Loading...
1/* IPv4 specific functions of netfilter core */
2#include <linux/kernel.h>
3#include <linux/netfilter.h>
4#include <linux/netfilter_ipv4.h>
5#include <linux/ip.h>
6#include <linux/skbuff.h>
7#include <linux/gfp.h>
8#include <net/route.h>
9#include <net/xfrm.h>
10#include <net/ip.h>
11#include <net/netfilter/nf_queue.h>
12
13/* route_me_harder function, used by iptable_nat, iptable_mangle + ip_queue */
14int ip_route_me_harder(struct sk_buff *skb, unsigned addr_type)
15{
16 struct net *net = dev_net(skb_dst(skb)->dev);
17 const struct iphdr *iph = ip_hdr(skb);
18 struct rtable *rt;
19 struct flowi4 fl4 = {};
20 __be32 saddr = iph->saddr;
21 __u8 flags = skb->sk ? inet_sk_flowi_flags(skb->sk) : 0;
22 unsigned int hh_len;
23
24 if (addr_type == RTN_UNSPEC)
25 addr_type = inet_addr_type(net, saddr);
26 if (addr_type == RTN_LOCAL || addr_type == RTN_UNICAST)
27 flags |= FLOWI_FLAG_ANYSRC;
28 else
29 saddr = 0;
30
31 /* some non-standard hacks like ipt_REJECT.c:send_reset() can cause
32 * packets with foreign saddr to appear on the NF_INET_LOCAL_OUT hook.
33 */
34 fl4.daddr = iph->daddr;
35 fl4.saddr = saddr;
36 fl4.flowi4_tos = RT_TOS(iph->tos);
37 fl4.flowi4_oif = skb->sk ? skb->sk->sk_bound_dev_if : 0;
38 fl4.flowi4_mark = skb->mark;
39 fl4.flowi4_flags = flags;
40 rt = ip_route_output_key(net, &fl4);
41 if (IS_ERR(rt))
42 return -1;
43
44 /* Drop old route. */
45 skb_dst_drop(skb);
46 skb_dst_set(skb, &rt->dst);
47
48 if (skb_dst(skb)->error)
49 return -1;
50
51#ifdef CONFIG_XFRM
52 if (!(IPCB(skb)->flags & IPSKB_XFRM_TRANSFORMED) &&
53 xfrm_decode_session(skb, flowi4_to_flowi(&fl4), AF_INET) == 0) {
54 struct dst_entry *dst = skb_dst(skb);
55 skb_dst_set(skb, NULL);
56 dst = xfrm_lookup(net, dst, flowi4_to_flowi(&fl4), skb->sk, 0);
57 if (IS_ERR(dst))
58 return -1;
59 skb_dst_set(skb, dst);
60 }
61#endif
62
63 /* Change in oif may mean change in hh_len. */
64 hh_len = skb_dst(skb)->dev->hard_header_len;
65 if (skb_headroom(skb) < hh_len &&
66 pskb_expand_head(skb, hh_len - skb_headroom(skb), 0, GFP_ATOMIC))
67 return -1;
68
69 return 0;
70}
71EXPORT_SYMBOL(ip_route_me_harder);
72
73#ifdef CONFIG_XFRM
74int ip_xfrm_me_harder(struct sk_buff *skb)
75{
76 struct flowi fl;
77 unsigned int hh_len;
78 struct dst_entry *dst;
79
80 if (IPCB(skb)->flags & IPSKB_XFRM_TRANSFORMED)
81 return 0;
82 if (xfrm_decode_session(skb, &fl, AF_INET) < 0)
83 return -1;
84
85 dst = skb_dst(skb);
86 if (dst->xfrm)
87 dst = ((struct xfrm_dst *)dst)->route;
88 dst_hold(dst);
89
90 dst = xfrm_lookup(dev_net(dst->dev), dst, &fl, skb->sk, 0);
91 if (IS_ERR(dst))
92 return -1;
93
94 skb_dst_drop(skb);
95 skb_dst_set(skb, dst);
96
97 /* Change in oif may mean change in hh_len. */
98 hh_len = skb_dst(skb)->dev->hard_header_len;
99 if (skb_headroom(skb) < hh_len &&
100 pskb_expand_head(skb, hh_len - skb_headroom(skb), 0, GFP_ATOMIC))
101 return -1;
102 return 0;
103}
104EXPORT_SYMBOL(ip_xfrm_me_harder);
105#endif
106
107void (*ip_nat_decode_session)(struct sk_buff *, struct flowi *);
108EXPORT_SYMBOL(ip_nat_decode_session);
109
110/*
111 * Extra routing may needed on local out, as the QUEUE target never
112 * returns control to the table.
113 */
114
115struct ip_rt_info {
116 __be32 daddr;
117 __be32 saddr;
118 u_int8_t tos;
119 u_int32_t mark;
120};
121
122static void nf_ip_saveroute(const struct sk_buff *skb,
123 struct nf_queue_entry *entry)
124{
125 struct ip_rt_info *rt_info = nf_queue_entry_reroute(entry);
126
127 if (entry->hook == NF_INET_LOCAL_OUT) {
128 const struct iphdr *iph = ip_hdr(skb);
129
130 rt_info->tos = iph->tos;
131 rt_info->daddr = iph->daddr;
132 rt_info->saddr = iph->saddr;
133 rt_info->mark = skb->mark;
134 }
135}
136
137static int nf_ip_reroute(struct sk_buff *skb,
138 const struct nf_queue_entry *entry)
139{
140 const struct ip_rt_info *rt_info = nf_queue_entry_reroute(entry);
141
142 if (entry->hook == NF_INET_LOCAL_OUT) {
143 const struct iphdr *iph = ip_hdr(skb);
144
145 if (!(iph->tos == rt_info->tos &&
146 skb->mark == rt_info->mark &&
147 iph->daddr == rt_info->daddr &&
148 iph->saddr == rt_info->saddr))
149 return ip_route_me_harder(skb, RTN_UNSPEC);
150 }
151 return 0;
152}
153
154__sum16 nf_ip_checksum(struct sk_buff *skb, unsigned int hook,
155 unsigned int dataoff, u_int8_t protocol)
156{
157 const struct iphdr *iph = ip_hdr(skb);
158 __sum16 csum = 0;
159
160 switch (skb->ip_summed) {
161 case CHECKSUM_COMPLETE:
162 if (hook != NF_INET_PRE_ROUTING && hook != NF_INET_LOCAL_IN)
163 break;
164 if ((protocol == 0 && !csum_fold(skb->csum)) ||
165 !csum_tcpudp_magic(iph->saddr, iph->daddr,
166 skb->len - dataoff, protocol,
167 skb->csum)) {
168 skb->ip_summed = CHECKSUM_UNNECESSARY;
169 break;
170 }
171 /* fall through */
172 case CHECKSUM_NONE:
173 if (protocol == 0)
174 skb->csum = 0;
175 else
176 skb->csum = csum_tcpudp_nofold(iph->saddr, iph->daddr,
177 skb->len - dataoff,
178 protocol, 0);
179 csum = __skb_checksum_complete(skb);
180 }
181 return csum;
182}
183EXPORT_SYMBOL(nf_ip_checksum);
184
185static __sum16 nf_ip_checksum_partial(struct sk_buff *skb, unsigned int hook,
186 unsigned int dataoff, unsigned int len,
187 u_int8_t protocol)
188{
189 const struct iphdr *iph = ip_hdr(skb);
190 __sum16 csum = 0;
191
192 switch (skb->ip_summed) {
193 case CHECKSUM_COMPLETE:
194 if (len == skb->len - dataoff)
195 return nf_ip_checksum(skb, hook, dataoff, protocol);
196 /* fall through */
197 case CHECKSUM_NONE:
198 skb->csum = csum_tcpudp_nofold(iph->saddr, iph->daddr, protocol,
199 skb->len - dataoff, 0);
200 skb->ip_summed = CHECKSUM_NONE;
201 return __skb_checksum_complete_head(skb, dataoff + len);
202 }
203 return csum;
204}
205
206static int nf_ip_route(struct net *net, struct dst_entry **dst,
207 struct flowi *fl, bool strict __always_unused)
208{
209 struct rtable *rt = ip_route_output_key(net, &fl->u.ip4);
210 if (IS_ERR(rt))
211 return PTR_ERR(rt);
212 *dst = &rt->dst;
213 return 0;
214}
215
216static const struct nf_afinfo nf_ip_afinfo = {
217 .family = AF_INET,
218 .checksum = nf_ip_checksum,
219 .checksum_partial = nf_ip_checksum_partial,
220 .route = nf_ip_route,
221 .saveroute = nf_ip_saveroute,
222 .reroute = nf_ip_reroute,
223 .route_key_size = sizeof(struct ip_rt_info),
224};
225
226static int ipv4_netfilter_init(void)
227{
228 return nf_register_afinfo(&nf_ip_afinfo);
229}
230
231static void ipv4_netfilter_fini(void)
232{
233 nf_unregister_afinfo(&nf_ip_afinfo);
234}
235
236module_init(ipv4_netfilter_init);
237module_exit(ipv4_netfilter_fini);
238
239#ifdef CONFIG_SYSCTL
240struct ctl_path nf_net_ipv4_netfilter_sysctl_path[] = {
241 { .procname = "net", },
242 { .procname = "ipv4", },
243 { .procname = "netfilter", },
244 { }
245};
246EXPORT_SYMBOL_GPL(nf_net_ipv4_netfilter_sysctl_path);
247#endif /* CONFIG_SYSCTL */
1/* IPv4 specific functions of netfilter core */
2#include <linux/kernel.h>
3#include <linux/netfilter.h>
4#include <linux/netfilter_ipv4.h>
5#include <linux/ip.h>
6#include <linux/skbuff.h>
7#include <linux/gfp.h>
8#include <linux/export.h>
9#include <net/route.h>
10#include <net/xfrm.h>
11#include <net/ip.h>
12#include <net/netfilter/nf_queue.h>
13
14/* route_me_harder function, used by iptable_nat, iptable_mangle + ip_queue */
15int ip_route_me_harder(struct sk_buff *skb, unsigned int addr_type)
16{
17 struct net *net = dev_net(skb_dst(skb)->dev);
18 const struct iphdr *iph = ip_hdr(skb);
19 struct rtable *rt;
20 struct flowi4 fl4 = {};
21 __be32 saddr = iph->saddr;
22 __u8 flags = skb->sk ? inet_sk_flowi_flags(skb->sk) : 0;
23 unsigned int hh_len;
24
25 if (addr_type == RTN_UNSPEC)
26 addr_type = inet_addr_type(net, saddr);
27 if (addr_type == RTN_LOCAL || addr_type == RTN_UNICAST)
28 flags |= FLOWI_FLAG_ANYSRC;
29 else
30 saddr = 0;
31
32 /* some non-standard hacks like ipt_REJECT.c:send_reset() can cause
33 * packets with foreign saddr to appear on the NF_INET_LOCAL_OUT hook.
34 */
35 fl4.daddr = iph->daddr;
36 fl4.saddr = saddr;
37 fl4.flowi4_tos = RT_TOS(iph->tos);
38 fl4.flowi4_oif = skb->sk ? skb->sk->sk_bound_dev_if : 0;
39 fl4.flowi4_mark = skb->mark;
40 fl4.flowi4_flags = flags;
41 rt = ip_route_output_key(net, &fl4);
42 if (IS_ERR(rt))
43 return -1;
44
45 /* Drop old route. */
46 skb_dst_drop(skb);
47 skb_dst_set(skb, &rt->dst);
48
49 if (skb_dst(skb)->error)
50 return -1;
51
52#ifdef CONFIG_XFRM
53 if (!(IPCB(skb)->flags & IPSKB_XFRM_TRANSFORMED) &&
54 xfrm_decode_session(skb, flowi4_to_flowi(&fl4), AF_INET) == 0) {
55 struct dst_entry *dst = skb_dst(skb);
56 skb_dst_set(skb, NULL);
57 dst = xfrm_lookup(net, dst, flowi4_to_flowi(&fl4), skb->sk, 0);
58 if (IS_ERR(dst))
59 return -1;
60 skb_dst_set(skb, dst);
61 }
62#endif
63
64 /* Change in oif may mean change in hh_len. */
65 hh_len = skb_dst(skb)->dev->hard_header_len;
66 if (skb_headroom(skb) < hh_len &&
67 pskb_expand_head(skb, HH_DATA_ALIGN(hh_len - skb_headroom(skb)),
68 0, GFP_ATOMIC))
69 return -1;
70
71 return 0;
72}
73EXPORT_SYMBOL(ip_route_me_harder);
74
75#ifdef CONFIG_XFRM
76int ip_xfrm_me_harder(struct sk_buff *skb)
77{
78 struct flowi fl;
79 unsigned int hh_len;
80 struct dst_entry *dst;
81
82 if (IPCB(skb)->flags & IPSKB_XFRM_TRANSFORMED)
83 return 0;
84 if (xfrm_decode_session(skb, &fl, AF_INET) < 0)
85 return -1;
86
87 dst = skb_dst(skb);
88 if (dst->xfrm)
89 dst = ((struct xfrm_dst *)dst)->route;
90 dst_hold(dst);
91
92 dst = xfrm_lookup(dev_net(dst->dev), dst, &fl, skb->sk, 0);
93 if (IS_ERR(dst))
94 return -1;
95
96 skb_dst_drop(skb);
97 skb_dst_set(skb, dst);
98
99 /* Change in oif may mean change in hh_len. */
100 hh_len = skb_dst(skb)->dev->hard_header_len;
101 if (skb_headroom(skb) < hh_len &&
102 pskb_expand_head(skb, hh_len - skb_headroom(skb), 0, GFP_ATOMIC))
103 return -1;
104 return 0;
105}
106EXPORT_SYMBOL(ip_xfrm_me_harder);
107#endif
108
109void (*ip_nat_decode_session)(struct sk_buff *, struct flowi *);
110EXPORT_SYMBOL(ip_nat_decode_session);
111
112/*
113 * Extra routing may needed on local out, as the QUEUE target never
114 * returns control to the table.
115 */
116
117struct ip_rt_info {
118 __be32 daddr;
119 __be32 saddr;
120 u_int8_t tos;
121 u_int32_t mark;
122};
123
124static void nf_ip_saveroute(const struct sk_buff *skb,
125 struct nf_queue_entry *entry)
126{
127 struct ip_rt_info *rt_info = nf_queue_entry_reroute(entry);
128
129 if (entry->hook == NF_INET_LOCAL_OUT) {
130 const struct iphdr *iph = ip_hdr(skb);
131
132 rt_info->tos = iph->tos;
133 rt_info->daddr = iph->daddr;
134 rt_info->saddr = iph->saddr;
135 rt_info->mark = skb->mark;
136 }
137}
138
139static int nf_ip_reroute(struct sk_buff *skb,
140 const struct nf_queue_entry *entry)
141{
142 const struct ip_rt_info *rt_info = nf_queue_entry_reroute(entry);
143
144 if (entry->hook == NF_INET_LOCAL_OUT) {
145 const struct iphdr *iph = ip_hdr(skb);
146
147 if (!(iph->tos == rt_info->tos &&
148 skb->mark == rt_info->mark &&
149 iph->daddr == rt_info->daddr &&
150 iph->saddr == rt_info->saddr))
151 return ip_route_me_harder(skb, RTN_UNSPEC);
152 }
153 return 0;
154}
155
156__sum16 nf_ip_checksum(struct sk_buff *skb, unsigned int hook,
157 unsigned int dataoff, u_int8_t protocol)
158{
159 const struct iphdr *iph = ip_hdr(skb);
160 __sum16 csum = 0;
161
162 switch (skb->ip_summed) {
163 case CHECKSUM_COMPLETE:
164 if (hook != NF_INET_PRE_ROUTING && hook != NF_INET_LOCAL_IN)
165 break;
166 if ((protocol == 0 && !csum_fold(skb->csum)) ||
167 !csum_tcpudp_magic(iph->saddr, iph->daddr,
168 skb->len - dataoff, protocol,
169 skb->csum)) {
170 skb->ip_summed = CHECKSUM_UNNECESSARY;
171 break;
172 }
173 /* fall through */
174 case CHECKSUM_NONE:
175 if (protocol == 0)
176 skb->csum = 0;
177 else
178 skb->csum = csum_tcpudp_nofold(iph->saddr, iph->daddr,
179 skb->len - dataoff,
180 protocol, 0);
181 csum = __skb_checksum_complete(skb);
182 }
183 return csum;
184}
185EXPORT_SYMBOL(nf_ip_checksum);
186
187static __sum16 nf_ip_checksum_partial(struct sk_buff *skb, unsigned int hook,
188 unsigned int dataoff, unsigned int len,
189 u_int8_t protocol)
190{
191 const struct iphdr *iph = ip_hdr(skb);
192 __sum16 csum = 0;
193
194 switch (skb->ip_summed) {
195 case CHECKSUM_COMPLETE:
196 if (len == skb->len - dataoff)
197 return nf_ip_checksum(skb, hook, dataoff, protocol);
198 /* fall through */
199 case CHECKSUM_NONE:
200 skb->csum = csum_tcpudp_nofold(iph->saddr, iph->daddr, protocol,
201 skb->len - dataoff, 0);
202 skb->ip_summed = CHECKSUM_NONE;
203 return __skb_checksum_complete_head(skb, dataoff + len);
204 }
205 return csum;
206}
207
208static int nf_ip_route(struct net *net, struct dst_entry **dst,
209 struct flowi *fl, bool strict __always_unused)
210{
211 struct rtable *rt = ip_route_output_key(net, &fl->u.ip4);
212 if (IS_ERR(rt))
213 return PTR_ERR(rt);
214 *dst = &rt->dst;
215 return 0;
216}
217
218static const struct nf_afinfo nf_ip_afinfo = {
219 .family = AF_INET,
220 .checksum = nf_ip_checksum,
221 .checksum_partial = nf_ip_checksum_partial,
222 .route = nf_ip_route,
223 .saveroute = nf_ip_saveroute,
224 .reroute = nf_ip_reroute,
225 .route_key_size = sizeof(struct ip_rt_info),
226};
227
228static int ipv4_netfilter_init(void)
229{
230 return nf_register_afinfo(&nf_ip_afinfo);
231}
232
233static void ipv4_netfilter_fini(void)
234{
235 nf_unregister_afinfo(&nf_ip_afinfo);
236}
237
238module_init(ipv4_netfilter_init);
239module_exit(ipv4_netfilter_fini);