summaryrefslogtreecommitdiffstats
path: root/net/sched/sch_red.c
blob: ee66c5ca80c6909b23281502bb3a29081cdbca12 (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
/*
 * net/sched/sch_red.c	Random Early Detection queue.
 *
 *		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>
 *
 * Changes:
 * J Hadi Salim 980914:	computation fixes
 * Alexey Makarenko <makar@phoenix.kharkov.ua> 990814: qave on idle link was calculated incorrectly.
 * J Hadi Salim 980816:  ECN support
 */

#include <linux/module.h>
#include <linux/types.h>
#include <linux/kernel.h>
#include <linux/netdevice.h>
#include <linux/skbuff.h>
#include <net/pkt_sched.h>
#include <net/inet_ecn.h>
#include <net/red.h>


/*	Parameters, settable by user:
	-----------------------------

	limit		- bytes (must be > qth_max + burst)

	Hard limit on queue length, should be chosen >qth_max
	to allow packet bursts. This parameter does not
	affect the algorithms behaviour and can be chosen
	arbitrarily high (well, less than ram size)
	Really, this limit will never be reached
	if RED works correctly.
 */

struct red_sched_data
{
	u32			limit;		/* HARD maximal queue length */
	unsigned char		flags;
	struct red_parms	parms;
	struct red_stats	stats;
	struct Qdisc		*qdisc;
};

static inline int red_use_ecn(struct red_sched_data *q)
{
	return q->flags & TC_RED_ECN;
}

static inline int red_use_harddrop(struct red_sched_data *q)
{
	return q->flags & TC_RED_HARDDROP;
}

static int red_enqueue(struct sk_buff *skb, struct Qdisc* sch)
{
	struct red_sched_data *q = qdisc_priv(sch);
	struct Qdisc *child = q->qdisc;
	int ret;

	q->parms.qavg = red_calc_qavg(&q->parms, child->qstats.backlog);

	if (red_is_idling(&q->parms))
		red_end_of_idle_period(&q->parms);

	switch (red_action(&q->parms, q->parms.qavg)) {
		case RED_DONT_MARK:
			break;

		case RED_PROB_MARK:
			sch->qstats.overlimits++;
			if (!red_use_ecn(q) || !INET_ECN_set_ce(skb)) {
				q->stats.prob_drop++;
				goto congestion_drop;
			}

			q->stats.prob_mark++;
			break;

		case RED_HARD_MARK:
			sch->qstats.overlimits++;
			if (red_use_harddrop(q) || !red_use_ecn(q) ||
			    !INET_ECN_set_ce(skb)) {
				q->stats.forced_drop++;
				goto congestion_drop;
			}

			q->stats.forced_mark++;
			break;
	}

	ret = child->enqueue(skb, child);
	if (likely(ret == NET_XMIT_SUCCESS)) {
		sch->bstats.bytes += skb->len;
		sch->bstats.packets++;
		sch->q.qlen++;
	} else {
		q->stats.pdrop++;
		sch->qstats.drops++;
	}
	return ret;

congestion_drop:
	qdisc_drop(skb, sch);
	return NET_XMIT_CN;
}

static int red_requeue(struct sk_buff *skb, struct Qdisc* sch)
{
	struct red_sched_data *q = qdisc_priv(sch);
	struct Qdisc *child = q->qdisc;
	int ret;

	if (red_is_idling(&q->parms))
		red_end_of_idle_period(&q->parms);

	ret = child->ops->requeue(skb, child);
	if (likely(ret == NET_XMIT_SUCCESS)) {
		sch->qstats.requeues++;
		sch->q.qlen++;
	}
	return ret;
}

static struct sk_buff * red_dequeue(struct Qdisc* sch)
{
	struct sk_buff *skb;
	struct red_sched_data *q = qdisc_priv(sch);
	struct Qdisc *child = q->qdisc;

	skb = child->dequeue(child);
	if (skb)
		sch->q.qlen--;
	else if (!red_is_idling(&q->parms))
		red_start_of_idle_period(&q->parms);

	return skb;
}

static unsigned int red_drop(struct Qdisc* sch)
{
	struct red_sched_data *q = qdisc_priv(sch);
	struct Qdisc *child = q->qdisc;
	unsigned int len;

	if (child->ops->drop && (len = child->ops->drop(child)) > 0) {
		q->stats.other++;
		sch->qstats.drops++;
		sch->q.qlen--;
		return len;
	}

	if (!red_is_idling(&q->parms))
		red_start_of_idle_period(&q->parms);

	return 0;
}

static void red_reset(struct Qdisc* sch)
{
	struct red_sched_data *q = qdisc_priv(sch);

	qdisc_reset(q->qdisc);
	sch->q.qlen = 0;
	red_restart(&q->parms);
}

static void red_destroy(struct Qdisc *sch)
{
	struct red_sched_data *q = qdisc_priv(sch);
	qdisc_destroy(q->qdisc);
}

static struct Qdisc *red_create_dflt(struct Qdisc *sch, u32 limit)
{
	struct Qdisc *q;
	struct rtattr *rta;
	int ret;

	q = qdisc_create_dflt(sch->dev, &bfifo_qdisc_ops,
			      TC_H_MAKE(sch->handle, 1));
	if (q) {
		rta = kmalloc(RTA_LENGTH(sizeof(struct tc_fifo_qopt)),
		              GFP_KERNEL);
		if (rta) {
			rta->rta_type = RTM_NEWQDISC;
			rta->rta_len = RTA_LENGTH(sizeof(struct tc_fifo_qopt));
			((struct tc_fifo_qopt *)RTA_DATA(rta))->limit = limit;

			ret = q->ops->change(q, rta);
			kfree(rta);

			if (ret == 0)
				return q;
		}
		qdisc_destroy(q);
	}
	return NULL;
}

static int red_change(struct Qdisc *sch, struct rtattr *opt)
{
	struct red_sched_data *q = qdisc_priv(sch);
	struct rtattr *tb[TCA_RED_MAX];
	struct tc_red_qopt *ctl;
	struct Qdisc *child = NULL;

	if (opt == NULL || rtattr_parse_nested(tb, TCA_RED_MAX, opt))
		return -EINVAL;

	if (tb[TCA_RED_PARMS-1] == NULL ||
	    RTA_PAYLOAD(tb[TCA_RED_PARMS-1]) < sizeof(*ctl) ||
	    tb[TCA_RED_STAB-1] == NULL ||
	    RTA_PAYLOAD(tb[TCA_RED_STAB-1]) < RED_STAB_SIZE)
		return -EINVAL;

	ctl = RTA_DATA(tb[TCA_RED_PARMS-1]);

	if (ctl->limit > 0) {
		child = red_create_dflt(sch, ctl->limit);
		if (child == NULL)
			return -ENOMEM;
	}

	sch_tree_lock(sch);
	q->flags = ctl->flags;
	q->limit = ctl->limit;
	if (child)
		qdisc_destroy(xchg(&q->qdisc, child));

	red_set_parms(&q->parms, ctl->qth_min, ctl->qth_max, ctl->Wlog,
				 ctl->Plog, ctl->Scell_log,
				 RTA_DATA(tb[TCA_RED_STAB-1]));

	if (skb_queue_empty(&sch->q))
		red_end_of_idle_period(&q->parms);

	sch_tree_unlock(sch);
	return 0;
}

static int red_init(struct Qdisc* sch, struct rtattr *opt)
{
	struct red_sched_data *q = qdisc_priv(sch);

	q->qdisc = &noop_qdisc;
	return red_change(sch, opt);
}

static int red_dump(struct Qdisc *sch, struct sk_buff *skb)
{
	struct red_sched_data *q = qdisc_priv(sch);
	struct rtattr *opts = NULL;
	struct tc_red_qopt opt = {
		.limit		= q->limit,
		.flags		= q->flags,
		.qth_min	= q->parms.qth_min >> q->parms.Wlog,
		.qth_max	= q->parms.qth_max >> q->parms.Wlog,
		.Wlog		= q->parms.Wlog,
		.Plog		= q->parms.Plog,
		.Scell_log	= q->parms.Scell_log,
	};

	opts = RTA_NEST(skb, TCA_OPTIONS);
	RTA_PUT(skb, TCA_RED_PARMS, sizeof(opt), &opt);
	return RTA_NEST_END(skb, opts);

rtattr_failure:
	return RTA_NEST_CANCEL(skb, opts);
}

static int red_dump_stats(struct Qdisc *sch, struct gnet_dump *d)
{
	struct red_sched_data *q = qdisc_priv(sch);
	struct tc_red_xstats st = {
		.early	= q->stats.prob_drop + q->stats.forced_drop,
		.pdrop	= q->stats.pdrop,
		.other	= q->stats.other,
		.marked	= q->stats.prob_mark + q->stats.forced_mark,
	};

	return gnet_stats_copy_app(d, &st, sizeof(st));
}

static int red_dump_class(struct Qdisc *sch, unsigned long cl,
			  struct sk_buff *skb, struct tcmsg *tcm)
{
	struct red_sched_data *q = qdisc_priv(sch);

	if (cl != 1)
		return -ENOENT;
	tcm->tcm_handle |= TC_H_MIN(1);
	tcm->tcm_info = q->qdisc->handle;
	return 0;
}

static int red_graft(struct Qdisc *sch, unsigned long arg, struct Qdisc *new,
		     struct Qdisc **old)
{
	struct red_sched_data *q = qdisc_priv(sch);

	if (new == NULL)
		new = &noop_qdisc;

	sch_tree_lock(sch);
	*old = xchg(&q->qdisc, new);
	qdisc_reset(*old);
	sch->q.qlen = 0;
	sch_tree_unlock(sch);
	return 0;
}

static struct Qdisc *red_leaf(struct Qdisc *sch, unsigned long arg)
{
	struct red_sched_data *q = qdisc_priv(sch);
	return q->qdisc;
}

static unsigned long red_get(struct Qdisc *sch, u32 classid)
{
	return 1;
}

static void red_put(struct Qdisc *sch, unsigned long arg)
{
	return;
}

static int red_change_class(struct Qdisc *sch, u32 classid, u32 parentid,
			    struct rtattr **tca, unsigned long *arg)
{
	return -ENOSYS;
}

static int red_delete(struct Qdisc *sch, unsigned long cl)
{
	return -ENOSYS;
}

static void red_walk(struct Qdisc *sch, struct qdisc_walker *walker)
{
	if (!walker->stop) {
		if (walker->count >= walker->skip)
			if (walker->fn(sch, 1, walker) < 0) {
				walker->stop = 1;
				return;
			}
		walker->count++;
	}
}

static struct tcf_proto **red_find_tcf(struct Qdisc *sch, unsigned long cl)
{
	return NULL;
}

static struct Qdisc_class_ops red_class_ops = {
	.graft		=	red_graft,
	.leaf		=	red_leaf,
	.get		=	red_get,
	.put		=	red_put,
	.change		=	red_change_class,
	.delete		=	red_delete,
	.walk		=	red_walk,
	.tcf_chain	=	red_find_tcf,
	.dump		=	red_dump_class,
};

static struct Qdisc_ops red_qdisc_ops = {
	.id		=	"red",
	.priv_size	=	sizeof(struct red_sched_data),
	.cl_ops		=	&red_class_ops,
	.enqueue	=	red_enqueue,
	.dequeue	=	red_dequeue,
	.requeue	=	red_requeue,
	.drop		=	red_drop,
	.init		=	red_init,
	.reset		=	red_reset,
	.destroy	=	red_destroy,
	.change		=	red_change,
	.dump		=	red_dump,
	.dump_stats	=	red_dump_stats,
	.owner		=	THIS_MODULE,
};

static int __init red_module_init(void)
{
	return register_qdisc(&red_qdisc_ops);
}

static void __exit red_module_exit(void)
{
	unregister_qdisc(&red_qdisc_ops);
}

module_init(red_module_init)
module_exit(red_module_exit)

MODULE_LICENSE("GPL");