ewah/ewah_io.c

   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 #include "git-compat-util.h"
  20 #include "ewok.h"
  21 #include "strbuf.h"
  22
  23 int ewah_serialize_native(struct ewah_bitmap *self, int fd)
  24 {
  25         uint32_t write32;
  26         size_t to_write = self->buffer_size * 8;
  27
  28         /* 32 bit -- bit size for the map */
  29         write32 = (uint32_t)self->bit_size;
  30         if (write(fd, &write32, 4) != 4)
  31                 return -1;
  32
  33         /** 32 bit -- number of compressed 64-bit words */
  34         write32 = (uint32_t)self->buffer_size;
  35         if (write(fd, &write32, 4) != 4)
  36                 return -1;
  37
  38         if (write(fd, self->buffer, to_write) != to_write)
  39                 return -1;
  40
  41         /** 32 bit -- position for the RLW */
  42         write32 = self->rlw - self->buffer;
  43         if (write(fd, &write32, 4) != 4)
  44                 return -1;
  45
  46         return (3 * 4) + to_write;
  47 }
  48
  49 int ewah_serialize_to(struct ewah_bitmap *self,
  50                       int (*write_fun)(void *, const void *, size_t),
  51                       void *data)
  52 {
  53         size_t i;
  54         eword_t dump[2048];
  55         const size_t words_per_dump = sizeof(dump) / sizeof(eword_t);
  56         uint32_t bitsize, word_count, rlw_pos;
  57
  58         const eword_t *buffer;
  59         size_t words_left;
  60
  61         /* 32 bit -- bit size for the map */
  62         bitsize =  htonl((uint32_t)self->bit_size);
  63         if (write_fun(data, &bitsize, 4) != 4)
  64                 return -1;
  65
  66         /** 32 bit -- number of compressed 64-bit words */
  67         word_count =  htonl((uint32_t)self->buffer_size);
  68         if (write_fun(data, &word_count, 4) != 4)
  69                 return -1;
  70
  71         /** 64 bit x N -- compressed words */
  72         buffer = self->buffer;
  73         words_left = self->buffer_size;
  74
  75         while (words_left >= words_per_dump) {
  76                 for (i = 0; i < words_per_dump; ++i, ++buffer)
  77                         dump[i] = htonll(*buffer);
  78
  79                 if (write_fun(data, dump, sizeof(dump)) != sizeof(dump))
  80                         return -1;
  81
  82                 words_left -= words_per_dump;
  83         }
  84
  85         if (words_left) {
  86                 for (i = 0; i < words_left; ++i, ++buffer)
  87                         dump[i] = htonll(*buffer);
  88
  89                 if (write_fun(data, dump, words_left * 8) != words_left * 8)
  90                         return -1;
  91         }
  92
  93         /** 32 bit -- position for the RLW */
  94         rlw_pos = (uint8_t*)self->rlw - (uint8_t *)self->buffer;
  95         rlw_pos = htonl(rlw_pos / sizeof(eword_t));
  96
  97         if (write_fun(data, &rlw_pos, 4) != 4)
  98                 return -1;
  99
 100         return (3 * 4) + (self->buffer_size * 8);
 101 }
 102
 103 static int write_helper(void *fd, const void *buf, size_t len)
 104 {
 105         return write((intptr_t)fd, buf, len);
 106 }
 107
 108 int ewah_serialize(struct ewah_bitmap *self, int fd)
 109 {
 110         return ewah_serialize_to(self, write_helper, (void *)(intptr_t)fd);
 111 }
 112
 113 static int write_strbuf(void *user_data, const void *data, size_t len)
 114 {
 115         struct strbuf *sb = user_data;
 116         strbuf_add(sb, data, len);
 117         return len;
 118 }
 119
 120 int ewah_serialize_strbuf(struct ewah_bitmap *self, struct strbuf *sb)
 121 {
 122         return ewah_serialize_to(self, write_strbuf, sb);
 123 }
 124
 125 ssize_t ewah_read_mmap(struct ewah_bitmap *self, const void *map, size_t len)
 126 {
 127         const uint8_t *ptr = map;
 128         size_t data_len;
 129         size_t i;
 130
 131         if (len < sizeof(uint32_t))
 132                 return error("corrupt ewah bitmap: eof before bit size");
 133         self->bit_size = get_be32(ptr);
 134         ptr += sizeof(uint32_t);
 135         len -= sizeof(uint32_t);
 136
 137         if (len < sizeof(uint32_t))
 138                 return error("corrupt ewah bitmap: eof before length");
 139         self->buffer_size = self->alloc_size = get_be32(ptr);
 140         ptr += sizeof(uint32_t);
 141         len -= sizeof(uint32_t);
 142
 143         REALLOC_ARRAY(self->buffer, self->alloc_size);
 144
 145         /*
 146          * Copy the raw data for the bitmap as a whole chunk;
 147          * if we're in a little-endian platform, we'll perform
 148          * the endianness conversion in a separate pass to ensure
 149          * we're loading 8-byte aligned words.
 150          */
 151         data_len = st_mult(self->buffer_size, sizeof(eword_t));
 152         if (len < data_len)
 153                 return error("corrupt ewah bitmap: eof in data "
 154                              "(%"PRIuMAX" bytes short)",
 155                              (uintmax_t)(data_len - len));
 156         memcpy(self->buffer, ptr, data_len);
 157         ptr += data_len;
 158         len -= data_len;
 159
 160         for (i = 0; i < self->buffer_size; ++i)
 161                 self->buffer[i] = ntohll(self->buffer[i]);
 162
 163         if (len < sizeof(uint32_t))
 164                 return error("corrupt ewah bitmap: eof before rlw");
 165         self->rlw = self->buffer + get_be32(ptr);
 166         ptr += sizeof(uint32_t);
 167         len -= sizeof(uint32_t);
 168
 169         return ptr - (const uint8_t *)map;
 170 }
 171
 172 int ewah_deserialize(struct ewah_bitmap *self, int fd)
 173 {
 174         size_t i;
 175         eword_t dump[2048];
 176         const size_t words_per_dump = sizeof(dump) / sizeof(eword_t);
 177         uint32_t bitsize, word_count, rlw_pos;
 178
 179         eword_t *buffer = NULL;
 180         size_t words_left;
 181
 182         ewah_clear(self);
 183
 184         /* 32 bit -- bit size for the map */
 185         if (read(fd, &bitsize, 4) != 4)
 186                 return -1;
 187
 188         self->bit_size = (size_t)ntohl(bitsize);
 189
 190         /** 32 bit -- number of compressed 64-bit words */
 191         if (read(fd, &word_count, 4) != 4)
 192                 return -1;
 193
 194         self->buffer_size = self->alloc_size = (size_t)ntohl(word_count);
 195         REALLOC_ARRAY(self->buffer, self->alloc_size);
 196
 197         /** 64 bit x N -- compressed words */
 198         buffer = self->buffer;
 199         words_left = self->buffer_size;
 200
 201         while (words_left >= words_per_dump) {
 202                 if (read(fd, dump, sizeof(dump)) != sizeof(dump))
 203                         return -1;
 204
 205                 for (i = 0; i < words_per_dump; ++i, ++buffer)
 206                         *buffer = ntohll(dump[i]);
 207
 208                 words_left -= words_per_dump;
 209         }
 210
 211         if (words_left) {
 212                 if (read(fd, dump, words_left * 8) != words_left * 8)
 213                         return -1;
 214
 215                 for (i = 0; i < words_left; ++i, ++buffer)
 216                         *buffer = ntohll(dump[i]);
 217         }
 218
 219         /** 32 bit -- position for the RLW */
 220         if (read(fd, &rlw_pos, 4) != 4)
 221                 return -1;
 222
 223         self->rlw = self->buffer + ntohl(rlw_pos);
 224         return 0;
 225 }