sch_prio.c 9.2 KB
Newer Older
L
Linus Torvalds 已提交
1 2 3 4 5 6 7 8 9
/*
 * net/sched/sch_prio.c	Simple 3-band priority "scheduler".
 *
 *		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>
10
 * Fixes:       19990609: J Hadi Salim <hadi@nortelnetworks.com>:
L
Linus Torvalds 已提交
11 12 13 14 15 16 17 18 19
 *              Init --  EINVAL when opt undefined
 */

#include <linux/module.h>
#include <linux/types.h>
#include <linux/kernel.h>
#include <linux/string.h>
#include <linux/errno.h>
#include <linux/skbuff.h>
20
#include <net/netlink.h>
L
Linus Torvalds 已提交
21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38
#include <net/pkt_sched.h>


struct prio_sched_data
{
	int bands;
	struct tcf_proto *filter_list;
	u8  prio2band[TC_PRIO_MAX+1];
	struct Qdisc *queues[TCQ_PRIO_BANDS];
};


static struct Qdisc *
prio_classify(struct sk_buff *skb, struct Qdisc *sch, int *qerr)
{
	struct prio_sched_data *q = qdisc_priv(sch);
	u32 band = skb->priority;
	struct tcf_result res;
39
	int err;
L
Linus Torvalds 已提交
40

41
	*qerr = NET_XMIT_BYPASS;
L
Linus Torvalds 已提交
42
	if (TC_H_MAJ(skb->priority) != sch->handle) {
43
		err = tc_classify(skb, q->filter_list, &res);
L
Linus Torvalds 已提交
44
#ifdef CONFIG_NET_CLS_ACT
45
		switch (err) {
L
Linus Torvalds 已提交
46 47
		case TC_ACT_STOLEN:
		case TC_ACT_QUEUED:
48
			*qerr = NET_XMIT_SUCCESS | __NET_XMIT_STOLEN;
L
Linus Torvalds 已提交
49 50
		case TC_ACT_SHOT:
			return NULL;
51
		}
L
Linus Torvalds 已提交
52
#endif
53
		if (!q->filter_list || err < 0) {
L
Linus Torvalds 已提交
54 55
			if (TC_H_MAJ(band))
				band = 0;
56
			return q->queues[q->prio2band[band&TC_PRIO_MAX]];
L
Linus Torvalds 已提交
57 58 59 60
		}
		band = res.classid;
	}
	band = TC_H_MIN(band) - 1;
61
	if (band >= q->bands)
62 63
		return q->queues[q->prio2band[0]];

L
Linus Torvalds 已提交
64 65 66 67 68 69 70 71 72 73 74 75
	return q->queues[band];
}

static int
prio_enqueue(struct sk_buff *skb, struct Qdisc *sch)
{
	struct Qdisc *qdisc;
	int ret;

	qdisc = prio_classify(skb, sch, &ret);
#ifdef CONFIG_NET_CLS_ACT
	if (qdisc == NULL) {
76 77

		if (ret == NET_XMIT_BYPASS)
L
Linus Torvalds 已提交
78 79 80 81 82 83
			sch->qstats.drops++;
		kfree_skb(skb);
		return ret;
	}
#endif

84 85
	ret = qdisc_enqueue(skb, qdisc);
	if (ret == NET_XMIT_SUCCESS) {
86
		sch->bstats.bytes += qdisc_pkt_len(skb);
L
Linus Torvalds 已提交
87 88 89 90
		sch->bstats.packets++;
		sch->q.qlen++;
		return NET_XMIT_SUCCESS;
	}
91 92
	if (net_xmit_drop_count(ret))
		sch->qstats.drops++;
93
	return ret;
L
Linus Torvalds 已提交
94 95 96 97 98 99 100 101 102 103 104 105
}


static int
prio_requeue(struct sk_buff *skb, struct Qdisc* sch)
{
	struct Qdisc *qdisc;
	int ret;

	qdisc = prio_classify(skb, sch, &ret);
#ifdef CONFIG_NET_CLS_ACT
	if (qdisc == NULL) {
106
		if (ret == NET_XMIT_BYPASS)
L
Linus Torvalds 已提交
107 108 109 110 111 112 113 114 115 116 117
			sch->qstats.drops++;
		kfree_skb(skb);
		return ret;
	}
#endif

	if ((ret = qdisc->ops->requeue(skb, qdisc)) == NET_XMIT_SUCCESS) {
		sch->q.qlen++;
		sch->qstats.requeues++;
		return 0;
	}
118 119
	if (net_xmit_drop_count(ret))
		sch->qstats.drops++;
L
Linus Torvalds 已提交
120 121 122 123
	return NET_XMIT_DROP;
}


124
static struct sk_buff *prio_dequeue(struct Qdisc* sch)
L
Linus Torvalds 已提交
125 126 127 128 129
{
	struct prio_sched_data *q = qdisc_priv(sch);
	int prio;

	for (prio = 0; prio < q->bands; prio++) {
130 131 132 133 134
		struct Qdisc *qdisc = q->queues[prio];
		struct sk_buff *skb = qdisc->dequeue(qdisc);
		if (skb) {
			sch->q.qlen--;
			return skb;
L
Linus Torvalds 已提交
135 136 137 138 139 140 141 142 143 144 145 146 147 148 149
		}
	}
	return NULL;

}

static unsigned int prio_drop(struct Qdisc* sch)
{
	struct prio_sched_data *q = qdisc_priv(sch);
	int prio;
	unsigned int len;
	struct Qdisc *qdisc;

	for (prio = q->bands-1; prio >= 0; prio--) {
		qdisc = q->queues[prio];
150
		if (qdisc->ops->drop && (len = qdisc->ops->drop(qdisc)) != 0) {
L
Linus Torvalds 已提交
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
			sch->q.qlen--;
			return len;
		}
	}
	return 0;
}


static void
prio_reset(struct Qdisc* sch)
{
	int prio;
	struct prio_sched_data *q = qdisc_priv(sch);

	for (prio=0; prio<q->bands; prio++)
		qdisc_reset(q->queues[prio]);
	sch->q.qlen = 0;
}

static void
prio_destroy(struct Qdisc* sch)
{
	int prio;
	struct prio_sched_data *q = qdisc_priv(sch);

176
	tcf_destroy_chain(&q->filter_list);
L
Linus Torvalds 已提交
177 178 179 180
	for (prio=0; prio<q->bands; prio++)
		qdisc_destroy(q->queues[prio]);
}

181
static int prio_tune(struct Qdisc *sch, struct nlattr *opt)
L
Linus Torvalds 已提交
182 183
{
	struct prio_sched_data *q = qdisc_priv(sch);
184
	struct tc_prio_qopt *qopt;
L
Linus Torvalds 已提交
185 186
	int i;

187 188 189
	if (nla_len(opt) < sizeof(*qopt))
		return -EINVAL;
	qopt = nla_data(opt);
190

191
	if (qopt->bands > TCQ_PRIO_BANDS || qopt->bands < 2)
L
Linus Torvalds 已提交
192 193 194
		return -EINVAL;

	for (i=0; i<=TC_PRIO_MAX; i++) {
195
		if (qopt->priomap[i] >= qopt->bands)
L
Linus Torvalds 已提交
196 197 198 199
			return -EINVAL;
	}

	sch_tree_lock(sch);
200
	q->bands = qopt->bands;
L
Linus Torvalds 已提交
201 202 203 204
	memcpy(q->prio2band, qopt->priomap, TC_PRIO_MAX+1);

	for (i=q->bands; i<TCQ_PRIO_BANDS; i++) {
		struct Qdisc *child = xchg(&q->queues[i], &noop_qdisc);
205 206
		if (child != &noop_qdisc) {
			qdisc_tree_decrease_qlen(child, child->q.qlen);
L
Linus Torvalds 已提交
207
			qdisc_destroy(child);
208
		}
L
Linus Torvalds 已提交
209 210 211
	}
	sch_tree_unlock(sch);

212 213
	for (i=0; i<q->bands; i++) {
		if (q->queues[i] == &noop_qdisc) {
L
Linus Torvalds 已提交
214
			struct Qdisc *child;
215
			child = qdisc_create_dflt(qdisc_dev(sch), sch->dev_queue,
216
						  &pfifo_qdisc_ops,
217
						  TC_H_MAKE(sch->handle, i + 1));
L
Linus Torvalds 已提交
218 219
			if (child) {
				sch_tree_lock(sch);
220
				child = xchg(&q->queues[i], child);
L
Linus Torvalds 已提交
221

222 223 224
				if (child != &noop_qdisc) {
					qdisc_tree_decrease_qlen(child,
								 child->q.qlen);
L
Linus Torvalds 已提交
225
					qdisc_destroy(child);
226
				}
L
Linus Torvalds 已提交
227 228 229 230 231 232 233
				sch_tree_unlock(sch);
			}
		}
	}
	return 0;
}

234
static int prio_init(struct Qdisc *sch, struct nlattr *opt)
L
Linus Torvalds 已提交
235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255
{
	struct prio_sched_data *q = qdisc_priv(sch);
	int i;

	for (i=0; i<TCQ_PRIO_BANDS; i++)
		q->queues[i] = &noop_qdisc;

	if (opt == NULL) {
		return -EINVAL;
	} else {
		int err;

		if ((err= prio_tune(sch, opt)) != 0)
			return err;
	}
	return 0;
}

static int prio_dump(struct Qdisc *sch, struct sk_buff *skb)
{
	struct prio_sched_data *q = qdisc_priv(sch);
256
	unsigned char *b = skb_tail_pointer(skb);
257
	struct nlattr *nest;
L
Linus Torvalds 已提交
258 259 260 261
	struct tc_prio_qopt opt;

	opt.bands = q->bands;
	memcpy(&opt.priomap, q->prio2band, TC_PRIO_MAX+1);
262

263 264 265 266
	nest = nla_nest_compat_start(skb, TCA_OPTIONS, sizeof(opt), &opt);
	if (nest == NULL)
		goto nla_put_failure;
	nla_nest_compat_end(skb, nest);
267

L
Linus Torvalds 已提交
268 269
	return skb->len;

270
nla_put_failure:
271
	nlmsg_trim(skb, b);
L
Linus Torvalds 已提交
272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289
	return -1;
}

static int prio_graft(struct Qdisc *sch, unsigned long arg, struct Qdisc *new,
		      struct Qdisc **old)
{
	struct prio_sched_data *q = qdisc_priv(sch);
	unsigned long band = arg - 1;

	if (band >= q->bands)
		return -EINVAL;

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

	sch_tree_lock(sch);
	*old = q->queues[band];
	q->queues[band] = new;
290
	qdisc_tree_decrease_qlen(*old, (*old)->q.qlen);
L
Linus Torvalds 已提交
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
	qdisc_reset(*old);
	sch_tree_unlock(sch);

	return 0;
}

static struct Qdisc *
prio_leaf(struct Qdisc *sch, unsigned long arg)
{
	struct prio_sched_data *q = qdisc_priv(sch);
	unsigned long band = arg - 1;

	if (band >= q->bands)
		return NULL;

	return q->queues[band];
}

static unsigned long prio_get(struct Qdisc *sch, u32 classid)
{
	struct prio_sched_data *q = qdisc_priv(sch);
	unsigned long band = TC_H_MIN(classid);

	if (band - 1 >= q->bands)
		return 0;
	return band;
}

static unsigned long prio_bind(struct Qdisc *sch, unsigned long parent, u32 classid)
{
	return prio_get(sch, classid);
}


static void prio_put(struct Qdisc *q, unsigned long cl)
{
	return;
}

330
static int prio_change(struct Qdisc *sch, u32 handle, u32 parent, struct nlattr **tca, unsigned long *arg)
L
Linus Torvalds 已提交
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
{
	unsigned long cl = *arg;
	struct prio_sched_data *q = qdisc_priv(sch);

	if (cl - 1 > q->bands)
		return -ENOENT;
	return 0;
}

static int prio_delete(struct Qdisc *sch, unsigned long cl)
{
	struct prio_sched_data *q = qdisc_priv(sch);
	if (cl - 1 > q->bands)
		return -ENOENT;
	return 0;
}


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

	if (cl - 1 > q->bands)
		return -ENOENT;
	tcm->tcm_handle |= TC_H_MIN(cl);
	if (q->queues[cl-1])
		tcm->tcm_info = q->queues[cl-1]->handle;
	return 0;
}

362 363 364 365 366 367 368 369 370 371 372 373 374 375
static int prio_dump_class_stats(struct Qdisc *sch, unsigned long cl,
				 struct gnet_dump *d)
{
	struct prio_sched_data *q = qdisc_priv(sch);
	struct Qdisc *cl_q;

	cl_q = q->queues[cl - 1];
	if (gnet_stats_copy_basic(d, &cl_q->bstats) < 0 ||
	    gnet_stats_copy_queue(d, &cl_q->qstats) < 0)
		return -1;

	return 0;
}

L
Linus Torvalds 已提交
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
static void prio_walk(struct Qdisc *sch, struct qdisc_walker *arg)
{
	struct prio_sched_data *q = qdisc_priv(sch);
	int prio;

	if (arg->stop)
		return;

	for (prio = 0; prio < q->bands; prio++) {
		if (arg->count < arg->skip) {
			arg->count++;
			continue;
		}
		if (arg->fn(sch, prio+1, arg) < 0) {
			arg->stop = 1;
			break;
		}
		arg->count++;
	}
}

static struct tcf_proto ** prio_find_tcf(struct Qdisc *sch, unsigned long cl)
{
	struct prio_sched_data *q = qdisc_priv(sch);

	if (cl)
		return NULL;
	return &q->filter_list;
}

406
static const struct Qdisc_class_ops prio_class_ops = {
L
Linus Torvalds 已提交
407 408 409 410 411 412 413 414 415 416 417
	.graft		=	prio_graft,
	.leaf		=	prio_leaf,
	.get		=	prio_get,
	.put		=	prio_put,
	.change		=	prio_change,
	.delete		=	prio_delete,
	.walk		=	prio_walk,
	.tcf_chain	=	prio_find_tcf,
	.bind_tcf	=	prio_bind,
	.unbind_tcf	=	prio_put,
	.dump		=	prio_dump_class,
418
	.dump_stats	=	prio_dump_class_stats,
L
Linus Torvalds 已提交
419 420
};

421
static struct Qdisc_ops prio_qdisc_ops __read_mostly = {
L
Linus Torvalds 已提交
422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439
	.next		=	NULL,
	.cl_ops		=	&prio_class_ops,
	.id		=	"prio",
	.priv_size	=	sizeof(struct prio_sched_data),
	.enqueue	=	prio_enqueue,
	.dequeue	=	prio_dequeue,
	.requeue	=	prio_requeue,
	.drop		=	prio_drop,
	.init		=	prio_init,
	.reset		=	prio_reset,
	.destroy	=	prio_destroy,
	.change		=	prio_tune,
	.dump		=	prio_dump,
	.owner		=	THIS_MODULE,
};

static int __init prio_module_init(void)
{
440
	return register_qdisc(&prio_qdisc_ops);
L
Linus Torvalds 已提交
441 442
}

443
static void __exit prio_module_exit(void)
L
Linus Torvalds 已提交
444 445 446 447 448 449 450 451
{
	unregister_qdisc(&prio_qdisc_ops);
}

module_init(prio_module_init)
module_exit(prio_module_exit)

MODULE_LICENSE("GPL");