summaryrefslogtreecommitdiff
path: root/sys/altq/altq_subr.c
blob: 1571bd625907c61fa7711005962185e7925f70b8 (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
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
1375
1376
1377
1378
1379
1380
1381
1382
1383
1384
1385
1386
1387
1388
1389
1390
1391
1392
1393
1394
1395
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
1410
1411
1412
1413
1414
1415
1416
1417
1418
1419
1420
1421
1422
1423
1424
1425
1426
1427
1428
1429
1430
1431
1432
1433
1434
1435
1436
1437
1438
1439
1440
1441
1442
1443
1444
1445
1446
1447
1448
1449
1450
1451
1452
1453
1454
1455
1456
1457
1458
1459
1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
1484
1485
1486
1487
1488
1489
1490
1491
1492
1493
1494
1495
1496
1497
1498
1499
1500
1501
1502
1503
1504
1505
1506
1507
1508
1509
1510
1511
1512
1513
1514
1515
1516
1517
1518
1519
1520
1521
1522
1523
1524
1525
1526
1527
1528
1529
1530
1531
1532
1533
1534
1535
1536
1537
1538
1539
1540
1541
1542
1543
1544
1545
1546
1547
1548
1549
1550
1551
1552
1553
1554
1555
1556
1557
1558
1559
1560
1561
1562
1563
1564
1565
1566
1567
1568
1569
1570
1571
1572
1573
1574
1575
1576
1577
1578
1579
1580
1581
1582
1583
1584
1585
1586
1587
1588
1589
1590
1591
/*	$OpenBSD: altq_subr.c,v 1.7 2002/07/25 20:42:53 deraadt Exp $	*/
/*	$KAME: altq_subr.c,v 1.11 2002/01/11 08:11:49 kjc Exp $	*/

/*
 * Copyright (C) 1997-2002
 *	Sony Computer Science Laboratories Inc.  All rights reserved.
 *
 * Redistribution and use in source and binary forms, with or without
 * modification, are permitted provided that the following conditions
 * are met:
 * 1. Redistributions of source code must retain the above copyright
 *    notice, this list of conditions and the following disclaimer.
 * 2. Redistributions in binary form must reproduce the above copyright
 *    notice, this list of conditions and the following disclaimer in the
 *    documentation and/or other materials provided with the distribution.
 *
 * THIS SOFTWARE IS PROVIDED BY SONY CSL AND CONTRIBUTORS ``AS IS'' AND
 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
 * ARE DISCLAIMED.  IN NO EVENT SHALL SONY CSL OR CONTRIBUTORS BE LIABLE
 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
 * SUCH DAMAGE.
 */

#include <sys/param.h>
#include <sys/malloc.h>
#include <sys/mbuf.h>
#include <sys/systm.h>
#include <sys/proc.h>
#include <sys/socket.h>
#include <sys/socketvar.h>
#include <sys/kernel.h>
#include <sys/errno.h>
#include <sys/syslog.h>
#include <uvm/uvm_extern.h>
#include <sys/sysctl.h>
#include <sys/queue.h>

#include <net/if.h>
#include <net/if_dl.h>
#include <net/if_types.h>

#include <netinet/in.h>
#include <netinet/in_systm.h>
#include <netinet/ip.h>
#ifdef INET6
#include <netinet/ip6.h>
#endif
#include <netinet/tcp.h>
#include <netinet/udp.h>

#include <altq/altq.h>
#include <altq/altq_conf.h>

/* machine dependent clock related includes */
#ifdef __FreeBSD__
#include "opt_cpu.h"	/* for FreeBSD-2.2.8 to get i586_ctr_freq */
#include <machine/clock.h>
#endif
#if defined(__i386__)
#include <machine/specialreg.h>		/* for CPUID_TSC */
#ifdef __FreeBSD__
#include <machine/md_var.h>		/* for cpu_feature */
#elif defined(__NetBSD__) || defined(__OpenBSD__)
#include <machine/cpu.h>		/* for cpu_feature */
#endif
#endif /* __i386__ */

/*
 * internal function prototypes
 */
static void	tbr_timeout(void *);
static int 	extract_ports4(struct mbuf *, struct ip *,
				    struct flowinfo_in *);
#ifdef INET6
static int 	extract_ports6(struct mbuf *, struct ip6_hdr *,
				    struct flowinfo_in6 *);
#endif
static int	apply_filter4(u_int32_t, struct flow_filter *,
				   struct flowinfo_in *);
static int	apply_ppfilter4(u_int32_t, struct flow_filter *,
				     struct flowinfo_in *);
#ifdef INET6
static int	apply_filter6(u_int32_t, struct flow_filter6 *,
					   struct flowinfo_in6 *);
#endif
static int	apply_tosfilter4(u_int32_t, struct flow_filter *,
					     struct flowinfo_in *);
static u_long	get_filt_handle(struct acc_classifier *, int);
static struct acc_filter *filth_to_filtp(struct acc_classifier *,
					      u_long);
static u_int32_t filt2fibmask(struct flow_filter *);

static void 	ip4f_cache(struct ip *, struct flowinfo_in *);
static int 	ip4f_lookup(struct ip *, struct flowinfo_in *);
static int 	ip4f_init(void);
static struct ip4_frag	*ip4f_alloc(void);
static void 	ip4f_free(struct ip4_frag *);

int (*altq_input)(struct mbuf *, int) = NULL;
static int tbr_timer = 0;	/* token bucket regulator timer */
static struct callout tbr_callout = CALLOUT_INITIALIZER;

/*
 * alternate queueing support routines
 */

/* look up the queue state by the interface name and the queuing type. */
void *
altq_lookup(name, type)
	char *name;
	int type;
{
	struct ifnet *ifp;

	if ((ifp = ifunit(name)) != NULL) {
		if (type != ALTQT_NONE && ifp->if_snd.altq_type == type)
			return (ifp->if_snd.altq_disc);
	}

	return NULL;
}

int
altq_attach(ifq, type, discipline, enqueue, dequeue, request, clfier, classify)
	struct ifaltq *ifq;
	int type;
	void *discipline;
	int (*enqueue)(struct ifaltq *, struct mbuf *, struct altq_pktattr *);
	struct mbuf *(*dequeue)(struct ifaltq *, int);
	int (*request)(struct ifaltq *, int, void *);
	void *clfier;
	void *(*classify)(void *, struct mbuf *, int);
{
	if (!ALTQ_IS_READY(ifq))
		return ENXIO;
	if (ALTQ_IS_ENABLED(ifq))
		return EBUSY;
	if (ALTQ_IS_ATTACHED(ifq))
		return EEXIST;
	ifq->altq_type     = type;
	ifq->altq_disc     = discipline;
	ifq->altq_enqueue  = enqueue;
	ifq->altq_dequeue  = dequeue;
	ifq->altq_request  = request;
	ifq->altq_clfier   = clfier;
	ifq->altq_classify = classify;
	ifq->altq_flags &= ALTQF_CANTCHANGE;
#ifdef ALTQ_KLD
	altq_module_incref(type);
#endif
	return 0;
}

int
altq_detach(ifq)
	struct ifaltq *ifq;
{
	if (!ALTQ_IS_READY(ifq))
		return ENXIO;
	if (ALTQ_IS_ENABLED(ifq))
		return EBUSY;
	if (!ALTQ_IS_ATTACHED(ifq))
		return (0);

#ifdef ALTQ_KLD
	altq_module_declref(ifq->altq_type);
#endif
	ifq->altq_type     = ALTQT_NONE;
	ifq->altq_disc     = NULL;
	ifq->altq_enqueue  = NULL;
	ifq->altq_dequeue  = NULL;
	ifq->altq_request  = NULL;
	ifq->altq_clfier   = NULL;
	ifq->altq_classify = NULL;
	ifq->altq_flags &= ALTQF_CANTCHANGE;
	return 0;
}

int
altq_enable(ifq)
	struct ifaltq *ifq;
{
	int s;
    
	if (!ALTQ_IS_READY(ifq))
		return ENXIO;
	if (ALTQ_IS_ENABLED(ifq))
		return 0;

	s = splimp();
	IFQ_PURGE(ifq);
	ASSERT(ifq->ifq_len == 0);
	ifq->altq_flags |= ALTQF_ENABLED;
	if (ifq->altq_clfier != NULL)
		ifq->altq_flags |= ALTQF_CLASSIFY;
	splx(s);

	return 0;
}

int
altq_disable(ifq)
	struct ifaltq *ifq;
{
	int s;
    
	if (!ALTQ_IS_ENABLED(ifq))
		return 0;

	s = splimp();
	IFQ_PURGE(ifq);
	ASSERT(ifq->ifq_len == 0);
	ifq->altq_flags &= ~(ALTQF_ENABLED|ALTQF_CLASSIFY);
	splx(s);
	return 0;
}

void
altq_assert(file, line, failedexpr)
	const char *file, *failedexpr;
	int line;
{
	(void)printf("altq assertion \"%s\" failed: file \"%s\", line %d\n",
		     failedexpr, file, line);
	panic("altq assertion");
	/* NOTREACHED */
}

/*
 * internal representation of token bucket parameters
 *	rate: 	byte_per_unittime << 32
 *		(((bits_per_sec) / 8) << 32) / machclk_freq
 *	depth:	byte << 32
 *
 */
#define	TBR_SHIFT	32
#define	TBR_SCALE(x)	((int64_t)(x) << TBR_SHIFT)
#define	TBR_UNSCALE(x)	((x) >> TBR_SHIFT)

struct mbuf *
tbr_dequeue(ifq, op)
	struct ifaltq *ifq;
	int op;
{
	struct tb_regulator *tbr;
	struct mbuf *m;
	int64_t interval;
	u_int64_t now;

	tbr = ifq->altq_tbr;
	if (op == ALTDQ_REMOVE && tbr->tbr_lastop == ALTDQ_POLL) {
		/* if this is a remove after poll, bypass tbr check */
	} else {
		/* update token only when it is negative */
		if (tbr->tbr_token <= 0) {
			now = read_machclk();
			interval = now - tbr->tbr_last;
			if (interval >= tbr->tbr_filluptime)
				tbr->tbr_token = tbr->tbr_depth;
			else {
				tbr->tbr_token += interval * tbr->tbr_rate;
				if (tbr->tbr_token > tbr->tbr_depth)
					tbr->tbr_token = tbr->tbr_depth;
			}
			tbr->tbr_last = now;
		}
		/* if token is still negative, don't allow dequeue */
		if (tbr->tbr_token <= 0)
			return (NULL);
	}

	if (ALTQ_IS_ENABLED(ifq))
		m = (*ifq->altq_dequeue)(ifq, op);
	else {
		if (op == ALTDQ_POLL)
			IF_POLL(ifq, m);
		else
			IF_DEQUEUE(ifq, m);
	}

	if (m != NULL && op == ALTDQ_REMOVE)
		tbr->tbr_token -= TBR_SCALE(m_pktlen(m));
	tbr->tbr_lastop = op;
	return (m);
}

/*
 * set a token bucket regulator.
 * if the specified rate is zero, the token bucket regulator is deleted.
 */
int
tbr_set(ifq, profile)
	struct ifaltq *ifq;
	struct tb_profile *profile;
{
	struct tb_regulator *tbr, *otbr;

	if (machclk_freq == 0)
		init_machclk();
	if (machclk_freq == 0) {
		printf("tbr_set: no cpu clock available!\n");
		return (ENXIO);
	}
	
	if (profile->rate == 0) {
		/* delete this tbr */
		if ((tbr = ifq->altq_tbr) == NULL)
			return (ENOENT);
		ifq->altq_tbr = NULL;
		FREE(tbr, M_DEVBUF);
		return (0);
	}

	MALLOC(tbr, struct tb_regulator *, sizeof(struct tb_regulator),
	       M_DEVBUF, M_WAITOK);
	if (tbr == NULL)
		return (ENOMEM);
	bzero(tbr, sizeof(struct tb_regulator));

	tbr->tbr_rate = TBR_SCALE(profile->rate / 8) / machclk_freq;
	tbr->tbr_depth = TBR_SCALE(profile->depth);
	if (tbr->tbr_rate > 0)
		tbr->tbr_filluptime = tbr->tbr_depth / tbr->tbr_rate;
	else
		tbr->tbr_filluptime = 0xffffffffffffffffLL;
	tbr->tbr_token = tbr->tbr_depth;
	tbr->tbr_last = read_machclk();
	tbr->tbr_lastop = ALTDQ_REMOVE;

	otbr = ifq->altq_tbr;
	ifq->altq_tbr = tbr;	/* set the new tbr */

	if (otbr != NULL)
		FREE(otbr, M_DEVBUF);
	else {
		if (tbr_timer == 0) {
			CALLOUT_RESET(&tbr_callout, 1, tbr_timeout, (void *)0);
			tbr_timer = 1;
		}
	}
	return (0);
}

/*
 * tbr_timeout goes through the interface list, and kicks the drivers
 * if necessary.
 */
static void
tbr_timeout(arg)
	void *arg;
{
	struct ifnet *ifp;
	int active, s;

	active = 0;
	s = splimp();
#ifdef __FreeBSD__
#if (__FreeBSD_version < 300000)
	for (ifp = ifnet; ifp; ifp = ifp->if_next)
#else
	for (ifp = ifnet.tqh_first; ifp != NULL; ifp = ifp->if_link.tqe_next)
#endif
#else /* !FreeBSD */
	for (ifp = ifnet.tqh_first; ifp != NULL; ifp = ifp->if_list.tqe_next)
#endif
	{
		if (!TBR_IS_ENABLED(&ifp->if_snd))
			continue;
		active++;
		if (!IFQ_IS_EMPTY(&ifp->if_snd) && ifp->if_start != NULL)
			(*ifp->if_start)(ifp);
	}
	splx(s);
	if (active > 0)
		CALLOUT_RESET(&tbr_callout, 1, tbr_timeout, (void *)0);
	else
		tbr_timer = 0;	/* don't need tbr_timer anymore */
#if defined(__alpha__) && !defined(ALTQ_NOPCC)
	{
		/*
		 * XXX read out the machine dependent clock once a second
		 * to detect counter wrap-around.
		 */
		static u_int cnt;

		if (++cnt >= hz) {
			(void)read_machclk();
			cnt = 0;
		}
	}
#endif /* __alpha__ && !ALTQ_NOPCC */
}

/*
 * get token bucket regulator profile
 */
int
tbr_get(ifq, profile)
	struct ifaltq *ifq;
	struct tb_profile *profile;
{
	struct tb_regulator *tbr;

	if ((tbr = ifq->altq_tbr) == NULL) {
		profile->rate = 0;
		profile->depth = 0;
	} else {
		profile->rate =
		    (u_int)TBR_UNSCALE(tbr->tbr_rate * 8 * machclk_freq);
		profile->depth = (u_int)TBR_UNSCALE(tbr->tbr_depth);
	}
	return (0);
}


#ifndef IPPROTO_ESP
#define	IPPROTO_ESP	50		/* encapsulating security payload */
#endif
#ifndef IPPROTO_AH
#define	IPPROTO_AH	51		/* authentication header */
#endif

/* 
 * extract flow information from a given packet.
 * filt_mask shows flowinfo fields required.
 * we assume the ip header is in one mbuf, and addresses and ports are
 * in network byte order.
 */
int 
altq_extractflow(m, af, flow, filt_bmask)
	struct mbuf *m;
	int af;
	struct flowinfo *flow;
	u_int32_t	filt_bmask;
{

	switch (af) {
	case PF_INET: {
		struct flowinfo_in *fin;
		struct ip *ip;

		ip = mtod(m, struct ip *);

		if (ip->ip_v != 4)
			break;

		fin = (struct flowinfo_in *)flow;
		fin->fi_len = sizeof(struct flowinfo_in);
		fin->fi_family = AF_INET;

		fin->fi_proto = ip->ip_p;
		fin->fi_tos = ip->ip_tos;

		fin->fi_src.s_addr = ip->ip_src.s_addr;
		fin->fi_dst.s_addr = ip->ip_dst.s_addr;

		if (filt_bmask & FIMB4_PORTS)
			/* if port info is required, extract port numbers */
			extract_ports4(m, ip, fin);
		else {
			fin->fi_sport = 0;
			fin->fi_dport = 0;
			fin->fi_gpi = 0;
		}
		return (1);
	}

#ifdef INET6
	case PF_INET6: {
		struct flowinfo_in6 *fin6;
		struct ip6_hdr *ip6;

		ip6 = mtod(m, struct ip6_hdr *);
		/* should we check the ip version? */

		fin6 = (struct flowinfo_in6 *)flow;
		fin6->fi6_len = sizeof(struct flowinfo_in6);
		fin6->fi6_family = AF_INET6;

		fin6->fi6_proto = ip6->ip6_nxt;
		fin6->fi6_tclass   = (ntohl(ip6->ip6_flow) >> 20) & 0xff;

		fin6->fi6_flowlabel = ip6->ip6_flow & htonl(0x000fffff);
		fin6->fi6_src = ip6->ip6_src;
		fin6->fi6_dst = ip6->ip6_dst;

		if ((filt_bmask & FIMB6_PORTS) ||
		    ((filt_bmask & FIMB6_PROTO)
		     && ip6->ip6_nxt > IPPROTO_IPV6))
			/*
			 * if port info is required, or proto is required
			 * but there are option headers, extract port
			 * and protocol numbers.
			 */
			extract_ports6(m, ip6, fin6);
		else {
			fin6->fi6_sport = 0;
			fin6->fi6_dport = 0;
			fin6->fi6_gpi = 0;
		}
		return (1);
	}
#endif /* INET6 */

	default:
		break;
	}

	/* failed */
	flow->fi_len = sizeof(struct flowinfo);
	flow->fi_family = AF_UNSPEC;
	return (0);
}

/*
 * helper routine to extract port numbers
 */
/* structure for ipsec and ipv6 option header template */
struct _opt6 {
	u_int8_t	opt6_nxt;	/* next header */
	u_int8_t	opt6_hlen;	/* header extension length */
	u_int16_t	_pad;
	u_int32_t	ah_spi;		/* security parameter index
					   for authentication header */
};

/*
 * extract port numbers from a ipv4 packet.
 */
static int
extract_ports4(m, ip, fin)
	struct mbuf *m;
	struct ip *ip;
	struct flowinfo_in *fin;
{
	struct mbuf *m0;
	u_short ip_off;
	u_int8_t proto;
	int 	off;

	fin->fi_sport = 0;
	fin->fi_dport = 0;
	fin->fi_gpi = 0;

	ip_off = ntohs(ip->ip_off);
	/* if it is a fragment, try cached fragment info */
	if (ip_off & IP_OFFMASK) {
		ip4f_lookup(ip, fin);
		return (1);
	}

	/* locate the mbuf containing the protocol header */
	for (m0 = m; m0 != NULL; m0 = m0->m_next)
		if (((caddr_t)ip >= m0->m_data) &&
		    ((caddr_t)ip < m0->m_data + m0->m_len))
			break;
	if (m0 == NULL) {
#ifdef ALTQ_DEBUG
		printf("extract_ports4: can't locate header! ip=%p\n", ip);
#endif
		return (0);
	}
	off = ((caddr_t)ip - m0->m_data) + (ip->ip_hl << 2);
	proto = ip->ip_p;

#ifdef ALTQ_IPSEC
 again:
#endif
	while (off >= m0->m_len) {
		off -= m0->m_len;
		m0 = m0->m_next;
		if (m0 == NULL)
			return (0);  /* bogus ip_hl! */
	}
	if (m0->m_len < off + 4)
		return (0);

	switch (proto) {
	case IPPROTO_TCP:
	case IPPROTO_UDP: {
		struct udphdr *udp;

		udp = (struct udphdr *)(mtod(m0, caddr_t) + off);
		fin->fi_sport = udp->uh_sport;
		fin->fi_dport = udp->uh_dport;
		fin->fi_proto = proto;
		}
		break;

#ifdef ALTQ_IPSEC
	case IPPROTO_ESP:
		if (fin->fi_gpi == 0){
			u_int32_t *gpi;

			gpi = (u_int32_t *)(mtod(m0, caddr_t) + off);
			fin->fi_gpi   = *gpi;
		}
		fin->fi_proto = proto;
		break;

	case IPPROTO_AH: {
			/* get next header and header length */
			struct _opt6 *opt6;

			opt6 = (struct _opt6 *)(mtod(m0, caddr_t) + off);
			proto = opt6->opt6_nxt;
			off += 8 + (opt6->opt6_hlen * 4);
			if (fin->fi_gpi == 0 && m0->m_len >= off + 8)
				fin->fi_gpi = opt6->ah_spi;
		}
		/* goto the next header */
		goto again;
#endif  /* ALTQ_IPSEC */

	default:
		fin->fi_proto = proto;
		return (0);
	}

	/* if this is a first fragment, cache it. */
	if (ip_off & IP_MF)
		ip4f_cache(ip, fin);

	return (1);
}

#ifdef INET6
static int
extract_ports6(m, ip6, fin6)
	struct mbuf *m;
	struct ip6_hdr *ip6;
	struct flowinfo_in6 *fin6;
{
	struct mbuf *m0;
	int	off;
	u_int8_t proto;

	fin6->fi6_gpi   = 0;
	fin6->fi6_sport = 0;
	fin6->fi6_dport = 0;

	/* locate the mbuf containing the protocol header */
	for (m0 = m; m0 != NULL; m0 = m0->m_next)
		if (((caddr_t)ip6 >= m0->m_data) &&
		    ((caddr_t)ip6 < m0->m_data + m0->m_len))
			break;
	if (m0 == NULL) {
#ifdef ALTQ_DEBUG
		printf("extract_ports6: can't locate header! ip6=%p\n", ip6);
#endif
		return (0);
	}
	off = ((caddr_t)ip6 - m0->m_data) + sizeof(struct ip6_hdr);

	proto = ip6->ip6_nxt;
	do {
		while (off >= m0->m_len) {
			off -= m0->m_len;
			m0 = m0->m_next;
			if (m0 == NULL)
				return (0);
		}
		if (m0->m_len < off + 4)
			return (0);

		switch (proto) {
		case IPPROTO_TCP:
		case IPPROTO_UDP: {
			struct udphdr *udp;

			udp = (struct udphdr *)(mtod(m0, caddr_t) + off);
			fin6->fi6_sport = udp->uh_sport;
			fin6->fi6_dport = udp->uh_dport;
			fin6->fi6_proto = proto;
			}
			return (1);

		case IPPROTO_ESP:
			if (fin6->fi6_gpi == 0) {
				u_int32_t *gpi;

				gpi = (u_int32_t *)(mtod(m0, caddr_t) + off);
				fin6->fi6_gpi   = *gpi;
			}
			fin6->fi6_proto = proto;
			return (1);

		case IPPROTO_AH: {
			/* get next header and header length */
			struct _opt6 *opt6;

			opt6 = (struct _opt6 *)(mtod(m0, caddr_t) + off);
			if (fin6->fi6_gpi == 0 && m0->m_len >= off + 8)
				fin6->fi6_gpi = opt6->ah_spi;
			proto = opt6->opt6_nxt;
			off += 8 + (opt6->opt6_hlen * 4);
			/* goto the next header */
			break;
			}

		case IPPROTO_HOPOPTS:
		case IPPROTO_ROUTING:
		case IPPROTO_DSTOPTS: {
			/* get next header and header length */
			struct _opt6 *opt6;

			opt6 = (struct _opt6 *)(mtod(m0, caddr_t) + off);
			proto = opt6->opt6_nxt;
			off += (opt6->opt6_hlen + 1) * 8;
			/* goto the next header */
			break;
			}

		case IPPROTO_FRAGMENT:
			/* ipv6 fragmentations are not supported yet */
		default:
			fin6->fi6_proto = proto;
			return (0);
		}
	} while (1);
	/*NOTREACHED*/
}
#endif /* INET6 */

/*
 * altq common classifier
 */
int
acc_add_filter(classifier, filter, class, phandle)
	struct acc_classifier *classifier;
	struct flow_filter *filter;
	void	*class;
	u_long	*phandle;
{
	struct acc_filter *afp, *prev, *tmp;
	int	i, s;

#ifdef INET6
	if (filter->ff_flow.fi_family != AF_INET &&
	    filter->ff_flow.fi_family != AF_INET6)
		return (EINVAL);
#else
	if (filter->ff_flow.fi_family != AF_INET)
		return (EINVAL);
#endif

	MALLOC(afp, struct acc_filter *, sizeof(struct acc_filter),
	       M_DEVBUF, M_WAITOK);
	if (afp == NULL)
		return (ENOMEM);
	bzero(afp, sizeof(struct acc_filter));

	afp->f_filter = *filter;
	afp->f_class = class;

	i = ACC_WILDCARD_INDEX;
	if (filter->ff_flow.fi_family == AF_INET) {
		struct flow_filter *filter4 = &afp->f_filter;

		/*
		 * if address is 0, it's a wildcard.  if address mask
		 * isn't set, use full mask.
		 */
		if (filter4->ff_flow.fi_dst.s_addr == 0)
			filter4->ff_mask.mask_dst.s_addr = 0;
		else if (filter4->ff_mask.mask_dst.s_addr == 0)
			filter4->ff_mask.mask_dst.s_addr = 0xffffffff;
		if (filter4->ff_flow.fi_src.s_addr == 0)
			filter4->ff_mask.mask_src.s_addr = 0;
		else if (filter4->ff_mask.mask_src.s_addr == 0)
			filter4->ff_mask.mask_src.s_addr = 0xffffffff;

		/* clear extra bits in addresses  */
		   filter4->ff_flow.fi_dst.s_addr &=
		       filter4->ff_mask.mask_dst.s_addr;
		   filter4->ff_flow.fi_src.s_addr &=
		       filter4->ff_mask.mask_src.s_addr;

		/*
		 * if dst address is a wildcard, use hash-entry
		 * ACC_WILDCARD_INDEX.
		 */
		if (filter4->ff_mask.mask_dst.s_addr != 0xffffffff)
			i = ACC_WILDCARD_INDEX;
		else
			i = ACC_GET_HASH_INDEX(filter4->ff_flow.fi_dst.s_addr);
	}
#ifdef INET6
	else if (filter->ff_flow.fi_family == AF_INET6) {
		struct flow_filter6 *filter6 =
			(struct flow_filter6 *)&afp->f_filter;
#ifndef IN6MASK0 /* taken from kame ipv6 */
#define	IN6MASK0	{{{ 0, 0, 0, 0 }}}
#define	IN6MASK128	{{{ 0xffffffff, 0xffffffff, 0xffffffff, 0xffffffff }}}
		const struct in6_addr in6mask0 = IN6MASK0;
		const struct in6_addr in6mask128 = IN6MASK128;
#endif

		if (IN6_IS_ADDR_UNSPECIFIED(&filter6->ff_flow6.fi6_dst))
			filter6->ff_mask6.mask6_dst = in6mask0;
		else if (IN6_IS_ADDR_UNSPECIFIED(&filter6->ff_mask6.mask6_dst))
			filter6->ff_mask6.mask6_dst = in6mask128;
		if (IN6_IS_ADDR_UNSPECIFIED(&filter6->ff_flow6.fi6_src))
			filter6->ff_mask6.mask6_src = in6mask0;
		else if (IN6_IS_ADDR_UNSPECIFIED(&filter6->ff_mask6.mask6_src))
			filter6->ff_mask6.mask6_src = in6mask128;

		/* clear extra bits in addresses  */
		for (i = 0; i < 16; i++)
			filter6->ff_flow6.fi6_dst.s6_addr[i] &=
			    filter6->ff_mask6.mask6_dst.s6_addr[i];
		for (i = 0; i < 16; i++)
			filter6->ff_flow6.fi6_src.s6_addr[i] &=
			    filter6->ff_mask6.mask6_src.s6_addr[i];

		if (filter6->ff_flow6.fi6_flowlabel == 0)
			i = ACC_WILDCARD_INDEX;
		else
			i = ACC_GET_HASH_INDEX(filter6->ff_flow6.fi6_flowlabel);
	}
#endif /* INET6 */

	afp->f_handle = get_filt_handle(classifier, i);

	/* update filter bitmask */
	afp->f_fbmask = filt2fibmask(filter);
	classifier->acc_fbmask |= afp->f_fbmask;

	/*
	 * add this filter to the filter list.
	 * filters are ordered from the highest rule number.
	 */
	s = splimp();
	prev = NULL;
	LIST_FOREACH(tmp, &classifier->acc_filters[i], f_chain) {
		if (tmp->f_filter.ff_ruleno > afp->f_filter.ff_ruleno)
			prev = tmp;
		else
			break;
	}
	if (prev == NULL)
		LIST_INSERT_HEAD(&classifier->acc_filters[i], afp, f_chain);
	else
		LIST_INSERT_AFTER(prev, afp, f_chain);
	splx(s);

	*phandle = afp->f_handle;
	return (0);
}

int
acc_delete_filter(classifier, handle)
	struct acc_classifier *classifier;
	u_long handle;
{
	struct acc_filter *afp;
	int	s;

	if ((afp = filth_to_filtp(classifier, handle)) == NULL)
		return (EINVAL);

	s = splimp();
	LIST_REMOVE(afp, f_chain);
	splx(s);

	FREE(afp, M_DEVBUF);

	/* todo: update filt_bmask */

	return (0);
}

/*
 * delete filters referencing to the specified class.
 * if the all flag is not 0, delete all the filters.
 */
int
acc_discard_filters(classifier, class, all)
	struct acc_classifier *classifier;
	void	*class;
	int	all;
{
	struct acc_filter *afp;
	int	i, s;

	s = splimp();
	for (i = 0; i < ACC_FILTER_TABLESIZE; i++) {
		do {
			LIST_FOREACH(afp, &classifier->acc_filters[i], f_chain)
				if (all || afp->f_class == class) {
					LIST_REMOVE(afp, f_chain);
					FREE(afp, M_DEVBUF);
					/* start again from the head */
					break;
				}
		} while (afp != NULL);
	}
	splx(s);

	if (all)
		classifier->acc_fbmask = 0;

	return (0);
}

void *
acc_classify(clfier, m, af)
	void *clfier;
	struct mbuf *m;
	int af;
{
	struct acc_classifier *classifier;
	struct flowinfo flow;
	struct acc_filter *afp;
	int	i;

	classifier = (struct acc_classifier *)clfier;
	altq_extractflow(m, af, &flow, classifier->acc_fbmask);

	if (flow.fi_family == AF_INET) {
		struct flowinfo_in *fp = (struct flowinfo_in *)&flow;

		if ((classifier->acc_fbmask & FIMB4_ALL) == FIMB4_TOS) {
			/* only tos is used */
			LIST_FOREACH(afp,
				 &classifier->acc_filters[ACC_WILDCARD_INDEX],
				 f_chain)
				if (apply_tosfilter4(afp->f_fbmask,
						     &afp->f_filter, fp))
					/* filter matched */
					return (afp->f_class);
		} else if ((classifier->acc_fbmask &
			(~(FIMB4_PROTO|FIMB4_SPORT|FIMB4_DPORT) & FIMB4_ALL))
		    == 0) {
			/* only proto and ports are used */
			LIST_FOREACH(afp,
				 &classifier->acc_filters[ACC_WILDCARD_INDEX],
				 f_chain)
				if (apply_ppfilter4(afp->f_fbmask,
						    &afp->f_filter, fp))
					/* filter matched */
					return (afp->f_class);
		} else {
			/* get the filter hash entry from its dest address */
			i = ACC_GET_HASH_INDEX(fp->fi_dst.s_addr);
			do {
				/*
				 * go through this loop twice.  first for dst
				 * hash, second for wildcards.
				 */
				LIST_FOREACH(afp, &classifier->acc_filters[i],
					     f_chain)
					if (apply_filter4(afp->f_fbmask,
							  &afp->f_filter, fp))
						/* filter matched */
						return (afp->f_class);

				/*
				 * check again for filters with a dst addr
				 * wildcard.
				 * (daddr == 0 || dmask != 0xffffffff).
				 */
				if (i != ACC_WILDCARD_INDEX)
					i = ACC_WILDCARD_INDEX;
				else
					break;
			} while (1);
		}
	}
#ifdef INET6
	else if (flow.fi_family == AF_INET6) {
		struct flowinfo_in6 *fp6 = (struct flowinfo_in6 *)&flow;

		/* get the filter hash entry from its flow ID */
		if (fp6->fi6_flowlabel != 0)
			i = ACC_GET_HASH_INDEX(fp6->fi6_flowlabel);
		else
			/* flowlable can be zero */
			i = ACC_WILDCARD_INDEX;

		/* go through this loop twice.  first for flow hash, second
		   for wildcards. */
		do {
			LIST_FOREACH(afp, &classifier->acc_filters[i], f_chain)
				if (apply_filter6(afp->f_fbmask,
					(struct flow_filter6 *)&afp->f_filter,
					fp6))
					/* filter matched */
					return (afp->f_class);

			/*
			 * check again for filters with a wildcard.
			 */
			if (i != ACC_WILDCARD_INDEX)
				i = ACC_WILDCARD_INDEX;
			else
				break;
		} while (1);
	}
#endif /* INET6 */

	/* no filter matched */
	return (NULL);
}

static int
apply_filter4(fbmask, filt, pkt)
	u_int32_t	fbmask;
	struct flow_filter *filt;
	struct flowinfo_in *pkt;
{
	if (filt->ff_flow.fi_family != AF_INET)
		return (0);
	if ((fbmask & FIMB4_SPORT) && filt->ff_flow.fi_sport != pkt->fi_sport)
		return (0);
	if ((fbmask & FIMB4_DPORT) && filt->ff_flow.fi_dport != pkt->fi_dport)
		return (0);
	if ((fbmask & FIMB4_DADDR) &&
	    filt->ff_flow.fi_dst.s_addr !=
	    (pkt->fi_dst.s_addr & filt->ff_mask.mask_dst.s_addr))
		return (0);
	if ((fbmask & FIMB4_SADDR) &&
	    filt->ff_flow.fi_src.s_addr !=
	    (pkt->fi_src.s_addr & filt->ff_mask.mask_src.s_addr))
		return (0);
	if ((fbmask & FIMB4_PROTO) && filt->ff_flow.fi_proto != pkt->fi_proto)
		return (0);
	if ((fbmask & FIMB4_TOS) && filt->ff_flow.fi_tos !=
	    (pkt->fi_tos & filt->ff_mask.mask_tos))
		return (0);
	if ((fbmask & FIMB4_GPI) && filt->ff_flow.fi_gpi != (pkt->fi_gpi))
		return (0);
	/* match */
	return (1);
}

/*
 * filter matching function optimized for a common case that checks
 * only protocol and port numbers
 */
static int
apply_ppfilter4(fbmask, filt, pkt)
	u_int32_t	fbmask;
	struct flow_filter *filt;
	struct flowinfo_in *pkt;
{
	if (filt->ff_flow.fi_family != AF_INET)
		return (0);
	if ((fbmask & FIMB4_SPORT) && filt->ff_flow.fi_sport != pkt->fi_sport)
		return (0);
	if ((fbmask & FIMB4_DPORT) && filt->ff_flow.fi_dport != pkt->fi_dport)
		return (0);
	if ((fbmask & FIMB4_PROTO) && filt->ff_flow.fi_proto != pkt->fi_proto)
		return (0);
	/* match */
	return (1);
}

/*
 * filter matching function only for tos field.
 */
static int
apply_tosfilter4(fbmask, filt, pkt)
	u_int32_t	fbmask;
	struct flow_filter *filt;
	struct flowinfo_in *pkt;
{
	if (filt->ff_flow.fi_family != AF_INET)
		return (0);
	if ((fbmask & FIMB4_TOS) && filt->ff_flow.fi_tos !=
	    (pkt->fi_tos & filt->ff_mask.mask_tos))
		return (0);
	/* match */
	return (1);
}

#ifdef INET6
static int
apply_filter6(fbmask, filt, pkt)
	u_int32_t	fbmask;
	struct flow_filter6 *filt;
	struct flowinfo_in6 *pkt;
{
	int i;

	if (filt->ff_flow6.fi6_family != AF_INET6)
		return (0);
	if ((fbmask & FIMB6_FLABEL) &&
	    filt->ff_flow6.fi6_flowlabel != pkt->fi6_flowlabel)
		return (0);
	if ((fbmask & FIMB6_PROTO) &&
	    filt->ff_flow6.fi6_proto != pkt->fi6_proto)
		return (0);
	if ((fbmask & FIMB6_SPORT) &&
	    filt->ff_flow6.fi6_sport != pkt->fi6_sport)
		return (0);
	if ((fbmask & FIMB6_DPORT) &&
	    filt->ff_flow6.fi6_dport != pkt->fi6_dport)
		return (0);
	if (fbmask & FIMB6_SADDR) {
		for (i = 0; i < 4; i++)
			if (filt->ff_flow6.fi6_src.s6_addr32[i] !=
			    (pkt->fi6_src.s6_addr32[i] &
			     filt->ff_mask6.mask6_src.s6_addr32[i]))
				return (0);
	}
	if (fbmask & FIMB6_DADDR) {
		for (i = 0; i < 4; i++)
			if (filt->ff_flow6.fi6_dst.s6_addr32[i] !=
			    (pkt->fi6_dst.s6_addr32[i] &
			     filt->ff_mask6.mask6_dst.s6_addr32[i]))
				return (0);
	}
	if ((fbmask & FIMB6_TCLASS) &&
	    filt->ff_flow6.fi6_tclass !=
	    (pkt->fi6_tclass & filt->ff_mask6.mask6_tclass))
		return (0);
	if ((fbmask & FIMB6_GPI) &&
	    filt->ff_flow6.fi6_gpi != pkt->fi6_gpi)
		return (0);
	/* match */
	return (1);
}
#endif /* INET6 */

/*
 *  filter handle:
 *	bit 20-28: index to the filter hash table
 *	bit  0-19: unique id in the hash bucket.
 */
static u_long
get_filt_handle(classifier, i)
	struct acc_classifier *classifier;
	int	i;
{
	static u_long handle_number = 1;
	u_long 	handle;
	struct acc_filter *afp;

	while (1) {
		handle = handle_number++ & 0x000fffff;

		if (LIST_EMPTY(&classifier->acc_filters[i]))
			break;

		LIST_FOREACH(afp, &classifier->acc_filters[i], f_chain)
			if ((afp->f_handle & 0x000fffff) == handle)
				break;
		if (afp == NULL)
			break;
		/* this handle is already used, try again */
	}

	return ((i << 20) | handle);
}

/* convert filter handle to filter pointer */
static struct acc_filter *
filth_to_filtp(classifier, handle)
	struct acc_classifier *classifier;
	u_long handle;
{
	struct acc_filter *afp;
	int	i;

	i = ACC_GET_HINDEX(handle);

	LIST_FOREACH(afp, &classifier->acc_filters[i], f_chain)
		if (afp->f_handle == handle)
			return (afp);

	return (NULL);
}

/* create flowinfo bitmask */
static u_int32_t
filt2fibmask(filt)
	struct flow_filter *filt;
{
	u_int32_t mask = 0;
#ifdef INET6
	struct flow_filter6 *filt6;
#endif

	switch (filt->ff_flow.fi_family) {
	case AF_INET:
		if (filt->ff_flow.fi_proto != 0)
			mask |= FIMB4_PROTO;
		if (filt->ff_flow.fi_tos != 0)
			mask |= FIMB4_TOS;
		if (filt->ff_flow.fi_dst.s_addr != 0)
			mask |= FIMB4_DADDR;
		if (filt->ff_flow.fi_src.s_addr != 0)
			mask |= FIMB4_SADDR;
		if (filt->ff_flow.fi_sport != 0)
			mask |= FIMB4_SPORT;
		if (filt->ff_flow.fi_dport != 0)
			mask |= FIMB4_DPORT;
		if (filt->ff_flow.fi_gpi != 0)
			mask |= FIMB4_GPI;
		break;
#ifdef INET6
	case AF_INET6:
		filt6 = (struct flow_filter6 *)filt;

		if (filt6->ff_flow6.fi6_proto != 0)
			mask |= FIMB6_PROTO;
		if (filt6->ff_flow6.fi6_tclass != 0)
			mask |= FIMB6_TCLASS;
		if (!IN6_IS_ADDR_UNSPECIFIED(&filt6->ff_flow6.fi6_dst))
			mask |= FIMB6_DADDR;
		if (!IN6_IS_ADDR_UNSPECIFIED(&filt6->ff_flow6.fi6_src))
			mask |= FIMB6_SADDR;
		if (filt6->ff_flow6.fi6_sport != 0)
			mask |= FIMB6_SPORT;
		if (filt6->ff_flow6.fi6_dport != 0)
			mask |= FIMB6_DPORT;
		if (filt6->ff_flow6.fi6_gpi != 0)
			mask |= FIMB6_GPI;
		if (filt6->ff_flow6.fi6_flowlabel != 0)
			mask |= FIMB6_FLABEL;
		break;
#endif /* INET6 */
	}
	return (mask);
}


/*
 * helper functions to handle IPv4 fragments.
 * currently only in-sequence fragments are handled.
 *	- fragment info is cached in a LRU list.
 *	- when a first fragment is found, cache its flow info.
 *	- when a non-first fragment is found, lookup the cache.
 */

struct ip4_frag {
    TAILQ_ENTRY(ip4_frag) ip4f_chain;
    char    ip4f_valid;
    u_short ip4f_id;
    struct flowinfo_in ip4f_info;
};

static TAILQ_HEAD(ip4f_list, ip4_frag) ip4f_list; /* IPv4 fragment cache */

#define	IP4F_TABSIZE		16	/* IPv4 fragment cache size */


static void
ip4f_cache(ip, fin)
	struct ip *ip;
	struct flowinfo_in *fin;
{
	struct ip4_frag *fp;

	if (TAILQ_EMPTY(&ip4f_list)) {
		/* first time call, allocate fragment cache entries. */
		if (ip4f_init() < 0)
			/* allocation failed! */
			return;
	}

	fp = ip4f_alloc();
	fp->ip4f_id = ip->ip_id;
	fp->ip4f_info.fi_proto = ip->ip_p;
	fp->ip4f_info.fi_src.s_addr = ip->ip_src.s_addr;
	fp->ip4f_info.fi_dst.s_addr = ip->ip_dst.s_addr;

	/* save port numbers */
	fp->ip4f_info.fi_sport = fin->fi_sport;
	fp->ip4f_info.fi_dport = fin->fi_dport;
	fp->ip4f_info.fi_gpi   = fin->fi_gpi;
}

static int
ip4f_lookup(ip, fin)
	struct ip *ip;
	struct flowinfo_in *fin;
{
	struct ip4_frag *fp;

	for (fp = TAILQ_FIRST(&ip4f_list); fp != NULL && fp->ip4f_valid;
	     fp = TAILQ_NEXT(fp, ip4f_chain))
		if (ip->ip_id == fp->ip4f_id &&
		    ip->ip_src.s_addr == fp->ip4f_info.fi_src.s_addr &&
		    ip->ip_dst.s_addr == fp->ip4f_info.fi_dst.s_addr &&
		    ip->ip_p == fp->ip4f_info.fi_proto) {

			/* found the matching entry */
			fin->fi_sport = fp->ip4f_info.fi_sport;
			fin->fi_dport = fp->ip4f_info.fi_dport;
			fin->fi_gpi   = fp->ip4f_info.fi_gpi;

			if ((ntohs(ip->ip_off) & IP_MF) == 0)
				/* this is the last fragment,
				   release the entry. */
				ip4f_free(fp);

			return (1);
		}

	/* no matching entry found */
	return (0);
}

static int
ip4f_init(void)
{
	struct ip4_frag *fp;
	int i;

	TAILQ_INIT(&ip4f_list);
	for (i=0; i<IP4F_TABSIZE; i++) {
		MALLOC(fp, struct ip4_frag *, sizeof(struct ip4_frag),
		       M_DEVBUF, M_NOWAIT);
		if (fp == NULL) {
			printf("ip4f_init: can't alloc %dth entry!\n", i);
			if (i == 0)
				return (-1);
			return (0);
		}
		fp->ip4f_valid = 0;
		TAILQ_INSERT_TAIL(&ip4f_list, fp, ip4f_chain);
	}
	return (0);
}

static struct ip4_frag *
ip4f_alloc(void)
{
	struct ip4_frag *fp;

	/* reclaim an entry at the tail, put it at the head */
	fp = TAILQ_LAST(&ip4f_list, ip4f_list);
	TAILQ_REMOVE(&ip4f_list, fp, ip4f_chain);
	fp->ip4f_valid = 1;
	TAILQ_INSERT_HEAD(&ip4f_list, fp, ip4f_chain);
	return (fp);
}

static void
ip4f_free(fp)
	struct ip4_frag *fp;
{
	TAILQ_REMOVE(&ip4f_list, fp, ip4f_chain);
	fp->ip4f_valid = 0;
	TAILQ_INSERT_TAIL(&ip4f_list, fp, ip4f_chain);
}

/*
 * read and write diffserv field in IPv4 or IPv6 header
 */
u_int8_t
read_dsfield(m, pktattr)
	struct mbuf *m;
	struct altq_pktattr *pktattr;
{
	struct mbuf *m0;
	u_int8_t ds_field = 0;

	if (pktattr == NULL ||
	    (pktattr->pattr_af != AF_INET && pktattr->pattr_af != AF_INET6))
		return ((u_int8_t)0);

	/* verify that pattr_hdr is within the mbuf data */
	for (m0 = m; m0 != NULL; m0 = m0->m_next)
		if ((pktattr->pattr_hdr >= m0->m_data) &&
		    (pktattr->pattr_hdr < m0->m_data + m0->m_len))
			break;
	if (m0 == NULL) {
		/* ick, pattr_hdr is stale */
		pktattr->pattr_af = AF_UNSPEC;
#ifdef ALTQ_DEBUG
		printf("read_dsfield: can't locate header!\n");
#endif
		return ((u_int8_t)0);
	}

	if (pktattr->pattr_af == AF_INET) {
		struct ip *ip = (struct ip *)pktattr->pattr_hdr;

		if (ip->ip_v != 4)
			return ((u_int8_t)0);	/* version mismatch! */
		ds_field = ip->ip_tos;
	}
#ifdef INET6
	else if (pktattr->pattr_af == AF_INET6) {
		struct ip6_hdr *ip6 = (struct ip6_hdr *)pktattr->pattr_hdr;
		u_int32_t flowlabel;

		flowlabel = ntohl(ip6->ip6_flow);
		if ((flowlabel >> 28) != 6)
			return ((u_int8_t)0);	/* version mismatch! */
		ds_field = (flowlabel >> 20) & 0xff;
	}
#endif
	return (ds_field);
}

void
write_dsfield(m, pktattr, dsfield)
	struct mbuf *m;
	struct altq_pktattr *pktattr;
	u_int8_t dsfield;
{
	struct mbuf *m0;

	if (pktattr == NULL ||
	    (pktattr->pattr_af != AF_INET && pktattr->pattr_af != AF_INET6))
		return;

	/* verify that pattr_hdr is within the mbuf data */
	for (m0 = m; m0 != NULL; m0 = m0->m_next)
		if ((pktattr->pattr_hdr >= m0->m_data) &&
		    (pktattr->pattr_hdr < m0->m_data + m0->m_len))
			break;
	if (m0 == NULL) {
		/* ick, pattr_hdr is stale */
		pktattr->pattr_af = AF_UNSPEC;
#ifdef ALTQ_DEBUG
		printf("write_dsfield: can't locate header!\n");
#endif
		return;
	}

	if (pktattr->pattr_af == AF_INET) {
		struct ip *ip = (struct ip *)pktattr->pattr_hdr;
		u_int8_t old;
		int32_t sum;

		if (ip->ip_v != 4)
			return;		/* version mismatch! */
		old = ip->ip_tos;
		dsfield |= old & 3;	/* leave CU bits */
		if (old == dsfield)
			return;
		ip->ip_tos = dsfield;
		/*
		 * update checksum (from RFC1624)
		 *	   HC' = ~(~HC + ~m + m')
		 */
		sum = ~ntohs(ip->ip_sum) & 0xffff;
		sum += 0xff00 + (~old & 0xff) + dsfield;
		sum = (sum >> 16) + (sum & 0xffff);
		sum += (sum >> 16);  /* add carry */

		ip->ip_sum = htons(~sum & 0xffff);
	}
#ifdef INET6
	else if (pktattr->pattr_af == AF_INET6) {
		struct ip6_hdr *ip6 = (struct ip6_hdr *)pktattr->pattr_hdr;
		u_int32_t flowlabel;

		flowlabel = ntohl(ip6->ip6_flow);
		if ((flowlabel >> 28) != 6)
			return;		/* version mismatch! */
		flowlabel = (flowlabel & 0xf03fffff) | (dsfield << 20);
		ip6->ip6_flow = htonl(flowlabel);
	}
#endif
	return;
}


/*
 * high resolution clock support taking advantage of a machine dependent
 * high resolution time counter (e.g., timestamp counter of intel pentium).
 * we assume
 *  - 64-bit-long monotonically-increasing counter
 *  - frequency range is 100M-4GHz (CPU speed)
 */
u_int32_t machclk_freq = 0;
u_int32_t machclk_per_tick = 0;

#if (defined(__i386__) || defined(__alpha__)) && !defined(ALTQ_NOPCC)

#if defined(__FreeBSD__) && defined(SMP)
#error SMP system!  use ALTQ_NOPCC option.
#endif

#ifdef __alpha__
#ifdef __FreeBSD__
extern u_int32_t cycles_per_sec;	/* alpha cpu clock frequency */
#elif defined(__NetBSD__) || defined(__OpenBSD__)
extern u_int64_t cycles_per_usec;	/* alpha cpu clock frequency */
#endif
#endif /* __alpha__ */
#if defined(__i386__) && defined(__NetBSD__)
extern u_int64_t cpu_tsc_freq;
#endif

void
init_machclk(void) 
{
	/* sanity check */
#ifdef __i386__
	/* check if TSC is available */
	if ((cpu_feature & CPUID_TSC) == 0) {
		printf("altq: TSC isn't available! use ALTQ_NOPCC option.\n");
		return;
	}
#endif

	/*
	 * if the clock frequency (of Pentium TSC or Alpha PCC) is
	 * accessible, just use it.
	 */
#ifdef __i386__
#ifdef __FreeBSD__
#if (__FreeBSD_version > 300000)
	machclk_freq = tsc_freq;
#else
	machclk_freq = i586_ctr_freq;
#endif
#elif defined(__NetBSD__)
	machclk_freq = (u_int32_t)cpu_tsc_freq;
#elif defined(__OpenBSD__) && (defined(I586_CPU) || defined(I686_CPU))
	machclk_freq = pentium_mhz * 1000000;
#endif
#elif defined(__alpha__)
#ifdef __FreeBSD__
	machclk_freq = cycles_per_sec;
#elif defined(__NetBSD__) || defined(__OpenBSD__)
	machclk_freq = (u_int32_t)(cycles_per_usec * 1000000);
#endif
#endif /* __alpha__ */

	/*
	 * if we don't know the clock frequency, measure it.
	 */
	if (machclk_freq == 0) {
		static int	wait;
		struct timeval	tv_start, tv_end;
		u_int64_t	start, end, diff;
		int		timo;

		microtime(&tv_start);
		start = read_machclk();
		timo = hz;	/* 1 sec */
		(void)tsleep(&wait, PWAIT | PCATCH, "init_machclk", timo);
		microtime(&tv_end);
		end = read_machclk();
		diff = (u_int64_t)(tv_end.tv_sec - tv_start.tv_sec) * 1000000
		    + tv_end.tv_usec - tv_start.tv_usec;
		if (diff != 0)
			machclk_freq = (u_int)((end - start) * 1000000 / diff);
	}

	machclk_per_tick = machclk_freq / hz;

#ifdef ALTQ_DEBUG
	printf("altq: CPU clock: %uHz\n", machclk_freq);
#endif
}

#ifdef __alpha__
/*
 * make a 64bit counter value out of the 32bit alpha processor cycle counter.
 * read_machclk must be called within a half of its wrap-around cycle
 * (about 5 sec for 400MHz cpu) to properly detect a counter wrap-around.
 * tbr_timeout calls read_machclk once a second.
 */
u_int64_t
read_machclk(void)
{
	static u_int32_t last_pcc, upper;
	u_int32_t pcc;

	pcc = (u_int32_t)alpha_rpcc();
	if (pcc <= last_pcc)
		upper++;
	last_pcc = pcc;
	return (((u_int64_t)upper << 32) + pcc);
}
#endif /* __alpha__ */
#else /* !i386  && !alpha */
/* use microtime() for now */
void
init_machclk(void)
{
	machclk_freq = 1000000 << MACHCLK_SHIFT;
	machclk_per_tick = machclk_freq / hz;
	printf("altq: emulate %uHz cpu clock\n", machclk_freq);
}
#endif /* !i386 && !alpha */