blob: 6c3c723e902bb42c194fbf1c979a2197a549ffc2 [file] [log] [blame]
Thomas Graf7e1e7762014-08-02 11:47:44 +02001/*
2 * Resizable, Scalable, Concurrent Hash Table
3 *
4 * Copyright (c) 2014 Thomas Graf <tgraf@suug.ch>
5 * Copyright (c) 2008-2014 Patrick McHardy <kaber@trash.net>
6 *
7 * Based on the following paper:
8 * https://www.usenix.org/legacy/event/atc11/tech/final_files/Triplett.pdf
9 *
10 * Code partially derived from nft_hash
11 *
12 * This program is free software; you can redistribute it and/or modify
13 * it under the terms of the GNU General Public License version 2 as
14 * published by the Free Software Foundation.
15 */
16
17#include <linux/kernel.h>
18#include <linux/init.h>
19#include <linux/log2.h>
20#include <linux/slab.h>
21#include <linux/vmalloc.h>
22#include <linux/mm.h>
Daniel Borkmann87545892014-12-10 16:33:11 +010023#include <linux/jhash.h>
Thomas Graf7e1e7762014-08-02 11:47:44 +020024#include <linux/random.h>
25#include <linux/rhashtable.h>
Thomas Graf7e1e7762014-08-02 11:47:44 +020026
27#define HASH_DEFAULT_SIZE 64UL
28#define HASH_MIN_SIZE 4UL
29
30#define ASSERT_RHT_MUTEX(HT) BUG_ON(!lockdep_rht_mutex_is_held(HT))
31
32#ifdef CONFIG_PROVE_LOCKING
33int lockdep_rht_mutex_is_held(const struct rhashtable *ht)
34{
Herbert Xu7b4ce232014-11-13 18:11:22 +080035 return ht->p.mutex_is_held(ht->p.parent);
Thomas Graf7e1e7762014-08-02 11:47:44 +020036}
37EXPORT_SYMBOL_GPL(lockdep_rht_mutex_is_held);
38#endif
39
Thomas Grafc91eee52014-08-13 16:38:30 +020040static void *rht_obj(const struct rhashtable *ht, const struct rhash_head *he)
Thomas Graf7e1e7762014-08-02 11:47:44 +020041{
42 return (void *) he - ht->p.head_offset;
43}
Thomas Graf7e1e7762014-08-02 11:47:44 +020044
45static u32 __hashfn(const struct rhashtable *ht, const void *key,
46 u32 len, u32 hsize)
47{
48 u32 h;
49
50 h = ht->p.hashfn(key, len, ht->p.hash_rnd);
51
52 return h & (hsize - 1);
53}
54
55/**
56 * rhashtable_hashfn - compute hash for key of given length
Geert Uytterhoeven45d5acd2014-08-08 17:19:14 +020057 * @ht: hash table to compute for
Thomas Graf7e1e7762014-08-02 11:47:44 +020058 * @key: pointer to key
59 * @len: length of key
60 *
61 * Computes the hash value using the hash function provided in the 'hashfn'
62 * of struct rhashtable_params. The returned value is guaranteed to be
63 * smaller than the number of buckets in the hash table.
64 */
65u32 rhashtable_hashfn(const struct rhashtable *ht, const void *key, u32 len)
66{
67 struct bucket_table *tbl = rht_dereference_rcu(ht->tbl, ht);
68
69 return __hashfn(ht, key, len, tbl->size);
70}
71EXPORT_SYMBOL_GPL(rhashtable_hashfn);
72
73static u32 obj_hashfn(const struct rhashtable *ht, const void *ptr, u32 hsize)
74{
75 if (unlikely(!ht->p.key_len)) {
76 u32 h;
77
78 h = ht->p.obj_hashfn(ptr, ht->p.hash_rnd);
79
80 return h & (hsize - 1);
81 }
82
83 return __hashfn(ht, ptr + ht->p.key_offset, ht->p.key_len, hsize);
84}
85
86/**
87 * rhashtable_obj_hashfn - compute hash for hashed object
Geert Uytterhoeven45d5acd2014-08-08 17:19:14 +020088 * @ht: hash table to compute for
Thomas Graf7e1e7762014-08-02 11:47:44 +020089 * @ptr: pointer to hashed object
90 *
91 * Computes the hash value using the hash function `hashfn` respectively
92 * 'obj_hashfn' depending on whether the hash table is set up to work with
93 * a fixed length key. The returned value is guaranteed to be smaller than
94 * the number of buckets in the hash table.
95 */
96u32 rhashtable_obj_hashfn(const struct rhashtable *ht, void *ptr)
97{
98 struct bucket_table *tbl = rht_dereference_rcu(ht->tbl, ht);
99
100 return obj_hashfn(ht, ptr, tbl->size);
101}
102EXPORT_SYMBOL_GPL(rhashtable_obj_hashfn);
103
104static u32 head_hashfn(const struct rhashtable *ht,
105 const struct rhash_head *he, u32 hsize)
106{
107 return obj_hashfn(ht, rht_obj(ht, he), hsize);
108}
109
Thomas Graf6eba8222014-11-13 13:45:46 +0100110static struct bucket_table *bucket_table_alloc(size_t nbuckets)
Thomas Graf7e1e7762014-08-02 11:47:44 +0200111{
112 struct bucket_table *tbl;
113 size_t size;
114
115 size = sizeof(*tbl) + nbuckets * sizeof(tbl->buckets[0]);
Thomas Graf6eba8222014-11-13 13:45:46 +0100116 tbl = kzalloc(size, GFP_KERNEL | __GFP_NOWARN);
Thomas Graf7e1e7762014-08-02 11:47:44 +0200117 if (tbl == NULL)
118 tbl = vzalloc(size);
119
120 if (tbl == NULL)
121 return NULL;
122
123 tbl->size = nbuckets;
124
125 return tbl;
126}
127
128static void bucket_table_free(const struct bucket_table *tbl)
129{
130 kvfree(tbl);
131}
132
133/**
134 * rht_grow_above_75 - returns true if nelems > 0.75 * table-size
135 * @ht: hash table
136 * @new_size: new table size
137 */
138bool rht_grow_above_75(const struct rhashtable *ht, size_t new_size)
139{
140 /* Expand table when exceeding 75% load */
141 return ht->nelems > (new_size / 4 * 3);
142}
143EXPORT_SYMBOL_GPL(rht_grow_above_75);
144
145/**
146 * rht_shrink_below_30 - returns true if nelems < 0.3 * table-size
147 * @ht: hash table
148 * @new_size: new table size
149 */
150bool rht_shrink_below_30(const struct rhashtable *ht, size_t new_size)
151{
152 /* Shrink table beneath 30% load */
153 return ht->nelems < (new_size * 3 / 10);
154}
155EXPORT_SYMBOL_GPL(rht_shrink_below_30);
156
157static void hashtable_chain_unzip(const struct rhashtable *ht,
158 const struct bucket_table *new_tbl,
159 struct bucket_table *old_tbl, size_t n)
160{
161 struct rhash_head *he, *p, *next;
162 unsigned int h;
163
164 /* Old bucket empty, no work needed. */
165 p = rht_dereference(old_tbl->buckets[n], ht);
166 if (!p)
167 return;
168
169 /* Advance the old bucket pointer one or more times until it
170 * reaches a node that doesn't hash to the same bucket as the
171 * previous node p. Call the previous node p;
172 */
173 h = head_hashfn(ht, p, new_tbl->size);
174 rht_for_each(he, p->next, ht) {
175 if (head_hashfn(ht, he, new_tbl->size) != h)
176 break;
177 p = he;
178 }
179 RCU_INIT_POINTER(old_tbl->buckets[n], p->next);
180
181 /* Find the subsequent node which does hash to the same
182 * bucket as node P, or NULL if no such node exists.
183 */
184 next = NULL;
185 if (he) {
186 rht_for_each(he, he->next, ht) {
187 if (head_hashfn(ht, he, new_tbl->size) == h) {
188 next = he;
189 break;
190 }
191 }
192 }
193
194 /* Set p's next pointer to that subsequent node pointer,
195 * bypassing the nodes which do not hash to p's bucket
196 */
197 RCU_INIT_POINTER(p->next, next);
198}
199
200/**
201 * rhashtable_expand - Expand hash table while allowing concurrent lookups
202 * @ht: the hash table to expand
Thomas Graf7e1e7762014-08-02 11:47:44 +0200203 *
204 * A secondary bucket array is allocated and the hash entries are migrated
205 * while keeping them on both lists until the end of the RCU grace period.
206 *
207 * This function may only be called in a context where it is safe to call
208 * synchronize_rcu(), e.g. not within a rcu_read_lock() section.
209 *
210 * The caller must ensure that no concurrent table mutations take place.
211 * It is however valid to have concurrent lookups if they are RCU protected.
212 */
Thomas Graf6eba8222014-11-13 13:45:46 +0100213int rhashtable_expand(struct rhashtable *ht)
Thomas Graf7e1e7762014-08-02 11:47:44 +0200214{
215 struct bucket_table *new_tbl, *old_tbl = rht_dereference(ht->tbl, ht);
216 struct rhash_head *he;
217 unsigned int i, h;
218 bool complete;
219
220 ASSERT_RHT_MUTEX(ht);
221
222 if (ht->p.max_shift && ht->shift >= ht->p.max_shift)
223 return 0;
224
Thomas Graf6eba8222014-11-13 13:45:46 +0100225 new_tbl = bucket_table_alloc(old_tbl->size * 2);
Thomas Graf7e1e7762014-08-02 11:47:44 +0200226 if (new_tbl == NULL)
227 return -ENOMEM;
228
229 ht->shift++;
230
231 /* For each new bucket, search the corresponding old bucket
Herbert Xu0c828f22014-11-13 13:10:48 +0800232 * for the first entry that hashes to the new bucket, and
Thomas Graf7e1e7762014-08-02 11:47:44 +0200233 * link the new bucket to that entry. Since all the entries
234 * which will end up in the new bucket appear in the same
235 * old bucket, this constructs an entirely valid new hash
236 * table, but with multiple buckets "zipped" together into a
237 * single imprecise chain.
238 */
239 for (i = 0; i < new_tbl->size; i++) {
240 h = i & (old_tbl->size - 1);
241 rht_for_each(he, old_tbl->buckets[h], ht) {
242 if (head_hashfn(ht, he, new_tbl->size) == i) {
243 RCU_INIT_POINTER(new_tbl->buckets[i], he);
244 break;
245 }
246 }
247 }
248
249 /* Publish the new table pointer. Lookups may now traverse
Herbert Xu0c828f22014-11-13 13:10:48 +0800250 * the new table, but they will not benefit from any
251 * additional efficiency until later steps unzip the buckets.
Thomas Graf7e1e7762014-08-02 11:47:44 +0200252 */
253 rcu_assign_pointer(ht->tbl, new_tbl);
254
255 /* Unzip interleaved hash chains */
256 do {
257 /* Wait for readers. All new readers will see the new
258 * table, and thus no references to the old table will
259 * remain.
260 */
261 synchronize_rcu();
262
263 /* For each bucket in the old table (each of which
264 * contains items from multiple buckets of the new
265 * table): ...
266 */
267 complete = true;
268 for (i = 0; i < old_tbl->size; i++) {
269 hashtable_chain_unzip(ht, new_tbl, old_tbl, i);
270 if (old_tbl->buckets[i] != NULL)
271 complete = false;
272 }
273 } while (!complete);
274
275 bucket_table_free(old_tbl);
276 return 0;
277}
278EXPORT_SYMBOL_GPL(rhashtable_expand);
279
280/**
281 * rhashtable_shrink - Shrink hash table while allowing concurrent lookups
282 * @ht: the hash table to shrink
Thomas Graf7e1e7762014-08-02 11:47:44 +0200283 *
284 * This function may only be called in a context where it is safe to call
285 * synchronize_rcu(), e.g. not within a rcu_read_lock() section.
286 *
287 * The caller must ensure that no concurrent table mutations take place.
288 * It is however valid to have concurrent lookups if they are RCU protected.
289 */
Thomas Graf6eba8222014-11-13 13:45:46 +0100290int rhashtable_shrink(struct rhashtable *ht)
Thomas Graf7e1e7762014-08-02 11:47:44 +0200291{
292 struct bucket_table *ntbl, *tbl = rht_dereference(ht->tbl, ht);
293 struct rhash_head __rcu **pprev;
294 unsigned int i;
295
296 ASSERT_RHT_MUTEX(ht);
297
Ying Xue94000172014-09-03 09:22:36 +0800298 if (ht->shift <= ht->p.min_shift)
Thomas Graf7e1e7762014-08-02 11:47:44 +0200299 return 0;
300
Thomas Graf6eba8222014-11-13 13:45:46 +0100301 ntbl = bucket_table_alloc(tbl->size / 2);
Thomas Graf7e1e7762014-08-02 11:47:44 +0200302 if (ntbl == NULL)
303 return -ENOMEM;
304
305 ht->shift--;
306
Herbert Xu0c828f22014-11-13 13:10:48 +0800307 /* Link each bucket in the new table to the first bucket
Thomas Graf7e1e7762014-08-02 11:47:44 +0200308 * in the old table that contains entries which will hash
309 * to the new bucket.
310 */
311 for (i = 0; i < ntbl->size; i++) {
312 ntbl->buckets[i] = tbl->buckets[i];
313
Herbert Xu0c828f22014-11-13 13:10:48 +0800314 /* Link each bucket in the new table to the first bucket
Thomas Graf7e1e7762014-08-02 11:47:44 +0200315 * in the old table that contains entries which will hash
316 * to the new bucket.
317 */
318 for (pprev = &ntbl->buckets[i]; *pprev != NULL;
319 pprev = &rht_dereference(*pprev, ht)->next)
320 ;
321 RCU_INIT_POINTER(*pprev, tbl->buckets[i + ntbl->size]);
322 }
323
324 /* Publish the new, valid hash table */
325 rcu_assign_pointer(ht->tbl, ntbl);
326
327 /* Wait for readers. No new readers will have references to the
328 * old hash table.
329 */
330 synchronize_rcu();
331
332 bucket_table_free(tbl);
333
334 return 0;
335}
336EXPORT_SYMBOL_GPL(rhashtable_shrink);
337
338/**
339 * rhashtable_insert - insert object into hash hash table
340 * @ht: hash table
341 * @obj: pointer to hash head inside object
Thomas Graf7e1e7762014-08-02 11:47:44 +0200342 *
343 * Will automatically grow the table via rhashtable_expand() if the the
344 * grow_decision function specified at rhashtable_init() returns true.
345 *
346 * The caller must ensure that no concurrent table mutations occur. It is
347 * however valid to have concurrent lookups if they are RCU protected.
348 */
Thomas Graf6eba8222014-11-13 13:45:46 +0100349void rhashtable_insert(struct rhashtable *ht, struct rhash_head *obj)
Thomas Graf7e1e7762014-08-02 11:47:44 +0200350{
351 struct bucket_table *tbl = rht_dereference(ht->tbl, ht);
352 u32 hash;
353
354 ASSERT_RHT_MUTEX(ht);
355
356 hash = head_hashfn(ht, obj, tbl->size);
357 RCU_INIT_POINTER(obj->next, tbl->buckets[hash]);
358 rcu_assign_pointer(tbl->buckets[hash], obj);
359 ht->nelems++;
360
361 if (ht->p.grow_decision && ht->p.grow_decision(ht, tbl->size))
Thomas Graf6eba8222014-11-13 13:45:46 +0100362 rhashtable_expand(ht);
Thomas Graf7e1e7762014-08-02 11:47:44 +0200363}
364EXPORT_SYMBOL_GPL(rhashtable_insert);
365
366/**
367 * rhashtable_remove_pprev - remove object from hash table given previous element
368 * @ht: hash table
369 * @obj: pointer to hash head inside object
370 * @pprev: pointer to previous element
Thomas Graf7e1e7762014-08-02 11:47:44 +0200371 *
372 * Identical to rhashtable_remove() but caller is alreayd aware of the element
373 * in front of the element to be deleted. This is in particular useful for
374 * deletion when combined with walking or lookup.
375 */
376void rhashtable_remove_pprev(struct rhashtable *ht, struct rhash_head *obj,
Thomas Graf6eba8222014-11-13 13:45:46 +0100377 struct rhash_head __rcu **pprev)
Thomas Graf7e1e7762014-08-02 11:47:44 +0200378{
379 struct bucket_table *tbl = rht_dereference(ht->tbl, ht);
380
381 ASSERT_RHT_MUTEX(ht);
382
383 RCU_INIT_POINTER(*pprev, obj->next);
384 ht->nelems--;
385
386 if (ht->p.shrink_decision &&
387 ht->p.shrink_decision(ht, tbl->size))
Thomas Graf6eba8222014-11-13 13:45:46 +0100388 rhashtable_shrink(ht);
Thomas Graf7e1e7762014-08-02 11:47:44 +0200389}
390EXPORT_SYMBOL_GPL(rhashtable_remove_pprev);
391
392/**
393 * rhashtable_remove - remove object from hash table
394 * @ht: hash table
395 * @obj: pointer to hash head inside object
Thomas Graf7e1e7762014-08-02 11:47:44 +0200396 *
397 * Since the hash chain is single linked, the removal operation needs to
398 * walk the bucket chain upon removal. The removal operation is thus
399 * considerable slow if the hash table is not correctly sized.
400 *
401 * Will automatically shrink the table via rhashtable_expand() if the the
402 * shrink_decision function specified at rhashtable_init() returns true.
403 *
404 * The caller must ensure that no concurrent table mutations occur. It is
405 * however valid to have concurrent lookups if they are RCU protected.
406 */
Thomas Graf6eba8222014-11-13 13:45:46 +0100407bool rhashtable_remove(struct rhashtable *ht, struct rhash_head *obj)
Thomas Graf7e1e7762014-08-02 11:47:44 +0200408{
409 struct bucket_table *tbl = rht_dereference(ht->tbl, ht);
410 struct rhash_head __rcu **pprev;
411 struct rhash_head *he;
412 u32 h;
413
414 ASSERT_RHT_MUTEX(ht);
415
416 h = head_hashfn(ht, obj, tbl->size);
417
418 pprev = &tbl->buckets[h];
419 rht_for_each(he, tbl->buckets[h], ht) {
420 if (he != obj) {
421 pprev = &he->next;
422 continue;
423 }
424
Thomas Graf6eba8222014-11-13 13:45:46 +0100425 rhashtable_remove_pprev(ht, he, pprev);
Thomas Graf7e1e7762014-08-02 11:47:44 +0200426 return true;
427 }
428
429 return false;
430}
431EXPORT_SYMBOL_GPL(rhashtable_remove);
432
433/**
434 * rhashtable_lookup - lookup key in hash table
435 * @ht: hash table
436 * @key: pointer to key
437 *
438 * Computes the hash value for the key and traverses the bucket chain looking
439 * for a entry with an identical key. The first matching entry is returned.
440 *
441 * This lookup function may only be used for fixed key hash table (key_len
442 * paramter set). It will BUG() if used inappropriately.
443 *
444 * Lookups may occur in parallel with hash mutations as long as the lookup is
445 * guarded by rcu_read_lock(). The caller must take care of this.
446 */
447void *rhashtable_lookup(const struct rhashtable *ht, const void *key)
448{
449 const struct bucket_table *tbl = rht_dereference_rcu(ht->tbl, ht);
450 struct rhash_head *he;
451 u32 h;
452
453 BUG_ON(!ht->p.key_len);
454
455 h = __hashfn(ht, key, ht->p.key_len, tbl->size);
456 rht_for_each_rcu(he, tbl->buckets[h], ht) {
457 if (memcmp(rht_obj(ht, he) + ht->p.key_offset, key,
458 ht->p.key_len))
459 continue;
460 return (void *) he - ht->p.head_offset;
461 }
462
463 return NULL;
464}
465EXPORT_SYMBOL_GPL(rhashtable_lookup);
466
467/**
468 * rhashtable_lookup_compare - search hash table with compare function
469 * @ht: hash table
470 * @hash: hash value of desired entry
471 * @compare: compare function, must return true on match
472 * @arg: argument passed on to compare function
473 *
474 * Traverses the bucket chain behind the provided hash value and calls the
475 * specified compare function for each entry.
476 *
477 * Lookups may occur in parallel with hash mutations as long as the lookup is
478 * guarded by rcu_read_lock(). The caller must take care of this.
479 *
480 * Returns the first entry on which the compare function returned true.
481 */
482void *rhashtable_lookup_compare(const struct rhashtable *ht, u32 hash,
483 bool (*compare)(void *, void *), void *arg)
484{
485 const struct bucket_table *tbl = rht_dereference_rcu(ht->tbl, ht);
486 struct rhash_head *he;
487
488 if (unlikely(hash >= tbl->size))
489 return NULL;
490
491 rht_for_each_rcu(he, tbl->buckets[hash], ht) {
492 if (!compare(rht_obj(ht, he), arg))
493 continue;
494 return (void *) he - ht->p.head_offset;
495 }
496
497 return NULL;
498}
499EXPORT_SYMBOL_GPL(rhashtable_lookup_compare);
500
Ying Xue94000172014-09-03 09:22:36 +0800501static size_t rounded_hashtable_size(struct rhashtable_params *params)
Thomas Graf7e1e7762014-08-02 11:47:44 +0200502{
Ying Xue94000172014-09-03 09:22:36 +0800503 return max(roundup_pow_of_two(params->nelem_hint * 4 / 3),
504 1UL << params->min_shift);
Thomas Graf7e1e7762014-08-02 11:47:44 +0200505}
506
507/**
508 * rhashtable_init - initialize a new hash table
509 * @ht: hash table to be initialized
510 * @params: configuration parameters
511 *
512 * Initializes a new hash table based on the provided configuration
513 * parameters. A table can be configured either with a variable or
514 * fixed length key:
515 *
516 * Configuration Example 1: Fixed length keys
517 * struct test_obj {
518 * int key;
519 * void * my_member;
520 * struct rhash_head node;
521 * };
522 *
523 * struct rhashtable_params params = {
524 * .head_offset = offsetof(struct test_obj, node),
525 * .key_offset = offsetof(struct test_obj, key),
526 * .key_len = sizeof(int),
Daniel Borkmann87545892014-12-10 16:33:11 +0100527 * .hashfn = jhash,
Herbert Xu1b2f3092014-11-13 18:11:20 +0800528 * #ifdef CONFIG_PROVE_LOCKING
Thomas Graf7e1e7762014-08-02 11:47:44 +0200529 * .mutex_is_held = &my_mutex_is_held,
Herbert Xu1b2f3092014-11-13 18:11:20 +0800530 * #endif
Thomas Graf7e1e7762014-08-02 11:47:44 +0200531 * };
532 *
533 * Configuration Example 2: Variable length keys
534 * struct test_obj {
535 * [...]
536 * struct rhash_head node;
537 * };
538 *
539 * u32 my_hash_fn(const void *data, u32 seed)
540 * {
541 * struct test_obj *obj = data;
542 *
543 * return [... hash ...];
544 * }
545 *
546 * struct rhashtable_params params = {
547 * .head_offset = offsetof(struct test_obj, node),
Daniel Borkmann87545892014-12-10 16:33:11 +0100548 * .hashfn = jhash,
Thomas Graf7e1e7762014-08-02 11:47:44 +0200549 * .obj_hashfn = my_hash_fn,
Herbert Xu1b2f3092014-11-13 18:11:20 +0800550 * #ifdef CONFIG_PROVE_LOCKING
Thomas Graf7e1e7762014-08-02 11:47:44 +0200551 * .mutex_is_held = &my_mutex_is_held,
Herbert Xu1b2f3092014-11-13 18:11:20 +0800552 * #endif
Thomas Graf7e1e7762014-08-02 11:47:44 +0200553 * };
554 */
555int rhashtable_init(struct rhashtable *ht, struct rhashtable_params *params)
556{
557 struct bucket_table *tbl;
558 size_t size;
559
560 size = HASH_DEFAULT_SIZE;
561
562 if ((params->key_len && !params->hashfn) ||
563 (!params->key_len && !params->obj_hashfn))
564 return -EINVAL;
565
Ying Xue94000172014-09-03 09:22:36 +0800566 params->min_shift = max_t(size_t, params->min_shift,
567 ilog2(HASH_MIN_SIZE));
568
Thomas Graf7e1e7762014-08-02 11:47:44 +0200569 if (params->nelem_hint)
Ying Xue94000172014-09-03 09:22:36 +0800570 size = rounded_hashtable_size(params);
Thomas Graf7e1e7762014-08-02 11:47:44 +0200571
Thomas Graf6eba8222014-11-13 13:45:46 +0100572 tbl = bucket_table_alloc(size);
Thomas Graf7e1e7762014-08-02 11:47:44 +0200573 if (tbl == NULL)
574 return -ENOMEM;
575
576 memset(ht, 0, sizeof(*ht));
577 ht->shift = ilog2(tbl->size);
578 memcpy(&ht->p, params, sizeof(*params));
579 RCU_INIT_POINTER(ht->tbl, tbl);
580
581 if (!ht->p.hash_rnd)
582 get_random_bytes(&ht->p.hash_rnd, sizeof(ht->p.hash_rnd));
583
584 return 0;
585}
586EXPORT_SYMBOL_GPL(rhashtable_init);
587
588/**
589 * rhashtable_destroy - destroy hash table
590 * @ht: the hash table to destroy
591 *
Pablo Neira Ayusoae82ddc2014-09-02 00:26:05 +0200592 * Frees the bucket array. This function is not rcu safe, therefore the caller
593 * has to make sure that no resizing may happen by unpublishing the hashtable
594 * and waiting for the quiescent cycle before releasing the bucket array.
Thomas Graf7e1e7762014-08-02 11:47:44 +0200595 */
596void rhashtable_destroy(const struct rhashtable *ht)
597{
Pablo Neira Ayusoae82ddc2014-09-02 00:26:05 +0200598 bucket_table_free(ht->tbl);
Thomas Graf7e1e7762014-08-02 11:47:44 +0200599}
600EXPORT_SYMBOL_GPL(rhashtable_destroy);
601
602/**************************************************************************
603 * Self Test
604 **************************************************************************/
605
606#ifdef CONFIG_TEST_RHASHTABLE
607
608#define TEST_HT_SIZE 8
609#define TEST_ENTRIES 2048
610#define TEST_PTR ((void *) 0xdeadbeef)
611#define TEST_NEXPANDS 4
612
Herbert Xu1b2f3092014-11-13 18:11:20 +0800613#ifdef CONFIG_PROVE_LOCKING
Herbert Xu7b4ce232014-11-13 18:11:22 +0800614static int test_mutex_is_held(void *parent)
Thomas Graf7e1e7762014-08-02 11:47:44 +0200615{
616 return 1;
617}
Herbert Xu1b2f3092014-11-13 18:11:20 +0800618#endif
Thomas Graf7e1e7762014-08-02 11:47:44 +0200619
620struct test_obj {
621 void *ptr;
622 int value;
623 struct rhash_head node;
624};
625
626static int __init test_rht_lookup(struct rhashtable *ht)
627{
628 unsigned int i;
629
630 for (i = 0; i < TEST_ENTRIES * 2; i++) {
631 struct test_obj *obj;
632 bool expected = !(i % 2);
633 u32 key = i;
634
635 obj = rhashtable_lookup(ht, &key);
636
637 if (expected && !obj) {
638 pr_warn("Test failed: Could not find key %u\n", key);
639 return -ENOENT;
640 } else if (!expected && obj) {
641 pr_warn("Test failed: Unexpected entry found for key %u\n",
642 key);
643 return -EEXIST;
644 } else if (expected && obj) {
645 if (obj->ptr != TEST_PTR || obj->value != i) {
646 pr_warn("Test failed: Lookup value mismatch %p!=%p, %u!=%u\n",
647 obj->ptr, TEST_PTR, obj->value, i);
648 return -EINVAL;
649 }
650 }
651 }
652
653 return 0;
654}
655
Thomas Graf3e7b2ec2014-11-24 12:37:58 +0100656static void test_bucket_stats(struct rhashtable *ht, bool quiet)
Thomas Graf7e1e7762014-08-02 11:47:44 +0200657{
Thomas Graf3e7b2ec2014-11-24 12:37:58 +0100658 unsigned int cnt, rcu_cnt, i, total = 0;
Thomas Graf7e1e7762014-08-02 11:47:44 +0200659 struct test_obj *obj;
Thomas Graf3e7b2ec2014-11-24 12:37:58 +0100660 struct bucket_table *tbl;
Thomas Graf7e1e7762014-08-02 11:47:44 +0200661
Thomas Graf3e7b2ec2014-11-24 12:37:58 +0100662 tbl = rht_dereference_rcu(ht->tbl, ht);
Thomas Graf7e1e7762014-08-02 11:47:44 +0200663 for (i = 0; i < tbl->size; i++) {
Thomas Graf3e7b2ec2014-11-24 12:37:58 +0100664 rcu_cnt = cnt = 0;
Thomas Graf7e1e7762014-08-02 11:47:44 +0200665
666 if (!quiet)
667 pr_info(" [%#4x/%zu]", i, tbl->size);
668
669 rht_for_each_entry_rcu(obj, tbl->buckets[i], node) {
670 cnt++;
671 total++;
672 if (!quiet)
673 pr_cont(" [%p],", obj);
674 }
675
Thomas Graf3e7b2ec2014-11-24 12:37:58 +0100676 rht_for_each_entry_rcu(obj, tbl->buckets[i], node)
677 rcu_cnt++;
678
679 if (rcu_cnt != cnt)
680 pr_warn("Test failed: Chain count mismach %d != %d",
681 cnt, rcu_cnt);
682
Thomas Graf7e1e7762014-08-02 11:47:44 +0200683 if (!quiet)
684 pr_cont("\n [%#x] first element: %p, chain length: %u\n",
685 i, tbl->buckets[i], cnt);
686 }
687
688 pr_info(" Traversal complete: counted=%u, nelems=%zu, entries=%d\n",
689 total, ht->nelems, TEST_ENTRIES);
Thomas Graf3e7b2ec2014-11-24 12:37:58 +0100690
691 if (total != ht->nelems || total != TEST_ENTRIES)
692 pr_warn("Test failed: Total count mismatch ^^^");
Thomas Graf7e1e7762014-08-02 11:47:44 +0200693}
694
695static int __init test_rhashtable(struct rhashtable *ht)
696{
697 struct bucket_table *tbl;
698 struct test_obj *obj, *next;
699 int err;
700 unsigned int i;
701
702 /*
703 * Insertion Test:
704 * Insert TEST_ENTRIES into table with all keys even numbers
705 */
706 pr_info(" Adding %d keys\n", TEST_ENTRIES);
707 for (i = 0; i < TEST_ENTRIES; i++) {
708 struct test_obj *obj;
709
710 obj = kzalloc(sizeof(*obj), GFP_KERNEL);
711 if (!obj) {
712 err = -ENOMEM;
713 goto error;
714 }
715
716 obj->ptr = TEST_PTR;
717 obj->value = i * 2;
718
Thomas Graf6eba8222014-11-13 13:45:46 +0100719 rhashtable_insert(ht, &obj->node);
Thomas Graf7e1e7762014-08-02 11:47:44 +0200720 }
721
722 rcu_read_lock();
Thomas Graf3e7b2ec2014-11-24 12:37:58 +0100723 test_bucket_stats(ht, true);
Thomas Graf7e1e7762014-08-02 11:47:44 +0200724 test_rht_lookup(ht);
725 rcu_read_unlock();
726
727 for (i = 0; i < TEST_NEXPANDS; i++) {
728 pr_info(" Table expansion iteration %u...\n", i);
Thomas Graf6eba8222014-11-13 13:45:46 +0100729 rhashtable_expand(ht);
Thomas Graf7e1e7762014-08-02 11:47:44 +0200730
731 rcu_read_lock();
732 pr_info(" Verifying lookups...\n");
733 test_rht_lookup(ht);
734 rcu_read_unlock();
735 }
736
737 for (i = 0; i < TEST_NEXPANDS; i++) {
738 pr_info(" Table shrinkage iteration %u...\n", i);
Thomas Graf6eba8222014-11-13 13:45:46 +0100739 rhashtable_shrink(ht);
Thomas Graf7e1e7762014-08-02 11:47:44 +0200740
741 rcu_read_lock();
742 pr_info(" Verifying lookups...\n");
743 test_rht_lookup(ht);
744 rcu_read_unlock();
745 }
746
Thomas Graf3e7b2ec2014-11-24 12:37:58 +0100747 rcu_read_lock();
748 test_bucket_stats(ht, true);
749 rcu_read_unlock();
750
Thomas Graf7e1e7762014-08-02 11:47:44 +0200751 pr_info(" Deleting %d keys\n", TEST_ENTRIES);
752 for (i = 0; i < TEST_ENTRIES; i++) {
753 u32 key = i * 2;
754
755 obj = rhashtable_lookup(ht, &key);
756 BUG_ON(!obj);
757
Thomas Graf6eba8222014-11-13 13:45:46 +0100758 rhashtable_remove(ht, &obj->node);
Thomas Graf7e1e7762014-08-02 11:47:44 +0200759 kfree(obj);
760 }
761
762 return 0;
763
764error:
765 tbl = rht_dereference_rcu(ht->tbl, ht);
766 for (i = 0; i < tbl->size; i++)
767 rht_for_each_entry_safe(obj, next, tbl->buckets[i], ht, node)
768 kfree(obj);
769
770 return err;
771}
772
773static int __init test_rht_init(void)
774{
775 struct rhashtable ht;
776 struct rhashtable_params params = {
777 .nelem_hint = TEST_HT_SIZE,
778 .head_offset = offsetof(struct test_obj, node),
779 .key_offset = offsetof(struct test_obj, value),
780 .key_len = sizeof(int),
Daniel Borkmann87545892014-12-10 16:33:11 +0100781 .hashfn = jhash,
Herbert Xu1b2f3092014-11-13 18:11:20 +0800782#ifdef CONFIG_PROVE_LOCKING
Thomas Graf7e1e7762014-08-02 11:47:44 +0200783 .mutex_is_held = &test_mutex_is_held,
Herbert Xu1b2f3092014-11-13 18:11:20 +0800784#endif
Thomas Graf7e1e7762014-08-02 11:47:44 +0200785 .grow_decision = rht_grow_above_75,
786 .shrink_decision = rht_shrink_below_30,
787 };
788 int err;
789
790 pr_info("Running resizable hashtable tests...\n");
791
792 err = rhashtable_init(&ht, &params);
793 if (err < 0) {
794 pr_warn("Test failed: Unable to initialize hashtable: %d\n",
795 err);
796 return err;
797 }
798
799 err = test_rhashtable(&ht);
800
801 rhashtable_destroy(&ht);
802
803 return err;
804}
805
806subsys_initcall(test_rht_init);
807
808#endif /* CONFIG_TEST_RHASHTABLE */