src/gromacs/fileio/md5.cpp

   1 /*
   2  * This file is part of the GROMACS molecular simulation package.
   3  *
   4  * Copyright (c) 2009,2010,2011,2012,2014,2015,2018,2019, by the GROMACS development team, led by
   5  * Mark Abraham, David van der Spoel, Berk Hess, and Erik Lindahl,
   6  * and including many others, as listed in the AUTHORS file in the
   7  * top-level source directory and at http://www.gromacs.org.
   8  *
   9  * GROMACS is free software; you can redistribute it and/or
  10  * modify it under the terms of the GNU Lesser General Public License
  11  * as published by the Free Software Foundation; either version 2.1
  12  * of the License, or (at your option) any later version.
  13  *
  14  * GROMACS is distributed in the hope that it will be useful,
  15  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  16  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
  17  * Lesser General Public License for more details.
  18  *
  19  * You should have received a copy of the GNU Lesser General Public
  20  * License along with GROMACS; if not, see
  21  * http://www.gnu.org/licenses, or write to the Free Software Foundation,
  22  * Inc., 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301  USA.
  23  *
  24  * If you want to redistribute modifications to GROMACS, please
  25  * consider that scientific software is very special. Version
  26  * control is crucial - bugs must be traceable. We will be happy to
  27  * consider code for inclusion in the official distribution, but
  28  * derived work must not be called official GROMACS. Details are found
  29  * in the README & COPYING files - if they are missing, get the
  30  * official version at http://www.gromacs.org.
  31  *
  32  * To help us fund GROMACS development, we humbly ask that you cite
  33  * the research papers on the package. Check out http://www.gromacs.org.
  34  */
  35 /* This software has been altered by GROMACS for its use, including
  36  * the use of GMX_INTEGER_BIG_ENDIAN, and the renaming of the
  37  * functions md5_init, md5_append and md5_finish to have a gmx_ prefix
  38  * (to avoid name clashes). */
  39 #include "gmxpre.h"
  40
  41 #include "config.h"
  42
  43 #include <cstring>
  44
  45 #if GMX_INTEGER_BIG_ENDIAN
  46 #    define ARCH_IS_BIG_ENDIAN 1
  47 #else
  48 #    define ARCH_IS_BIG_ENDIAN 0
  49 #endif
  50
  51 /*
  52    Copyright (C) 1999, 2002 Aladdin Enterprises.  All rights reserved.
  53
  54    This software is provided 'as-is', without any express or implied
  55    warranty.  In no event will the authors be held liable for any damages
  56    arising from the use of this software.
  57
  58    Permission is granted to anyone to use this software for any purpose,
  59    including commercial applications, and to alter it and redistribute it
  60    freely, subject to the following restrictions:
  61
  62    1. The origin of this software must not be misrepresented; you must not
  63      claim that you wrote the original software. If you use this software
  64      in a product, an acknowledgment in the product documentation would be
  65      appreciated but is not required.
  66    2. Altered source versions must be plainly marked as such, and must not be
  67      misrepresented as being the original software.
  68    3. This notice may not be removed or altered from any source distribution.
  69
  70    L. Peter Deutsch
  71    ghost@aladdin.com
  72
  73  */
  74 /*
  75    Independent implementation of MD5 (RFC 1321).
  76
  77    This code implements the MD5 Algorithm defined in RFC 1321, whose
  78    text is available at
  79     http://www.ietf.org/rfc/rfc1321.txt
  80    The code is derived from the text of the RFC, including the test suite
  81    (section A.5) but excluding the rest of Appendix A.  It does not include
  82    any code or documentation that is identified in the RFC as being
  83    copyrighted.
  84
  85    The original and principal author of md5.c is L. Peter Deutsch
  86    <ghost@aladdin.com>.  Other authors are noted in the change history
  87    that follows (in reverse chronological order):
  88
  89    2002-04-13 lpd Clarified derivation from RFC 1321; now handles byte order
  90     either statically or dynamically; added missing #include <string.h>
  91     in library.
  92    2002-03-11 lpd Corrected argument list for main(), and added int return
  93     type, in test program and T value program.
  94    2002-02-21 lpd Added missing #include <stdio.h> in test program.
  95    2000-07-03 lpd Patched to eliminate warnings about "constant is
  96     unsigned in ANSI C, signed in traditional"; made test program
  97     self-checking.
  98    1999-11-04 lpd Edited comments slightly for automatic TOC extraction.
  99    1999-10-18 lpd Fixed typo in header comment (ansi2knr rather than md5).
 100    1999-05-03 lpd Original version.
 101  */
 102
 103 #include "md5.h"
 104
 105 #undef BYTE_ORDER /* 1 = big-endian, -1 = little-endian, 0 = unknown */
 106 #ifdef ARCH_IS_BIG_ENDIAN
 107 #    define BYTE_ORDER (ARCH_IS_BIG_ENDIAN ? 1 : -1)
 108 #else
 109 #    define BYTE_ORDER 0
 110 #endif
 111
 112 #define T_MASK (static_cast<md5_word_t>(~0))
 113 #define T1 /* 0xd76aa478 */ (T_MASK ^ 0x28955b87)
 114 #define T2 /* 0xe8c7b756 */ (T_MASK ^ 0x173848a9)
 115 #define T3 0x242070db
 116 #define T4 /* 0xc1bdceee */ (T_MASK ^ 0x3e423111)
 117 #define T5 /* 0xf57c0faf */ (T_MASK ^ 0x0a83f050)
 118 #define T6 0x4787c62a
 119 #define T7 /* 0xa8304613 */ (T_MASK ^ 0x57cfb9ec)
 120 #define T8 /* 0xfd469501 */ (T_MASK ^ 0x02b96afe)
 121 #define T9 0x698098d8
 122 #define T10 /* 0x8b44f7af */ (T_MASK ^ 0x74bb0850)
 123 #define T11 /* 0xffff5bb1 */ (T_MASK ^ 0x0000a44e)
 124 #define T12 /* 0x895cd7be */ (T_MASK ^ 0x76a32841)
 125 #define T13 0x6b901122
 126 #define T14 /* 0xfd987193 */ (T_MASK ^ 0x02678e6c)
 127 #define T15 /* 0xa679438e */ (T_MASK ^ 0x5986bc71)
 128 #define T16 0x49b40821
 129 #define T17 /* 0xf61e2562 */ (T_MASK ^ 0x09e1da9d)
 130 #define T18 /* 0xc040b340 */ (T_MASK ^ 0x3fbf4cbf)
 131 #define T19 0x265e5a51
 132 #define T20 /* 0xe9b6c7aa */ (T_MASK ^ 0x16493855)
 133 #define T21 /* 0xd62f105d */ (T_MASK ^ 0x29d0efa2)
 134 #define T22 0x02441453
 135 #define T23 /* 0xd8a1e681 */ (T_MASK ^ 0x275e197e)
 136 #define T24 /* 0xe7d3fbc8 */ (T_MASK ^ 0x182c0437)
 137 #define T25 0x21e1cde6
 138 #define T26 /* 0xc33707d6 */ (T_MASK ^ 0x3cc8f829)
 139 #define T27 /* 0xf4d50d87 */ (T_MASK ^ 0x0b2af278)
 140 #define T28 0x455a14ed
 141 #define T29 /* 0xa9e3e905 */ (T_MASK ^ 0x561c16fa)
 142 #define T30 /* 0xfcefa3f8 */ (T_MASK ^ 0x03105c07)
 143 #define T31 0x676f02d9
 144 #define T32 /* 0x8d2a4c8a */ (T_MASK ^ 0x72d5b375)
 145 #define T33 /* 0xfffa3942 */ (T_MASK ^ 0x0005c6bd)
 146 #define T34 /* 0x8771f681 */ (T_MASK ^ 0x788e097e)
 147 #define T35 0x6d9d6122
 148 #define T36 /* 0xfde5380c */ (T_MASK ^ 0x021ac7f3)
 149 #define T37 /* 0xa4beea44 */ (T_MASK ^ 0x5b4115bb)
 150 #define T38 0x4bdecfa9
 151 #define T39 /* 0xf6bb4b60 */ (T_MASK ^ 0x0944b49f)
 152 #define T40 /* 0xbebfbc70 */ (T_MASK ^ 0x4140438f)
 153 #define T41 0x289b7ec6
 154 #define T42 /* 0xeaa127fa */ (T_MASK ^ 0x155ed805)
 155 #define T43 /* 0xd4ef3085 */ (T_MASK ^ 0x2b10cf7a)
 156 #define T44 0x04881d05
 157 #define T45 /* 0xd9d4d039 */ (T_MASK ^ 0x262b2fc6)
 158 #define T46 /* 0xe6db99e5 */ (T_MASK ^ 0x1924661a)
 159 #define T47 0x1fa27cf8
 160 #define T48 /* 0xc4ac5665 */ (T_MASK ^ 0x3b53a99a)
 161 #define T49 /* 0xf4292244 */ (T_MASK ^ 0x0bd6ddbb)
 162 #define T50 0x432aff97
 163 #define T51 /* 0xab9423a7 */ (T_MASK ^ 0x546bdc58)
 164 #define T52 /* 0xfc93a039 */ (T_MASK ^ 0x036c5fc6)
 165 #define T53 0x655b59c3
 166 #define T54 /* 0x8f0ccc92 */ (T_MASK ^ 0x70f3336d)
 167 #define T55 /* 0xffeff47d */ (T_MASK ^ 0x00100b82)
 168 #define T56 /* 0x85845dd1 */ (T_MASK ^ 0x7a7ba22e)
 169 #define T57 0x6fa87e4f
 170 #define T58 /* 0xfe2ce6e0 */ (T_MASK ^ 0x01d3191f)
 171 #define T59 /* 0xa3014314 */ (T_MASK ^ 0x5cfebceb)
 172 #define T60 0x4e0811a1
 173 #define T61 /* 0xf7537e82 */ (T_MASK ^ 0x08ac817d)
 174 #define T62 /* 0xbd3af235 */ (T_MASK ^ 0x42c50dca)
 175 #define T63 0x2ad7d2bb
 176 #define T64 /* 0xeb86d391 */ (T_MASK ^ 0x14792c6e)
 177
 178
 179 static void md5_process(md5_state_t* pms, const md5_byte_t* data /*[64]*/)
 180 {
 181     md5_word_t a = pms->abcd[0], b = pms->abcd[1], c = pms->abcd[2], d = pms->abcd[3];
 182     md5_word_t t;
 183 #if BYTE_ORDER > 0
 184     /* Define storage only for big-endian CPUs. */
 185     md5_word_t X[16];
 186 #else
 187     /* Define storage for little-endian or both types of CPUs. */
 188     md5_word_t        xbuf[16];
 189     const md5_word_t* X;
 190 #endif
 191
 192     {
 193 #if BYTE_ORDER == 0
 194         /*
 195          * Determine dynamically whether this is a big-endian or
 196          * little-endian machine, since we can use a more efficient
 197          * algorithm on the latter.
 198          */
 199         static const int w = 1;
 200
 201         if (*(reinterpret_cast<const md5_byte_t*>(&w))) /* dynamic little-endian */
 202 #endif
 203 #if BYTE_ORDER <= 0 /* little-endian */
 204         {
 205             /*
 206              * On little-endian machines, we can process properly aligned
 207              * data without copying it.
 208              */
 209             if (!((data - reinterpret_cast<const md5_byte_t*>(0)) & 3))
 210             {
 211                 /* data are properly aligned */
 212                 X = reinterpret_cast<const md5_word_t*>(data);
 213             }
 214             else
 215             {
 216                 /* not aligned */
 217                 std::memcpy(xbuf, data, 64);
 218                 X = xbuf;
 219             }
 220         }
 221 #endif
 222 #if BYTE_ORDER == 0
 223         else /* dynamic big-endian */
 224 #endif
 225 #if BYTE_ORDER >= 0 /* big-endian */
 226         {
 227             /*
 228              * On big-endian machines, we must arrange the bytes in the
 229              * right order.
 230              */
 231             const md5_byte_t* xp = data;
 232             int               i;
 233
 234 #    if BYTE_ORDER == 0
 235             X = xbuf; /* (dynamic only) */
 236 #    else
 237 #        define xbuf X /* (static only) */
 238 #    endif
 239             for (i = 0; i < 16; ++i, xp += 4)
 240             {
 241                 xbuf[i] = xp[0] + (xp[1] << 8) + (xp[2] << 16) + (xp[3] << 24);
 242             }
 243         }
 244 #endif
 245     }
 246
 247 #define ROTATE_LEFT(x, n) (((x) << (n)) | ((x) >> (32 - (n))))
 248
 249     /* Round 1. */
 250     /* Let [abcd k s i] denote the operation
 251        a = b + ((a + F(b,c,d) + X[k] + T[i]) <<< s). */
 252 #define F(x, y, z) (((x) & (y)) | (~(x) & (z)))
 253 #define SET(a, b, c, d, k, s, Ti)         \
 254     t   = (a) + F(b, c, d) + X[k] + (Ti); \
 255     (a) = ROTATE_LEFT(t, s) + b
 256     /* Do the following 16 operations. */
 257     SET(a, b, c, d, 0, 7, T1);
 258     SET(d, a, b, c, 1, 12, T2);
 259     SET(c, d, a, b, 2, 17, T3);
 260     SET(b, c, d, a, 3, 22, T4);
 261     SET(a, b, c, d, 4, 7, T5);
 262     SET(d, a, b, c, 5, 12, T6);
 263     SET(c, d, a, b, 6, 17, T7);
 264     SET(b, c, d, a, 7, 22, T8);
 265     SET(a, b, c, d, 8, 7, T9);
 266     SET(d, a, b, c, 9, 12, T10);
 267     SET(c, d, a, b, 10, 17, T11);
 268     SET(b, c, d, a, 11, 22, T12);
 269     SET(a, b, c, d, 12, 7, T13);
 270     SET(d, a, b, c, 13, 12, T14);
 271     SET(c, d, a, b, 14, 17, T15);
 272     SET(b, c, d, a, 15, 22, T16);
 273 #undef SET
 274
 275     /* Round 2. */
 276     /* Let [abcd k s i] denote the operation
 277          a = b + ((a + G(b,c,d) + X[k] + T[i]) <<< s). */
 278 #define G(x, y, z) (((x) & (z)) | ((y) & ~(z)))
 279 #define SET(a, b, c, d, k, s, Ti)         \
 280     t   = (a) + G(b, c, d) + X[k] + (Ti); \
 281     (a) = ROTATE_LEFT(t, s) + b
 282     /* Do the following 16 operations. */
 283     SET(a, b, c, d, 1, 5, T17);
 284     SET(d, a, b, c, 6, 9, T18);
 285     SET(c, d, a, b, 11, 14, T19);
 286     SET(b, c, d, a, 0, 20, T20);
 287     SET(a, b, c, d, 5, 5, T21);
 288     SET(d, a, b, c, 10, 9, T22);
 289     SET(c, d, a, b, 15, 14, T23);
 290     SET(b, c, d, a, 4, 20, T24);
 291     SET(a, b, c, d, 9, 5, T25);
 292     SET(d, a, b, c, 14, 9, T26);
 293     SET(c, d, a, b, 3, 14, T27);
 294     SET(b, c, d, a, 8, 20, T28);
 295     SET(a, b, c, d, 13, 5, T29);
 296     SET(d, a, b, c, 2, 9, T30);
 297     SET(c, d, a, b, 7, 14, T31);
 298     SET(b, c, d, a, 12, 20, T32);
 299 #undef SET
 300
 301     /* Round 3. */
 302     /* Let [abcd k s t] denote the operation
 303          a = b + ((a + H(b,c,d) + X[k] + T[i]) <<< s). */
 304 #define H(x, y, z) ((x) ^ (y) ^ (z))
 305 #define SET(a, b, c, d, k, s, Ti)         \
 306     t   = (a) + H(b, c, d) + X[k] + (Ti); \
 307     (a) = ROTATE_LEFT(t, s) + b
 308     /* Do the following 16 operations. */
 309     SET(a, b, c, d, 5, 4, T33);
 310     SET(d, a, b, c, 8, 11, T34);
 311     SET(c, d, a, b, 11, 16, T35);
 312     SET(b, c, d, a, 14, 23, T36);
 313     SET(a, b, c, d, 1, 4, T37);
 314     SET(d, a, b, c, 4, 11, T38);
 315     SET(c, d, a, b, 7, 16, T39);
 316     SET(b, c, d, a, 10, 23, T40);
 317     SET(a, b, c, d, 13, 4, T41);
 318     SET(d, a, b, c, 0, 11, T42);
 319     SET(c, d, a, b, 3, 16, T43);
 320     SET(b, c, d, a, 6, 23, T44);
 321     SET(a, b, c, d, 9, 4, T45);
 322     SET(d, a, b, c, 12, 11, T46);
 323     SET(c, d, a, b, 15, 16, T47);
 324     SET(b, c, d, a, 2, 23, T48);
 325 #undef SET
 326
 327     /* Round 4. */
 328     /* Let [abcd k s t] denote the operation
 329          a = b + ((a + I(b,c,d) + X[k] + T[i]) <<< s). */
 330 #define I(x, y, z) ((y) ^ ((x) | ~(z)))
 331 #define SET(a, b, c, d, k, s, Ti)         \
 332     t   = (a) + I(b, c, d) + X[k] + (Ti); \
 333     (a) = ROTATE_LEFT(t, s) + b
 334     /* Do the following 16 operations. */
 335     SET(a, b, c, d, 0, 6, T49);
 336     SET(d, a, b, c, 7, 10, T50);
 337     SET(c, d, a, b, 14, 15, T51);
 338     SET(b, c, d, a, 5, 21, T52);
 339     SET(a, b, c, d, 12, 6, T53);
 340     SET(d, a, b, c, 3, 10, T54);
 341     SET(c, d, a, b, 10, 15, T55);
 342     SET(b, c, d, a, 1, 21, T56);
 343     SET(a, b, c, d, 8, 6, T57);
 344     SET(d, a, b, c, 15, 10, T58);
 345     SET(c, d, a, b, 6, 15, T59);
 346     SET(b, c, d, a, 13, 21, T60);
 347     SET(a, b, c, d, 4, 6, T61);
 348     SET(d, a, b, c, 11, 10, T62);
 349     SET(c, d, a, b, 2, 15, T63);
 350     SET(b, c, d, a, 9, 21, T64);
 351 #undef SET
 352
 353     /* Then perform the following additions. (That is increment each
 354        of the four registers by the value it had before this block
 355        was started.) */
 356     pms->abcd[0] += a;
 357     pms->abcd[1] += b;
 358     pms->abcd[2] += c;
 359     pms->abcd[3] += d;
 360 }
 361
 362 void gmx_md5_init(md5_state_t* pms)
 363 {
 364     pms->count[0] = pms->count[1] = 0;
 365     pms->abcd[0]                  = 0x67452301;
 366     pms->abcd[1]                  = /*0xefcdab89*/ T_MASK ^ 0x10325476;
 367     pms->abcd[2]                  = /*0x98badcfe*/ T_MASK ^ 0x67452301;
 368     pms->abcd[3]                  = 0x10325476;
 369 }
 370
 371 void gmx_md5_append(md5_state_t* pms, const md5_byte_t* data, int nbytes)
 372 {
 373     const md5_byte_t* p      = data;
 374     int               left   = nbytes;
 375     int               offset = (pms->count[0] >> 3) & 63;
 376     md5_word_t        nbits  = static_cast<md5_word_t>(nbytes << 3);
 377
 378     if (nbytes <= 0)
 379     {
 380         return;
 381     }
 382
 383     /* Update the message length. */
 384     pms->count[1] += nbytes >> 29;
 385     pms->count[0] += nbits;
 386     if (pms->count[0] < nbits)
 387     {
 388         pms->count[1]++;
 389     }
 390
 391     /* Process an initial partial block. */
 392     if (offset)
 393     {
 394         int copy = (offset + nbytes > 64 ? 64 - offset : nbytes);
 395
 396         std::memcpy(pms->buf + offset, p, copy);
 397         if (offset + copy < 64)
 398         {
 399             return;
 400         }
 401         p += copy;
 402         left -= copy;
 403         md5_process(pms, pms->buf);
 404     }
 405
 406     /* Process full blocks. */
 407     for (; left >= 64; p += 64, left -= 64)
 408     {
 409         md5_process(pms, p);
 410     }
 411
 412     /* Process a final partial block. */
 413     if (left)
 414     {
 415         memcpy(pms->buf, p, left);
 416     }
 417 }
 418
 419 std::array<unsigned char, 16> gmx_md5_finish(md5_state_t* pms)
 420 {
 421     static const md5_byte_t pad[64] = { 0x80, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
 422                                         0,    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
 423                                         0,    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
 424                                         0,    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 };
 425     md5_byte_t              data[8];
 426     int                     i;
 427
 428     /* Save the length before padding. */
 429     for (i = 0; i < 8; ++i)
 430     {
 431         data[i] = static_cast<md5_byte_t>(pms->count[i >> 2] >> ((i & 3) << 3));
 432     }
 433     /* Pad to 56 bytes mod 64. */
 434     gmx_md5_append(pms, pad, ((55 - (pms->count[0] >> 3)) & 63) + 1);
 435     /* Append the length. */
 436     gmx_md5_append(pms, data, 8);
 437     std::array<unsigned char, 16> digest;
 438     for (size_t i = 0; i < digest.size(); ++i)
 439     {
 440         digest[i] = static_cast<unsigned char>(pms->abcd[i >> 2] >> ((i & 3) << 3));
 441     }
 442     return digest;
 443 }