Blame view
net/ipv4/ip_input.c
18.3 KB
2874c5fd2
|
1 |
// SPDX-License-Identifier: GPL-2.0-or-later |
1da177e4c
|
2 3 4 5 6 7 8 |
/* * INET An implementation of the TCP/IP protocol suite for the LINUX * operating system. INET is implemented using the BSD Socket * interface as the means of communication with the user level. * * The Internet Protocol (IP) module. * |
02c30a84e
|
9 |
* Authors: Ross Biro |
1da177e4c
|
10 11 |
* Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG> * Donald Becker, <becker@super.org> |
113aa838e
|
12 |
* Alan Cox, <alan@lxorguk.ukuu.org.uk> |
1da177e4c
|
13 14 15 16 |
* Richard Underwood * Stefan Becker, <stefanb@yello.ping.de> * Jorge Cwik, <jorge@laser.satlink.net> * Arnt Gulbrandsen, <agulbra@nvg.unit.no> |
e905a9eda
|
17 |
* |
1da177e4c
|
18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 |
* Fixes: * Alan Cox : Commented a couple of minor bits of surplus code * Alan Cox : Undefining IP_FORWARD doesn't include the code * (just stops a compiler warning). * Alan Cox : Frames with >=MAX_ROUTE record routes, strict routes or loose routes * are junked rather than corrupting things. * Alan Cox : Frames to bad broadcast subnets are dumped * We used to process them non broadcast and * boy could that cause havoc. * Alan Cox : ip_forward sets the free flag on the * new frame it queues. Still crap because * it copies the frame but at least it * doesn't eat memory too. * Alan Cox : Generic queue code and memory fixes. * Fred Van Kempen : IP fragment support (borrowed from NET2E) * Gerhard Koerting: Forward fragmented frames correctly. * Gerhard Koerting: Fixes to my fix of the above 8-). * Gerhard Koerting: IP interface addressing fix. * Linus Torvalds : More robustness checks * Alan Cox : Even more checks: Still not as robust as it ought to be * Alan Cox : Save IP header pointer for later * Alan Cox : ip option setting * Alan Cox : Use ip_tos/ip_ttl settings * Alan Cox : Fragmentation bogosity removed * (Thanks to Mark.Bush@prg.ox.ac.uk) * Dmitry Gorodchanin : Send of a raw packet crash fix. * Alan Cox : Silly ip bug when an overlength * fragment turns up. Now frees the * queue. * Linus Torvalds/ : Memory leakage on fragmentation * Alan Cox : handling. * Gerhard Koerting: Forwarding uses IP priority hints * Teemu Rantanen : Fragment problems. * Alan Cox : General cleanup, comments and reformat * Alan Cox : SNMP statistics * Alan Cox : BSD address rule semantics. Also see * UDP as there is a nasty checksum issue * if you do things the wrong way. * Alan Cox : Always defrag, moved IP_FORWARD to the config.in file * Alan Cox : IP options adjust sk->priority. * Pedro Roque : Fix mtu/length error in ip_forward. * Alan Cox : Avoid ip_chk_addr when possible. * Richard Underwood : IP multicasting. * Alan Cox : Cleaned up multicast handlers. * Alan Cox : RAW sockets demultiplex in the BSD style. * Gunther Mayer : Fix the SNMP reporting typo * Alan Cox : Always in group 224.0.0.1 * Pauline Middelink : Fast ip_checksum update when forwarding * Masquerading support. * Alan Cox : Multicast loopback error for 224.0.0.1 * Alan Cox : IP_MULTICAST_LOOP option. * Alan Cox : Use notifiers. * Bjorn Ekwall : Removed ip_csum (from slhc.c too) * Bjorn Ekwall : Moved ip_fast_csum to ip.h (inline!) * Stefan Becker : Send out ICMP HOST REDIRECT * Arnt Gulbrandsen : ip_build_xmit * Alan Cox : Per socket routing cache * Alan Cox : Fixed routing cache, added header cache. * Alan Cox : Loopback didn't work right in original ip_build_xmit - fixed it. * Alan Cox : Only send ICMP_REDIRECT if src/dest are the same net. * Alan Cox : Incoming IP option handling. * Alan Cox : Set saddr on raw output frames as per BSD. * Alan Cox : Stopped broadcast source route explosions. * Alan Cox : Can disable source routing * Takeshi Sone : Masquerading didn't work. * Dave Bonn,Alan Cox : Faster IP forwarding whenever possible. * Alan Cox : Memory leaks, tramples, misc debugging. * Alan Cox : Fixed multicast (by popular demand 8)) * Alan Cox : Fixed forwarding (by even more popular demand 8)) * Alan Cox : Fixed SNMP statistics [I think] * Gerhard Koerting : IP fragmentation forwarding fix * Alan Cox : Device lock against page fault. * Alan Cox : IP_HDRINCL facility. * Werner Almesberger : Zero fragment bug * Alan Cox : RAW IP frame length bug * Alan Cox : Outgoing firewall on build_xmit * A.N.Kuznetsov : IP_OPTIONS support throughout the kernel * Alan Cox : Multicast routing hooks * Jos Vos : Do accounting *before* call_in_firewall * Willy Konynenberg : Transparent proxying support * |
1da177e4c
|
99 100 101 102 |
* To Fix: * IP fragmentation wants rewriting cleanly. The RFC815 algorithm is much more efficient * and could be made very efficient with the addition of some virtual memory hacks to permit * the allocation of a buffer that can then be 'grown' by twiddling page tables. |
e905a9eda
|
103 |
* Output fragmentation wants updating along with the buffer management to use a single |
1da177e4c
|
104 105 106 |
* interleaved copy algorithm so that fragmenting has a one copy overhead. Actual packet * output should probably do its own fragmentation at the UDP/RAW layer. TCP shouldn't cause * fragmentation anyway. |
1da177e4c
|
107 |
*/ |
afd465030
|
108 |
#define pr_fmt(fmt) "IPv4: " fmt |
1da177e4c
|
109 110 111 112 113 |
#include <linux/module.h> #include <linux/types.h> #include <linux/kernel.h> #include <linux/string.h> #include <linux/errno.h> |
5a0e3ad6a
|
114 |
#include <linux/slab.h> |
1da177e4c
|
115 116 117 118 119 120 |
#include <linux/net.h> #include <linux/socket.h> #include <linux/sockios.h> #include <linux/in.h> #include <linux/inet.h> |
14c850212
|
121 |
#include <linux/inetdevice.h> |
1da177e4c
|
122 123 |
#include <linux/netdevice.h> #include <linux/etherdevice.h> |
0e219ae48
|
124 |
#include <linux/indirect_call_wrapper.h> |
1da177e4c
|
125 126 127 128 129 130 131 132 133 134 135 |
#include <net/snmp.h> #include <net/ip.h> #include <net/protocol.h> #include <net/route.h> #include <linux/skbuff.h> #include <net/sock.h> #include <net/arp.h> #include <net/icmp.h> #include <net/raw.h> #include <net/checksum.h> |
1f07d03e2
|
136 |
#include <net/inet_ecn.h> |
1da177e4c
|
137 138 139 140 |
#include <linux/netfilter_ipv4.h> #include <net/xfrm.h> #include <linux/mroute.h> #include <linux/netlink.h> |
f38a9eb1f
|
141 |
#include <net/dst_metadata.h> |
1da177e4c
|
142 143 |
/* |
66018506e
|
144 |
* Process Router Attention IP option (RFC 2113) |
e905a9eda
|
145 |
*/ |
ba57b4db2
|
146 |
bool ip_call_ra_chain(struct sk_buff *skb) |
1da177e4c
|
147 148 |
{ struct ip_ra_chain *ra; |
eddc9ec53
|
149 |
u8 protocol = ip_hdr(skb)->protocol; |
1da177e4c
|
150 |
struct sock *last = NULL; |
cb84663e4
|
151 |
struct net_device *dev = skb->dev; |
37fcbab61
|
152 |
struct net *net = dev_net(dev); |
1da177e4c
|
153 |
|
5796ef75e
|
154 |
for (ra = rcu_dereference(net->ipv4.ra_chain); ra; ra = rcu_dereference(ra->next)) { |
1da177e4c
|
155 156 157 158 159 |
struct sock *sk = ra->sk; /* If socket is bound to an interface, only report * the packet if it came from that interface. */ |
c720c7e83
|
160 |
if (sk && inet_sk(sk)->inet_num == protocol && |
1da177e4c
|
161 |
(!sk->sk_bound_dev_if || |
5796ef75e
|
162 |
sk->sk_bound_dev_if == dev->ifindex)) { |
56f8a75c1
|
163 |
if (ip_is_fragment(ip_hdr(skb))) { |
19bcf9f20
|
164 |
if (ip_defrag(net, skb, IP_DEFRAG_CALL_RA_CHAIN)) |
ba57b4db2
|
165 |
return true; |
1da177e4c
|
166 167 168 169 170 171 172 173 174 175 176 177 |
} if (last) { struct sk_buff *skb2 = skb_clone(skb, GFP_ATOMIC); if (skb2) raw_rcv(last, skb2); } last = sk; } } if (last) { raw_rcv(last, skb); |
ba57b4db2
|
178 |
return true; |
1da177e4c
|
179 |
} |
ba57b4db2
|
180 |
return false; |
1da177e4c
|
181 |
} |
0e219ae48
|
182 183 |
INDIRECT_CALLABLE_DECLARE(int udp_rcv(struct sk_buff *)); INDIRECT_CALLABLE_DECLARE(int tcp_v4_rcv(struct sk_buff *)); |
68cb7d531
|
184 |
void ip_protocol_deliver_rcu(struct net *net, struct sk_buff *skb, int protocol) |
1da177e4c
|
185 |
{ |
68cb7d531
|
186 187 |
const struct net_protocol *ipprot; int raw, ret; |
1da177e4c
|
188 |
|
68cb7d531
|
189 190 191 192 193 194 195 196 197 |
resubmit: raw = raw_local_deliver(skb, protocol); ipprot = rcu_dereference(inet_protos[protocol]); if (ipprot) { if (!ipprot->no_policy) { if (!xfrm4_policy_check(NULL, XFRM_POLICY_IN, skb)) { kfree_skb(skb); return; |
1da177e4c
|
198 |
} |
895b5c9f2
|
199 |
nf_reset_ct(skb); |
68cb7d531
|
200 |
} |
0e219ae48
|
201 202 |
ret = INDIRECT_CALL_2(ipprot->handler, tcp_v4_rcv, udp_rcv, skb); |
68cb7d531
|
203 204 205 206 207 208 209 210 211 212 213 |
if (ret < 0) { protocol = -ret; goto resubmit; } __IP_INC_STATS(net, IPSTATS_MIB_INDELIVERS); } else { if (!raw) { if (xfrm4_policy_check(NULL, XFRM_POLICY_IN, skb)) { __IP_INC_STATS(net, IPSTATS_MIB_INUNKNOWNPROTOS); icmp_send(skb, ICMP_DEST_UNREACH, ICMP_PROT_UNREACH, 0); |
1da177e4c
|
214 |
} |
68cb7d531
|
215 |
kfree_skb(skb); |
1da177e4c
|
216 |
} else { |
68cb7d531
|
217 218 |
__IP_INC_STATS(net, IPSTATS_MIB_INDELIVERS); consume_skb(skb); |
1da177e4c
|
219 220 |
} } |
68cb7d531
|
221 222 223 224 225 226 227 228 |
} static int ip_local_deliver_finish(struct net *net, struct sock *sk, struct sk_buff *skb) { __skb_pull(skb, skb_network_header_len(skb)); rcu_read_lock(); ip_protocol_deliver_rcu(net, skb, ip_hdr(skb)->protocol); |
1da177e4c
|
229 230 231 232 233 234 235 |
rcu_read_unlock(); return 0; } /* * Deliver IP Packets to the higher protocol layers. |
e905a9eda
|
236 |
*/ |
1da177e4c
|
237 238 239 240 241 |
int ip_local_deliver(struct sk_buff *skb) { /* * Reassemble IP fragments. */ |
19bcf9f20
|
242 |
struct net *net = dev_net(skb->dev); |
1da177e4c
|
243 |
|
56f8a75c1
|
244 |
if (ip_is_fragment(ip_hdr(skb))) { |
19bcf9f20
|
245 |
if (ip_defrag(net, skb, IP_DEFRAG_LOCAL_DELIVER)) |
1da177e4c
|
246 247 |
return 0; } |
29a26a568
|
248 |
return NF_HOOK(NFPROTO_IPV4, NF_INET_LOCAL_IN, |
19bcf9f20
|
249 |
net, NULL, skb, skb->dev, NULL, |
1da177e4c
|
250 251 |
ip_local_deliver_finish); } |
8c83f2df9
|
252 |
static inline bool ip_rcv_options(struct sk_buff *skb, struct net_device *dev) |
d245407e7
|
253 254 |
{ struct ip_options *opt; |
b71d1d426
|
255 |
const struct iphdr *iph; |
d245407e7
|
256 257 258 259 260 261 262 263 264 |
/* It looks as overkill, because not all IP options require packet mangling. But it is the easiest for now, especially taking into account that combination of IP options and running sniffer is extremely rare condition. --ANK (980813) */ if (skb_cow(skb, skb_headroom(skb))) { |
b45386efa
|
265 |
__IP_INC_STATS(dev_net(dev), IPSTATS_MIB_INDISCARDS); |
d245407e7
|
266 267 |
goto drop; } |
eddc9ec53
|
268 |
iph = ip_hdr(skb); |
22aba383c
|
269 270 |
opt = &(IPCB(skb)->opt); opt->optlen = iph->ihl*4 - sizeof(struct iphdr); |
d245407e7
|
271 |
|
c346dca10
|
272 |
if (ip_options_compile(dev_net(dev), opt, skb)) { |
b45386efa
|
273 |
__IP_INC_STATS(dev_net(dev), IPSTATS_MIB_INHDRERRORS); |
d245407e7
|
274 275 |
goto drop; } |
d245407e7
|
276 |
if (unlikely(opt->srr)) { |
6e8b11b43
|
277 |
struct in_device *in_dev = __in_dev_get_rcu(dev); |
d245407e7
|
278 279 |
if (in_dev) { if (!IN_DEV_SOURCE_ROUTE(in_dev)) { |
e87cc4728
|
280 281 282 283 284 |
if (IN_DEV_LOG_MARTIANS(in_dev)) net_info_ratelimited("source route option %pI4 -> %pI4 ", &iph->saddr, &iph->daddr); |
d245407e7
|
285 286 |
goto drop; } |
d245407e7
|
287 |
} |
8c83f2df9
|
288 |
if (ip_options_rcv_srr(skb, dev)) |
d245407e7
|
289 290 |
goto drop; } |
6a91395f2
|
291 |
return false; |
d245407e7
|
292 |
drop: |
6a91395f2
|
293 |
return true; |
d245407e7
|
294 |
} |
02b249416
|
295 296 297 298 299 300 |
static bool ip_can_use_hint(const struct sk_buff *skb, const struct iphdr *iph, const struct sk_buff *hint) { return hint && !skb_dst(skb) && ip_hdr(hint)->daddr == iph->daddr && ip_hdr(hint)->tos == iph->tos; } |
97ff7ffb1
|
301 302 |
INDIRECT_CALLABLE_DECLARE(int udp_v4_early_demux(struct sk_buff *)); INDIRECT_CALLABLE_DECLARE(int tcp_v4_early_demux(struct sk_buff *)); |
5fa12739a
|
303 |
static int ip_rcv_finish_core(struct net *net, struct sock *sk, |
02b249416
|
304 305 |
struct sk_buff *skb, struct net_device *dev, const struct sk_buff *hint) |
1da177e4c
|
306 |
{ |
eddc9ec53
|
307 |
const struct iphdr *iph = ip_hdr(skb); |
7487449c8
|
308 |
int (*edemux)(struct sk_buff *skb); |
7487449c8
|
309 310 |
struct rtable *rt; int err; |
1da177e4c
|
311 |
|
02b249416
|
312 313 314 315 316 317 |
if (ip_can_use_hint(skb, iph, hint)) { err = ip_route_use_hint(skb, iph->daddr, iph->saddr, iph->tos, dev, hint); if (unlikely(err)) goto drop_error; } |
e21145a98
|
318 |
if (net->ipv4.sysctl_ip_early_demux && |
63e51b6a2
|
319 320 321 |
!skb_dst(skb) && !skb->sk && !ip_is_fragment(iph)) { |
160eb5a6b
|
322 323 |
const struct net_protocol *ipprot; int protocol = iph->protocol; |
160eb5a6b
|
324 |
ipprot = rcu_dereference(inet_protos[protocol]); |
dddb64bcb
|
325 |
if (ipprot && (edemux = READ_ONCE(ipprot->early_demux))) { |
97ff7ffb1
|
326 327 |
err = INDIRECT_CALL_2(edemux, tcp_v4_early_demux, udp_v4_early_demux, skb); |
7487449c8
|
328 329 |
if (unlikely(err)) goto drop_error; |
9cb429d69
|
330 331 332 |
/* must reload iph, skb->head might have changed */ iph = ip_hdr(skb); } |
160eb5a6b
|
333 |
} |
1da177e4c
|
334 335 336 |
/* * Initialise the virtual path cache for the packet. It describes * how the packet travels inside Linux networking. |
e905a9eda
|
337 |
*/ |
f38a9eb1f
|
338 |
if (!skb_valid_dst(skb)) { |
7487449c8
|
339 340 341 342 |
err = ip_route_input_noref(skb, iph->daddr, iph->saddr, iph->tos, dev); if (unlikely(err)) goto drop_error; |
1da177e4c
|
343 |
} |
c7066f70d
|
344 |
#ifdef CONFIG_IP_ROUTE_CLASSID |
adf30907d
|
345 |
if (unlikely(skb_dst(skb)->tclassid)) { |
7a9b2d595
|
346 |
struct ip_rt_acct *st = this_cpu_ptr(ip_rt_acct); |
adf30907d
|
347 |
u32 idx = skb_dst(skb)->tclassid; |
1da177e4c
|
348 |
st[idx&0xFF].o_packets++; |
fd3f8c4cb
|
349 |
st[idx&0xFF].o_bytes += skb->len; |
1da177e4c
|
350 |
st[(idx>>16)&0xFF].i_packets++; |
fd3f8c4cb
|
351 |
st[(idx>>16)&0xFF].i_bytes += skb->len; |
1da177e4c
|
352 353 |
} #endif |
8c83f2df9
|
354 |
if (iph->ihl > 5 && ip_rcv_options(skb, dev)) |
d245407e7
|
355 |
goto drop; |
1da177e4c
|
356 |
|
511c3f92a
|
357 |
rt = skb_rtable(skb); |
edf391ff1
|
358 |
if (rt->rt_type == RTN_MULTICAST) { |
b15084ec7
|
359 |
__IP_UPD_PO_STATS(net, IPSTATS_MIB_INMCAST, skb->len); |
12b74dfad
|
360 |
} else if (rt->rt_type == RTN_BROADCAST) { |
b15084ec7
|
361 |
__IP_UPD_PO_STATS(net, IPSTATS_MIB_INBCAST, skb->len); |
12b74dfad
|
362 363 |
} else if (skb->pkt_type == PACKET_BROADCAST || skb->pkt_type == PACKET_MULTICAST) { |
d6f64d725
|
364 |
struct in_device *in_dev = __in_dev_get_rcu(dev); |
12b74dfad
|
365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 |
/* RFC 1122 3.3.6: * * When a host sends a datagram to a link-layer broadcast * address, the IP destination address MUST be a legal IP * broadcast or IP multicast address. * * A host SHOULD silently discard a datagram that is received * via a link-layer broadcast (see Section 2.4) but does not * specify an IP multicast or broadcast destination address. * * This doesn't explicitly say L2 *broadcast*, but broadcast is * in a way a form of multicast and the most common use case for * this is 802.11 protecting against cross-station spoofing (the * so-called "hole-196" attack) so do it for both. */ if (in_dev && IN_DEV_ORCONF(in_dev, DROP_UNICAST_IN_L2_MULTICAST)) goto drop; } |
5506b54b3
|
385 |
|
5fa12739a
|
386 |
return NET_RX_SUCCESS; |
1da177e4c
|
387 |
|
1da177e4c
|
388 |
drop: |
e905a9eda
|
389 390 |
kfree_skb(skb); return NET_RX_DROP; |
7487449c8
|
391 392 393 394 395 |
drop_error: if (err == -EXDEV) __NET_INC_STATS(net, LINUX_MIB_IPRPFILTER); goto drop; |
1da177e4c
|
396 |
} |
5fa12739a
|
397 398 |
static int ip_rcv_finish(struct net *net, struct sock *sk, struct sk_buff *skb) { |
a1fd1ad25
|
399 |
struct net_device *dev = skb->dev; |
efe6aaca6
|
400 401 402 403 404 405 406 407 |
int ret; /* if ingress device is enslaved to an L3 master device pass the * skb to its handler for processing */ skb = l3mdev_ip_rcv(skb); if (!skb) return NET_RX_SUCCESS; |
5fa12739a
|
408 |
|
02b249416
|
409 |
ret = ip_rcv_finish_core(net, sk, skb, dev, NULL); |
5fa12739a
|
410 411 412 413 |
if (ret != NET_RX_DROP) ret = dst_input(skb); return ret; } |
1da177e4c
|
414 415 |
/* * Main IP Receive routine. |
e905a9eda
|
416 |
*/ |
17266ee93
|
417 |
static struct sk_buff *ip_rcv_core(struct sk_buff *skb, struct net *net) |
1da177e4c
|
418 |
{ |
b71d1d426
|
419 |
const struct iphdr *iph; |
586152424
|
420 |
u32 len; |
1da177e4c
|
421 422 423 424 425 426 |
/* When the interface is in promisc. mode, drop all the crap * that it receives, do not try to analyse it. */ if (skb->pkt_type == PACKET_OTHERHOST) goto drop; |
b15084ec7
|
427 |
__IP_UPD_PO_STATS(net, IPSTATS_MIB_IN, skb->len); |
1da177e4c
|
428 |
|
51456b291
|
429 430 |
skb = skb_share_check(skb, GFP_ATOMIC); if (!skb) { |
b45386efa
|
431 |
__IP_INC_STATS(net, IPSTATS_MIB_INDISCARDS); |
1da177e4c
|
432 433 434 435 436 |
goto out; } if (!pskb_may_pull(skb, sizeof(struct iphdr))) goto inhdr_error; |
eddc9ec53
|
437 |
iph = ip_hdr(skb); |
1da177e4c
|
438 439 |
/* |
c67fa0279
|
440 |
* RFC1122: 3.2.1.2 MUST silently discard any IP frame that fails the checksum. |
1da177e4c
|
441 442 443 444 445 446 447 448 449 450 |
* * Is the datagram acceptable? * * 1. Length at least the size of an ip header * 2. Version of 4 * 3. Checksums correctly. [Speed optimisation for later, skip loopback checksums] * 4. Doesn't have a bogus length */ if (iph->ihl < 5 || iph->version != 4) |
586152424
|
451 |
goto inhdr_error; |
1da177e4c
|
452 |
|
1f07d03e2
|
453 454 455 |
BUILD_BUG_ON(IPSTATS_MIB_ECT1PKTS != IPSTATS_MIB_NOECTPKTS + INET_ECN_ECT_1); BUILD_BUG_ON(IPSTATS_MIB_ECT0PKTS != IPSTATS_MIB_NOECTPKTS + INET_ECN_ECT_0); BUILD_BUG_ON(IPSTATS_MIB_CEPKTS != IPSTATS_MIB_NOECTPKTS + INET_ECN_CE); |
98f619957
|
456 457 458 |
__IP_ADD_STATS(net, IPSTATS_MIB_NOECTPKTS + (iph->tos & INET_ECN_MASK), max_t(unsigned short, 1, skb_shinfo(skb)->gso_segs)); |
1f07d03e2
|
459 |
|
1da177e4c
|
460 461 |
if (!pskb_may_pull(skb, iph->ihl*4)) goto inhdr_error; |
eddc9ec53
|
462 |
iph = ip_hdr(skb); |
1da177e4c
|
463 |
|
e9c604227
|
464 |
if (unlikely(ip_fast_csum((u8 *)iph, iph->ihl))) |
6a5dc9e59
|
465 |
goto csum_error; |
1da177e4c
|
466 |
|
586152424
|
467 |
len = ntohs(iph->tot_len); |
704aed53b
|
468 |
if (skb->len < len) { |
b45386efa
|
469 |
__IP_INC_STATS(net, IPSTATS_MIB_INTRUNCATEDPKTS); |
704aed53b
|
470 471 |
goto drop; } else if (len < (iph->ihl*4)) |
586152424
|
472 |
goto inhdr_error; |
1da177e4c
|
473 |
|
586152424
|
474 475 476 477 478 |
/* Our transport medium may have padded the buffer out. Now we know it * is IP we can trim to the true length of the frame. * Note this now means skb->len holds ntohs(iph->tot_len). */ if (pskb_trim_rcsum(skb, len)) { |
b45386efa
|
479 |
__IP_INC_STATS(net, IPSTATS_MIB_INDISCARDS); |
586152424
|
480 |
goto drop; |
1da177e4c
|
481 |
} |
6c57f0458
|
482 |
iph = ip_hdr(skb); |
21d1196a3
|
483 |
skb->transport_header = skb->network_header + iph->ihl*4; |
53602f92d
|
484 |
/* Remove any debris in the socket control block */ |
d569f1d72
|
485 |
memset(IPCB(skb), 0, sizeof(struct inet_skb_parm)); |
0b922b7a8
|
486 |
IPCB(skb)->iif = skb->skb_iif; |
53602f92d
|
487 |
|
71f9dacd2
|
488 |
/* Must drop socket now because of tproxy. */ |
cf7fbe660
|
489 490 |
if (!skb_sk_is_prefetched(skb)) skb_orphan(skb); |
71f9dacd2
|
491 |
|
17266ee93
|
492 |
return skb; |
1da177e4c
|
493 |
|
6a5dc9e59
|
494 |
csum_error: |
b45386efa
|
495 |
__IP_INC_STATS(net, IPSTATS_MIB_CSUMERRORS); |
1da177e4c
|
496 |
inhdr_error: |
b45386efa
|
497 |
__IP_INC_STATS(net, IPSTATS_MIB_INHDRERRORS); |
1da177e4c
|
498 |
drop: |
e905a9eda
|
499 |
kfree_skb(skb); |
1da177e4c
|
500 |
out: |
17266ee93
|
501 502 503 504 505 506 507 508 509 510 511 512 513 514 |
return NULL; } /* * IP receive entry point */ int ip_rcv(struct sk_buff *skb, struct net_device *dev, struct packet_type *pt, struct net_device *orig_dev) { struct net *net = dev_net(dev); skb = ip_rcv_core(skb, net); if (skb == NULL) return NET_RX_DROP; |
fb1b69999
|
515 |
|
17266ee93
|
516 517 518 519 |
return NF_HOOK(NFPROTO_IPV4, NF_INET_PRE_ROUTING, net, NULL, skb, dev, NULL, ip_rcv_finish); } |
5fa12739a
|
520 |
static void ip_sublist_rcv_finish(struct list_head *head) |
17266ee93
|
521 522 |
{ struct sk_buff *skb, *next; |
0761680d5
|
523 |
list_for_each_entry_safe(skb, next, head, list) { |
992cba7e2
|
524 |
skb_list_del_init(skb); |
5fa12739a
|
525 |
dst_input(skb); |
0761680d5
|
526 |
} |
5fa12739a
|
527 |
} |
02b249416
|
528 529 530 531 532 533 534 535 |
static struct sk_buff *ip_extract_route_hint(const struct net *net, struct sk_buff *skb, int rt_type) { if (fib4_has_custom_rules(net) || rt_type == RTN_BROADCAST) return NULL; return skb; } |
5fa12739a
|
536 537 538 |
static void ip_list_rcv_finish(struct net *net, struct sock *sk, struct list_head *head) { |
02b249416
|
539 |
struct sk_buff *skb, *next, *hint = NULL; |
5fa12739a
|
540 |
struct dst_entry *curr_dst = NULL; |
5fa12739a
|
541 |
struct list_head sublist; |
a4ca8b7df
|
542 |
INIT_LIST_HEAD(&sublist); |
5fa12739a
|
543 |
list_for_each_entry_safe(skb, next, head, list) { |
a1fd1ad25
|
544 |
struct net_device *dev = skb->dev; |
5fa12739a
|
545 |
struct dst_entry *dst; |
22f6bbb7b
|
546 |
skb_list_del_init(skb); |
efe6aaca6
|
547 548 549 550 551 552 |
/* if ingress device is enslaved to an L3 master device pass the * skb to its handler for processing */ skb = l3mdev_ip_rcv(skb); if (!skb) continue; |
02b249416
|
553 |
if (ip_rcv_finish_core(net, sk, skb, dev, hint) == NET_RX_DROP) |
5fa12739a
|
554 555 556 557 |
continue; dst = skb_dst(skb); if (curr_dst != dst) { |
02b249416
|
558 559 |
hint = ip_extract_route_hint(net, skb, ((struct rtable *)dst)->rt_type); |
5fa12739a
|
560 |
/* dispatch old sublist */ |
5fa12739a
|
561 562 563 |
if (!list_empty(&sublist)) ip_sublist_rcv_finish(&sublist); /* start new sublist */ |
a4ca8b7df
|
564 |
INIT_LIST_HEAD(&sublist); |
5fa12739a
|
565 566 |
curr_dst = dst; } |
a4ca8b7df
|
567 |
list_add_tail(&skb->list, &sublist); |
5fa12739a
|
568 569 |
} /* dispatch final sublist */ |
a4ca8b7df
|
570 |
ip_sublist_rcv_finish(&sublist); |
5fa12739a
|
571 572 573 574 575 |
} static void ip_sublist_rcv(struct list_head *head, struct net_device *dev, struct net *net) { |
17266ee93
|
576 577 |
NF_HOOK_LIST(NFPROTO_IPV4, NF_INET_PRE_ROUTING, net, NULL, head, dev, NULL, ip_rcv_finish); |
5fa12739a
|
578 |
ip_list_rcv_finish(net, NULL, head); |
17266ee93
|
579 580 581 582 583 584 585 586 587 588 |
} /* Receive a list of IP packets */ void ip_list_rcv(struct list_head *head, struct packet_type *pt, struct net_device *orig_dev) { struct net_device *curr_dev = NULL; struct net *curr_net = NULL; struct sk_buff *skb, *next; struct list_head sublist; |
a4ca8b7df
|
589 |
INIT_LIST_HEAD(&sublist); |
17266ee93
|
590 591 592 |
list_for_each_entry_safe(skb, next, head, list) { struct net_device *dev = skb->dev; struct net *net = dev_net(dev); |
22f6bbb7b
|
593 |
skb_list_del_init(skb); |
17266ee93
|
594 595 596 597 598 599 |
skb = ip_rcv_core(skb, net); if (skb == NULL) continue; if (curr_dev != dev || curr_net != net) { /* dispatch old sublist */ |
17266ee93
|
600 |
if (!list_empty(&sublist)) |
a4ca8b7df
|
601 |
ip_sublist_rcv(&sublist, curr_dev, curr_net); |
17266ee93
|
602 |
/* start new sublist */ |
a4ca8b7df
|
603 |
INIT_LIST_HEAD(&sublist); |
17266ee93
|
604 605 606 |
curr_dev = dev; curr_net = net; } |
a4ca8b7df
|
607 |
list_add_tail(&skb->list, &sublist); |
17266ee93
|
608 609 |
} /* dispatch final sublist */ |
51210ad5a
|
610 611 |
if (!list_empty(&sublist)) ip_sublist_rcv(&sublist, curr_dev, curr_net); |
1da177e4c
|
612 |
} |