src/mesa/main/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 "main/hash_table.h"
  47 #include "ralloc.h"
  48
  49 #define ARRAY_SIZE(array) (sizeof(array) / sizeof(array[0]))
  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                         bool key_equals_function(const void *a,
 115                                                  const void *b))
 116 {
 117    struct hash_table *ht;
 118
 119    ht = ralloc(mem_ctx, struct hash_table);
 120    if (ht == NULL)
 121       return NULL;
 122
 123    ht->mem_ctx = mem_ctx;
 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_equals_function = key_equals_function;
 129    ht->table = rzalloc_array(ht, struct hash_entry, ht->size);
 130    ht->entries = 0;
 131    ht->deleted_entries = 0;
 132    ht->deleted_key = &deleted_key_value;
 133
 134    if (ht->table == NULL) {
 135       ralloc_free(ht);
 136       return NULL;
 137    }
 138
 139    return ht;
 140 }
 141
 142 /**
 143  * Frees the given hash table.
 144  *
 145  * If delete_function is passed, it gets called on each entry present before
 146  * freeing.
 147  */
 148 void
 149 _mesa_hash_table_destroy(struct hash_table *ht,
 150                          void (*delete_function)(struct hash_entry *entry))
 151 {
 152    if (!ht)
 153       return;
 154
 155    if (delete_function) {
 156       struct hash_entry *entry;
 157
 158       hash_table_foreach(ht, entry) {
 159          delete_function(entry);
 160       }
 161    }
 162    ralloc_free(ht);
 163 }
 164
 165 /** Sets the value of the key pointer used for deleted entries in the table.
 166  *
 167  * The assumption is that usually keys are actual pointers, so we use a
 168  * default value of a pointer to an arbitrary piece of storage in the library.
 169  * But in some cases a consumer wants to store some other sort of value in the
 170  * table, like a uint32_t, in which case that pointer may conflict with one of
 171  * their valid keys.  This lets that user select a safe value.
 172  *
 173  * This must be called before any keys are actually deleted from the table.
 174  */
 175 void
 176 _mesa_hash_table_set_deleted_key(struct hash_table *ht, const void *deleted_key)
 177 {
 178    ht->deleted_key = deleted_key;
 179 }
 180
 181 /**
 182  * Finds a hash table entry with the given key and hash of that key.
 183  *
 184  * Returns NULL if no entry is found.  Note that the data pointer may be
 185  * modified by the user.
 186  */
 187 struct hash_entry *
 188 _mesa_hash_table_search(struct hash_table *ht, uint32_t hash,
 189                         const void *key)
 190 {
 191    uint32_t start_hash_address = hash % ht->size;
 192    uint32_t hash_address = start_hash_address;
 193
 194    do {
 195       uint32_t double_hash;
 196
 197       struct hash_entry *entry = ht->table + hash_address;
 198
 199       if (entry_is_free(entry)) {
 200          return NULL;
 201       } else if (entry_is_present(ht, entry) && entry->hash == hash) {
 202          if (ht->key_equals_function(key, entry->key)) {
 203             return entry;
 204          }
 205       }
 206
 207       double_hash = 1 + hash % ht->rehash;
 208
 209       hash_address = (hash_address + double_hash) % ht->size;
 210    } while (hash_address != start_hash_address);
 211
 212    return NULL;
 213 }
 214
 215 static void
 216 _mesa_hash_table_rehash(struct hash_table *ht, int new_size_index)
 217 {
 218    struct hash_table old_ht;
 219    struct hash_entry *table, *entry;
 220
 221    if (new_size_index >= ARRAY_SIZE(hash_sizes))
 222       return;
 223
 224    table = rzalloc_array(ht, struct hash_entry,
 225                          hash_sizes[new_size_index].size);
 226    if (table == NULL)
 227       return;
 228
 229    old_ht = *ht;
 230
 231    ht->table = table;
 232    ht->size_index = new_size_index;
 233    ht->size = hash_sizes[ht->size_index].size;
 234    ht->rehash = hash_sizes[ht->size_index].rehash;
 235    ht->max_entries = hash_sizes[ht->size_index].max_entries;
 236    ht->entries = 0;
 237    ht->deleted_entries = 0;
 238
 239    hash_table_foreach(&old_ht, entry) {
 240       _mesa_hash_table_insert(ht, entry->hash,
 241                               entry->key, entry->data);
 242    }
 243
 244    ralloc_free(old_ht.table);
 245 }
 246
 247 /**
 248  * Inserts the key with the given hash into the table.
 249  *
 250  * Note that insertion may rearrange the table on a resize or rehash,
 251  * so previously found hash_entries are no longer valid after this function.
 252  */
 253 struct hash_entry *
 254 _mesa_hash_table_insert(struct hash_table *ht, uint32_t hash,
 255                         const void *key, void *data)
 256 {
 257    uint32_t start_hash_address, hash_address;
 258
 259    if (ht->entries >= ht->max_entries) {
 260       _mesa_hash_table_rehash(ht, ht->size_index + 1);
 261    } else if (ht->deleted_entries + ht->entries >= ht->max_entries) {
 262       _mesa_hash_table_rehash(ht, ht->size_index);
 263    }
 264
 265    start_hash_address = hash % ht->size;
 266    hash_address = start_hash_address;
 267    do {
 268       struct hash_entry *entry = ht->table + hash_address;
 269       uint32_t double_hash;
 270
 271       if (!entry_is_present(ht, entry)) {
 272          if (entry_is_deleted(ht, entry))
 273             ht->deleted_entries--;
 274          entry->hash = hash;
 275          entry->key = key;
 276          entry->data = data;
 277          ht->entries++;
 278          return entry;
 279       }
 280
 281       /* Implement replacement when another insert happens
 282        * with a matching key.  This is a relatively common
 283        * feature of hash tables, with the alternative
 284        * generally being "insert the new value as well, and
 285        * return it first when the key is searched for".
 286        *
 287        * Note that the hash table doesn't have a delete
 288        * callback.  If freeing of old data pointers is
 289        * required to avoid memory leaks, perform a search
 290        * before inserting.
 291        */
 292       if (entry->hash == hash &&
 293           ht->key_equals_function(key, entry->key)) {
 294          entry->key = key;
 295          entry->data = data;
 296          return entry;
 297       }
 298
 299
 300       double_hash = 1 + hash % ht->rehash;
 301
 302       hash_address = (hash_address + double_hash) % ht->size;
 303    } while (hash_address != start_hash_address);
 304
 305    /* We could hit here if a required resize failed. An unchecked-malloc
 306     * application could ignore this result.
 307     */
 308    return NULL;
 309 }
 310
 311 /**
 312  * This function deletes the given hash table entry.
 313  *
 314  * Note that deletion doesn't otherwise modify the table, so an iteration over
 315  * the table deleting entries is safe.
 316  */
 317 void
 318 _mesa_hash_table_remove(struct hash_table *ht,
 319                         struct hash_entry *entry)
 320 {
 321    if (!entry)
 322       return;
 323
 324    entry->key = ht->deleted_key;
 325    ht->entries--;
 326    ht->deleted_entries++;
 327 }
 328
 329 /**
 330  * This function is an iterator over the hash table.
 331  *
 332  * Pass in NULL for the first entry, as in the start of a for loop.  Note that
 333  * an iteration over the table is O(table_size) not O(entries).
 334  */
 335 struct hash_entry *
 336 _mesa_hash_table_next_entry(struct hash_table *ht,
 337                             struct hash_entry *entry)
 338 {
 339    if (entry == NULL)
 340       entry = ht->table;
 341    else
 342       entry = entry + 1;
 343
 344    for (; entry != ht->table + ht->size; entry++) {
 345       if (entry_is_present(ht, entry)) {
 346          return entry;
 347       }
 348    }
 349
 350    return NULL;
 351 }
 352
 353 /**
 354  * Returns a random entry from the hash table.
 355  *
 356  * This may be useful in implementing random replacement (as opposed
 357  * to just removing everything) in caches based on this hash table
 358  * implementation.  @predicate may be used to filter entries, or may
 359  * be set to NULL for no filtering.
 360  */
 361 struct hash_entry *
 362 _mesa_hash_table_random_entry(struct hash_table *ht,
 363                               bool (*predicate)(struct hash_entry *entry))
 364 {
 365    struct hash_entry *entry;
 366    uint32_t i = rand() % ht->size;
 367
 368    if (ht->entries == 0)
 369       return NULL;
 370
 371    for (entry = ht->table + i; entry != ht->table + ht->size; entry++) {
 372       if (entry_is_present(ht, entry) &&
 373           (!predicate || predicate(entry))) {
 374          return entry;
 375       }
 376    }
 377
 378    for (entry = ht->table; entry != ht->table + i; entry++) {
 379       if (entry_is_present(ht, entry) &&
 380           (!predicate || predicate(entry))) {
 381          return entry;
 382       }
 383    }
 384
 385    return NULL;
 386 }
 387
 388
 389 /**
 390  * Quick FNV-1 hash implementation based on:
 391  * http://www.isthe.com/chongo/tech/comp/fnv/
 392  *
 393  * FNV-1 is not be the best hash out there -- Jenkins's lookup3 is supposed to
 394  * be quite good, and it probably beats FNV.  But FNV has the advantage that
 395  * it involves almost no code.  For an improvement on both, see Paul
 396  * Hsieh's http://www.azillionmonkeys.com/qed/hash.html
 397  */
 398 uint32_t
 399 _mesa_hash_data(const void *data, size_t size)
 400 {
 401    uint32_t hash = 2166136261ul;
 402    const uint8_t *bytes = data;
 403
 404    while (size-- != 0) {
 405       hash ^= *bytes;
 406       hash = hash * 0x01000193;
 407       bytes++;
 408    }
 409
 410    return hash;
 411 }
 412
 413 /** FNV-1 string hash implementation */
 414 uint32_t
 415 _mesa_hash_string(const char *key)
 416 {
 417    uint32_t hash = 2166136261ul;
 418
 419    while (*key != 0) {
 420       hash ^= *key;
 421       hash = hash * 0x01000193;
 422       key++;
 423    }
 424
 425    return hash;
 426 }
 427
 428 /**
 429  * String compare function for use as the comparison callback in
 430  * _mesa_hash_table_create().
 431  */
 432 bool
 433 _mesa_key_string_equal(const void *a, const void *b)
 434 {
 435    return strcmp(a, b) == 0;
 436 }
 437
 438 bool
 439 _mesa_key_pointer_equal(const void *a, const void *b)
 440 {
 441    return a == b;
 442 }