ewah / ewok.hon commit ewah_io: delete unused 'ewah_serialize()' (83ea4e1)
   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, see <http://www.gnu.org/licenses/>.
  18 */
  19#ifndef __EWOK_BITMAP_H__
  20#define __EWOK_BITMAP_H__
  21
  22struct strbuf;
  23typedef uint64_t eword_t;
  24#define BITS_IN_EWORD (sizeof(eword_t) * 8)
  25
  26/**
  27 * Do not use __builtin_popcountll. The GCC implementation
  28 * is notoriously slow on all platforms.
  29 *
  30 * See: http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36041
  31 */
  32static inline uint32_t ewah_bit_popcount64(uint64_t x)
  33{
  34        x = (x & 0x5555555555555555ULL) + ((x >>  1) & 0x5555555555555555ULL);
  35        x = (x & 0x3333333333333333ULL) + ((x >>  2) & 0x3333333333333333ULL);
  36        x = (x & 0x0F0F0F0F0F0F0F0FULL) + ((x >>  4) & 0x0F0F0F0F0F0F0F0FULL);
  37        return (x * 0x0101010101010101ULL) >> 56;
  38}
  39
  40/* __builtin_ctzll was not available until 3.4.0 */
  41#if defined(__GNUC__) && (__GNUC__ > 3 || (__GNUC__ == 3  && __GNUC_MINOR > 3))
  42#define ewah_bit_ctz64(x) __builtin_ctzll(x)
  43#else
  44static inline int ewah_bit_ctz64(uint64_t x)
  45{
  46        int n = 0;
  47        if ((x & 0xffffffff) == 0) { x >>= 32; n += 32; }
  48        if ((x &     0xffff) == 0) { x >>= 16; n += 16; }
  49        if ((x &       0xff) == 0) { x >>=  8; n +=  8; }
  50        if ((x &        0xf) == 0) { x >>=  4; n +=  4; }
  51        if ((x &        0x3) == 0) { x >>=  2; n +=  2; }
  52        if ((x &        0x1) == 0) { x >>=  1; n +=  1; }
  53        return n + !x;
  54}
  55#endif
  56
  57struct ewah_bitmap {
  58        eword_t *buffer;
  59        size_t buffer_size;
  60        size_t alloc_size;
  61        size_t bit_size;
  62        eword_t *rlw;
  63};
  64
  65typedef void (*ewah_callback)(size_t pos, void *);
  66
  67struct ewah_bitmap *ewah_pool_new(void);
  68void ewah_pool_free(struct ewah_bitmap *self);
  69
  70/**
  71 * Allocate a new EWAH Compressed bitmap
  72 */
  73struct ewah_bitmap *ewah_new(void);
  74
  75/**
  76 * Clear all the bits in the bitmap. Does not free or resize
  77 * memory.
  78 */
  79void ewah_clear(struct ewah_bitmap *self);
  80
  81/**
  82 * Free all the memory of the bitmap
  83 */
  84void ewah_free(struct ewah_bitmap *self);
  85
  86int ewah_serialize_to(struct ewah_bitmap *self,
  87                      int (*write_fun)(void *out, const void *buf, size_t len),
  88                      void *out);
  89int ewah_serialize_native(struct ewah_bitmap *self, int fd);
  90int ewah_serialize_strbuf(struct ewah_bitmap *self, struct strbuf *);
  91
  92int ewah_deserialize(struct ewah_bitmap *self, int fd);
  93ssize_t ewah_read_mmap(struct ewah_bitmap *self, const void *map, size_t len);
  94
  95uint32_t ewah_checksum(struct ewah_bitmap *self);
  96
  97/**
  98 * Call the given callback with the position of every single bit
  99 * that has been set on the bitmap.
 100 *
 101 * This is an efficient operation that does not fully decompress
 102 * the bitmap.
 103 */
 104void ewah_each_bit(struct ewah_bitmap *self, ewah_callback callback, void *payload);
 105
 106/**
 107 * Set a given bit on the bitmap.
 108 *
 109 * The bit at position `pos` will be set to true. Because of the
 110 * way that the bitmap is compressed, a set bit cannot be unset
 111 * later on.
 112 *
 113 * Furthermore, since the bitmap uses streaming compression, bits
 114 * can only set incrementally.
 115 *
 116 * E.g.
 117 *              ewah_set(bitmap, 1); // ok
 118 *              ewah_set(bitmap, 76); // ok
 119 *              ewah_set(bitmap, 77); // ok
 120 *              ewah_set(bitmap, 8712800127); // ok
 121 *              ewah_set(bitmap, 25); // failed, assert raised
 122 */
 123void ewah_set(struct ewah_bitmap *self, size_t i);
 124
 125struct ewah_iterator {
 126        const eword_t *buffer;
 127        size_t buffer_size;
 128
 129        size_t pointer;
 130        eword_t compressed, literals;
 131        eword_t rl, lw;
 132        int b;
 133};
 134
 135/**
 136 * Initialize a new iterator to run through the bitmap in uncompressed form.
 137 *
 138 * The iterator can be stack allocated. The underlying bitmap must not be freed
 139 * before the iteration is over.
 140 *
 141 * E.g.
 142 *
 143 *              struct ewah_bitmap *bitmap = ewah_new();
 144 *              struct ewah_iterator it;
 145 *
 146 *              ewah_iterator_init(&it, bitmap);
 147 */
 148void ewah_iterator_init(struct ewah_iterator *it, struct ewah_bitmap *parent);
 149
 150/**
 151 * Yield every single word in the bitmap in uncompressed form. This is:
 152 * yield single words (32-64 bits) where each bit represents an actual
 153 * bit from the bitmap.
 154 *
 155 * Return: true if a word was yield, false if there are no words left
 156 */
 157int ewah_iterator_next(eword_t *next, struct ewah_iterator *it);
 158
 159void ewah_xor(
 160        struct ewah_bitmap *ewah_i,
 161        struct ewah_bitmap *ewah_j,
 162        struct ewah_bitmap *out);
 163
 164/**
 165 * Direct word access
 166 */
 167size_t ewah_add_empty_words(struct ewah_bitmap *self, int v, size_t number);
 168void ewah_add_dirty_words(
 169        struct ewah_bitmap *self, const eword_t *buffer, size_t number, int negate);
 170size_t ewah_add(struct ewah_bitmap *self, eword_t word);
 171
 172
 173/**
 174 * Uncompressed, old-school bitmap that can be efficiently compressed
 175 * into an `ewah_bitmap`.
 176 */
 177struct bitmap {
 178        eword_t *words;
 179        size_t word_alloc;
 180};
 181
 182struct bitmap *bitmap_new(void);
 183void bitmap_set(struct bitmap *self, size_t pos);
 184int bitmap_get(struct bitmap *self, size_t pos);
 185void bitmap_reset(struct bitmap *self);
 186void bitmap_free(struct bitmap *self);
 187int bitmap_equals(struct bitmap *self, struct bitmap *other);
 188int bitmap_is_subset(struct bitmap *self, struct bitmap *super);
 189
 190struct ewah_bitmap * bitmap_to_ewah(struct bitmap *bitmap);
 191struct bitmap *ewah_to_bitmap(struct ewah_bitmap *ewah);
 192
 193void bitmap_and_not(struct bitmap *self, struct bitmap *other);
 194void bitmap_or_ewah(struct bitmap *self, struct ewah_bitmap *other);
 195void bitmap_or(struct bitmap *self, const struct bitmap *other);
 196
 197size_t bitmap_popcount(struct bitmap *self);
 198
 199#endif