rhashtable.h 27 KB
Newer Older
1 2 3
/*
 * Resizable, Scalable, Concurrent Hash Table
 *
4
 * Copyright (c) 2015 Herbert Xu <herbert@gondor.apana.org.au>
5
 * Copyright (c) 2014-2015 Thomas Graf <tgraf@suug.ch>
6 7 8
 * Copyright (c) 2008-2014 Patrick McHardy <kaber@trash.net>
 *
 * Code partially derived from nft_hash
9 10
 * Rewritten with rehash code from br_multicast plus single list
 * pointer as suggested by Josh Triplett
11 12 13 14 15 16 17 18 19
 *
 * This program is free software; you can redistribute it and/or modify
 * it under the terms of the GNU General Public License version 2 as
 * published by the Free Software Foundation.
 */

#ifndef _LINUX_RHASHTABLE_H
#define _LINUX_RHASHTABLE_H

20
#include <linux/atomic.h>
21
#include <linux/compiler.h>
22
#include <linux/err.h>
23
#include <linux/errno.h>
24
#include <linux/jhash.h>
25
#include <linux/list_nulls.h>
26
#include <linux/workqueue.h>
27
#include <linux/mutex.h>
28
#include <linux/rcupdate.h>
29

30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48
/*
 * The end of the chain is marked with a special nulls marks which has
 * the following format:
 *
 * +-------+-----------------------------------------------------+-+
 * | Base  |                      Hash                           |1|
 * +-------+-----------------------------------------------------+-+
 *
 * Base (4 bits) : Reserved to distinguish between multiple tables.
 *                 Specified via &struct rhashtable_params.nulls_base.
 * Hash (27 bits): Full hash (unmasked) of first element added to bucket
 * 1 (1 bit)     : Nulls marker (always set)
 *
 * The remaining bits of the next pointer remain unused for now.
 */
#define RHT_BASE_BITS		4
#define RHT_HASH_BITS		27
#define RHT_BASE_SHIFT		RHT_HASH_BITS

49 50 51
/* Base bits plus 1 bit for nulls marker */
#define RHT_HASH_RESERVED_SPACE	(RHT_BASE_BITS + 1)

52
struct rhash_head {
53
	struct rhash_head __rcu		*next;
54 55
};

56 57 58
/**
 * struct bucket_table - Table of hash buckets
 * @size: Number of hash buckets
59
 * @rehash: Current bucket being rehashed
60
 * @hash_rnd: Random seed to fold into hash
61 62
 * @locks_mask: Mask to apply before accessing locks[]
 * @locks: Array of spinlocks protecting individual buckets
63
 * @walkers: List of active walkers
64
 * @rcu: RCU structure for freeing the table
65
 * @future_tbl: Table under construction during rehashing
66 67
 * @buckets: size * hash buckets
 */
68
struct bucket_table {
69 70
	unsigned int		size;
	unsigned int		rehash;
71
	u32			hash_rnd;
72 73
	unsigned int		locks_mask;
	spinlock_t		*locks;
74
	struct list_head	walkers;
75
	struct rcu_head		rcu;
76

77 78
	struct bucket_table __rcu *future_tbl;

79
	struct rhash_head __rcu	*buckets[] ____cacheline_aligned_in_smp;
80 81
};

82 83 84 85 86 87 88 89 90 91
/**
 * struct rhashtable_compare_arg - Key for the function rhashtable_compare
 * @ht: Hash table
 * @key: Key to compare against
 */
struct rhashtable_compare_arg {
	struct rhashtable *ht;
	const void *key;
};

92
typedef u32 (*rht_hashfn_t)(const void *data, u32 len, u32 seed);
93
typedef u32 (*rht_obj_hashfn_t)(const void *data, u32 len, u32 seed);
94 95
typedef int (*rht_obj_cmpfn_t)(struct rhashtable_compare_arg *arg,
			       const void *obj);
96 97 98 99 100 101 102 103 104

struct rhashtable;

/**
 * struct rhashtable_params - Hash table construction parameters
 * @nelem_hint: Hint on number of elements, should be 75% of desired size
 * @key_len: Length of key
 * @key_offset: Offset of key in struct to be hashed
 * @head_offset: Offset of rhash_head in struct to be hashed
105
 * @insecure_max_entries: Maximum number of entries (may be exceeded)
106 107
 * @max_size: Maximum size while expanding
 * @min_size: Minimum size while shrinking
108
 * @nulls_base: Base value to generate nulls marker
109
 * @insecure_elasticity: Set to true to disable chain length checks
110
 * @automatic_shrinking: Enable automatic shrinking of tables
111
 * @locks_mul: Number of bucket locks to allocate per cpu (default: 128)
112
 * @hashfn: Hash function (default: jhash2 if !(key_len % 4), or jhash)
113
 * @obj_hashfn: Function to hash object
114
 * @obj_cmpfn: Function to compare key with object
115 116 117 118 119 120
 */
struct rhashtable_params {
	size_t			nelem_hint;
	size_t			key_len;
	size_t			key_offset;
	size_t			head_offset;
121
	unsigned int		insecure_max_entries;
122 123
	unsigned int		max_size;
	unsigned int		min_size;
124
	u32			nulls_base;
125
	bool			insecure_elasticity;
126
	bool			automatic_shrinking;
127
	size_t			locks_mul;
128 129
	rht_hashfn_t		hashfn;
	rht_obj_hashfn_t	obj_hashfn;
130
	rht_obj_cmpfn_t		obj_cmpfn;
131 132 133 134 135 136
};

/**
 * struct rhashtable - Hash table handle
 * @tbl: Bucket table
 * @nelems: Number of elements in table
137
 * @key_len: Key length for hashfn
138
 * @elasticity: Maximum chain length before rehash
139
 * @p: Configuration parameters
140 141
 * @run_work: Deferred worker to expand/shrink asynchronously
 * @mutex: Mutex to protect current/future table swapping
142
 * @lock: Spin lock to protect walker list
143 144 145
 */
struct rhashtable {
	struct bucket_table __rcu	*tbl;
146
	atomic_t			nelems;
147
	unsigned int			key_len;
148
	unsigned int			elasticity;
149
	struct rhashtable_params	p;
150
	struct work_struct		run_work;
151
	struct mutex                    mutex;
152
	spinlock_t			lock;
153 154
};

155 156 157
/**
 * struct rhashtable_walker - Hash table walker
 * @list: List entry on list of walkers
158
 * @tbl: The table that we were walking over
159 160 161
 */
struct rhashtable_walker {
	struct list_head list;
162
	struct bucket_table *tbl;
163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180
};

/**
 * struct rhashtable_iter - Hash table iterator, fits into netlink cb
 * @ht: Table to iterate through
 * @p: Current pointer
 * @walker: Associated rhashtable walker
 * @slot: Current slot
 * @skip: Number of entries to skip in slot
 */
struct rhashtable_iter {
	struct rhashtable *ht;
	struct rhash_head *p;
	struct rhashtable_walker *walker;
	unsigned int slot;
	unsigned int skip;
};

181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198
static inline unsigned long rht_marker(const struct rhashtable *ht, u32 hash)
{
	return NULLS_MARKER(ht->p.nulls_base + hash);
}

#define INIT_RHT_NULLS_HEAD(ptr, ht, hash) \
	((ptr) = (typeof(ptr)) rht_marker(ht, hash))

static inline bool rht_is_a_nulls(const struct rhash_head *ptr)
{
	return ((unsigned long) ptr & 1);
}

static inline unsigned long rht_get_nulls_value(const struct rhash_head *ptr)
{
	return ((unsigned long) ptr) >> 1;
}

199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214
static inline void *rht_obj(const struct rhashtable *ht,
			    const struct rhash_head *he)
{
	return (char *)he - ht->p.head_offset;
}

static inline unsigned int rht_bucket_index(const struct bucket_table *tbl,
					    unsigned int hash)
{
	return (hash >> RHT_HASH_RESERVED_SPACE) & (tbl->size - 1);
}

static inline unsigned int rht_key_hashfn(
	struct rhashtable *ht, const struct bucket_table *tbl,
	const void *key, const struct rhashtable_params params)
{
215
	unsigned int hash;
216

217 218 219 220
	/* params must be equal to ht->p if it isn't constant. */
	if (!__builtin_constant_p(params.key_len))
		hash = ht->p.hashfn(key, ht->key_len, tbl->hash_rnd);
	else if (params.key_len) {
221
		unsigned int key_len = params.key_len;
222 223 224 225 226 227 228 229 230

		if (params.hashfn)
			hash = params.hashfn(key, key_len, tbl->hash_rnd);
		else if (key_len & (sizeof(u32) - 1))
			hash = jhash(key, key_len, tbl->hash_rnd);
		else
			hash = jhash2(key, key_len / sizeof(u32),
				      tbl->hash_rnd);
	} else {
231
		unsigned int key_len = ht->p.key_len;
232 233 234 235 236 237 238 239

		if (params.hashfn)
			hash = params.hashfn(key, key_len, tbl->hash_rnd);
		else
			hash = jhash(key, key_len, tbl->hash_rnd);
	}

	return rht_bucket_index(tbl, hash);
240 241 242 243 244 245 246 247 248
}

static inline unsigned int rht_head_hashfn(
	struct rhashtable *ht, const struct bucket_table *tbl,
	const struct rhash_head *he, const struct rhashtable_params params)
{
	const char *ptr = rht_obj(ht, he);

	return likely(params.obj_hashfn) ?
249 250 251
	       rht_bucket_index(tbl, params.obj_hashfn(ptr, params.key_len ?:
							    ht->p.key_len,
						       tbl->hash_rnd)) :
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
	       rht_key_hashfn(ht, tbl, ptr + params.key_offset, params);
}

/**
 * rht_grow_above_75 - returns true if nelems > 0.75 * table-size
 * @ht:		hash table
 * @tbl:	current table
 */
static inline bool rht_grow_above_75(const struct rhashtable *ht,
				     const struct bucket_table *tbl)
{
	/* Expand table when exceeding 75% load */
	return atomic_read(&ht->nelems) > (tbl->size / 4 * 3) &&
	       (!ht->p.max_size || tbl->size < ht->p.max_size);
}

/**
 * rht_shrink_below_30 - returns true if nelems < 0.3 * table-size
 * @ht:		hash table
 * @tbl:	current table
 */
static inline bool rht_shrink_below_30(const struct rhashtable *ht,
				       const struct bucket_table *tbl)
{
	/* Shrink table beneath 30% load */
	return atomic_read(&ht->nelems) < (tbl->size * 3 / 10) &&
	       tbl->size > ht->p.min_size;
}

281 282 283 284 285 286 287 288
/**
 * rht_grow_above_100 - returns true if nelems > table-size
 * @ht:		hash table
 * @tbl:	current table
 */
static inline bool rht_grow_above_100(const struct rhashtable *ht,
				      const struct bucket_table *tbl)
{
289 290
	return atomic_read(&ht->nelems) > tbl->size &&
		(!ht->p.max_size || tbl->size < ht->p.max_size);
291 292
}

293 294 295 296 297 298 299 300 301 302 303 304
/**
 * rht_grow_above_max - returns true if table is above maximum
 * @ht:		hash table
 * @tbl:	current table
 */
static inline bool rht_grow_above_max(const struct rhashtable *ht,
				      const struct bucket_table *tbl)
{
	return ht->p.insecure_max_entries &&
	       atomic_read(&ht->nelems) >= ht->p.insecure_max_entries;
}

305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323
/* The bucket lock is selected based on the hash and protects mutations
 * on a group of hash buckets.
 *
 * A maximum of tbl->size/2 bucket locks is allocated. This ensures that
 * a single lock always covers both buckets which may both contains
 * entries which link to the same bucket of the old table during resizing.
 * This allows to simplify the locking as locking the bucket in both
 * tables during resize always guarantee protection.
 *
 * IMPORTANT: When holding the bucket lock of both the old and new table
 * during expansions and shrinking, the old bucket lock must always be
 * acquired first.
 */
static inline spinlock_t *rht_bucket_lock(const struct bucket_table *tbl,
					  unsigned int hash)
{
	return &tbl->locks[hash & tbl->locks_mask];
}

324
#ifdef CONFIG_PROVE_LOCKING
325
int lockdep_rht_mutex_is_held(struct rhashtable *ht);
326
int lockdep_rht_bucket_is_held(const struct bucket_table *tbl, u32 hash);
327
#else
328
static inline int lockdep_rht_mutex_is_held(struct rhashtable *ht)
329 330 331
{
	return 1;
}
332 333 334 335 336 337

static inline int lockdep_rht_bucket_is_held(const struct bucket_table *tbl,
					     u32 hash)
{
	return 1;
}
338 339
#endif /* CONFIG_PROVE_LOCKING */

340 341
int rhashtable_init(struct rhashtable *ht,
		    const struct rhashtable_params *params);
342

343 344 345 346 347
struct bucket_table *rhashtable_insert_slow(struct rhashtable *ht,
					    const void *key,
					    struct rhash_head *obj,
					    struct bucket_table *old_tbl);
int rhashtable_insert_rehash(struct rhashtable *ht, struct bucket_table *tbl);
348

349 350
int rhashtable_walk_init(struct rhashtable *ht, struct rhashtable_iter *iter,
			 gfp_t gfp);
351 352 353 354 355
void rhashtable_walk_exit(struct rhashtable_iter *iter);
int rhashtable_walk_start(struct rhashtable_iter *iter) __acquires(RCU);
void *rhashtable_walk_next(struct rhashtable_iter *iter);
void rhashtable_walk_stop(struct rhashtable_iter *iter) __releases(RCU);

356 357 358
void rhashtable_free_and_destroy(struct rhashtable *ht,
				 void (*free_fn)(void *ptr, void *arg),
				 void *arg);
359
void rhashtable_destroy(struct rhashtable *ht);
360 361 362 363 364 365 366

#define rht_dereference(p, ht) \
	rcu_dereference_protected(p, lockdep_rht_mutex_is_held(ht))

#define rht_dereference_rcu(p, ht) \
	rcu_dereference_check(p, lockdep_rht_mutex_is_held(ht))

367 368
#define rht_dereference_bucket(p, tbl, hash) \
	rcu_dereference_protected(p, lockdep_rht_bucket_is_held(tbl, hash))
369

370 371 372 373 374
#define rht_dereference_bucket_rcu(p, tbl, hash) \
	rcu_dereference_check(p, lockdep_rht_bucket_is_held(tbl, hash))

#define rht_entry(tpos, pos, member) \
	({ tpos = container_of(pos, typeof(*tpos), member); 1; })
375 376

/**
377 378 379 380 381
 * rht_for_each_continue - continue iterating over hash chain
 * @pos:	the &struct rhash_head to use as a loop cursor.
 * @head:	the previous &struct rhash_head to continue from
 * @tbl:	the &struct bucket_table
 * @hash:	the hash value / bucket index
382
 */
383 384
#define rht_for_each_continue(pos, head, tbl, hash) \
	for (pos = rht_dereference_bucket(head, tbl, hash); \
385
	     !rht_is_a_nulls(pos); \
386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407
	     pos = rht_dereference_bucket((pos)->next, tbl, hash))

/**
 * rht_for_each - iterate over hash chain
 * @pos:	the &struct rhash_head to use as a loop cursor.
 * @tbl:	the &struct bucket_table
 * @hash:	the hash value / bucket index
 */
#define rht_for_each(pos, tbl, hash) \
	rht_for_each_continue(pos, (tbl)->buckets[hash], tbl, hash)

/**
 * rht_for_each_entry_continue - continue iterating over hash chain
 * @tpos:	the type * to use as a loop cursor.
 * @pos:	the &struct rhash_head to use as a loop cursor.
 * @head:	the previous &struct rhash_head to continue from
 * @tbl:	the &struct bucket_table
 * @hash:	the hash value / bucket index
 * @member:	name of the &struct rhash_head within the hashable struct.
 */
#define rht_for_each_entry_continue(tpos, pos, head, tbl, hash, member)	\
	for (pos = rht_dereference_bucket(head, tbl, hash);		\
408
	     (!rht_is_a_nulls(pos)) && rht_entry(tpos, pos, member);	\
409
	     pos = rht_dereference_bucket((pos)->next, tbl, hash))
410 411 412

/**
 * rht_for_each_entry - iterate over hash chain of given type
413 414 415 416 417
 * @tpos:	the type * to use as a loop cursor.
 * @pos:	the &struct rhash_head to use as a loop cursor.
 * @tbl:	the &struct bucket_table
 * @hash:	the hash value / bucket index
 * @member:	name of the &struct rhash_head within the hashable struct.
418
 */
419 420 421
#define rht_for_each_entry(tpos, pos, tbl, hash, member)		\
	rht_for_each_entry_continue(tpos, pos, (tbl)->buckets[hash],	\
				    tbl, hash, member)
422 423 424

/**
 * rht_for_each_entry_safe - safely iterate over hash chain of given type
425 426 427 428 429 430
 * @tpos:	the type * to use as a loop cursor.
 * @pos:	the &struct rhash_head to use as a loop cursor.
 * @next:	the &struct rhash_head to use as next in loop cursor.
 * @tbl:	the &struct bucket_table
 * @hash:	the hash value / bucket index
 * @member:	name of the &struct rhash_head within the hashable struct.
431 432 433 434
 *
 * This hash chain list-traversal primitive allows for the looped code to
 * remove the loop cursor from the list.
 */
435 436
#define rht_for_each_entry_safe(tpos, pos, next, tbl, hash, member)	    \
	for (pos = rht_dereference_bucket((tbl)->buckets[hash], tbl, hash), \
437 438 439
	     next = !rht_is_a_nulls(pos) ?				    \
		       rht_dereference_bucket(pos->next, tbl, hash) : NULL; \
	     (!rht_is_a_nulls(pos)) && rht_entry(tpos, pos, member);	    \
440 441 442
	     pos = next,						    \
	     next = !rht_is_a_nulls(pos) ?				    \
		       rht_dereference_bucket(pos->next, tbl, hash) : NULL)
443 444 445 446 447 448 449 450 451 452 453 454 455 456 457

/**
 * rht_for_each_rcu_continue - continue iterating over rcu hash chain
 * @pos:	the &struct rhash_head to use as a loop cursor.
 * @head:	the previous &struct rhash_head to continue from
 * @tbl:	the &struct bucket_table
 * @hash:	the hash value / bucket index
 *
 * This hash chain list-traversal primitive may safely run concurrently with
 * the _rcu mutation primitives such as rhashtable_insert() as long as the
 * traversal is guarded by rcu_read_lock().
 */
#define rht_for_each_rcu_continue(pos, head, tbl, hash)			\
	for (({barrier(); }),						\
	     pos = rht_dereference_bucket_rcu(head, tbl, hash);		\
458
	     !rht_is_a_nulls(pos);					\
459
	     pos = rcu_dereference_raw(pos->next))
460 461 462

/**
 * rht_for_each_rcu - iterate over rcu hash chain
463 464 465
 * @pos:	the &struct rhash_head to use as a loop cursor.
 * @tbl:	the &struct bucket_table
 * @hash:	the hash value / bucket index
466 467
 *
 * This hash chain list-traversal primitive may safely run concurrently with
468
 * the _rcu mutation primitives such as rhashtable_insert() as long as the
469 470
 * traversal is guarded by rcu_read_lock().
 */
471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489
#define rht_for_each_rcu(pos, tbl, hash)				\
	rht_for_each_rcu_continue(pos, (tbl)->buckets[hash], tbl, hash)

/**
 * rht_for_each_entry_rcu_continue - continue iterating over rcu hash chain
 * @tpos:	the type * to use as a loop cursor.
 * @pos:	the &struct rhash_head to use as a loop cursor.
 * @head:	the previous &struct rhash_head to continue from
 * @tbl:	the &struct bucket_table
 * @hash:	the hash value / bucket index
 * @member:	name of the &struct rhash_head within the hashable struct.
 *
 * This hash chain list-traversal primitive may safely run concurrently with
 * the _rcu mutation primitives such as rhashtable_insert() as long as the
 * traversal is guarded by rcu_read_lock().
 */
#define rht_for_each_entry_rcu_continue(tpos, pos, head, tbl, hash, member) \
	for (({barrier(); }),						    \
	     pos = rht_dereference_bucket_rcu(head, tbl, hash);		    \
490
	     (!rht_is_a_nulls(pos)) && rht_entry(tpos, pos, member);	    \
491
	     pos = rht_dereference_bucket_rcu(pos->next, tbl, hash))
492 493 494

/**
 * rht_for_each_entry_rcu - iterate over rcu hash chain of given type
495 496 497 498 499
 * @tpos:	the type * to use as a loop cursor.
 * @pos:	the &struct rhash_head to use as a loop cursor.
 * @tbl:	the &struct bucket_table
 * @hash:	the hash value / bucket index
 * @member:	name of the &struct rhash_head within the hashable struct.
500 501
 *
 * This hash chain list-traversal primitive may safely run concurrently with
502
 * the _rcu mutation primitives such as rhashtable_insert() as long as the
503 504
 * traversal is guarded by rcu_read_lock().
 */
505 506 507
#define rht_for_each_entry_rcu(tpos, pos, tbl, hash, member)		\
	rht_for_each_entry_rcu_continue(tpos, pos, (tbl)->buckets[hash],\
					tbl, hash, member)
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
static inline int rhashtable_compare(struct rhashtable_compare_arg *arg,
				     const void *obj)
{
	struct rhashtable *ht = arg->ht;
	const char *ptr = obj;

	return memcmp(ptr + ht->p.key_offset, arg->key, ht->p.key_len);
}

/**
 * rhashtable_lookup_fast - search hash table, inlined version
 * @ht:		hash table
 * @key:	the pointer to the key
 * @params:	hash table parameters
 *
 * Computes the hash value for the key and traverses the bucket chain looking
 * for a entry with an identical key. The first matching entry is returned.
 *
 * Returns the first entry on which the compare function returned true.
 */
static inline void *rhashtable_lookup_fast(
	struct rhashtable *ht, const void *key,
	const struct rhashtable_params params)
{
	struct rhashtable_compare_arg arg = {
		.ht = ht,
		.key = key,
	};
	const struct bucket_table *tbl;
	struct rhash_head *he;
539
	unsigned int hash;
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

	rcu_read_lock();

	tbl = rht_dereference_rcu(ht->tbl, ht);
restart:
	hash = rht_key_hashfn(ht, tbl, key, params);
	rht_for_each_rcu(he, tbl, hash) {
		if (params.obj_cmpfn ?
		    params.obj_cmpfn(&arg, rht_obj(ht, he)) :
		    rhashtable_compare(&arg, rht_obj(ht, he)))
			continue;
		rcu_read_unlock();
		return rht_obj(ht, he);
	}

	/* Ensure we see any new tables. */
	smp_rmb();

	tbl = rht_dereference_rcu(tbl->future_tbl, ht);
	if (unlikely(tbl))
		goto restart;
	rcu_read_unlock();

	return NULL;
}

566
/* Internal function, please use rhashtable_insert_fast() instead */
567 568 569 570 571 572 573 574 575 576 577
static inline int __rhashtable_insert_fast(
	struct rhashtable *ht, const void *key, struct rhash_head *obj,
	const struct rhashtable_params params)
{
	struct rhashtable_compare_arg arg = {
		.ht = ht,
		.key = key,
	};
	struct bucket_table *tbl, *new_tbl;
	struct rhash_head *head;
	spinlock_t *lock;
578 579
	unsigned int elasticity;
	unsigned int hash;
580
	int err;
581

582
restart:
583 584 585 586
	rcu_read_lock();

	tbl = rht_dereference_rcu(ht->tbl, ht);

587 588
	/* All insertions must grab the oldest table containing
	 * the hashed bucket that is yet to be rehashed.
589
	 */
590 591 592 593 594 595 596 597 598 599 600 601
	for (;;) {
		hash = rht_head_hashfn(ht, tbl, obj, params);
		lock = rht_bucket_lock(tbl, hash);
		spin_lock_bh(lock);

		if (tbl->rehash <= hash)
			break;

		spin_unlock_bh(lock);
		tbl = rht_dereference_rcu(tbl->future_tbl, ht);
	}

602 603
	new_tbl = rht_dereference_rcu(tbl->future_tbl, ht);
	if (unlikely(new_tbl)) {
604 605
		tbl = rhashtable_insert_slow(ht, key, obj, new_tbl);
		if (!IS_ERR_OR_NULL(tbl))
606
			goto slow_path;
607 608

		err = PTR_ERR(tbl);
609 610 611
		goto out;
	}

612 613 614 615
	err = -E2BIG;
	if (unlikely(rht_grow_above_max(ht, tbl)))
		goto out;

616 617 618
	if (unlikely(rht_grow_above_100(ht, tbl))) {
slow_path:
		spin_unlock_bh(lock);
619
		err = rhashtable_insert_rehash(ht, tbl);
620
		rcu_read_unlock();
621 622 623 624 625
		if (err)
			return err;

		goto restart;
	}
626

627 628
	err = -EEXIST;
	elasticity = ht->elasticity;
629
	rht_for_each(head, tbl, hash) {
630 631
		if (key &&
		    unlikely(!(params.obj_cmpfn ?
632 633 634
			       params.obj_cmpfn(&arg, rht_obj(ht, head)) :
			       rhashtable_compare(&arg, rht_obj(ht, head)))))
			goto out;
635 636
		if (!--elasticity)
			goto slow_path;
637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744
	}

	err = 0;

	head = rht_dereference_bucket(tbl->buckets[hash], tbl, hash);

	RCU_INIT_POINTER(obj->next, head);

	rcu_assign_pointer(tbl->buckets[hash], obj);

	atomic_inc(&ht->nelems);
	if (rht_grow_above_75(ht, tbl))
		schedule_work(&ht->run_work);

out:
	spin_unlock_bh(lock);
	rcu_read_unlock();

	return err;
}

/**
 * rhashtable_insert_fast - insert object into hash table
 * @ht:		hash table
 * @obj:	pointer to hash head inside object
 * @params:	hash table parameters
 *
 * Will take a per bucket spinlock to protect against mutual mutations
 * on the same bucket. Multiple insertions may occur in parallel unless
 * they map to the same bucket lock.
 *
 * It is safe to call this function from atomic context.
 *
 * Will trigger an automatic deferred table resizing if the size grows
 * beyond the watermark indicated by grow_decision() which can be passed
 * to rhashtable_init().
 */
static inline int rhashtable_insert_fast(
	struct rhashtable *ht, struct rhash_head *obj,
	const struct rhashtable_params params)
{
	return __rhashtable_insert_fast(ht, NULL, obj, params);
}

/**
 * rhashtable_lookup_insert_fast - lookup and insert object into hash table
 * @ht:		hash table
 * @obj:	pointer to hash head inside object
 * @params:	hash table parameters
 *
 * Locks down the bucket chain in both the old and new table if a resize
 * is in progress to ensure that writers can't remove from the old table
 * and can't insert to the new table during the atomic operation of search
 * and insertion. Searches for duplicates in both the old and new table if
 * a resize is in progress.
 *
 * This lookup function may only be used for fixed key hash table (key_len
 * parameter set). It will BUG() if used inappropriately.
 *
 * It is safe to call this function from atomic context.
 *
 * Will trigger an automatic deferred table resizing if the size grows
 * beyond the watermark indicated by grow_decision() which can be passed
 * to rhashtable_init().
 */
static inline int rhashtable_lookup_insert_fast(
	struct rhashtable *ht, struct rhash_head *obj,
	const struct rhashtable_params params)
{
	const char *key = rht_obj(ht, obj);

	BUG_ON(ht->p.obj_hashfn);

	return __rhashtable_insert_fast(ht, key + ht->p.key_offset, obj,
					params);
}

/**
 * rhashtable_lookup_insert_key - search and insert object to hash table
 *				  with explicit key
 * @ht:		hash table
 * @key:	key
 * @obj:	pointer to hash head inside object
 * @params:	hash table parameters
 *
 * Locks down the bucket chain in both the old and new table if a resize
 * is in progress to ensure that writers can't remove from the old table
 * and can't insert to the new table during the atomic operation of search
 * and insertion. Searches for duplicates in both the old and new table if
 * a resize is in progress.
 *
 * Lookups may occur in parallel with hashtable mutations and resizing.
 *
 * Will trigger an automatic deferred table resizing if the size grows
 * beyond the watermark indicated by grow_decision() which can be passed
 * to rhashtable_init().
 *
 * Returns zero on success.
 */
static inline int rhashtable_lookup_insert_key(
	struct rhashtable *ht, const void *key, struct rhash_head *obj,
	const struct rhashtable_params params)
{
	BUG_ON(!ht->p.obj_hashfn || !key);

	return __rhashtable_insert_fast(ht, key, obj, params);
}

745
/* Internal function, please use rhashtable_remove_fast() instead */
746 747 748 749 750 751 752
static inline int __rhashtable_remove_fast(
	struct rhashtable *ht, struct bucket_table *tbl,
	struct rhash_head *obj, const struct rhashtable_params params)
{
	struct rhash_head __rcu **pprev;
	struct rhash_head *he;
	spinlock_t * lock;
753
	unsigned int hash;
754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816
	int err = -ENOENT;

	hash = rht_head_hashfn(ht, tbl, obj, params);
	lock = rht_bucket_lock(tbl, hash);

	spin_lock_bh(lock);

	pprev = &tbl->buckets[hash];
	rht_for_each(he, tbl, hash) {
		if (he != obj) {
			pprev = &he->next;
			continue;
		}

		rcu_assign_pointer(*pprev, obj->next);
		err = 0;
		break;
	}

	spin_unlock_bh(lock);

	return err;
}

/**
 * rhashtable_remove_fast - remove object from hash table
 * @ht:		hash table
 * @obj:	pointer to hash head inside object
 * @params:	hash table parameters
 *
 * Since the hash chain is single linked, the removal operation needs to
 * walk the bucket chain upon removal. The removal operation is thus
 * considerable slow if the hash table is not correctly sized.
 *
 * Will automatically shrink the table via rhashtable_expand() if the
 * shrink_decision function specified at rhashtable_init() returns true.
 *
 * Returns zero on success, -ENOENT if the entry could not be found.
 */
static inline int rhashtable_remove_fast(
	struct rhashtable *ht, struct rhash_head *obj,
	const struct rhashtable_params params)
{
	struct bucket_table *tbl;
	int err;

	rcu_read_lock();

	tbl = rht_dereference_rcu(ht->tbl, ht);

	/* Because we have already taken (and released) the bucket
	 * lock in old_tbl, if we find that future_tbl is not yet
	 * visible then that guarantees the entry to still be in
	 * the old tbl if it exists.
	 */
	while ((err = __rhashtable_remove_fast(ht, tbl, obj, params)) &&
	       (tbl = rht_dereference_rcu(tbl->future_tbl, ht)))
		;

	if (err)
		goto out;

	atomic_dec(&ht->nelems);
817 818
	if (unlikely(ht->p.automatic_shrinking &&
		     rht_shrink_below_30(ht, tbl)))
819 820 821 822 823 824 825 826
		schedule_work(&ht->run_work);

out:
	rcu_read_unlock();

	return err;
}

827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908
/* Internal function, please use rhashtable_replace_fast() instead */
static inline int __rhashtable_replace_fast(
	struct rhashtable *ht, struct bucket_table *tbl,
	struct rhash_head *obj_old, struct rhash_head *obj_new,
	const struct rhashtable_params params)
{
	struct rhash_head __rcu **pprev;
	struct rhash_head *he;
	spinlock_t *lock;
	unsigned int hash;
	int err = -ENOENT;

	/* Minimally, the old and new objects must have same hash
	 * (which should mean identifiers are the same).
	 */
	hash = rht_head_hashfn(ht, tbl, obj_old, params);
	if (hash != rht_head_hashfn(ht, tbl, obj_new, params))
		return -EINVAL;

	lock = rht_bucket_lock(tbl, hash);

	spin_lock_bh(lock);

	pprev = &tbl->buckets[hash];
	rht_for_each(he, tbl, hash) {
		if (he != obj_old) {
			pprev = &he->next;
			continue;
		}

		rcu_assign_pointer(obj_new->next, obj_old->next);
		rcu_assign_pointer(*pprev, obj_new);
		err = 0;
		break;
	}

	spin_unlock_bh(lock);

	return err;
}

/**
 * rhashtable_replace_fast - replace an object in hash table
 * @ht:		hash table
 * @obj_old:	pointer to hash head inside object being replaced
 * @obj_new:	pointer to hash head inside object which is new
 * @params:	hash table parameters
 *
 * Replacing an object doesn't affect the number of elements in the hash table
 * or bucket, so we don't need to worry about shrinking or expanding the
 * table here.
 *
 * Returns zero on success, -ENOENT if the entry could not be found,
 * -EINVAL if hash is not the same for the old and new objects.
 */
static inline int rhashtable_replace_fast(
	struct rhashtable *ht, struct rhash_head *obj_old,
	struct rhash_head *obj_new,
	const struct rhashtable_params params)
{
	struct bucket_table *tbl;
	int err;

	rcu_read_lock();

	tbl = rht_dereference_rcu(ht->tbl, ht);

	/* Because we have already taken (and released) the bucket
	 * lock in old_tbl, if we find that future_tbl is not yet
	 * visible then that guarantees the entry to still be in
	 * the old tbl if it exists.
	 */
	while ((err = __rhashtable_replace_fast(ht, tbl, obj_old,
						obj_new, params)) &&
	       (tbl = rht_dereference_rcu(tbl->future_tbl, ht)))
		;

	rcu_read_unlock();

	return err;
}

909
#endif /* _LINUX_RHASHTABLE_H */