ewah / ewok.hon commit Merge branch 'nd/no-more-fnmatch' (650c90a)
   1/**
   2 * Copyright 2013, GitHub, Inc
   3 * Copyright 2009-2013, Daniel Lemire, Cliff Moon,
   4 *      David McIntosh, Robert Becho, Google Inc. and Veronika Zenz
   5 *
   6 * This program is free software; you can redistribute it and/or
   7 * modify it under the terms of the GNU General Public License
   8 * as published by the Free Software Foundation; either version 2
   9 * of the License, or (at your option) any later version.
  10 *
  11 * This program is distributed in the hope that it will be useful,
  12 * but WITHOUT ANY WARRANTY; without even the implied warranty of
  13 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  14 * GNU General Public License for more details.
  15 *
  16 * You should have received a copy of the GNU General Public License
  17 * along with this program; if not, write to the Free Software
  18 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
  19 */
  20#ifndef __EWOK_BITMAP_H__
  21#define __EWOK_BITMAP_H__
  22
  23#ifndef ewah_malloc
  24#       define ewah_malloc xmalloc
  25#endif
  26#ifndef ewah_realloc
  27#       define ewah_realloc xrealloc
  28#endif
  29#ifndef ewah_calloc
  30#       define ewah_calloc xcalloc
  31#endif
  32
  33typedef uint64_t eword_t;
  34#define BITS_IN_WORD (sizeof(eword_t) * 8)
  35
  36/**
  37 * Do not use __builtin_popcountll. The GCC implementation
  38 * is notoriously slow on all platforms.
  39 *
  40 * See: http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36041
  41 */
  42static inline uint32_t ewah_bit_popcount64(uint64_t x)
  43{
  44        x = (x & 0x5555555555555555ULL) + ((x >>  1) & 0x5555555555555555ULL);
  45        x = (x & 0x3333333333333333ULL) + ((x >>  2) & 0x3333333333333333ULL);
  46        x = (x & 0x0F0F0F0F0F0F0F0FULL) + ((x >>  4) & 0x0F0F0F0F0F0F0F0FULL);
  47        return (x * 0x0101010101010101ULL) >> 56;
  48}
  49
  50#ifdef __GNUC__
  51#define ewah_bit_ctz64(x) __builtin_ctzll(x)
  52#else
  53static inline int ewah_bit_ctz64(uint64_t x)
  54{
  55        int n = 0;
  56        if ((x & 0xffffffff) == 0) { x >>= 32; n += 32; }
  57        if ((x &     0xffff) == 0) { x >>= 16; n += 16; }
  58        if ((x &       0xff) == 0) { x >>=  8; n +=  8; }
  59        if ((x &        0xf) == 0) { x >>=  4; n +=  4; }
  60        if ((x &        0x3) == 0) { x >>=  2; n +=  2; }
  61        if ((x &        0x1) == 0) { x >>=  1; n +=  1; }
  62        return n + !x;
  63}
  64#endif
  65
  66struct ewah_bitmap {
  67        eword_t *buffer;
  68        size_t buffer_size;
  69        size_t alloc_size;
  70        size_t bit_size;
  71        eword_t *rlw;
  72};
  73
  74typedef void (*ewah_callback)(size_t pos, void *);
  75
  76struct ewah_bitmap *ewah_pool_new(void);
  77void ewah_pool_free(struct ewah_bitmap *self);
  78
  79/**
  80 * Allocate a new EWAH Compressed bitmap
  81 */
  82struct ewah_bitmap *ewah_new(void);
  83
  84/**
  85 * Clear all the bits in the bitmap. Does not free or resize
  86 * memory.
  87 */
  88void ewah_clear(struct ewah_bitmap *self);
  89
  90/**
  91 * Free all the memory of the bitmap
  92 */
  93void ewah_free(struct ewah_bitmap *self);
  94
  95int ewah_serialize_to(struct ewah_bitmap *self,
  96                      int (*write_fun)(void *out, const void *buf, size_t len),
  97                      void *out);
  98int ewah_serialize(struct ewah_bitmap *self, int fd);
  99int ewah_serialize_native(struct ewah_bitmap *self, int fd);
 100
 101int ewah_deserialize(struct ewah_bitmap *self, int fd);
 102int ewah_read_mmap(struct ewah_bitmap *self, void *map, size_t len);
 103int ewah_read_mmap_native(struct ewah_bitmap *self, void *map, size_t len);
 104
 105uint32_t ewah_checksum(struct ewah_bitmap *self);
 106
 107/**
 108 * Logical not (bitwise negation) in-place on the bitmap
 109 *
 110 * This operation is linear time based on the size of the bitmap.
 111 */
 112void ewah_not(struct ewah_bitmap *self);
 113
 114/**
 115 * Call the given callback with the position of every single bit
 116 * that has been set on the bitmap.
 117 *
 118 * This is an efficient operation that does not fully decompress
 119 * the bitmap.
 120 */
 121void ewah_each_bit(struct ewah_bitmap *self, ewah_callback callback, void *payload);
 122
 123/**
 124 * Set a given bit on the bitmap.
 125 *
 126 * The bit at position `pos` will be set to true. Because of the
 127 * way that the bitmap is compressed, a set bit cannot be unset
 128 * later on.
 129 *
 130 * Furthermore, since the bitmap uses streaming compression, bits
 131 * can only set incrementally.
 132 *
 133 * E.g.
 134 *              ewah_set(bitmap, 1); // ok
 135 *              ewah_set(bitmap, 76); // ok
 136 *              ewah_set(bitmap, 77); // ok
 137 *              ewah_set(bitmap, 8712800127); // ok
 138 *              ewah_set(bitmap, 25); // failed, assert raised
 139 */
 140void ewah_set(struct ewah_bitmap *self, size_t i);
 141
 142struct ewah_iterator {
 143        const eword_t *buffer;
 144        size_t buffer_size;
 145
 146        size_t pointer;
 147        eword_t compressed, literals;
 148        eword_t rl, lw;
 149        int b;
 150};
 151
 152/**
 153 * Initialize a new iterator to run through the bitmap in uncompressed form.
 154 *
 155 * The iterator can be stack allocated. The underlying bitmap must not be freed
 156 * before the iteration is over.
 157 *
 158 * E.g.
 159 *
 160 *              struct ewah_bitmap *bitmap = ewah_new();
 161 *              struct ewah_iterator it;
 162 *
 163 *              ewah_iterator_init(&it, bitmap);
 164 */
 165void ewah_iterator_init(struct ewah_iterator *it, struct ewah_bitmap *parent);
 166
 167/**
 168 * Yield every single word in the bitmap in uncompressed form. This is:
 169 * yield single words (32-64 bits) where each bit represents an actual
 170 * bit from the bitmap.
 171 *
 172 * Return: true if a word was yield, false if there are no words left
 173 */
 174int ewah_iterator_next(eword_t *next, struct ewah_iterator *it);
 175
 176void ewah_or(
 177        struct ewah_bitmap *ewah_i,
 178        struct ewah_bitmap *ewah_j,
 179        struct ewah_bitmap *out);
 180
 181void ewah_and_not(
 182        struct ewah_bitmap *ewah_i,
 183        struct ewah_bitmap *ewah_j,
 184        struct ewah_bitmap *out);
 185
 186void ewah_xor(
 187        struct ewah_bitmap *ewah_i,
 188        struct ewah_bitmap *ewah_j,
 189        struct ewah_bitmap *out);
 190
 191void ewah_and(
 192        struct ewah_bitmap *ewah_i,
 193        struct ewah_bitmap *ewah_j,
 194        struct ewah_bitmap *out);
 195
 196/**
 197 * Direct word access
 198 */
 199size_t ewah_add_empty_words(struct ewah_bitmap *self, int v, size_t number);
 200void ewah_add_dirty_words(
 201        struct ewah_bitmap *self, const eword_t *buffer, size_t number, int negate);
 202size_t ewah_add(struct ewah_bitmap *self, eword_t word);
 203
 204
 205/**
 206 * Uncompressed, old-school bitmap that can be efficiently compressed
 207 * into an `ewah_bitmap`.
 208 */
 209struct bitmap {
 210        eword_t *words;
 211        size_t word_alloc;
 212};
 213
 214struct bitmap *bitmap_new(void);
 215void bitmap_set(struct bitmap *self, size_t pos);
 216void bitmap_clear(struct bitmap *self, size_t pos);
 217int bitmap_get(struct bitmap *self, size_t pos);
 218void bitmap_reset(struct bitmap *self);
 219void bitmap_free(struct bitmap *self);
 220int bitmap_equals(struct bitmap *self, struct bitmap *other);
 221int bitmap_is_subset(struct bitmap *self, struct bitmap *super);
 222
 223struct ewah_bitmap * bitmap_to_ewah(struct bitmap *bitmap);
 224struct bitmap *ewah_to_bitmap(struct ewah_bitmap *ewah);
 225
 226void bitmap_and_not(struct bitmap *self, struct bitmap *other);
 227void bitmap_or_ewah(struct bitmap *self, struct ewah_bitmap *other);
 228void bitmap_or(struct bitmap *self, const struct bitmap *other);
 229
 230void bitmap_each_bit(struct bitmap *self, ewah_callback callback, void *data);
 231size_t bitmap_popcount(struct bitmap *self);
 232
 233#endif