Ulrich Drepper | 16303e8 | 2008-02-01 02:19:56 +0000 | [diff] [blame] | 1 | /* Functions to compute MD5 message digest of files or memory blocks. |
| 2 | according to the definition of MD5 in RFC 1321 from April 1992. |
| 3 | Copyright (C) 1995,1996,1997,1999,2000,2001,2005 Red Hat, Inc. |
| 4 | This file is part of Red Hat elfutils. |
| 5 | Written by Ulrich Drepper <drepper@redhat.com>, 1995. |
| 6 | |
| 7 | Red Hat elfutils is free software; you can redistribute it and/or modify |
| 8 | it under the terms of the GNU General Public License as published by the |
| 9 | Free Software Foundation; version 2 of the License. |
| 10 | |
| 11 | Red Hat elfutils is distributed in the hope that it will be useful, but |
| 12 | WITHOUT ANY WARRANTY; without even the implied warranty of |
| 13 | MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU |
| 14 | General Public License for more details. |
| 15 | |
| 16 | You should have received a copy of the GNU General Public License along |
| 17 | with Red Hat elfutils; if not, write to the Free Software Foundation, |
| 18 | Inc., 51 Franklin Street, Fifth Floor, Boston MA 02110-1301 USA. |
| 19 | |
| 20 | Red Hat elfutils is an included package of the Open Invention Network. |
| 21 | An included package of the Open Invention Network is a package for which |
| 22 | Open Invention Network licensees cross-license their patents. No patent |
| 23 | license is granted, either expressly or impliedly, by designation as an |
| 24 | included package. Should you wish to participate in the Open Invention |
| 25 | Network licensing program, please visit www.openinventionnetwork.com |
| 26 | <http://www.openinventionnetwork.com>. */ |
| 27 | |
| 28 | #ifdef HAVE_CONFIG_H |
| 29 | # include <config.h> |
| 30 | #endif |
| 31 | |
| 32 | #include <endian.h> |
| 33 | #include <stdlib.h> |
| 34 | #include <string.h> |
| 35 | #include <sys/types.h> |
| 36 | |
| 37 | #include "md5.h" |
| 38 | |
| 39 | #if __BYTE_ORDER == __BIG_ENDIAN |
| 40 | # define SWAP(n) \ |
| 41 | (((n) << 24) | (((n) & 0xff00) << 8) | (((n) >> 8) & 0xff00) | ((n) >> 24)) |
| 42 | #else |
| 43 | # define SWAP(n) (n) |
| 44 | #endif |
| 45 | |
| 46 | |
| 47 | /* This array contains the bytes used to pad the buffer to the next |
| 48 | 64-byte boundary. (RFC 1321, 3.1: Step 1) */ |
| 49 | static const unsigned char fillbuf[64] = { 0x80, 0 /* , 0, 0, ... */ }; |
| 50 | |
| 51 | |
| 52 | /* Initialize structure containing state of computation. |
| 53 | (RFC 1321, 3.3: Step 3) */ |
| 54 | void |
| 55 | md5_init_ctx (ctx) |
| 56 | struct md5_ctx *ctx; |
| 57 | { |
| 58 | ctx->A = 0x67452301; |
| 59 | ctx->B = 0xefcdab89; |
| 60 | ctx->C = 0x98badcfe; |
| 61 | ctx->D = 0x10325476; |
| 62 | |
| 63 | ctx->total[0] = ctx->total[1] = 0; |
| 64 | ctx->buflen = 0; |
| 65 | } |
| 66 | |
| 67 | /* Put result from CTX in first 16 bytes following RESBUF. The result |
| 68 | must be in little endian byte order. |
| 69 | |
| 70 | IMPORTANT: On some systems it is required that RESBUF is correctly |
| 71 | aligned for a 32 bits value. */ |
| 72 | void * |
| 73 | md5_read_ctx (ctx, resbuf) |
| 74 | const struct md5_ctx *ctx; |
| 75 | void *resbuf; |
| 76 | { |
| 77 | ((md5_uint32 *) resbuf)[0] = SWAP (ctx->A); |
| 78 | ((md5_uint32 *) resbuf)[1] = SWAP (ctx->B); |
| 79 | ((md5_uint32 *) resbuf)[2] = SWAP (ctx->C); |
| 80 | ((md5_uint32 *) resbuf)[3] = SWAP (ctx->D); |
| 81 | |
| 82 | return resbuf; |
| 83 | } |
| 84 | |
| 85 | /* Process the remaining bytes in the internal buffer and the usual |
| 86 | prolog according to the standard and write the result to RESBUF. |
| 87 | |
| 88 | IMPORTANT: On some systems it is required that RESBUF is correctly |
| 89 | aligned for a 32 bits value. */ |
| 90 | void * |
| 91 | md5_finish_ctx (ctx, resbuf) |
| 92 | struct md5_ctx *ctx; |
| 93 | void *resbuf; |
| 94 | { |
| 95 | /* Take yet unprocessed bytes into account. */ |
| 96 | md5_uint32 bytes = ctx->buflen; |
| 97 | size_t pad; |
| 98 | |
| 99 | /* Now count remaining bytes. */ |
| 100 | ctx->total[0] += bytes; |
| 101 | if (ctx->total[0] < bytes) |
| 102 | ++ctx->total[1]; |
| 103 | |
| 104 | pad = bytes >= 56 ? 64 + 56 - bytes : 56 - bytes; |
| 105 | memcpy (&ctx->buffer[bytes], fillbuf, pad); |
| 106 | |
| 107 | /* Put the 64-bit file length in *bits* at the end of the buffer. */ |
| 108 | *(md5_uint32 *) &ctx->buffer[bytes + pad] = SWAP (ctx->total[0] << 3); |
| 109 | *(md5_uint32 *) &ctx->buffer[bytes + pad + 4] = SWAP ((ctx->total[1] << 3) | |
| 110 | (ctx->total[0] >> 29)); |
| 111 | |
| 112 | /* Process last bytes. */ |
| 113 | md5_process_block (ctx->buffer, bytes + pad + 8, ctx); |
| 114 | |
| 115 | return md5_read_ctx (ctx, resbuf); |
| 116 | } |
| 117 | |
| 118 | |
| 119 | #ifdef NEED_MD5_STREAM |
| 120 | /* Compute MD5 message digest for bytes read from STREAM. The |
| 121 | resulting message digest number will be written into the 16 bytes |
| 122 | beginning at RESBLOCK. */ |
| 123 | int |
| 124 | md5_stream (stream, resblock) |
| 125 | FILE *stream; |
| 126 | void *resblock; |
| 127 | { |
| 128 | /* Important: BLOCKSIZE must be a multiple of 64. */ |
| 129 | #define BLOCKSIZE 4096 |
| 130 | struct md5_ctx ctx; |
| 131 | char buffer[BLOCKSIZE + 72]; |
| 132 | size_t sum; |
| 133 | |
| 134 | /* Initialize the computation context. */ |
| 135 | md5_init_ctx (&ctx); |
| 136 | |
| 137 | /* Iterate over full file contents. */ |
| 138 | while (1) |
| 139 | { |
| 140 | /* We read the file in blocks of BLOCKSIZE bytes. One call of the |
| 141 | computation function processes the whole buffer so that with the |
| 142 | next round of the loop another block can be read. */ |
| 143 | size_t n; |
| 144 | sum = 0; |
| 145 | |
| 146 | /* Read block. Take care for partial reads. */ |
| 147 | do |
| 148 | { |
| 149 | n = fread (buffer + sum, 1, BLOCKSIZE - sum, stream); |
| 150 | |
| 151 | sum += n; |
| 152 | } |
| 153 | while (sum < BLOCKSIZE && n != 0); |
| 154 | if (n == 0 && ferror (stream)) |
| 155 | return 1; |
| 156 | |
| 157 | /* If end of file is reached, end the loop. */ |
| 158 | if (n == 0) |
| 159 | break; |
| 160 | |
| 161 | /* Process buffer with BLOCKSIZE bytes. Note that |
| 162 | BLOCKSIZE % 64 == 0 |
| 163 | */ |
| 164 | md5_process_block (buffer, BLOCKSIZE, &ctx); |
| 165 | } |
| 166 | |
| 167 | /* Add the last bytes if necessary. */ |
| 168 | if (sum > 0) |
| 169 | md5_process_bytes (buffer, sum, &ctx); |
| 170 | |
| 171 | /* Construct result in desired memory. */ |
| 172 | md5_finish_ctx (&ctx, resblock); |
| 173 | return 0; |
| 174 | } |
| 175 | #endif |
| 176 | |
| 177 | |
| 178 | #ifdef NEED_MD5_BUFFER |
| 179 | /* Compute MD5 message digest for LEN bytes beginning at BUFFER. The |
| 180 | result is always in little endian byte order, so that a byte-wise |
| 181 | output yields to the wanted ASCII representation of the message |
| 182 | digest. */ |
| 183 | void * |
| 184 | md5_buffer (buffer, len, resblock) |
| 185 | const char *buffer; |
| 186 | size_t len; |
| 187 | void *resblock; |
| 188 | { |
| 189 | struct md5_ctx ctx; |
| 190 | |
| 191 | /* Initialize the computation context. */ |
| 192 | md5_init_ctx (&ctx); |
| 193 | |
| 194 | /* Process whole buffer but last len % 64 bytes. */ |
| 195 | md5_process_bytes (buffer, len, &ctx); |
| 196 | |
| 197 | /* Put result in desired memory area. */ |
| 198 | return md5_finish_ctx (&ctx, resblock); |
| 199 | } |
| 200 | #endif |
| 201 | |
| 202 | |
| 203 | void |
| 204 | md5_process_bytes (buffer, len, ctx) |
| 205 | const void *buffer; |
| 206 | size_t len; |
| 207 | struct md5_ctx *ctx; |
| 208 | { |
| 209 | /* When we already have some bits in our internal buffer concatenate |
| 210 | both inputs first. */ |
| 211 | if (ctx->buflen != 0) |
| 212 | { |
| 213 | size_t left_over = ctx->buflen; |
| 214 | size_t add = 128 - left_over > len ? len : 128 - left_over; |
| 215 | |
| 216 | memcpy (&ctx->buffer[left_over], buffer, add); |
| 217 | ctx->buflen += add; |
| 218 | |
| 219 | if (ctx->buflen > 64) |
| 220 | { |
| 221 | md5_process_block (ctx->buffer, ctx->buflen & ~63, ctx); |
| 222 | |
| 223 | ctx->buflen &= 63; |
| 224 | /* The regions in the following copy operation cannot overlap. */ |
| 225 | memcpy (ctx->buffer, &ctx->buffer[(left_over + add) & ~63], |
| 226 | ctx->buflen); |
| 227 | } |
| 228 | |
| 229 | buffer = (const char *) buffer + add; |
| 230 | len -= add; |
| 231 | } |
| 232 | |
| 233 | /* Process available complete blocks. */ |
| 234 | if (len >= 64) |
| 235 | { |
| 236 | #if !_STRING_ARCH_unaligned |
| 237 | /* To check alignment gcc has an appropriate operator. Other |
| 238 | compilers don't. */ |
| 239 | # if __GNUC__ >= 2 |
| 240 | # define UNALIGNED_P(p) (((md5_uintptr) p) % __alignof__ (md5_uint32) != 0) |
| 241 | # else |
| 242 | # define UNALIGNED_P(p) (((md5_uintptr) p) % sizeof (md5_uint32) != 0) |
| 243 | # endif |
| 244 | if (UNALIGNED_P (buffer)) |
| 245 | while (len > 64) |
| 246 | { |
| 247 | md5_process_block (memcpy (ctx->buffer, buffer, 64), 64, ctx); |
| 248 | buffer = (const char *) buffer + 64; |
| 249 | len -= 64; |
| 250 | } |
| 251 | else |
| 252 | #endif |
| 253 | { |
| 254 | md5_process_block (buffer, len & ~63, ctx); |
| 255 | buffer = (const char *) buffer + (len & ~63); |
| 256 | len &= 63; |
| 257 | } |
| 258 | } |
| 259 | |
| 260 | /* Move remaining bytes in internal buffer. */ |
| 261 | if (len > 0) |
| 262 | { |
| 263 | size_t left_over = ctx->buflen; |
| 264 | |
| 265 | memcpy (&ctx->buffer[left_over], buffer, len); |
| 266 | left_over += len; |
| 267 | if (left_over >= 64) |
| 268 | { |
| 269 | md5_process_block (ctx->buffer, 64, ctx); |
| 270 | left_over -= 64; |
| 271 | memcpy (ctx->buffer, &ctx->buffer[64], left_over); |
| 272 | } |
| 273 | ctx->buflen = left_over; |
| 274 | } |
| 275 | } |
| 276 | |
| 277 | |
| 278 | /* These are the four functions used in the four steps of the MD5 algorithm |
| 279 | and defined in the RFC 1321. The first function is a little bit optimized |
| 280 | (as found in Colin Plumbs public domain implementation). */ |
| 281 | /* #define FF(b, c, d) ((b & c) | (~b & d)) */ |
| 282 | #define FF(b, c, d) (d ^ (b & (c ^ d))) |
| 283 | #define FG(b, c, d) FF (d, b, c) |
| 284 | #define FH(b, c, d) (b ^ c ^ d) |
| 285 | #define FI(b, c, d) (c ^ (b | ~d)) |
| 286 | |
| 287 | /* Process LEN bytes of BUFFER, accumulating context into CTX. |
| 288 | It is assumed that LEN % 64 == 0. */ |
| 289 | |
| 290 | void |
| 291 | md5_process_block (buffer, len, ctx) |
| 292 | const void *buffer; |
| 293 | size_t len; |
| 294 | struct md5_ctx *ctx; |
| 295 | { |
| 296 | md5_uint32 correct_words[16]; |
| 297 | const md5_uint32 *words = buffer; |
| 298 | size_t nwords = len / sizeof (md5_uint32); |
| 299 | const md5_uint32 *endp = words + nwords; |
| 300 | md5_uint32 A = ctx->A; |
| 301 | md5_uint32 B = ctx->B; |
| 302 | md5_uint32 C = ctx->C; |
| 303 | md5_uint32 D = ctx->D; |
| 304 | |
| 305 | /* First increment the byte count. RFC 1321 specifies the possible |
| 306 | length of the file up to 2^64 bits. Here we only compute the |
| 307 | number of bytes. Do a double word increment. */ |
| 308 | ctx->total[0] += len; |
| 309 | if (ctx->total[0] < len) |
| 310 | ++ctx->total[1]; |
| 311 | |
| 312 | /* Process all bytes in the buffer with 64 bytes in each round of |
| 313 | the loop. */ |
| 314 | while (words < endp) |
| 315 | { |
| 316 | md5_uint32 *cwp = correct_words; |
| 317 | md5_uint32 A_save = A; |
| 318 | md5_uint32 B_save = B; |
| 319 | md5_uint32 C_save = C; |
| 320 | md5_uint32 D_save = D; |
| 321 | |
| 322 | /* First round: using the given function, the context and a constant |
| 323 | the next context is computed. Because the algorithms processing |
| 324 | unit is a 32-bit word and it is determined to work on words in |
| 325 | little endian byte order we perhaps have to change the byte order |
| 326 | before the computation. To reduce the work for the next steps |
| 327 | we store the swapped words in the array CORRECT_WORDS. */ |
| 328 | |
| 329 | #define OP(a, b, c, d, s, T) \ |
| 330 | do \ |
| 331 | { \ |
| 332 | a += FF (b, c, d) + (*cwp++ = SWAP (*words)) + T; \ |
| 333 | ++words; \ |
| 334 | CYCLIC (a, s); \ |
| 335 | a += b; \ |
| 336 | } \ |
| 337 | while (0) |
| 338 | |
| 339 | /* It is unfortunate that C does not provide an operator for |
| 340 | cyclic rotation. Hope the C compiler is smart enough. */ |
| 341 | #define CYCLIC(w, s) (w = (w << s) | (w >> (32 - s))) |
| 342 | |
| 343 | /* Before we start, one word to the strange constants. |
| 344 | They are defined in RFC 1321 as |
| 345 | |
| 346 | T[i] = (int) (4294967296.0 * fabs (sin (i))), i=1..64 |
| 347 | */ |
| 348 | |
| 349 | /* Round 1. */ |
| 350 | OP (A, B, C, D, 7, 0xd76aa478); |
| 351 | OP (D, A, B, C, 12, 0xe8c7b756); |
| 352 | OP (C, D, A, B, 17, 0x242070db); |
| 353 | OP (B, C, D, A, 22, 0xc1bdceee); |
| 354 | OP (A, B, C, D, 7, 0xf57c0faf); |
| 355 | OP (D, A, B, C, 12, 0x4787c62a); |
| 356 | OP (C, D, A, B, 17, 0xa8304613); |
| 357 | OP (B, C, D, A, 22, 0xfd469501); |
| 358 | OP (A, B, C, D, 7, 0x698098d8); |
| 359 | OP (D, A, B, C, 12, 0x8b44f7af); |
| 360 | OP (C, D, A, B, 17, 0xffff5bb1); |
| 361 | OP (B, C, D, A, 22, 0x895cd7be); |
| 362 | OP (A, B, C, D, 7, 0x6b901122); |
| 363 | OP (D, A, B, C, 12, 0xfd987193); |
| 364 | OP (C, D, A, B, 17, 0xa679438e); |
| 365 | OP (B, C, D, A, 22, 0x49b40821); |
| 366 | |
| 367 | /* For the second to fourth round we have the possibly swapped words |
| 368 | in CORRECT_WORDS. Redefine the macro to take an additional first |
| 369 | argument specifying the function to use. */ |
| 370 | #undef OP |
| 371 | #define OP(f, a, b, c, d, k, s, T) \ |
| 372 | do \ |
| 373 | { \ |
| 374 | a += f (b, c, d) + correct_words[k] + T; \ |
| 375 | CYCLIC (a, s); \ |
| 376 | a += b; \ |
| 377 | } \ |
| 378 | while (0) |
| 379 | |
| 380 | /* Round 2. */ |
| 381 | OP (FG, A, B, C, D, 1, 5, 0xf61e2562); |
| 382 | OP (FG, D, A, B, C, 6, 9, 0xc040b340); |
| 383 | OP (FG, C, D, A, B, 11, 14, 0x265e5a51); |
| 384 | OP (FG, B, C, D, A, 0, 20, 0xe9b6c7aa); |
| 385 | OP (FG, A, B, C, D, 5, 5, 0xd62f105d); |
| 386 | OP (FG, D, A, B, C, 10, 9, 0x02441453); |
| 387 | OP (FG, C, D, A, B, 15, 14, 0xd8a1e681); |
| 388 | OP (FG, B, C, D, A, 4, 20, 0xe7d3fbc8); |
| 389 | OP (FG, A, B, C, D, 9, 5, 0x21e1cde6); |
| 390 | OP (FG, D, A, B, C, 14, 9, 0xc33707d6); |
| 391 | OP (FG, C, D, A, B, 3, 14, 0xf4d50d87); |
| 392 | OP (FG, B, C, D, A, 8, 20, 0x455a14ed); |
| 393 | OP (FG, A, B, C, D, 13, 5, 0xa9e3e905); |
| 394 | OP (FG, D, A, B, C, 2, 9, 0xfcefa3f8); |
| 395 | OP (FG, C, D, A, B, 7, 14, 0x676f02d9); |
| 396 | OP (FG, B, C, D, A, 12, 20, 0x8d2a4c8a); |
| 397 | |
| 398 | /* Round 3. */ |
| 399 | OP (FH, A, B, C, D, 5, 4, 0xfffa3942); |
| 400 | OP (FH, D, A, B, C, 8, 11, 0x8771f681); |
| 401 | OP (FH, C, D, A, B, 11, 16, 0x6d9d6122); |
| 402 | OP (FH, B, C, D, A, 14, 23, 0xfde5380c); |
| 403 | OP (FH, A, B, C, D, 1, 4, 0xa4beea44); |
| 404 | OP (FH, D, A, B, C, 4, 11, 0x4bdecfa9); |
| 405 | OP (FH, C, D, A, B, 7, 16, 0xf6bb4b60); |
| 406 | OP (FH, B, C, D, A, 10, 23, 0xbebfbc70); |
| 407 | OP (FH, A, B, C, D, 13, 4, 0x289b7ec6); |
| 408 | OP (FH, D, A, B, C, 0, 11, 0xeaa127fa); |
| 409 | OP (FH, C, D, A, B, 3, 16, 0xd4ef3085); |
| 410 | OP (FH, B, C, D, A, 6, 23, 0x04881d05); |
| 411 | OP (FH, A, B, C, D, 9, 4, 0xd9d4d039); |
| 412 | OP (FH, D, A, B, C, 12, 11, 0xe6db99e5); |
| 413 | OP (FH, C, D, A, B, 15, 16, 0x1fa27cf8); |
| 414 | OP (FH, B, C, D, A, 2, 23, 0xc4ac5665); |
| 415 | |
| 416 | /* Round 4. */ |
| 417 | OP (FI, A, B, C, D, 0, 6, 0xf4292244); |
| 418 | OP (FI, D, A, B, C, 7, 10, 0x432aff97); |
| 419 | OP (FI, C, D, A, B, 14, 15, 0xab9423a7); |
| 420 | OP (FI, B, C, D, A, 5, 21, 0xfc93a039); |
| 421 | OP (FI, A, B, C, D, 12, 6, 0x655b59c3); |
| 422 | OP (FI, D, A, B, C, 3, 10, 0x8f0ccc92); |
| 423 | OP (FI, C, D, A, B, 10, 15, 0xffeff47d); |
| 424 | OP (FI, B, C, D, A, 1, 21, 0x85845dd1); |
| 425 | OP (FI, A, B, C, D, 8, 6, 0x6fa87e4f); |
| 426 | OP (FI, D, A, B, C, 15, 10, 0xfe2ce6e0); |
| 427 | OP (FI, C, D, A, B, 6, 15, 0xa3014314); |
| 428 | OP (FI, B, C, D, A, 13, 21, 0x4e0811a1); |
| 429 | OP (FI, A, B, C, D, 4, 6, 0xf7537e82); |
| 430 | OP (FI, D, A, B, C, 11, 10, 0xbd3af235); |
| 431 | OP (FI, C, D, A, B, 2, 15, 0x2ad7d2bb); |
| 432 | OP (FI, B, C, D, A, 9, 21, 0xeb86d391); |
| 433 | |
| 434 | /* Add the starting values of the context. */ |
| 435 | A += A_save; |
| 436 | B += B_save; |
| 437 | C += C_save; |
| 438 | D += D_save; |
| 439 | } |
| 440 | |
| 441 | /* Put checksum in context given as argument. */ |
| 442 | ctx->A = A; |
| 443 | ctx->B = B; |
| 444 | ctx->C = C; |
| 445 | ctx->D = D; |
| 446 | } |