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 typedef 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  */
  42 static 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
  53 static 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
  66 struct 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
  74 typedef void (*ewah_callback)(size_t pos, void *);
  75
  76 struct ewah_bitmap *ewah_pool_new(void);
  77 void ewah_pool_free(struct ewah_bitmap *self);
  78
  79 /**
  80  * Allocate a new EWAH Compressed bitmap
  81  */
  82 struct ewah_bitmap *ewah_new(void);
  83
  84 /**
  85  * Clear all the bits in the bitmap. Does not free or resize
  86  * memory.
  87  */
  88 void ewah_clear(struct ewah_bitmap *self);
  89
  90 /**
  91  * Free all the memory of the bitmap
  92  */
  93 void ewah_free(struct ewah_bitmap *self);
  94
  95 int ewah_serialize_to(struct ewah_bitmap *self,
  96                       int (*write_fun)(void *out, const void *buf, size_t len),
  97                       void *out);
  98 int ewah_serialize(struct ewah_bitmap *self, int fd);
  99 int ewah_serialize_native(struct ewah_bitmap *self, int fd);
 100
 101 int ewah_deserialize(struct ewah_bitmap *self, int fd);
 102 int ewah_read_mmap(struct ewah_bitmap *self, const void *map, size_t len);
 103
 104 uint32_t ewah_checksum(struct ewah_bitmap *self);
 105
 106 /**
 107  * Logical not (bitwise negation) in-place on the bitmap
 108  *
 109  * This operation is linear time based on the size of the bitmap.
 110  */
 111 void ewah_not(struct ewah_bitmap *self);
 112
 113 /**
 114  * Call the given callback with the position of every single bit
 115  * that has been set on the bitmap.
 116  *
 117  * This is an efficient operation that does not fully decompress
 118  * the bitmap.
 119  */
 120 void ewah_each_bit(struct ewah_bitmap *self, ewah_callback callback, void *payload);
 121
 122 /**
 123  * Set a given bit on the bitmap.
 124  *
 125  * The bit at position `pos` will be set to true. Because of the
 126  * way that the bitmap is compressed, a set bit cannot be unset
 127  * later on.
 128  *
 129  * Furthermore, since the bitmap uses streaming compression, bits
 130  * can only set incrementally.
 131  *
 132  * E.g.
 133  *              ewah_set(bitmap, 1); // ok
 134  *              ewah_set(bitmap, 76); // ok
 135  *              ewah_set(bitmap, 77); // ok
 136  *              ewah_set(bitmap, 8712800127); // ok
 137  *              ewah_set(bitmap, 25); // failed, assert raised
 138  */
 139 void ewah_set(struct ewah_bitmap *self, size_t i);
 140
 141 struct ewah_iterator {
 142         const eword_t *buffer;
 143         size_t buffer_size;
 144
 145         size_t pointer;
 146         eword_t compressed, literals;
 147         eword_t rl, lw;
 148         int b;
 149 };
 150
 151 /**
 152  * Initialize a new iterator to run through the bitmap in uncompressed form.
 153  *
 154  * The iterator can be stack allocated. The underlying bitmap must not be freed
 155  * before the iteration is over.
 156  *
 157  * E.g.
 158  *
 159  *              struct ewah_bitmap *bitmap = ewah_new();
 160  *              struct ewah_iterator it;
 161  *
 162  *              ewah_iterator_init(&it, bitmap);
 163  */
 164 void ewah_iterator_init(struct ewah_iterator *it, struct ewah_bitmap *parent);
 165
 166 /**
 167  * Yield every single word in the bitmap in uncompressed form. This is:
 168  * yield single words (32-64 bits) where each bit represents an actual
 169  * bit from the bitmap.
 170  *
 171  * Return: true if a word was yield, false if there are no words left
 172  */
 173 int ewah_iterator_next(eword_t *next, struct ewah_iterator *it);
 174
 175 void ewah_or(
 176         struct ewah_bitmap *ewah_i,
 177         struct ewah_bitmap *ewah_j,
 178         struct ewah_bitmap *out);
 179
 180 void ewah_and_not(
 181         struct ewah_bitmap *ewah_i,
 182         struct ewah_bitmap *ewah_j,
 183         struct ewah_bitmap *out);
 184
 185 void ewah_xor(
 186         struct ewah_bitmap *ewah_i,
 187         struct ewah_bitmap *ewah_j,
 188         struct ewah_bitmap *out);
 189
 190 void ewah_and(
 191         struct ewah_bitmap *ewah_i,
 192         struct ewah_bitmap *ewah_j,
 193         struct ewah_bitmap *out);
 194
 195 /**
 196  * Direct word access
 197  */
 198 size_t ewah_add_empty_words(struct ewah_bitmap *self, int v, size_t number);
 199 void ewah_add_dirty_words(
 200         struct ewah_bitmap *self, const eword_t *buffer, size_t number, int negate);
 201 size_t ewah_add(struct ewah_bitmap *self, eword_t word);
 202
 203
 204 /**
 205  * Uncompressed, old-school bitmap that can be efficiently compressed
 206  * into an `ewah_bitmap`.
 207  */
 208 struct bitmap {
 209         eword_t *words;
 210         size_t word_alloc;
 211 };
 212
 213 struct bitmap *bitmap_new(void);
 214 void bitmap_set(struct bitmap *self, size_t pos);
 215 void bitmap_clear(struct bitmap *self, size_t pos);
 216 int bitmap_get(struct bitmap *self, size_t pos);
 217 void bitmap_reset(struct bitmap *self);
 218 void bitmap_free(struct bitmap *self);
 219 int bitmap_equals(struct bitmap *self, struct bitmap *other);
 220 int bitmap_is_subset(struct bitmap *self, struct bitmap *super);
 221
 222 struct ewah_bitmap * bitmap_to_ewah(struct bitmap *bitmap);
 223 struct bitmap *ewah_to_bitmap(struct ewah_bitmap *ewah);
 224
 225 void bitmap_and_not(struct bitmap *self, struct bitmap *other);
 226 void bitmap_or_ewah(struct bitmap *self, struct ewah_bitmap *other);
 227 void bitmap_or(struct bitmap *self, const struct bitmap *other);
 228
 229 void bitmap_each_bit(struct bitmap *self, ewah_callback callback, void *data);
 230 size_t bitmap_popcount(struct bitmap *self);
 231
 232 #endif