1 files changed, 147 insertions, 162 deletions

diff --git a/net/ipv4/fib_rules.c b/net/ipv4/fib_rules.c
index 52b2adae4f2..f2e15738534 100644
--- a/net/ipv4/fib_rules.c
+++ b/net/ipv4/fib_rules.c
@@ -6,7 +6,7 @@
  *		IPv4 Forwarding Information Base: policy rules.
  *
  * Authors:	Alexey Kuznetsov, <kuznet@ms2.inr.ac.ru>
- * 		Thomas Graf <tgraf@suug.ch>
+ *		Thomas Graf <tgraf@suug.ch>
  *
  *		This program is free software; you can redistribute it and/or
  *		modify it under the terms of the GNU General Public License
@@ -14,7 +14,7 @@
  *		2 of the License, or (at your option) any later version.
  *
  * Fixes:
- * 		Rani Assaf	:	local_rule cannot be deleted
+ *		Rani Assaf	:	local_rule cannot be deleted
  *		Marc Boucher	:	routing by fwmark
  */
 
@@ -26,81 +26,45 @@
 #include <linux/init.h>
 #include <linux/list.h>
 #include <linux/rcupdate.h>
+#include <linux/export.h>
 #include <net/ip.h>
 #include <net/route.h>
 #include <net/tcp.h>
 #include <net/ip_fib.h>
 #include <net/fib_rules.h>
 
-static struct fib_rules_ops fib4_rules_ops;
-
-struct fib4_rule
-{
+struct fib4_rule {
 	struct fib_rule		common;
 	u8			dst_len;
 	u8			src_len;
 	u8			tos;
-	u32			src;
-	u32			srcmask;
-	u32			dst;
-	u32			dstmask;
-#ifdef CONFIG_IP_ROUTE_FWMARK
-	u32			fwmark;
-	u32			fwmask;
-#endif
-#ifdef CONFIG_NET_CLS_ROUTE
+	__be32			src;
+	__be32			srcmask;
+	__be32			dst;
+	__be32			dstmask;
+#ifdef CONFIG_IP_ROUTE_CLASSID
 	u32			tclassid;
 #endif
 };
 
-static struct fib4_rule default_rule = {
-	.common = {
-		.refcnt =	ATOMIC_INIT(2),
-		.pref =		0x7FFF,
-		.table =	RT_TABLE_DEFAULT,
-		.action =	FR_ACT_TO_TBL,
-	},
-};
-
-static struct fib4_rule main_rule = {
-	.common = {
-		.refcnt =	ATOMIC_INIT(2),
-		.pref =		0x7FFE,
-		.table =	RT_TABLE_MAIN,
-		.action =	FR_ACT_TO_TBL,
-	},
-};
-
-static struct fib4_rule local_rule = {
-	.common = {
-		.refcnt =	ATOMIC_INIT(2),
-		.table =	RT_TABLE_LOCAL,
-		.action =	FR_ACT_TO_TBL,
-		.flags =	FIB_RULE_PERMANENT,
-	},
-};
-
-static LIST_HEAD(fib4_rules);
-
-#ifdef CONFIG_NET_CLS_ROUTE
-u32 fib_rules_tclass(struct fib_result *res)
-{
-	return res->r ? ((struct fib4_rule *) res->r)->tclassid : 0;
-}
-#endif
-
-int fib_lookup(struct flowi *flp, struct fib_result *res)
+int __fib_lookup(struct net *net, struct flowi4 *flp, struct fib_result *res)
 {
 	struct fib_lookup_arg arg = {
 		.result = res,
+		.flags = FIB_LOOKUP_NOREF,
 	};
 	int err;
 
-	err = fib_rules_lookup(&fib4_rules_ops, flp, 0, &arg);
-	res->r = arg.rule;
-
+	err = fib_rules_lookup(net->ipv4.rules_ops, flowi4_to_flowi(flp), 0, &arg);
+#ifdef CONFIG_IP_ROUTE_CLASSID
+	if (arg.rule)
+		res->tclassid = ((struct fib4_rule *)arg.rule)->tclassid;
+	else
+		res->tclassid = 0;
+#endif
 	return err;
 }
+EXPORT_SYMBOL_GPL(__fib_lookup);
 
 static int fib4_rule_action(struct fib_rule *rule, struct flowi *flp,
 			    int flags, struct fib_lookup_arg *arg)
@@ -126,85 +90,93 @@ static int fib4_rule_action(struct fib_rule *rule, struct flowi *flp,
 		goto errout;
 	}
 
-	if ((tbl = fib_get_table(rule->table)) == NULL)
+	tbl = fib_get_table(rule->fr_net, rule->table);
+	if (!tbl)
 		goto errout;
 
-	err = tbl->tb_lookup(tbl, flp, (struct fib_result *) arg->result);
+	err = fib_table_lookup(tbl, &flp->u.ip4, (struct fib_result *) arg->result, arg->flags);
 	if (err > 0)
 		err = -EAGAIN;
 errout:
 	return err;
 }
 
-
-void fib_select_default(const struct flowi *flp, struct fib_result *res)
+static bool fib4_rule_suppress(struct fib_rule *rule, struct fib_lookup_arg *arg)
 {
-	if (res->r && res->r->action == FR_ACT_TO_TBL &&
-	    FIB_RES_GW(*res) && FIB_RES_NH(*res).nh_scope == RT_SCOPE_LINK) {
-		struct fib_table *tb;
-		if ((tb = fib_get_table(res->r->table)) != NULL)
-			tb->tb_select_default(tb, flp, res);
-	}
+	struct fib_result *result = (struct fib_result *) arg->result;
+	struct net_device *dev = NULL;
+
+	if (result->fi)
+		dev = result->fi->fib_dev;
+
+	/* do not accept result if the route does
+	 * not meet the required prefix length
+	 */
+	if (result->prefixlen <= rule->suppress_prefixlen)
+		goto suppress_route;
+
+	/* do not accept result if the route uses a device
+	 * belonging to a forbidden interface group
+	 */
+	if (rule->suppress_ifgroup != -1 && dev && dev->group == rule->suppress_ifgroup)
+		goto suppress_route;
+
+	return false;
+
+suppress_route:
+	if (!(arg->flags & FIB_LOOKUP_NOREF))
+		fib_info_put(result->fi);
+	return true;
 }
 
 static int fib4_rule_match(struct fib_rule *rule, struct flowi *fl, int flags)
 {
 	struct fib4_rule *r = (struct fib4_rule *) rule;
-	u32 daddr = fl->fl4_dst;
-	u32 saddr = fl->fl4_src;
+	struct flowi4 *fl4 = &fl->u.ip4;
+	__be32 daddr = fl4->daddr;
+	__be32 saddr = fl4->saddr;
 
 	if (((saddr ^ r->src) & r->srcmask) ||
 	    ((daddr ^ r->dst) & r->dstmask))
 		return 0;
 
-	if (r->tos && (r->tos != fl->fl4_tos))
-		return 0;
-
-#ifdef CONFIG_IP_ROUTE_FWMARK
-	if ((r->fwmark ^ fl->fl4_fwmark) & r->fwmask)
+	if (r->tos && (r->tos != fl4->flowi4_tos))
 		return 0;
-#endif
 
 	return 1;
 }
 
-static struct fib_table *fib_empty_table(void)
+static struct fib_table *fib_empty_table(struct net *net)
 {
 	u32 id;
 
 	for (id = 1; id <= RT_TABLE_MAX; id++)
-		if (fib_get_table(id) == NULL)
-			return fib_new_table(id);
+		if (fib_get_table(net, id) == NULL)
+			return fib_new_table(net, id);
 	return NULL;
 }
 
-static struct nla_policy fib4_rule_policy[FRA_MAX+1] __read_mostly = {
-	[FRA_IFNAME]	= { .type = NLA_STRING, .len = IFNAMSIZ - 1 },
-	[FRA_PRIORITY]	= { .type = NLA_U32 },
-	[FRA_SRC]	= { .type = NLA_U32 },
-	[FRA_DST]	= { .type = NLA_U32 },
-	[FRA_FWMARK]	= { .type = NLA_U32 },
-	[FRA_FWMASK]	= { .type = NLA_U32 },
+static const struct nla_policy fib4_rule_policy[FRA_MAX+1] = {
+	FRA_GENERIC_POLICY,
 	[FRA_FLOW]	= { .type = NLA_U32 },
-	[FRA_TABLE]	= { .type = NLA_U32 },
 };
 
 static int fib4_rule_configure(struct fib_rule *rule, struct sk_buff *skb,
-			       struct nlmsghdr *nlh, struct fib_rule_hdr *frh,
+			       struct fib_rule_hdr *frh,
 			       struct nlattr **tb)
 {
+	struct net *net = sock_net(skb->sk);
 	int err = -EINVAL;
 	struct fib4_rule *rule4 = (struct fib4_rule *) rule;
 
-	if (frh->src_len > 32 || frh->dst_len > 32 ||
-	    (frh->tos & ~IPTOS_TOS_MASK))
+	if (frh->tos & ~IPTOS_TOS_MASK)
 		goto errout;
 
 	if (rule->table == RT_TABLE_UNSPEC) {
 		if (rule->action == FR_ACT_TO_TBL) {
 			struct fib_table *table;
 
-			table = fib_empty_table();
+			table = fib_empty_table(net);
 			if (table == NULL) {
 				err = -ENOBUFS;
 				goto errout;
@@ -214,29 +186,18 @@ static int fib4_rule_configure(struct fib_rule *rule, struct sk_buff *skb,
 		}
 	}
 
-	if (tb[FRA_SRC])
-		rule4->src = nla_get_u32(tb[FRA_SRC]);
-
-	if (tb[FRA_DST])
-		rule4->dst = nla_get_u32(tb[FRA_DST]);
+	if (frh->src_len)
+		rule4->src = nla_get_be32(tb[FRA_SRC]);
 
-#ifdef CONFIG_IP_ROUTE_FWMARK
-	if (tb[FRA_FWMARK]) {
-		rule4->fwmark = nla_get_u32(tb[FRA_FWMARK]);
-		if (rule4->fwmark)
-			/* compatibility: if the mark value is non-zero all bits
-			 * are compared unless a mask is explicitly specified.
-			 */
-			rule4->fwmask = 0xFFFFFFFF;
-	}
+	if (frh->dst_len)
+		rule4->dst = nla_get_be32(tb[FRA_DST]);
 
-	if (tb[FRA_FWMASK])
-		rule4->fwmask = nla_get_u32(tb[FRA_FWMASK]);
-#endif
-
-#ifdef CONFIG_NET_CLS_ROUTE
-	if (tb[FRA_FLOW])
+#ifdef CONFIG_IP_ROUTE_CLASSID
+	if (tb[FRA_FLOW]) {
 		rule4->tclassid = nla_get_u32(tb[FRA_FLOW]);
+		if (rule4->tclassid)
+			net->ipv4.fib_num_tclassid_users++;
+	}
 #endif
 
 	rule4->src_len = frh->src_len;
@@ -245,11 +206,24 @@ static int fib4_rule_configure(struct fib_rule *rule, struct sk_buff *skb,
 	rule4->dstmask = inet_make_mask(rule4->dst_len);
 	rule4->tos = frh->tos;
 
+	net->ipv4.fib_has_custom_rules = true;
 	err = 0;
 errout:
 	return err;
 }
 
+static void fib4_rule_delete(struct fib_rule *rule)
+{
+	struct net *net = rule->fr_net;
+#ifdef CONFIG_IP_ROUTE_CLASSID
+	struct fib4_rule *rule4 = (struct fib4_rule *) rule;
+
+	if (rule4->tclassid)
+		net->ipv4.fib_num_tclassid_users--;
+#endif
+	net->ipv4.fib_has_custom_rules = true;
+}
+
 static int fib4_rule_compare(struct fib_rule *rule, struct fib_rule_hdr *frh,
 			     struct nlattr **tb)
 {
@@ -264,55 +238,38 @@ static int fib4_rule_compare(struct fib_rule *rule, struct fib_rule_hdr *frh,
 	if (frh->tos && (rule4->tos != frh->tos))
 		return 0;
 
-#ifdef CONFIG_IP_ROUTE_FWMARK
-	if (tb[FRA_FWMARK] && (rule4->fwmark != nla_get_u32(tb[FRA_FWMARK])))
-		return 0;
-
-	if (tb[FRA_FWMASK] && (rule4->fwmask != nla_get_u32(tb[FRA_FWMASK])))
-		return 0;
-#endif
-
-#ifdef CONFIG_NET_CLS_ROUTE
+#ifdef CONFIG_IP_ROUTE_CLASSID
 	if (tb[FRA_FLOW] && (rule4->tclassid != nla_get_u32(tb[FRA_FLOW])))
 		return 0;
 #endif
 
-	if (tb[FRA_SRC] && (rule4->src != nla_get_u32(tb[FRA_SRC])))
+	if (frh->src_len && (rule4->src != nla_get_be32(tb[FRA_SRC])))
 		return 0;
 
-	if (tb[FRA_DST] && (rule4->dst != nla_get_u32(tb[FRA_DST])))
+	if (frh->dst_len && (rule4->dst != nla_get_be32(tb[FRA_DST])))
 		return 0;
 
 	return 1;
 }
 
 static int fib4_rule_fill(struct fib_rule *rule, struct sk_buff *skb,
-			  struct nlmsghdr *nlh, struct fib_rule_hdr *frh)
+			  struct fib_rule_hdr *frh)
 {
 	struct fib4_rule *rule4 = (struct fib4_rule *) rule;
 
-	frh->family = AF_INET;
 	frh->dst_len = rule4->dst_len;
 	frh->src_len = rule4->src_len;
 	frh->tos = rule4->tos;
 
-#ifdef CONFIG_IP_ROUTE_FWMARK
-	if (rule4->fwmark)
-		NLA_PUT_U32(skb, FRA_FWMARK, rule4->fwmark);
-
-	if (rule4->fwmask || rule4->fwmark)
-		NLA_PUT_U32(skb, FRA_FWMASK, rule4->fwmask);
-#endif
-
-	if (rule4->dst_len)
-		NLA_PUT_U32(skb, FRA_DST, rule4->dst);
-
-	if (rule4->src_len)
-		NLA_PUT_U32(skb, FRA_SRC, rule4->src);
-
-#ifdef CONFIG_NET_CLS_ROUTE
-	if (rule4->tclassid)
-		NLA_PUT_U32(skb, FRA_FLOW, rule4->tclassid);
+	if ((rule4->dst_len &&
+	     nla_put_be32(skb, FRA_DST, rule4->dst)) ||
+	    (rule4->src_len &&
+	     nla_put_be32(skb, FRA_SRC, rule4->src)))
+		goto nla_put_failure;
+#ifdef CONFIG_IP_ROUTE_CLASSID
+	if (rule4->tclassid &&
+	    nla_put_u32(skb, FRA_FLOW, rule4->tclassid))
+		goto nla_put_failure;
 #endif
 	return 0;
 
@@ -320,48 +277,76 @@ nla_put_failure:
 	return -ENOBUFS;
 }
 
-int fib4_rules_dump(struct sk_buff *skb, struct netlink_callback *cb)
+static size_t fib4_rule_nlmsg_payload(struct fib_rule *rule)
 {
-	return fib_rules_dump(skb, cb, AF_INET);
+	return nla_total_size(4) /* dst */
+	       + nla_total_size(4) /* src */
+	       + nla_total_size(4); /* flow */
 }
 
-static u32 fib4_rule_default_pref(void)
+static void fib4_rule_flush_cache(struct fib_rules_ops *ops)
 {
-	struct list_head *pos;
-	struct fib_rule *rule;
-
-	if (!list_empty(&fib4_rules)) {
-		pos = fib4_rules.next;
-		if (pos->next != &fib4_rules) {
-			rule = list_entry(pos->next, struct fib_rule, list);
-			if (rule->pref)
-				return rule->pref - 1;
-		}
-	}
-
-	return 0;
+	rt_cache_flush(ops->fro_net);
 }
 
-static struct fib_rules_ops fib4_rules_ops = {
+static const struct fib_rules_ops __net_initconst fib4_rules_ops_template = {
 	.family		= AF_INET,
 	.rule_size	= sizeof(struct fib4_rule),
+	.addr_size	= sizeof(u32),
 	.action		= fib4_rule_action,
+	.suppress	= fib4_rule_suppress,
 	.match		= fib4_rule_match,
 	.configure	= fib4_rule_configure,
+	.delete		= fib4_rule_delete,
 	.compare	= fib4_rule_compare,
 	.fill		= fib4_rule_fill,
-	.default_pref	= fib4_rule_default_pref,
+	.default_pref	= fib_default_rule_pref,
+	.nlmsg_payload	= fib4_rule_nlmsg_payload,
+	.flush_cache	= fib4_rule_flush_cache,
 	.nlgroup	= RTNLGRP_IPV4_RULE,
 	.policy		= fib4_rule_policy,
-	.rules_list	= &fib4_rules,
 	.owner		= THIS_MODULE,
 };
 
-void __init fib4_rules_init(void)
+static int fib_default_rules_init(struct fib_rules_ops *ops)
+{
+	int err;
+
+	err = fib_default_rule_add(ops, 0, RT_TABLE_LOCAL, 0);
+	if (err < 0)
+		return err;
+	err = fib_default_rule_add(ops, 0x7FFE, RT_TABLE_MAIN, 0);
+	if (err < 0)
+		return err;
+	err = fib_default_rule_add(ops, 0x7FFF, RT_TABLE_DEFAULT, 0);
+	if (err < 0)
+		return err;
+	return 0;
+}
+
+int __net_init fib4_rules_init(struct net *net)
 {
-	list_add_tail(&local_rule.common.list, &fib4_rules);
-	list_add_tail(&main_rule.common.list, &fib4_rules);
-	list_add_tail(&default_rule.common.list, &fib4_rules);
+	int err;
+	struct fib_rules_ops *ops;
 
-	fib_rules_register(&fib4_rules_ops);
+	ops = fib_rules_register(&fib4_rules_ops_template, net);
+	if (IS_ERR(ops))
+		return PTR_ERR(ops);
+
+	err = fib_default_rules_init(ops);
+	if (err < 0)
+		goto fail;
+	net->ipv4.rules_ops = ops;
+	net->ipv4.fib_has_custom_rules = false;
+	return 0;
+
+fail:
+	/* also cleans all rules already added */
+	fib_rules_unregister(ops);
+	return err;
+}
+
+void __net_exit fib4_rules_exit(struct net *net)
+{
+	fib_rules_unregister(net->ipv4.rules_ops);
 }