Netfilter/IPVS updates for net-next

index 8230fefff9f5..29b6313f0557 100644
@@ -2,10 +2,8 @@
 #ifndef _NF_SOCK_H_
 #define _NF_SOCK_H_
 
-struct net_device;
-struct sk_buff;
-struct sock;
-struct net;
+#include <net/sock.h>
+#include <net/inet_timewait_sock.h>
 
 static inline bool nf_sk_is_transparent(struct sock *sk)
 {

index f538c5001547..ad3aeff152ed 100644
@@ -7,7 +7,6 @@
  */
 
 #include <linux/types.h>
-#include <linux/module.h>
 #include <linux/atomic.h>
 #include <linux/inetdevice.h>
 #include <linux/ip.h>

@@ -157,6 +156,3 @@ void nf_nat_masquerade_ipv4_unregister_notifier(void)
 	unregister_inetaddr_notifier(&masq_inet_notifier);
 }
 EXPORT_SYMBOL_GPL(nf_nat_masquerade_ipv4_unregister_notifier);
-
-MODULE_LICENSE("GPL");
-MODULE_AUTHOR("Rusty Russell <[email protected]>");

index 9dfc2b90c362..e6eb7cf9b54f 100644
@@ -10,7 +10,6 @@
  */
 
 #include <linux/kernel.h>
-#include <linux/module.h>
 #include <linux/atomic.h>
 #include <linux/netdevice.h>
 #include <linux/ipv6.h>

@@ -186,6 +185,3 @@ void nf_nat_masquerade_ipv6_unregister_notifier(void)
 	unregister_netdevice_notifier(&masq_dev_notifier);
 }
 EXPORT_SYMBOL_GPL(nf_nat_masquerade_ipv6_unregister_notifier);
-
-MODULE_LICENSE("GPL");
-MODULE_AUTHOR("Patrick McHardy <[email protected]>");

index 7c4bb0a773ca..adee04af8d43 100644
@@ -15,7 +15,6 @@
 #include <linux/inetdevice.h>
 #include <linux/ip.h>
 #include <linux/kernel.h>
-#include <linux/module.h>
 #include <linux/netdevice.h>
 #include <linux/netfilter.h>
 #include <linux/types.h>

@@ -124,6 +123,3 @@ nf_nat_redirect_ipv6(struct sk_buff *skb, const struct nf_nat_range2 *range,
 	return nf_nat_setup_info(ct, &newrange, NF_NAT_MANIP_DST);
 }
 EXPORT_SYMBOL_GPL(nf_nat_redirect_ipv6);
-
-MODULE_LICENSE("GPL");
-MODULE_AUTHOR("Patrick McHardy <[email protected]>");

index 03ead8a9e90c..88c9e222b670 100644
@@ -441,7 +441,14 @@ static void nfnetlink_rcv_batch(struct sk_buff *skb, struct nlmsghdr *nlh,
 		kfree_skb(skb);
 		goto replay;
 	} else if (status == NFNL_BATCH_DONE) {
-		ss->commit(net, oskb);
+		err = ss->commit(net, oskb);
+		if (err == -EAGAIN) {
+			status |= NFNL_BATCH_REPLAY;
+			goto done;
+		} else if (err) {
+			ss->abort(net, oskb);
+			netlink_ack(oskb, nlmsg_hdr(oskb), err, NULL);
+		}
 	} else {
 		ss->abort(net, oskb);
 	}

index 603b51401deb..6b9ddb99f3a8 100644
@@ -858,6 +858,8 @@ enum nft_chain_flags {
  *	@name: name of the chain
  */
 struct nft_chain {
+	struct nft_rule			*__rcu *rules_gen_0;
+	struct nft_rule			*__rcu *rules_gen_1;
 	struct list_head		rules;
 	struct list_head		list;
 	struct nft_table		*table;

@@ -867,6 +869,9 @@ struct nft_chain {
 	u8				flags:6,
 					genmask:2;
 	char				*name;
+
+	/* Only used during control plane commit phase: */
+	struct nft_rule			**rules_next;
 };
 
 enum nft_chain_types {

248 lines not on this page · raw message

index 4f46d2f4e167..0548cd50ec26 100644
@@ -133,7 +133,7 @@ static noinline void nft_update_chain_stats(const struct nft_chain *chain,
 
 struct nft_jumpstack {
 	const struct nft_chain	*chain;
-	const struct nft_rule	*rule;
+	struct nft_rule	*const *rules;
 };
 
 unsigned int

@@ -141,27 +141,29 @@ nft_do_chain(struct nft_pktinfo *pkt, void *priv)
 {
 	const struct nft_chain *chain = priv, *basechain = chain;
 	const struct net *net = nft_net(pkt);
+	struct nft_rule *const *rules;
 	const struct nft_rule *rule;
 	const struct nft_expr *expr, *last;
 	struct nft_regs regs;
 	unsigned int stackptr = 0;
 	struct nft_jumpstack jumpstack[NFT_JUMP_STACK_SIZE];
-	unsigned int gencursor = nft_genmask_cur(net);
+	bool genbit = READ_ONCE(net->nft.gencursor);
 	struct nft_traceinfo info;
 
 	info.trace = false;
 	if (static_branch_unlikely(&nft_trace_enabled))
 		nft_trace_init(&info, pkt, &regs.verdict, basechain);
 do_chain:
-	rule = list_entry(&chain->rules, struct nft_rule, list);
+	if (genbit)
+		rules = rcu_dereference(chain->rules_gen_1);
+	else
+		rules = rcu_dereference(chain->rules_gen_0);
+
 next_rule:
+	rule = *rules;
 	regs.verdict.code = NFT_CONTINUE;
-	list_for_each_entry_continue_rcu(rule, &chain->rules, list) {
-
-		/* This rule is not active, skip. */
-		if (unlikely(rule->genmask & gencursor))
-			continue;
-
+	for (; *rules ; rules++) {
+		rule = *rules;
 		nft_rule_for_each_expr(expr, last, rule) {
 			if (expr->ops == &nft_cmp_fast_ops)
 				nft_cmp_fast_eval(expr, &regs);

@@ -199,7 +201,7 @@ nft_do_chain(struct nft_pktinfo *pkt, void *priv)
 	case NFT_JUMP:
 		BUG_ON(stackptr >= NFT_JUMP_STACK_SIZE);
 		jumpstack[stackptr].chain = chain;
-		jumpstack[stackptr].rule  = rule;
+		jumpstack[stackptr].rules = rules + 1;
 		stackptr++;
 		/* fall through */
 	case NFT_GOTO:

@@ -221,7 +223,7 @@ nft_do_chain(struct nft_pktinfo *pkt, void *priv)
 	if (stackptr > 0) {
 		stackptr--;
 		chain = jumpstack[stackptr].chain;
-		rule  = jumpstack[stackptr].rule;
+		rules = jumpstack[stackptr].rules;
 		goto next_rule;
 	}
 

index 821f8d835f7a..b7df32a56e7e 100644
@@ -1036,7 +1036,7 @@ static struct pernet_operations nat_net_ops = {
 	.size = sizeof(struct nat_net),
 };
 
-struct nf_nat_hook nat_hook = {
+static struct nf_nat_hook nat_hook = {
 	.parse_nat_setup	= nfnetlink_parse_nat_setup,
 #ifdef CONFIG_XFRM
 	.decode_session		= __nf_nat_decode_session,

index 1d99a1efdafc..8d1ff654e5af 100644
@@ -611,10 +611,10 @@ nfnl_compat_fill_info(struct sk_buff *skb, u32 portid, u32 seq, u32 type,
 	return -1;
 }
 
-static int nfnl_compat_get(struct net *net, struct sock *nfnl,
-			   struct sk_buff *skb, const struct nlmsghdr *nlh,
-			   const struct nlattr * const tb[],
-			   struct netlink_ext_ack *extack)
+static int nfnl_compat_get_rcu(struct net *net, struct sock *nfnl,
+			       struct sk_buff *skb, const struct nlmsghdr *nlh,
+			       const struct nlattr * const tb[],
+			       struct netlink_ext_ack *extack)
 {
 	int ret = 0, target;
 	struct nfgenmsg *nfmsg;

@@ -653,16 +653,21 @@ static int nfnl_compat_get(struct net *net, struct sock *nfnl,
 		return -EINVAL;
 	}
 
+	if (!try_module_get(THIS_MODULE))
+		return -EINVAL;
+
+	rcu_read_unlock();
 	try_then_request_module(xt_find_revision(nfmsg->nfgen_family, name,
 						 rev, target, &ret),
 						 fmt, name);
-
 	if (ret < 0)
-		return ret;
+		goto out_put;
 
 	skb2 = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL);
-	if (skb2 == NULL)
-		return -ENOMEM;
+	if (skb2 == NULL) {
+		ret = -ENOMEM;
+		goto out_put;
+	}
 
 	/* include the best revision for this extension in the message */
 	if (nfnl_compat_fill_info(skb2, NETLINK_CB(skb).portid,

@@ -672,14 +677,16 @@ static int nfnl_compat_get(struct net *net, struct sock *nfnl,
 				  nfmsg->nfgen_family,
 				  name, ret, target) <= 0) {
 		kfree_skb(skb2);
-		return -ENOSPC;
+		goto out_put;
 	}
 
 	ret = netlink_unicast(nfnl, skb2, NETLINK_CB(skb).portid,
 				MSG_DONTWAIT);
 	if (ret > 0)
 		ret = 0;
-
+out_put:
+	rcu_read_lock();
+	module_put(THIS_MODULE);
 	return ret == -EAGAIN ? -ENOBUFS : ret;
 }
 

@@ -691,7 +698,7 @@ static const struct nla_policy nfnl_compat_policy_get[NFTA_COMPAT_MAX+1] = {
 };
 
 static const struct nfnl_callback nfnl_nft_compat_cb[NFNL_MSG_COMPAT_MAX] = {
-	[NFNL_MSG_COMPAT_GET]		= { .call = nfnl_compat_get,
+	[NFNL_MSG_COMPAT_GET]		= { .call_rcu = nfnl_compat_get_rcu,
 					    .attr_count = NFTA_COMPAT_MAX,
 					    .policy = nfnl_compat_policy_get },
 };

index 583673743648..8f04bfc41bf9 100644
@@ -2830,7 +2830,7 @@ static int nf_msecs_to_jiffies64(const struct nlattr *nla, u64 *result)
 	return 0;
 }
 
-static u64 nf_jiffies64_to_msecs(u64 input)
+static __be64 nf_jiffies64_to_msecs(u64 input)
 {
 	u64 ms = jiffies64_to_nsecs(input);
 

index 88c9e222b670..5a1bd23af1a3 100644
@@ -25,6 +25,7 @@
 #include <linux/uaccess.h>
 #include <net/sock.h>
 #include <linux/init.h>
+#include <linux/sched/signal.h>
 
 #include <net/netlink.h>
 #include <linux/netfilter/nfnetlink.h>

@@ -330,6 +331,13 @@ static void nfnetlink_rcv_batch(struct sk_buff *skb, struct nlmsghdr *nlh,
 	while (skb->len >= nlmsg_total_size(0)) {
 		int msglen, type;
 
+		if (fatal_signal_pending(current)) {
+			nfnl_err_reset(&err_list);
+			err = -EINTR;
+			status = NFNL_BATCH_FAILURE;
+			goto done;
+		}
+
 		memset(&extack, 0, sizeof(extack));
 		nlh = nlmsg_hdr(skb);
 		err = 0;

index 8f04bfc41bf9..3b2ad96a9a05 100644
@@ -373,7 +373,7 @@ static struct nft_table *nft_table_lookup(const struct net *net,
 	if (nla == NULL)
 		return ERR_PTR(-EINVAL);
 
-	list_for_each_entry(table, &net->nft.tables, list) {
+	list_for_each_entry_rcu(table, &net->nft.tables, list) {
 		if (!nla_strcmp(nla, table->name) &&
 		    table->family == family &&
 		    nft_active_genmask(table, genmask))

@@ -546,6 +546,24 @@ static int nf_tables_dump_tables(struct sk_buff *skb,
 	return skb->len;
 }
 
+static int nft_netlink_dump_start_rcu(struct sock *nlsk, struct sk_buff *skb,
+				      const struct nlmsghdr *nlh,
+				      struct netlink_dump_control *c)
+{
+	int err;
+
+	if (!try_module_get(THIS_MODULE))
+		return -EINVAL;
+
+	rcu_read_unlock();
+	err = netlink_dump_start(nlsk, skb, nlh, c);
+	rcu_read_lock();
+	module_put(THIS_MODULE);
+
+	return err;
+}
+
+/* called with rcu_read_lock held */
 static int nf_tables_gettable(struct net *net, struct sock *nlsk,
 			      struct sk_buff *skb, const struct nlmsghdr *nlh,
 			      const struct nlattr * const nla[],

@@ -561,8 +579,10 @@ static int nf_tables_gettable(struct net *net, struct sock *nlsk,
 	if (nlh->nlmsg_flags & NLM_F_DUMP) {
 		struct netlink_dump_control c = {
 			.dump = nf_tables_dump_tables,
+			.module = THIS_MODULE,
 		};
-		return netlink_dump_start(nlsk, skb, nlh, &c);
+
+		return nft_netlink_dump_start_rcu(nlsk, skb, nlh, &c);
 	}
 
 	table = nft_table_lookup(net, nla[NFTA_TABLE_NAME], family, genmask);

@@ -571,7 +591,7 @@ static int nf_tables_gettable(struct net *net, struct sock *nlsk,
 		return PTR_ERR(table);
 	}
 
-	skb2 = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL);
+	skb2 = alloc_skb(NLMSG_GOODSIZE, GFP_ATOMIC);
 	if (!skb2)
 		return -ENOMEM;
 

@@ -933,7 +953,7 @@ static struct nft_chain *nft_chain_lookup(const struct nft_table *table,
 	if (nla == NULL)
 		return ERR_PTR(-EINVAL);
 
-	list_for_each_entry(chain, &table->chains, list) {
+	list_for_each_entry_rcu(chain, &table->chains, list) {
 		if (!nla_strcmp(nla, chain->name) &&
 		    nft_active_genmask(chain, genmask))
 			return chain;

@@ -1135,6 +1155,7 @@ static int nf_tables_dump_chains(struct sk_buff *skb,
 	return skb->len;
 }
 
+/* called with rcu_read_lock held */
 static int nf_tables_getchain(struct net *net, struct sock *nlsk,
 			      struct sk_buff *skb, const struct nlmsghdr *nlh,
 			      const struct nlattr * const nla[],

@@ -1151,8 +1172,10 @@ static int nf_tables_getchain(struct net *net, struct sock *nlsk,
 	if (nlh->nlmsg_flags & NLM_F_DUMP) {
 		struct netlink_dump_control c = {
 			.dump = nf_tables_dump_chains,
+			.module = THIS_MODULE,
 		};
-		return netlink_dump_start(nlsk, skb, nlh, &c);
+
+		return nft_netlink_dump_start_rcu(nlsk, skb, nlh, &c);
 	}
 
 	table = nft_table_lookup(net, nla[NFTA_CHAIN_TABLE], family, genmask);

@@ -1167,7 +1190,7 @@ static int nf_tables_getchain(struct net *net, struct sock *nlsk,
 		return PTR_ERR(chain);
 	}
 
-	skb2 = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL);
+	skb2 = alloc_skb(NLMSG_GOODSIZE, GFP_ATOMIC);
 	if (!skb2)
 		return -ENOMEM;
 

@@ -1969,7 +1992,7 @@ static struct nft_rule *__nft_rule_lookup(const struct nft_chain *chain,
 	struct nft_rule *rule;
 
 	// FIXME: this sucks
-	list_for_each_entry(rule, &chain->rules, list) {
+	list_for_each_entry_rcu(rule, &chain->rules, list) {
 		if (handle == rule->handle)
 			return rule;
 	}

@@ -2165,6 +2188,7 @@ static int nf_tables_dump_rules_done(struct netlink_callback *cb)
 	return 0;
 }
 
+/* called with rcu_read_lock held */
 static int nf_tables_getrule(struct net *net, struct sock *nlsk,
 			     struct sk_buff *skb, const struct nlmsghdr *nlh,
 			     const struct nlattr * const nla[],

@@ -2183,18 +2207,19 @@ static int nf_tables_getrule(struct net *net, struct sock *nlsk,
 		struct netlink_dump_control c = {
 			.dump = nf_tables_dump_rules,
 			.done = nf_tables_dump_rules_done,
+			.module = THIS_MODULE,
 		};
 
 		if (nla[NFTA_RULE_TABLE] || nla[NFTA_RULE_CHAIN]) {
 			struct nft_rule_dump_ctx *ctx;
 
-			ctx = kzalloc(sizeof(*ctx), GFP_KERNEL);
+			ctx = kzalloc(sizeof(*ctx), GFP_ATOMIC);
 			if (!ctx)
 				return -ENOMEM;
 
 			if (nla[NFTA_RULE_TABLE]) {
 				ctx->table = nla_strdup(nla[NFTA_RULE_TABLE],
-							GFP_KERNEL);
+							GFP_ATOMIC);
 				if (!ctx->table) {
 					kfree(ctx);
 					return -ENOMEM;

@@ -2202,7 +2227,7 @@ static int nf_tables_getrule(struct net *net, struct sock *nlsk,
 			}
 			if (nla[NFTA_RULE_CHAIN]) {
 				ctx->chain = nla_strdup(nla[NFTA_RULE_CHAIN],
-							GFP_KERNEL);
+							GFP_ATOMIC);
 				if (!ctx->chain) {
 					kfree(ctx->table);
 					kfree(ctx);

@@ -2212,7 +2237,7 @@ static int nf_tables_getrule(struct net *net, struct sock *nlsk,
 			c.data = ctx;
 		}
 
-		return netlink_dump_start(nlsk, skb, nlh, &c);
+		return nft_netlink_dump_start_rcu(nlsk, skb, nlh, &c);
 	}
 
 	table = nft_table_lookup(net, nla[NFTA_RULE_TABLE], family, genmask);

@@ -2233,7 +2258,7 @@ static int nf_tables_getrule(struct net *net, struct sock *nlsk,
 		return PTR_ERR(rule);
 	}
 
-	skb2 = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL);
+	skb2 = alloc_skb(NLMSG_GOODSIZE, GFP_ATOMIC);
 	if (!skb2)
 		return -ENOMEM;
 

@@ -2704,7 +2729,7 @@ static struct nft_set *nft_set_lookup(const struct nft_table *table,
 	if (nla == NULL)
 		return ERR_PTR(-EINVAL);
 
-	list_for_each_entry(set, &table->sets, list) {
+	list_for_each_entry_rcu(set, &table->sets, list) {
 		if (!nla_strcmp(nla, set->name) &&
 		    nft_active_genmask(set, genmask))
 			return set;

@@ -3009,6 +3034,7 @@ static int nf_tables_dump_sets_done(struct netlink_callback *cb)
 	return 0;
 }
 
+/* called with rcu_read_lock held */
 static int nf_tables_getset(struct net *net, struct sock *nlsk,
 			    struct sk_buff *skb, const struct nlmsghdr *nlh,
 			    const struct nlattr * const nla[],

@@ -3031,17 +3057,18 @@ static int nf_tables_getset(struct net *net, struct sock *nlsk,
 		struct netlink_dump_control c = {
 			.dump = nf_tables_dump_sets,
 			.done = nf_tables_dump_sets_done,
+			.module = THIS_MODULE,
 		};
 		struct nft_ctx *ctx_dump;
 
-		ctx_dump = kmalloc(sizeof(*ctx_dump), GFP_KERNEL);
+		ctx_dump = kmalloc(sizeof(*ctx_dump), GFP_ATOMIC);
 		if (ctx_dump == NULL)
 			return -ENOMEM;
 
 		*ctx_dump = ctx;
 		c.data = ctx_dump;
 
-		return netlink_dump_start(nlsk, skb, nlh, &c);
+		return nft_netlink_dump_start_rcu(nlsk, skb, nlh, &c);
 	}
 
 	/* Only accept unspec with dump */

@@ -3054,7 +3081,7 @@ static int nf_tables_getset(struct net *net, struct sock *nlsk,
 	if (IS_ERR(set))
 		return PTR_ERR(set);
 
-	skb2 = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL);
+	skb2 = alloc_skb(NLMSG_GOODSIZE, GFP_ATOMIC);
 	if (skb2 == NULL)
 		return -ENOMEM;
 

@@ -3795,7 +3822,7 @@ static int nft_get_set_elem(struct nft_ctx *ctx, struct nft_set *set,
 	ext = nft_set_elem_ext(set, &elem);
 
 	err = -ENOMEM;
-	skb = nlmsg_new(NLMSG_GOODSIZE, GFP_KERNEL);
+	skb = nlmsg_new(NLMSG_GOODSIZE, GFP_ATOMIC);
 	if (skb == NULL)
 		goto err1;
 

@@ -3817,6 +3844,7 @@ static int nft_get_set_elem(struct nft_ctx *ctx, struct nft_set *set,
 	return err == -EAGAIN ? -ENOBUFS : err;
 }
 
+/* called with rcu_read_lock held */
 static int nf_tables_getsetelem(struct net *net, struct sock *nlsk,
 				struct sk_buff *skb, const struct nlmsghdr *nlh,
 				const struct nlattr * const nla[],

@@ -3841,10 +3869,11 @@ static int nf_tables_getsetelem(struct net *net, struct sock *nlsk,
 		struct netlink_dump_control c = {
 			.dump = nf_tables_dump_set,
 			.done = nf_tables_dump_set_done,
+			.module = THIS_MODULE,
 		};
 		struct nft_set_dump_ctx *dump_ctx;
 
-		dump_ctx = kmalloc(sizeof(*dump_ctx), GFP_KERNEL);
+		dump_ctx = kmalloc(sizeof(*dump_ctx), GFP_ATOMIC);
 		if (!dump_ctx)
 			return -ENOMEM;
 

@@ -3852,7 +3881,7 @@ static int nf_tables_getsetelem(struct net *net, struct sock *nlsk,
 		dump_ctx->ctx = ctx;
 
 		c.data = dump_ctx;
-		return netlink_dump_start(nlsk, skb, nlh, &c);
+		return nft_netlink_dump_start_rcu(nlsk, skb, nlh, &c);
 	}
 
 	if (!nla[NFTA_SET_ELEM_LIST_ELEMENTS])

@@ -4475,7 +4504,7 @@ struct nft_object *nft_obj_lookup(const struct nft_table *table,
 {
 	struct nft_object *obj;
 
-	list_for_each_entry(obj, &table->objects, list) {
+	list_for_each_entry_rcu(obj, &table->objects, list) {
 		if (!nla_strcmp(nla, obj->name) &&
 		    objtype == obj->ops->type->type &&
 		    nft_active_genmask(obj, genmask))

@@ -4805,12 +4834,12 @@ nft_obj_filter_alloc(const struct nlattr * const nla[])
 {
 	struct nft_obj_filter *filter;
 
-	filter = kzalloc(sizeof(*filter), GFP_KERNEL);
+	filter = kzalloc(sizeof(*filter), GFP_ATOMIC);
 	if (!filter)
 		return ERR_PTR(-ENOMEM);
 
 	if (nla[NFTA_OBJ_TABLE]) {
-		filter->table = nla_strdup(nla[NFTA_OBJ_TABLE], GFP_KERNEL);
+		filter->table = nla_strdup(nla[NFTA_OBJ_TABLE], GFP_ATOMIC);
 		if (!filter->table) {
 			kfree(filter);
 			return ERR_PTR(-ENOMEM);

@@ -4822,6 +4851,7 @@ nft_obj_filter_alloc(const struct nlattr * const nla[])
 	return filter;
 }
 
+/* called with rcu_read_lock held */
 static int nf_tables_getobj(struct net *net, struct sock *nlsk,
 			    struct sk_buff *skb, const struct nlmsghdr *nlh,
 			    const struct nlattr * const nla[],

@@ -4841,6 +4871,7 @@ static int nf_tables_getobj(struct net *net, struct sock *nlsk,
 		struct netlink_dump_control c = {
 			.dump = nf_tables_dump_obj,
 			.done = nf_tables_dump_obj_done,
+			.module = THIS_MODULE,
 		};
 
 		if (nla[NFTA_OBJ_TABLE] ||

@@ -4853,7 +4884,7 @@ static int nf_tables_getobj(struct net *net, struct sock *nlsk,
 
 			c.data = filter;
 		}
-		return netlink_dump_start(nlsk, skb, nlh, &c);
+		return nft_netlink_dump_start_rcu(nlsk, skb, nlh, &c);
 	}
 
 	if (!nla[NFTA_OBJ_NAME] ||

@@ -4873,7 +4904,7 @@ static int nf_tables_getobj(struct net *net, struct sock *nlsk,
 		return PTR_ERR(obj);
 	}
 
-	skb2 = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL);
+	skb2 = alloc_skb(NLMSG_GOODSIZE, GFP_ATOMIC);
 	if (!skb2)
 		return -ENOMEM;
 

@@ -5018,7 +5049,7 @@ struct nft_flowtable *nft_flowtable_lookup(const struct nft_table *table,
 {
 	struct nft_flowtable *flowtable;
 
-	list_for_each_entry(flowtable, &table->flowtables, list) {
+	list_for_each_entry_rcu(flowtable, &table->flowtables, list) {
 		if (!nla_strcmp(nla, flowtable->name) &&
 		    nft_active_genmask(flowtable, genmask))
 			return flowtable;

@@ -5479,13 +5510,13 @@ nft_flowtable_filter_alloc(const struct nlattr * const nla[])
 {
 	struct nft_flowtable_filter *filter;
 
-	filter = kzalloc(sizeof(*filter), GFP_KERNEL);
+	filter = kzalloc(sizeof(*filter), GFP_ATOMIC);
 	if (!filter)
 		return ERR_PTR(-ENOMEM);
 
 	if (nla[NFTA_FLOWTABLE_TABLE]) {
 		filter->table = nla_strdup(nla[NFTA_FLOWTABLE_TABLE],
-					   GFP_KERNEL);
+					   GFP_ATOMIC);
 		if (!filter->table) {
 			kfree(filter);
 			return ERR_PTR(-ENOMEM);

@@ -5494,6 +5525,7 @@ nft_flowtable_filter_alloc(const struct nlattr * const nla[])
 	return filter;
 }
 
+/* called with rcu_read_lock held */
 static int nf_tables_getflowtable(struct net *net, struct sock *nlsk,
 				  struct sk_buff *skb,
 				  const struct nlmsghdr *nlh,

@@ -5512,6 +5544,7 @@ static int nf_tables_getflowtable(struct net *net, struct sock *nlsk,
 		struct netlink_dump_control c = {
 			.dump = nf_tables_dump_flowtable,
 			.done = nf_tables_dump_flowtable_done,
+			.module = THIS_MODULE,
 		};
 
 		if (nla[NFTA_FLOWTABLE_TABLE]) {

@@ -5523,7 +5556,7 @@ static int nf_tables_getflowtable(struct net *net, struct sock *nlsk,
 
 			c.data = filter;
 		}
-		return netlink_dump_start(nlsk, skb, nlh, &c);
+		return nft_netlink_dump_start_rcu(nlsk, skb, nlh, &c);
 	}
 
 	if (!nla[NFTA_FLOWTABLE_NAME])

@@ -5539,7 +5572,7 @@ static int nf_tables_getflowtable(struct net *net, struct sock *nlsk,
 	if (IS_ERR(flowtable))
 		return PTR_ERR(flowtable);
 
-	skb2 = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL);
+	skb2 = alloc_skb(NLMSG_GOODSIZE, GFP_ATOMIC);
 	if (!skb2)
 		return -ENOMEM;
 

@@ -5703,7 +5736,7 @@ static int nf_tables_getgen(struct net *net, struct sock *nlsk,
 	struct sk_buff *skb2;
 	int err;
 
-	skb2 = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL);
+	skb2 = alloc_skb(NLMSG_GOODSIZE, GFP_ATOMIC);
 	if (skb2 == NULL)
 		return -ENOMEM;
 

@@ -5725,7 +5758,7 @@ static const struct nfnl_callback nf_tables_cb[NFT_MSG_MAX] = {
 		.policy		= nft_table_policy,
 	},
 	[NFT_MSG_GETTABLE] = {
-		.call		= nf_tables_gettable,
+		.call_rcu	= nf_tables_gettable,
 		.attr_count	= NFTA_TABLE_MAX,
 		.policy		= nft_table_policy,
 	},

@@ -5740,7 +5773,7 @@ static const struct nfnl_callback nf_tables_cb[NFT_MSG_MAX] = {
 		.policy		= nft_chain_policy,
 	},
 	[NFT_MSG_GETCHAIN] = {
-		.call		= nf_tables_getchain,
+		.call_rcu	= nf_tables_getchain,
 		.attr_count	= NFTA_CHAIN_MAX,
 		.policy		= nft_chain_policy,
 	},

@@ -5755,7 +5788,7 @@ static const struct nfnl_callback nf_tables_cb[NFT_MSG_MAX] = {
 		.policy		= nft_rule_policy,
 	},
 	[NFT_MSG_GETRULE] = {
-		.call		= nf_tables_getrule,
+		.call_rcu	= nf_tables_getrule,
 		.attr_count	= NFTA_RULE_MAX,
 		.policy		= nft_rule_policy,
 	},

@@ -5770,7 +5803,7 @@ static const struct nfnl_callback nf_tables_cb[NFT_MSG_MAX] = {
 		.policy		= nft_set_policy,
 	},
 	[NFT_MSG_GETSET] = {
-		.call		= nf_tables_getset,
+		.call_rcu	= nf_tables_getset,
 		.attr_count	= NFTA_SET_MAX,
 		.policy		= nft_set_policy,
 	},

@@ -5785,7 +5818,7 @@ static const struct nfnl_callback nf_tables_cb[NFT_MSG_MAX] = {
 		.policy		= nft_set_elem_list_policy,
 	},
 	[NFT_MSG_GETSETELEM] = {
-		.call		= nf_tables_getsetelem,
+		.call_rcu	= nf_tables_getsetelem,
 		.attr_count	= NFTA_SET_ELEM_LIST_MAX,
 		.policy		= nft_set_elem_list_policy,
 	},

@@ -5795,7 +5828,7 @@ static const struct nfnl_callback nf_tables_cb[NFT_MSG_MAX] = {
 		.policy		= nft_set_elem_list_policy,
 	},
 	[NFT_MSG_GETGEN] = {
-		.call		= nf_tables_getgen,
+		.call_rcu	= nf_tables_getgen,
 	},
 	[NFT_MSG_NEWOBJ] = {
 		.call_batch	= nf_tables_newobj,

@@ -5803,7 +5836,7 @@ static const struct nfnl_callback nf_tables_cb[NFT_MSG_MAX] = {
 		.policy		= nft_obj_policy,
 	},
 	[NFT_MSG_GETOBJ] = {
-		.call		= nf_tables_getobj,
+		.call_rcu	= nf_tables_getobj,
 		.attr_count	= NFTA_OBJ_MAX,
 		.policy		= nft_obj_policy,
 	},

@@ -5813,7 +5846,7 @@ static const struct nfnl_callback nf_tables_cb[NFT_MSG_MAX] = {
 		.policy		= nft_obj_policy,
 	},
 	[NFT_MSG_GETOBJ_RESET] = {
-		.call		= nf_tables_getobj,
+		.call_rcu	= nf_tables_getobj,
 		.attr_count	= NFTA_OBJ_MAX,
 		.policy		= nft_obj_policy,
 	},

@@ -5823,7 +5856,7 @@ static const struct nfnl_callback nf_tables_cb[NFT_MSG_MAX] = {
 		.policy		= nft_flowtable_policy,
 	},
 	[NFT_MSG_GETFLOWTABLE] = {
-		.call		= nf_tables_getflowtable,
+		.call_rcu	= nf_tables_getflowtable,
 		.attr_count	= NFTA_FLOWTABLE_MAX,
 		.policy		= nft_flowtable_policy,
 	},

index 0548cd50ec26..47cf667b15ca 100644
@@ -23,22 +23,6 @@
 #include <net/netfilter/nf_tables.h>
 #include <net/netfilter/nf_log.h>
 
-static const char *const comments[__NFT_TRACETYPE_MAX] = {
-	[NFT_TRACETYPE_POLICY]	= "policy",
-	[NFT_TRACETYPE_RETURN]	= "return",
-	[NFT_TRACETYPE_RULE]	= "rule",
-};
-
-static const struct nf_loginfo trace_loginfo = {
-	.type = NF_LOG_TYPE_LOG,
-	.u = {
-		.log = {
-			.level = LOGLEVEL_WARNING,
-			.logflags = NF_LOG_DEFAULT_MASK,
-	        },
-	},
-};
-
 static noinline void __nft_trace_packet(struct nft_traceinfo *info,
 					const struct nft_chain *chain,
 					enum nft_trace_types type)

index f0fc21f88775..c2d237144f74 100644
@@ -177,10 +177,7 @@ static int nft_jhash_map_init(const struct nft_ctx *ctx,
 	priv->map = nft_set_lookup_global(ctx->net, ctx->table,
 					  tb[NFTA_HASH_SET_NAME],
 					  tb[NFTA_HASH_SET_ID], genmask);
-	if (IS_ERR(priv->map))
-		return PTR_ERR(priv->map);
-
-	return 0;
+	return PTR_ERR_OR_ZERO(priv->map);
 }
 
 static int nft_symhash_init(const struct nft_ctx *ctx,

@@ -220,10 +217,7 @@ static int nft_symhash_map_init(const struct nft_ctx *ctx,
 	priv->map = nft_set_lookup_global(ctx->net, ctx->table,
 					  tb[NFTA_HASH_SET_NAME],
 					  tb[NFTA_HASH_SET_ID], genmask);
-	if (IS_ERR(priv->map))
-		return PTR_ERR(priv->map);
-
-	return 0;
+	return PTR_ERR_OR_ZERO(priv->map);
 }
 
 static int nft_jhash_dump(struct sk_buff *skb,

index cdbc62a53933..1f4d0854cf70 100644
@@ -114,10 +114,7 @@ static int nft_ng_inc_map_init(const struct nft_ctx *ctx,
 					  tb[NFTA_NG_SET_NAME],
 					  tb[NFTA_NG_SET_ID], genmask);
 
-	if (IS_ERR(priv->map))
-		return PTR_ERR(priv->map);
-
-	return 0;
+	return PTR_ERR_OR_ZERO(priv->map);
 }
 
 static int nft_ng_dump(struct sk_buff *skb, enum nft_registers dreg,

index 9c71f024f9cc..3d46c82a5ebd 100644
@@ -905,6 +905,31 @@ enum nft_rt_attributes {
 #define NFTA_RT_MAX		(__NFTA_RT_MAX - 1)
 
 /**
+ * enum nft_socket_attributes - nf_tables socket expression netlink attributes
+ *
+ * @NFTA_SOCKET_KEY: socket key to match
+ * @NFTA_SOCKET_DREG: destination register
+ */
+enum nft_socket_attributes {
+	NFTA_SOCKET_UNSPEC,
+	NFTA_SOCKET_KEY,
+	NFTA_SOCKET_DREG,
+	__NFTA_SOCKET_MAX
+};
+#define NFTA_SOCKET_MAX		(__NFTA_SOCKET_MAX - 1)
+
+/*
+ * enum nft_socket_keys - nf_tables socket expression keys
+ *
+ * @NFT_SOCKET_TRANSPARENT: Value of the IP(V6)_TRANSPARENT socket option_
+ */
+enum nft_socket_keys {
+	NFT_SOCKET_TRANSPARENT,
+	__NFT_SOCKET_MAX
+};
+#define NFT_SOCKET_MAX	(__NFT_SOCKET_MAX - 1)
+
+/**
  * enum nft_ct_keys - nf_tables ct expression keys
  *
  * @NFT_CT_STATE: conntrack state (bitmask of enum ip_conntrack_info)

143 lines not on this page · raw message

index 3d46c82a5ebd..5c7eb9b9f6d6 100644
@@ -1081,6 +1081,11 @@ enum nft_log_attributes {
 #define NFTA_LOG_MAX		(__NFTA_LOG_MAX - 1)
 
 /**
+ * LOGLEVEL_AUDIT - a pseudo log level enabling audit logging
+ */
+#define LOGLEVEL_AUDIT		8
+
+/**
  * enum nft_queue_attributes - nf_tables queue expression netlink attributes
  *
  * @NFTA_QUEUE_NUM: netlink queue to send messages to (NLA_U16)

134 lines not on this page · raw message

index 34551f8aaf9d..3ecc3050be0e 100644
@@ -31,6 +31,7 @@ struct nfnetlink_subsystem {
 	const struct nfnl_callback *cb;	/* callback for individual types */
 	int (*commit)(struct net *net, struct sk_buff *skb);
 	int (*abort)(struct net *net, struct sk_buff *skb);
+	void (*cleanup)(struct net *net);
 	bool (*valid_genid)(struct net *net, u32 genid);
 };
 

index 6b9ddb99f3a8..435c32d8a995 100644
@@ -874,6 +874,8 @@ struct nft_chain {
 	struct nft_rule			**rules_next;
 };
 
+int nft_chain_validate(const struct nft_ctx *ctx, const struct nft_chain *chain);
+
 enum nft_chain_types {
 	NFT_CHAIN_T_DEFAULT = 0,
 	NFT_CHAIN_T_ROUTE,

index cd6915b6c054..e0c0c2558ec4 100644
@@ -2,6 +2,8 @@
 #ifndef _NET_NF_TABLES_CORE_H
 #define _NET_NF_TABLES_CORE_H
 
+#include <net/netfilter/nf_tables.h>
+
 extern struct nft_expr_type nft_imm_type;
 extern struct nft_expr_type nft_cmp_type;
 extern struct nft_expr_type nft_lookup_type;

@@ -23,6 +25,12 @@ struct nft_cmp_fast_expr {
 	u8			len;
 };
 
+struct nft_immediate_expr {
+	struct nft_data		data;
+	enum nft_registers	dreg:8;
+	u8			dlen;
+};
+
 /* Calculate the mask for the nft_cmp_fast expression. On big endian the
  * mask needs to include the *upper* bytes when interpreting that data as
  * something smaller than the full u32, therefore a cpu_to_le32 is done.

index 29c3851b486a..94767ea3a490 100644
@@ -9,6 +9,7 @@ struct netns_nftables {
 	struct list_head	commit_list;
 	unsigned int		base_seq;
 	u8			gencursor;
+	u8			validate_state;
 };
 
 #endif

245 lines not on this page · raw message

index 5a1bd23af1a3..261820627711 100644
@@ -460,6 +460,8 @@ static void nfnetlink_rcv_batch(struct sk_buff *skb, struct nlmsghdr *nlh,
 	} else {
 		ss->abort(net, oskb);
 	}
+	if (ss->cleanup)
+		ss->cleanup(net);
 
 	nfnl_err_deliver(&err_list, oskb);
 	nfnl_unlock(subsys_id);

index aa87ff8beae8..15adf8ca82c3 100644
@@ -17,12 +17,6 @@
 #include <net/netfilter/nf_tables_core.h>
 #include <net/netfilter/nf_tables.h>
 
-struct nft_immediate_expr {
-	struct nft_data		data;
-	enum nft_registers	dreg:8;
-	u8			dlen;
-};
-
 static void nft_immediate_eval(const struct nft_expr *expr,
 			       struct nft_regs *regs,
 			       const struct nft_pktinfo *pkt)

@@ -101,12 +95,27 @@ static int nft_immediate_dump(struct sk_buff *skb, const struct nft_expr *expr)
 
 static int nft_immediate_validate(const struct nft_ctx *ctx,
 				  const struct nft_expr *expr,
-				  const struct nft_data **data)
+				  const struct nft_data **d)
 {
 	const struct nft_immediate_expr *priv = nft_expr_priv(expr);
+	const struct nft_data *data;
+	int err;
 
-	if (priv->dreg == NFT_REG_VERDICT)
-		*data = &priv->data;
+	if (priv->dreg != NFT_REG_VERDICT)
+		return 0;
+
+	data = &priv->data;
+
+	switch (data->verdict.code) {
+	case NFT_JUMP:
+	case NFT_GOTO:
+		err = nft_chain_validate(ctx, data->verdict.chain);
+		if (err < 0)
+			return err;
+		break;
+	default:
+		break;
+	}
 
 	return 0;
 }

59 lines not on this page · raw message

index 82451b7e0acb..15ed91309992 100644
@@ -220,7 +220,7 @@ nf_flow_offload_ip_hook(void *priv, struct sk_buff *skb,
 	enum flow_offload_tuple_dir dir;
 	struct flow_offload *flow;
 	struct net_device *outdev;
-	const struct rtable *rt;
+	struct rtable *rt;
 	unsigned int thoff;
 	struct iphdr *iph;
 	__be32 nexthop;

@@ -241,7 +241,7 @@ nf_flow_offload_ip_hook(void *priv, struct sk_buff *skb,
 
 	dir = tuplehash->tuple.dir;
 	flow = container_of(tuplehash, struct flow_offload, tuplehash[dir]);
-	rt = (const struct rtable *)flow->tuplehash[dir].tuple.dst_cache;
+	rt = (struct rtable *)flow->tuplehash[dir].tuple.dst_cache;
 
 	if (unlikely(nf_flow_exceeds_mtu(skb, flow->tuplehash[dir].tuple.mtu)) &&
 	    (ip_hdr(skb)->frag_off & htons(IP_DF)) != 0)

@@ -264,6 +264,7 @@ nf_flow_offload_ip_hook(void *priv, struct sk_buff *skb,
 
 	skb->dev = outdev;
 	nexthop = rt_nexthop(rt, flow->tuplehash[!dir].tuple.src_v4.s_addr);
+	skb_dst_set_noref(skb, &rt->dst);
 	neigh_xmit(NEIGH_ARP_TABLE, outdev, &nexthop, skb);
 
 	return NF_STOLEN;

@@ -480,6 +481,7 @@ nf_flow_offload_ipv6_hook(void *priv, struct sk_buff *skb,
 
 	skb->dev = outdev;
 	nexthop = rt6_nexthop(rt, &flow->tuplehash[!dir].tuple.src_v6);
+	skb_dst_set_noref(skb, &rt->dst);
 	neigh_xmit(NEIGH_ND_TABLE, outdev, nexthop, skb);
 
 	return NF_STOLEN;

index 261820627711..4d0da7042aff 100644
@@ -38,6 +38,8 @@ MODULE_ALIAS_NET_PF_PROTO(PF_NETLINK, NETLINK_NETFILTER);
 	rcu_dereference_protected(table[(id)].subsys, \
 				  lockdep_nfnl_is_held((id)))
 
+#define NFNL_MAX_ATTR_COUNT	32
+
 static struct {
 	struct mutex				mutex;
 	const struct nfnetlink_subsystem __rcu	*subsys;

@@ -77,6 +79,13 @@ EXPORT_SYMBOL_GPL(lockdep_nfnl_is_held);
 
 int nfnetlink_subsys_register(const struct nfnetlink_subsystem *n)
 {
+	u8 cb_id;
+
+	/* Sanity-check attr_count size to avoid stack buffer overflow. */
+	for (cb_id = 0; cb_id < n->cb_count; cb_id++)
+		if (WARN_ON(n->cb[cb_id].attr_count > NFNL_MAX_ATTR_COUNT))
+			return -EINVAL;
+
 	nfnl_lock(n->subsys_id);
 	if (table[n->subsys_id].subsys) {
 		nfnl_unlock(n->subsys_id);

@@ -186,11 +195,17 @@ static int nfnetlink_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh,
 	{
 		int min_len = nlmsg_total_size(sizeof(struct nfgenmsg));
 		u8 cb_id = NFNL_MSG_TYPE(nlh->nlmsg_type);
-		struct nlattr *cda[ss->cb[cb_id].attr_count + 1];
+		struct nlattr *cda[NFNL_MAX_ATTR_COUNT + 1];
 		struct nlattr *attr = (void *)nlh + min_len;
 		int attrlen = nlh->nlmsg_len - min_len;
 		__u8 subsys_id = NFNL_SUBSYS_ID(type);
 
+		/* Sanity-check NFNL_MAX_ATTR_COUNT */
+		if (ss->cb[cb_id].attr_count > NFNL_MAX_ATTR_COUNT) {
+			rcu_read_unlock();
+			return -ENOMEM;
+		}
+
 		err = nla_parse(cda, ss->cb[cb_id].attr_count, attr, attrlen,
 				ss->cb[cb_id].policy, extack);
 		if (err < 0) {

@@ -387,10 +402,16 @@ static void nfnetlink_rcv_batch(struct sk_buff *skb, struct nlmsghdr *nlh,
 		{
 			int min_len = nlmsg_total_size(sizeof(struct nfgenmsg));
 			u8 cb_id = NFNL_MSG_TYPE(nlh->nlmsg_type);
-			struct nlattr *cda[ss->cb[cb_id].attr_count + 1];
+			struct nlattr *cda[NFNL_MAX_ATTR_COUNT + 1];
 			struct nlattr *attr = (void *)nlh + min_len;
 			int attrlen = nlh->nlmsg_len - min_len;
 
+			/* Sanity-check NFTA_MAX_ATTR */
+			if (ss->cb[cb_id].attr_count > NFNL_MAX_ATTR_COUNT) {
+				err = -ENOMEM;
+				goto ack;
+			}
+
 			err = nla_parse(cda, ss->cb[cb_id].attr_count, attr,
 					attrlen, ss->cb[cb_id].policy, NULL);
 			if (err < 0)

index 5c7eb9b9f6d6..a089af092a29 100644
@@ -1260,10 +1260,14 @@ enum nft_dup_attributes {
  * enum nft_fwd_attributes - nf_tables fwd expression netlink attributes
  *
  * @NFTA_FWD_SREG_DEV: source register of output interface (NLA_U32: nft_register)
+ * @NFTA_FWD_SREG_ADDR: source register of destination address (NLA_U32: nft_register)
+ * @NFTA_FWD_NFPROTO: layer 3 family of source register address (NLA_U32: enum nfproto)
  */
 enum nft_fwd_attributes {
 	NFTA_FWD_UNSPEC,
 	NFTA_FWD_SREG_DEV,
+	NFTA_FWD_SREG_ADDR,
+	NFTA_FWD_NFPROTO,
 	__NFTA_FWD_MAX
 };
 #define NFTA_FWD_MAX	(__NFTA_FWD_MAX - 1)

176 lines not on this page · raw message

index 6cf3fd81a5ec..eb8b9c883889 100644
@@ -67,15 +67,20 @@
 #include <net/netfilter/nf_conntrack_zones.h>
 
 
-#define FMT_TUPLE	"%pI4:%u->%pI4:%u/%u"
-#define ARG_TUPLE(T)	&(T)->src.u3.ip, ntohs((T)->src.u.all), \
-			&(T)->dst.u3.ip, ntohs((T)->dst.u.all), \
+#define FMT_TUPLE	"%s:%u->%s:%u/%u"
+#define ARG_TUPLE(T)	IP_VS_DBG_ADDR((T)->src.l3num, &(T)->src.u3),	\
+			ntohs((T)->src.u.all),				\
+			IP_VS_DBG_ADDR((T)->src.l3num, &(T)->dst.u3),	\
+			ntohs((T)->dst.u.all),				\
 			(T)->dst.protonum
 
-#define FMT_CONN	"%pI4:%u->%pI4:%u->%pI4:%u/%u:%u"
-#define ARG_CONN(C)	&((C)->caddr.ip), ntohs((C)->cport), \
-			&((C)->vaddr.ip), ntohs((C)->vport), \
-			&((C)->daddr.ip), ntohs((C)->dport), \
+#define FMT_CONN	"%s:%u->%s:%u->%s:%u/%u:%u"
+#define ARG_CONN(C)	IP_VS_DBG_ADDR((C)->af, &((C)->caddr)),		\
+			ntohs((C)->cport),				\
+			IP_VS_DBG_ADDR((C)->af, &((C)->vaddr)),		\
+			ntohs((C)->vport),				\
+			IP_VS_DBG_ADDR((C)->daf, &((C)->daddr)),	\
+			ntohs((C)->dport),				\
 			(C)->protocol, (C)->state
 
 void

@@ -127,13 +132,17 @@ ip_vs_update_conntrack(struct sk_buff *skb, struct ip_vs_conn *cp, int outin)
 		    new_tuple.dst.protonum != IPPROTO_ICMPV6)
 			new_tuple.dst.u.tcp.port = cp->vport;
 	}
-	IP_VS_DBG(7, "%s: Updating conntrack ct=%p, status=0x%lX, "
-		  "ctinfo=%d, old reply=" FMT_TUPLE
-		  ", new reply=" FMT_TUPLE ", cp=" FMT_CONN "\n",
-		  __func__, ct, ct->status, ctinfo,
-		  ARG_TUPLE(&ct->tuplehash[IP_CT_DIR_REPLY].tuple),
-		  ARG_TUPLE(&new_tuple), ARG_CONN(cp));
+	IP_VS_DBG_BUF(7, "%s: Updating conntrack ct=%p, status=0x%lX, "
+		      "ctinfo=%d, old reply=" FMT_TUPLE "\n",
+		      __func__, ct, ct->status, ctinfo,
+		      ARG_TUPLE(&ct->tuplehash[IP_CT_DIR_REPLY].tuple));
+	IP_VS_DBG_BUF(7, "%s: Updating conntrack ct=%p, status=0x%lX, "
+		      "ctinfo=%d, new reply=" FMT_TUPLE "\n",
+		      __func__, ct, ct->status, ctinfo,
+		      ARG_TUPLE(&new_tuple));
 	nf_conntrack_alter_reply(ct, &new_tuple);
+	IP_VS_DBG_BUF(7, "%s: Updated conntrack ct=%p for cp=" FMT_CONN "\n",
+		      __func__, ct, ARG_CONN(cp));
 }
 
 int ip_vs_confirm_conntrack(struct sk_buff *skb)

@@ -152,9 +161,6 @@ static void ip_vs_nfct_expect_callback(struct nf_conn *ct,
 	struct ip_vs_conn_param p;
 	struct net *net = nf_ct_net(ct);
 
-	if (exp->tuple.src.l3num != PF_INET)
-		return;
-
 	/*
 	 * We assume that no NF locks are held before this callback.
 	 * ip_vs_conn_out_get and ip_vs_conn_in_get should match their

@@ -171,19 +177,15 @@ static void ip_vs_nfct_expect_callback(struct nf_conn *ct,
 	cp = ip_vs_conn_out_get(&p);
 	if (cp) {
 		/* Change reply CLIENT->RS to CLIENT->VS */
+		IP_VS_DBG_BUF(7, "%s: for ct=%p, status=0x%lX found inout cp="
+			      FMT_CONN "\n",
+			      __func__, ct, ct->status, ARG_CONN(cp));
 		new_reply = ct->tuplehash[IP_CT_DIR_REPLY].tuple;
-		IP_VS_DBG(7, "%s: ct=%p, status=0x%lX, tuples=" FMT_TUPLE ", "
-			  FMT_TUPLE ", found inout cp=" FMT_CONN "\n",
-			  __func__, ct, ct->status,
-			  ARG_TUPLE(orig), ARG_TUPLE(&new_reply),
-			  ARG_CONN(cp));
+		IP_VS_DBG_BUF(7, "%s: ct=%p before alter: reply tuple="
+			      FMT_TUPLE "\n",
+			      __func__, ct, ARG_TUPLE(&new_reply));
 		new_reply.dst.u3 = cp->vaddr;
 		new_reply.dst.u.tcp.port = cp->vport;
-		IP_VS_DBG(7, "%s: ct=%p, new tuples=" FMT_TUPLE ", " FMT_TUPLE
-			  ", inout cp=" FMT_CONN "\n",
-			  __func__, ct,
-			  ARG_TUPLE(orig), ARG_TUPLE(&new_reply),
-			  ARG_CONN(cp));
 		goto alter;
 	}
 

@@ -191,25 +193,21 @@ static void ip_vs_nfct_expect_callback(struct nf_conn *ct,
 	cp = ip_vs_conn_in_get(&p);
 	if (cp) {
 		/* Change reply VS->CLIENT to RS->CLIENT */
+		IP_VS_DBG_BUF(7, "%s: for ct=%p, status=0x%lX found outin cp="
+			      FMT_CONN "\n",
+			      __func__, ct, ct->status, ARG_CONN(cp));
 		new_reply = ct->tuplehash[IP_CT_DIR_REPLY].tuple;
-		IP_VS_DBG(7, "%s: ct=%p, status=0x%lX, tuples=" FMT_TUPLE ", "
-			  FMT_TUPLE ", found outin cp=" FMT_CONN "\n",
-			  __func__, ct, ct->status,
-			  ARG_TUPLE(orig), ARG_TUPLE(&new_reply),
-			  ARG_CONN(cp));
+		IP_VS_DBG_BUF(7, "%s: ct=%p before alter: reply tuple="
+			      FMT_TUPLE "\n",
+			      __func__, ct, ARG_TUPLE(&new_reply));
 		new_reply.src.u3 = cp->daddr;
 		new_reply.src.u.tcp.port = cp->dport;
-		IP_VS_DBG(7, "%s: ct=%p, new tuples=" FMT_TUPLE ", "
-			  FMT_TUPLE ", outin cp=" FMT_CONN "\n",
-			  __func__, ct,
-			  ARG_TUPLE(orig), ARG_TUPLE(&new_reply),
-			  ARG_CONN(cp));
 		goto alter;
 	}
 
-	IP_VS_DBG(7, "%s: ct=%p, status=0x%lX, tuple=" FMT_TUPLE
-		  " - unknown expect\n",
-		  __func__, ct, ct->status, ARG_TUPLE(orig));
+	IP_VS_DBG_BUF(7, "%s: ct=%p, status=0x%lX, tuple=" FMT_TUPLE
+		      " - unknown expect\n",
+		      __func__, ct, ct->status, ARG_TUPLE(orig));
 	return;
 
 alter:

@@ -247,8 +245,8 @@ void ip_vs_nfct_expect_related(struct sk_buff *skb, struct nf_conn *ct,
 
 	exp->expectfn = ip_vs_nfct_expect_callback;
 
-	IP_VS_DBG(7, "%s: ct=%p, expect tuple=" FMT_TUPLE "\n",
-		__func__, ct, ARG_TUPLE(&exp->tuple));
+	IP_VS_DBG_BUF(7, "%s: ct=%p, expect tuple=" FMT_TUPLE "\n",
+		      __func__, ct, ARG_TUPLE(&exp->tuple));
 	nf_ct_expect_related(exp);
 	nf_ct_expect_put(exp);
 }

@@ -274,26 +272,25 @@ void ip_vs_conn_drop_conntrack(struct ip_vs_conn *cp)
 	tuple.dst.u3 = cp->vaddr;
 	tuple.dst.u.all = cp->vport;
 
-	IP_VS_DBG(7, "%s: dropping conntrack with tuple=" FMT_TUPLE
-		" for conn " FMT_CONN "\n",
-		__func__, ARG_TUPLE(&tuple), ARG_CONN(cp));
+	IP_VS_DBG_BUF(7, "%s: dropping conntrack for conn " FMT_CONN "\n",
+		      __func__, ARG_CONN(cp));
 
 	h = nf_conntrack_find_get(cp->ipvs->net, &nf_ct_zone_dflt, &tuple);
 	if (h) {
 		ct = nf_ct_tuplehash_to_ctrack(h);
 		if (nf_ct_kill(ct)) {
-			IP_VS_DBG(7, "%s: ct=%p, deleted conntrack for tuple="
-				FMT_TUPLE "\n",
-				__func__, ct, ARG_TUPLE(&tuple));
+			IP_VS_DBG_BUF(7, "%s: ct=%p deleted for tuple="
+				      FMT_TUPLE "\n",
+				      __func__, ct, ARG_TUPLE(&tuple));
 		} else {
-			IP_VS_DBG(7, "%s: ct=%p, no conntrack timer for tuple="
-				FMT_TUPLE "\n",
-				__func__, ct, ARG_TUPLE(&tuple));
+			IP_VS_DBG_BUF(7, "%s: ct=%p, no conntrack for tuple="
+				      FMT_TUPLE "\n",
+				      __func__, ct, ARG_TUPLE(&tuple));
 		}
 		nf_ct_put(ct);
 	} else {
-		IP_VS_DBG(7, "%s: no conntrack for tuple=" FMT_TUPLE "\n",
-			__func__, ARG_TUPLE(&tuple));
+		IP_VS_DBG_BUF(7, "%s: no conntrack for tuple=" FMT_TUPLE "\n",
+			      __func__, ARG_TUPLE(&tuple));
 	}
 }
 

index 0ac795b41ab8..03f567eb9536 100644
@@ -763,14 +763,14 @@ struct ip_vs_app {
 	 *	   2=Mangled but checksum was not updated
 	 */
 	int (*pkt_out)(struct ip_vs_app *, struct ip_vs_conn *,
-		       struct sk_buff *, int *diff);
+		       struct sk_buff *, int *diff, struct ip_vs_iphdr *ipvsh);
 
 	/* input hook: Process packet in outin direction, diff set for TCP.
 	 * Return: 0=Error, 1=Payload Not Mangled/Mangled but checksum is ok,
 	 *	   2=Mangled but checksum was not updated
 	 */
 	int (*pkt_in)(struct ip_vs_app *, struct ip_vs_conn *,
-		      struct sk_buff *, int *diff);
+		      struct sk_buff *, int *diff, struct ip_vs_iphdr *ipvsh);
 
 	/* ip_vs_app initializer */
 	int (*init_conn)(struct ip_vs_app *, struct ip_vs_conn *);

@@ -1328,8 +1328,10 @@ int register_ip_vs_app_inc(struct netns_ipvs *ipvs, struct ip_vs_app *app, __u16
 int ip_vs_app_inc_get(struct ip_vs_app *inc);
 void ip_vs_app_inc_put(struct ip_vs_app *inc);
 
-int ip_vs_app_pkt_out(struct ip_vs_conn *, struct sk_buff *skb);
-int ip_vs_app_pkt_in(struct ip_vs_conn *, struct sk_buff *skb);
+int ip_vs_app_pkt_out(struct ip_vs_conn *, struct sk_buff *skb,
+		      struct ip_vs_iphdr *ipvsh);
+int ip_vs_app_pkt_in(struct ip_vs_conn *, struct sk_buff *skb,
+		     struct ip_vs_iphdr *ipvsh);
 
 int register_ip_vs_pe(struct ip_vs_pe *pe);
 int unregister_ip_vs_pe(struct ip_vs_pe *pe);

index 1c98c907bc63..12d74896556a 100644
@@ -355,7 +355,8 @@ static inline void vs_seq_update(struct ip_vs_conn *cp, struct ip_vs_seq *vseq,
 }
 
 static inline int app_tcp_pkt_out(struct ip_vs_conn *cp, struct sk_buff *skb,
-				  struct ip_vs_app *app)
+				  struct ip_vs_app *app,
+				  struct ip_vs_iphdr *ipvsh)
 {
 	int diff;
 	const unsigned int tcp_offset = ip_hdrlen(skb);

@@ -386,7 +387,7 @@ static inline int app_tcp_pkt_out(struct ip_vs_conn *cp, struct sk_buff *skb,
 	if (app->pkt_out == NULL)
 		return 1;
 
-	if (!app->pkt_out(app, cp, skb, &diff))
+	if (!app->pkt_out(app, cp, skb, &diff, ipvsh))
 		return 0;
 
 	/*

@@ -404,7 +405,8 @@ static inline int app_tcp_pkt_out(struct ip_vs_conn *cp, struct sk_buff *skb,
  *	called by ipvs packet handler, assumes previously checked cp!=NULL
  *	returns false if it can't handle packet (oom)
  */
-int ip_vs_app_pkt_out(struct ip_vs_conn *cp, struct sk_buff *skb)
+int ip_vs_app_pkt_out(struct ip_vs_conn *cp, struct sk_buff *skb,
+		      struct ip_vs_iphdr *ipvsh)
 {
 	struct ip_vs_app *app;
 

@@ -417,7 +419,7 @@ int ip_vs_app_pkt_out(struct ip_vs_conn *cp, struct sk_buff *skb)
 
 	/* TCP is complicated */
 	if (cp->protocol == IPPROTO_TCP)
-		return app_tcp_pkt_out(cp, skb, app);
+		return app_tcp_pkt_out(cp, skb, app, ipvsh);
 
 	/*
 	 *	Call private output hook function

@@ -425,12 +427,13 @@ int ip_vs_app_pkt_out(struct ip_vs_conn *cp, struct sk_buff *skb)
 	if (app->pkt_out == NULL)
 		return 1;
 
-	return app->pkt_out(app, cp, skb, NULL);
+	return app->pkt_out(app, cp, skb, NULL, ipvsh);
 }
 
 
 static inline int app_tcp_pkt_in(struct ip_vs_conn *cp, struct sk_buff *skb,
-				 struct ip_vs_app *app)
+				 struct ip_vs_app *app,
+				 struct ip_vs_iphdr *ipvsh)
 {
 	int diff;
 	const unsigned int tcp_offset = ip_hdrlen(skb);

@@ -461,7 +464,7 @@ static inline int app_tcp_pkt_in(struct ip_vs_conn *cp, struct sk_buff *skb,
 	if (app->pkt_in == NULL)
 		return 1;
 
-	if (!app->pkt_in(app, cp, skb, &diff))
+	if (!app->pkt_in(app, cp, skb, &diff, ipvsh))
 		return 0;
 
 	/*

@@ -479,7 +482,8 @@ static inline int app_tcp_pkt_in(struct ip_vs_conn *cp, struct sk_buff *skb,
  *	called by ipvs packet handler, assumes previously checked cp!=NULL.
  *	returns false if can't handle packet (oom).
  */
-int ip_vs_app_pkt_in(struct ip_vs_conn *cp, struct sk_buff *skb)
+int ip_vs_app_pkt_in(struct ip_vs_conn *cp, struct sk_buff *skb,
+		     struct ip_vs_iphdr *ipvsh)
 {
 	struct ip_vs_app *app;
 

@@ -492,7 +496,7 @@ int ip_vs_app_pkt_in(struct ip_vs_conn *cp, struct sk_buff *skb)
 
 	/* TCP is complicated */
 	if (cp->protocol == IPPROTO_TCP)
-		return app_tcp_pkt_in(cp, skb, app);
+		return app_tcp_pkt_in(cp, skb, app, ipvsh);
 
 	/*
 	 *	Call private input hook function

@@ -500,7 +504,7 @@ int ip_vs_app_pkt_in(struct ip_vs_conn *cp, struct sk_buff *skb)
 	if (app->pkt_in == NULL)
 		return 1;
 
-	return app->pkt_in(app, cp, skb, NULL);
+	return app->pkt_in(app, cp, skb, NULL, ipvsh);
 }
 
 

618 lines not on this page · raw message

index eff7569824e5..3250c4a1111e 100644
@@ -109,7 +109,7 @@ sctp_snat_handler(struct sk_buff *skb, struct ip_vs_protocol *pp,
 			return 0;
 
 		/* Call application helper if needed */
-		ret = ip_vs_app_pkt_out(cp, skb);
+		ret = ip_vs_app_pkt_out(cp, skb, iph);
 		if (ret == 0)
 			return 0;
 		/* ret=2: csum update is needed after payload mangling */

@@ -156,7 +156,7 @@ sctp_dnat_handler(struct sk_buff *skb, struct ip_vs_protocol *pp,
 			return 0;
 
 		/* Call application helper if needed */
-		ret = ip_vs_app_pkt_in(cp, skb);
+		ret = ip_vs_app_pkt_in(cp, skb, iph);
 		if (ret == 0)
 			return 0;
 		/* ret=2: csum update is needed after payload mangling */

index 569631d2b2a1..80d10ad12a15 100644
@@ -170,7 +170,7 @@ tcp_snat_handler(struct sk_buff *skb, struct ip_vs_protocol *pp,
 			return 0;
 
 		/* Call application helper if needed */
-		if (!(ret = ip_vs_app_pkt_out(cp, skb)))
+		if (!(ret = ip_vs_app_pkt_out(cp, skb, iph)))
 			return 0;
 		/* ret=2: csum update is needed after payload mangling */
 		if (ret == 1)

@@ -251,7 +251,7 @@ tcp_dnat_handler(struct sk_buff *skb, struct ip_vs_protocol *pp,
 		 *	Attempt ip_vs_app call.
 		 *	It will fix ip_vs_conn and iph ack_seq stuff
 		 */
-		if (!(ret = ip_vs_app_pkt_in(cp, skb)))
+		if (!(ret = ip_vs_app_pkt_in(cp, skb, iph)))
 			return 0;
 		/* ret=2: csum update is needed after payload mangling */
 		if (ret == 1)

index c15ef7c2a1fa..e0ef11c3691e 100644
@@ -162,7 +162,7 @@ udp_snat_handler(struct sk_buff *skb, struct ip_vs_protocol *pp,
 		/*
 		 *	Call application helper if needed
 		 */
-		if (!(ret = ip_vs_app_pkt_out(cp, skb)))
+		if (!(ret = ip_vs_app_pkt_out(cp, skb, iph)))
 			return 0;
 		/* ret=2: csum update is needed after payload mangling */
 		if (ret == 1)

@@ -246,7 +246,7 @@ udp_dnat_handler(struct sk_buff *skb, struct ip_vs_protocol *pp,
 		 *	Attempt ip_vs_app call.
 		 *	It will fix ip_vs_conn
 		 */
-		if (!(ret = ip_vs_app_pkt_in(cp, skb)))
+		if (!(ret = ip_vs_app_pkt_in(cp, skb, iph)))
 			return 0;
 		/* ret=2: csum update is needed after payload mangling */
 		if (ret == 1)