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 #include <string.h>
  38
  39 #include "hash_table.h"
  40 #include "macros.h"
  41 #include "ralloc.h"
  42 #include "set.h"
  43
  44 /*
  45  * From Knuth -- a good choice for hash/rehash values is p, p-2 where
  46  * p and p-2 are both prime.  These tables are sized to have an extra 10%
  47  * free to avoid exponential performance degradation as the hash table fills
  48  */
  49
  50 static const uint32_t deleted_key_value;
  51 static const void *deleted_key = &deleted_key_value;
  52
  53 static const struct {
  54    uint32_t max_entries, size, rehash;
  55 } hash_sizes[] = {
  56    { 2,            5,            3            },
  57    { 4,            7,            5            },
  58    { 8,            13,           11           },
  59    { 16,           19,           17           },
  60    { 32,           43,           41           },
  61    { 64,           73,           71           },
  62    { 128,          151,          149          },
  63    { 256,          283,          281          },
  64    { 512,          571,          569          },
  65    { 1024,         1153,         1151         },
  66    { 2048,         2269,         2267         },
  67    { 4096,         4519,         4517         },
  68    { 8192,         9013,         9011         },
  69    { 16384,        18043,        18041        },
  70    { 32768,        36109,        36107        },
  71    { 65536,        72091,        72089        },
  72    { 131072,       144409,       144407       },
  73    { 262144,       288361,       288359       },
  74    { 524288,       576883,       576881       },
  75    { 1048576,      1153459,      1153457      },
  76    { 2097152,      2307163,      2307161      },
  77    { 4194304,      4613893,      4613891      },
  78    { 8388608,      9227641,      9227639      },
  79    { 16777216,     18455029,     18455027     },
  80    { 33554432,     36911011,     36911009     },
  81    { 67108864,     73819861,     73819859     },
  82    { 134217728,    147639589,    147639587    },
  83    { 268435456,    295279081,    295279079    },
  84    { 536870912,    590559793,    590559791    },
  85    { 1073741824,   1181116273,   1181116271   },
  86    { 2147483648ul, 2362232233ul, 2362232231ul }
  87 };
  88
  89 static int
  90 entry_is_free(struct set_entry *entry)
  91 {
  92    return entry->key == NULL;
  93 }
  94
  95 static int
  96 entry_is_deleted(struct set_entry *entry)
  97 {
  98    return entry->key == deleted_key;
  99 }
 100
 101 static int
 102 entry_is_present(struct set_entry *entry)
 103 {
 104    return entry->key != NULL && entry->key != deleted_key;
 105 }
 106
 107 struct set *
 108 _mesa_set_create(void *mem_ctx,
 109                  uint32_t (*key_hash_function)(const void *key),
 110                  bool (*key_equals_function)(const void *a,
 111                                              const void *b))
 112 {
 113    struct set *ht;
 114
 115    ht = ralloc(mem_ctx, struct set);
 116    if (ht == NULL)
 117       return NULL;
 118
 119    ht->size_index = 0;
 120    ht->size = hash_sizes[ht->size_index].size;
 121    ht->rehash = hash_sizes[ht->size_index].rehash;
 122    ht->max_entries = hash_sizes[ht->size_index].max_entries;
 123    ht->key_hash_function = key_hash_function;
 124    ht->key_equals_function = key_equals_function;
 125    ht->table = rzalloc_array(ht, struct set_entry, ht->size);
 126    ht->entries = 0;
 127    ht->deleted_entries = 0;
 128
 129    if (ht->table == NULL) {
 130       ralloc_free(ht);
 131       return NULL;
 132    }
 133
 134    return ht;
 135 }
 136
 137 struct set *
 138 _mesa_set_clone(struct set *set, void *dst_mem_ctx)
 139 {
 140    struct set *clone;
 141
 142    clone = ralloc(dst_mem_ctx, struct set);
 143    if (clone == NULL)
 144       return NULL;
 145
 146    memcpy(clone, set, sizeof(struct set));
 147
 148    clone->table = ralloc_array(clone, struct set_entry, clone->size);
 149    if (clone->table == NULL) {
 150       ralloc_free(clone);
 151       return NULL;
 152    }
 153
 154    memcpy(clone->table, set->table, clone->size * sizeof(struct set_entry));
 155
 156    return clone;
 157 }
 158
 159 /**
 160  * Frees the given set.
 161  *
 162  * If delete_function is passed, it gets called on each entry present before
 163  * freeing.
 164  */
 165 void
 166 _mesa_set_destroy(struct set *ht, void (*delete_function)(struct set_entry *entry))
 167 {
 168    if (!ht)
 169       return;
 170
 171    if (delete_function) {
 172       set_foreach (ht, entry) {
 173          delete_function(entry);
 174       }
 175    }
 176    ralloc_free(ht->table);
 177    ralloc_free(ht);
 178 }
 179
 180 /**
 181  * Clears all values from the given set.
 182  *
 183  * If delete_function is passed, it gets called on each entry present before
 184  * the set is cleared.
 185  */
 186 void
 187 _mesa_set_clear(struct set *set, void (*delete_function)(struct set_entry *entry))
 188 {
 189    if (!set)
 190       return;
 191
 192    set_foreach (set, entry) {
 193       if (delete_function)
 194          delete_function(entry);
 195       entry->key = deleted_key;
 196    }
 197
 198    set->entries = set->deleted_entries = 0;
 199 }
 200
 201 /**
 202  * Finds a set entry with the given key and hash of that key.
 203  *
 204  * Returns NULL if no entry is found.
 205  */
 206 static struct set_entry *
 207 set_search(const struct set *ht, uint32_t hash, const void *key)
 208 {
 209    uint32_t hash_address;
 210
 211    hash_address = hash % ht->size;
 212    do {
 213       uint32_t double_hash;
 214
 215       struct set_entry *entry = ht->table + hash_address;
 216
 217       if (entry_is_free(entry)) {
 218          return NULL;
 219       } else if (entry_is_present(entry) && entry->hash == hash) {
 220          if (ht->key_equals_function(key, entry->key)) {
 221             return entry;
 222          }
 223       }
 224
 225       double_hash = 1 + hash % ht->rehash;
 226
 227       hash_address = (hash_address + double_hash) % ht->size;
 228    } while (hash_address != hash % ht->size);
 229
 230    return NULL;
 231 }
 232
 233 struct set_entry *
 234 _mesa_set_search(const struct set *set, const void *key)
 235 {
 236    assert(set->key_hash_function);
 237    return set_search(set, set->key_hash_function(key), key);
 238 }
 239
 240 struct set_entry *
 241 _mesa_set_search_pre_hashed(const struct set *set, uint32_t hash,
 242                             const void *key)
 243 {
 244    assert(set->key_hash_function == NULL ||
 245           hash == set->key_hash_function(key));
 246    return set_search(set, hash, key);
 247 }
 248
 249 static struct set_entry *
 250 set_add(struct set *ht, uint32_t hash, const void *key);
 251
 252 static void
 253 set_rehash(struct set *ht, unsigned new_size_index)
 254 {
 255    struct set old_ht;
 256    struct set_entry *table;
 257
 258    if (new_size_index >= ARRAY_SIZE(hash_sizes))
 259       return;
 260
 261    table = rzalloc_array(ht, struct set_entry,
 262                          hash_sizes[new_size_index].size);
 263    if (table == NULL)
 264       return;
 265
 266    old_ht = *ht;
 267
 268    ht->table = table;
 269    ht->size_index = new_size_index;
 270    ht->size = hash_sizes[ht->size_index].size;
 271    ht->rehash = hash_sizes[ht->size_index].rehash;
 272    ht->max_entries = hash_sizes[ht->size_index].max_entries;
 273    ht->entries = 0;
 274    ht->deleted_entries = 0;
 275
 276    set_foreach(&old_ht, entry) {
 277       set_add(ht, entry->hash, entry->key);
 278    }
 279
 280    ralloc_free(old_ht.table);
 281 }
 282
 283 /**
 284  * Inserts the key with the given hash into the table.
 285  *
 286  * Note that insertion may rearrange the table on a resize or rehash,
 287  * so previously found hash_entries are no longer valid after this function.
 288  */
 289 static struct set_entry *
 290 set_add(struct set *ht, uint32_t hash, const void *key)
 291 {
 292    uint32_t hash_address;
 293    struct set_entry *available_entry = NULL;
 294
 295    if (ht->entries >= ht->max_entries) {
 296       set_rehash(ht, ht->size_index + 1);
 297    } else if (ht->deleted_entries + ht->entries >= ht->max_entries) {
 298       set_rehash(ht, ht->size_index);
 299    }
 300
 301    hash_address = hash % ht->size;
 302    do {
 303       struct set_entry *entry = ht->table + hash_address;
 304       uint32_t double_hash;
 305
 306       if (!entry_is_present(entry)) {
 307          /* Stash the first available entry we find */
 308          if (available_entry == NULL)
 309             available_entry = entry;
 310          if (entry_is_free(entry))
 311             break;
 312       }
 313
 314       /* Implement replacement when another insert happens
 315        * with a matching key.  This is a relatively common
 316        * feature of hash tables, with the alternative
 317        * generally being "insert the new value as well, and
 318        * return it first when the key is searched for".
 319        *
 320        * Note that the hash table doesn't have a delete callback.
 321        * If freeing of old keys is required to avoid memory leaks,
 322        * perform a search before inserting.
 323        */
 324       if (!entry_is_deleted(entry) &&
 325           entry->hash == hash &&
 326           ht->key_equals_function(key, entry->key)) {
 327          entry->key = key;
 328          return entry;
 329       }
 330
 331       double_hash = 1 + hash % ht->rehash;
 332
 333       hash_address = (hash_address + double_hash) % ht->size;
 334    } while (hash_address != hash % ht->size);
 335
 336    if (available_entry) {
 337       if (entry_is_deleted(available_entry))
 338          ht->deleted_entries--;
 339       available_entry->hash = hash;
 340       available_entry->key = key;
 341       ht->entries++;
 342       return available_entry;
 343    }
 344
 345    /* We could hit here if a required resize failed. An unchecked-malloc
 346     * application could ignore this result.
 347     */
 348    return NULL;
 349 }
 350
 351 struct set_entry *
 352 _mesa_set_add(struct set *set, const void *key)
 353 {
 354    assert(set->key_hash_function);
 355    return set_add(set, set->key_hash_function(key), key);
 356 }
 357
 358 struct set_entry *
 359 _mesa_set_add_pre_hashed(struct set *set, uint32_t hash, const void *key)
 360 {
 361    assert(set->key_hash_function == NULL ||
 362           hash == set->key_hash_function(key));
 363    return set_add(set, hash, key);
 364 }
 365
 366 /**
 367  * This function deletes the given hash table entry.
 368  *
 369  * Note that deletion doesn't otherwise modify the table, so an iteration over
 370  * the table deleting entries is safe.
 371  */
 372 void
 373 _mesa_set_remove(struct set *ht, struct set_entry *entry)
 374 {
 375    if (!entry)
 376       return;
 377
 378    entry->key = deleted_key;
 379    ht->entries--;
 380    ht->deleted_entries++;
 381 }
 382
 383 /**
 384  * Removes the entry with the corresponding key, if exists.
 385  */
 386 void
 387 _mesa_set_remove_key(struct set *set, const void *key)
 388 {
 389    _mesa_set_remove(set, _mesa_set_search(set, key));
 390 }
 391
 392 /**
 393  * This function is an iterator over the hash table.
 394  *
 395  * Pass in NULL for the first entry, as in the start of a for loop.  Note that
 396  * an iteration over the table is O(table_size) not O(entries).
 397  */
 398 struct set_entry *
 399 _mesa_set_next_entry(const struct set *ht, struct set_entry *entry)
 400 {
 401    if (entry == NULL)
 402       entry = ht->table;
 403    else
 404       entry = entry + 1;
 405
 406    for (; entry != ht->table + ht->size; entry++) {
 407       if (entry_is_present(entry)) {
 408          return entry;
 409       }
 410    }
 411
 412    return NULL;
 413 }
 414
 415 struct set_entry *
 416 _mesa_set_random_entry(struct set *ht,
 417                        int (*predicate)(struct set_entry *entry))
 418 {
 419    struct set_entry *entry;
 420    uint32_t i = rand() % ht->size;
 421
 422    if (ht->entries == 0)
 423       return NULL;
 424
 425    for (entry = ht->table + i; entry != ht->table + ht->size; entry++) {
 426       if (entry_is_present(entry) &&
 427           (!predicate || predicate(entry))) {
 428          return entry;
 429       }
 430    }
 431
 432    for (entry = ht->table; entry != ht->table + i; entry++) {
 433       if (entry_is_present(entry) &&
 434           (!predicate || predicate(entry))) {
 435          return entry;
 436       }
 437    }
 438
 439    return NULL;
 440 }
 441
 442 /**
 443  * Helper to create a set with pointer keys.
 444  */
 445 struct set *
 446 _mesa_pointer_set_create(void *mem_ctx)
 447 {
 448    return _mesa_set_create(mem_ctx, _mesa_hash_pointer,
 449                            _mesa_key_pointer_equal);
 450 }