ewah/ewok.h

   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
  33 struct strbuf;
  34 typedef uint64_t eword_t;
  35 #define BITS_IN_WORD (sizeof(eword_t) * 8)
  36
  37 /**
  38  * Do not use __builtin_popcountll. The GCC implementation
  39  * is notoriously slow on all platforms.
  40  *
  41  * See: http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36041
  42  */
  43 static inline uint32_t ewah_bit_popcount64(uint64_t x)
  44 {
  45         x = (x & 0x5555555555555555ULL) + ((x >>  1) & 0x5555555555555555ULL);
  46         x = (x & 0x3333333333333333ULL) + ((x >>  2) & 0x3333333333333333ULL);
  47         x = (x & 0x0F0F0F0F0F0F0F0FULL) + ((x >>  4) & 0x0F0F0F0F0F0F0F0FULL);
  48         return (x * 0x0101010101010101ULL) >> 56;
  49 }
  50
  51 /* __builtin_ctzll was not available until 3.4.0 */
  52 #if defined(__GNUC__) && (__GNUC__ > 3 || (__GNUC__ == 3  && __GNUC_MINOR > 3))
  53 #define ewah_bit_ctz64(x) __builtin_ctzll(x)
  54 #else
  55 static inline int ewah_bit_ctz64(uint64_t x)
  56 {
  57         int n = 0;
  58         if ((x & 0xffffffff) == 0) { x >>= 32; n += 32; }
  59         if ((x &     0xffff) == 0) { x >>= 16; n += 16; }
  60         if ((x &       0xff) == 0) { x >>=  8; n +=  8; }
  61         if ((x &        0xf) == 0) { x >>=  4; n +=  4; }
  62         if ((x &        0x3) == 0) { x >>=  2; n +=  2; }
  63         if ((x &        0x1) == 0) { x >>=  1; n +=  1; }
  64         return n + !x;
  65 }
  66 #endif
  67
  68 struct ewah_bitmap {
  69         eword_t *buffer;
  70         size_t buffer_size;
  71         size_t alloc_size;
  72         size_t bit_size;
  73         eword_t *rlw;
  74 };
  75
  76 typedef void (*ewah_callback)(size_t pos, void *);
  77
  78 struct ewah_bitmap *ewah_pool_new(void);
  79 void ewah_pool_free(struct ewah_bitmap *self);
  80
  81 /**
  82  * Allocate a new EWAH Compressed bitmap
  83  */
  84 struct ewah_bitmap *ewah_new(void);
  85
  86 /**
  87  * Clear all the bits in the bitmap. Does not free or resize
  88  * memory.
  89  */
  90 void ewah_clear(struct ewah_bitmap *self);
  91
  92 /**
  93  * Free all the memory of the bitmap
  94  */
  95 void ewah_free(struct ewah_bitmap *self);
  96
  97 int ewah_serialize_to(struct ewah_bitmap *self,
  98                       int (*write_fun)(void *out, const void *buf, size_t len),
  99                       void *out);
 100 int ewah_serialize(struct ewah_bitmap *self, int fd);
 101 int ewah_serialize_native(struct ewah_bitmap *self, int fd);
 102 int ewah_serialize_strbuf(struct ewah_bitmap *self, struct strbuf *);
 103
 104 int ewah_deserialize(struct ewah_bitmap *self, int fd);
 105 int ewah_read_mmap(struct ewah_bitmap *self, const void *map, size_t len);
 106
 107 uint32_t ewah_checksum(struct ewah_bitmap *self);
 108
 109 /**
 110  * Logical not (bitwise negation) in-place on the bitmap
 111  *
 112  * This operation is linear time based on the size of the bitmap.
 113  */
 114 void ewah_not(struct ewah_bitmap *self);
 115
 116 /**
 117  * Call the given callback with the position of every single bit
 118  * that has been set on the bitmap.
 119  *
 120  * This is an efficient operation that does not fully decompress
 121  * the bitmap.
 122  */
 123 void ewah_each_bit(struct ewah_bitmap *self, ewah_callback callback, void *payload);
 124
 125 /**
 126  * Set a given bit on the bitmap.
 127  *
 128  * The bit at position `pos` will be set to true. Because of the
 129  * way that the bitmap is compressed, a set bit cannot be unset
 130  * later on.
 131  *
 132  * Furthermore, since the bitmap uses streaming compression, bits
 133  * can only set incrementally.
 134  *
 135  * E.g.
 136  *              ewah_set(bitmap, 1); // ok
 137  *              ewah_set(bitmap, 76); // ok
 138  *              ewah_set(bitmap, 77); // ok
 139  *              ewah_set(bitmap, 8712800127); // ok
 140  *              ewah_set(bitmap, 25); // failed, assert raised
 141  */
 142 void ewah_set(struct ewah_bitmap *self, size_t i);
 143
 144 struct ewah_iterator {
 145         const eword_t *buffer;
 146         size_t buffer_size;
 147
 148         size_t pointer;
 149         eword_t compressed, literals;
 150         eword_t rl, lw;
 151         int b;
 152 };
 153
 154 /**
 155  * Initialize a new iterator to run through the bitmap in uncompressed form.
 156  *
 157  * The iterator can be stack allocated. The underlying bitmap must not be freed
 158  * before the iteration is over.
 159  *
 160  * E.g.
 161  *
 162  *              struct ewah_bitmap *bitmap = ewah_new();
 163  *              struct ewah_iterator it;
 164  *
 165  *              ewah_iterator_init(&it, bitmap);
 166  */
 167 void ewah_iterator_init(struct ewah_iterator *it, struct ewah_bitmap *parent);
 168
 169 /**
 170  * Yield every single word in the bitmap in uncompressed form. This is:
 171  * yield single words (32-64 bits) where each bit represents an actual
 172  * bit from the bitmap.
 173  *
 174  * Return: true if a word was yield, false if there are no words left
 175  */
 176 int ewah_iterator_next(eword_t *next, struct ewah_iterator *it);
 177
 178 void ewah_or(
 179         struct ewah_bitmap *ewah_i,
 180         struct ewah_bitmap *ewah_j,
 181         struct ewah_bitmap *out);
 182
 183 void ewah_and_not(
 184         struct ewah_bitmap *ewah_i,
 185         struct ewah_bitmap *ewah_j,
 186         struct ewah_bitmap *out);
 187
 188 void ewah_xor(
 189         struct ewah_bitmap *ewah_i,
 190         struct ewah_bitmap *ewah_j,
 191         struct ewah_bitmap *out);
 192
 193 void ewah_and(
 194         struct ewah_bitmap *ewah_i,
 195         struct ewah_bitmap *ewah_j,
 196         struct ewah_bitmap *out);
 197
 198 /**
 199  * Direct word access
 200  */
 201 size_t ewah_add_empty_words(struct ewah_bitmap *self, int v, size_t number);
 202 void ewah_add_dirty_words(
 203         struct ewah_bitmap *self, const eword_t *buffer, size_t number, int negate);
 204 size_t ewah_add(struct ewah_bitmap *self, eword_t word);
 205
 206
 207 /**
 208  * Uncompressed, old-school bitmap that can be efficiently compressed
 209  * into an `ewah_bitmap`.
 210  */
 211 struct bitmap {
 212         eword_t *words;
 213         size_t word_alloc;
 214 };
 215
 216 struct bitmap *bitmap_new(void);
 217 void bitmap_set(struct bitmap *self, size_t pos);
 218 void bitmap_clear(struct bitmap *self, size_t pos);
 219 int bitmap_get(struct bitmap *self, size_t pos);
 220 void bitmap_reset(struct bitmap *self);
 221 void bitmap_free(struct bitmap *self);
 222 int bitmap_equals(struct bitmap *self, struct bitmap *other);
 223 int bitmap_is_subset(struct bitmap *self, struct bitmap *super);
 224
 225 struct ewah_bitmap * bitmap_to_ewah(struct bitmap *bitmap);
 226 struct bitmap *ewah_to_bitmap(struct ewah_bitmap *ewah);
 227
 228 void bitmap_and_not(struct bitmap *self, struct bitmap *other);
 229 void bitmap_or_ewah(struct bitmap *self, struct ewah_bitmap *other);
 230 void bitmap_or(struct bitmap *self, const struct bitmap *other);
 231
 232 void bitmap_each_bit(struct bitmap *self, ewah_callback callback, void *data);
 233 size_t bitmap_popcount(struct bitmap *self);
 234
 235 #endif