sch_generic.h 14.0 KB
Newer Older
L
Linus Torvalds 已提交
1 2 3 4 5 6 7 8 9 10
#ifndef __NET_SCHED_GENERIC_H
#define __NET_SCHED_GENERIC_H

#include <linux/netdevice.h>
#include <linux/types.h>
#include <linux/rcupdate.h>
#include <linux/module.h>
#include <linux/pkt_sched.h>
#include <linux/pkt_cls.h>
#include <net/gen_stats.h>
11
#include <net/rtnetlink.h>
L
Linus Torvalds 已提交
12 13 14 15 16 17 18 19 20 21 22 23 24 25

struct Qdisc_ops;
struct qdisc_walker;
struct tcf_walker;
struct module;

struct qdisc_rate_table
{
	struct tc_ratespec rate;
	u32		data[256];
	struct qdisc_rate_table *next;
	int		refcnt;
};

26 27 28
enum qdisc_state_t
{
	__QDISC_STATE_RUNNING,
29
	__QDISC_STATE_SCHED,
30
	__QDISC_STATE_DEACTIVATED,
31 32
};

33 34 35 36 37 38 39
struct qdisc_size_table {
	struct list_head	list;
	struct tc_sizespec	szopts;
	int			refcnt;
	u16			data[];
};

L
Linus Torvalds 已提交
40 41 42 43 44
struct Qdisc
{
	int 			(*enqueue)(struct sk_buff *skb, struct Qdisc *dev);
	struct sk_buff *	(*dequeue)(struct Qdisc *dev);
	unsigned		flags;
45 46 47
#define TCQ_F_BUILTIN		1
#define TCQ_F_THROTTLED		2
#define TCQ_F_INGRESS		4
48
#define TCQ_F_CAN_BYPASS	8
49
#define TCQ_F_MQROOT		16
50
#define TCQ_F_WARN_NONWC	(1 << 16)
L
Linus Torvalds 已提交
51 52
	int			padded;
	struct Qdisc_ops	*ops;
53
	struct qdisc_size_table	*stab;
54
	struct list_head	list;
L
Linus Torvalds 已提交
55 56 57 58 59 60 61
	u32			handle;
	u32			parent;
	atomic_t		refcnt;
	struct gnet_stats_rate_est	rate_est;
	int			(*reshape_fail)(struct sk_buff *skb,
					struct Qdisc *q);

62 63
	void			*u32_node;

L
Linus Torvalds 已提交
64 65 66 67
	/* This field is deprecated, but it is still used by CBQ
	 * and it will live until better solution will be invented.
	 */
	struct Qdisc		*__parent;
68 69 70 71 72 73 74 75 76
	struct netdev_queue	*dev_queue;
	struct Qdisc		*next_sched;

	struct sk_buff		*gso_skb;
	/*
	 * For performance sake on SMP, we put highly modified fields at the end
	 */
	unsigned long		state;
	struct sk_buff_head	q;
77
	struct gnet_stats_basic_packed bstats;
78
	struct gnet_stats_queue	qstats;
L
Linus Torvalds 已提交
79 80 81 82 83
};

struct Qdisc_class_ops
{
	/* Child qdisc manipulation */
84
	unsigned int		(*select_queue)(struct Qdisc *, struct tcmsg *);
L
Linus Torvalds 已提交
85 86 87
	int			(*graft)(struct Qdisc *, unsigned long cl,
					struct Qdisc *, struct Qdisc **);
	struct Qdisc *		(*leaf)(struct Qdisc *, unsigned long cl);
88
	void			(*qlen_notify)(struct Qdisc *, unsigned long);
L
Linus Torvalds 已提交
89 90 91 92 93

	/* Class manipulation routines */
	unsigned long		(*get)(struct Qdisc *, u32 classid);
	void			(*put)(struct Qdisc *, unsigned long);
	int			(*change)(struct Qdisc *, u32, u32,
94
					struct nlattr **, unsigned long *);
L
Linus Torvalds 已提交
95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113
	int			(*delete)(struct Qdisc *, unsigned long);
	void			(*walk)(struct Qdisc *, struct qdisc_walker * arg);

	/* Filter manipulation */
	struct tcf_proto **	(*tcf_chain)(struct Qdisc *, unsigned long);
	unsigned long		(*bind_tcf)(struct Qdisc *, unsigned long,
					u32 classid);
	void			(*unbind_tcf)(struct Qdisc *, unsigned long);

	/* rtnetlink specific */
	int			(*dump)(struct Qdisc *, unsigned long,
					struct sk_buff *skb, struct tcmsg*);
	int			(*dump_stats)(struct Qdisc *, unsigned long,
					struct gnet_dump *);
};

struct Qdisc_ops
{
	struct Qdisc_ops	*next;
114
	const struct Qdisc_class_ops	*cl_ops;
L
Linus Torvalds 已提交
115 116 117 118 119
	char			id[IFNAMSIZ];
	int			priv_size;

	int 			(*enqueue)(struct sk_buff *, struct Qdisc *);
	struct sk_buff *	(*dequeue)(struct Qdisc *);
120
	struct sk_buff *	(*peek)(struct Qdisc *);
L
Linus Torvalds 已提交
121 122
	unsigned int		(*drop)(struct Qdisc *);

123
	int			(*init)(struct Qdisc *, struct nlattr *arg);
L
Linus Torvalds 已提交
124 125
	void			(*reset)(struct Qdisc *);
	void			(*destroy)(struct Qdisc *);
126
	int			(*change)(struct Qdisc *, struct nlattr *arg);
127
	void			(*attach)(struct Qdisc *);
L
Linus Torvalds 已提交
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

	int			(*dump)(struct Qdisc *, struct sk_buff *);
	int			(*dump_stats)(struct Qdisc *, struct gnet_dump *);

	struct module		*owner;
};


struct tcf_result
{
	unsigned long	class;
	u32		classid;
};

struct tcf_proto_ops
{
	struct tcf_proto_ops	*next;
	char			kind[IFNAMSIZ];

	int			(*classify)(struct sk_buff*, struct tcf_proto*,
					struct tcf_result *);
	int			(*init)(struct tcf_proto*);
	void			(*destroy)(struct tcf_proto*);

	unsigned long		(*get)(struct tcf_proto*, u32 handle);
	void			(*put)(struct tcf_proto*, unsigned long);
	int			(*change)(struct tcf_proto*, unsigned long,
155
					u32 handle, struct nlattr **,
L
Linus Torvalds 已提交
156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173
					unsigned long *);
	int			(*delete)(struct tcf_proto*, unsigned long);
	void			(*walk)(struct tcf_proto*, struct tcf_walker *arg);

	/* rtnetlink specific */
	int			(*dump)(struct tcf_proto*, unsigned long,
					struct sk_buff *skb, struct tcmsg*);

	struct module		*owner;
};

struct tcf_proto
{
	/* Fast access part */
	struct tcf_proto	*next;
	void			*root;
	int			(*classify)(struct sk_buff*, struct tcf_proto*,
					struct tcf_result *);
A
Al Viro 已提交
174
	__be16			protocol;
L
Linus Torvalds 已提交
175 176 177 178 179 180 181 182 183

	/* All the rest */
	u32			prio;
	u32			classid;
	struct Qdisc		*q;
	void			*data;
	struct tcf_proto_ops	*ops;
};

184 185 186 187 188
struct qdisc_skb_cb {
	unsigned int		pkt_len;
	char			data[];
};

189 190 191 192 193
static inline int qdisc_qlen(struct Qdisc *q)
{
	return q->q.qlen;
}

194 195 196 197 198
static inline struct qdisc_skb_cb *qdisc_skb_cb(struct sk_buff *skb)
{
	return (struct qdisc_skb_cb *)skb->cb;
}

199 200 201 202 203
static inline spinlock_t *qdisc_lock(struct Qdisc *qdisc)
{
	return &qdisc->q.lock;
}

204 205 206 207 208
static inline struct Qdisc *qdisc_root(struct Qdisc *qdisc)
{
	return qdisc->dev_queue->qdisc;
}

209 210 211 212 213
static inline struct Qdisc *qdisc_root_sleeping(struct Qdisc *qdisc)
{
	return qdisc->dev_queue->qdisc_sleeping;
}

214 215 216 217 218 219 220 221 222 223 224
/* The qdisc root lock is a mechanism by which to top level
 * of a qdisc tree can be locked from any qdisc node in the
 * forest.  This allows changing the configuration of some
 * aspect of the qdisc tree while blocking out asynchronous
 * qdisc access in the packet processing paths.
 *
 * It is only legal to do this when the root will not change
 * on us.  Otherwise we'll potentially lock the wrong qdisc
 * root.  This is enforced by holding the RTNL semaphore, which
 * all users of this lock accessor must do.
 */
225 226 227 228
static inline spinlock_t *qdisc_root_lock(struct Qdisc *qdisc)
{
	struct Qdisc *root = qdisc_root(qdisc);

229
	ASSERT_RTNL();
230
	return qdisc_lock(root);
231 232
}

233 234 235 236 237 238 239 240
static inline spinlock_t *qdisc_root_sleeping_lock(struct Qdisc *qdisc)
{
	struct Qdisc *root = qdisc_root_sleeping(qdisc);

	ASSERT_RTNL();
	return qdisc_lock(root);
}

241 242 243 244
static inline struct net_device *qdisc_dev(struct Qdisc *qdisc)
{
	return qdisc->dev_queue->dev;
}
L
Linus Torvalds 已提交
245

246 247
static inline void sch_tree_lock(struct Qdisc *q)
{
J
Jarek Poplawski 已提交
248
	spin_lock_bh(qdisc_root_sleeping_lock(q));
249 250 251 252
}

static inline void sch_tree_unlock(struct Qdisc *q)
{
J
Jarek Poplawski 已提交
253
	spin_unlock_bh(qdisc_root_sleeping_lock(q));
254 255 256 257
}

#define tcf_tree_lock(tp)	sch_tree_lock((tp)->q)
#define tcf_tree_unlock(tp)	sch_tree_unlock((tp)->q)
L
Linus Torvalds 已提交
258

259 260
extern struct Qdisc noop_qdisc;
extern struct Qdisc_ops noop_qdisc_ops;
261 262
extern struct Qdisc_ops pfifo_fast_ops;
extern struct Qdisc_ops mq_qdisc_ops;
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
struct Qdisc_class_common
{
	u32			classid;
	struct hlist_node	hnode;
};

struct Qdisc_class_hash
{
	struct hlist_head	*hash;
	unsigned int		hashsize;
	unsigned int		hashmask;
	unsigned int		hashelems;
};

static inline unsigned int qdisc_class_hash(u32 id, u32 mask)
{
	id ^= id >> 8;
	id ^= id >> 4;
	return id & mask;
}

static inline struct Qdisc_class_common *
qdisc_class_find(struct Qdisc_class_hash *hash, u32 id)
{
	struct Qdisc_class_common *cl;
	struct hlist_node *n;
	unsigned int h;

	h = qdisc_class_hash(id, hash->hashmask);
	hlist_for_each_entry(cl, n, &hash->hash[h], hnode) {
		if (cl->classid == id)
			return cl;
	}
	return NULL;
}

extern int qdisc_class_hash_init(struct Qdisc_class_hash *);
extern void qdisc_class_hash_insert(struct Qdisc_class_hash *, struct Qdisc_class_common *);
extern void qdisc_class_hash_remove(struct Qdisc_class_hash *, struct Qdisc_class_common *);
extern void qdisc_class_hash_grow(struct Qdisc *, struct Qdisc_class_hash *);
extern void qdisc_class_hash_destroy(struct Qdisc_class_hash *);

306 307 308 309
extern void dev_init_scheduler(struct net_device *dev);
extern void dev_shutdown(struct net_device *dev);
extern void dev_activate(struct net_device *dev);
extern void dev_deactivate(struct net_device *dev);
310 311
extern struct Qdisc *dev_graft_qdisc(struct netdev_queue *dev_queue,
				     struct Qdisc *qdisc);
312 313
extern void qdisc_reset(struct Qdisc *qdisc);
extern void qdisc_destroy(struct Qdisc *qdisc);
314
extern void qdisc_tree_decrease_qlen(struct Qdisc *qdisc, unsigned int n);
315
extern struct Qdisc *qdisc_alloc(struct netdev_queue *dev_queue,
316
				 struct Qdisc_ops *ops);
317
extern struct Qdisc *qdisc_create_dflt(struct net_device *dev,
318
				       struct netdev_queue *dev_queue,
319
				       struct Qdisc_ops *ops, u32 parentid);
320 321
extern void qdisc_calculate_pkt_len(struct sk_buff *skb,
				   struct qdisc_size_table *stab);
322
extern void tcf_destroy(struct tcf_proto *tp);
323
extern void tcf_destroy_chain(struct tcf_proto **fl);
L
Linus Torvalds 已提交
324

325 326 327
/* Reset all TX qdiscs of a device.  */
static inline void qdisc_reset_all_tx(struct net_device *dev)
{
328 329 330
	unsigned int i;
	for (i = 0; i < dev->num_tx_queues; i++)
		qdisc_reset(netdev_get_tx_queue(dev, i)->qdisc);
331 332
}

333 334 335
/* Are all TX queues of the device empty?  */
static inline bool qdisc_all_tx_empty(const struct net_device *dev)
{
336 337 338 339
	unsigned int i;
	for (i = 0; i < dev->num_tx_queues; i++) {
		struct netdev_queue *txq = netdev_get_tx_queue(dev, i);
		const struct Qdisc *q = txq->qdisc;
340

341 342 343 344
		if (q->q.qlen)
			return false;
	}
	return true;
345 346
}

347 348 349
/* Are any of the TX qdiscs changing?  */
static inline bool qdisc_tx_changing(struct net_device *dev)
{
350 351 352 353 354 355 356
	unsigned int i;
	for (i = 0; i < dev->num_tx_queues; i++) {
		struct netdev_queue *txq = netdev_get_tx_queue(dev, i);
		if (txq->qdisc != txq->qdisc_sleeping)
			return true;
	}
	return false;
357 358
}

359
/* Is the device using the noop qdisc on all queues?  */
360 361
static inline bool qdisc_tx_is_noop(const struct net_device *dev)
{
362 363 364 365 366 367 368
	unsigned int i;
	for (i = 0; i < dev->num_tx_queues; i++) {
		struct netdev_queue *txq = netdev_get_tx_queue(dev, i);
		if (txq->qdisc != &noop_qdisc)
			return false;
	}
	return true;
369 370
}

371 372
static inline unsigned int qdisc_pkt_len(struct sk_buff *skb)
{
373
	return qdisc_skb_cb(skb)->pkt_len;
374 375
}

376
/* additional qdisc xmit flags (NET_XMIT_MASK in linux/netdevice.h) */
377 378
enum net_xmit_qdisc_t {
	__NET_XMIT_STOLEN = 0x00010000,
379
	__NET_XMIT_BYPASS = 0x00020000,
380 381
};

382
#ifdef CONFIG_NET_CLS_ACT
383 384 385 386 387
#define net_xmit_drop_count(e)	((e) & __NET_XMIT_STOLEN ? 0 : 1)
#else
#define net_xmit_drop_count(e)	(1)
#endif

388 389
static inline int qdisc_enqueue(struct sk_buff *skb, struct Qdisc *sch)
{
390
#ifdef CONFIG_NET_SCHED
391 392
	if (sch->stab)
		qdisc_calculate_pkt_len(skb, sch->stab);
393
#endif
394 395 396 397 398
	return sch->enqueue(skb, sch);
}

static inline int qdisc_enqueue_root(struct sk_buff *skb, struct Qdisc *sch)
{
399
	qdisc_skb_cb(skb)->pkt_len = skb->len;
400
	return qdisc_enqueue(skb, sch) & NET_XMIT_MASK;
401 402
}

403 404 405 406 407 408
static inline void __qdisc_update_bstats(struct Qdisc *sch, unsigned int len)
{
	sch->bstats.bytes += len;
	sch->bstats.packets++;
}

409 410 411 412
static inline int __qdisc_enqueue_tail(struct sk_buff *skb, struct Qdisc *sch,
				       struct sk_buff_head *list)
{
	__skb_queue_tail(list, skb);
413
	sch->qstats.backlog += qdisc_pkt_len(skb);
414
	__qdisc_update_bstats(sch, qdisc_pkt_len(skb));
415 416 417 418 419 420 421 422 423 424 425 426 427 428 429

	return NET_XMIT_SUCCESS;
}

static inline int qdisc_enqueue_tail(struct sk_buff *skb, struct Qdisc *sch)
{
	return __qdisc_enqueue_tail(skb, sch, &sch->q);
}

static inline struct sk_buff *__qdisc_dequeue_head(struct Qdisc *sch,
						   struct sk_buff_head *list)
{
	struct sk_buff *skb = __skb_dequeue(list);

	if (likely(skb != NULL))
430
		sch->qstats.backlog -= qdisc_pkt_len(skb);
431 432 433 434 435 436 437 438 439 440 441 442 443 444 445

	return skb;
}

static inline struct sk_buff *qdisc_dequeue_head(struct Qdisc *sch)
{
	return __qdisc_dequeue_head(sch, &sch->q);
}

static inline struct sk_buff *__qdisc_dequeue_tail(struct Qdisc *sch,
						   struct sk_buff_head *list)
{
	struct sk_buff *skb = __skb_dequeue_tail(list);

	if (likely(skb != NULL))
446
		sch->qstats.backlog -= qdisc_pkt_len(skb);
447 448 449 450 451 452 453 454 455

	return skb;
}

static inline struct sk_buff *qdisc_dequeue_tail(struct Qdisc *sch)
{
	return __qdisc_dequeue_tail(sch, &sch->q);
}

456 457 458 459 460
static inline struct sk_buff *qdisc_peek_head(struct Qdisc *sch)
{
	return skb_peek(&sch->q);
}

461 462 463 464
/* generic pseudo peek method for non-work-conserving qdisc */
static inline struct sk_buff *qdisc_peek_dequeued(struct Qdisc *sch)
{
	/* we can reuse ->gso_skb because peek isn't called for root qdiscs */
465
	if (!sch->gso_skb) {
466
		sch->gso_skb = sch->dequeue(sch);
467 468 469 470
		if (sch->gso_skb)
			/* it's still part of the queue */
			sch->q.qlen++;
	}
471 472 473 474 475 476 477 478 479

	return sch->gso_skb;
}

/* use instead of qdisc->dequeue() for all qdiscs queried with ->peek() */
static inline struct sk_buff *qdisc_dequeue_peeked(struct Qdisc *sch)
{
	struct sk_buff *skb = sch->gso_skb;

480
	if (skb) {
481
		sch->gso_skb = NULL;
482 483
		sch->q.qlen--;
	} else {
484
		skb = sch->dequeue(sch);
485
	}
486 487 488 489

	return skb;
}

490 491 492 493 494 495 496
static inline void __qdisc_reset_queue(struct Qdisc *sch,
				       struct sk_buff_head *list)
{
	/*
	 * We do not know the backlog in bytes of this list, it
	 * is up to the caller to correct it
	 */
497
	__skb_queue_purge(list);
498 499 500 501 502 503 504 505 506 507 508 509 510 511
}

static inline void qdisc_reset_queue(struct Qdisc *sch)
{
	__qdisc_reset_queue(sch, &sch->q);
	sch->qstats.backlog = 0;
}

static inline unsigned int __qdisc_queue_drop(struct Qdisc *sch,
					      struct sk_buff_head *list)
{
	struct sk_buff *skb = __qdisc_dequeue_tail(sch, list);

	if (likely(skb != NULL)) {
512
		unsigned int len = qdisc_pkt_len(skb);
513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536
		kfree_skb(skb);
		return len;
	}

	return 0;
}

static inline unsigned int qdisc_queue_drop(struct Qdisc *sch)
{
	return __qdisc_queue_drop(sch, &sch->q);
}

static inline int qdisc_drop(struct sk_buff *skb, struct Qdisc *sch)
{
	kfree_skb(skb);
	sch->qstats.drops++;

	return NET_XMIT_DROP;
}

static inline int qdisc_reshape_fail(struct sk_buff *skb, struct Qdisc *sch)
{
	sch->qstats.drops++;

537
#ifdef CONFIG_NET_CLS_ACT
538 539 540 541 542 543 544 545 546 547 548
	if (sch->reshape_fail == NULL || sch->reshape_fail(skb, sch))
		goto drop;

	return NET_XMIT_SUCCESS;

drop:
#endif
	kfree_skb(skb);
	return NET_XMIT_DROP;
}

549 550 551 552 553
/* Length to Time (L2T) lookup in a qdisc_rate_table, to determine how
   long it will take to send a packet given its size.
 */
static inline u32 qdisc_l2t(struct qdisc_rate_table* rtab, unsigned int pktlen)
{
554 555 556
	int slot = pktlen + rtab->rate.cell_align + rtab->rate.overhead;
	if (slot < 0)
		slot = 0;
557 558 559 560 561 562
	slot >>= rtab->rate.cell_log;
	if (slot > 255)
		return (rtab->data[255]*(slot >> 8) + rtab->data[slot & 0xFF]);
	return rtab->data[slot];
}

563 564 565 566 567 568 569 570 571 572 573 574 575 576
#ifdef CONFIG_NET_CLS_ACT
static inline struct sk_buff *skb_act_clone(struct sk_buff *skb, gfp_t gfp_mask)
{
	struct sk_buff *n = skb_clone(skb, gfp_mask);

	if (n) {
		n->tc_verd = SET_TC_VERD(n->tc_verd, 0);
		n->tc_verd = CLR_TC_OK2MUNGE(n->tc_verd);
		n->tc_verd = CLR_TC_MUNGED(n->tc_verd);
	}
	return n;
}
#endif

L
Linus Torvalds 已提交
577
#endif