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
  46 #include "hash_table.h"
  47 #include "ralloc.h"
  48 #include "macros.h"
  49
  50 static const uint32_t deleted_key_value;
  51
  52 /**
  53  * From Knuth -- a good choice for hash/rehash values is p, p-2 where
  54  * p and p-2 are both prime.  These tables are sized to have an extra 10%
  55  * free to avoid exponential performance degradation as the hash table fills
  56  */
  57 static const struct {
  58    uint32_t max_entries, size, rehash;
  59 } hash_sizes[] = {
  60    { 2,                 5,              3         },
  61    { 4,                 7,              5         },
  62    { 8,                 13,             11        },
  63    { 16,                19,             17        },
  64    { 32,                43,             41        },
  65    { 64,                73,             71        },
  66    { 128,               151,            149       },
  67    { 256,               283,            281       },
  68    { 512,               571,            569       },
  69    { 1024,              1153,           1151      },
  70    { 2048,              2269,           2267      },
  71    { 4096,              4519,           4517      },
  72    { 8192,              9013,           9011      },
  73    { 16384,             18043,          18041     },
  74    { 32768,             36109,          36107     },
  75    { 65536,             72091,          72089     },
  76    { 131072,            144409,         144407    },
  77    { 262144,            288361,         288359    },
  78    { 524288,            576883,         576881    },
  79    { 1048576,           1153459,        1153457   },
  80    { 2097152,           2307163,        2307161   },
  81    { 4194304,           4613893,        4613891   },
  82    { 8388608,           9227641,        9227639   },
  83    { 16777216,          18455029,       18455027  },
  84    { 33554432,          36911011,       36911009  },
  85    { 67108864,          73819861,       73819859  },
  86    { 134217728,         147639589,      147639587 },
  87    { 268435456,         295279081,      295279079 },
  88    { 536870912,         590559793,      590559791 },
  89    { 1073741824,        1181116273,     1181116271},
  90    { 2147483648ul,      2362232233ul,   2362232231ul}
  91 };
  92
  93 static int
  94 entry_is_free(const struct hash_entry *entry)
  95 {
  96    return entry->key == NULL;
  97 }
  98
  99 static int
 100 entry_is_deleted(const struct hash_table *ht, struct hash_entry *entry)
 101 {
 102    return entry->key == ht->deleted_key;
 103 }
 104
 105 static int
 106 entry_is_present(const struct hash_table *ht, struct hash_entry *entry)
 107 {
 108    return entry->key != NULL && entry->key != ht->deleted_key;
 109 }
 110
 111 struct hash_table *
 112 _mesa_hash_table_create(void *mem_ctx,
 113                         bool (*key_equals_function)(const void *a,
 114                                                     const void *b))
 115 {
 116    struct hash_table *ht;
 117
 118    ht = ralloc(mem_ctx, struct hash_table);
 119    if (ht == NULL)
 120       return NULL;
 121
 122    ht->size_index = 0;
 123    ht->size = hash_sizes[ht->size_index].size;
 124    ht->rehash = hash_sizes[ht->size_index].rehash;
 125    ht->max_entries = hash_sizes[ht->size_index].max_entries;
 126    ht->key_equals_function = key_equals_function;
 127    ht->table = rzalloc_array(ht, struct hash_entry, ht->size);
 128    ht->entries = 0;
 129    ht->deleted_entries = 0;
 130    ht->deleted_key = &deleted_key_value;
 131
 132    if (ht->table == NULL) {
 133       ralloc_free(ht);
 134       return NULL;
 135    }
 136
 137    return ht;
 138 }
 139
 140 /**
 141  * Frees the given hash table.
 142  *
 143  * If delete_function is passed, it gets called on each entry present before
 144  * freeing.
 145  */
 146 void
 147 _mesa_hash_table_destroy(struct hash_table *ht,
 148                          void (*delete_function)(struct hash_entry *entry))
 149 {
 150    if (!ht)
 151       return;
 152
 153    if (delete_function) {
 154       struct hash_entry *entry;
 155
 156       hash_table_foreach(ht, entry) {
 157          delete_function(entry);
 158       }
 159    }
 160    ralloc_free(ht);
 161 }
 162
 163 /** Sets the value of the key pointer used for deleted entries in the table.
 164  *
 165  * The assumption is that usually keys are actual pointers, so we use a
 166  * default value of a pointer to an arbitrary piece of storage in the library.
 167  * But in some cases a consumer wants to store some other sort of value in the
 168  * table, like a uint32_t, in which case that pointer may conflict with one of
 169  * their valid keys.  This lets that user select a safe value.
 170  *
 171  * This must be called before any keys are actually deleted from the table.
 172  */
 173 void
 174 _mesa_hash_table_set_deleted_key(struct hash_table *ht, const void *deleted_key)
 175 {
 176    ht->deleted_key = deleted_key;
 177 }
 178
 179 /**
 180  * Finds a hash table entry with the given key and hash of that key.
 181  *
 182  * Returns NULL if no entry is found.  Note that the data pointer may be
 183  * modified by the user.
 184  */
 185 struct hash_entry *
 186 _mesa_hash_table_search(struct hash_table *ht, uint32_t hash,
 187                         const void *key)
 188 {
 189    uint32_t start_hash_address = hash % ht->size;
 190    uint32_t hash_address = start_hash_address;
 191
 192    do {
 193       uint32_t double_hash;
 194
 195       struct hash_entry *entry = ht->table + hash_address;
 196
 197       if (entry_is_free(entry)) {
 198          return NULL;
 199       } else if (entry_is_present(ht, entry) && entry->hash == hash) {
 200          if (ht->key_equals_function(key, entry->key)) {
 201             return entry;
 202          }
 203       }
 204
 205       double_hash = 1 + hash % ht->rehash;
 206
 207       hash_address = (hash_address + double_hash) % ht->size;
 208    } while (hash_address != start_hash_address);
 209
 210    return NULL;
 211 }
 212
 213 static void
 214 _mesa_hash_table_rehash(struct hash_table *ht, int new_size_index)
 215 {
 216    struct hash_table old_ht;
 217    struct hash_entry *table, *entry;
 218
 219    if (new_size_index >= ARRAY_SIZE(hash_sizes))
 220       return;
 221
 222    table = rzalloc_array(ht, struct hash_entry,
 223                          hash_sizes[new_size_index].size);
 224    if (table == NULL)
 225       return;
 226
 227    old_ht = *ht;
 228
 229    ht->table = table;
 230    ht->size_index = new_size_index;
 231    ht->size = hash_sizes[ht->size_index].size;
 232    ht->rehash = hash_sizes[ht->size_index].rehash;
 233    ht->max_entries = hash_sizes[ht->size_index].max_entries;
 234    ht->entries = 0;
 235    ht->deleted_entries = 0;
 236
 237    hash_table_foreach(&old_ht, entry) {
 238       _mesa_hash_table_insert(ht, entry->hash,
 239                               entry->key, entry->data);
 240    }
 241
 242    ralloc_free(old_ht.table);
 243 }
 244
 245 /**
 246  * Inserts the key with the given hash into the table.
 247  *
 248  * Note that insertion may rearrange the table on a resize or rehash,
 249  * so previously found hash_entries are no longer valid after this function.
 250  */
 251 struct hash_entry *
 252 _mesa_hash_table_insert(struct hash_table *ht, uint32_t hash,
 253                         const void *key, void *data)
 254 {
 255    uint32_t start_hash_address, hash_address;
 256
 257    if (ht->entries >= ht->max_entries) {
 258       _mesa_hash_table_rehash(ht, ht->size_index + 1);
 259    } else if (ht->deleted_entries + ht->entries >= ht->max_entries) {
 260       _mesa_hash_table_rehash(ht, ht->size_index);
 261    }
 262
 263    start_hash_address = hash % ht->size;
 264    hash_address = start_hash_address;
 265    do {
 266       struct hash_entry *entry = ht->table + hash_address;
 267       uint32_t double_hash;
 268
 269       if (!entry_is_present(ht, entry)) {
 270          if (entry_is_deleted(ht, entry))
 271             ht->deleted_entries--;
 272          entry->hash = hash;
 273          entry->key = key;
 274          entry->data = data;
 275          ht->entries++;
 276          return entry;
 277       }
 278
 279       /* Implement replacement when another insert happens
 280        * with a matching key.  This is a relatively common
 281        * feature of hash tables, with the alternative
 282        * generally being "insert the new value as well, and
 283        * return it first when the key is searched for".
 284        *
 285        * Note that the hash table doesn't have a delete
 286        * callback.  If freeing of old data pointers is
 287        * required to avoid memory leaks, perform a search
 288        * before inserting.
 289        */
 290       if (entry->hash == hash &&
 291           ht->key_equals_function(key, entry->key)) {
 292          entry->key = key;
 293          entry->data = data;
 294          return entry;
 295       }
 296
 297
 298       double_hash = 1 + hash % ht->rehash;
 299
 300       hash_address = (hash_address + double_hash) % ht->size;
 301    } while (hash_address != start_hash_address);
 302
 303    /* We could hit here if a required resize failed. An unchecked-malloc
 304     * application could ignore this result.
 305     */
 306    return NULL;
 307 }
 308
 309 /**
 310  * This function deletes the given hash table entry.
 311  *
 312  * Note that deletion doesn't otherwise modify the table, so an iteration over
 313  * the table deleting entries is safe.
 314  */
 315 void
 316 _mesa_hash_table_remove(struct hash_table *ht,
 317                         struct hash_entry *entry)
 318 {
 319    if (!entry)
 320       return;
 321
 322    entry->key = ht->deleted_key;
 323    ht->entries--;
 324    ht->deleted_entries++;
 325 }
 326
 327 /**
 328  * This function is an iterator over the hash table.
 329  *
 330  * Pass in NULL for the first entry, as in the start of a for loop.  Note that
 331  * an iteration over the table is O(table_size) not O(entries).
 332  */
 333 struct hash_entry *
 334 _mesa_hash_table_next_entry(struct hash_table *ht,
 335                             struct hash_entry *entry)
 336 {
 337    if (entry == NULL)
 338       entry = ht->table;
 339    else
 340       entry = entry + 1;
 341
 342    for (; entry != ht->table + ht->size; entry++) {
 343       if (entry_is_present(ht, entry)) {
 344          return entry;
 345       }
 346    }
 347
 348    return NULL;
 349 }
 350
 351 /**
 352  * Returns a random entry from the hash table.
 353  *
 354  * This may be useful in implementing random replacement (as opposed
 355  * to just removing everything) in caches based on this hash table
 356  * implementation.  @predicate may be used to filter entries, or may
 357  * be set to NULL for no filtering.
 358  */
 359 struct hash_entry *
 360 _mesa_hash_table_random_entry(struct hash_table *ht,
 361                               bool (*predicate)(struct hash_entry *entry))
 362 {
 363    struct hash_entry *entry;
 364    uint32_t i = rand() % ht->size;
 365
 366    if (ht->entries == 0)
 367       return NULL;
 368
 369    for (entry = ht->table + i; entry != ht->table + ht->size; entry++) {
 370       if (entry_is_present(ht, entry) &&
 371           (!predicate || predicate(entry))) {
 372          return entry;
 373       }
 374    }
 375
 376    for (entry = ht->table; entry != ht->table + i; entry++) {
 377       if (entry_is_present(ht, entry) &&
 378           (!predicate || predicate(entry))) {
 379          return entry;
 380       }
 381    }
 382
 383    return NULL;
 384 }
 385
 386
 387 /**
 388  * Quick FNV-1 hash implementation based on:
 389  * http://www.isthe.com/chongo/tech/comp/fnv/
 390  *
 391  * FNV-1 is not be the best hash out there -- Jenkins's lookup3 is supposed to
 392  * be quite good, and it probably beats FNV.  But FNV has the advantage that
 393  * it involves almost no code.  For an improvement on both, see Paul
 394  * Hsieh's http://www.azillionmonkeys.com/qed/hash.html
 395  */
 396 uint32_t
 397 _mesa_hash_data(const void *data, size_t size)
 398 {
 399    uint32_t hash = 2166136261ul;
 400    const uint8_t *bytes = data;
 401
 402    while (size-- != 0) {
 403       hash ^= *bytes;
 404       hash = hash * 0x01000193;
 405       bytes++;
 406    }
 407
 408    return hash;
 409 }
 410
 411 /** FNV-1 string hash implementation */
 412 uint32_t
 413 _mesa_hash_string(const char *key)
 414 {
 415    uint32_t hash = 2166136261ul;
 416
 417    while (*key != 0) {
 418       hash ^= *key;
 419       hash = hash * 0x01000193;
 420       key++;
 421    }
 422
 423    return hash;
 424 }
 425
 426 /**
 427  * String compare function for use as the comparison callback in
 428  * _mesa_hash_table_create().
 429  */
 430 bool
 431 _mesa_key_string_equal(const void *a, const void *b)
 432 {
 433    return strcmp(a, b) == 0;
 434 }
 435
 436 bool
 437 _mesa_key_pointer_equal(const void *a, const void *b)
 438 {
 439    return a == b;
 440 }