src/util/hash_table.c

   1 /*
   2  * Copyright © 2009,2012 Intel Corporation
   3  * Copyright © 1988-2004 Keith Packard and Bart Massey.
   4  *
   5  * Permission is hereby granted, free of charge, to any person obtaining a
   6  * copy of this software and associated documentation files (the "Software"),
   7  * to deal in the Software without restriction, including without limitation
   8  * the rights to use, copy, modify, merge, publish, distribute, sublicense,
   9  * and/or sell copies of the Software, and to permit persons to whom the
  10  * Software is furnished to do so, subject to the following conditions:
  11  *
  12  * The above copyright notice and this permission notice (including the next
  13  * paragraph) shall be included in all copies or substantial portions of the
  14  * Software.
  15  *
  16  * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
  17  * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
  18  * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
  19  * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
  20  * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
  21  * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
  22  * IN THE SOFTWARE.
  23  *
  24  * Except as contained in this notice, the names of the authors
  25  * or their institutions shall not be used in advertising or
  26  * otherwise to promote the sale, use or other dealings in this
  27  * Software without prior written authorization from the
  28  * authors.
  29  *
  30  * Authors:
  31  *    Eric Anholt <eric@anholt.net>
  32  *    Keith Packard <keithp@keithp.com>
  33  */
  34
  35 /**
  36  * Implements an open-addressing, linear-reprobing hash table.
  37  *
  38  * For more information, see:
  39  *
  40  * http://cgit.freedesktop.org/~anholt/hash_table/tree/README
  41  */
  42
  43 #include <stdlib.h>
  44 #include <string.h>
  45 #include <assert.h>
  46
  47 #include "hash_table.h"
  48 #include "ralloc.h"
  49 #include "macros.h"
  50
  51 static const uint32_t deleted_key_value;
  52
  53 /**
  54  * From Knuth -- a good choice for hash/rehash values is p, p-2 where
  55  * p and p-2 are both prime.  These tables are sized to have an extra 10%
  56  * free to avoid exponential performance degradation as the hash table fills
  57  */
  58 static const struct {
  59    uint32_t max_entries, size, rehash;
  60 } hash_sizes[] = {
  61    { 2,                 5,              3         },
  62    { 4,                 7,              5         },
  63    { 8,                 13,             11        },
  64    { 16,                19,             17        },
  65    { 32,                43,             41        },
  66    { 64,                73,             71        },
  67    { 128,               151,            149       },
  68    { 256,               283,            281       },
  69    { 512,               571,            569       },
  70    { 1024,              1153,           1151      },
  71    { 2048,              2269,           2267      },
  72    { 4096,              4519,           4517      },
  73    { 8192,              9013,           9011      },
  74    { 16384,             18043,          18041     },
  75    { 32768,             36109,          36107     },
  76    { 65536,             72091,          72089     },
  77    { 131072,            144409,         144407    },
  78    { 262144,            288361,         288359    },
  79    { 524288,            576883,         576881    },
  80    { 1048576,           1153459,        1153457   },
  81    { 2097152,           2307163,        2307161   },
  82    { 4194304,           4613893,        4613891   },
  83    { 8388608,           9227641,        9227639   },
  84    { 16777216,          18455029,       18455027  },
  85    { 33554432,          36911011,       36911009  },
  86    { 67108864,          73819861,       73819859  },
  87    { 134217728,         147639589,      147639587 },
  88    { 268435456,         295279081,      295279079 },
  89    { 536870912,         590559793,      590559791 },
  90    { 1073741824,        1181116273,     1181116271},
  91    { 2147483648ul,      2362232233ul,   2362232231ul}
  92 };
  93
  94 static int
  95 entry_is_free(const struct hash_entry *entry)
  96 {
  97    return entry->key == NULL;
  98 }
  99
 100 static int
 101 entry_is_deleted(const struct hash_table *ht, struct hash_entry *entry)
 102 {
 103    return entry->key == ht->deleted_key;
 104 }
 105
 106 static int
 107 entry_is_present(const struct hash_table *ht, struct hash_entry *entry)
 108 {
 109    return entry->key != NULL && entry->key != ht->deleted_key;
 110 }
 111
 112 struct hash_table *
 113 _mesa_hash_table_create(void *mem_ctx,
 114                         uint32_t (*key_hash_function)(const void *key),
 115                         bool (*key_equals_function)(const void *a,
 116                                                     const void *b))
 117 {
 118    struct hash_table *ht;
 119
 120    ht = ralloc(mem_ctx, struct hash_table);
 121    if (ht == NULL)
 122       return NULL;
 123
 124    ht->size_index = 0;
 125    ht->size = hash_sizes[ht->size_index].size;
 126    ht->rehash = hash_sizes[ht->size_index].rehash;
 127    ht->max_entries = hash_sizes[ht->size_index].max_entries;
 128    ht->key_hash_function = key_hash_function;
 129    ht->key_equals_function = key_equals_function;
 130    ht->table = rzalloc_array(ht, struct hash_entry, ht->size);
 131    ht->entries = 0;
 132    ht->deleted_entries = 0;
 133    ht->deleted_key = &deleted_key_value;
 134
 135    if (ht->table == NULL) {
 136       ralloc_free(ht);
 137       return NULL;
 138    }
 139
 140    return ht;
 141 }
 142
 143 /**
 144  * Frees the given hash table.
 145  *
 146  * If delete_function is passed, it gets called on each entry present before
 147  * freeing.
 148  */
 149 void
 150 _mesa_hash_table_destroy(struct hash_table *ht,
 151                          void (*delete_function)(struct hash_entry *entry))
 152 {
 153    if (!ht)
 154       return;
 155
 156    if (delete_function) {
 157       struct hash_entry *entry;
 158
 159       hash_table_foreach(ht, entry) {
 160          delete_function(entry);
 161       }
 162    }
 163    ralloc_free(ht);
 164 }
 165
 166 /**
 167  * Deletes all entries of the given hash table without deleting the table
 168  * itself or changing its structure.
 169  *
 170  * If delete_function is passed, it gets called on each entry present.
 171  */
 172 void
 173 _mesa_hash_table_clear(struct hash_table *ht,
 174                        void (*delete_function)(struct hash_entry *entry))
 175 {
 176    struct hash_entry *entry;
 177
 178    for (entry = ht->table; entry != ht->table + ht->size; entry++) {
 179       if (entry->key == NULL)
 180          continue;
 181
 182       if (delete_function != NULL && entry->key != ht->deleted_key)
 183          delete_function(entry);
 184
 185       entry->key = NULL;
 186    }
 187
 188    ht->entries = 0;
 189    ht->deleted_entries = 0;
 190 }
 191
 192 /** Sets the value of the key pointer used for deleted entries in the table.
 193  *
 194  * The assumption is that usually keys are actual pointers, so we use a
 195  * default value of a pointer to an arbitrary piece of storage in the library.
 196  * But in some cases a consumer wants to store some other sort of value in the
 197  * table, like a uint32_t, in which case that pointer may conflict with one of
 198  * their valid keys.  This lets that user select a safe value.
 199  *
 200  * This must be called before any keys are actually deleted from the table.
 201  */
 202 void
 203 _mesa_hash_table_set_deleted_key(struct hash_table *ht, const void *deleted_key)
 204 {
 205    ht->deleted_key = deleted_key;
 206 }
 207
 208 static struct hash_entry *
 209 hash_table_search(struct hash_table *ht, uint32_t hash, const void *key)
 210 {
 211    uint32_t start_hash_address = hash % ht->size;
 212    uint32_t hash_address = start_hash_address;
 213
 214    do {
 215       uint32_t double_hash;
 216
 217       struct hash_entry *entry = ht->table + hash_address;
 218
 219       if (entry_is_free(entry)) {
 220          return NULL;
 221       } else if (entry_is_present(ht, entry) && entry->hash == hash) {
 222          if (ht->key_equals_function(key, entry->key)) {
 223             return entry;
 224          }
 225       }
 226
 227       double_hash = 1 + hash % ht->rehash;
 228
 229       hash_address = (hash_address + double_hash) % ht->size;
 230    } while (hash_address != start_hash_address);
 231
 232    return NULL;
 233 }
 234
 235 /**
 236  * Finds a hash table entry with the given key and hash of that key.
 237  *
 238  * Returns NULL if no entry is found.  Note that the data pointer may be
 239  * modified by the user.
 240  */
 241 struct hash_entry *
 242 _mesa_hash_table_search(struct hash_table *ht, const void *key)
 243 {
 244    assert(ht->key_hash_function);
 245    return hash_table_search(ht, ht->key_hash_function(key), key);
 246 }
 247
 248 struct hash_entry *
 249 _mesa_hash_table_search_pre_hashed(struct hash_table *ht, uint32_t hash,
 250                                   const void *key)
 251 {
 252    assert(ht->key_hash_function == NULL || hash == ht->key_hash_function(key));
 253    return hash_table_search(ht, hash, key);
 254 }
 255
 256 static struct hash_entry *
 257 hash_table_insert(struct hash_table *ht, uint32_t hash,
 258                   const void *key, void *data);
 259
 260 static void
 261 _mesa_hash_table_rehash(struct hash_table *ht, unsigned new_size_index)
 262 {
 263    struct hash_table old_ht;
 264    struct hash_entry *table, *entry;
 265
 266    if (new_size_index >= ARRAY_SIZE(hash_sizes))
 267       return;
 268
 269    table = rzalloc_array(ht, struct hash_entry,
 270                          hash_sizes[new_size_index].size);
 271    if (table == NULL)
 272       return;
 273
 274    old_ht = *ht;
 275
 276    ht->table = table;
 277    ht->size_index = new_size_index;
 278    ht->size = hash_sizes[ht->size_index].size;
 279    ht->rehash = hash_sizes[ht->size_index].rehash;
 280    ht->max_entries = hash_sizes[ht->size_index].max_entries;
 281    ht->entries = 0;
 282    ht->deleted_entries = 0;
 283
 284    hash_table_foreach(&old_ht, entry) {
 285       hash_table_insert(ht, entry->hash, entry->key, entry->data);
 286    }
 287
 288    ralloc_free(old_ht.table);
 289 }
 290
 291 static struct hash_entry *
 292 hash_table_insert(struct hash_table *ht, uint32_t hash,
 293                   const void *key, void *data)
 294 {
 295    uint32_t start_hash_address, hash_address;
 296    struct hash_entry *available_entry = NULL;
 297
 298    assert(key != NULL);
 299
 300    if (ht->entries >= ht->max_entries) {
 301       _mesa_hash_table_rehash(ht, ht->size_index + 1);
 302    } else if (ht->deleted_entries + ht->entries >= ht->max_entries) {
 303       _mesa_hash_table_rehash(ht, ht->size_index);
 304    }
 305
 306    start_hash_address = hash % ht->size;
 307    hash_address = start_hash_address;
 308    do {
 309       struct hash_entry *entry = ht->table + hash_address;
 310       uint32_t double_hash;
 311
 312       if (!entry_is_present(ht, entry)) {
 313          /* Stash the first available entry we find */
 314          if (available_entry == NULL)
 315             available_entry = entry;
 316          if (entry_is_free(entry))
 317             break;
 318       }
 319
 320       /* Implement replacement when another insert happens
 321        * with a matching key.  This is a relatively common
 322        * feature of hash tables, with the alternative
 323        * generally being "insert the new value as well, and
 324        * return it first when the key is searched for".
 325        *
 326        * Note that the hash table doesn't have a delete
 327        * callback.  If freeing of old data pointers is
 328        * required to avoid memory leaks, perform a search
 329        * before inserting.
 330        */
 331       if (!entry_is_deleted(ht, entry) &&
 332           entry->hash == hash &&
 333           ht->key_equals_function(key, entry->key)) {
 334          entry->key = key;
 335          entry->data = data;
 336          return entry;
 337       }
 338
 339
 340       double_hash = 1 + hash % ht->rehash;
 341
 342       hash_address = (hash_address + double_hash) % ht->size;
 343    } while (hash_address != start_hash_address);
 344
 345    if (available_entry) {
 346       if (entry_is_deleted(ht, available_entry))
 347          ht->deleted_entries--;
 348       available_entry->hash = hash;
 349       available_entry->key = key;
 350       available_entry->data = data;
 351       ht->entries++;
 352       return available_entry;
 353    }
 354
 355    /* We could hit here if a required resize failed. An unchecked-malloc
 356     * application could ignore this result.
 357     */
 358    return NULL;
 359 }
 360
 361 /**
 362  * Inserts the key with the given hash into the table.
 363  *
 364  * Note that insertion may rearrange the table on a resize or rehash,
 365  * so previously found hash_entries are no longer valid after this function.
 366  */
 367 struct hash_entry *
 368 _mesa_hash_table_insert(struct hash_table *ht, const void *key, void *data)
 369 {
 370    assert(ht->key_hash_function);
 371    return hash_table_insert(ht, ht->key_hash_function(key), key, data);
 372 }
 373
 374 struct hash_entry *
 375 _mesa_hash_table_insert_pre_hashed(struct hash_table *ht, uint32_t hash,
 376                                    const void *key, void *data)
 377 {
 378    assert(ht->key_hash_function == NULL || hash == ht->key_hash_function(key));
 379    return hash_table_insert(ht, hash, key, data);
 380 }
 381
 382 /**
 383  * This function deletes the given hash table entry.
 384  *
 385  * Note that deletion doesn't otherwise modify the table, so an iteration over
 386  * the table deleting entries is safe.
 387  */
 388 void
 389 _mesa_hash_table_remove(struct hash_table *ht,
 390                         struct hash_entry *entry)
 391 {
 392    if (!entry)
 393       return;
 394
 395    entry->key = ht->deleted_key;
 396    ht->entries--;
 397    ht->deleted_entries++;
 398 }
 399
 400 /**
 401  * This function is an iterator over the hash table.
 402  *
 403  * Pass in NULL for the first entry, as in the start of a for loop.  Note that
 404  * an iteration over the table is O(table_size) not O(entries).
 405  */
 406 struct hash_entry *
 407 _mesa_hash_table_next_entry(struct hash_table *ht,
 408                             struct hash_entry *entry)
 409 {
 410    if (entry == NULL)
 411       entry = ht->table;
 412    else
 413       entry = entry + 1;
 414
 415    for (; entry != ht->table + ht->size; entry++) {
 416       if (entry_is_present(ht, entry)) {
 417          return entry;
 418       }
 419    }
 420
 421    return NULL;
 422 }
 423
 424 /**
 425  * Returns a random entry from the hash table.
 426  *
 427  * This may be useful in implementing random replacement (as opposed
 428  * to just removing everything) in caches based on this hash table
 429  * implementation.  @predicate may be used to filter entries, or may
 430  * be set to NULL for no filtering.
 431  */
 432 struct hash_entry *
 433 _mesa_hash_table_random_entry(struct hash_table *ht,
 434                               bool (*predicate)(struct hash_entry *entry))
 435 {
 436    struct hash_entry *entry;
 437    uint32_t i = rand() % ht->size;
 438
 439    if (ht->entries == 0)
 440       return NULL;
 441
 442    for (entry = ht->table + i; entry != ht->table + ht->size; entry++) {
 443       if (entry_is_present(ht, entry) &&
 444           (!predicate || predicate(entry))) {
 445          return entry;
 446       }
 447    }
 448
 449    for (entry = ht->table; entry != ht->table + i; entry++) {
 450       if (entry_is_present(ht, entry) &&
 451           (!predicate || predicate(entry))) {
 452          return entry;
 453       }
 454    }
 455
 456    return NULL;
 457 }
 458
 459
 460 /**
 461  * Quick FNV-1a hash implementation based on:
 462  * http://www.isthe.com/chongo/tech/comp/fnv/
 463  *
 464  * FNV-1a is not be the best hash out there -- Jenkins's lookup3 is supposed
 465  * to be quite good, and it probably beats FNV.  But FNV has the advantage
 466  * that it involves almost no code.  For an improvement on both, see Paul
 467  * Hsieh's http://www.azillionmonkeys.com/qed/hash.html
 468  */
 469 uint32_t
 470 _mesa_hash_data(const void *data, size_t size)
 471 {
 472    return _mesa_fnv32_1a_accumulate_block(_mesa_fnv32_1a_offset_bias,
 473                                           data, size);
 474 }
 475
 476 /** FNV-1a string hash implementation */
 477 uint32_t
 478 _mesa_hash_string(const char *key)
 479 {
 480    uint32_t hash = _mesa_fnv32_1a_offset_bias;
 481
 482    while (*key != 0) {
 483       hash = _mesa_fnv32_1a_accumulate(hash, *key);
 484       key++;
 485    }
 486
 487    return hash;
 488 }
 489
 490 /**
 491  * String compare function for use as the comparison callback in
 492  * _mesa_hash_table_create().
 493  */
 494 bool
 495 _mesa_key_string_equal(const void *a, const void *b)
 496 {
 497    return strcmp(a, b) == 0;
 498 }
 499
 500 bool
 501 _mesa_key_pointer_equal(const void *a, const void *b)
 502 {
 503    return a == b;
 504 }