xref: /linux/net/sched/sch_tbf.c (revision ca55b2fef3a9373fcfc30f82fd26bc7fccbda732)
1 /*
2  * net/sched/sch_tbf.c	Token Bucket Filter queue.
3  *
4  *		This program is free software; you can redistribute it and/or
5  *		modify it under the terms of the GNU General Public License
6  *		as published by the Free Software Foundation; either version
7  *		2 of the License, or (at your option) any later version.
8  *
9  * Authors:	Alexey Kuznetsov, <kuznet@ms2.inr.ac.ru>
10  *		Dmitry Torokhov <dtor@mail.ru> - allow attaching inner qdiscs -
11  *						 original idea by Martin Devera
12  *
13  */
14 
15 #include <linux/module.h>
16 #include <linux/types.h>
17 #include <linux/kernel.h>
18 #include <linux/string.h>
19 #include <linux/errno.h>
20 #include <linux/skbuff.h>
21 #include <net/netlink.h>
22 #include <net/sch_generic.h>
23 #include <net/pkt_sched.h>
24 
25 
26 /*	Simple Token Bucket Filter.
27 	=======================================
28 
29 	SOURCE.
30 	-------
31 
32 	None.
33 
34 	Description.
35 	------------
36 
37 	A data flow obeys TBF with rate R and depth B, if for any
38 	time interval t_i...t_f the number of transmitted bits
39 	does not exceed B + R*(t_f-t_i).
40 
41 	Packetized version of this definition:
42 	The sequence of packets of sizes s_i served at moments t_i
43 	obeys TBF, if for any i<=k:
44 
45 	s_i+....+s_k <= B + R*(t_k - t_i)
46 
47 	Algorithm.
48 	----------
49 
50 	Let N(t_i) be B/R initially and N(t) grow continuously with time as:
51 
52 	N(t+delta) = min{B/R, N(t) + delta}
53 
54 	If the first packet in queue has length S, it may be
55 	transmitted only at the time t_* when S/R <= N(t_*),
56 	and in this case N(t) jumps:
57 
58 	N(t_* + 0) = N(t_* - 0) - S/R.
59 
60 
61 
62 	Actually, QoS requires two TBF to be applied to a data stream.
63 	One of them controls steady state burst size, another
64 	one with rate P (peak rate) and depth M (equal to link MTU)
65 	limits bursts at a smaller time scale.
66 
67 	It is easy to see that P>R, and B>M. If P is infinity, this double
68 	TBF is equivalent to a single one.
69 
70 	When TBF works in reshaping mode, latency is estimated as:
71 
72 	lat = max ((L-B)/R, (L-M)/P)
73 
74 
75 	NOTES.
76 	------
77 
78 	If TBF throttles, it starts a watchdog timer, which will wake it up
79 	when it is ready to transmit.
80 	Note that the minimal timer resolution is 1/HZ.
81 	If no new packets arrive during this period,
82 	or if the device is not awaken by EOI for some previous packet,
83 	TBF can stop its activity for 1/HZ.
84 
85 
86 	This means, that with depth B, the maximal rate is
87 
88 	R_crit = B*HZ
89 
90 	F.e. for 10Mbit ethernet and HZ=100 the minimal allowed B is ~10Kbytes.
91 
92 	Note that the peak rate TBF is much more tough: with MTU 1500
93 	P_crit = 150Kbytes/sec. So, if you need greater peak
94 	rates, use alpha with HZ=1000 :-)
95 
96 	With classful TBF, limit is just kept for backwards compatibility.
97 	It is passed to the default bfifo qdisc - if the inner qdisc is
98 	changed the limit is not effective anymore.
99 */
100 
101 struct tbf_sched_data {
102 /* Parameters */
103 	u32		limit;		/* Maximal length of backlog: bytes */
104 	u32		max_size;
105 	s64		buffer;		/* Token bucket depth/rate: MUST BE >= MTU/B */
106 	s64		mtu;
107 	struct psched_ratecfg rate;
108 	struct psched_ratecfg peak;
109 
110 /* Variables */
111 	s64	tokens;			/* Current number of B tokens */
112 	s64	ptokens;		/* Current number of P tokens */
113 	s64	t_c;			/* Time check-point */
114 	struct Qdisc	*qdisc;		/* Inner qdisc, default - bfifo queue */
115 	struct qdisc_watchdog watchdog;	/* Watchdog timer */
116 };
117 
118 
119 /* Time to Length, convert time in ns to length in bytes
120  * to determinate how many bytes can be sent in given time.
121  */
122 static u64 psched_ns_t2l(const struct psched_ratecfg *r,
123 			 u64 time_in_ns)
124 {
125 	/* The formula is :
126 	 * len = (time_in_ns * r->rate_bytes_ps) / NSEC_PER_SEC
127 	 */
128 	u64 len = time_in_ns * r->rate_bytes_ps;
129 
130 	do_div(len, NSEC_PER_SEC);
131 
132 	if (unlikely(r->linklayer == TC_LINKLAYER_ATM)) {
133 		do_div(len, 53);
134 		len = len * 48;
135 	}
136 
137 	if (len > r->overhead)
138 		len -= r->overhead;
139 	else
140 		len = 0;
141 
142 	return len;
143 }
144 
145 /*
146  * Return length of individual segments of a gso packet,
147  * including all headers (MAC, IP, TCP/UDP)
148  */
149 static unsigned int skb_gso_mac_seglen(const struct sk_buff *skb)
150 {
151 	unsigned int hdr_len = skb_transport_header(skb) - skb_mac_header(skb);
152 	return hdr_len + skb_gso_transport_seglen(skb);
153 }
154 
155 /* GSO packet is too big, segment it so that tbf can transmit
156  * each segment in time
157  */
158 static int tbf_segment(struct sk_buff *skb, struct Qdisc *sch)
159 {
160 	struct tbf_sched_data *q = qdisc_priv(sch);
161 	struct sk_buff *segs, *nskb;
162 	netdev_features_t features = netif_skb_features(skb);
163 	int ret, nb;
164 
165 	segs = skb_gso_segment(skb, features & ~NETIF_F_GSO_MASK);
166 
167 	if (IS_ERR_OR_NULL(segs))
168 		return qdisc_reshape_fail(skb, sch);
169 
170 	nb = 0;
171 	while (segs) {
172 		nskb = segs->next;
173 		segs->next = NULL;
174 		qdisc_skb_cb(segs)->pkt_len = segs->len;
175 		ret = qdisc_enqueue(segs, q->qdisc);
176 		if (ret != NET_XMIT_SUCCESS) {
177 			if (net_xmit_drop_count(ret))
178 				qdisc_qstats_drop(sch);
179 		} else {
180 			nb++;
181 		}
182 		segs = nskb;
183 	}
184 	sch->q.qlen += nb;
185 	if (nb > 1)
186 		qdisc_tree_decrease_qlen(sch, 1 - nb);
187 	consume_skb(skb);
188 	return nb > 0 ? NET_XMIT_SUCCESS : NET_XMIT_DROP;
189 }
190 
191 static int tbf_enqueue(struct sk_buff *skb, struct Qdisc *sch)
192 {
193 	struct tbf_sched_data *q = qdisc_priv(sch);
194 	int ret;
195 
196 	if (qdisc_pkt_len(skb) > q->max_size) {
197 		if (skb_is_gso(skb) && skb_gso_mac_seglen(skb) <= q->max_size)
198 			return tbf_segment(skb, sch);
199 		return qdisc_reshape_fail(skb, sch);
200 	}
201 	ret = qdisc_enqueue(skb, q->qdisc);
202 	if (ret != NET_XMIT_SUCCESS) {
203 		if (net_xmit_drop_count(ret))
204 			qdisc_qstats_drop(sch);
205 		return ret;
206 	}
207 
208 	sch->q.qlen++;
209 	return NET_XMIT_SUCCESS;
210 }
211 
212 static unsigned int tbf_drop(struct Qdisc *sch)
213 {
214 	struct tbf_sched_data *q = qdisc_priv(sch);
215 	unsigned int len = 0;
216 
217 	if (q->qdisc->ops->drop && (len = q->qdisc->ops->drop(q->qdisc)) != 0) {
218 		sch->q.qlen--;
219 		qdisc_qstats_drop(sch);
220 	}
221 	return len;
222 }
223 
224 static bool tbf_peak_present(const struct tbf_sched_data *q)
225 {
226 	return q->peak.rate_bytes_ps;
227 }
228 
229 static struct sk_buff *tbf_dequeue(struct Qdisc *sch)
230 {
231 	struct tbf_sched_data *q = qdisc_priv(sch);
232 	struct sk_buff *skb;
233 
234 	skb = q->qdisc->ops->peek(q->qdisc);
235 
236 	if (skb) {
237 		s64 now;
238 		s64 toks;
239 		s64 ptoks = 0;
240 		unsigned int len = qdisc_pkt_len(skb);
241 
242 		now = ktime_get_ns();
243 		toks = min_t(s64, now - q->t_c, q->buffer);
244 
245 		if (tbf_peak_present(q)) {
246 			ptoks = toks + q->ptokens;
247 			if (ptoks > q->mtu)
248 				ptoks = q->mtu;
249 			ptoks -= (s64) psched_l2t_ns(&q->peak, len);
250 		}
251 		toks += q->tokens;
252 		if (toks > q->buffer)
253 			toks = q->buffer;
254 		toks -= (s64) psched_l2t_ns(&q->rate, len);
255 
256 		if ((toks|ptoks) >= 0) {
257 			skb = qdisc_dequeue_peeked(q->qdisc);
258 			if (unlikely(!skb))
259 				return NULL;
260 
261 			q->t_c = now;
262 			q->tokens = toks;
263 			q->ptokens = ptoks;
264 			sch->q.qlen--;
265 			qdisc_unthrottled(sch);
266 			qdisc_bstats_update(sch, skb);
267 			return skb;
268 		}
269 
270 		qdisc_watchdog_schedule_ns(&q->watchdog,
271 					   now + max_t(long, -toks, -ptoks),
272 					   true);
273 
274 		/* Maybe we have a shorter packet in the queue,
275 		   which can be sent now. It sounds cool,
276 		   but, however, this is wrong in principle.
277 		   We MUST NOT reorder packets under these circumstances.
278 
279 		   Really, if we split the flow into independent
280 		   subflows, it would be a very good solution.
281 		   This is the main idea of all FQ algorithms
282 		   (cf. CSZ, HPFQ, HFSC)
283 		 */
284 
285 		qdisc_qstats_overlimit(sch);
286 	}
287 	return NULL;
288 }
289 
290 static void tbf_reset(struct Qdisc *sch)
291 {
292 	struct tbf_sched_data *q = qdisc_priv(sch);
293 
294 	qdisc_reset(q->qdisc);
295 	sch->q.qlen = 0;
296 	q->t_c = ktime_get_ns();
297 	q->tokens = q->buffer;
298 	q->ptokens = q->mtu;
299 	qdisc_watchdog_cancel(&q->watchdog);
300 }
301 
302 static const struct nla_policy tbf_policy[TCA_TBF_MAX + 1] = {
303 	[TCA_TBF_PARMS]	= { .len = sizeof(struct tc_tbf_qopt) },
304 	[TCA_TBF_RTAB]	= { .type = NLA_BINARY, .len = TC_RTAB_SIZE },
305 	[TCA_TBF_PTAB]	= { .type = NLA_BINARY, .len = TC_RTAB_SIZE },
306 	[TCA_TBF_RATE64]	= { .type = NLA_U64 },
307 	[TCA_TBF_PRATE64]	= { .type = NLA_U64 },
308 	[TCA_TBF_BURST] = { .type = NLA_U32 },
309 	[TCA_TBF_PBURST] = { .type = NLA_U32 },
310 };
311 
312 static int tbf_change(struct Qdisc *sch, struct nlattr *opt)
313 {
314 	int err;
315 	struct tbf_sched_data *q = qdisc_priv(sch);
316 	struct nlattr *tb[TCA_TBF_MAX + 1];
317 	struct tc_tbf_qopt *qopt;
318 	struct Qdisc *child = NULL;
319 	struct psched_ratecfg rate;
320 	struct psched_ratecfg peak;
321 	u64 max_size;
322 	s64 buffer, mtu;
323 	u64 rate64 = 0, prate64 = 0;
324 
325 	err = nla_parse_nested(tb, TCA_TBF_MAX, opt, tbf_policy);
326 	if (err < 0)
327 		return err;
328 
329 	err = -EINVAL;
330 	if (tb[TCA_TBF_PARMS] == NULL)
331 		goto done;
332 
333 	qopt = nla_data(tb[TCA_TBF_PARMS]);
334 	if (qopt->rate.linklayer == TC_LINKLAYER_UNAWARE)
335 		qdisc_put_rtab(qdisc_get_rtab(&qopt->rate,
336 					      tb[TCA_TBF_RTAB]));
337 
338 	if (qopt->peakrate.linklayer == TC_LINKLAYER_UNAWARE)
339 			qdisc_put_rtab(qdisc_get_rtab(&qopt->peakrate,
340 						      tb[TCA_TBF_PTAB]));
341 
342 	buffer = min_t(u64, PSCHED_TICKS2NS(qopt->buffer), ~0U);
343 	mtu = min_t(u64, PSCHED_TICKS2NS(qopt->mtu), ~0U);
344 
345 	if (tb[TCA_TBF_RATE64])
346 		rate64 = nla_get_u64(tb[TCA_TBF_RATE64]);
347 	psched_ratecfg_precompute(&rate, &qopt->rate, rate64);
348 
349 	if (tb[TCA_TBF_BURST]) {
350 		max_size = nla_get_u32(tb[TCA_TBF_BURST]);
351 		buffer = psched_l2t_ns(&rate, max_size);
352 	} else {
353 		max_size = min_t(u64, psched_ns_t2l(&rate, buffer), ~0U);
354 	}
355 
356 	if (qopt->peakrate.rate) {
357 		if (tb[TCA_TBF_PRATE64])
358 			prate64 = nla_get_u64(tb[TCA_TBF_PRATE64]);
359 		psched_ratecfg_precompute(&peak, &qopt->peakrate, prate64);
360 		if (peak.rate_bytes_ps <= rate.rate_bytes_ps) {
361 			pr_warn_ratelimited("sch_tbf: peakrate %llu is lower than or equals to rate %llu !\n",
362 					peak.rate_bytes_ps, rate.rate_bytes_ps);
363 			err = -EINVAL;
364 			goto done;
365 		}
366 
367 		if (tb[TCA_TBF_PBURST]) {
368 			u32 pburst = nla_get_u32(tb[TCA_TBF_PBURST]);
369 			max_size = min_t(u32, max_size, pburst);
370 			mtu = psched_l2t_ns(&peak, pburst);
371 		} else {
372 			max_size = min_t(u64, max_size, psched_ns_t2l(&peak, mtu));
373 		}
374 	} else {
375 		memset(&peak, 0, sizeof(peak));
376 	}
377 
378 	if (max_size < psched_mtu(qdisc_dev(sch)))
379 		pr_warn_ratelimited("sch_tbf: burst %llu is lower than device %s mtu (%u) !\n",
380 				    max_size, qdisc_dev(sch)->name,
381 				    psched_mtu(qdisc_dev(sch)));
382 
383 	if (!max_size) {
384 		err = -EINVAL;
385 		goto done;
386 	}
387 
388 	if (q->qdisc != &noop_qdisc) {
389 		err = fifo_set_limit(q->qdisc, qopt->limit);
390 		if (err)
391 			goto done;
392 	} else if (qopt->limit > 0) {
393 		child = fifo_create_dflt(sch, &bfifo_qdisc_ops, qopt->limit);
394 		if (IS_ERR(child)) {
395 			err = PTR_ERR(child);
396 			goto done;
397 		}
398 	}
399 
400 	sch_tree_lock(sch);
401 	if (child) {
402 		qdisc_tree_decrease_qlen(q->qdisc, q->qdisc->q.qlen);
403 		qdisc_destroy(q->qdisc);
404 		q->qdisc = child;
405 	}
406 	q->limit = qopt->limit;
407 	if (tb[TCA_TBF_PBURST])
408 		q->mtu = mtu;
409 	else
410 		q->mtu = PSCHED_TICKS2NS(qopt->mtu);
411 	q->max_size = max_size;
412 	if (tb[TCA_TBF_BURST])
413 		q->buffer = buffer;
414 	else
415 		q->buffer = PSCHED_TICKS2NS(qopt->buffer);
416 	q->tokens = q->buffer;
417 	q->ptokens = q->mtu;
418 
419 	memcpy(&q->rate, &rate, sizeof(struct psched_ratecfg));
420 	memcpy(&q->peak, &peak, sizeof(struct psched_ratecfg));
421 
422 	sch_tree_unlock(sch);
423 	err = 0;
424 done:
425 	return err;
426 }
427 
428 static int tbf_init(struct Qdisc *sch, struct nlattr *opt)
429 {
430 	struct tbf_sched_data *q = qdisc_priv(sch);
431 
432 	if (opt == NULL)
433 		return -EINVAL;
434 
435 	q->t_c = ktime_get_ns();
436 	qdisc_watchdog_init(&q->watchdog, sch);
437 	q->qdisc = &noop_qdisc;
438 
439 	return tbf_change(sch, opt);
440 }
441 
442 static void tbf_destroy(struct Qdisc *sch)
443 {
444 	struct tbf_sched_data *q = qdisc_priv(sch);
445 
446 	qdisc_watchdog_cancel(&q->watchdog);
447 	qdisc_destroy(q->qdisc);
448 }
449 
450 static int tbf_dump(struct Qdisc *sch, struct sk_buff *skb)
451 {
452 	struct tbf_sched_data *q = qdisc_priv(sch);
453 	struct nlattr *nest;
454 	struct tc_tbf_qopt opt;
455 
456 	sch->qstats.backlog = q->qdisc->qstats.backlog;
457 	nest = nla_nest_start(skb, TCA_OPTIONS);
458 	if (nest == NULL)
459 		goto nla_put_failure;
460 
461 	opt.limit = q->limit;
462 	psched_ratecfg_getrate(&opt.rate, &q->rate);
463 	if (tbf_peak_present(q))
464 		psched_ratecfg_getrate(&opt.peakrate, &q->peak);
465 	else
466 		memset(&opt.peakrate, 0, sizeof(opt.peakrate));
467 	opt.mtu = PSCHED_NS2TICKS(q->mtu);
468 	opt.buffer = PSCHED_NS2TICKS(q->buffer);
469 	if (nla_put(skb, TCA_TBF_PARMS, sizeof(opt), &opt))
470 		goto nla_put_failure;
471 	if (q->rate.rate_bytes_ps >= (1ULL << 32) &&
472 	    nla_put_u64(skb, TCA_TBF_RATE64, q->rate.rate_bytes_ps))
473 		goto nla_put_failure;
474 	if (tbf_peak_present(q) &&
475 	    q->peak.rate_bytes_ps >= (1ULL << 32) &&
476 	    nla_put_u64(skb, TCA_TBF_PRATE64, q->peak.rate_bytes_ps))
477 		goto nla_put_failure;
478 
479 	return nla_nest_end(skb, nest);
480 
481 nla_put_failure:
482 	nla_nest_cancel(skb, nest);
483 	return -1;
484 }
485 
486 static int tbf_dump_class(struct Qdisc *sch, unsigned long cl,
487 			  struct sk_buff *skb, struct tcmsg *tcm)
488 {
489 	struct tbf_sched_data *q = qdisc_priv(sch);
490 
491 	tcm->tcm_handle |= TC_H_MIN(1);
492 	tcm->tcm_info = q->qdisc->handle;
493 
494 	return 0;
495 }
496 
497 static int tbf_graft(struct Qdisc *sch, unsigned long arg, struct Qdisc *new,
498 		     struct Qdisc **old)
499 {
500 	struct tbf_sched_data *q = qdisc_priv(sch);
501 
502 	if (new == NULL)
503 		new = &noop_qdisc;
504 
505 	sch_tree_lock(sch);
506 	*old = q->qdisc;
507 	q->qdisc = new;
508 	qdisc_tree_decrease_qlen(*old, (*old)->q.qlen);
509 	qdisc_reset(*old);
510 	sch_tree_unlock(sch);
511 
512 	return 0;
513 }
514 
515 static struct Qdisc *tbf_leaf(struct Qdisc *sch, unsigned long arg)
516 {
517 	struct tbf_sched_data *q = qdisc_priv(sch);
518 	return q->qdisc;
519 }
520 
521 static unsigned long tbf_get(struct Qdisc *sch, u32 classid)
522 {
523 	return 1;
524 }
525 
526 static void tbf_put(struct Qdisc *sch, unsigned long arg)
527 {
528 }
529 
530 static void tbf_walk(struct Qdisc *sch, struct qdisc_walker *walker)
531 {
532 	if (!walker->stop) {
533 		if (walker->count >= walker->skip)
534 			if (walker->fn(sch, 1, walker) < 0) {
535 				walker->stop = 1;
536 				return;
537 			}
538 		walker->count++;
539 	}
540 }
541 
542 static const struct Qdisc_class_ops tbf_class_ops = {
543 	.graft		=	tbf_graft,
544 	.leaf		=	tbf_leaf,
545 	.get		=	tbf_get,
546 	.put		=	tbf_put,
547 	.walk		=	tbf_walk,
548 	.dump		=	tbf_dump_class,
549 };
550 
551 static struct Qdisc_ops tbf_qdisc_ops __read_mostly = {
552 	.next		=	NULL,
553 	.cl_ops		=	&tbf_class_ops,
554 	.id		=	"tbf",
555 	.priv_size	=	sizeof(struct tbf_sched_data),
556 	.enqueue	=	tbf_enqueue,
557 	.dequeue	=	tbf_dequeue,
558 	.peek		=	qdisc_peek_dequeued,
559 	.drop		=	tbf_drop,
560 	.init		=	tbf_init,
561 	.reset		=	tbf_reset,
562 	.destroy	=	tbf_destroy,
563 	.change		=	tbf_change,
564 	.dump		=	tbf_dump,
565 	.owner		=	THIS_MODULE,
566 };
567 
568 static int __init tbf_module_init(void)
569 {
570 	return register_qdisc(&tbf_qdisc_ops);
571 }
572 
573 static void __exit tbf_module_exit(void)
574 {
575 	unregister_qdisc(&tbf_qdisc_ops);
576 }
577 module_init(tbf_module_init)
578 module_exit(tbf_module_exit)
579 MODULE_LICENSE("GPL");
580