summaryrefslogtreecommitdiffstats
path: root/net/sched/cls_rsvp.h
blob: 61286a0f7a3e9bb23ecc0a46014193976bdb637c (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
/*
 * net/sched/cls_rsvp.h	Template file for RSVPv[46] classifiers.
 *
 *		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.
 *
 * Authors:	Alexey Kuznetsov, <kuznet@ms2.inr.ac.ru>
 */

/*
   Comparing to general packet classification problem,
   RSVP needs only sevaral relatively simple rules:

   * (dst, protocol) are always specified,
     so that we are able to hash them.
   * src may be exact, or may be wildcard, so that
     we can keep a hash table plus one wildcard entry.
   * source port (or flow label) is important only if src is given.

   IMPLEMENTATION.

   We use a two level hash table: The top level is keyed by
   destination address and protocol ID, every bucket contains a list
   of "rsvp sessions", identified by destination address, protocol and
   DPI(="Destination Port ID"): triple (key, mask, offset).

   Every bucket has a smaller hash table keyed by source address
   (cf. RSVP flowspec) and one wildcard entry for wildcard reservations.
   Every bucket is again a list of "RSVP flows", selected by
   source address and SPI(="Source Port ID" here rather than
   "security parameter index"): triple (key, mask, offset).


   NOTE 1. All the packets with IPv6 extension headers (but AH and ESP)
   and all fragmented packets go to the best-effort traffic class.


   NOTE 2. Two "port id"'s seems to be redundant, rfc2207 requires
   only one "Generalized Port Identifier". So that for classic
   ah, esp (and udp,tcp) both *pi should coincide or one of them
   should be wildcard.

   At first sight, this redundancy is just a waste of CPU
   resources. But DPI and SPI add the possibility to assign different
   priorities to GPIs. Look also at note 4 about tunnels below.


   NOTE 3. One complication is the case of tunneled packets.
   We implement it as following: if the first lookup
   matches a special session with "tunnelhdr" value not zero,
   flowid doesn't contain the true flow ID, but the tunnel ID (1...255).
   In this case, we pull tunnelhdr bytes and restart lookup
   with tunnel ID added to the list of keys. Simple and stupid 8)8)
   It's enough for PIMREG and IPIP.


   NOTE 4. Two GPIs make it possible to parse even GRE packets.
   F.e. DPI can select ETH_P_IP (and necessary flags to make
   tunnelhdr correct) in GRE protocol field and SPI matches
   GRE key. Is it not nice? 8)8)


   Well, as result, despite its simplicity, we get a pretty
   powerful classification engine.  */


struct rsvp_head
{
	u32			tmap[256/32];
	u32			hgenerator;
	u8			tgenerator;
	struct rsvp_session	*ht[256];
};

struct rsvp_session
{
	struct rsvp_session	*next;
	__be32			dst[RSVP_DST_LEN];
	struct tc_rsvp_gpi 	dpi;
	u8			protocol;
	u8			tunnelid;
	/* 16 (src,sport) hash slots, and one wildcard source slot */
	struct rsvp_filter	*ht[16+1];
};


struct rsvp_filter
{
	struct rsvp_filter	*next;
	__be32			src[RSVP_DST_LEN];
	struct tc_rsvp_gpi	spi;
	u8			tunnelhdr;

	struct tcf_result	res;
	struct tcf_exts		exts;

	u32			handle;
	struct rsvp_session	*sess;
};

static __inline__ unsigned hash_dst(__be32 *dst, u8 protocol, u8 tunnelid)
{
	unsigned h = (__force __u32)dst[RSVP_DST_LEN-1];
	h ^= h>>16;
	h ^= h>>8;
	return (h ^ protocol ^ tunnelid) & 0xFF;
}

static __inline__ unsigned hash_src(__be32 *src)
{
	unsigned h = (__force __u32)src[RSVP_DST_LEN-1];
	h ^= h>>16;
	h ^= h>>8;
	h ^= h>>4;
	return h & 0xF;
}

static struct tcf_ext_map rsvp_ext_map = {
	.police = TCA_RSVP_POLICE,
	.action = TCA_RSVP_ACT
};

#define RSVP_APPLY_RESULT()				\
{							\
	int r = tcf_exts_exec(skb, &f->exts, res);	\
	if (r < 0)					\
		continue;				\
	else if (r > 0)					\
		return r;				\
}

static int rsvp_classify(struct sk_buff *skb, struct tcf_proto *tp,
			 struct tcf_result *res)
{
	struct rsvp_session **sht = ((struct rsvp_head*)tp->root)->ht;
	struct rsvp_session *s;
	struct rsvp_filter *f;
	unsigned h1, h2;
	__be32 *dst, *src;
	u8 protocol;
	u8 tunnelid = 0;
	u8 *xprt;
#if RSVP_DST_LEN == 4
	struct ipv6hdr *nhptr = ipv6_hdr(skb);
#else
	struct iphdr *nhptr = ip_hdr(skb);
#endif

restart:

#if RSVP_DST_LEN == 4
	src = &nhptr->saddr.s6_addr32[0];
	dst = &nhptr->daddr.s6_addr32[0];
	protocol = nhptr->nexthdr;
	xprt = ((u8*)nhptr) + sizeof(struct ipv6hdr);
#else
	src = &nhptr->saddr;
	dst = &nhptr->daddr;
	protocol = nhptr->protocol;
	xprt = ((u8*)nhptr) + (nhptr->ihl<<2);
	if (nhptr->frag_off & htons(IP_MF|IP_OFFSET))
		return -1;
#endif

	h1 = hash_dst(dst, protocol, tunnelid);
	h2 = hash_src(src);

	for (s = sht[h1]; s; s = s->next) {
		if (dst[RSVP_DST_LEN-1] == s->dst[RSVP_DST_LEN-1] &&
		    protocol == s->protocol &&
		    !(s->dpi.mask & (*(u32*)(xprt+s->dpi.offset)^s->dpi.key))
#if RSVP_DST_LEN == 4
		    && dst[0] == s->dst[0]
		    && dst[1] == s->dst[1]
		    && dst[2] == s->dst[2]
#endif
		    && tunnelid == s->tunnelid) {

			for (f = s->ht[h2]; f; f = f->next) {
				if (src[RSVP_DST_LEN-1] == f->src[RSVP_DST_LEN-1] &&
				    !(f->spi.mask & (*(u32*)(xprt+f->spi.offset)^f->spi.key))
#if RSVP_DST_LEN == 4
				    && src[0] == f->src[0]
				    && src[1] == f->src[1]
				    && src[2] == f->src[2]
#endif
				    ) {
					*res = f->res;
					RSVP_APPLY_RESULT();

matched:
					if (f->tunnelhdr == 0)
						return 0;

					tunnelid = f->res.classid;
					nhptr = (void*)(xprt + f->tunnelhdr - sizeof(*nhptr));
					goto restart;
				}
			}

			/* And wildcard bucket... */
			for (f = s->ht[16]; f; f = f->next) {
				*res = f->res;
				RSVP_APPLY_RESULT();
				goto matched;
			}
			return -1;
		}
	}
	return -1;
}

static unsigned long rsvp_get(struct tcf_proto *tp, u32 handle)
{
	struct rsvp_session **sht = ((struct rsvp_head*)tp->root)->ht;
	struct rsvp_session *s;
	struct rsvp_filter *f;
	unsigned h1 = handle&0xFF;
	unsigned h2 = (handle>>8)&0xFF;

	if (h2 > 16)
		return 0;

	for (s = sht[h1]; s; s = s->next) {
		for (f = s->ht[h2]; f; f = f->next) {
			if (f->handle == handle)
				return (unsigned long)f;
		}
	}
	return 0;
}

static void rsvp_put(struct tcf_proto *tp, unsigned long f)
{
}

static int rsvp_init(struct tcf_proto *tp)
{
	struct rsvp_head *data;

	data = kzalloc(sizeof(struct rsvp_head), GFP_KERNEL);
	if (data) {
		tp->root = data;
		return 0;
	}
	return -ENOBUFS;
}

static inline void
rsvp_delete_filter(struct tcf_proto *tp, struct rsvp_filter *f)
{
	tcf_unbind_filter(tp, &f->res);
	tcf_exts_destroy(tp, &f->exts);
	kfree(f);
}

static void rsvp_destroy(struct tcf_proto *tp)
{
	struct rsvp_head *data = xchg(&tp->root, NULL);
	struct rsvp_session **sht;
	int h1, h2;

	if (data == NULL)
		return;

	sht = data->ht;

	for (h1=0; h1<256; h1++) {
		struct rsvp_session *s;

		while ((s = sht[h1]) != NULL) {
			sht[h1] = s->next;

			for (h2=0; h2<=16; h2++) {
				struct rsvp_filter *f;

				while ((f = s->ht[h2]) != NULL) {
					s->ht[h2] = f->next;
					rsvp_delete_filter(tp, f);
				}
			}
			kfree(s);
		}
	}
	kfree(data);
}

static int rsvp_delete(struct tcf_proto *tp, unsigned long arg)
{
	struct rsvp_filter **fp, *f = (struct rsvp_filter*)arg;
	unsigned h = f->handle;
	struct rsvp_session **sp;
	struct rsvp_session *s = f->sess;
	int i;

	for (fp = &s->ht[(h>>8)&0xFF]; *fp; fp = &(*fp)->next) {
		if (*fp == f) {
			tcf_tree_lock(tp);
			*fp = f->next;
			tcf_tree_unlock(tp);
			rsvp_delete_filter(tp, f);

			/* Strip tree */

			for (i=0; i<=16; i++)
				if (s->ht[i])
					return 0;

			/* OK, session has no flows */
			for (sp = &((struct rsvp_head*)tp->root)->ht[h&0xFF];
			     *sp; sp = &(*sp)->next) {
				if (*sp == s) {
					tcf_tree_lock(tp);
					*sp = s->next;
					tcf_tree_unlock(tp);

					kfree(s);
					return 0;
				}
			}

			return 0;
		}
	}
	return 0;
}

static unsigned gen_handle(struct tcf_proto *tp, unsigned salt)
{
	struct rsvp_head *data = tp->root;
	int i = 0xFFFF;

	while (i-- > 0) {
		u32 h;
		if ((data->hgenerator += 0x10000) == 0)
			data->hgenerator = 0x10000;
		h = data->hgenerator|salt;
		if (rsvp_get(tp, h) == 0)
			return h;
	}
	return 0;
}

static int tunnel_bts(struct rsvp_head *data)
{
	int n = data->tgenerator>>5;
	u32 b = 1<<(data->tgenerator&0x1F);

	if (data->tmap[n]&b)
		return 0;
	data->tmap[n] |= b;
	return 1;
}

static void tunnel_recycle(struct rsvp_head *data)
{
	struct rsvp_session **sht = data->ht;
	u32 tmap[256/32];
	int h1, h2;

	memset(tmap, 0, sizeof(tmap));

	for (h1=0; h1<256; h1++) {
		struct rsvp_session *s;
		for (s = sht[h1]; s; s = s->next) {
			for (h2=0; h2<=16; h2++) {
				struct rsvp_filter *f;

				for (f = s->ht[h2]; f; f = f->next) {
					if (f->tunnelhdr == 0)
						continue;
					data->tgenerator = f->res.classid;
					tunnel_bts(data);
				}
			}
		}
	}

	memcpy(data->tmap, tmap, sizeof(tmap));
}

static u32 gen_tunnel(struct rsvp_head *data)
{
	int i, k;

	for (k=0; k<2; k++) {
		for (i=255; i>0; i--) {
			if (++data->tgenerator == 0)
				data->tgenerator = 1;
			if (tunnel_bts(data))
				return data->tgenerator;
		}
		tunnel_recycle(data);
	}
	return 0;
}

static int rsvp_change(struct tcf_proto *tp, unsigned long base,
		       u32 handle,
		       struct nlattr **tca,
		       unsigned long *arg)
{
	struct rsvp_head *data = tp->root;
	struct rsvp_filter *f, **fp;
	struct rsvp_session *s, **sp;
	struct tc_rsvp_pinfo *pinfo = NULL;
	struct nlattr *opt = tca[TCA_OPTIONS-1];
	struct nlattr *tb[TCA_RSVP_MAX + 1];
	struct tcf_exts e;
	unsigned h1, h2;
	__be32 *dst;
	int err;

	if (opt == NULL)
		return handle ? -EINVAL : 0;

	err = nla_parse_nested(tb, TCA_RSVP_MAX, opt, NULL);
	if (err < 0)
		return err;

	err = tcf_exts_validate(tp, tb, tca[TCA_RATE-1], &e, &rsvp_ext_map);
	if (err < 0)
		return err;

	if ((f = (struct rsvp_filter*)*arg) != NULL) {
		/* Node exists: adjust only classid */

		if (f->handle != handle && handle)
			goto errout2;
		if (tb[TCA_RSVP_CLASSID-1]) {
			f->res.classid = nla_get_u32(tb[TCA_RSVP_CLASSID-1]);
			tcf_bind_filter(tp, &f->res, base);
		}

		tcf_exts_change(tp, &f->exts, &e);
		return 0;
	}

	/* Now more serious part... */
	err = -EINVAL;
	if (handle)
		goto errout2;
	if (tb[TCA_RSVP_DST-1] == NULL)
		goto errout2;

	err = -ENOBUFS;
	f = kzalloc(sizeof(struct rsvp_filter), GFP_KERNEL);
	if (f == NULL)
		goto errout2;

	h2 = 16;
	if (tb[TCA_RSVP_SRC-1]) {
		err = -EINVAL;
		if (nla_len(tb[TCA_RSVP_SRC-1]) != sizeof(f->src))
			goto errout;
		memcpy(f->src, nla_data(tb[TCA_RSVP_SRC-1]), sizeof(f->src));
		h2 = hash_src(f->src);
	}
	if (tb[TCA_RSVP_PINFO-1]) {
		err = -EINVAL;
		if (nla_len(tb[TCA_RSVP_PINFO-1]) < sizeof(struct tc_rsvp_pinfo))
			goto errout;
		pinfo = nla_data(tb[TCA_RSVP_PINFO-1]);
		f->spi = pinfo->spi;
		f->tunnelhdr = pinfo->tunnelhdr;
	}
	if (tb[TCA_RSVP_CLASSID-1]) {
		err = -EINVAL;
		if (nla_len(tb[TCA_RSVP_CLASSID-1]) != 4)
			goto errout;
		f->res.classid = nla_get_u32(tb[TCA_RSVP_CLASSID-1]);
	}

	err = -EINVAL;
	if (nla_len(tb[TCA_RSVP_DST-1]) != sizeof(f->src))
		goto errout;
	dst = nla_data(tb[TCA_RSVP_DST-1]);
	h1 = hash_dst(dst, pinfo ? pinfo->protocol : 0, pinfo ? pinfo->tunnelid : 0);

	err = -ENOMEM;
	if ((f->handle = gen_handle(tp, h1 | (h2<<8))) == 0)
		goto errout;

	if (f->tunnelhdr) {
		err = -EINVAL;
		if (f->res.classid > 255)
			goto errout;

		err = -ENOMEM;
		if (f->res.classid == 0 &&
		    (f->res.classid = gen_tunnel(data)) == 0)
			goto errout;
	}

	for (sp = &data->ht[h1]; (s=*sp) != NULL; sp = &s->next) {
		if (dst[RSVP_DST_LEN-1] == s->dst[RSVP_DST_LEN-1] &&
		    pinfo && pinfo->protocol == s->protocol &&
		    memcmp(&pinfo->dpi, &s->dpi, sizeof(s->dpi)) == 0
#if RSVP_DST_LEN == 4
		    && dst[0] == s->dst[0]
		    && dst[1] == s->dst[1]
		    && dst[2] == s->dst[2]
#endif
		    && pinfo->tunnelid == s->tunnelid) {

insert:
			/* OK, we found appropriate session */

			fp = &s->ht[h2];

			f->sess = s;
			if (f->tunnelhdr == 0)
				tcf_bind_filter(tp, &f->res, base);

			tcf_exts_change(tp, &f->exts, &e);

			for (fp = &s->ht[h2]; *fp; fp = &(*fp)->next)
				if (((*fp)->spi.mask&f->spi.mask) != f->spi.mask)
					break;
			f->next = *fp;
			wmb();
			*fp = f;

			*arg = (unsigned long)f;
			return 0;
		}
	}

	/* No session found. Create new one. */

	err = -ENOBUFS;
	s = kzalloc(sizeof(struct rsvp_session), GFP_KERNEL);
	if (s == NULL)
		goto errout;
	memcpy(s->dst, dst, sizeof(s->dst));

	if (pinfo) {
		s->dpi = pinfo->dpi;
		s->protocol = pinfo->protocol;
		s->tunnelid = pinfo->tunnelid;
	}
	for (sp = &data->ht[h1]; *sp; sp = &(*sp)->next) {
		if (((*sp)->dpi.mask&s->dpi.mask) != s->dpi.mask)
			break;
	}
	s->next = *sp;
	wmb();
	*sp = s;

	goto insert;

errout:
	kfree(f);
errout2:
	tcf_exts_destroy(tp, &e);
	return err;
}

static void rsvp_walk(struct tcf_proto *tp, struct tcf_walker *arg)
{
	struct rsvp_head *head = tp->root;
	unsigned h, h1;

	if (arg->stop)
		return;

	for (h = 0; h < 256; h++) {
		struct rsvp_session *s;

		for (s = head->ht[h]; s; s = s->next) {
			for (h1 = 0; h1 <= 16; h1++) {
				struct rsvp_filter *f;

				for (f = s->ht[h1]; f; f = f->next) {
					if (arg->count < arg->skip) {
						arg->count++;
						continue;
					}
					if (arg->fn(tp, (unsigned long)f, arg) < 0) {
						arg->stop = 1;
						return;
					}
					arg->count++;
				}
			}
		}
	}
}

static int rsvp_dump(struct tcf_proto *tp, unsigned long fh,
		     struct sk_buff *skb, struct tcmsg *t)
{
	struct rsvp_filter *f = (struct rsvp_filter*)fh;
	struct rsvp_session *s;
	unsigned char *b = skb_tail_pointer(skb);
	struct nlattr *nest;
	struct tc_rsvp_pinfo pinfo;

	if (f == NULL)
		return skb->len;
	s = f->sess;

	t->tcm_handle = f->handle;

	nest = nla_nest_start(skb, TCA_OPTIONS);
	if (nest == NULL)
		goto nla_put_failure;

	NLA_PUT(skb, TCA_RSVP_DST, sizeof(s->dst), &s->dst);
	pinfo.dpi = s->dpi;
	pinfo.spi = f->spi;
	pinfo.protocol = s->protocol;
	pinfo.tunnelid = s->tunnelid;
	pinfo.tunnelhdr = f->tunnelhdr;
	pinfo.pad = 0;
	NLA_PUT(skb, TCA_RSVP_PINFO, sizeof(pinfo), &pinfo);
	if (f->res.classid)
		NLA_PUT_U32(skb, TCA_RSVP_CLASSID, f->res.classid);
	if (((f->handle>>8)&0xFF) != 16)
		NLA_PUT(skb, TCA_RSVP_SRC, sizeof(f->src), f->src);

	if (tcf_exts_dump(skb, &f->exts, &rsvp_ext_map) < 0)
		goto nla_put_failure;

	nla_nest_end(skb, nest);

	if (tcf_exts_dump_stats(skb, &f->exts, &rsvp_ext_map) < 0)
		goto nla_put_failure;
	return skb->len;

nla_put_failure:
	nlmsg_trim(skb, b);
	return -1;
}

static struct tcf_proto_ops RSVP_OPS = {
	.next		=	NULL,
	.kind		=	RSVP_ID,
	.classify	=	rsvp_classify,
	.init		=	rsvp_init,
	.destroy	=	rsvp_destroy,
	.get		=	rsvp_get,
	.put		=	rsvp_put,
	.change		=	rsvp_change,
	.delete		=	rsvp_delete,
	.walk		=	rsvp_walk,
	.dump		=	rsvp_dump,
	.owner		=	THIS_MODULE,
};

static int __init init_rsvp(void)
{
	return register_tcf_proto_ops(&RSVP_OPS);
}

static void __exit exit_rsvp(void)
{
	unregister_tcf_proto_ops(&RSVP_OPS);
}

module_init(init_rsvp)
module_exit(exit_rsvp)