| 1 |
/* |
| 2 |
* The GOST 28147-89 cipher |
| 3 |
* |
| 4 |
* This is based on the 25 Movember 1993 draft translation |
| 5 |
* by Aleksandr Malchik, with Whitfield Diffie, of the Government |
| 6 |
* Standard of the U.S.S.R. GOST 28149-89, "Cryptographic Transformation |
| 7 |
* Algorithm", effective 1 July 1990. (Whitfield.Diffie@eng.sun.com) |
| 8 |
* |
| 9 |
* That is a draft, and may contain errors, which will be faithfully |
| 10 |
* reflected here, along with possible exciting new bugs. |
| 11 |
* |
| 12 |
* Some details have been cleared up by the paper "Soviet Encryption |
| 13 |
* Algorithm" by Josef Pieprzyk and Leonid Tombak of the University |
| 14 |
* of Wollongong, New South Wales. (josef/leo@cs.adfa.oz.au) |
| 15 |
* |
| 16 |
* The standard is written by A. Zabotin (project leader), G.P. Glazkov, |
| 17 |
* and V.B. Isaeva. It was accepted and introduced into use by the |
| 18 |
* action of the State Standards Committee of the USSR on 2 June 89 as |
| 19 |
* No. 1409. It was to be reviewed in 1993, but whether anyone wishes |
| 20 |
* to take on this obligation from the USSR is questionable. |
| 21 |
* |
| 22 |
* This code is placed in the public domain. |
| 23 |
*/ |
| 24 |
|
| 25 |
/* modified in order to use the libmcrypt API by Nikos Mavroyanopoulos |
| 26 |
* All modifications are placed under the license of libmcrypt. |
| 27 |
*/ |
| 28 |
|
| 29 |
/* $Id: gost.c,v 1.6 2000/02/02 20:15:50 nikos Exp $ */ |
| 30 |
|
| 31 |
#include "libdefs.h" |
| 32 |
#include "swap.h" |
| 33 |
/* |
| 34 |
* If you read the standard, it belabors the point of copying corresponding |
| 35 |
* bits from point A to point B quite a bit. It helps to understand that |
| 36 |
* the standard is uniformly little-endian, although it numbers bits from |
| 37 |
* 1 rather than 0, so bit n has value 2^(n-1). The least significant bit |
| 38 |
* of the 32-bit words that are manipulated in the algorithm is the first, |
| 39 |
* lowest-numbered, in the bit string. |
| 40 |
*/ |
| 41 |
|
| 42 |
void _mcrypt_kboxinit(void); |
| 43 |
|
| 44 |
/* |
| 45 |
* The standard does not specify the contents of the 8 4 bit->4 bit |
| 46 |
* substitution boxes, saying they're a parameter of the network |
| 47 |
* being set up. For illustration purposes here, I have used |
| 48 |
* the first rows of the 8 S-boxes from the DES. (Note that the |
| 49 |
* DES S-boxes are numbered starting from 1 at the msb. In keeping |
| 50 |
* with the rest of the GOST, I have used little-endian numbering. |
| 51 |
* Thus, k8 is S-box 1. |
| 52 |
* |
| 53 |
* Obviously, a careful look at the cryptographic properties of the cipher |
| 54 |
* must be undertaken before "production" substitution boxes are defined. |
| 55 |
* |
| 56 |
* The standard also does not specify a standard bit-string representation |
| 57 |
* for the contents of these blocks. |
| 58 |
*/ |
| 59 |
|
| 60 |
/* These are NOT the original s-boxes. I replaced them with the ones |
| 61 |
* found in Applied Cryptography book by Bruce Schneier. These were |
| 62 |
* used in an application for the Central Bank of the Russian Federation |
| 63 |
* --Nikos |
| 64 |
*/ |
| 65 |
static int init=0; |
| 66 |
|
| 67 |
static unsigned char const k1[16] = { |
| 68 |
1, 15, 13, 0, 5, 7, 10, 4, 9, 2, 3, 14, 6, 11, 8, 2 |
| 69 |
}; |
| 70 |
static unsigned char const k2[16] = { |
| 71 |
13, 11, 4, 1, 3, 15, 5, 9, 0, 10, 14, 7, 6, 8, 2, 12 |
| 72 |
}; |
| 73 |
static unsigned char const k3[16] = { |
| 74 |
4, 11, 10, 0, 7, 2, 1, 13, 3, 6, 8, 5, 9, 12, 15, 14 |
| 75 |
}; |
| 76 |
static unsigned char const k4[16] = { |
| 77 |
6, 12, 7, 1, 5, 15, 13, 8, 4, 10, 9, 14, 0, 3, 11, 2 |
| 78 |
}; |
| 79 |
static unsigned char const k5[16] = { |
| 80 |
7, 13, 10, 1, 0, 8, 9, 15, 14, 4, 6, 12, 11, 2, 5, 3 |
| 81 |
}; |
| 82 |
static unsigned char const k6[16] = { |
| 83 |
5, 8, 1, 13, 10, 3, 4, 2, 14, 15, 12, 7, 6, 0, 9, 11 |
| 84 |
}; |
| 85 |
static unsigned char const k7[16] = { |
| 86 |
14, 11, 4, 12, 6, 13, 15, 10, 2, 3, 8, 1, 0, 7, 5, 9 |
| 87 |
}; |
| 88 |
static unsigned char const k8[16] = { |
| 89 |
4, 10, 9, 2, 13, 8, 0, 14, 6, 11, 1, 12, 7, 15, 5, 3 |
| 90 |
}; |
| 91 |
|
| 92 |
/* Byte-at-a-time substitution boxes */ |
| 93 |
static unsigned char k87[256]; |
| 94 |
static unsigned char k65[256]; |
| 95 |
static unsigned char k43[256]; |
| 96 |
static unsigned char k21[256]; |
| 97 |
|
| 98 |
/* |
| 99 |
* Build byte-at-a-time subtitution tables. |
| 100 |
* This must be called once for global setup. |
| 101 |
*/ |
| 102 |
|
| 103 |
void _mcrypt_gost_set_key (word32 *inst, word32* key, int len) { |
| 104 |
|
| 105 |
_mcrypt_kboxinit(); |
| 106 |
inst[0] = 0; inst[1] = 0; |
| 107 |
inst[2] = 0; inst[3] = 0; |
| 108 |
inst[4] = 0; inst[5] = 0; |
| 109 |
inst[6] = 0; inst[7] = 0; |
| 110 |
memmove( inst, key, len); |
| 111 |
#ifdef WORDS_BIGENDIAN |
| 112 |
inst[0] = byteswap(inst[0]); |
| 113 |
inst[1] = byteswap(inst[1]); |
| 114 |
inst[2] = byteswap(inst[2]); |
| 115 |
inst[3] = byteswap(inst[3]); |
| 116 |
inst[4] = byteswap(inst[4]); |
| 117 |
inst[5] = byteswap(inst[5]); |
| 118 |
inst[6] = byteswap(inst[6]); |
| 119 |
inst[7] = byteswap(inst[7]); |
| 120 |
#endif |
| 121 |
} |
| 122 |
|
| 123 |
void _mcrypt_kboxinit(void) |
| 124 |
{ |
| 125 |
int i; |
| 126 |
|
| 127 |
if (init==0) { |
| 128 |
init=1; |
| 129 |
for (i = 0; i < 256; i++) { |
| 130 |
k87[i] = k8[i >> 4] << 4 | k7[i & 15]; |
| 131 |
k65[i] = k6[i >> 4] << 4 | k5[i & 15]; |
| 132 |
k43[i] = k4[i >> 4] << 4 | k3[i & 15]; |
| 133 |
k21[i] = k2[i >> 4] << 4 | k1[i & 15]; |
| 134 |
} |
| 135 |
} |
| 136 |
|
| 137 |
} |
| 138 |
|
| 139 |
/* |
| 140 |
* Do the substitution and rotation that are the core of the operation, |
| 141 |
* like the expansion, substitution and permutation of the DES. |
| 142 |
* It would be possible to perform DES-like optimisations and store |
| 143 |
* the table entries as 32-bit words, already rotated, but the |
| 144 |
* efficiency gain is questionable. |
| 145 |
* |
| 146 |
* This should be inlined for maximum speed |
| 147 |
*/ |
| 148 |
static word32 f(word32 x) |
| 149 |
{ |
| 150 |
/* Do substitutions */ |
| 151 |
# if 0 |
| 152 |
/* This is annoyingly slow */ |
| 153 |
x = k8[x >> 28 & 15] << 28 | k7[x >> 24 & 15] << 24 | |
| 154 |
k6[x >> 20 & 15] << 20 | k5[x >> 16 & 15] << 16 | |
| 155 |
k4[x >> 12 & 15] << 12 | k3[x >> 8 & 15] << 8 | |
| 156 |
k2[x >> 4 & 15] << 4 | k1[x & 15]; |
| 157 |
# else |
| 158 |
/* This is faster */ |
| 159 |
x = k87[x >> 24 & 255] << 24 | k65[x >> 16 & 255] << 16 | |
| 160 |
k43[x >> 8 & 255] << 8 | k21[x & 255]; |
| 161 |
# endif |
| 162 |
|
| 163 |
/* Rotate left 11 bits */ |
| 164 |
return x << 11 | x >> (32 - 11); |
| 165 |
|
| 166 |
} |
| 167 |
|
| 168 |
/* |
| 169 |
* The GOST standard defines the input in terms of bits 1..64, with |
| 170 |
* bit 1 being the lsb of in[0] and bit 64 being the msb of in[1]. |
| 171 |
* |
| 172 |
* The keys are defined similarly, with bit 256 being the msb of key[7]. |
| 173 |
*/ |
| 174 |
void _mcrypt_gost_encrypt(word32 const key[8], word32 * in) |
| 175 |
{ |
| 176 |
register word32 n1, n2; /* As named in the GOST */ |
| 177 |
|
| 178 |
/* Added to make it compatible with bigendian machines |
| 179 |
* --nikos |
| 180 |
*/ |
| 181 |
|
| 182 |
#ifndef WORDS_BIGENDIAN |
| 183 |
n1 = byteswap(in[0]); |
| 184 |
n2 = byteswap(in[1]); |
| 185 |
#else |
| 186 |
n1 = in[0]; |
| 187 |
n2 = in[1]; |
| 188 |
#endif |
| 189 |
|
| 190 |
/* Instead of swapping halves, swap names each round */ |
| 191 |
n2 ^= f(n1 + key[0]); |
| 192 |
n1 ^= f(n2 + key[1]); |
| 193 |
n2 ^= f(n1 + key[2]); |
| 194 |
n1 ^= f(n2 + key[3]); |
| 195 |
n2 ^= f(n1 + key[4]); |
| 196 |
n1 ^= f(n2 + key[5]); |
| 197 |
n2 ^= f(n1 + key[6]); |
| 198 |
n1 ^= f(n2 + key[7]); |
| 199 |
|
| 200 |
n2 ^= f(n1 + key[0]); |
| 201 |
n1 ^= f(n2 + key[1]); |
| 202 |
n2 ^= f(n1 + key[2]); |
| 203 |
n1 ^= f(n2 + key[3]); |
| 204 |
n2 ^= f(n1 + key[4]); |
| 205 |
n1 ^= f(n2 + key[5]); |
| 206 |
n2 ^= f(n1 + key[6]); |
| 207 |
n1 ^= f(n2 + key[7]); |
| 208 |
|
| 209 |
n2 ^= f(n1 + key[0]); |
| 210 |
n1 ^= f(n2 + key[1]); |
| 211 |
n2 ^= f(n1 + key[2]); |
| 212 |
n1 ^= f(n2 + key[3]); |
| 213 |
n2 ^= f(n1 + key[4]); |
| 214 |
n1 ^= f(n2 + key[5]); |
| 215 |
n2 ^= f(n1 + key[6]); |
| 216 |
n1 ^= f(n2 + key[7]); |
| 217 |
|
| 218 |
n2 ^= f(n1 + key[7]); |
| 219 |
n1 ^= f(n2 + key[6]); |
| 220 |
n2 ^= f(n1 + key[5]); |
| 221 |
n1 ^= f(n2 + key[4]); |
| 222 |
n2 ^= f(n1 + key[3]); |
| 223 |
n1 ^= f(n2 + key[2]); |
| 224 |
n2 ^= f(n1 + key[1]); |
| 225 |
n1 ^= f(n2 + key[0]); |
| 226 |
|
| 227 |
/* There is no swap after the last round */ |
| 228 |
|
| 229 |
#ifndef WORDS_BIGENDIAN |
| 230 |
in[0] = byteswap(n2); |
| 231 |
in[1] = byteswap(n1); |
| 232 |
#else |
| 233 |
in[0] = n2; |
| 234 |
in[1] = n1; |
| 235 |
#endif |
| 236 |
} |
| 237 |
|
| 238 |
|
| 239 |
/* |
| 240 |
* The key schedule is somewhat different for decryption. |
| 241 |
* (The key table is used once forward and three times backward.) |
| 242 |
* You could define an expanded key, or just write the code twice, |
| 243 |
* as done here. |
| 244 |
*/ |
| 245 |
void _mcrypt_gost_decrypt(word32 const key[8], word32 * in) |
| 246 |
{ |
| 247 |
register word32 n1, n2; /* As named in the GOST */ |
| 248 |
|
| 249 |
#ifndef WORDS_BIGENDIAN |
| 250 |
n1 = byteswap(in[0]); |
| 251 |
n2 = byteswap(in[1]); |
| 252 |
#else |
| 253 |
n1 = in[0]; |
| 254 |
n2 = in[1]; |
| 255 |
#endif |
| 256 |
|
| 257 |
n2 ^= f(n1 + key[0]); |
| 258 |
n1 ^= f(n2 + key[1]); |
| 259 |
n2 ^= f(n1 + key[2]); |
| 260 |
n1 ^= f(n2 + key[3]); |
| 261 |
n2 ^= f(n1 + key[4]); |
| 262 |
n1 ^= f(n2 + key[5]); |
| 263 |
n2 ^= f(n1 + key[6]); |
| 264 |
n1 ^= f(n2 + key[7]); |
| 265 |
|
| 266 |
n2 ^= f(n1 + key[7]); |
| 267 |
n1 ^= f(n2 + key[6]); |
| 268 |
n2 ^= f(n1 + key[5]); |
| 269 |
n1 ^= f(n2 + key[4]); |
| 270 |
n2 ^= f(n1 + key[3]); |
| 271 |
n1 ^= f(n2 + key[2]); |
| 272 |
n2 ^= f(n1 + key[1]); |
| 273 |
n1 ^= f(n2 + key[0]); |
| 274 |
|
| 275 |
n2 ^= f(n1 + key[7]); |
| 276 |
n1 ^= f(n2 + key[6]); |
| 277 |
n2 ^= f(n1 + key[5]); |
| 278 |
n1 ^= f(n2 + key[4]); |
| 279 |
n2 ^= f(n1 + key[3]); |
| 280 |
n1 ^= f(n2 + key[2]); |
| 281 |
n2 ^= f(n1 + key[1]); |
| 282 |
n1 ^= f(n2 + key[0]); |
| 283 |
|
| 284 |
n2 ^= f(n1 + key[7]); |
| 285 |
n1 ^= f(n2 + key[6]); |
| 286 |
n2 ^= f(n1 + key[5]); |
| 287 |
n1 ^= f(n2 + key[4]); |
| 288 |
n2 ^= f(n1 + key[3]); |
| 289 |
n1 ^= f(n2 + key[2]); |
| 290 |
n2 ^= f(n1 + key[1]); |
| 291 |
n1 ^= f(n2 + key[0]); |
| 292 |
|
| 293 |
#ifndef WORDS_BIGENDIAN |
| 294 |
in[0] = byteswap(n2); |
| 295 |
in[1] = byteswap(n1); |
| 296 |
#else |
| 297 |
in[0] = n2; |
| 298 |
in[1] = n1; |
| 299 |
#endif |
| 300 |
|
| 301 |
} |
Parent Directory
| 
Revision Log