src/util/set.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 #include <stdlib.h>
  36 #include <assert.h>
  37
  38 #include "macros.h"
  39 #include "ralloc.h"
  40 #include "set.h"
  41
  42 /*
  43  * From Knuth -- a good choice for hash/rehash values is p, p-2 where
  44  * p and p-2 are both prime.  These tables are sized to have an extra 10%
  45  * free to avoid exponential performance degradation as the hash table fills
  46  */
  47
  48 static const uint32_t deleted_key_value;
  49 static const void *deleted_key = &deleted_key_value;
  50
  51 static const struct {
  52    uint32_t max_entries, size, rehash;
  53 } hash_sizes[] = {
  54    { 2,            5,            3            },
  55    { 4,            7,            5            },
  56    { 8,            13,           11           },
  57    { 16,           19,           17           },
  58    { 32,           43,           41           },
  59    { 64,           73,           71           },
  60    { 128,          151,          149          },
  61    { 256,          283,          281          },
  62    { 512,          571,          569          },
  63    { 1024,         1153,         1151         },
  64    { 2048,         2269,         2267         },
  65    { 4096,         4519,         4517         },
  66    { 8192,         9013,         9011         },
  67    { 16384,        18043,        18041        },
  68    { 32768,        36109,        36107        },
  69    { 65536,        72091,        72089        },
  70    { 131072,       144409,       144407       },
  71    { 262144,       288361,       288359       },
  72    { 524288,       576883,       576881       },
  73    { 1048576,      1153459,      1153457      },
  74    { 2097152,      2307163,      2307161      },
  75    { 4194304,      4613893,      4613891      },
  76    { 8388608,      9227641,      9227639      },
  77    { 16777216,     18455029,     18455027     },
  78    { 33554432,     36911011,     36911009     },
  79    { 67108864,     73819861,     73819859     },
  80    { 134217728,    147639589,    147639587    },
  81    { 268435456,    295279081,    295279079    },
  82    { 536870912,    590559793,    590559791    },
  83    { 1073741824,   1181116273,   1181116271   },
  84    { 2147483648ul, 2362232233ul, 2362232231ul }
  85 };
  86
  87 static int
  88 entry_is_free(struct set_entry *entry)
  89 {
  90    return entry->key == NULL;
  91 }
  92
  93 static int
  94 entry_is_deleted(struct set_entry *entry)
  95 {
  96    return entry->key == deleted_key;
  97 }
  98
  99 static int
 100 entry_is_present(struct set_entry *entry)
 101 {
 102    return entry->key != NULL && entry->key != deleted_key;
 103 }
 104
 105 struct set *
 106 _mesa_set_create(void *mem_ctx,
 107                  uint32_t (*key_hash_function)(const void *key),
 108                  bool (*key_equals_function)(const void *a,
 109                                              const void *b))
 110 {
 111    struct set *ht;
 112
 113    ht = ralloc(mem_ctx, struct set);
 114    if (ht == NULL)
 115       return NULL;
 116
 117    ht->size_index = 0;
 118    ht->size = hash_sizes[ht->size_index].size;
 119    ht->rehash = hash_sizes[ht->size_index].rehash;
 120    ht->max_entries = hash_sizes[ht->size_index].max_entries;
 121    ht->key_hash_function = key_hash_function;
 122    ht->key_equals_function = key_equals_function;
 123    ht->table = rzalloc_array(ht, struct set_entry, ht->size);
 124    ht->entries = 0;
 125    ht->deleted_entries = 0;
 126
 127    if (ht->table == NULL) {
 128       ralloc_free(ht);
 129       return NULL;
 130    }
 131
 132    return ht;
 133 }
 134
 135 /**
 136  * Frees the given set.
 137  *
 138  * If delete_function is passed, it gets called on each entry present before
 139  * freeing.
 140  */
 141 void
 142 _mesa_set_destroy(struct set *ht, void (*delete_function)(struct set_entry *entry))
 143 {
 144    if (!ht)
 145       return;
 146
 147    if (delete_function) {
 148       struct set_entry *entry;
 149
 150       set_foreach (ht, entry) {
 151          delete_function(entry);
 152       }
 153    }
 154    ralloc_free(ht->table);
 155    ralloc_free(ht);
 156 }
 157
 158 /**
 159  * Clears all values from the given set.
 160  *
 161  * If delete_function is passed, it gets called on each entry present before
 162  * the set is cleared.
 163  */
 164 void
 165 _mesa_set_clear(struct set *set, void (*delete_function)(struct set_entry *entry))
 166 {
 167    struct set_entry *entry;
 168
 169    if (!set)
 170       return;
 171
 172    set_foreach (set, entry) {
 173       if (delete_function)
 174          delete_function(entry);
 175       entry->key = deleted_key;
 176    }
 177
 178    set->entries = set->deleted_entries = 0;
 179 }
 180
 181 /**
 182  * Finds a set entry with the given key and hash of that key.
 183  *
 184  * Returns NULL if no entry is found.
 185  */
 186 static struct set_entry *
 187 set_search(const struct set *ht, uint32_t hash, const void *key)
 188 {
 189    uint32_t hash_address;
 190
 191    hash_address = hash % ht->size;
 192    do {
 193       uint32_t double_hash;
 194
 195       struct set_entry *entry = ht->table + hash_address;
 196
 197       if (entry_is_free(entry)) {
 198          return NULL;
 199       } else if (entry_is_present(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 != hash % ht->size);
 209
 210    return NULL;
 211 }
 212
 213 struct set_entry *
 214 _mesa_set_search(const struct set *set, const void *key)
 215 {
 216    assert(set->key_hash_function);
 217    return set_search(set, set->key_hash_function(key), key);
 218 }
 219
 220 struct set_entry *
 221 _mesa_set_search_pre_hashed(const struct set *set, uint32_t hash,
 222                             const void *key)
 223 {
 224    assert(set->key_hash_function == NULL ||
 225           hash == set->key_hash_function(key));
 226    return set_search(set, hash, key);
 227 }
 228
 229 static struct set_entry *
 230 set_add(struct set *ht, uint32_t hash, const void *key);
 231
 232 static void
 233 set_rehash(struct set *ht, unsigned new_size_index)
 234 {
 235    struct set old_ht;
 236    struct set_entry *table, *entry;
 237
 238    if (new_size_index >= ARRAY_SIZE(hash_sizes))
 239       return;
 240
 241    table = rzalloc_array(ht, struct set_entry,
 242                          hash_sizes[new_size_index].size);
 243    if (table == NULL)
 244       return;
 245
 246    old_ht = *ht;
 247
 248    ht->table = table;
 249    ht->size_index = new_size_index;
 250    ht->size = hash_sizes[ht->size_index].size;
 251    ht->rehash = hash_sizes[ht->size_index].rehash;
 252    ht->max_entries = hash_sizes[ht->size_index].max_entries;
 253    ht->entries = 0;
 254    ht->deleted_entries = 0;
 255
 256    set_foreach(&old_ht, entry) {
 257       set_add(ht, entry->hash, entry->key);
 258    }
 259
 260    ralloc_free(old_ht.table);
 261 }
 262
 263 /**
 264  * Inserts the key with the given hash into the table.
 265  *
 266  * Note that insertion may rearrange the table on a resize or rehash,
 267  * so previously found hash_entries are no longer valid after this function.
 268  */
 269 static struct set_entry *
 270 set_add(struct set *ht, uint32_t hash, const void *key)
 271 {
 272    uint32_t hash_address;
 273    struct set_entry *available_entry = NULL;
 274
 275    if (ht->entries >= ht->max_entries) {
 276       set_rehash(ht, ht->size_index + 1);
 277    } else if (ht->deleted_entries + ht->entries >= ht->max_entries) {
 278       set_rehash(ht, ht->size_index);
 279    }
 280
 281    hash_address = hash % ht->size;
 282    do {
 283       struct set_entry *entry = ht->table + hash_address;
 284       uint32_t double_hash;
 285
 286       if (!entry_is_present(entry)) {
 287          /* Stash the first available entry we find */
 288          if (available_entry == NULL)
 289             available_entry = entry;
 290          if (entry_is_free(entry))
 291             break;
 292       }
 293
 294       /* Implement replacement when another insert happens
 295        * with a matching key.  This is a relatively common
 296        * feature of hash tables, with the alternative
 297        * generally being "insert the new value as well, and
 298        * return it first when the key is searched for".
 299        *
 300        * Note that the hash table doesn't have a delete callback.
 301        * If freeing of old keys is required to avoid memory leaks,
 302        * perform a search before inserting.
 303        */
 304       if (!entry_is_deleted(entry) &&
 305           entry->hash == hash &&
 306           ht->key_equals_function(key, entry->key)) {
 307          entry->key = key;
 308          return entry;
 309       }
 310
 311       double_hash = 1 + hash % ht->rehash;
 312
 313       hash_address = (hash_address + double_hash) % ht->size;
 314    } while (hash_address != hash % ht->size);
 315
 316    if (available_entry) {
 317       if (entry_is_deleted(available_entry))
 318          ht->deleted_entries--;
 319       available_entry->hash = hash;
 320       available_entry->key = key;
 321       ht->entries++;
 322       return available_entry;
 323    }
 324
 325    /* We could hit here if a required resize failed. An unchecked-malloc
 326     * application could ignore this result.
 327     */
 328    return NULL;
 329 }
 330
 331 struct set_entry *
 332 _mesa_set_add(struct set *set, const void *key)
 333 {
 334    assert(set->key_hash_function);
 335    return set_add(set, set->key_hash_function(key), key);
 336 }
 337
 338 struct set_entry *
 339 _mesa_set_add_pre_hashed(struct set *set, uint32_t hash, const void *key)
 340 {
 341    assert(set->key_hash_function == NULL ||
 342           hash == set->key_hash_function(key));
 343    return set_add(set, hash, key);
 344 }
 345
 346 /**
 347  * This function deletes the given hash table entry.
 348  *
 349  * Note that deletion doesn't otherwise modify the table, so an iteration over
 350  * the table deleting entries is safe.
 351  */
 352 void
 353 _mesa_set_remove(struct set *ht, struct set_entry *entry)
 354 {
 355    if (!entry)
 356       return;
 357
 358    entry->key = deleted_key;
 359    ht->entries--;
 360    ht->deleted_entries++;
 361 }
 362
 363 /**
 364  * This function is an iterator over the hash table.
 365  *
 366  * Pass in NULL for the first entry, as in the start of a for loop.  Note that
 367  * an iteration over the table is O(table_size) not O(entries).
 368  */
 369 struct set_entry *
 370 _mesa_set_next_entry(const struct set *ht, struct set_entry *entry)
 371 {
 372    if (entry == NULL)
 373       entry = ht->table;
 374    else
 375       entry = entry + 1;
 376
 377    for (; entry != ht->table + ht->size; entry++) {
 378       if (entry_is_present(entry)) {
 379          return entry;
 380       }
 381    }
 382
 383    return NULL;
 384 }
 385
 386 struct set_entry *
 387 _mesa_set_random_entry(struct set *ht,
 388                        int (*predicate)(struct set_entry *entry))
 389 {
 390    struct set_entry *entry;
 391    uint32_t i = rand() % ht->size;
 392
 393    if (ht->entries == 0)
 394       return NULL;
 395
 396    for (entry = ht->table + i; entry != ht->table + ht->size; entry++) {
 397       if (entry_is_present(entry) &&
 398           (!predicate || predicate(entry))) {
 399          return entry;
 400       }
 401    }
 402
 403    for (entry = ht->table; entry != ht->table + i; entry++) {
 404       if (entry_is_present(entry) &&
 405           (!predicate || predicate(entry))) {
 406          return entry;
 407       }
 408    }
 409
 410    return NULL;
 411 }