| 1 |
/* Copyright (C) 1995-1998 Eric Young ([email protected]) |
| 2 |
* All rights reserved. |
| 3 |
* |
| 4 |
* This package is an SSL implementation written |
| 5 |
* by Eric Young ([email protected]). |
| 6 |
* The implementation was written so as to conform with Netscapes SSL. |
| 7 |
* |
| 8 |
* This library is free for commercial and non-commercial use as long as |
| 9 |
* the following conditions are aheared to. The following conditions |
| 10 |
* apply to all code found in this distribution, be it the RC4, RSA, |
| 11 |
* lhash, DES, etc., code; not just the SSL code. The SSL documentation |
| 12 |
* included with this distribution is covered by the same copyright terms |
| 13 |
* except that the holder is Tim Hudson ([email protected]). |
| 14 |
* |
| 15 |
* Copyright remains Eric Young's, and as such any Copyright notices in |
| 16 |
* the code are not to be removed. |
| 17 |
* If this package is used in a product, Eric Young should be given attribution |
| 18 |
* as the author of the parts of the library used. |
| 19 |
* This can be in the form of a textual message at program startup or |
| 20 |
* in documentation (online or textual) provided with the package. |
| 21 |
* |
| 22 |
* Redistribution and use in source and binary forms, with or without |
| 23 |
* modification, are permitted provided that the following conditions |
| 24 |
* are met: |
| 25 |
* 1. Redistributions of source code must retain the copyright |
| 26 |
* notice, this list of conditions and the following disclaimer. |
| 27 |
* 2. Redistributions in binary form must reproduce the above copyright |
| 28 |
* notice, this list of conditions and the following disclaimer in the |
| 29 |
* documentation and/or other materials provided with the distribution. |
| 30 |
* 3. All advertising materials mentioning features or use of this software |
| 31 |
* must display the following acknowledgement: |
| 32 |
* "This product includes cryptographic software written by |
| 33 |
* Eric Young ([email protected])" |
| 34 |
* The word 'cryptographic' can be left out if the rouines from the library |
| 35 |
* being used are not cryptographic related :-). |
| 36 |
* 4. If you include any Windows specific code (or a derivative thereof) from |
| 37 |
* the apps directory (application code) you must include an acknowledgement: |
| 38 |
* "This product includes software written by Tim Hudson ([email protected])" |
| 39 |
* |
| 40 |
* THIS SOFTWARE IS PROVIDED BY ERIC YOUNG ``AS IS'' AND |
| 41 |
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE |
| 42 |
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE |
| 43 |
* ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE |
| 44 |
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL |
| 45 |
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS |
| 46 |
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) |
| 47 |
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT |
| 48 |
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY |
| 49 |
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF |
| 50 |
* SUCH DAMAGE. |
| 51 |
* |
| 52 |
* The licence and distribution terms for any publically available version or |
| 53 |
* derivative of this code cannot be changed. i.e. this code cannot simply be |
| 54 |
* copied and put under another distribution licence |
| 55 |
* [including the GNU Public Licence.] */ |
| 56 |
|
| 57 |
#include <openssl/bn.h> |
| 58 |
|
| 59 |
#include <assert.h> |
| 60 |
|
| 61 |
#include "internal.h" |
| 62 |
|
| 63 |
|
| 64 |
/* This file has two other implementations: x86 assembly language in |
| 65 |
* asm/bn-586.pl and x86_64 inline assembly in asm/x86_64-gcc.c. */ |
| 66 |
#if defined(OPENSSL_NO_ASM) || \ |
| 67 |
!(defined(OPENSSL_X86) || (defined(OPENSSL_X86_64) && defined(__GNUC__))) |
| 68 |
|
| 69 |
#ifdef BN_ULLONG |
| 70 |
#define mul_add(r, a, w, c) \ |
| 71 |
{ \ |
| 72 |
BN_ULLONG t; \ |
| 73 |
t = (BN_ULLONG)w * (a) + (r) + (c); \ |
| 74 |
(r) = Lw(t); \ |
| 75 |
(c) = Hw(t); \ |
| 76 |
} |
| 77 |
|
| 78 |
#define mul(r, a, w, c) \ |
| 79 |
{ \ |
| 80 |
BN_ULLONG t; \ |
| 81 |
t = (BN_ULLONG)w * (a) + (c); \ |
| 82 |
(r) = Lw(t); \ |
| 83 |
(c) = Hw(t); \ |
| 84 |
} |
| 85 |
|
| 86 |
#define sqr(r0, r1, a) \ |
| 87 |
{ \ |
| 88 |
BN_ULLONG t; \ |
| 89 |
t = (BN_ULLONG)(a) * (a); \ |
| 90 |
(r0) = Lw(t); \ |
| 91 |
(r1) = Hw(t); \ |
| 92 |
} |
| 93 |
|
| 94 |
#else |
| 95 |
|
| 96 |
#define mul_add(r, a, w, c) \ |
| 97 |
{ \ |
| 98 |
BN_ULONG high, low, ret, tmp = (a); \ |
| 99 |
ret = (r); \ |
| 100 |
BN_UMULT_LOHI(low, high, w, tmp); \ |
| 101 |
ret += (c); \ |
| 102 |
(c) = (ret < (c)) ? 1 : 0; \ |
| 103 |
(c) += high; \ |
| 104 |
ret += low; \ |
| 105 |
(c) += (ret < low) ? 1 : 0; \ |
| 106 |
(r) = ret; \ |
| 107 |
} |
| 108 |
|
| 109 |
#define mul(r, a, w, c) \ |
| 110 |
{ \ |
| 111 |
BN_ULONG high, low, ret, ta = (a); \ |
| 112 |
BN_UMULT_LOHI(low, high, w, ta); \ |
| 113 |
ret = low + (c); \ |
| 114 |
(c) = high; \ |
| 115 |
(c) += (ret < low) ? 1 : 0; \ |
| 116 |
(r) = ret; \ |
| 117 |
} |
| 118 |
|
| 119 |
#define sqr(r0, r1, a) \ |
| 120 |
{ \ |
| 121 |
BN_ULONG tmp = (a); \ |
| 122 |
BN_UMULT_LOHI(r0, r1, tmp, tmp); \ |
| 123 |
} |
| 124 |
|
| 125 |
#endif /* !BN_ULLONG */ |
| 126 |
|
| 127 |
BN_ULONG bn_mul_add_words(BN_ULONG *rp, const BN_ULONG *ap, int num, |
| 128 |
BN_ULONG w) { |
| 129 |
BN_ULONG c1 = 0; |
| 130 |
|
| 131 |
assert(num >= 0); |
| 132 |
if (num <= 0) { |
| 133 |
return c1; |
| 134 |
} |
| 135 |
|
| 136 |
while (num & ~3) { |
| 137 |
mul_add(rp[0], ap[0], w, c1); |
| 138 |
mul_add(rp[1], ap[1], w, c1); |
| 139 |
mul_add(rp[2], ap[2], w, c1); |
| 140 |
mul_add(rp[3], ap[3], w, c1); |
| 141 |
ap += 4; |
| 142 |
rp += 4; |
| 143 |
num -= 4; |
| 144 |
} |
| 145 |
|
| 146 |
while (num) { |
| 147 |
mul_add(rp[0], ap[0], w, c1); |
| 148 |
ap++; |
| 149 |
rp++; |
| 150 |
num--; |
| 151 |
} |
| 152 |
|
| 153 |
return c1; |
| 154 |
} |
| 155 |
|
| 156 |
BN_ULONG bn_mul_words(BN_ULONG *rp, const BN_ULONG *ap, int num, BN_ULONG w) { |
| 157 |
BN_ULONG c1 = 0; |
| 158 |
|
| 159 |
assert(num >= 0); |
| 160 |
if (num <= 0) { |
| 161 |
return c1; |
| 162 |
} |
| 163 |
|
| 164 |
while (num & ~3) { |
| 165 |
mul(rp[0], ap[0], w, c1); |
| 166 |
mul(rp[1], ap[1], w, c1); |
| 167 |
mul(rp[2], ap[2], w, c1); |
| 168 |
mul(rp[3], ap[3], w, c1); |
| 169 |
ap += 4; |
| 170 |
rp += 4; |
| 171 |
num -= 4; |
| 172 |
} |
| 173 |
while (num) { |
| 174 |
mul(rp[0], ap[0], w, c1); |
| 175 |
ap++; |
| 176 |
rp++; |
| 177 |
num--; |
| 178 |
} |
| 179 |
return c1; |
| 180 |
} |
| 181 |
|
| 182 |
void bn_sqr_words(BN_ULONG *r, const BN_ULONG *a, int n) { |
| 183 |
assert(n >= 0); |
| 184 |
if (n <= 0) { |
| 185 |
return; |
| 186 |
} |
| 187 |
|
| 188 |
while (n & ~3) { |
| 189 |
sqr(r[0], r[1], a[0]); |
| 190 |
sqr(r[2], r[3], a[1]); |
| 191 |
sqr(r[4], r[5], a[2]); |
| 192 |
sqr(r[6], r[7], a[3]); |
| 193 |
a += 4; |
| 194 |
r += 8; |
| 195 |
n -= 4; |
| 196 |
} |
| 197 |
while (n) { |
| 198 |
sqr(r[0], r[1], a[0]); |
| 199 |
a++; |
| 200 |
r += 2; |
| 201 |
n--; |
| 202 |
} |
| 203 |
} |
| 204 |
|
| 205 |
#ifdef BN_ULLONG |
| 206 |
BN_ULONG bn_add_words(BN_ULONG *r, const BN_ULONG *a, const BN_ULONG *b, |
| 207 |
int n) { |
| 208 |
BN_ULLONG ll = 0; |
| 209 |
|
| 210 |
assert(n >= 0); |
| 211 |
if (n <= 0) { |
| 212 |
return (BN_ULONG)0; |
| 213 |
} |
| 214 |
|
| 215 |
while (n & ~3) { |
| 216 |
ll += (BN_ULLONG)a[0] + b[0]; |
| 217 |
r[0] = (BN_ULONG)ll & BN_MASK2; |
| 218 |
ll >>= BN_BITS2; |
| 219 |
ll += (BN_ULLONG)a[1] + b[1]; |
| 220 |
r[1] = (BN_ULONG)ll & BN_MASK2; |
| 221 |
ll >>= BN_BITS2; |
| 222 |
ll += (BN_ULLONG)a[2] + b[2]; |
| 223 |
r[2] = (BN_ULONG)ll & BN_MASK2; |
| 224 |
ll >>= BN_BITS2; |
| 225 |
ll += (BN_ULLONG)a[3] + b[3]; |
| 226 |
r[3] = (BN_ULONG)ll & BN_MASK2; |
| 227 |
ll >>= BN_BITS2; |
| 228 |
a += 4; |
| 229 |
b += 4; |
| 230 |
r += 4; |
| 231 |
n -= 4; |
| 232 |
} |
| 233 |
while (n) { |
| 234 |
ll += (BN_ULLONG)a[0] + b[0]; |
| 235 |
r[0] = (BN_ULONG)ll & BN_MASK2; |
| 236 |
ll >>= BN_BITS2; |
| 237 |
a++; |
| 238 |
b++; |
| 239 |
r++; |
| 240 |
n--; |
| 241 |
} |
| 242 |
return (BN_ULONG)ll; |
| 243 |
} |
| 244 |
|
| 245 |
#else /* !BN_ULLONG */ |
| 246 |
|
| 247 |
BN_ULONG bn_add_words(BN_ULONG *r, const BN_ULONG *a, const BN_ULONG *b, |
| 248 |
int n) { |
| 249 |
BN_ULONG c, l, t; |
| 250 |
|
| 251 |
assert(n >= 0); |
| 252 |
if (n <= 0) { |
| 253 |
return (BN_ULONG)0; |
| 254 |
} |
| 255 |
|
| 256 |
c = 0; |
| 257 |
while (n & ~3) { |
| 258 |
t = a[0]; |
| 259 |
t = (t + c) & BN_MASK2; |
| 260 |
c = (t < c); |
| 261 |
l = (t + b[0]) & BN_MASK2; |
| 262 |
c += (l < t); |
| 263 |
r[0] = l; |
| 264 |
t = a[1]; |
| 265 |
t = (t + c) & BN_MASK2; |
| 266 |
c = (t < c); |
| 267 |
l = (t + b[1]) & BN_MASK2; |
| 268 |
c += (l < t); |
| 269 |
r[1] = l; |
| 270 |
t = a[2]; |
| 271 |
t = (t + c) & BN_MASK2; |
| 272 |
c = (t < c); |
| 273 |
l = (t + b[2]) & BN_MASK2; |
| 274 |
c += (l < t); |
| 275 |
r[2] = l; |
| 276 |
t = a[3]; |
| 277 |
t = (t + c) & BN_MASK2; |
| 278 |
c = (t < c); |
| 279 |
l = (t + b[3]) & BN_MASK2; |
| 280 |
c += (l < t); |
| 281 |
r[3] = l; |
| 282 |
a += 4; |
| 283 |
b += 4; |
| 284 |
r += 4; |
| 285 |
n -= 4; |
| 286 |
} |
| 287 |
while (n) { |
| 288 |
t = a[0]; |
| 289 |
t = (t + c) & BN_MASK2; |
| 290 |
c = (t < c); |
| 291 |
l = (t + b[0]) & BN_MASK2; |
| 292 |
c += (l < t); |
| 293 |
r[0] = l; |
| 294 |
a++; |
| 295 |
b++; |
| 296 |
r++; |
| 297 |
n--; |
| 298 |
} |
| 299 |
return (BN_ULONG)c; |
| 300 |
} |
| 301 |
|
| 302 |
#endif /* !BN_ULLONG */ |
| 303 |
|
| 304 |
BN_ULONG bn_sub_words(BN_ULONG *r, const BN_ULONG *a, const BN_ULONG *b, |
| 305 |
int n) { |
| 306 |
BN_ULONG t1, t2; |
| 307 |
int c = 0; |
| 308 |
|
| 309 |
assert(n >= 0); |
| 310 |
if (n <= 0) { |
| 311 |
return (BN_ULONG)0; |
| 312 |
} |
| 313 |
|
| 314 |
while (n & ~3) { |
| 315 |
t1 = a[0]; |
| 316 |
t2 = b[0]; |
| 317 |
r[0] = (t1 - t2 - c) & BN_MASK2; |
| 318 |
if (t1 != t2) { |
| 319 |
c = (t1 < t2); |
| 320 |
} |
| 321 |
t1 = a[1]; |
| 322 |
t2 = b[1]; |
| 323 |
r[1] = (t1 - t2 - c) & BN_MASK2; |
| 324 |
if (t1 != t2) { |
| 325 |
c = (t1 < t2); |
| 326 |
} |
| 327 |
t1 = a[2]; |
| 328 |
t2 = b[2]; |
| 329 |
r[2] = (t1 - t2 - c) & BN_MASK2; |
| 330 |
if (t1 != t2) { |
| 331 |
c = (t1 < t2); |
| 332 |
} |
| 333 |
t1 = a[3]; |
| 334 |
t2 = b[3]; |
| 335 |
r[3] = (t1 - t2 - c) & BN_MASK2; |
| 336 |
if (t1 != t2) { |
| 337 |
c = (t1 < t2); |
| 338 |
} |
| 339 |
a += 4; |
| 340 |
b += 4; |
| 341 |
r += 4; |
| 342 |
n -= 4; |
| 343 |
} |
| 344 |
while (n) { |
| 345 |
t1 = a[0]; |
| 346 |
t2 = b[0]; |
| 347 |
r[0] = (t1 - t2 - c) & BN_MASK2; |
| 348 |
if (t1 != t2) { |
| 349 |
c = (t1 < t2); |
| 350 |
} |
| 351 |
a++; |
| 352 |
b++; |
| 353 |
r++; |
| 354 |
n--; |
| 355 |
} |
| 356 |
return c; |
| 357 |
} |
| 358 |
|
| 359 |
/* mul_add_c(a,b,c0,c1,c2) -- c+=a*b for three word number c=(c2,c1,c0) */ |
| 360 |
/* mul_add_c2(a,b,c0,c1,c2) -- c+=2*a*b for three word number c=(c2,c1,c0) */ |
| 361 |
/* sqr_add_c(a,i,c0,c1,c2) -- c+=a[i]^2 for three word number c=(c2,c1,c0) */ |
| 362 |
/* sqr_add_c2(a,i,c0,c1,c2) -- c+=2*a[i]*a[j] for three word number c=(c2,c1,c0) */ |
| 363 |
|
| 364 |
#ifdef BN_ULLONG |
| 365 |
|
| 366 |
/* Keep in mind that additions to multiplication result can not overflow, |
| 367 |
* because its high half cannot be all-ones. */ |
| 368 |
#define mul_add_c(a, b, c0, c1, c2) \ |
| 369 |
do { \ |
| 370 |
BN_ULONG hi; \ |
| 371 |
BN_ULLONG t = (BN_ULLONG)(a) * (b); \ |
| 372 |
t += c0; /* no carry */ \ |
| 373 |
c0 = (BN_ULONG)Lw(t); \ |
| 374 |
hi = (BN_ULONG)Hw(t); \ |
| 375 |
c1 = (c1 + hi) & BN_MASK2; \ |
| 376 |
if (c1 < hi) \ |
| 377 |
c2++; \ |
| 378 |
} while (0) |
| 379 |
|
| 380 |
#define mul_add_c2(a, b, c0, c1, c2) \ |
| 381 |
do { \ |
| 382 |
BN_ULONG hi; \ |
| 383 |
BN_ULLONG t = (BN_ULLONG)(a) * (b); \ |
| 384 |
BN_ULLONG tt = t + c0; /* no carry */ \ |
| 385 |
c0 = (BN_ULONG)Lw(tt); \ |
| 386 |
hi = (BN_ULONG)Hw(tt); \ |
| 387 |
c1 = (c1 + hi) & BN_MASK2; \ |
| 388 |
if (c1 < hi) \ |
| 389 |
c2++; \ |
| 390 |
t += c0; /* no carry */ \ |
| 391 |
c0 = (BN_ULONG)Lw(t); \ |
| 392 |
hi = (BN_ULONG)Hw(t); \ |
| 393 |
c1 = (c1 + hi) & BN_MASK2; \ |
| 394 |
if (c1 < hi) \ |
| 395 |
c2++; \ |
| 396 |
} while (0) |
| 397 |
|
| 398 |
#define sqr_add_c(a, i, c0, c1, c2) \ |
| 399 |
do { \ |
| 400 |
BN_ULONG hi; \ |
| 401 |
BN_ULLONG t = (BN_ULLONG)a[i] * a[i]; \ |
| 402 |
t += c0; /* no carry */ \ |
| 403 |
c0 = (BN_ULONG)Lw(t); \ |
| 404 |
hi = (BN_ULONG)Hw(t); \ |
| 405 |
c1 = (c1 + hi) & BN_MASK2; \ |
| 406 |
if (c1 < hi) \ |
| 407 |
c2++; \ |
| 408 |
} while (0) |
| 409 |
|
| 410 |
#define sqr_add_c2(a, i, j, c0, c1, c2) mul_add_c2((a)[i], (a)[j], c0, c1, c2) |
| 411 |
|
| 412 |
#else |
| 413 |
|
| 414 |
/* Keep in mind that additions to hi can not overflow, because the high word of |
| 415 |
* a multiplication result cannot be all-ones. */ |
| 416 |
#define mul_add_c(a, b, c0, c1, c2) \ |
| 417 |
do { \ |
| 418 |
BN_ULONG ta = (a), tb = (b); \ |
| 419 |
BN_ULONG lo, hi; \ |
| 420 |
BN_UMULT_LOHI(lo, hi, ta, tb); \ |
| 421 |
c0 += lo; \ |
| 422 |
hi += (c0 < lo) ? 1 : 0; \ |
| 423 |
c1 += hi; \ |
| 424 |
c2 += (c1 < hi) ? 1 : 0; \ |
| 425 |
} while (0) |
| 426 |
|
| 427 |
#define mul_add_c2(a, b, c0, c1, c2) \ |
| 428 |
do { \ |
| 429 |
BN_ULONG ta = (a), tb = (b); \ |
| 430 |
BN_ULONG lo, hi, tt; \ |
| 431 |
BN_UMULT_LOHI(lo, hi, ta, tb); \ |
| 432 |
c0 += lo; \ |
| 433 |
tt = hi + ((c0 < lo) ? 1 : 0); \ |
| 434 |
c1 += tt; \ |
| 435 |
c2 += (c1 < tt) ? 1 : 0; \ |
| 436 |
c0 += lo; \ |
| 437 |
hi += (c0 < lo) ? 1 : 0; \ |
| 438 |
c1 += hi; \ |
| 439 |
c2 += (c1 < hi) ? 1 : 0; \ |
| 440 |
} while (0) |
| 441 |
|
| 442 |
#define sqr_add_c(a, i, c0, c1, c2) \ |
| 443 |
do { \ |
| 444 |
BN_ULONG ta = (a)[i]; \ |
| 445 |
BN_ULONG lo, hi; \ |
| 446 |
BN_UMULT_LOHI(lo, hi, ta, ta); \ |
| 447 |
c0 += lo; \ |
| 448 |
hi += (c0 < lo) ? 1 : 0; \ |
| 449 |
c1 += hi; \ |
| 450 |
c2 += (c1 < hi) ? 1 : 0; \ |
| 451 |
} while (0) |
| 452 |
|
| 453 |
#define sqr_add_c2(a, i, j, c0, c1, c2) mul_add_c2((a)[i], (a)[j], c0, c1, c2) |
| 454 |
|
| 455 |
#endif /* !BN_ULLONG */ |
| 456 |
|
| 457 |
void bn_mul_comba8(BN_ULONG *r, BN_ULONG *a, BN_ULONG *b) { |
| 458 |
BN_ULONG c1, c2, c3; |
| 459 |
|
| 460 |
c1 = 0; |
| 461 |
c2 = 0; |
| 462 |
c3 = 0; |
| 463 |
mul_add_c(a[0], b[0], c1, c2, c3); |
| 464 |
r[0] = c1; |
| 465 |
c1 = 0; |
| 466 |
mul_add_c(a[0], b[1], c2, c3, c1); |
| 467 |
mul_add_c(a[1], b[0], c2, c3, c1); |
| 468 |
r[1] = c2; |
| 469 |
c2 = 0; |
| 470 |
mul_add_c(a[2], b[0], c3, c1, c2); |
| 471 |
mul_add_c(a[1], b[1], c3, c1, c2); |
| 472 |
mul_add_c(a[0], b[2], c3, c1, c2); |
| 473 |
r[2] = c3; |
| 474 |
c3 = 0; |
| 475 |
mul_add_c(a[0], b[3], c1, c2, c3); |
| 476 |
mul_add_c(a[1], b[2], c1, c2, c3); |
| 477 |
mul_add_c(a[2], b[1], c1, c2, c3); |
| 478 |
mul_add_c(a[3], b[0], c1, c2, c3); |
| 479 |
r[3] = c1; |
| 480 |
c1 = 0; |
| 481 |
mul_add_c(a[4], b[0], c2, c3, c1); |
| 482 |
mul_add_c(a[3], b[1], c2, c3, c1); |
| 483 |
mul_add_c(a[2], b[2], c2, c3, c1); |
| 484 |
mul_add_c(a[1], b[3], c2, c3, c1); |
| 485 |
mul_add_c(a[0], b[4], c2, c3, c1); |
| 486 |
r[4] = c2; |
| 487 |
c2 = 0; |
| 488 |
mul_add_c(a[0], b[5], c3, c1, c2); |
| 489 |
mul_add_c(a[1], b[4], c3, c1, c2); |
| 490 |
mul_add_c(a[2], b[3], c3, c1, c2); |
| 491 |
mul_add_c(a[3], b[2], c3, c1, c2); |
| 492 |
mul_add_c(a[4], b[1], c3, c1, c2); |
| 493 |
mul_add_c(a[5], b[0], c3, c1, c2); |
| 494 |
r[5] = c3; |
| 495 |
c3 = 0; |
| 496 |
mul_add_c(a[6], b[0], c1, c2, c3); |
| 497 |
mul_add_c(a[5], b[1], c1, c2, c3); |
| 498 |
mul_add_c(a[4], b[2], c1, c2, c3); |
| 499 |
mul_add_c(a[3], b[3], c1, c2, c3); |
| 500 |
mul_add_c(a[2], b[4], c1, c2, c3); |
| 501 |
mul_add_c(a[1], b[5], c1, c2, c3); |
| 502 |
mul_add_c(a[0], b[6], c1, c2, c3); |
| 503 |
r[6] = c1; |
| 504 |
c1 = 0; |
| 505 |
mul_add_c(a[0], b[7], c2, c3, c1); |
| 506 |
mul_add_c(a[1], b[6], c2, c3, c1); |
| 507 |
mul_add_c(a[2], b[5], c2, c3, c1); |
| 508 |
mul_add_c(a[3], b[4], c2, c3, c1); |
| 509 |
mul_add_c(a[4], b[3], c2, c3, c1); |
| 510 |
mul_add_c(a[5], b[2], c2, c3, c1); |
| 511 |
mul_add_c(a[6], b[1], c2, c3, c1); |
| 512 |
mul_add_c(a[7], b[0], c2, c3, c1); |
| 513 |
r[7] = c2; |
| 514 |
c2 = 0; |
| 515 |
mul_add_c(a[7], b[1], c3, c1, c2); |
| 516 |
mul_add_c(a[6], b[2], c3, c1, c2); |
| 517 |
mul_add_c(a[5], b[3], c3, c1, c2); |
| 518 |
mul_add_c(a[4], b[4], c3, c1, c2); |
| 519 |
mul_add_c(a[3], b[5], c3, c1, c2); |
| 520 |
mul_add_c(a[2], b[6], c3, c1, c2); |
| 521 |
mul_add_c(a[1], b[7], c3, c1, c2); |
| 522 |
r[8] = c3; |
| 523 |
c3 = 0; |
| 524 |
mul_add_c(a[2], b[7], c1, c2, c3); |
| 525 |
mul_add_c(a[3], b[6], c1, c2, c3); |
| 526 |
mul_add_c(a[4], b[5], c1, c2, c3); |
| 527 |
mul_add_c(a[5], b[4], c1, c2, c3); |
| 528 |
mul_add_c(a[6], b[3], c1, c2, c3); |
| 529 |
mul_add_c(a[7], b[2], c1, c2, c3); |
| 530 |
r[9] = c1; |
| 531 |
c1 = 0; |
| 532 |
mul_add_c(a[7], b[3], c2, c3, c1); |
| 533 |
mul_add_c(a[6], b[4], c2, c3, c1); |
| 534 |
mul_add_c(a[5], b[5], c2, c3, c1); |
| 535 |
mul_add_c(a[4], b[6], c2, c3, c1); |
| 536 |
mul_add_c(a[3], b[7], c2, c3, c1); |
| 537 |
r[10] = c2; |
| 538 |
c2 = 0; |
| 539 |
mul_add_c(a[4], b[7], c3, c1, c2); |
| 540 |
mul_add_c(a[5], b[6], c3, c1, c2); |
| 541 |
mul_add_c(a[6], b[5], c3, c1, c2); |
| 542 |
mul_add_c(a[7], b[4], c3, c1, c2); |
| 543 |
r[11] = c3; |
| 544 |
c3 = 0; |
| 545 |
mul_add_c(a[7], b[5], c1, c2, c3); |
| 546 |
mul_add_c(a[6], b[6], c1, c2, c3); |
| 547 |
mul_add_c(a[5], b[7], c1, c2, c3); |
| 548 |
r[12] = c1; |
| 549 |
c1 = 0; |
| 550 |
mul_add_c(a[6], b[7], c2, c3, c1); |
| 551 |
mul_add_c(a[7], b[6], c2, c3, c1); |
| 552 |
r[13] = c2; |
| 553 |
c2 = 0; |
| 554 |
mul_add_c(a[7], b[7], c3, c1, c2); |
| 555 |
r[14] = c3; |
| 556 |
r[15] = c1; |
| 557 |
} |
| 558 |
|
| 559 |
void bn_mul_comba4(BN_ULONG *r, BN_ULONG *a, BN_ULONG *b) { |
| 560 |
BN_ULONG c1, c2, c3; |
| 561 |
|
| 562 |
c1 = 0; |
| 563 |
c2 = 0; |
| 564 |
c3 = 0; |
| 565 |
mul_add_c(a[0], b[0], c1, c2, c3); |
| 566 |
r[0] = c1; |
| 567 |
c1 = 0; |
| 568 |
mul_add_c(a[0], b[1], c2, c3, c1); |
| 569 |
mul_add_c(a[1], b[0], c2, c3, c1); |
| 570 |
r[1] = c2; |
| 571 |
c2 = 0; |
| 572 |
mul_add_c(a[2], b[0], c3, c1, c2); |
| 573 |
mul_add_c(a[1], b[1], c3, c1, c2); |
| 574 |
mul_add_c(a[0], b[2], c3, c1, c2); |
| 575 |
r[2] = c3; |
| 576 |
c3 = 0; |
| 577 |
mul_add_c(a[0], b[3], c1, c2, c3); |
| 578 |
mul_add_c(a[1], b[2], c1, c2, c3); |
| 579 |
mul_add_c(a[2], b[1], c1, c2, c3); |
| 580 |
mul_add_c(a[3], b[0], c1, c2, c3); |
| 581 |
r[3] = c1; |
| 582 |
c1 = 0; |
| 583 |
mul_add_c(a[3], b[1], c2, c3, c1); |
| 584 |
mul_add_c(a[2], b[2], c2, c3, c1); |
| 585 |
mul_add_c(a[1], b[3], c2, c3, c1); |
| 586 |
r[4] = c2; |
| 587 |
c2 = 0; |
| 588 |
mul_add_c(a[2], b[3], c3, c1, c2); |
| 589 |
mul_add_c(a[3], b[2], c3, c1, c2); |
| 590 |
r[5] = c3; |
| 591 |
c3 = 0; |
| 592 |
mul_add_c(a[3], b[3], c1, c2, c3); |
| 593 |
r[6] = c1; |
| 594 |
r[7] = c2; |
| 595 |
} |
| 596 |
|
| 597 |
void bn_sqr_comba8(BN_ULONG *r, const BN_ULONG *a) { |
| 598 |
BN_ULONG c1, c2, c3; |
| 599 |
|
| 600 |
c1 = 0; |
| 601 |
c2 = 0; |
| 602 |
c3 = 0; |
| 603 |
sqr_add_c(a, 0, c1, c2, c3); |
| 604 |
r[0] = c1; |
| 605 |
c1 = 0; |
| 606 |
sqr_add_c2(a, 1, 0, c2, c3, c1); |
| 607 |
r[1] = c2; |
| 608 |
c2 = 0; |
| 609 |
sqr_add_c(a, 1, c3, c1, c2); |
| 610 |
sqr_add_c2(a, 2, 0, c3, c1, c2); |
| 611 |
r[2] = c3; |
| 612 |
c3 = 0; |
| 613 |
sqr_add_c2(a, 3, 0, c1, c2, c3); |
| 614 |
sqr_add_c2(a, 2, 1, c1, c2, c3); |
| 615 |
r[3] = c1; |
| 616 |
c1 = 0; |
| 617 |
sqr_add_c(a, 2, c2, c3, c1); |
| 618 |
sqr_add_c2(a, 3, 1, c2, c3, c1); |
| 619 |
sqr_add_c2(a, 4, 0, c2, c3, c1); |
| 620 |
r[4] = c2; |
| 621 |
c2 = 0; |
| 622 |
sqr_add_c2(a, 5, 0, c3, c1, c2); |
| 623 |
sqr_add_c2(a, 4, 1, c3, c1, c2); |
| 624 |
sqr_add_c2(a, 3, 2, c3, c1, c2); |
| 625 |
r[5] = c3; |
| 626 |
c3 = 0; |
| 627 |
sqr_add_c(a, 3, c1, c2, c3); |
| 628 |
sqr_add_c2(a, 4, 2, c1, c2, c3); |
| 629 |
sqr_add_c2(a, 5, 1, c1, c2, c3); |
| 630 |
sqr_add_c2(a, 6, 0, c1, c2, c3); |
| 631 |
r[6] = c1; |
| 632 |
c1 = 0; |
| 633 |
sqr_add_c2(a, 7, 0, c2, c3, c1); |
| 634 |
sqr_add_c2(a, 6, 1, c2, c3, c1); |
| 635 |
sqr_add_c2(a, 5, 2, c2, c3, c1); |
| 636 |
sqr_add_c2(a, 4, 3, c2, c3, c1); |
| 637 |
r[7] = c2; |
| 638 |
c2 = 0; |
| 639 |
sqr_add_c(a, 4, c3, c1, c2); |
| 640 |
sqr_add_c2(a, 5, 3, c3, c1, c2); |
| 641 |
sqr_add_c2(a, 6, 2, c3, c1, c2); |
| 642 |
sqr_add_c2(a, 7, 1, c3, c1, c2); |
| 643 |
r[8] = c3; |
| 644 |
c3 = 0; |
| 645 |
sqr_add_c2(a, 7, 2, c1, c2, c3); |
| 646 |
sqr_add_c2(a, 6, 3, c1, c2, c3); |
| 647 |
sqr_add_c2(a, 5, 4, c1, c2, c3); |
| 648 |
r[9] = c1; |
| 649 |
c1 = 0; |
| 650 |
sqr_add_c(a, 5, c2, c3, c1); |
| 651 |
sqr_add_c2(a, 6, 4, c2, c3, c1); |
| 652 |
sqr_add_c2(a, 7, 3, c2, c3, c1); |
| 653 |
r[10] = c2; |
| 654 |
c2 = 0; |
| 655 |
sqr_add_c2(a, 7, 4, c3, c1, c2); |
| 656 |
sqr_add_c2(a, 6, 5, c3, c1, c2); |
| 657 |
r[11] = c3; |
| 658 |
c3 = 0; |
| 659 |
sqr_add_c(a, 6, c1, c2, c3); |
| 660 |
sqr_add_c2(a, 7, 5, c1, c2, c3); |
| 661 |
r[12] = c1; |
| 662 |
c1 = 0; |
| 663 |
sqr_add_c2(a, 7, 6, c2, c3, c1); |
| 664 |
r[13] = c2; |
| 665 |
c2 = 0; |
| 666 |
sqr_add_c(a, 7, c3, c1, c2); |
| 667 |
r[14] = c3; |
| 668 |
r[15] = c1; |
| 669 |
} |
| 670 |
|
| 671 |
void bn_sqr_comba4(BN_ULONG *r, const BN_ULONG *a) { |
| 672 |
BN_ULONG c1, c2, c3; |
| 673 |
|
| 674 |
c1 = 0; |
| 675 |
c2 = 0; |
| 676 |
c3 = 0; |
| 677 |
sqr_add_c(a, 0, c1, c2, c3); |
| 678 |
r[0] = c1; |
| 679 |
c1 = 0; |
| 680 |
sqr_add_c2(a, 1, 0, c2, c3, c1); |
| 681 |
r[1] = c2; |
| 682 |
c2 = 0; |
| 683 |
sqr_add_c(a, 1, c3, c1, c2); |
| 684 |
sqr_add_c2(a, 2, 0, c3, c1, c2); |
| 685 |
r[2] = c3; |
| 686 |
c3 = 0; |
| 687 |
sqr_add_c2(a, 3, 0, c1, c2, c3); |
| 688 |
sqr_add_c2(a, 2, 1, c1, c2, c3); |
| 689 |
r[3] = c1; |
| 690 |
c1 = 0; |
| 691 |
sqr_add_c(a, 2, c2, c3, c1); |
| 692 |
sqr_add_c2(a, 3, 1, c2, c3, c1); |
| 693 |
r[4] = c2; |
| 694 |
c2 = 0; |
| 695 |
sqr_add_c2(a, 3, 2, c3, c1, c2); |
| 696 |
r[5] = c3; |
| 697 |
c3 = 0; |
| 698 |
sqr_add_c(a, 3, c1, c2, c3); |
| 699 |
r[6] = c1; |
| 700 |
r[7] = c2; |
| 701 |
} |
| 702 |
|
| 703 |
#endif |