IPv4 FIB configuration rework

PATCH New Thomas Graf 2 participants 6 messages Aug 8, 2006

Subsystems (3)

Networking Network drivers IPv4/IPv6

lore

Share & tools

lore

Thomas Graf

author

2006-08-08 19:46 UTC #0 lore raw

PATCH 1/1 IPv4 FIB configuration rework

Replaces the rather ugly struct kern_rta + rtmsg with a new struct fib_config while converting everything to the new netlink api. Simplifies the FIB module interface quite a bit and allows changing internals while keeping a stable netlink interface.

Gets rid of things like passing on netlink_skb_parms all around the fib code etc.

Also fixes some memory corruptions that could be triggered via netlink messages, not sure which would actually lead to an oops though.

Thomas Graf

author

2006-08-08 19:46 UTC #1 lore raw

PATCH 1/3 : FIB configuration using struct fib_config

Introduces struct fib_config replacing the ugly struct kern_rta prone to ordering issues. Avoids creating faked netlink messages for auto generated routes or requests via ioctl.

A new interface net/nexthop.h is added to help navigate through nexthop configuration arrays.

Since the netlink source pid of the requesting process is stored in fib_config, it is no longer required to pass netlink_skb_parms all along just for event notification.

Signed-off-by Thomas Graf <[email protected]>

include/net/ip_fib.h +42 ~~-27~~

Index: net-2.6.19.git/include/net/ip_fib.h
===================================================================
@@ -20,25 +20,46 @@
 #include <linux/seq_file.h>
 #include <net/fib_rules.h>
 
-/* WARNING: The ordering of these elements must match ordering
- *          of RTA_* rtnetlink attribute numbers.
- */
-struct kern_rta {
-	void		*rta_dst;
-	void		*rta_src;
-	int		*rta_iif;
-	int		*rta_oif;
-	void		*rta_gw;
-	u32		*rta_priority;
-	void		*rta_prefsrc;
-	struct rtattr	*rta_mx;
-	struct rtattr	*rta_mp;
-	unsigned char	*rta_protoinfo;
-	u32		*rta_flow;
-	struct rta_cacheinfo *rta_ci;
-	struct rta_session *rta_sess;
-	u32		*rta_mp_alg;
-};
+struct fib_config {
+	u8			fc_family;
+	u8			fc_dst_len;
+	u8			fc_src_len;
+	u8			fc_tos;
+	u8			fc_protocol;
+	u8			fc_scope;
+	u8			fc_type;
+	/* 1 byte unused */
+	u32			fc_table;
+	u32			fc_flags;
+	u32			fc_dst;
+	u32			fc_src;
+	u32			fc_gw;
+	int			fc_oif;
+	u32			fc_priority;
+	u32			fc_prefsrc;
+	struct nlattr		*fc_mx;
+	struct rtnexthop	*fc_mp;
+	int			fc_mx_len;
+	int			fc_mp_len;
+	u32			fc_flow;
+	u32			fc_mp_alg;
+
+	/*
+	 * Request was issued via netlink, points to netlink header
+	 * to answer request
+	 */
+	struct nlmsghdr		*fc_nlh;
+
+	/*
+	 * Use separate field for netlink flags so non-netlink
+	 * sources are able to emulate it without providing a
+	 * complete header.
+	 */
+	u32			fc_nlmsg_flags;
+
+	/* Netlink PID of the process requesting the change */
+	u32			fc_pid;
+ };
 
 struct fib_info;

@@ -153,12 +174,8 @@ struct fib_table {
 	unsigned char	tb_id;
 	unsigned	tb_stamp;
 	int		(*tb_lookup)(struct fib_table *tb, const struct flowi *flp, struct fib_result *res);
-	int		(*tb_insert)(struct fib_table *table, struct rtmsg *r,
-				     struct kern_rta *rta, struct nlmsghdr *n,
-				     struct netlink_skb_parms *req);
-	int		(*tb_delete)(struct fib_table *table, struct rtmsg *r,
-				     struct kern_rta *rta, struct nlmsghdr *n,
-				     struct netlink_skb_parms *req);
+	int		(*tb_insert)(struct fib_table *, struct fib_config *);
+	int		(*tb_delete)(struct fib_table *, struct fib_config *);
 	int		(*tb_dump)(struct fib_table *table, struct sk_buff *skb,
 				     struct netlink_callback *cb);
 	int		(*tb_flush)(struct fib_table *table);

@@ -243,8 +260,6 @@ struct rtentry;
 extern int ip_fib_check_default(u32 gw, struct net_device *dev);
 extern int fib_sync_down(u32 local, struct net_device *dev, int force);
 extern int fib_sync_up(struct net_device *dev);
-extern int fib_convert_rtentry(int cmd, struct nlmsghdr *nl, struct rtmsg *rtm,
-			       struct kern_rta *rta, struct rtentry *r);
 extern u32  __fib_res_prefsrc(struct fib_result *res);
 
 /* Exported by fib_hash.c */

net/ipv4/fib_frontend.c +289 ~~-70~~

424 lines not on this page · raw message

net/ipv4/fib_semantics.c +130 ~~-276~~

583 lines not on this page · raw message

include/net/nexthop.h new file +33 -0

Index: net-2.6.19.git/include/net/nexthop.h
===================================================================
@@ -0,0 +1,33 @@
+#ifndef __NET_NEXTHOP_H
+#define __NET_NEXTHOP_H
+
+#include <linux/rtnetlink.h>
+#include <net/netlink.h>
+
+static inline int rtnh_ok(const struct rtnexthop *rtnh, int remaining)
+{
+	return remaining >= sizeof(*rtnh) &&
+	       rtnh->rtnh_len >= sizeof(*rtnh) &&
+	       rtnh->rtnh_len <= remaining;
+}
+
+static inline struct rtnexthop *rtnh_next(const struct rtnexthop *rtnh,
+                                         int *remaining)
+{
+	int totlen = NLA_ALIGN(rtnh->rtnh_len);
+
+	*remaining -= totlen;
+	return (struct rtnexthop *) ((char *) rtnh + totlen);
+}
+
+static inline struct nlattr *rtnh_attrs(const struct rtnexthop *rtnh)
+{
+	return (struct nlattr *) ((char *) rtnh + NLA_ALIGN(sizeof(*rtnh)));
+}
+
+static inline int rtnh_attrlen(const struct rtnexthop *rtnh)
+{
+	return rtnh->rtnh_len - NLA_ALIGN(sizeof(*rtnh));
+}
+
+#endif

net/ipv4/fib_hash.c +43 ~~-50~~

195 lines not on this page · raw message

net/ipv4/fib_lookup.h +3 -8

Index: net-2.6.19.git/net/ipv4/fib_lookup.h
===================================================================
@@ -23,19 +23,14 @@ extern int fib_semantic_match(struct lis
 			      struct fib_result *res, __u32 zone, __u32 mask,
 				int prefixlen);
 extern void fib_release_info(struct fib_info *);
-extern struct fib_info *fib_create_info(const struct rtmsg *r,
-					struct kern_rta *rta,
-					const struct nlmsghdr *,
-					int *err);
-extern int fib_nh_match(struct rtmsg *r, struct nlmsghdr *,
-			struct kern_rta *rta, struct fib_info *fi);
+extern struct fib_info *fib_create_info(struct fib_config *cfg);
+extern int fib_nh_match(struct fib_config *cfg, struct fib_info *fi);
 extern int fib_dump_info(struct sk_buff *skb, u32 pid, u32 seq, int event,
 			 u8 tb_id, u8 type, u8 scope, void *dst,
 			 int dst_len, u8 tos, struct fib_info *fi,
 			 unsigned int);
 extern void rtmsg_fib(int event, u32 key, struct fib_alias *fa,
-		      int z, int tb_id,
-		      struct nlmsghdr *n, struct netlink_skb_parms *req);
+		      int dst_len, int tb_id, struct fib_config *cfg);
 extern struct fib_alias *fib_find_alias(struct list_head *fah,
 					u8 tos, u32 prio);
 extern int fib_detect_death(struct fib_info *fi, int order,

net/ipv4/fib_trie.c +28 ~~-42~~

155 lines not on this page · raw message

Thomas Graf

author

2006-08-08 23:10 UTC #2 lore raw

Re: [PATCH 1/3] [IPv4]: FIB configuration using struct fib_config

* Thomas Graf <[email protected]> 2006-08-08 00:00

Quoted 10 lines

Introduces struct fib_config replacing the ugly struct kern_rta prone to ordering issues. Avoids creating faked netlink messages for auto generated routes or requests via ioctl.
A new interface net/nexthop.h is added to help navigate through nexthop configuration arrays.
Since the netlink source pid of the requesting process is stored in fib_config, it is no longer required to pass netlink_skb_parms all along just for event notification.

Hold on with this patch, netlink pid handling for netlink events is completely inconsistent. I want to figure out the right way to do it first.

David Miller

maintainer

2006-08-08 23:12 UTC #3 lore raw

Re: [PATCH 1/3] [IPv4]: FIB configuration using struct fib_config

From: Thomas Graf <[email protected]>
Date: Wed, 9 Aug 2006 01:11:03 +0200

Quoted 3 lines

Hold on with this patch, netlink pid handling for netlink events is completely inconsistent. I want to figure out the right way to do it first.

Ok.

Thomas Graf

author

2006-08-08 19:46 UTC #4 lore raw

PATCH 2/3 : Convert FIB dumping to use new netlink api

In reply to Thomas Graf

Signed-off-by Thomas Graf <[email protected]>

net/ipv4/fib_semantics.c +47 ~~-39~~

Index: net-2.6.19.git/net/ipv4/fib_semantics.c
===================================================================
@@ -286,7 +286,7 @@ void rtmsg_fib(int event, u32 key, struc
 		return;
 
 	if (fib_dump_info(skb, pid, seq, event, tb_id, fa->fa_type,
-			  fa->fa_scope, &key, dst_len, fa->fa_tos,
+			  fa->fa_scope, key, dst_len, fa->fa_tos,
 			  fa->fa_info, 0) <= 0) {
 		kfree_skb(skb);
 		return;

@@ -935,17 +935,18 @@ u32 __fib_res_prefsrc(struct fib_result 
 	return inet_select_addr(FIB_RES_DEV(*res), FIB_RES_GW(*res), res->scope);
 }
 
-int
-fib_dump_info(struct sk_buff *skb, u32 pid, u32 seq, int event,
-	      u8 tb_id, u8 type, u8 scope, void *dst, int dst_len, u8 tos,
-	      struct fib_info *fi, unsigned int flags)
+int fib_dump_info(struct sk_buff *skb, u32 pid, u32 seq, int event,
+		  u8 tb_id, u8 type, u8 scope, u32 dst, int dst_len, u8 tos,
+		  struct fib_info *fi, unsigned int flags)
 {
+	struct nlmsghdr *nlh;
 	struct rtmsg *rtm;
-	struct nlmsghdr  *nlh;
-	unsigned char	 *b = skb->tail;
 
-	nlh = NLMSG_NEW(skb, pid, seq, event, sizeof(*rtm), flags);
-	rtm = NLMSG_DATA(nlh);
+	nlh = nlmsg_put(skb, pid, seq, event, sizeof(*rtm), flags);
+	if (nlh == NULL)
+		return -ENOBUFS;
+
+	rtm = nlmsg_data(nlh);
 	rtm->rtm_family = AF_INET;
 	rtm->rtm_dst_len = dst_len;
 	rtm->rtm_src_len = 0;

@@ -954,59 +955,66 @@ fib_dump_info(struct sk_buff *skb, u32 p
 	rtm->rtm_type = type;
 	rtm->rtm_flags = fi->fib_flags;
 	rtm->rtm_scope = scope;
-	if (rtm->rtm_dst_len)
-		RTA_PUT(skb, RTA_DST, 4, dst);
 	rtm->rtm_protocol = fi->fib_protocol;
+
+	if (rtm->rtm_dst_len)
+		NLA_PUT_U32(skb, RTA_DST, dst);
+
 	if (fi->fib_priority)
-		RTA_PUT(skb, RTA_PRIORITY, 4, &fi->fib_priority);
+		NLA_PUT_U32(skb, RTA_PRIORITY, fi->fib_priority);
+
 	if (rtnetlink_put_metrics(skb, fi->fib_metrics) < 0)
-		goto rtattr_failure;
+		goto nla_put_failure;
+
 	if (fi->fib_prefsrc)
-		RTA_PUT(skb, RTA_PREFSRC, 4, &fi->fib_prefsrc);
+		NLA_PUT_U32(skb, RTA_PREFSRC, fi->fib_prefsrc);
+
 	if (fi->fib_nhs == 1) {
 		if (fi->fib_nh->nh_gw)
-			RTA_PUT(skb, RTA_GATEWAY, 4, &fi->fib_nh->nh_gw);
+			NLA_PUT_U32(skb, RTA_GATEWAY, fi->fib_nh->nh_gw);
+
 		if (fi->fib_nh->nh_oif)
-			RTA_PUT(skb, RTA_OIF, sizeof(int), &fi->fib_nh->nh_oif);
+			NLA_PUT_U32(skb, RTA_OIF, fi->fib_nh->nh_oif);
 #ifdef CONFIG_NET_CLS_ROUTE
 		if (fi->fib_nh[0].nh_tclassid)
-			RTA_PUT(skb, RTA_FLOW, 4, &fi->fib_nh[0].nh_tclassid);
+			NLA_PUT_U32(skb, RTA_FLOW, fi->fib_nh[0].nh_tclassid);
 #endif
 	}
 #ifdef CONFIG_IP_ROUTE_MULTIPATH
 	if (fi->fib_nhs > 1) {
-		struct rtnexthop *nhp;
-		struct rtattr *mp_head;
-		if (skb_tailroom(skb) <= RTA_SPACE(0))
-			goto rtattr_failure;
-		mp_head = (struct rtattr*)skb_put(skb, RTA_SPACE(0));
+		struct rtnexthop *rtnh;
+		struct nlattr *mp;
+
+		mp = nla_nest_start(skb, RTA_MULTIPATH);
+		if (mp == NULL)
+			goto nla_put_failure;
 
 		for_nexthops(fi) {
-			if (skb_tailroom(skb) < RTA_ALIGN(RTA_ALIGN(sizeof(*nhp)) + 4))
-				goto rtattr_failure;
-			nhp = (struct rtnexthop*)skb_put(skb, RTA_ALIGN(sizeof(*nhp)));
-			nhp->rtnh_flags = nh->nh_flags & 0xFF;
-			nhp->rtnh_hops = nh->nh_weight-1;
-			nhp->rtnh_ifindex = nh->nh_oif;
+			rtnh = nla_reserve_nohdr(skb, sizeof(*rtnh));
+			if (rtnh == NULL)
+				goto nla_put_failure;
+
+			rtnh->rtnh_flags = nh->nh_flags & 0xFF;
+			rtnh->rtnh_hops = nh->nh_weight - 1;
+			rtnh->rtnh_ifindex = nh->nh_oif;
+
 			if (nh->nh_gw)
-				RTA_PUT(skb, RTA_GATEWAY, 4, &nh->nh_gw);
+				NLA_PUT_U32(skb, RTA_GATEWAY, nh->nh_gw);
 #ifdef CONFIG_NET_CLS_ROUTE
 			if (nh->nh_tclassid)
-				RTA_PUT(skb, RTA_FLOW, 4, &nh->nh_tclassid);
+				NLA_PUT_U32(skb, RTA_FLOW, nh->nh_tclassid);
 #endif
-			nhp->rtnh_len = skb->tail - (unsigned char*)nhp;
+			/* length of rtnetlink header + attributes */
+			rtnh->rtnh_len = nlmsg_get_pos(skb) - (void *) rtnh;
 		} endfor_nexthops(fi);
-		mp_head->rta_type = RTA_MULTIPATH;
-		mp_head->rta_len = skb->tail - (u8*)mp_head;
+
+		nla_nest_end(skb, mp);
 	}
 #endif
-	nlh->nlmsg_len = skb->tail - b;
-	return skb->len;
+	return nlmsg_end(skb, nlh);
 
-nlmsg_failure:
-rtattr_failure:
-	skb_trim(skb, b - skb->data);
-	return -1;
+nla_put_failure:
+	return nlmsg_cancel(skb, nlh);
 }
 
 /*

net/ipv4/fib_hash.c +1 -1

Index: net-2.6.19.git/net/ipv4/fib_hash.c
===================================================================
@@ -692,7 +692,7 @@ fn_hash_dump_bucket(struct sk_buff *skb,
 					  tb->tb_id,
 					  fa->fa_type,
 					  fa->fa_scope,
-					  &f->fn_key,
+					  f->fn_key,
 					  fz->fz_order,
 					  fa->fa_tos,
 					  fa->fa_info,

net/ipv4/fib_trie.c +1 -1

Index: net-2.6.19.git/net/ipv4/fib_trie.c
===================================================================
@@ -1852,7 +1852,7 @@ static int fn_trie_dump_fa(t_key key, in
 				  tb->tb_id,
 				  fa->fa_type,
 				  fa->fa_scope,
-				  &xkey,
+				  xkey,
 				  plen,
 				  fa->fa_tos,
 				  fa->fa_info, 0) < 0) {

net/ipv4/fib_lookup.h +1 -1

Index: net-2.6.19.git/net/ipv4/fib_lookup.h
===================================================================
@@ -26,7 +26,7 @@ extern void fib_release_info(struct fib_
 extern struct fib_info *fib_create_info(struct fib_config *cfg);
 extern int fib_nh_match(struct fib_config *cfg, struct fib_info *fi);
 extern int fib_dump_info(struct sk_buff *skb, u32 pid, u32 seq, int event,
-			 u8 tb_id, u8 type, u8 scope, void *dst,
+			 u8 tb_id, u8 type, u8 scope, u32 dst,
 			 int dst_len, u8 tos, struct fib_info *fi,
 			 unsigned int);
 extern void rtmsg_fib(int event, u32 key, struct fib_alias *fa,

net/ipv4/route.c +32 ~~-34~~

113 lines not on this page · raw message

net/ipv4/fib_frontend.c +2 -2

Index: net-2.6.19.git/net/ipv4/fib_frontend.c
===================================================================
@@ -563,8 +563,8 @@ int inet_dump_fib(struct sk_buff *skb, s
 	int s_t;
 	struct fib_table *tb;
 
-	if (NLMSG_PAYLOAD(cb->nlh, 0) >= sizeof(struct rtmsg) &&
-	    ((struct rtmsg*)NLMSG_DATA(cb->nlh))->rtm_flags&RTM_F_CLONED)
+	if (nlmsg_len(cb->nlh) >= sizeof(struct rtmsg) &&
+	    ((struct rtmsg *) nlmsg_data(cb->nlh))->rtm_flags & RTM_F_CLONED)
 		return ip_rt_dump(skb, cb);
 
 	s_t = cb->args[0];

Thomas Graf

author

2006-08-08 19:47 UTC #5 lore raw

PATCH 3/3 : Convert route get to new netlink api

In reply to Thomas Graf

Fixes various unvalidated netlink attributes causing memory corruptions when left empty by userspace applications.

Signed-off-by Thomas Graf <[email protected]>

include/net/ip_fib.h +1 -0

Index: net-2.6.19.git/include/net/ip_fib.h
===================================================================
@@ -245,6 +245,7 @@ extern void fib_select_default(const str
 #endif /* CONFIG_IP_MULTIPLE_TABLES */
 
 /* Exported by fib_frontend.c */
+extern struct nla_policy rtm_ipv4_policy[];
 extern void		ip_fib_init(void);
 extern int inet_rtm_delroute(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg);
 extern int inet_rtm_newroute(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg);

net/ipv4/fib_frontend.c +1 -1

Index: net-2.6.19.git/net/ipv4/fib_frontend.c
===================================================================
@@ -433,7 +433,7 @@ int ip_rt_ioctl(unsigned int cmd, void *
 
 #endif
 
-static struct nla_policy rtm_ipv4_policy[RTA_MAX+1] __read_mostly = {
+struct nla_policy rtm_ipv4_policy[RTA_MAX+1] __read_mostly = {
 	[RTA_DST]		= { .type = NLA_U32 },
 	[RTA_SRC]		= { .type = NLA_U32 },
 	[RTA_IIF]		= { .type = NLA_U32 },

net/ipv4/route.c +47 ~~-39~~

127 lines not on this page · raw message