1 files changed, 264 insertions, 59 deletions

diff --git a/net/ipv4/fib_rules.c b/net/ipv4/fib_rules.c
index 26aa65d1fce4..51f0193092f0 100644
--- a/net/ipv4/fib_rules.c
+++ b/net/ipv4/fib_rules.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
 /*
  * INET		An implementation of the TCP/IP protocol suite for the LINUX
  *		operating system.  INET is implemented using the  BSD Socket
@@ -8,11 +9,6 @@
  * Authors:	Alexey Kuznetsov, <kuznet@ms2.inr.ac.ru>
  *		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
- *		as published by the Free Software Foundation; either version
- *		2 of the License, or (at your option) any later version.
- *
  * Fixes:
  *		Rani Assaf	:	local_rule cannot be deleted
  *		Marc Boucher	:	routing by fwmark
@@ -27,17 +23,23 @@
 #include <linux/list.h>
 #include <linux/rcupdate.h>
 #include <linux/export.h>
+#include <net/flow.h>
+#include <net/inet_dscp.h>
 #include <net/ip.h>
 #include <net/route.h>
 #include <net/tcp.h>
 #include <net/ip_fib.h>
+#include <net/nexthop.h>
 #include <net/fib_rules.h>
+#include <linux/indirect_call_wrapper.h>
 
 struct fib4_rule {
 	struct fib_rule		common;
 	u8			dst_len;
 	u8			src_len;
-	u8			tos;
+	dscp_t			dscp;
+	dscp_t			dscp_mask;
+	u8			dscp_full:1;	/* DSCP or TOS selector */
 	__be32			src;
 	__be32			srcmask;
 	__be32			dst;
@@ -47,14 +49,50 @@ struct fib4_rule {
 #endif
 };
 
-int __fib_lookup(struct net *net, struct flowi4 *flp, struct fib_result *res)
+static bool fib4_rule_matchall(const struct fib_rule *rule)
+{
+	struct fib4_rule *r = container_of(rule, struct fib4_rule, common);
+
+	if (r->dst_len || r->src_len || r->dscp)
+		return false;
+	return fib_rule_matchall(rule);
+}
+
+bool fib4_rule_default(const struct fib_rule *rule)
+{
+	if (!fib4_rule_matchall(rule) || rule->action != FR_ACT_TO_TBL ||
+	    rule->l3mdev)
+		return false;
+	if (rule->table != RT_TABLE_LOCAL && rule->table != RT_TABLE_MAIN &&
+	    rule->table != RT_TABLE_DEFAULT)
+		return false;
+	return true;
+}
+EXPORT_SYMBOL_GPL(fib4_rule_default);
+
+int fib4_rules_dump(struct net *net, struct notifier_block *nb,
+		    struct netlink_ext_ack *extack)
+{
+	return fib_rules_dump(net, nb, AF_INET, extack);
+}
+
+unsigned int fib4_rules_seq_read(const struct net *net)
+{
+	return fib_rules_seq_read(net, AF_INET);
+}
+
+int __fib_lookup(struct net *net, struct flowi4 *flp,
+		 struct fib_result *res, unsigned int flags)
 {
 	struct fib_lookup_arg arg = {
 		.result = res,
-		.flags = FIB_LOOKUP_NOREF,
+		.flags = flags,
 	};
 	int err;
 
+	/* update flow if oif or iif point to device enslaved to l3mdev */
+	l3mdev_update_flow(net, flowi4_to_flowi(flp));
+
 	err = fib_rules_lookup(net->ipv4.rules_ops, flowi4_to_flowi(flp), 0, &arg);
 #ifdef CONFIG_IP_ROUTE_CLASSID
 	if (arg.rule)
@@ -62,47 +100,85 @@ int __fib_lookup(struct net *net, struct flowi4 *flp, struct fib_result *res)
 	else
 		res->tclassid = 0;
 #endif
+
+	if (err == -ESRCH)
+		err = -ENETUNREACH;
+
 	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)
+INDIRECT_CALLABLE_SCOPE int fib4_rule_action(struct fib_rule *rule,
+					     struct flowi *flp, int flags,
+					     struct fib_lookup_arg *arg)
 {
 	int err = -EAGAIN;
 	struct fib_table *tbl;
+	u32 tb_id;
 
 	switch (rule->action) {
 	case FR_ACT_TO_TBL:
 		break;
 
 	case FR_ACT_UNREACHABLE:
-		err = -ENETUNREACH;
-		goto errout;
+		return -ENETUNREACH;
 
 	case FR_ACT_PROHIBIT:
-		err = -EACCES;
-		goto errout;
+		return -EACCES;
 
 	case FR_ACT_BLACKHOLE:
 	default:
-		err = -EINVAL;
-		goto errout;
+		return -EINVAL;
 	}
 
-	tbl = fib_get_table(rule->fr_net, rule->table);
-	if (!tbl)
-		goto errout;
+	rcu_read_lock();
 
-	err = fib_table_lookup(tbl, &flp->u.ip4, (struct fib_result *) arg->result, arg->flags);
-	if (err > 0)
-		err = -EAGAIN;
-errout:
+	tb_id = fib_rule_get_table(rule, arg);
+	tbl = fib_get_table(rule->fr_net, tb_id);
+	if (tbl)
+		err = fib_table_lookup(tbl, &flp->u.ip4,
+				       (struct fib_result *)arg->result,
+				       arg->flags);
+
+	rcu_read_unlock();
 	return err;
 }
 
+INDIRECT_CALLABLE_SCOPE bool fib4_rule_suppress(struct fib_rule *rule,
+						int flags,
+						struct fib_lookup_arg *arg)
+{
+	struct fib_result *result = arg->result;
+	struct net_device *dev = NULL;
+
+	if (result->fi) {
+		struct fib_nh_common *nhc = fib_info_nhc(result->fi, 0);
+
+		dev = nhc->nhc_dev;
+	}
+
+	/* do not accept result if the route does
+	 * not meet the required prefix length
+	 */
+	if (result->prefixlen <= rule->suppress_prefixlen)
+		goto suppress_route;
 
-static int fib4_rule_match(struct fib_rule *rule, struct flowi *fl, int flags)
+	/* 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;
+}
+
+INDIRECT_CALLABLE_SCOPE int fib4_rule_match(struct fib_rule *rule,
+					    struct flowi *fl, int flags)
 {
 	struct fib4_rule *r = (struct fib4_rule *) rule;
 	struct flowi4 *fl4 = &fl->u.ip4;
@@ -113,7 +189,26 @@ static int fib4_rule_match(struct fib_rule *rule, struct flowi *fl, int flags)
 	    ((daddr ^ r->dst) & r->dstmask))
 		return 0;
 
-	if (r->tos && (r->tos != fl4->flowi4_tos))
+	/* When DSCP selector is used we need to match on the entire DSCP field
+	 * in the flow information structure. When TOS selector is used we need
+	 * to mask the upper three DSCP bits prior to matching to maintain
+	 * legacy behavior.
+	 */
+	if (r->dscp_full && (r->dscp ^ fl4->flowi4_dscp) & r->dscp_mask)
+		return 0;
+	else if (!r->dscp_full && r->dscp &&
+		 !fib_dscp_masked_match(r->dscp, fl4))
+		return 0;
+
+	if (rule->ip_proto && (rule->ip_proto != fl4->flowi4_proto))
+		return 0;
+
+	if (!fib_rule_port_match(&rule->sport_range, rule->sport_mask,
+				 fl4->fl4_sport))
+		return 0;
+
+	if (!fib_rule_port_match(&rule->dport_range, rule->dport_mask,
+				 fl4->fl4_dport))
 		return 0;
 
 	return 1;
@@ -121,36 +216,102 @@ static int fib4_rule_match(struct fib_rule *rule, struct flowi *fl, int flags)
 
 static struct fib_table *fib_empty_table(struct net *net)
 {
-	u32 id;
+	u32 id = 1;
 
-	for (id = 1; id <= RT_TABLE_MAX; id++)
-		if (fib_get_table(net, id) == NULL)
+	while (1) {
+		if (!fib_get_table(net, id))
 			return fib_new_table(net, id);
+
+		if (id++ == RT_TABLE_MAX)
+			break;
+	}
 	return NULL;
 }
 
-static const struct nla_policy fib4_rule_policy[FRA_MAX+1] = {
-	FRA_GENERIC_POLICY,
-	[FRA_FLOW]	= { .type = NLA_U32 },
-};
+static int fib4_nl2rule_dscp(const struct nlattr *nla, struct fib4_rule *rule4,
+			     struct netlink_ext_ack *extack)
+{
+	if (rule4->dscp) {
+		NL_SET_ERR_MSG(extack, "Cannot specify both TOS and DSCP");
+		return -EINVAL;
+	}
+
+	rule4->dscp = inet_dsfield_to_dscp(nla_get_u8(nla) << 2);
+	rule4->dscp_mask = inet_dsfield_to_dscp(INET_DSCP_MASK);
+	rule4->dscp_full = true;
+
+	return 0;
+}
+
+static int fib4_nl2rule_dscp_mask(const struct nlattr *nla,
+				  struct fib4_rule *rule4,
+				  struct netlink_ext_ack *extack)
+{
+	dscp_t dscp_mask;
+
+	if (!rule4->dscp_full) {
+		NL_SET_ERR_MSG_ATTR(extack, nla,
+				    "Cannot specify DSCP mask without DSCP value");
+		return -EINVAL;
+	}
+
+	dscp_mask = inet_dsfield_to_dscp(nla_get_u8(nla) << 2);
+	if (rule4->dscp & ~dscp_mask) {
+		NL_SET_ERR_MSG_ATTR(extack, nla, "Invalid DSCP mask");
+		return -EINVAL;
+	}
+
+	rule4->dscp_mask = dscp_mask;
+
+	return 0;
+}
 
 static int fib4_rule_configure(struct fib_rule *rule, struct sk_buff *skb,
 			       struct fib_rule_hdr *frh,
-			       struct nlattr **tb)
+			       struct nlattr **tb,
+			       struct netlink_ext_ack *extack)
 {
-	struct net *net = sock_net(skb->sk);
+	struct fib4_rule *rule4 = (struct fib4_rule *)rule;
+	struct net *net = rule->fr_net;
 	int err = -EINVAL;
-	struct fib4_rule *rule4 = (struct fib4_rule *) rule;
 
-	if (frh->tos & ~IPTOS_TOS_MASK)
+	if (tb[FRA_FLOWLABEL] || tb[FRA_FLOWLABEL_MASK]) {
+		NL_SET_ERR_MSG(extack,
+			       "Flow label cannot be specified for IPv4 FIB rules");
+		goto errout;
+	}
+
+	if (!inet_validate_dscp(frh->tos)) {
+		NL_SET_ERR_MSG(extack,
+			       "Invalid dsfield (tos): ECN bits must be 0");
+		goto errout;
+	}
+	/* IPv4 currently doesn't handle high order DSCP bits correctly */
+	if (frh->tos & ~IPTOS_TOS_MASK) {
+		NL_SET_ERR_MSG(extack, "Invalid tos");
+		goto errout;
+	}
+	rule4->dscp = inet_dsfield_to_dscp(frh->tos);
+
+	if (tb[FRA_DSCP] &&
+	    fib4_nl2rule_dscp(tb[FRA_DSCP], rule4, extack) < 0)
 		goto errout;
 
-	if (rule->table == RT_TABLE_UNSPEC) {
+	if (tb[FRA_DSCP_MASK] &&
+	    fib4_nl2rule_dscp_mask(tb[FRA_DSCP_MASK], rule4, extack) < 0)
+		goto errout;
+
+	/* split local/main if they are not already split */
+	err = fib_unmerge(net);
+	if (err)
+		goto errout;
+
+	if (rule->table == RT_TABLE_UNSPEC && !rule->l3mdev) {
 		if (rule->action == FR_ACT_TO_TBL) {
 			struct fib_table *table;
 
 			table = fib_empty_table(net);
-			if (table == NULL) {
+			if (!table) {
 				err = -ENOBUFS;
 				goto errout;
 			}
@@ -160,41 +321,55 @@ static int fib4_rule_configure(struct fib_rule *rule, struct sk_buff *skb,
 	}
 
 	if (frh->src_len)
-		rule4->src = nla_get_be32(tb[FRA_SRC]);
+		rule4->src = nla_get_in_addr(tb[FRA_SRC]);
 
 	if (frh->dst_len)
-		rule4->dst = nla_get_be32(tb[FRA_DST]);
+		rule4->dst = nla_get_in_addr(tb[FRA_DST]);
 
 #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++;
+			atomic_inc(&net->ipv4.fib_num_tclassid_users);
 	}
 #endif
 
+	if (fib_rule_requires_fldissect(rule))
+		net->ipv4.fib_rules_require_fldissect++;
+
 	rule4->src_len = frh->src_len;
 	rule4->srcmask = inet_make_mask(rule4->src_len);
 	rule4->dst_len = frh->dst_len;
 	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)
+static int 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;
+	int err;
 
-	if (rule4->tclassid)
-		net->ipv4.fib_num_tclassid_users--;
+	/* split local/main if they are not already split */
+	err = fib_unmerge(net);
+	if (err)
+		goto errout;
+
+#ifdef CONFIG_IP_ROUTE_CLASSID
+	if (((struct fib4_rule *)rule)->tclassid)
+		atomic_dec(&net->ipv4.fib_num_tclassid_users);
 #endif
 	net->ipv4.fib_has_custom_rules = true;
+
+	if (net->ipv4.fib_rules_require_fldissect &&
+	    fib_rule_requires_fldissect(rule))
+		net->ipv4.fib_rules_require_fldissect--;
+errout:
+	return err;
 }
 
 static int fib4_rule_compare(struct fib_rule *rule, struct fib_rule_hdr *frh,
@@ -208,18 +383,36 @@ static int fib4_rule_compare(struct fib_rule *rule, struct fib_rule_hdr *frh,
 	if (frh->dst_len && (rule4->dst_len != frh->dst_len))
 		return 0;
 
-	if (frh->tos && (rule4->tos != frh->tos))
+	if (frh->tos &&
+	    (rule4->dscp_full ||
+	     inet_dscp_to_dsfield(rule4->dscp) != frh->tos))
 		return 0;
 
+	if (tb[FRA_DSCP]) {
+		dscp_t dscp;
+
+		dscp = inet_dsfield_to_dscp(nla_get_u8(tb[FRA_DSCP]) << 2);
+		if (!rule4->dscp_full || rule4->dscp != dscp)
+			return 0;
+	}
+
+	if (tb[FRA_DSCP_MASK]) {
+		dscp_t dscp_mask;
+
+		dscp_mask = inet_dsfield_to_dscp(nla_get_u8(tb[FRA_DSCP_MASK]) << 2);
+		if (!rule4->dscp_full || rule4->dscp_mask != dscp_mask)
+			return 0;
+	}
+
 #ifdef CONFIG_IP_ROUTE_CLASSID
 	if (tb[FRA_FLOW] && (rule4->tclassid != nla_get_u32(tb[FRA_FLOW])))
 		return 0;
 #endif
 
-	if (frh->src_len && (rule4->src != nla_get_be32(tb[FRA_SRC])))
+	if (frh->src_len && (rule4->src != nla_get_in_addr(tb[FRA_SRC])))
 		return 0;
 
-	if (frh->dst_len && (rule4->dst != nla_get_be32(tb[FRA_DST])))
+	if (frh->dst_len && (rule4->dst != nla_get_in_addr(tb[FRA_DST])))
 		return 0;
 
 	return 1;
@@ -232,12 +425,22 @@ static int fib4_rule_fill(struct fib_rule *rule, struct sk_buff *skb,
 
 	frh->dst_len = rule4->dst_len;
 	frh->src_len = rule4->src_len;
-	frh->tos = rule4->tos;
+
+	if (rule4->dscp_full) {
+		frh->tos = 0;
+		if (nla_put_u8(skb, FRA_DSCP,
+			       inet_dscp_to_dsfield(rule4->dscp) >> 2) ||
+		    nla_put_u8(skb, FRA_DSCP_MASK,
+			       inet_dscp_to_dsfield(rule4->dscp_mask) >> 2))
+			goto nla_put_failure;
+	} else {
+		frh->tos = inet_dscp_to_dsfield(rule4->dscp);
+	}
 
 	if ((rule4->dst_len &&
-	     nla_put_be32(skb, FRA_DST, rule4->dst)) ||
+	     nla_put_in_addr(skb, FRA_DST, rule4->dst)) ||
 	    (rule4->src_len &&
-	     nla_put_be32(skb, FRA_SRC, rule4->src)))
+	     nla_put_in_addr(skb, FRA_SRC, rule4->src)))
 		goto nla_put_failure;
 #ifdef CONFIG_IP_ROUTE_CLASSID
 	if (rule4->tclassid &&
@@ -254,7 +457,9 @@ static size_t fib4_rule_nlmsg_payload(struct fib_rule *rule)
 {
 	return nla_total_size(4) /* dst */
 	       + nla_total_size(4) /* src */
-	       + nla_total_size(4); /* flow */
+	       + nla_total_size(4) /* flow */
+	       + nla_total_size(1) /* dscp */
+	       + nla_total_size(1); /* dscp mask */
 }
 
 static void fib4_rule_flush_cache(struct fib_rules_ops *ops)
@@ -267,16 +472,15 @@ static const struct fib_rules_ops __net_initconst fib4_rules_ops_template = {
 	.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	= fib_default_rule_pref,
 	.nlmsg_payload	= fib4_rule_nlmsg_payload,
 	.flush_cache	= fib4_rule_flush_cache,
 	.nlgroup	= RTNLGRP_IPV4_RULE,
-	.policy		= fib4_rule_policy,
 	.owner		= THIS_MODULE,
 };
 
@@ -284,13 +488,13 @@ static int fib_default_rules_init(struct fib_rules_ops *ops)
 {
 	int err;
 
-	err = fib_default_rule_add(ops, 0, RT_TABLE_LOCAL, 0);
+	err = fib_default_rule_add(ops, 0, RT_TABLE_LOCAL);
 	if (err < 0)
 		return err;
-	err = fib_default_rule_add(ops, 0x7FFE, RT_TABLE_MAIN, 0);
+	err = fib_default_rule_add(ops, 0x7FFE, RT_TABLE_MAIN);
 	if (err < 0)
 		return err;
-	err = fib_default_rule_add(ops, 0x7FFF, RT_TABLE_DEFAULT, 0);
+	err = fib_default_rule_add(ops, 0x7FFF, RT_TABLE_DEFAULT);
 	if (err < 0)
 		return err;
 	return 0;
@@ -310,6 +514,7 @@ int __net_init fib4_rules_init(struct net *net)
 		goto fail;
 	net->ipv4.rules_ops = ops;
 	net->ipv4.fib_has_custom_rules = false;
+	net->ipv4.fib_rules_require_fldissect = 0;
 	return 0;
 
 fail: