(act)) {
-			entry->id = FLOW_ACTION_TUNNEL_DECAP;
-		} else if (is_tcf_pedit(act)) {
-			for (k = 0; k < tcf_pedit_nkeys(act); k++) {
-				switch (tcf_pedit_cmd(act, k)) {
-				case TCA_PEDIT_KEY_EX_CMD_SET:
-					entry->id = FLOW_ACTION_MANGLE;
-					break;
-				case TCA_PEDIT_KEY_EX_CMD_ADD:
-					entry->id = FLOW_ACTION_ADD;
-					break;
-				default:
-					err = -EOPNOTSUPP;
-					goto err_out_locked;
-				}
-				entry->mangle.htype = tcf_pedit_htype(act, k);
-				entry->mangle.mask = tcf_pedit_mask(act, k);
-				entry->mangle.val = tcf_pedit_val(act, k);
-				entry->mangle.offset = tcf_pedit_offset(act, k);
-				entry->hw_stats = tc_act_hw_stats(act->hw_stats);
-				entry = &flow_action->entries[++j];
-			}
-		} else if (is_tcf_csum(act)) {
-			entry->id = FLOW_ACTION_CSUM;
-			entry->csum_flags = tcf_csum_update_flags(act);
-		} else if (is_tcf_skbedit_mark(act)) {
-			entry->id = FLOW_ACTION_MARK;
-			entry->mark = tcf_skbedit_mark(act);
-		} else if (is_tcf_sample(act)) {
-			entry->id = FLOW_ACTION_SAMPLE;
-			entry->sample.trunc_size = tcf_sample_trunc_size(act);
-			entry->sample.truncate = tcf_sample_truncate(act);
-			entry->sample.rate = tcf_sample_rate(act);
-			tcf_sample_get_group(entry, act);
-		} else if (is_tcf_police(act)) {
-			entry->id = FLOW_ACTION_POLICE;
-			entry->police.burst = tcf_police_burst(act);
-			entry->police.rate_bytes_ps =
-				tcf_police_rate_bytes_ps(act);
-			entry->police.burst_pkt = tcf_police_burst_pkt(act);
-			entry->police.rate_pkt_ps =
-				tcf_police_rate_pkt_ps(act);
-			entry->police.mtu = tcf_police_tcfp_mtu(act);
-			entry->police.index = act->tcfa_index;
-		} else if (is_tcf_ct(act)) {
-			entry->id = FLOW_ACTION_CT;
-			entry->ct.action = tcf_ct_action(act);
-			entry->ct.zone = tcf_ct_zone(act);
-			entry->ct.flow_table = tcf_ct_ft(act);
-		} else if (is_tcf_mpls(act)) {
-			switch (tcf_mpls_action(act)) {
-			case TCA_MPLS_ACT_PUSH:
-				entry->id = FLOW_ACTION_MPLS_PUSH;
-				entry->mpls_push.proto = tcf_mpls_proto(act);
-				entry->mpls_push.label = tcf_mpls_label(act);
-				entry->mpls_push.tc = tcf_mpls_tc(act);
-				entry->mpls_push.bos = tcf_mpls_bos(act);
-				entry->mpls_push.ttl = tcf_mpls_ttl(act);
-				break;
-			case TCA_MPLS_ACT_POP:
-				entry->id = FLOW_ACTION_MPLS_POP;
-				entry->mpls_pop.proto = tcf_mpls_proto(act);
-				break;
-			case TCA_MPLS_ACT_MODIFY:
-				entry->id = FLOW_ACTION_MPLS_MANGLE;
-				entry->mpls_mangle.label = tcf_mpls_label(act);
-				entry->mpls_mangle.tc = tcf_mpls_tc(act);
-				entry->mpls_mangle.bos = tcf_mpls_bos(act);
-				entry->mpls_mangle.ttl = tcf_mpls_ttl(act);
-				break;
-			default:
-				goto err_out_locked;
-			}
-		} else if (is_tcf_skbedit_ptype(act)) {
-			entry->id = FLOW_ACTION_PTYPE;
-			entry->ptype = tcf_skbedit_ptype(act);
-		} else if (is_tcf_skbedit_priority(act)) {
-			entry->id = FLOW_ACTION_PRIORITY;
-			entry->priority = tcf_skbedit_priority(act);
-		} else if (is_tcf_gate(act)) {
-			entry->id = FLOW_ACTION_GATE;
-			entry->gate.index = tcf_gate_index(act);
-			entry->gate.prio = tcf_gate_prio(act);
-			entry->gate.basetime = tcf_gate_basetime(act);
-			entry->gate.cycletime = tcf_gate_cycletime(act);
-			entry->gate.cycletimeext = tcf_gate_cycletimeext(act);
-			entry->gate.num_entries = tcf_gate_num_entries(act);
-			err = tcf_gate_get_entries(entry, act);
-			if (err)
-				goto err_out_locked;
-		} else {
-			err = -EOPNOTSUPP;
+		index = 0;
+		err = tc_setup_offload_act(act, entry, &index, extack);
+		if (err)
 			goto err_out_locked;
+
+		for (k = 0; k < index ; k++) {
+			entry[k].hw_stats = tc_act_hw_stats(act->hw_stats);
+			entry[k].hw_index = act->tcfa_index;
 		}
-		spin_unlock_bh(&act->tcfa_lock);
 
-		if (!is_tcf_pedit(act))
-			j++;
+		j += index;
+
+		spin_unlock_bh(&act->tcfa_lock);
 	}
 
 err_out:
 	if (err)
-		tc_cleanup_flow_action(flow_action);
+		tc_cleanup_offload_action(flow_action);
 
 	return err;
 err_out_locked:
 	spin_unlock_bh(&act->tcfa_lock);
 	goto err_out;
 }
-EXPORT_SYMBOL(tc_setup_flow_action);
+
+int tc_setup_offload_action(struct flow_action *flow_action,
+			    const struct tcf_exts *exts,
+			    struct netlink_ext_ack *extack)
+{
+#ifdef CONFIG_NET_CLS_ACT
+	if (!exts)
+		return 0;
+
+	return tc_setup_action(flow_action, exts->actions, extack);
+#else
+	return 0;
+#endif
+}
+EXPORT_SYMBOL(tc_setup_offload_action);
 
 unsigned int tcf_exts_num_actions(struct tcf_exts *exts)
 {
@@ -3772,9 +3630,6 @@ int tcf_qevent_init(struct tcf_qevent *qe, struct Qdisc *sch,
 	if (err)
 		return err;
 
-	if (!block_index)
-		return 0;
-
 	qe->info.binder_type = binder_type;
 	qe->info.chain_head_change = tcf_chain_head_change_dflt;
 	qe->info.chain_head_change_priv = &qe->filter_chain;