summaryrefslogtreecommitdiff
path: root/pfinet/linux-src/net/ipv4/icmp.c
blob: 7fad4783571ed081ee2ebe232459624b0310b380 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
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
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
/*
 *	NET3:	Implementation of the ICMP protocol layer. 
 *	
 *		Alan Cox, <alan@redhat.com>
 *
 *	Version: $Id: icmp.c,v 1.52.2.4 1999/11/16 02:28:40 davem Exp $
 *
 *	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.
 *
 *	Some of the function names and the icmp unreach table for this
 *	module were derived from [icmp.c 1.0.11 06/02/93] by
 *	Ross Biro, Fred N. van Kempen, Mark Evans, Alan Cox, Gerhard Koerting.
 *	Other than that this module is a complete rewrite.
 *
 *	Fixes:
 *		Mike Shaver	:	RFC1122 checks.
 *		Alan Cox	:	Multicast ping reply as self.
 *		Alan Cox	:	Fix atomicity lockup in ip_build_xmit 
 *					call.
 *		Alan Cox	:	Added 216,128 byte paths to the MTU 
 *					code.
 *		Martin Mares	:	RFC1812 checks.
 *		Martin Mares	:	Can be configured to follow redirects 
 *					if acting as a router _without_ a
 *					routing protocol (RFC 1812).
 *		Martin Mares	:	Echo requests may be configured to 
 *					be ignored (RFC 1812).
 *		Martin Mares	:	Limitation of ICMP error message 
 *					transmit rate (RFC 1812).
 *		Martin Mares	:	TOS and Precedence set correctly 
 *					(RFC 1812).
 *		Martin Mares	:	Now copying as much data from the 
 *					original packet as we can without
 *					exceeding 576 bytes (RFC 1812).
 *	Willy Konynenberg	:	Transparent proxying support.
 *		Keith Owens	:	RFC1191 correction for 4.2BSD based 
 *					path MTU bug.
 *		Thomas Quinot	:	ICMP Dest Unreach codes up to 15 are
 *					valid (RFC 1812).
 *		Andi Kleen	:	Check all packet lengths properly
 *					and moved all kfree_skb() up to
 *					icmp_rcv.
 *		Andi Kleen	:	Move the rate limit bookkeeping
 *					into the dest entry and use a token
 *					bucket filter (thanks to ANK). Make
 *					the rates sysctl configurable.
 *		Yu Tianli	:	Fixed two ugly bugs in icmp_send
 *					- IP option length was accounted wrongly
 *					- ICMP header length was not accounted at all.
 *              Tristan Greaves :       Added sysctl option to ignore bogus broadcast
 *                                      responses from broken routers.
 *
 * To Fix:
 *
 *	- Should use skb_pull() instead of all the manual checking.
 *	  This would also greatly simply some upper layer error handlers. --AK
 *
 * RFC1122 (Host Requirements -- Comm. Layer) Status:
 * (boy, are there a lot of rules for ICMP)
 *  3.2.2 (Generic ICMP stuff)
 *   MUST discard messages of unknown type. (OK)
 *   MUST copy at least the first 8 bytes from the offending packet
 *     when sending ICMP errors. (OBSOLETE -- see RFC1812)
 *   MUST pass received ICMP errors up to protocol level. (OK)
 *   SHOULD send ICMP errors with TOS == 0. (OBSOLETE -- see RFC1812)
 *   MUST NOT send ICMP errors in reply to:
 *     ICMP errors (OK)
 *     Broadcast/multicast datagrams (OK)
 *     MAC broadcasts (OK)
 *     Non-initial fragments (OK)
 *     Datagram with a source address that isn't a single host. (OK)
 *  3.2.2.1 (Destination Unreachable)
 *   All the rules govern the IP layer, and are dealt with in ip.c, not here.
 *  3.2.2.2 (Redirect)
 *   Host SHOULD NOT send ICMP_REDIRECTs.  (OK)
 *   MUST update routing table in response to host or network redirects.
 *     (host OK, network OBSOLETE)
 *   SHOULD drop redirects if they're not from directly connected gateway
 *     (OK -- we drop it if it's not from our old gateway, which is close
 *      enough)
 * 3.2.2.3 (Source Quench)
 *   MUST pass incoming SOURCE_QUENCHs to transport layer (OK)
 *   Other requirements are dealt with at the transport layer.
 * 3.2.2.4 (Time Exceeded)
 *   MUST pass TIME_EXCEEDED to transport layer (OK)
 *   Other requirements dealt with at IP (generating TIME_EXCEEDED).
 * 3.2.2.5 (Parameter Problem)
 *   SHOULD generate these (OK)
 *   MUST pass received PARAMPROBLEM to transport layer (NOT YET)
 *   	[Solaris 2.X seems to assert EPROTO when this occurs] -- AC
 * 3.2.2.6 (Echo Request/Reply)
 *   MUST reply to ECHO_REQUEST, and give app to do ECHO stuff (OK, OK)
 *   MAY discard broadcast ECHO_REQUESTs. (Configurable with a sysctl.)
 *   MUST reply using same source address as the request was sent to.
 *     We're OK for unicast ECHOs, and it doesn't say anything about
 *     how to handle broadcast ones, since it's optional.
 *   MUST copy data from REQUEST to REPLY (OK)
 *     unless it would require illegal fragmentation (OK)
 *   MUST pass REPLYs to transport/user layer (OK)
 *   MUST use any provided source route (reversed) for REPLY. (NOT YET)
 * 3.2.2.7 (Information Request/Reply)
 *   MUST NOT implement this. (I guess that means silently discard...?) (OK)
 * 3.2.2.8 (Timestamp Request/Reply)
 *   MAY implement (OK)
 *   SHOULD be in-kernel for "minimum variability" (OK)
 *   MAY discard broadcast REQUESTs.  (OK, but see source for inconsistency)
 *   MUST reply using same source address as the request was sent to. (OK)
 *   MUST reverse source route, as per ECHO (NOT YET)
 *   MUST pass REPLYs to transport/user layer (requires RAW, just like 
 *	ECHO) (OK)
 *   MUST update clock for timestamp at least 15 times/sec (OK)
 *   MUST be "correct within a few minutes" (OK)
 * 3.2.2.9 (Address Mask Request/Reply)
 *   MAY implement (OK)
 *   MUST send a broadcast REQUEST if using this system to set netmask
 *     (OK... we don't use it)
 *   MUST discard received REPLYs if not using this system (OK)
 *   MUST NOT send replies unless specifically made agent for this sort
 *     of thing. (OK)
 *
 *
 * RFC 1812 (IPv4 Router Requirements) Status (even longer):
 *  4.3.2.1 (Unknown Message Types)
 *   MUST pass messages of unknown type to ICMP user iface or silently discard
 *     them (OK)
 *  4.3.2.2 (ICMP Message TTL)
 *   MUST initialize TTL when originating an ICMP message (OK)
 *  4.3.2.3 (Original Message Header)
 *   SHOULD copy as much data from the offending packet as possible without
 *     the length of the ICMP datagram exceeding 576 bytes (OK)
 *   MUST leave original IP header of the offending packet, but we're not
 *     required to undo modifications made (OK)
 *  4.3.2.4 (Original Message Source Address)
 *   MUST use one of addresses for the interface the orig. packet arrived as
 *     source address (OK)
 *  4.3.2.5 (TOS and Precedence)
 *   SHOULD leave TOS set to the same value unless the packet would be 
 *     discarded for that reason (OK)
 *   MUST use TOS=0 if not possible to leave original value (OK)
 *   MUST leave IP Precedence for Source Quench messages (OK -- not sent 
 *	at all)
 *   SHOULD use IP Precedence = 6 (Internetwork Control) or 7 (Network Control)
 *     for all other error messages (OK, we use 6)
 *   MAY allow configuration of IP Precedence (OK -- not done)
 *   MUST leave IP Precedence and TOS for reply messages (OK)
 *  4.3.2.6 (Source Route)
 *   SHOULD use reverse source route UNLESS sending Parameter Problem on source
 *     routing and UNLESS the packet would be immediately discarded (NOT YET)
 *  4.3.2.7 (When Not to Send ICMP Errors)
 *   MUST NOT send ICMP errors in reply to:
 *     ICMP errors (OK)
 *     Packets failing IP header validation tests unless otherwise noted (OK)
 *     Broadcast/multicast datagrams (OK)
 *     MAC broadcasts (OK)
 *     Non-initial fragments (OK)
 *     Datagram with a source address that isn't a single host. (OK)
 *  4.3.2.8 (Rate Limiting)
 *   SHOULD be able to limit error message rate (OK)
 *   SHOULD allow setting of rate limits (OK, in the source)
 *  4.3.3.1 (Destination Unreachable)
 *   All the rules govern the IP layer, and are dealt with in ip.c, not here.
 *  4.3.3.2 (Redirect)
 *   MAY ignore ICMP Redirects if running a routing protocol or if forwarding
 *     is enabled on the interface (OK -- ignores)
 *  4.3.3.3 (Source Quench)
 *   SHOULD NOT originate SQ messages (OK)
 *   MUST be able to limit SQ rate if originates them (OK as we don't 
 *	send them)
 *   MAY ignore SQ messages it receives (OK -- we don't)
 *  4.3.3.4 (Time Exceeded)
 *   Requirements dealt with at IP (generating TIME_EXCEEDED).
 *  4.3.3.5 (Parameter Problem)
 *   MUST generate these for all errors not covered by other messages (OK)
 *   MUST include original value of the value pointed by (OK)
 *  4.3.3.6 (Echo Request)
 *   MUST implement echo server function (OK)
 *   MUST process at ER of at least max(576, MTU) (OK)
 *   MAY reject broadcast/multicast ER's (We don't, but that's OK)
 *   SHOULD have a config option for silently ignoring ER's (OK)
 *   MUST have a default value for the above switch = NO (OK)
 *   MUST have application layer interface for Echo Request/Reply (OK)
 *   MUST reply using same source address as the request was sent to.
 *     We're OK for unicast ECHOs, and it doesn't say anything about
 *     how to handle broadcast ones, since it's optional.
 *   MUST copy data from Request to Reply (OK)
 *   SHOULD update Record Route / Timestamp options (??)
 *   MUST use reversed Source Route for Reply if possible (NOT YET)
 *  4.3.3.7 (Information Request/Reply)
 *   SHOULD NOT originate or respond to these (OK)
 *  4.3.3.8 (Timestamp / Timestamp Reply)
 *   MAY implement (OK)
 *   MUST reply to every Timestamp message received (OK)
 *   MAY discard broadcast REQUESTs.  (OK, but see source for inconsistency)
 *   MUST reply using same source address as the request was sent to. (OK)
 *   MUST use reversed Source Route if possible (NOT YET)
 *   SHOULD update Record Route / Timestamp options (??)
 *   MUST pass REPLYs to transport/user layer (requires RAW, just like 
 *	ECHO) (OK)
 *   MUST update clock for timestamp at least 16 times/sec (OK)
 *   MUST be "correct within a few minutes" (OK)
 * 4.3.3.9 (Address Mask Request/Reply)
 *   MUST have support for receiving AMRq and responding with AMRe (OK, 
 *	but only as a compile-time option)
 *   SHOULD have option for each interface for AMRe's, MUST default to 
 *	NO (NOT YET)
 *   MUST NOT reply to AMRq before knows the correct AM (OK)
 *   MUST NOT respond to AMRq with source address 0.0.0.0 on physical
 *    	interfaces having multiple logical i-faces with different masks
 *	(NOT YET)
 *   SHOULD examine all AMRe's it receives and check them (NOT YET)
 *   SHOULD log invalid AMRe's (AM+sender) (NOT YET)
 *   MUST NOT use contents of AMRe to determine correct AM (OK)
 *   MAY broadcast AMRe's after having configured address masks (OK -- doesn't)
 *   MUST NOT do broadcast AMRe's if not set by extra option (OK, no option)
 *   MUST use the { <NetPrefix>, -1 } form of broadcast addresses (OK)
 * 4.3.3.10 (Router Advertisement and Solicitations)
 *   MUST support router part of Router Discovery Protocol on all networks we
 *     support broadcast or multicast addressing. (OK -- done by gated)
 *   MUST have all config parameters with the respective defaults (OK)
 * 5.2.7.1 (Destination Unreachable)
 *   MUST generate DU's (OK)
 *   SHOULD choose a best-match response code (OK)
 *   SHOULD NOT generate Host Isolated codes (OK)
 *   SHOULD use Communication Administratively Prohibited when administratively
 *     filtering packets (NOT YET -- bug-to-bug compatibility)
 *   MAY include config option for not generating the above and silently
 *	discard the packets instead (OK)
 *   MAY include config option for not generating Precedence Violation and
 *     Precedence Cutoff messages (OK as we don't generate them at all)
 *   MUST use Host Unreachable or Dest. Host Unknown codes whenever other hosts
 *     on the same network might be reachable (OK -- no net unreach's at all)
 *   MUST use new form of Fragmentation Needed and DF Set messages (OK)
 * 5.2.7.2 (Redirect)
 *   MUST NOT generate network redirects (OK)
 *   MUST be able to generate host redirects (OK)
 *   SHOULD be able to generate Host+TOS redirects (NO as we don't use TOS)
 *   MUST have an option to use Host redirects instead of Host+TOS ones (OK as
 *     no Host+TOS Redirects are used)
 *   MUST NOT generate redirects unless forwarding to the same i-face and the
 *     dest. address is on the same subnet as the src. address and no source
 *     routing is in use. (OK)
 *   MUST NOT follow redirects when using a routing protocol (OK)
 *   MAY use redirects if not using a routing protocol (OK, compile-time option)
 *   MUST comply to Host Requirements when not acting as a router (OK)
 *  5.2.7.3 (Time Exceeded)
 *   MUST generate Time Exceeded Code 0 when discarding packet due to TTL=0 (OK)
 *   MAY have a per-interface option to disable origination of TE messages, but
 *     it MUST default to "originate" (OK -- we don't support it)
 */

#include <linux/config.h>
#include <linux/types.h>
#include <linux/sched.h>
#include <linux/kernel.h>
#include <linux/fcntl.h>
#include <linux/socket.h>
#include <linux/in.h>
#include <linux/inet.h>
#include <linux/netdevice.h>
#include <linux/string.h>
#include <net/snmp.h>
#include <net/ip.h>
#include <net/route.h>
#include <net/protocol.h>
#include <net/icmp.h>
#include <net/tcp.h>
#include <net/udp.h>
#include <net/raw.h>
#include <net/snmp.h>
#include <linux/skbuff.h>
#include <net/sock.h>
#include <linux/errno.h>
#include <linux/timer.h>
#include <linux/init.h>
#include <asm/system.h>
#include <asm/uaccess.h>
#include <net/checksum.h>

#ifdef CONFIG_IP_MASQUERADE
#include <net/ip_masq.h>
#endif

#define min(a,b)	((a)<(b)?(a):(b))

/*
 *	Statistics
 */
 
struct icmp_mib icmp_statistics;

/* An array of errno for error messages from dest unreach. */
/* RFC 1122: 3.2.2.1 States that NET_UNREACH, HOS_UNREACH and SR_FAIELD MUST be considered 'transient errs'. */

struct icmp_err icmp_err_convert[] = {
  { ENETUNREACH,	0 },	/*	ICMP_NET_UNREACH	*/
  { EHOSTUNREACH,	0 },	/*	ICMP_HOST_UNREACH	*/
  { ENOPROTOOPT,	1 },	/*	ICMP_PROT_UNREACH	*/
  { ECONNREFUSED,	1 },	/*	ICMP_PORT_UNREACH	*/
  { EMSGSIZE,		0 },	/*	ICMP_FRAG_NEEDED	*/
  { EOPNOTSUPP,		0 },	/*	ICMP_SR_FAILED		*/
  { ENETUNREACH,	1 },	/* 	ICMP_NET_UNKNOWN	*/
  { EHOSTDOWN,		1 },	/*	ICMP_HOST_UNKNOWN	*/
  { ENONET,		1 },	/*	ICMP_HOST_ISOLATED	*/
  { ENETUNREACH,	1 },	/*	ICMP_NET_ANO		*/
  { EHOSTUNREACH,	1 },	/*	ICMP_HOST_ANO		*/
  { ENETUNREACH,	0 },	/*	ICMP_NET_UNR_TOS	*/
  { EHOSTUNREACH,	0 },	/*	ICMP_HOST_UNR_TOS	*/
  { EHOSTUNREACH,	1 },	/*	ICMP_PKT_FILTERED	*/
  { EHOSTUNREACH,	1 },	/*	ICMP_PREC_VIOLATION	*/
  { EHOSTUNREACH,	1 }	/*	ICMP_PREC_CUTOFF	*/
};

/* Control parameters for ECHO relies. */
int sysctl_icmp_echo_ignore_all = 0;
int sysctl_icmp_echo_ignore_broadcasts = 0;

/* Control parameter - ignore bogus broadcast responses? */
int sysctl_icmp_ignore_bogus_error_responses =0;

extern int sysctl_ip_always_defrag;

/*
 *	ICMP control array. This specifies what to do with each ICMP.
 */

struct icmp_control
{
	unsigned long *output;		/* Address to increment on output */
	unsigned long *input;		/* Address to increment on input */
	void (*handler)(struct icmphdr *icmph, struct sk_buff *skb, int len);
	short	error;		/* This ICMP is classed as an error message */
	int *timeout; /* Rate limit */
};

static struct icmp_control icmp_pointers[NR_ICMP_TYPES+1];

/*
 *	Build xmit assembly blocks
 */

struct icmp_bxm
{
	void *data_ptr;
	int data_len;
	struct icmphdr icmph;
	unsigned long csum;
	struct ip_options replyopts;
	unsigned char  optbuf[40];
};

/*
 *	The ICMP socket. This is the most convenient way to flow control
 *	our ICMP output as well as maintain a clean interface throughout
 *	all layers. All Socketless IP sends will soon be gone.
 */
	
struct inode icmp_inode;
struct socket *icmp_socket=&icmp_inode.u.socket_i;

/*
 *	Send an ICMP frame.
 */

/*
 *	Check transmit rate limitation for given message.
 *	The rate information is held in the destination cache now.
 *	This function is generic and could be used for other purposes
 *	too. It uses a Token bucket filter as suggested by Alexey Kuznetsov.
 *
 *	Note that the same dst_entry fields are modified by functions in 
 *	route.c too, but these work for packet destinations while xrlim_allow
 *	works for icmp destinations. This means the rate limiting information
 *	for one "ip object" is shared.
 *
 *	Note that the same dst_entry fields are modified by functions in 
 *	route.c too, but these work for packet destinations while xrlim_allow
 *	works for icmp destinations. This means the rate limiting information
 *	for one "ip object" is shared - and these ICMPs are twice limited:
 *	by source and by destination.
 *
 *	RFC 1812: 4.3.2.8 SHOULD be able to limit error message rate
 *			  SHOULD allow setting of rate limits 
 *
 * 	Shared between ICMPv4 and ICMPv6.
 */
#define XRLIM_BURST_FACTOR 6
int xrlim_allow(struct dst_entry *dst, int timeout)
{
	unsigned long now;

	now = jiffies;
	dst->rate_tokens += now - dst->rate_last;
	dst->rate_last = now;
	if (dst->rate_tokens > XRLIM_BURST_FACTOR*timeout)
		dst->rate_tokens = XRLIM_BURST_FACTOR*timeout;
	if (dst->rate_tokens >= timeout) {
		dst->rate_tokens -= timeout;
		return 1;
	}
	return 0; 
}

static inline int icmpv4_xrlim_allow(struct rtable *rt, int type, int code)
{
	struct dst_entry *dst = &rt->u.dst; 

	if (type > NR_ICMP_TYPES || !icmp_pointers[type].timeout)
		return 1;

	/* Don't limit PMTU discovery. */
	if (type == ICMP_DEST_UNREACH && code == ICMP_FRAG_NEEDED)
		return 1;

	/* Redirect has its own rate limit mechanism */
	if (type == ICMP_REDIRECT)
		return 1;

	/* No rate limit on loopback */
	if (dst->dev && (dst->dev->flags&IFF_LOOPBACK))
 		return 1;

	return xrlim_allow(dst, *(icmp_pointers[type].timeout));
}

/*
 *	Maintain the counters used in the SNMP statistics for outgoing ICMP
 */
 
static void icmp_out_count(int type)
{
	if (type>NR_ICMP_TYPES)
		return;
	(*icmp_pointers[type].output)++;
	icmp_statistics.IcmpOutMsgs++;
}
 
/*
 *	Checksum each fragment, and on the first include the headers and final checksum.
 */
 
static int icmp_glue_bits(const void *p, char *to, unsigned int offset, unsigned int fraglen)
{
	struct icmp_bxm *icmp_param = (struct icmp_bxm *)p;
	struct icmphdr *icmph;
	unsigned long csum;

	if (offset) {
		icmp_param->csum=csum_partial_copy(icmp_param->data_ptr+offset-sizeof(struct icmphdr), 
				to, fraglen,icmp_param->csum);
		return 0;
	}

	/*
	 *	First fragment includes header. Note that we've done
	 *	the other fragments first, so that we get the checksum
	 *	for the whole packet here.
	 */
	csum = csum_partial_copy((void *)&icmp_param->icmph,
		to, sizeof(struct icmphdr), 
		icmp_param->csum);
	csum = csum_partial_copy(icmp_param->data_ptr,
		to+sizeof(struct icmphdr),
		fraglen-sizeof(struct icmphdr), csum);
	icmph=(struct icmphdr *)to;
	icmph->checksum = csum_fold(csum);
	return 0;
}
 
/*
 *	Driving logic for building and sending ICMP messages.
 */

static void icmp_reply(struct icmp_bxm *icmp_param, struct sk_buff *skb)
{
	struct sock *sk=icmp_socket->sk;
	struct ipcm_cookie ipc;
	struct rtable *rt = (struct rtable*)skb->dst;
	u32 daddr;

	if (ip_options_echo(&icmp_param->replyopts, skb))
		return;

	icmp_param->icmph.checksum=0;
	icmp_param->csum=0;
	icmp_out_count(icmp_param->icmph.type);

	sk->ip_tos = skb->nh.iph->tos;
	daddr = ipc.addr = rt->rt_src;
	ipc.opt = &icmp_param->replyopts;
	if (ipc.opt->srr)
		daddr = icmp_param->replyopts.faddr;
	if (ip_route_output(&rt, daddr, rt->rt_spec_dst, RT_TOS(skb->nh.iph->tos), 0))
		return;
	ip_build_xmit(sk, icmp_glue_bits, icmp_param, 
		icmp_param->data_len+sizeof(struct icmphdr),
		&ipc, rt, MSG_DONTWAIT);
	ip_rt_put(rt);
}


/*
 *	Send an ICMP message in response to a situation
 *
 *	RFC 1122: 3.2.2	MUST send at least the IP header and 8 bytes of header. MAY send more (we do).
 *			MUST NOT change this header information.
 *			MUST NOT reply to a multicast/broadcast IP address.
 *			MUST NOT reply to a multicast/broadcast MAC address.
 *			MUST reply to only the first fragment.
 */

void icmp_send(struct sk_buff *skb_in, int type, int code, unsigned long info)
{
	struct iphdr *iph;
	struct icmphdr *icmph;
	int room;
	struct icmp_bxm icmp_param;
	struct rtable *rt = (struct rtable*)skb_in->dst;
	struct ipcm_cookie ipc;
	u32 saddr;
	u8  tos;
	
	/*
	 *	Find the original header
	 */
	 
	iph = skb_in->nh.iph;
	
	/*
	 *	No replies to physical multicast/broadcast
	 */
	 
	if (skb_in->pkt_type!=PACKET_HOST)
		return;
		
	/*
	 *	Now check at the protocol level
	 */
	if (!rt) {
                if (sysctl_ip_always_defrag == 0 &&
                    net_ratelimit())
			printk(KERN_DEBUG "icmp_send: destinationless packet\n");
		return;
	}
	if (rt->rt_flags&(RTCF_BROADCAST|RTCF_MULTICAST))
		return;
	 
		
	/*
	 *	Only reply to fragment 0. We byte re-order the constant
	 *	mask for efficiency.
	 */
	 
	if (iph->frag_off&htons(IP_OFFSET))
		return;
		
	/* 
	 *	If we send an ICMP error to an ICMP error a mess would result..
	 */
	 
	if (icmp_pointers[type].error) {
		/*
		 *	We are an error, check if we are replying to an ICMP error
		 */
		 
		if (iph->protocol==IPPROTO_ICMP) {
			icmph = (struct icmphdr *)((char *)iph + (iph->ihl<<2));
			/*
			 *	Assume any unknown ICMP type is an error. This isn't
			 *	specified by the RFC, but think about it..
			 */
			if (icmph->type>NR_ICMP_TYPES || icmp_pointers[icmph->type].error)
				return;
		}
	}


	/*
	 *	Construct source address and options.
	 */

#ifdef CONFIG_IP_ROUTE_NAT	
	/*
	 *	Restore original addresses if packet has been translated.
	 */
	if (rt->rt_flags&RTCF_NAT && IPCB(skb_in)->flags&IPSKB_TRANSLATED) {
		iph->daddr = rt->key.dst;
		iph->saddr = rt->key.src;
	}
#endif
#ifdef CONFIG_IP_MASQUERADE
	if (type==ICMP_DEST_UNREACH && IPCB(skb_in)->flags&IPSKB_MASQUERADED) {
			ip_fw_unmasq_icmp(skb_in);
	}
#endif

	saddr = iph->daddr;
	if (!(rt->rt_flags & RTCF_LOCAL))
		saddr = 0;

	tos = icmp_pointers[type].error ?
		((iph->tos & IPTOS_TOS_MASK) | IPTOS_PREC_INTERNETCONTROL) :
			iph->tos;

	/* XXX: use a more aggressive expire for routes created by 
	 * this call (not longer than the rate limit timeout). 
	 * It could be also worthwhile to not put them into ipv4
	 * fast routing cache at first. Otherwise an attacker can
	 * grow the routing table.
	 */
	if (ip_route_output(&rt, iph->saddr, saddr, RT_TOS(tos), 0))
		return;
	
	if (ip_options_echo(&icmp_param.replyopts, skb_in)) 
		goto ende;


	/*
	 *	Prepare data for ICMP header.
	 */

	icmp_param.icmph.type=type;
	icmp_param.icmph.code=code;
	icmp_param.icmph.un.gateway = info;
	icmp_param.icmph.checksum=0;
	icmp_param.csum=0;
	icmp_param.data_ptr=iph;
	icmp_out_count(icmp_param.icmph.type);
	icmp_socket->sk->ip_tos = tos;
	ipc.addr = iph->saddr;
	ipc.opt = &icmp_param.replyopts;
	if (icmp_param.replyopts.srr) {
		ip_rt_put(rt);
		if (ip_route_output(&rt, icmp_param.replyopts.faddr, saddr, RT_TOS(tos), 0))
			return;
	}

	if (!icmpv4_xrlim_allow(rt, type, code))
		goto ende;

	/* RFC says return as much as we can without exceeding 576 bytes. */

	room = rt->u.dst.pmtu;
	if (room > 576)
		room = 576;
	room -= sizeof(struct iphdr) + icmp_param.replyopts.optlen;
	room -= sizeof(struct icmphdr);

	icmp_param.data_len=(iph->ihl<<2)+skb_in->len;
	if (icmp_param.data_len > room)
		icmp_param.data_len = room;
	
	ip_build_xmit(icmp_socket->sk, icmp_glue_bits, &icmp_param, 
		icmp_param.data_len+sizeof(struct icmphdr),
		&ipc, rt, MSG_DONTWAIT);

ende:
	ip_rt_put(rt);
}


/* 
 *	Handle ICMP_DEST_UNREACH, ICMP_TIME_EXCEED, and ICMP_QUENCH. 
 */

static void icmp_unreach(struct icmphdr *icmph, struct sk_buff *skb, int len)
{
	struct iphdr *iph;
	int hash;
	struct inet_protocol *ipprot;
	unsigned char *dp;
	struct sock *raw_sk;
	
	/*
	 *	Incomplete header ?
	 * 	Only checks for the IP header, there should be an
	 *	additional check for longer headers in upper levels.
	 */

	if(len<sizeof(struct iphdr)) {
		icmp_statistics.IcmpInErrors++;
		return;
	}
		
	iph = (struct iphdr *) (icmph + 1);
	dp = (unsigned char*)iph;
	
	if(icmph->type==ICMP_DEST_UNREACH) {
		switch(icmph->code & 15) {
			case ICMP_NET_UNREACH:
				break;
			case ICMP_HOST_UNREACH:
				break;
			case ICMP_PROT_UNREACH:
				break;
			case ICMP_PORT_UNREACH:
				break;
			case ICMP_FRAG_NEEDED:
				if (ipv4_config.no_pmtu_disc) {
					if (sysctl_ip_always_defrag == 0 && net_ratelimit())
						printk(KERN_INFO "ICMP: %d.%d.%d.%d: fragmentation needed and DF set.\n",
					       NIPQUAD(iph->daddr));
				} else {
					unsigned short new_mtu;
					new_mtu = ip_rt_frag_needed(iph, ntohs(icmph->un.frag.mtu));
					if (!new_mtu) 
						return;
					icmph->un.frag.mtu = htons(new_mtu);
				}
				break;
			case ICMP_SR_FAILED:
				if (sysctl_ip_always_defrag == 0 && net_ratelimit())
					printk(KERN_INFO "ICMP: %d.%d.%d.%d: Source Route Failed.\n", NIPQUAD(iph->daddr));
				break;
			default:
				break;
		}
		if (icmph->code>NR_ICMP_UNREACH) 
			return;
	}
	
	/*
	 *	Throw it at our lower layers
	 *
	 *	RFC 1122: 3.2.2 MUST extract the protocol ID from the passed header.
	 *	RFC 1122: 3.2.2.1 MUST pass ICMP unreach messages to the transport layer.
	 *	RFC 1122: 3.2.2.2 MUST pass ICMP time expired messages to transport layer.
	 */
	 
	/*
	 *	Check the other end isn't violating RFC 1122. Some routers send
	 *	bogus responses to broadcast frames. If you see this message
	 *	first check your netmask matches at both ends, if it does then
	 *	get the other vendor to fix their kit.
	 */

	if (!sysctl_icmp_ignore_bogus_error_responses)
	{
	
		if (inet_addr_type(iph->daddr) == RTN_BROADCAST)
		{
			if (net_ratelimit())
				printk(KERN_WARNING "%d.%d.%d.%d sent an invalid ICMP error to a broadcast.\n",
			       	NIPQUAD(skb->nh.iph->saddr));
			return; 
		}
	}

	/*
	 *	Deliver ICMP message to raw sockets. Pretty useless feature?
	 */

	/* Note: See raw.c and net/raw.h, RAWV4_HTABLE_SIZE==MAX_INET_PROTOS */
	hash = iph->protocol & (MAX_INET_PROTOS - 1);
	if ((raw_sk = raw_v4_htable[hash]) != NULL) 
	{
		while ((raw_sk = raw_v4_lookup(raw_sk, iph->protocol, iph->saddr,
					       iph->daddr, skb->dev->ifindex)) != NULL) {
			raw_err(raw_sk, skb);
			raw_sk = raw_sk->next;
		}
	}

	/*
	 *	This can't change while we are doing it. 
	 */

	ipprot = (struct inet_protocol *) inet_protos[hash];
	while(ipprot != NULL) {
		struct inet_protocol *nextip;

		nextip = (struct inet_protocol *) ipprot->next;
	
		/* 
		 *	Pass it off to everyone who wants it. 
		 */

		/* RFC1122: OK. Passes appropriate ICMP errors to the */
		/* appropriate protocol layer (MUST), as per 3.2.2. */

		if (iph->protocol == ipprot->protocol && ipprot->err_handler)
 			ipprot->err_handler(skb, dp, len);

		ipprot = nextip;
  	}
}


/*
 *	Handle ICMP_REDIRECT. 
 */

static void icmp_redirect(struct icmphdr *icmph, struct sk_buff *skb, int len)
{
	struct iphdr *iph;
	unsigned long ip;

	if (len < sizeof(struct iphdr)) {
		icmp_statistics.IcmpInErrors++;
		return; 
	}
		
	/*
	 *	Get the copied header of the packet that caused the redirect
	 */
	 
	iph = (struct iphdr *) (icmph + 1);
	ip = iph->daddr;

	switch(icmph->code & 7) {
		case ICMP_REDIR_NET:
		case ICMP_REDIR_NETTOS:
			/*
			 *	As per RFC recommendations now handle it as
			 *	a host redirect.
			 */
			 
		case ICMP_REDIR_HOST:
		case ICMP_REDIR_HOSTTOS:
			ip_rt_redirect(skb->nh.iph->saddr, ip, icmph->un.gateway, iph->saddr, iph->tos, skb->dev);
			break;
		default:
			break;
  	}
}

/*
 *	Handle ICMP_ECHO ("ping") requests. 
 *
 *	RFC 1122: 3.2.2.6 MUST have an echo server that answers ICMP echo requests.
 *	RFC 1122: 3.2.2.6 Data received in the ICMP_ECHO request MUST be included in the reply.
 *	RFC 1812: 4.3.3.6 SHOULD have a config option for silently ignoring echo requests, MUST have default=NOT.
 *	See also WRT handling of options once they are done and working.
 */

static void icmp_echo(struct icmphdr *icmph, struct sk_buff *skb, int len)
{
	if (!sysctl_icmp_echo_ignore_all) {
		struct icmp_bxm icmp_param;

		icmp_param.icmph=*icmph;
		icmp_param.icmph.type=ICMP_ECHOREPLY;
		icmp_param.data_ptr=(icmph+1);
		icmp_param.data_len=len;
		icmp_reply(&icmp_param, skb);
	}
}

/*
 *	Handle ICMP Timestamp requests. 
 *	RFC 1122: 3.2.2.8 MAY implement ICMP timestamp requests.
 *		  SHOULD be in the kernel for minimum random latency.
 *		  MUST be accurate to a few minutes.
 *		  MUST be updated at least at 15Hz.
 */
 
static void icmp_timestamp(struct icmphdr *icmph, struct sk_buff *skb, int len)
{
	struct timeval tv;
	__u32 times[3];		/* So the new timestamp works on ALPHA's.. */
	struct icmp_bxm icmp_param;
	
	/*
	 *	Too short.
	 */
	 
	if(len<12) {
		icmp_statistics.IcmpInErrors++;
		return;
	}
	
	/*
	 *	Fill in the current time as ms since midnight UT: 
	 */
	 
	do_gettimeofday(&tv);
	times[1] = htonl((tv.tv_sec % 86400) * 1000 + tv.tv_usec / 1000);
	times[2] = times[1];
	memcpy((void *)&times[0], icmph+1, 4);		/* Incoming stamp */
	icmp_param.icmph=*icmph;
	icmp_param.icmph.type=ICMP_TIMESTAMPREPLY;
	icmp_param.icmph.code=0;
	icmp_param.data_ptr=&times;
	icmp_param.data_len=12;
	icmp_reply(&icmp_param, skb);
}


/* 
 *	Handle ICMP_ADDRESS_MASK requests.  (RFC950)
 *
 * RFC1122 (3.2.2.9).  A host MUST only send replies to 
 * ADDRESS_MASK requests if it's been configured as an address mask 
 * agent.  Receiving a request doesn't constitute implicit permission to 
 * act as one. Of course, implementing this correctly requires (SHOULD) 
 * a way to turn the functionality on and off.  Another one for sysctl(), 
 * I guess. -- MS
 *
 * RFC1812 (4.3.3.9).	A router MUST implement it.
 *			A router SHOULD have switch turning it on/off.
 *		      	This switch MUST be ON by default.
 *
 * Gratuitous replies, zero-source replies are not implemented,
 * that complies with RFC. DO NOT implement them!!! All the idea
 * of broadcast addrmask replies as specified in RFC950 is broken.
 * The problem is that it is not uncommon to have several prefixes
 * on one physical interface. Moreover, addrmask agent can even be
 * not aware of existing another prefixes.
 * If source is zero, addrmask agent cannot choose correct prefix.
 * Gratuitous mask announcements suffer from the same problem.
 * RFC1812 explains it, but still allows to use ADDRMASK,
 * that is pretty silly. --ANK
 *
 * All these rules are so bizarre, that I removed kernel addrmask
 * support at all. It is wrong, it is obsolete, nobody uses it in
 * any case. --ANK
 *
 * Furthermore you can do it with a usermode address agent program
 * anyway...
 */

static void icmp_address(struct icmphdr *icmph, struct sk_buff *skb, int len)
{
#if 0
	if (sysctl_ip_always_defrag == 0 && net_ratelimit())
		printk(KERN_DEBUG "a guy asks for address mask. Who is it?\n");
#endif		
}

/*
 * RFC1812 (4.3.3.9).	A router SHOULD listen all replies, and complain
 *			loudly if an inconsistency is found.
 */

static void icmp_address_reply(struct icmphdr *icmph, struct sk_buff *skb, int len)
{
	struct rtable *rt = (struct rtable*)skb->dst;
	struct device *dev = skb->dev;
	struct in_device *in_dev = dev->ip_ptr;
	struct in_ifaddr *ifa;
	u32 mask;

	if (!in_dev || !in_dev->ifa_list ||
	    !IN_DEV_LOG_MARTIANS(in_dev) ||
	    !IN_DEV_FORWARD(in_dev) ||
	    len < 4 ||
	    !(rt->rt_flags&RTCF_DIRECTSRC))
		return;

	mask = *(u32*)&icmph[1];
	for (ifa=in_dev->ifa_list; ifa; ifa = ifa->ifa_next) {
		if (mask == ifa->ifa_mask && inet_ifa_match(rt->rt_src, ifa))
			return;
	}
	if (sysctl_ip_always_defrag == 0 && net_ratelimit())
		printk(KERN_INFO "Wrong address mask %08X from %08X/%s\n",
		       ntohl(mask), ntohl(rt->rt_src), dev->name);
}

static void icmp_discard(struct icmphdr *icmph, struct sk_buff *skb, int len)
{
}

#ifdef CONFIG_IP_TRANSPARENT_PROXY
/*
 *	Check incoming icmp packets not addressed locally, to check whether
 *	they relate to a (proxying) socket on our system.
 *	Needed for transparent proxying.
 *
 *	This code is presently ugly and needs cleanup.
 *	Probably should add a chkaddr entry to ipprot to call a chk routine
 *	in udp.c or tcp.c...
 */

/* This should work with the new hashes now. -DaveM */
extern struct sock *tcp_v4_lookup(u32 saddr, u16 sport, u32 daddr, u16 dport, int dif);
extern struct sock *udp_v4_lookup(u32 saddr, u16 sport, u32 daddr, u16 dport, int dif);

int icmp_chkaddr(struct sk_buff *skb)
{
	struct icmphdr *icmph=(struct icmphdr *)(skb->nh.raw + skb->nh.iph->ihl*4);
	struct iphdr *iph = (struct iphdr *) (icmph + 1);
	void (*handler)(struct icmphdr *icmph, struct sk_buff *skb, int len) = icmp_pointers[icmph->type].handler;

	if (handler == icmp_unreach || handler == icmp_redirect) {
		struct sock *sk;

		switch (iph->protocol) {
		case IPPROTO_TCP:
			{
			struct tcphdr *th = (struct tcphdr *)(((unsigned char *)iph)+(iph->ihl<<2));

			sk = tcp_v4_lookup(iph->daddr, th->dest, iph->saddr, th->source, skb->dev->ifindex);
			if (!sk || (sk->state == TCP_LISTEN))
				return 0;
			/*
			 * This packet came from us.
			 */
			return 1;
			}
		case IPPROTO_UDP:
			{
			struct udphdr *uh = (struct udphdr *)(((unsigned char *)iph)+(iph->ihl<<2));

			sk = udp_v4_lookup(iph->daddr, uh->dest, iph->saddr, uh->source, skb->dev->ifindex);
			if (!sk) return 0;
			if (sk->saddr != iph->saddr && inet_addr_type(iph->saddr) != RTN_LOCAL)
				return 0;
			/*
			 * This packet may have come from us.
			 * Assume it did.
			 */
			return 1;
			}
		}
	}
	return 0;
}

#endif

/* 
 *	Deal with incoming ICMP packets.
 */
 
int icmp_rcv(struct sk_buff *skb, unsigned short len)
{
	struct icmphdr *icmph = skb->h.icmph;
	struct rtable *rt = (struct rtable*)skb->dst;

	icmp_statistics.IcmpInMsgs++;

	/*
	 *	18 is the highest 'known' ICMP type. Anything else is a mystery
	 *
	 *	RFC 1122: 3.2.2  Unknown ICMP messages types MUST be silently discarded.
	 */
	if(len < sizeof(struct icmphdr) ||
	   ip_compute_csum((unsigned char *) icmph, len) ||
	   icmph->type > NR_ICMP_TYPES)
		goto error;
	 
	/*
	 *	Parse the ICMP message 
	 */

 	if (rt->rt_flags&(RTCF_BROADCAST|RTCF_MULTICAST)) {
		/*
		 *	RFC 1122: 3.2.2.6 An ICMP_ECHO to broadcast MAY be
		 *	  silently ignored (we let user decide with a sysctl).
		 *	RFC 1122: 3.2.2.8 An ICMP_TIMESTAMP MAY be silently
		 *	  discarded if to broadcast/multicast.
		 */
		if (icmph->type == ICMP_ECHO &&
		    sysctl_icmp_echo_ignore_broadcasts) {
			goto error;
		}
		if (icmph->type != ICMP_ECHO &&
		    icmph->type != ICMP_TIMESTAMP &&
		    icmph->type != ICMP_ADDRESS &&
		    icmph->type != ICMP_ADDRESSREPLY) {
			goto error;
  		}
	}

	len -= sizeof(struct icmphdr);
	(*icmp_pointers[icmph->type].input)++;
	(icmp_pointers[icmph->type].handler)(icmph, skb, len);

drop:
	kfree_skb(skb);
	return 0;
error:
	icmp_statistics.IcmpInErrors++;
	goto drop;
}

/*
 *	A spare long used to speed up statistics updating
 */
 
static unsigned long dummy;

/* 
 * 	Configurable rate limits.
 *	Someone should check if these default values are correct.
 *	Note that these values interact with the routing cache GC timeout.
 *	If you chose them too high they won't take effect, because the
 *	dst_entry gets expired too early. The same should happen when
 *	the cache grows too big.
 */
int sysctl_icmp_destunreach_time = 1*HZ;
int sysctl_icmp_timeexceed_time = 1*HZ;
int sysctl_icmp_paramprob_time = 1*HZ;
int sysctl_icmp_echoreply_time = 0; /* don't limit it per default. */

/*
 *	This table is the definition of how we handle ICMP.
 */
 
static struct icmp_control icmp_pointers[NR_ICMP_TYPES+1] = {
/* ECHO REPLY (0) */
 { &icmp_statistics.IcmpOutEchoReps, &icmp_statistics.IcmpInEchoReps, icmp_discard, 0, &sysctl_icmp_echoreply_time},
 { &dummy, &icmp_statistics.IcmpInErrors, icmp_discard, 1, },
 { &dummy, &icmp_statistics.IcmpInErrors, icmp_discard, 1, },
/* DEST UNREACH (3) */
 { &icmp_statistics.IcmpOutDestUnreachs, &icmp_statistics.IcmpInDestUnreachs, icmp_unreach, 1, &sysctl_icmp_destunreach_time },
/* SOURCE QUENCH (4) */
 { &icmp_statistics.IcmpOutSrcQuenchs, &icmp_statistics.IcmpInSrcQuenchs, icmp_unreach, 1, },
/* REDIRECT (5) */
 { &icmp_statistics.IcmpOutRedirects, &icmp_statistics.IcmpInRedirects, icmp_redirect, 1, },
 { &dummy, &icmp_statistics.IcmpInErrors, icmp_discard, 1, },
 { &dummy, &icmp_statistics.IcmpInErrors, icmp_discard, 1, },
/* ECHO (8) */
 { &icmp_statistics.IcmpOutEchos, &icmp_statistics.IcmpInEchos, icmp_echo, 0, },
 { &dummy, &icmp_statistics.IcmpInErrors, icmp_discard, 1, },
 { &dummy, &icmp_statistics.IcmpInErrors, icmp_discard, 1, },
/* TIME EXCEEDED (11) */
 { &icmp_statistics.IcmpOutTimeExcds, &icmp_statistics.IcmpInTimeExcds, icmp_unreach, 1, &sysctl_icmp_timeexceed_time },
/* PARAMETER PROBLEM (12) */
 { &icmp_statistics.IcmpOutParmProbs, &icmp_statistics.IcmpInParmProbs, icmp_unreach, 1, &sysctl_icmp_paramprob_time },
/* TIMESTAMP (13) */
 { &icmp_statistics.IcmpOutTimestamps, &icmp_statistics.IcmpInTimestamps, icmp_timestamp, 0,  },
/* TIMESTAMP REPLY (14) */
 { &icmp_statistics.IcmpOutTimestampReps, &icmp_statistics.IcmpInTimestampReps, icmp_discard, 0, },
/* INFO (15) */
 { &dummy, &dummy, icmp_discard, 0, },
/* INFO REPLY (16) */
 { &dummy, &dummy, icmp_discard, 0, },
/* ADDR MASK (17) */
 { &icmp_statistics.IcmpOutAddrMasks, &icmp_statistics.IcmpInAddrMasks, icmp_address, 0,  },
/* ADDR MASK REPLY (18) */
 { &icmp_statistics.IcmpOutAddrMaskReps, &icmp_statistics.IcmpInAddrMaskReps, icmp_address_reply, 0, }
};

__initfunc(void icmp_init(struct net_proto_family *ops))
{
	int err;

	icmp_inode.i_mode = S_IFSOCK;
	icmp_inode.i_sock = 1;
	icmp_inode.i_uid = 0;
	icmp_inode.i_gid = 0;

	icmp_socket->inode = &icmp_inode;
	icmp_socket->state = SS_UNCONNECTED;
	icmp_socket->type=SOCK_RAW;

	if ((err=ops->create(icmp_socket, IPPROTO_ICMP))<0)
		panic("Failed to create the ICMP control socket.\n");
	icmp_socket->sk->allocation=GFP_ATOMIC;
	icmp_socket->sk->num = 256;		/* Don't receive any data */
	icmp_socket->sk->ip_ttl = MAXTTL;
}