1/* 2Implementation by the Keccak, Keyak and Ketje Teams, namely, Guido Bertoni, 3Joan Daemen, Michaël Peeters, Gilles Van Assche and Ronny Van Keer, hereby 4denoted as "the implementer". 5 6For more information, feedback or questions, please refer to our websites: 7http://keccak.noekeon.org/ 8http://keyak.noekeon.org/ 9http://ketje.noekeon.org/ 10 11To the extent possible under law, the implementer has waived all copyright 12and related or neighboring rights to the source code in this file. 13http://creativecommons.org/publicdomain/zero/1.0/ 14*/ 15 16#define declareABCDE \ 17 UINT64 Aba, Abe, Abi, Abo, Abu; \ 18 UINT64 Aga, Age, Agi, Ago, Agu; \ 19 UINT64 Aka, Ake, Aki, Ako, Aku; \ 20 UINT64 Ama, Ame, Ami, Amo, Amu; \ 21 UINT64 Asa, Ase, Asi, Aso, Asu; \ 22 UINT64 Bba, Bbe, Bbi, Bbo, Bbu; \ 23 UINT64 Bga, Bge, Bgi, Bgo, Bgu; \ 24 UINT64 Bka, Bke, Bki, Bko, Bku; \ 25 UINT64 Bma, Bme, Bmi, Bmo, Bmu; \ 26 UINT64 Bsa, Bse, Bsi, Bso, Bsu; \ 27 UINT64 Ca, Ce, Ci, Co, Cu; \ 28 UINT64 Da, De, Di, Do, Du; \ 29 UINT64 Eba, Ebe, Ebi, Ebo, Ebu; \ 30 UINT64 Ega, Ege, Egi, Ego, Egu; \ 31 UINT64 Eka, Eke, Eki, Eko, Eku; \ 32 UINT64 Ema, Eme, Emi, Emo, Emu; \ 33 UINT64 Esa, Ese, Esi, Eso, Esu; \ 34 35#define prepareTheta \ 36 Ca = Aba^Aga^Aka^Ama^Asa; \ 37 Ce = Abe^Age^Ake^Ame^Ase; \ 38 Ci = Abi^Agi^Aki^Ami^Asi; \ 39 Co = Abo^Ago^Ako^Amo^Aso; \ 40 Cu = Abu^Agu^Aku^Amu^Asu; \ 41 42#ifdef UseBebigokimisa 43/* --- Code for round, with prepare-theta (lane complementing pattern 'bebigokimisa') */ 44/* --- 64-bit lanes mapped to 64-bit words */ 45#define thetaRhoPiChiIotaPrepareTheta(i, A, E) \ 46 Da = Cu^ROL64(Ce, 1); \ 47 De = Ca^ROL64(Ci, 1); \ 48 Di = Ce^ROL64(Co, 1); \ 49 Do = Ci^ROL64(Cu, 1); \ 50 Du = Co^ROL64(Ca, 1); \ 51\ 52 A##ba ^= Da; \ 53 Bba = A##ba; \ 54 A##ge ^= De; \ 55 Bbe = ROL64(A##ge, 44); \ 56 A##ki ^= Di; \ 57 Bbi = ROL64(A##ki, 43); \ 58 A##mo ^= Do; \ 59 Bbo = ROL64(A##mo, 21); \ 60 A##su ^= Du; \ 61 Bbu = ROL64(A##su, 14); \ 62 E##ba = Bba ^( Bbe | Bbi ); \ 63 E##ba ^= KeccakF1600RoundConstants[i]; \ 64 Ca = E##ba; \ 65 E##be = Bbe ^((~Bbi)| Bbo ); \ 66 Ce = E##be; \ 67 E##bi = Bbi ^( Bbo & Bbu ); \ 68 Ci = E##bi; \ 69 E##bo = Bbo ^( Bbu | Bba ); \ 70 Co = E##bo; \ 71 E##bu = Bbu ^( Bba & Bbe ); \ 72 Cu = E##bu; \ 73\ 74 A##bo ^= Do; \ 75 Bga = ROL64(A##bo, 28); \ 76 A##gu ^= Du; \ 77 Bge = ROL64(A##gu, 20); \ 78 A##ka ^= Da; \ 79 Bgi = ROL64(A##ka, 3); \ 80 A##me ^= De; \ 81 Bgo = ROL64(A##me, 45); \ 82 A##si ^= Di; \ 83 Bgu = ROL64(A##si, 61); \ 84 E##ga = Bga ^( Bge | Bgi ); \ 85 Ca ^= E##ga; \ 86 E##ge = Bge ^( Bgi & Bgo ); \ 87 Ce ^= E##ge; \ 88 E##gi = Bgi ^( Bgo |(~Bgu)); \ 89 Ci ^= E##gi; \ 90 E##go = Bgo ^( Bgu | Bga ); \ 91 Co ^= E##go; \ 92 E##gu = Bgu ^( Bga & Bge ); \ 93 Cu ^= E##gu; \ 94\ 95 A##be ^= De; \ 96 Bka = ROL64(A##be, 1); \ 97 A##gi ^= Di; \ 98 Bke = ROL64(A##gi, 6); \ 99 A##ko ^= Do; \ 100 Bki = ROL64(A##ko, 25); \ 101 A##mu ^= Du; \ 102 Bko = ROL64(A##mu, 8); \ 103 A##sa ^= Da; \ 104 Bku = ROL64(A##sa, 18); \ 105 E##ka = Bka ^( Bke | Bki ); \ 106 Ca ^= E##ka; \ 107 E##ke = Bke ^( Bki & Bko ); \ 108 Ce ^= E##ke; \ 109 E##ki = Bki ^((~Bko)& Bku ); \ 110 Ci ^= E##ki; \ 111 E##ko = (~Bko)^( Bku | Bka ); \ 112 Co ^= E##ko; \ 113 E##ku = Bku ^( Bka & Bke ); \ 114 Cu ^= E##ku; \ 115\ 116 A##bu ^= Du; \ 117 Bma = ROL64(A##bu, 27); \ 118 A##ga ^= Da; \ 119 Bme = ROL64(A##ga, 36); \ 120 A##ke ^= De; \ 121 Bmi = ROL64(A##ke, 10); \ 122 A##mi ^= Di; \ 123 Bmo = ROL64(A##mi, 15); \ 124 A##so ^= Do; \ 125 Bmu = ROL64(A##so, 56); \ 126 E##ma = Bma ^( Bme & Bmi ); \ 127 Ca ^= E##ma; \ 128 E##me = Bme ^( Bmi | Bmo ); \ 129 Ce ^= E##me; \ 130 E##mi = Bmi ^((~Bmo)| Bmu ); \ 131 Ci ^= E##mi; \ 132 E##mo = (~Bmo)^( Bmu & Bma ); \ 133 Co ^= E##mo; \ 134 E##mu = Bmu ^( Bma | Bme ); \ 135 Cu ^= E##mu; \ 136\ 137 A##bi ^= Di; \ 138 Bsa = ROL64(A##bi, 62); \ 139 A##go ^= Do; \ 140 Bse = ROL64(A##go, 55); \ 141 A##ku ^= Du; \ 142 Bsi = ROL64(A##ku, 39); \ 143 A##ma ^= Da; \ 144 Bso = ROL64(A##ma, 41); \ 145 A##se ^= De; \ 146 Bsu = ROL64(A##se, 2); \ 147 E##sa = Bsa ^((~Bse)& Bsi ); \ 148 Ca ^= E##sa; \ 149 E##se = (~Bse)^( Bsi | Bso ); \ 150 Ce ^= E##se; \ 151 E##si = Bsi ^( Bso & Bsu ); \ 152 Ci ^= E##si; \ 153 E##so = Bso ^( Bsu | Bsa ); \ 154 Co ^= E##so; \ 155 E##su = Bsu ^( Bsa & Bse ); \ 156 Cu ^= E##su; \ 157\ 158 159/* --- Code for round (lane complementing pattern 'bebigokimisa') */ 160/* --- 64-bit lanes mapped to 64-bit words */ 161#define thetaRhoPiChiIota(i, A, E) \ 162 Da = Cu^ROL64(Ce, 1); \ 163 De = Ca^ROL64(Ci, 1); \ 164 Di = Ce^ROL64(Co, 1); \ 165 Do = Ci^ROL64(Cu, 1); \ 166 Du = Co^ROL64(Ca, 1); \ 167\ 168 A##ba ^= Da; \ 169 Bba = A##ba; \ 170 A##ge ^= De; \ 171 Bbe = ROL64(A##ge, 44); \ 172 A##ki ^= Di; \ 173 Bbi = ROL64(A##ki, 43); \ 174 A##mo ^= Do; \ 175 Bbo = ROL64(A##mo, 21); \ 176 A##su ^= Du; \ 177 Bbu = ROL64(A##su, 14); \ 178 E##ba = Bba ^( Bbe | Bbi ); \ 179 E##ba ^= KeccakF1600RoundConstants[i]; \ 180 E##be = Bbe ^((~Bbi)| Bbo ); \ 181 E##bi = Bbi ^( Bbo & Bbu ); \ 182 E##bo = Bbo ^( Bbu | Bba ); \ 183 E##bu = Bbu ^( Bba & Bbe ); \ 184\ 185 A##bo ^= Do; \ 186 Bga = ROL64(A##bo, 28); \ 187 A##gu ^= Du; \ 188 Bge = ROL64(A##gu, 20); \ 189 A##ka ^= Da; \ 190 Bgi = ROL64(A##ka, 3); \ 191 A##me ^= De; \ 192 Bgo = ROL64(A##me, 45); \ 193 A##si ^= Di; \ 194 Bgu = ROL64(A##si, 61); \ 195 E##ga = Bga ^( Bge | Bgi ); \ 196 E##ge = Bge ^( Bgi & Bgo ); \ 197 E##gi = Bgi ^( Bgo |(~Bgu)); \ 198 E##go = Bgo ^( Bgu | Bga ); \ 199 E##gu = Bgu ^( Bga & Bge ); \ 200\ 201 A##be ^= De; \ 202 Bka = ROL64(A##be, 1); \ 203 A##gi ^= Di; \ 204 Bke = ROL64(A##gi, 6); \ 205 A##ko ^= Do; \ 206 Bki = ROL64(A##ko, 25); \ 207 A##mu ^= Du; \ 208 Bko = ROL64(A##mu, 8); \ 209 A##sa ^= Da; \ 210 Bku = ROL64(A##sa, 18); \ 211 E##ka = Bka ^( Bke | Bki ); \ 212 E##ke = Bke ^( Bki & Bko ); \ 213 E##ki = Bki ^((~Bko)& Bku ); \ 214 E##ko = (~Bko)^( Bku | Bka ); \ 215 E##ku = Bku ^( Bka & Bke ); \ 216\ 217 A##bu ^= Du; \ 218 Bma = ROL64(A##bu, 27); \ 219 A##ga ^= Da; \ 220 Bme = ROL64(A##ga, 36); \ 221 A##ke ^= De; \ 222 Bmi = ROL64(A##ke, 10); \ 223 A##mi ^= Di; \ 224 Bmo = ROL64(A##mi, 15); \ 225 A##so ^= Do; \ 226 Bmu = ROL64(A##so, 56); \ 227 E##ma = Bma ^( Bme & Bmi ); \ 228 E##me = Bme ^( Bmi | Bmo ); \ 229 E##mi = Bmi ^((~Bmo)| Bmu ); \ 230 E##mo = (~Bmo)^( Bmu & Bma ); \ 231 E##mu = Bmu ^( Bma | Bme ); \ 232\ 233 A##bi ^= Di; \ 234 Bsa = ROL64(A##bi, 62); \ 235 A##go ^= Do; \ 236 Bse = ROL64(A##go, 55); \ 237 A##ku ^= Du; \ 238 Bsi = ROL64(A##ku, 39); \ 239 A##ma ^= Da; \ 240 Bso = ROL64(A##ma, 41); \ 241 A##se ^= De; \ 242 Bsu = ROL64(A##se, 2); \ 243 E##sa = Bsa ^((~Bse)& Bsi ); \ 244 E##se = (~Bse)^( Bsi | Bso ); \ 245 E##si = Bsi ^( Bso & Bsu ); \ 246 E##so = Bso ^( Bsu | Bsa ); \ 247 E##su = Bsu ^( Bsa & Bse ); \ 248\ 249 250#else /* UseBebigokimisa */ 251/* --- Code for round, with prepare-theta */ 252/* --- 64-bit lanes mapped to 64-bit words */ 253#define thetaRhoPiChiIotaPrepareTheta(i, A, E) \ 254 Da = Cu^ROL64(Ce, 1); \ 255 De = Ca^ROL64(Ci, 1); \ 256 Di = Ce^ROL64(Co, 1); \ 257 Do = Ci^ROL64(Cu, 1); \ 258 Du = Co^ROL64(Ca, 1); \ 259\ 260 A##ba ^= Da; \ 261 Bba = A##ba; \ 262 A##ge ^= De; \ 263 Bbe = ROL64(A##ge, 44); \ 264 A##ki ^= Di; \ 265 Bbi = ROL64(A##ki, 43); \ 266 A##mo ^= Do; \ 267 Bbo = ROL64(A##mo, 21); \ 268 A##su ^= Du; \ 269 Bbu = ROL64(A##su, 14); \ 270 E##ba = Bba ^((~Bbe)& Bbi ); \ 271 E##ba ^= KeccakF1600RoundConstants[i]; \ 272 Ca = E##ba; \ 273 E##be = Bbe ^((~Bbi)& Bbo ); \ 274 Ce = E##be; \ 275 E##bi = Bbi ^((~Bbo)& Bbu ); \ 276 Ci = E##bi; \ 277 E##bo = Bbo ^((~Bbu)& Bba ); \ 278 Co = E##bo; \ 279 E##bu = Bbu ^((~Bba)& Bbe ); \ 280 Cu = E##bu; \ 281\ 282 A##bo ^= Do; \ 283 Bga = ROL64(A##bo, 28); \ 284 A##gu ^= Du; \ 285 Bge = ROL64(A##gu, 20); \ 286 A##ka ^= Da; \ 287 Bgi = ROL64(A##ka, 3); \ 288 A##me ^= De; \ 289 Bgo = ROL64(A##me, 45); \ 290 A##si ^= Di; \ 291 Bgu = ROL64(A##si, 61); \ 292 E##ga = Bga ^((~Bge)& Bgi ); \ 293 Ca ^= E##ga; \ 294 E##ge = Bge ^((~Bgi)& Bgo ); \ 295 Ce ^= E##ge; \ 296 E##gi = Bgi ^((~Bgo)& Bgu ); \ 297 Ci ^= E##gi; \ 298 E##go = Bgo ^((~Bgu)& Bga ); \ 299 Co ^= E##go; \ 300 E##gu = Bgu ^((~Bga)& Bge ); \ 301 Cu ^= E##gu; \ 302\ 303 A##be ^= De; \ 304 Bka = ROL64(A##be, 1); \ 305 A##gi ^= Di; \ 306 Bke = ROL64(A##gi, 6); \ 307 A##ko ^= Do; \ 308 Bki = ROL64(A##ko, 25); \ 309 A##mu ^= Du; \ 310 Bko = ROL64(A##mu, 8); \ 311 A##sa ^= Da; \ 312 Bku = ROL64(A##sa, 18); \ 313 E##ka = Bka ^((~Bke)& Bki ); \ 314 Ca ^= E##ka; \ 315 E##ke = Bke ^((~Bki)& Bko ); \ 316 Ce ^= E##ke; \ 317 E##ki = Bki ^((~Bko)& Bku ); \ 318 Ci ^= E##ki; \ 319 E##ko = Bko ^((~Bku)& Bka ); \ 320 Co ^= E##ko; \ 321 E##ku = Bku ^((~Bka)& Bke ); \ 322 Cu ^= E##ku; \ 323\ 324 A##bu ^= Du; \ 325 Bma = ROL64(A##bu, 27); \ 326 A##ga ^= Da; \ 327 Bme = ROL64(A##ga, 36); \ 328 A##ke ^= De; \ 329 Bmi = ROL64(A##ke, 10); \ 330 A##mi ^= Di; \ 331 Bmo = ROL64(A##mi, 15); \ 332 A##so ^= Do; \ 333 Bmu = ROL64(A##so, 56); \ 334 E##ma = Bma ^((~Bme)& Bmi ); \ 335 Ca ^= E##ma; \ 336 E##me = Bme ^((~Bmi)& Bmo ); \ 337 Ce ^= E##me; \ 338 E##mi = Bmi ^((~Bmo)& Bmu ); \ 339 Ci ^= E##mi; \ 340 E##mo = Bmo ^((~Bmu)& Bma ); \ 341 Co ^= E##mo; \ 342 E##mu = Bmu ^((~Bma)& Bme ); \ 343 Cu ^= E##mu; \ 344\ 345 A##bi ^= Di; \ 346 Bsa = ROL64(A##bi, 62); \ 347 A##go ^= Do; \ 348 Bse = ROL64(A##go, 55); \ 349 A##ku ^= Du; \ 350 Bsi = ROL64(A##ku, 39); \ 351 A##ma ^= Da; \ 352 Bso = ROL64(A##ma, 41); \ 353 A##se ^= De; \ 354 Bsu = ROL64(A##se, 2); \ 355 E##sa = Bsa ^((~Bse)& Bsi ); \ 356 Ca ^= E##sa; \ 357 E##se = Bse ^((~Bsi)& Bso ); \ 358 Ce ^= E##se; \ 359 E##si = Bsi ^((~Bso)& Bsu ); \ 360 Ci ^= E##si; \ 361 E##so = Bso ^((~Bsu)& Bsa ); \ 362 Co ^= E##so; \ 363 E##su = Bsu ^((~Bsa)& Bse ); \ 364 Cu ^= E##su; \ 365\ 366 367/* --- Code for round */ 368/* --- 64-bit lanes mapped to 64-bit words */ 369#define thetaRhoPiChiIota(i, A, E) \ 370 Da = Cu^ROL64(Ce, 1); \ 371 De = Ca^ROL64(Ci, 1); \ 372 Di = Ce^ROL64(Co, 1); \ 373 Do = Ci^ROL64(Cu, 1); \ 374 Du = Co^ROL64(Ca, 1); \ 375\ 376 A##ba ^= Da; \ 377 Bba = A##ba; \ 378 A##ge ^= De; \ 379 Bbe = ROL64(A##ge, 44); \ 380 A##ki ^= Di; \ 381 Bbi = ROL64(A##ki, 43); \ 382 A##mo ^= Do; \ 383 Bbo = ROL64(A##mo, 21); \ 384 A##su ^= Du; \ 385 Bbu = ROL64(A##su, 14); \ 386 E##ba = Bba ^((~Bbe)& Bbi ); \ 387 E##ba ^= KeccakF1600RoundConstants[i]; \ 388 E##be = Bbe ^((~Bbi)& Bbo ); \ 389 E##bi = Bbi ^((~Bbo)& Bbu ); \ 390 E##bo = Bbo ^((~Bbu)& Bba ); \ 391 E##bu = Bbu ^((~Bba)& Bbe ); \ 392\ 393 A##bo ^= Do; \ 394 Bga = ROL64(A##bo, 28); \ 395 A##gu ^= Du; \ 396 Bge = ROL64(A##gu, 20); \ 397 A##ka ^= Da; \ 398 Bgi = ROL64(A##ka, 3); \ 399 A##me ^= De; \ 400 Bgo = ROL64(A##me, 45); \ 401 A##si ^= Di; \ 402 Bgu = ROL64(A##si, 61); \ 403 E##ga = Bga ^((~Bge)& Bgi ); \ 404 E##ge = Bge ^((~Bgi)& Bgo ); \ 405 E##gi = Bgi ^((~Bgo)& Bgu ); \ 406 E##go = Bgo ^((~Bgu)& Bga ); \ 407 E##gu = Bgu ^((~Bga)& Bge ); \ 408\ 409 A##be ^= De; \ 410 Bka = ROL64(A##be, 1); \ 411 A##gi ^= Di; \ 412 Bke = ROL64(A##gi, 6); \ 413 A##ko ^= Do; \ 414 Bki = ROL64(A##ko, 25); \ 415 A##mu ^= Du; \ 416 Bko = ROL64(A##mu, 8); \ 417 A##sa ^= Da; \ 418 Bku = ROL64(A##sa, 18); \ 419 E##ka = Bka ^((~Bke)& Bki ); \ 420 E##ke = Bke ^((~Bki)& Bko ); \ 421 E##ki = Bki ^((~Bko)& Bku ); \ 422 E##ko = Bko ^((~Bku)& Bka ); \ 423 E##ku = Bku ^((~Bka)& Bke ); \ 424\ 425 A##bu ^= Du; \ 426 Bma = ROL64(A##bu, 27); \ 427 A##ga ^= Da; \ 428 Bme = ROL64(A##ga, 36); \ 429 A##ke ^= De; \ 430 Bmi = ROL64(A##ke, 10); \ 431 A##mi ^= Di; \ 432 Bmo = ROL64(A##mi, 15); \ 433 A##so ^= Do; \ 434 Bmu = ROL64(A##so, 56); \ 435 E##ma = Bma ^((~Bme)& Bmi ); \ 436 E##me = Bme ^((~Bmi)& Bmo ); \ 437 E##mi = Bmi ^((~Bmo)& Bmu ); \ 438 E##mo = Bmo ^((~Bmu)& Bma ); \ 439 E##mu = Bmu ^((~Bma)& Bme ); \ 440\ 441 A##bi ^= Di; \ 442 Bsa = ROL64(A##bi, 62); \ 443 A##go ^= Do; \ 444 Bse = ROL64(A##go, 55); \ 445 A##ku ^= Du; \ 446 Bsi = ROL64(A##ku, 39); \ 447 A##ma ^= Da; \ 448 Bso = ROL64(A##ma, 41); \ 449 A##se ^= De; \ 450 Bsu = ROL64(A##se, 2); \ 451 E##sa = Bsa ^((~Bse)& Bsi ); \ 452 E##se = Bse ^((~Bsi)& Bso ); \ 453 E##si = Bsi ^((~Bso)& Bsu ); \ 454 E##so = Bso ^((~Bsu)& Bsa ); \ 455 E##su = Bsu ^((~Bsa)& Bse ); \ 456\ 457 458#endif /* UseBebigokimisa */ 459 460#define copyFromState(X, state) \ 461 X##ba = state[ 0]; \ 462 X##be = state[ 1]; \ 463 X##bi = state[ 2]; \ 464 X##bo = state[ 3]; \ 465 X##bu = state[ 4]; \ 466 X##ga = state[ 5]; \ 467 X##ge = state[ 6]; \ 468 X##gi = state[ 7]; \ 469 X##go = state[ 8]; \ 470 X##gu = state[ 9]; \ 471 X##ka = state[10]; \ 472 X##ke = state[11]; \ 473 X##ki = state[12]; \ 474 X##ko = state[13]; \ 475 X##ku = state[14]; \ 476 X##ma = state[15]; \ 477 X##me = state[16]; \ 478 X##mi = state[17]; \ 479 X##mo = state[18]; \ 480 X##mu = state[19]; \ 481 X##sa = state[20]; \ 482 X##se = state[21]; \ 483 X##si = state[22]; \ 484 X##so = state[23]; \ 485 X##su = state[24]; \ 486 487#define copyToState(state, X) \ 488 state[ 0] = X##ba; \ 489 state[ 1] = X##be; \ 490 state[ 2] = X##bi; \ 491 state[ 3] = X##bo; \ 492 state[ 4] = X##bu; \ 493 state[ 5] = X##ga; \ 494 state[ 6] = X##ge; \ 495 state[ 7] = X##gi; \ 496 state[ 8] = X##go; \ 497 state[ 9] = X##gu; \ 498 state[10] = X##ka; \ 499 state[11] = X##ke; \ 500 state[12] = X##ki; \ 501 state[13] = X##ko; \ 502 state[14] = X##ku; \ 503 state[15] = X##ma; \ 504 state[16] = X##me; \ 505 state[17] = X##mi; \ 506 state[18] = X##mo; \ 507 state[19] = X##mu; \ 508 state[20] = X##sa; \ 509 state[21] = X##se; \ 510 state[22] = X##si; \ 511 state[23] = X##so; \ 512 state[24] = X##su; \ 513 514#define copyStateVariables(X, Y) \ 515 X##ba = Y##ba; \ 516 X##be = Y##be; \ 517 X##bi = Y##bi; \ 518 X##bo = Y##bo; \ 519 X##bu = Y##bu; \ 520 X##ga = Y##ga; \ 521 X##ge = Y##ge; \ 522 X##gi = Y##gi; \ 523 X##go = Y##go; \ 524 X##gu = Y##gu; \ 525 X##ka = Y##ka; \ 526 X##ke = Y##ke; \ 527 X##ki = Y##ki; \ 528 X##ko = Y##ko; \ 529 X##ku = Y##ku; \ 530 X##ma = Y##ma; \ 531 X##me = Y##me; \ 532 X##mi = Y##mi; \ 533 X##mo = Y##mo; \ 534 X##mu = Y##mu; \ 535 X##sa = Y##sa; \ 536 X##se = Y##se; \ 537 X##si = Y##si; \ 538 X##so = Y##so; \ 539 X##su = Y##su; \ 540 541#define copyFromStateAndAdd(X, state, input, laneCount) \ 542 if (laneCount < 16) { \ 543 if (laneCount < 8) { \ 544 if (laneCount < 4) { \ 545 if (laneCount < 2) { \ 546 if (laneCount < 1) { \ 547 X##ba = state[ 0]; \ 548 } \ 549 else { \ 550 X##ba = state[ 0]^input[ 0]; \ 551 } \ 552 X##be = state[ 1]; \ 553 X##bi = state[ 2]; \ 554 } \ 555 else { \ 556 X##ba = state[ 0]^input[ 0]; \ 557 X##be = state[ 1]^input[ 1]; \ 558 if (laneCount < 3) { \ 559 X##bi = state[ 2]; \ 560 } \ 561 else { \ 562 X##bi = state[ 2]^input[ 2]; \ 563 } \ 564 } \ 565 X##bo = state[ 3]; \ 566 X##bu = state[ 4]; \ 567 X##ga = state[ 5]; \ 568 X##ge = state[ 6]; \ 569 } \ 570 else { \ 571 X##ba = state[ 0]^input[ 0]; \ 572 X##be = state[ 1]^input[ 1]; \ 573 X##bi = state[ 2]^input[ 2]; \ 574 X##bo = state[ 3]^input[ 3]; \ 575 if (laneCount < 6) { \ 576 if (laneCount < 5) { \ 577 X##bu = state[ 4]; \ 578 } \ 579 else { \ 580 X##bu = state[ 4]^input[ 4]; \ 581 } \ 582 X##ga = state[ 5]; \ 583 X##ge = state[ 6]; \ 584 } \ 585 else { \ 586 X##bu = state[ 4]^input[ 4]; \ 587 X##ga = state[ 5]^input[ 5]; \ 588 if (laneCount < 7) { \ 589 X##ge = state[ 6]; \ 590 } \ 591 else { \ 592 X##ge = state[ 6]^input[ 6]; \ 593 } \ 594 } \ 595 } \ 596 X##gi = state[ 7]; \ 597 X##go = state[ 8]; \ 598 X##gu = state[ 9]; \ 599 X##ka = state[10]; \ 600 X##ke = state[11]; \ 601 X##ki = state[12]; \ 602 X##ko = state[13]; \ 603 X##ku = state[14]; \ 604 } \ 605 else { \ 606 X##ba = state[ 0]^input[ 0]; \ 607 X##be = state[ 1]^input[ 1]; \ 608 X##bi = state[ 2]^input[ 2]; \ 609 X##bo = state[ 3]^input[ 3]; \ 610 X##bu = state[ 4]^input[ 4]; \ 611 X##ga = state[ 5]^input[ 5]; \ 612 X##ge = state[ 6]^input[ 6]; \ 613 X##gi = state[ 7]^input[ 7]; \ 614 if (laneCount < 12) { \ 615 if (laneCount < 10) { \ 616 if (laneCount < 9) { \ 617 X##go = state[ 8]; \ 618 } \ 619 else { \ 620 X##go = state[ 8]^input[ 8]; \ 621 } \ 622 X##gu = state[ 9]; \ 623 X##ka = state[10]; \ 624 } \ 625 else { \ 626 X##go = state[ 8]^input[ 8]; \ 627 X##gu = state[ 9]^input[ 9]; \ 628 if (laneCount < 11) { \ 629 X##ka = state[10]; \ 630 } \ 631 else { \ 632 X##ka = state[10]^input[10]; \ 633 } \ 634 } \ 635 X##ke = state[11]; \ 636 X##ki = state[12]; \ 637 X##ko = state[13]; \ 638 X##ku = state[14]; \ 639 } \ 640 else { \ 641 X##go = state[ 8]^input[ 8]; \ 642 X##gu = state[ 9]^input[ 9]; \ 643 X##ka = state[10]^input[10]; \ 644 X##ke = state[11]^input[11]; \ 645 if (laneCount < 14) { \ 646 if (laneCount < 13) { \ 647 X##ki = state[12]; \ 648 } \ 649 else { \ 650 X##ki = state[12]^input[12]; \ 651 } \ 652 X##ko = state[13]; \ 653 X##ku = state[14]; \ 654 } \ 655 else { \ 656 X##ki = state[12]^input[12]; \ 657 X##ko = state[13]^input[13]; \ 658 if (laneCount < 15) { \ 659 X##ku = state[14]; \ 660 } \ 661 else { \ 662 X##ku = state[14]^input[14]; \ 663 } \ 664 } \ 665 } \ 666 } \ 667 X##ma = state[15]; \ 668 X##me = state[16]; \ 669 X##mi = state[17]; \ 670 X##mo = state[18]; \ 671 X##mu = state[19]; \ 672 X##sa = state[20]; \ 673 X##se = state[21]; \ 674 X##si = state[22]; \ 675 X##so = state[23]; \ 676 X##su = state[24]; \ 677 } \ 678 else { \ 679 X##ba = state[ 0]^input[ 0]; \ 680 X##be = state[ 1]^input[ 1]; \ 681 X##bi = state[ 2]^input[ 2]; \ 682 X##bo = state[ 3]^input[ 3]; \ 683 X##bu = state[ 4]^input[ 4]; \ 684 X##ga = state[ 5]^input[ 5]; \ 685 X##ge = state[ 6]^input[ 6]; \ 686 X##gi = state[ 7]^input[ 7]; \ 687 X##go = state[ 8]^input[ 8]; \ 688 X##gu = state[ 9]^input[ 9]; \ 689 X##ka = state[10]^input[10]; \ 690 X##ke = state[11]^input[11]; \ 691 X##ki = state[12]^input[12]; \ 692 X##ko = state[13]^input[13]; \ 693 X##ku = state[14]^input[14]; \ 694 X##ma = state[15]^input[15]; \ 695 if (laneCount < 24) { \ 696 if (laneCount < 20) { \ 697 if (laneCount < 18) { \ 698 if (laneCount < 17) { \ 699 X##me = state[16]; \ 700 } \ 701 else { \ 702 X##me = state[16]^input[16]; \ 703 } \ 704 X##mi = state[17]; \ 705 X##mo = state[18]; \ 706 } \ 707 else { \ 708 X##me = state[16]^input[16]; \ 709 X##mi = state[17]^input[17]; \ 710 if (laneCount < 19) { \ 711 X##mo = state[18]; \ 712 } \ 713 else { \ 714 X##mo = state[18]^input[18]; \ 715 } \ 716 } \ 717 X##mu = state[19]; \ 718 X##sa = state[20]; \ 719 X##se = state[21]; \ 720 X##si = state[22]; \ 721 } \ 722 else { \ 723 X##me = state[16]^input[16]; \ 724 X##mi = state[17]^input[17]; \ 725 X##mo = state[18]^input[18]; \ 726 X##mu = state[19]^input[19]; \ 727 if (laneCount < 22) { \ 728 if (laneCount < 21) { \ 729 X##sa = state[20]; \ 730 } \ 731 else { \ 732 X##sa = state[20]^input[20]; \ 733 } \ 734 X##se = state[21]; \ 735 X##si = state[22]; \ 736 } \ 737 else { \ 738 X##sa = state[20]^input[20]; \ 739 X##se = state[21]^input[21]; \ 740 if (laneCount < 23) { \ 741 X##si = state[22]; \ 742 } \ 743 else { \ 744 X##si = state[22]^input[22]; \ 745 } \ 746 } \ 747 } \ 748 X##so = state[23]; \ 749 X##su = state[24]; \ 750 } \ 751 else { \ 752 X##me = state[16]^input[16]; \ 753 X##mi = state[17]^input[17]; \ 754 X##mo = state[18]^input[18]; \ 755 X##mu = state[19]^input[19]; \ 756 X##sa = state[20]^input[20]; \ 757 X##se = state[21]^input[21]; \ 758 X##si = state[22]^input[22]; \ 759 X##so = state[23]^input[23]; \ 760 if (laneCount < 25) { \ 761 X##su = state[24]; \ 762 } \ 763 else { \ 764 X##su = state[24]^input[24]; \ 765 } \ 766 } \ 767 } 768 769#define addInput(X, input, laneCount) \ 770 if (laneCount == 21) { \ 771 X##ba ^= input[ 0]; \ 772 X##be ^= input[ 1]; \ 773 X##bi ^= input[ 2]; \ 774 X##bo ^= input[ 3]; \ 775 X##bu ^= input[ 4]; \ 776 X##ga ^= input[ 5]; \ 777 X##ge ^= input[ 6]; \ 778 X##gi ^= input[ 7]; \ 779 X##go ^= input[ 8]; \ 780 X##gu ^= input[ 9]; \ 781 X##ka ^= input[10]; \ 782 X##ke ^= input[11]; \ 783 X##ki ^= input[12]; \ 784 X##ko ^= input[13]; \ 785 X##ku ^= input[14]; \ 786 X##ma ^= input[15]; \ 787 X##me ^= input[16]; \ 788 X##mi ^= input[17]; \ 789 X##mo ^= input[18]; \ 790 X##mu ^= input[19]; \ 791 X##sa ^= input[20]; \ 792 } \ 793 else if (laneCount < 16) { \ 794 if (laneCount < 8) { \ 795 if (laneCount < 4) { \ 796 if (laneCount < 2) { \ 797 if (laneCount < 1) { \ 798 } \ 799 else { \ 800 X##ba ^= input[ 0]; \ 801 } \ 802 } \ 803 else { \ 804 X##ba ^= input[ 0]; \ 805 X##be ^= input[ 1]; \ 806 if (laneCount < 3) { \ 807 } \ 808 else { \ 809 X##bi ^= input[ 2]; \ 810 } \ 811 } \ 812 } \ 813 else { \ 814 X##ba ^= input[ 0]; \ 815 X##be ^= input[ 1]; \ 816 X##bi ^= input[ 2]; \ 817 X##bo ^= input[ 3]; \ 818 if (laneCount < 6) { \ 819 if (laneCount < 5) { \ 820 } \ 821 else { \ 822 X##bu ^= input[ 4]; \ 823 } \ 824 } \ 825 else { \ 826 X##bu ^= input[ 4]; \ 827 X##ga ^= input[ 5]; \ 828 if (laneCount < 7) { \ 829 } \ 830 else { \ 831 X##ge ^= input[ 6]; \ 832 } \ 833 } \ 834 } \ 835 } \ 836 else { \ 837 X##ba ^= input[ 0]; \ 838 X##be ^= input[ 1]; \ 839 X##bi ^= input[ 2]; \ 840 X##bo ^= input[ 3]; \ 841 X##bu ^= input[ 4]; \ 842 X##ga ^= input[ 5]; \ 843 X##ge ^= input[ 6]; \ 844 X##gi ^= input[ 7]; \ 845 if (laneCount < 12) { \ 846 if (laneCount < 10) { \ 847 if (laneCount < 9) { \ 848 } \ 849 else { \ 850 X##go ^= input[ 8]; \ 851 } \ 852 } \ 853 else { \ 854 X##go ^= input[ 8]; \ 855 X##gu ^= input[ 9]; \ 856 if (laneCount < 11) { \ 857 } \ 858 else { \ 859 X##ka ^= input[10]; \ 860 } \ 861 } \ 862 } \ 863 else { \ 864 X##go ^= input[ 8]; \ 865 X##gu ^= input[ 9]; \ 866 X##ka ^= input[10]; \ 867 X##ke ^= input[11]; \ 868 if (laneCount < 14) { \ 869 if (laneCount < 13) { \ 870 } \ 871 else { \ 872 X##ki ^= input[12]; \ 873 } \ 874 } \ 875 else { \ 876 X##ki ^= input[12]; \ 877 X##ko ^= input[13]; \ 878 if (laneCount < 15) { \ 879 } \ 880 else { \ 881 X##ku ^= input[14]; \ 882 } \ 883 } \ 884 } \ 885 } \ 886 } \ 887 else { \ 888 X##ba ^= input[ 0]; \ 889 X##be ^= input[ 1]; \ 890 X##bi ^= input[ 2]; \ 891 X##bo ^= input[ 3]; \ 892 X##bu ^= input[ 4]; \ 893 X##ga ^= input[ 5]; \ 894 X##ge ^= input[ 6]; \ 895 X##gi ^= input[ 7]; \ 896 X##go ^= input[ 8]; \ 897 X##gu ^= input[ 9]; \ 898 X##ka ^= input[10]; \ 899 X##ke ^= input[11]; \ 900 X##ki ^= input[12]; \ 901 X##ko ^= input[13]; \ 902 X##ku ^= input[14]; \ 903 X##ma ^= input[15]; \ 904 if (laneCount < 24) { \ 905 if (laneCount < 20) { \ 906 if (laneCount < 18) { \ 907 if (laneCount < 17) { \ 908 } \ 909 else { \ 910 X##me ^= input[16]; \ 911 } \ 912 } \ 913 else { \ 914 X##me ^= input[16]; \ 915 X##mi ^= input[17]; \ 916 if (laneCount < 19) { \ 917 } \ 918 else { \ 919 X##mo ^= input[18]; \ 920 } \ 921 } \ 922 } \ 923 else { \ 924 X##me ^= input[16]; \ 925 X##mi ^= input[17]; \ 926 X##mo ^= input[18]; \ 927 X##mu ^= input[19]; \ 928 if (laneCount < 22) { \ 929 if (laneCount < 21) { \ 930 } \ 931 else { \ 932 X##sa ^= input[20]; \ 933 } \ 934 } \ 935 else { \ 936 X##sa ^= input[20]; \ 937 X##se ^= input[21]; \ 938 if (laneCount < 23) { \ 939 } \ 940 else { \ 941 X##si ^= input[22]; \ 942 } \ 943 } \ 944 } \ 945 } \ 946 else { \ 947 X##me ^= input[16]; \ 948 X##mi ^= input[17]; \ 949 X##mo ^= input[18]; \ 950 X##mu ^= input[19]; \ 951 X##sa ^= input[20]; \ 952 X##se ^= input[21]; \ 953 X##si ^= input[22]; \ 954 X##so ^= input[23]; \ 955 if (laneCount < 25) { \ 956 } \ 957 else { \ 958 X##su ^= input[24]; \ 959 } \ 960 } \ 961 } 962 963#ifdef UseBebigokimisa 964 965#define copyToStateAndOutput(X, state, output, laneCount) \ 966 if (laneCount < 16) { \ 967 if (laneCount < 8) { \ 968 if (laneCount < 4) { \ 969 if (laneCount < 2) { \ 970 state[ 0] = X##ba; \ 971 if (laneCount >= 1) { \ 972 output[ 0] = X##ba; \ 973 } \ 974 state[ 1] = X##be; \ 975 state[ 2] = X##bi; \ 976 } \ 977 else { \ 978 state[ 0] = X##ba; \ 979 output[ 0] = X##ba; \ 980 state[ 1] = X##be; \ 981 output[ 1] = ~X##be; \ 982 state[ 2] = X##bi; \ 983 if (laneCount >= 3) { \ 984 output[ 2] = ~X##bi; \ 985 } \ 986 } \ 987 state[ 3] = X##bo; \ 988 state[ 4] = X##bu; \ 989 state[ 5] = X##ga; \ 990 state[ 6] = X##ge; \ 991 } \ 992 else { \ 993 state[ 0] = X##ba; \ 994 output[ 0] = X##ba; \ 995 state[ 1] = X##be; \ 996 output[ 1] = ~X##be; \ 997 state[ 2] = X##bi; \ 998 output[ 2] = ~X##bi; \ 999 state[ 3] = X##bo; \ 1000 output[ 3] = X##bo; \ 1001 if (laneCount < 6) { \ 1002 state[ 4] = X##bu; \ 1003 if (laneCount >= 5) { \ 1004 output[ 4] = X##bu; \ 1005 } \ 1006 state[ 5] = X##ga; \ 1007 state[ 6] = X##ge; \ 1008 } \ 1009 else { \ 1010 state[ 4] = X##bu; \ 1011 output[ 4] = X##bu; \ 1012 state[ 5] = X##ga; \ 1013 output[ 5] = X##ga; \ 1014 state[ 6] = X##ge; \ 1015 if (laneCount >= 7) { \ 1016 output[ 6] = X##ge; \ 1017 } \ 1018 } \ 1019 } \ 1020 state[ 7] = X##gi; \ 1021 state[ 8] = X##go; \ 1022 state[ 9] = X##gu; \ 1023 state[10] = X##ka; \ 1024 state[11] = X##ke; \ 1025 state[12] = X##ki; \ 1026 state[13] = X##ko; \ 1027 state[14] = X##ku; \ 1028 } \ 1029 else { \ 1030 state[ 0] = X##ba; \ 1031 output[ 0] = X##ba; \ 1032 state[ 1] = X##be; \ 1033 output[ 1] = ~X##be; \ 1034 state[ 2] = X##bi; \ 1035 output[ 2] = ~X##bi; \ 1036 state[ 3] = X##bo; \ 1037 output[ 3] = X##bo; \ 1038 state[ 4] = X##bu; \ 1039 output[ 4] = X##bu; \ 1040 state[ 5] = X##ga; \ 1041 output[ 5] = X##ga; \ 1042 state[ 6] = X##ge; \ 1043 output[ 6] = X##ge; \ 1044 state[ 7] = X##gi; \ 1045 output[ 7] = X##gi; \ 1046 if (laneCount < 12) { \ 1047 if (laneCount < 10) { \ 1048 state[ 8] = X##go; \ 1049 if (laneCount >= 9) { \ 1050 output[ 8] = ~X##go; \ 1051 } \ 1052 state[ 9] = X##gu; \ 1053 state[10] = X##ka; \ 1054 } \ 1055 else { \ 1056 state[ 8] = X##go; \ 1057 output[ 8] = ~X##go; \ 1058 state[ 9] = X##gu; \ 1059 output[ 9] = X##gu; \ 1060 state[10] = X##ka; \ 1061 if (laneCount >= 11) { \ 1062 output[10] = X##ka; \ 1063 } \ 1064 } \ 1065 state[11] = X##ke; \ 1066 state[12] = X##ki; \ 1067 state[13] = X##ko; \ 1068 state[14] = X##ku; \ 1069 } \ 1070 else { \ 1071 state[ 8] = X##go; \ 1072 output[ 8] = ~X##go; \ 1073 state[ 9] = X##gu; \ 1074 output[ 9] = X##gu; \ 1075 state[10] = X##ka; \ 1076 output[10] = X##ka; \ 1077 state[11] = X##ke; \ 1078 output[11] = X##ke; \ 1079 if (laneCount < 14) { \ 1080 state[12] = X##ki; \ 1081 if (laneCount >= 13) { \ 1082 output[12] = ~X##ki; \ 1083 } \ 1084 state[13] = X##ko; \ 1085 state[14] = X##ku; \ 1086 } \ 1087 else { \ 1088 state[12] = X##ki; \ 1089 output[12] = ~X##ki; \ 1090 state[13] = X##ko; \ 1091 output[13] = X##ko; \ 1092 state[14] = X##ku; \ 1093 if (laneCount >= 15) { \ 1094 output[14] = X##ku; \ 1095 } \ 1096 } \ 1097 } \ 1098 } \ 1099 state[15] = X##ma; \ 1100 state[16] = X##me; \ 1101 state[17] = X##mi; \ 1102 state[18] = X##mo; \ 1103 state[19] = X##mu; \ 1104 state[20] = X##sa; \ 1105 state[21] = X##se; \ 1106 state[22] = X##si; \ 1107 state[23] = X##so; \ 1108 state[24] = X##su; \ 1109 } \ 1110 else { \ 1111 state[ 0] = X##ba; \ 1112 output[ 0] = X##ba; \ 1113 state[ 1] = X##be; \ 1114 output[ 1] = ~X##be; \ 1115 state[ 2] = X##bi; \ 1116 output[ 2] = ~X##bi; \ 1117 state[ 3] = X##bo; \ 1118 output[ 3] = X##bo; \ 1119 state[ 4] = X##bu; \ 1120 output[ 4] = X##bu; \ 1121 state[ 5] = X##ga; \ 1122 output[ 5] = X##ga; \ 1123 state[ 6] = X##ge; \ 1124 output[ 6] = X##ge; \ 1125 state[ 7] = X##gi; \ 1126 output[ 7] = X##gi; \ 1127 state[ 8] = X##go; \ 1128 output[ 8] = ~X##go; \ 1129 state[ 9] = X##gu; \ 1130 output[ 9] = X##gu; \ 1131 state[10] = X##ka; \ 1132 output[10] = X##ka; \ 1133 state[11] = X##ke; \ 1134 output[11] = X##ke; \ 1135 state[12] = X##ki; \ 1136 output[12] = ~X##ki; \ 1137 state[13] = X##ko; \ 1138 output[13] = X##ko; \ 1139 state[14] = X##ku; \ 1140 output[14] = X##ku; \ 1141 state[15] = X##ma; \ 1142 output[15] = X##ma; \ 1143 if (laneCount < 24) { \ 1144 if (laneCount < 20) { \ 1145 if (laneCount < 18) { \ 1146 state[16] = X##me; \ 1147 if (laneCount >= 17) { \ 1148 output[16] = X##me; \ 1149 } \ 1150 state[17] = X##mi; \ 1151 state[18] = X##mo; \ 1152 } \ 1153 else { \ 1154 state[16] = X##me; \ 1155 output[16] = X##me; \ 1156 state[17] = X##mi; \ 1157 output[17] = ~X##mi; \ 1158 state[18] = X##mo; \ 1159 if (laneCount >= 19) { \ 1160 output[18] = X##mo; \ 1161 } \ 1162 } \ 1163 state[19] = X##mu; \ 1164 state[20] = X##sa; \ 1165 state[21] = X##se; \ 1166 state[22] = X##si; \ 1167 } \ 1168 else { \ 1169 state[16] = X##me; \ 1170 output[16] = X##me; \ 1171 state[17] = X##mi; \ 1172 output[17] = ~X##mi; \ 1173 state[18] = X##mo; \ 1174 output[18] = X##mo; \ 1175 state[19] = X##mu; \ 1176 output[19] = X##mu; \ 1177 if (laneCount < 22) { \ 1178 state[20] = X##sa; \ 1179 if (laneCount >= 21) { \ 1180 output[20] = ~X##sa; \ 1181 } \ 1182 state[21] = X##se; \ 1183 state[22] = X##si; \ 1184 } \ 1185 else { \ 1186 state[20] = X##sa; \ 1187 output[20] = ~X##sa; \ 1188 state[21] = X##se; \ 1189 output[21] = X##se; \ 1190 state[22] = X##si; \ 1191 if (laneCount >= 23) { \ 1192 output[22] = X##si; \ 1193 } \ 1194 } \ 1195 } \ 1196 state[23] = X##so; \ 1197 state[24] = X##su; \ 1198 } \ 1199 else { \ 1200 state[16] = X##me; \ 1201 output[16] = X##me; \ 1202 state[17] = X##mi; \ 1203 output[17] = ~X##mi; \ 1204 state[18] = X##mo; \ 1205 output[18] = X##mo; \ 1206 state[19] = X##mu; \ 1207 output[19] = X##mu; \ 1208 state[20] = X##sa; \ 1209 output[20] = ~X##sa; \ 1210 state[21] = X##se; \ 1211 output[21] = X##se; \ 1212 state[22] = X##si; \ 1213 output[22] = X##si; \ 1214 state[23] = X##so; \ 1215 output[23] = X##so; \ 1216 state[24] = X##su; \ 1217 if (laneCount >= 25) { \ 1218 output[24] = X##su; \ 1219 } \ 1220 } \ 1221 } 1222 1223#define output(X, output, laneCount) \ 1224 if (laneCount < 16) { \ 1225 if (laneCount < 8) { \ 1226 if (laneCount < 4) { \ 1227 if (laneCount < 2) { \ 1228 if (laneCount >= 1) { \ 1229 output[ 0] = X##ba; \ 1230 } \ 1231 } \ 1232 else { \ 1233 output[ 0] = X##ba; \ 1234 output[ 1] = ~X##be; \ 1235 if (laneCount >= 3) { \ 1236 output[ 2] = ~X##bi; \ 1237 } \ 1238 } \ 1239 } \ 1240 else { \ 1241 output[ 0] = X##ba; \ 1242 output[ 1] = ~X##be; \ 1243 output[ 2] = ~X##bi; \ 1244 output[ 3] = X##bo; \ 1245 if (laneCount < 6) { \ 1246 if (laneCount >= 5) { \ 1247 output[ 4] = X##bu; \ 1248 } \ 1249 } \ 1250 else { \ 1251 output[ 4] = X##bu; \ 1252 output[ 5] = X##ga; \ 1253 if (laneCount >= 7) { \ 1254 output[ 6] = X##ge; \ 1255 } \ 1256 } \ 1257 } \ 1258 } \ 1259 else { \ 1260 output[ 0] = X##ba; \ 1261 output[ 1] = ~X##be; \ 1262 output[ 2] = ~X##bi; \ 1263 output[ 3] = X##bo; \ 1264 output[ 4] = X##bu; \ 1265 output[ 5] = X##ga; \ 1266 output[ 6] = X##ge; \ 1267 output[ 7] = X##gi; \ 1268 if (laneCount < 12) { \ 1269 if (laneCount < 10) { \ 1270 if (laneCount >= 9) { \ 1271 output[ 8] = ~X##go; \ 1272 } \ 1273 } \ 1274 else { \ 1275 output[ 8] = ~X##go; \ 1276 output[ 9] = X##gu; \ 1277 if (laneCount >= 11) { \ 1278 output[10] = X##ka; \ 1279 } \ 1280 } \ 1281 } \ 1282 else { \ 1283 output[ 8] = ~X##go; \ 1284 output[ 9] = X##gu; \ 1285 output[10] = X##ka; \ 1286 output[11] = X##ke; \ 1287 if (laneCount < 14) { \ 1288 if (laneCount >= 13) { \ 1289 output[12] = ~X##ki; \ 1290 } \ 1291 } \ 1292 else { \ 1293 output[12] = ~X##ki; \ 1294 output[13] = X##ko; \ 1295 if (laneCount >= 15) { \ 1296 output[14] = X##ku; \ 1297 } \ 1298 } \ 1299 } \ 1300 } \ 1301 } \ 1302 else { \ 1303 output[ 0] = X##ba; \ 1304 output[ 1] = ~X##be; \ 1305 output[ 2] = ~X##bi; \ 1306 output[ 3] = X##bo; \ 1307 output[ 4] = X##bu; \ 1308 output[ 5] = X##ga; \ 1309 output[ 6] = X##ge; \ 1310 output[ 7] = X##gi; \ 1311 output[ 8] = ~X##go; \ 1312 output[ 9] = X##gu; \ 1313 output[10] = X##ka; \ 1314 output[11] = X##ke; \ 1315 output[12] = ~X##ki; \ 1316 output[13] = X##ko; \ 1317 output[14] = X##ku; \ 1318 output[15] = X##ma; \ 1319 if (laneCount < 24) { \ 1320 if (laneCount < 20) { \ 1321 if (laneCount < 18) { \ 1322 if (laneCount >= 17) { \ 1323 output[16] = X##me; \ 1324 } \ 1325 } \ 1326 else { \ 1327 output[16] = X##me; \ 1328 output[17] = ~X##mi; \ 1329 if (laneCount >= 19) { \ 1330 output[18] = X##mo; \ 1331 } \ 1332 } \ 1333 } \ 1334 else { \ 1335 output[16] = X##me; \ 1336 output[17] = ~X##mi; \ 1337 output[18] = X##mo; \ 1338 output[19] = X##mu; \ 1339 if (laneCount < 22) { \ 1340 if (laneCount >= 21) { \ 1341 output[20] = ~X##sa; \ 1342 } \ 1343 } \ 1344 else { \ 1345 output[20] = ~X##sa; \ 1346 output[21] = X##se; \ 1347 if (laneCount >= 23) { \ 1348 output[22] = X##si; \ 1349 } \ 1350 } \ 1351 } \ 1352 } \ 1353 else { \ 1354 output[16] = X##me; \ 1355 output[17] = ~X##mi; \ 1356 output[18] = X##mo; \ 1357 output[19] = X##mu; \ 1358 output[20] = ~X##sa; \ 1359 output[21] = X##se; \ 1360 output[22] = X##si; \ 1361 output[23] = X##so; \ 1362 if (laneCount >= 25) { \ 1363 output[24] = X##su; \ 1364 } \ 1365 } \ 1366 } 1367 1368#define wrapOne(X, input, output, index, name) \ 1369 X##name ^= input[index]; \ 1370 output[index] = X##name; 1371 1372#define wrapOneInvert(X, input, output, index, name) \ 1373 X##name ^= input[index]; \ 1374 output[index] = ~X##name; 1375 1376#define unwrapOne(X, input, output, index, name) \ 1377 output[index] = input[index] ^ X##name; \ 1378 X##name ^= output[index]; 1379 1380#define unwrapOneInvert(X, input, output, index, name) \ 1381 output[index] = ~(input[index] ^ X##name); \ 1382 X##name ^= output[index]; \ 1383 1384#else /* UseBebigokimisa */ 1385 1386#define copyToStateAndOutput(X, state, output, laneCount) \ 1387 if (laneCount < 16) { \ 1388 if (laneCount < 8) { \ 1389 if (laneCount < 4) { \ 1390 if (laneCount < 2) { \ 1391 state[ 0] = X##ba; \ 1392 if (laneCount >= 1) { \ 1393 output[ 0] = X##ba; \ 1394 } \ 1395 state[ 1] = X##be; \ 1396 state[ 2] = X##bi; \ 1397 } \ 1398 else { \ 1399 state[ 0] = X##ba; \ 1400 output[ 0] = X##ba; \ 1401 state[ 1] = X##be; \ 1402 output[ 1] = X##be; \ 1403 state[ 2] = X##bi; \ 1404 if (laneCount >= 3) { \ 1405 output[ 2] = X##bi; \ 1406 } \ 1407 } \ 1408 state[ 3] = X##bo; \ 1409 state[ 4] = X##bu; \ 1410 state[ 5] = X##ga; \ 1411 state[ 6] = X##ge; \ 1412 } \ 1413 else { \ 1414 state[ 0] = X##ba; \ 1415 output[ 0] = X##ba; \ 1416 state[ 1] = X##be; \ 1417 output[ 1] = X##be; \ 1418 state[ 2] = X##bi; \ 1419 output[ 2] = X##bi; \ 1420 state[ 3] = X##bo; \ 1421 output[ 3] = X##bo; \ 1422 if (laneCount < 6) { \ 1423 state[ 4] = X##bu; \ 1424 if (laneCount >= 5) { \ 1425 output[ 4] = X##bu; \ 1426 } \ 1427 state[ 5] = X##ga; \ 1428 state[ 6] = X##ge; \ 1429 } \ 1430 else { \ 1431 state[ 4] = X##bu; \ 1432 output[ 4] = X##bu; \ 1433 state[ 5] = X##ga; \ 1434 output[ 5] = X##ga; \ 1435 state[ 6] = X##ge; \ 1436 if (laneCount >= 7) { \ 1437 output[ 6] = X##ge; \ 1438 } \ 1439 } \ 1440 } \ 1441 state[ 7] = X##gi; \ 1442 state[ 8] = X##go; \ 1443 state[ 9] = X##gu; \ 1444 state[10] = X##ka; \ 1445 state[11] = X##ke; \ 1446 state[12] = X##ki; \ 1447 state[13] = X##ko; \ 1448 state[14] = X##ku; \ 1449 } \ 1450 else { \ 1451 state[ 0] = X##ba; \ 1452 output[ 0] = X##ba; \ 1453 state[ 1] = X##be; \ 1454 output[ 1] = X##be; \ 1455 state[ 2] = X##bi; \ 1456 output[ 2] = X##bi; \ 1457 state[ 3] = X##bo; \ 1458 output[ 3] = X##bo; \ 1459 state[ 4] = X##bu; \ 1460 output[ 4] = X##bu; \ 1461 state[ 5] = X##ga; \ 1462 output[ 5] = X##ga; \ 1463 state[ 6] = X##ge; \ 1464 output[ 6] = X##ge; \ 1465 state[ 7] = X##gi; \ 1466 output[ 7] = X##gi; \ 1467 if (laneCount < 12) { \ 1468 if (laneCount < 10) { \ 1469 state[ 8] = X##go; \ 1470 if (laneCount >= 9) { \ 1471 output[ 8] = X##go; \ 1472 } \ 1473 state[ 9] = X##gu; \ 1474 state[10] = X##ka; \ 1475 } \ 1476 else { \ 1477 state[ 8] = X##go; \ 1478 output[ 8] = X##go; \ 1479 state[ 9] = X##gu; \ 1480 output[ 9] = X##gu; \ 1481 state[10] = X##ka; \ 1482 if (laneCount >= 11) { \ 1483 output[10] = X##ka; \ 1484 } \ 1485 } \ 1486 state[11] = X##ke; \ 1487 state[12] = X##ki; \ 1488 state[13] = X##ko; \ 1489 state[14] = X##ku; \ 1490 } \ 1491 else { \ 1492 state[ 8] = X##go; \ 1493 output[ 8] = X##go; \ 1494 state[ 9] = X##gu; \ 1495 output[ 9] = X##gu; \ 1496 state[10] = X##ka; \ 1497 output[10] = X##ka; \ 1498 state[11] = X##ke; \ 1499 output[11] = X##ke; \ 1500 if (laneCount < 14) { \ 1501 state[12] = X##ki; \ 1502 if (laneCount >= 13) { \ 1503 output[12]= X##ki; \ 1504 } \ 1505 state[13] = X##ko; \ 1506 state[14] = X##ku; \ 1507 } \ 1508 else { \ 1509 state[12] = X##ki; \ 1510 output[12]= X##ki; \ 1511 state[13] = X##ko; \ 1512 output[13] = X##ko; \ 1513 state[14] = X##ku; \ 1514 if (laneCount >= 15) { \ 1515 output[14] = X##ku; \ 1516 } \ 1517 } \ 1518 } \ 1519 } \ 1520 state[15] = X##ma; \ 1521 state[16] = X##me; \ 1522 state[17] = X##mi; \ 1523 state[18] = X##mo; \ 1524 state[19] = X##mu; \ 1525 state[20] = X##sa; \ 1526 state[21] = X##se; \ 1527 state[22] = X##si; \ 1528 state[23] = X##so; \ 1529 state[24] = X##su; \ 1530 } \ 1531 else { \ 1532 state[ 0] = X##ba; \ 1533 output[ 0] = X##ba; \ 1534 state[ 1] = X##be; \ 1535 output[ 1] = X##be; \ 1536 state[ 2] = X##bi; \ 1537 output[ 2] = X##bi; \ 1538 state[ 3] = X##bo; \ 1539 output[ 3] = X##bo; \ 1540 state[ 4] = X##bu; \ 1541 output[ 4] = X##bu; \ 1542 state[ 5] = X##ga; \ 1543 output[ 5] = X##ga; \ 1544 state[ 6] = X##ge; \ 1545 output[ 6] = X##ge; \ 1546 state[ 7] = X##gi; \ 1547 output[ 7] = X##gi; \ 1548 state[ 8] = X##go; \ 1549 output[ 8] = X##go; \ 1550 state[ 9] = X##gu; \ 1551 output[ 9] = X##gu; \ 1552 state[10] = X##ka; \ 1553 output[10] = X##ka; \ 1554 state[11] = X##ke; \ 1555 output[11] = X##ke; \ 1556 state[12] = X##ki; \ 1557 output[12]= X##ki; \ 1558 state[13] = X##ko; \ 1559 output[13] = X##ko; \ 1560 state[14] = X##ku; \ 1561 output[14] = X##ku; \ 1562 state[15] = X##ma; \ 1563 output[15] = X##ma; \ 1564 if (laneCount < 24) { \ 1565 if (laneCount < 20) { \ 1566 if (laneCount < 18) { \ 1567 state[16] = X##me; \ 1568 if (laneCount >= 17) { \ 1569 output[16] = X##me; \ 1570 } \ 1571 state[17] = X##mi; \ 1572 state[18] = X##mo; \ 1573 } \ 1574 else { \ 1575 state[16] = X##me; \ 1576 output[16] = X##me; \ 1577 state[17] = X##mi; \ 1578 output[17] = X##mi; \ 1579 state[18] = X##mo; \ 1580 if (laneCount >= 19) { \ 1581 output[18] = X##mo; \ 1582 } \ 1583 } \ 1584 state[19] = X##mu; \ 1585 state[20] = X##sa; \ 1586 state[21] = X##se; \ 1587 state[22] = X##si; \ 1588 } \ 1589 else { \ 1590 state[16] = X##me; \ 1591 output[16] = X##me; \ 1592 state[17] = X##mi; \ 1593 output[17] = X##mi; \ 1594 state[18] = X##mo; \ 1595 output[18] = X##mo; \ 1596 state[19] = X##mu; \ 1597 output[19] = X##mu; \ 1598 if (laneCount < 22) { \ 1599 state[20] = X##sa; \ 1600 if (laneCount >= 21) { \ 1601 output[20] = X##sa; \ 1602 } \ 1603 state[21] = X##se; \ 1604 state[22] = X##si; \ 1605 } \ 1606 else { \ 1607 state[20] = X##sa; \ 1608 output[20] = X##sa; \ 1609 state[21] = X##se; \ 1610 output[21] = X##se; \ 1611 state[22] = X##si; \ 1612 if (laneCount >= 23) { \ 1613 output[22] = X##si; \ 1614 } \ 1615 } \ 1616 } \ 1617 state[23] = X##so; \ 1618 state[24] = X##su; \ 1619 } \ 1620 else { \ 1621 state[16] = X##me; \ 1622 output[16] = X##me; \ 1623 state[17] = X##mi; \ 1624 output[17] = X##mi; \ 1625 state[18] = X##mo; \ 1626 output[18] = X##mo; \ 1627 state[19] = X##mu; \ 1628 output[19] = X##mu; \ 1629 state[20] = X##sa; \ 1630 output[20] = X##sa; \ 1631 state[21] = X##se; \ 1632 output[21] = X##se; \ 1633 state[22] = X##si; \ 1634 output[22] = X##si; \ 1635 state[23] = X##so; \ 1636 output[23] = X##so; \ 1637 state[24] = X##su; \ 1638 if (laneCount >= 25) { \ 1639 output[24] = X##su; \ 1640 } \ 1641 } \ 1642 } 1643 1644#define output(X, output, laneCount) \ 1645 if (laneCount < 16) { \ 1646 if (laneCount < 8) { \ 1647 if (laneCount < 4) { \ 1648 if (laneCount < 2) { \ 1649 if (laneCount >= 1) { \ 1650 output[ 0] = X##ba; \ 1651 } \ 1652 } \ 1653 else { \ 1654 output[ 0] = X##ba; \ 1655 output[ 1] = X##be; \ 1656 if (laneCount >= 3) { \ 1657 output[ 2] = X##bi; \ 1658 } \ 1659 } \ 1660 } \ 1661 else { \ 1662 output[ 0] = X##ba; \ 1663 output[ 1] = X##be; \ 1664 output[ 2] = X##bi; \ 1665 output[ 3] = X##bo; \ 1666 if (laneCount < 6) { \ 1667 if (laneCount >= 5) { \ 1668 output[ 4] = X##bu; \ 1669 } \ 1670 } \ 1671 else { \ 1672 output[ 4] = X##bu; \ 1673 output[ 5] = X##ga; \ 1674 if (laneCount >= 7) { \ 1675 output[ 6] = X##ge; \ 1676 } \ 1677 } \ 1678 } \ 1679 } \ 1680 else { \ 1681 output[ 0] = X##ba; \ 1682 output[ 1] = X##be; \ 1683 output[ 2] = X##bi; \ 1684 output[ 3] = X##bo; \ 1685 output[ 4] = X##bu; \ 1686 output[ 5] = X##ga; \ 1687 output[ 6] = X##ge; \ 1688 output[ 7] = X##gi; \ 1689 if (laneCount < 12) { \ 1690 if (laneCount < 10) { \ 1691 if (laneCount >= 9) { \ 1692 output[ 8] = X##go; \ 1693 } \ 1694 } \ 1695 else { \ 1696 output[ 8] = X##go; \ 1697 output[ 9] = X##gu; \ 1698 if (laneCount >= 11) { \ 1699 output[10] = X##ka; \ 1700 } \ 1701 } \ 1702 } \ 1703 else { \ 1704 output[ 8] = X##go; \ 1705 output[ 9] = X##gu; \ 1706 output[10] = X##ka; \ 1707 output[11] = X##ke; \ 1708 if (laneCount < 14) { \ 1709 if (laneCount >= 13) { \ 1710 output[12] = X##ki; \ 1711 } \ 1712 } \ 1713 else { \ 1714 output[12] = X##ki; \ 1715 output[13] = X##ko; \ 1716 if (laneCount >= 15) { \ 1717 output[14] = X##ku; \ 1718 } \ 1719 } \ 1720 } \ 1721 } \ 1722 } \ 1723 else { \ 1724 output[ 0] = X##ba; \ 1725 output[ 1] = X##be; \ 1726 output[ 2] = X##bi; \ 1727 output[ 3] = X##bo; \ 1728 output[ 4] = X##bu; \ 1729 output[ 5] = X##ga; \ 1730 output[ 6] = X##ge; \ 1731 output[ 7] = X##gi; \ 1732 output[ 8] = X##go; \ 1733 output[ 9] = X##gu; \ 1734 output[10] = X##ka; \ 1735 output[11] = X##ke; \ 1736 output[12] = X##ki; \ 1737 output[13] = X##ko; \ 1738 output[14] = X##ku; \ 1739 output[15] = X##ma; \ 1740 if (laneCount < 24) { \ 1741 if (laneCount < 20) { \ 1742 if (laneCount < 18) { \ 1743 if (laneCount >= 17) { \ 1744 output[16] = X##me; \ 1745 } \ 1746 } \ 1747 else { \ 1748 output[16] = X##me; \ 1749 output[17] = X##mi; \ 1750 if (laneCount >= 19) { \ 1751 output[18] = X##mo; \ 1752 } \ 1753 } \ 1754 } \ 1755 else { \ 1756 output[16] = X##me; \ 1757 output[17] = X##mi; \ 1758 output[18] = X##mo; \ 1759 output[19] = X##mu; \ 1760 if (laneCount < 22) { \ 1761 if (laneCount >= 21) { \ 1762 output[20] = X##sa; \ 1763 } \ 1764 } \ 1765 else { \ 1766 output[20] = X##sa; \ 1767 output[21] = X##se; \ 1768 if (laneCount >= 23) { \ 1769 output[22] = X##si; \ 1770 } \ 1771 } \ 1772 } \ 1773 } \ 1774 else { \ 1775 output[16] = X##me; \ 1776 output[17] = X##mi; \ 1777 output[18] = X##mo; \ 1778 output[19] = X##mu; \ 1779 output[20] = X##sa; \ 1780 output[21] = X##se; \ 1781 output[22] = X##si; \ 1782 output[23] = X##so; \ 1783 if (laneCount >= 25) { \ 1784 output[24] = X##su; \ 1785 } \ 1786 } \ 1787 } 1788 1789#define wrapOne(X, input, output, index, name) \ 1790 X##name ^= input[index]; \ 1791 output[index] = X##name; 1792 1793#define wrapOneInvert(X, input, output, index, name) \ 1794 X##name ^= input[index]; \ 1795 output[index] = X##name; 1796 1797#define unwrapOne(X, input, output, index, name) \ 1798 output[index] = input[index] ^ X##name; \ 1799 X##name ^= output[index]; 1800 1801#define unwrapOneInvert(X, input, output, index, name) \ 1802 output[index] = input[index] ^ X##name; \ 1803 X##name ^= output[index]; 1804 1805#endif 1806 1807#define wrap(X, input, output, laneCount, trailingBits) \ 1808 if (laneCount < 16) { \ 1809 if (laneCount < 8) { \ 1810 if (laneCount < 4) { \ 1811 if (laneCount < 2) { \ 1812 if (laneCount < 1) { \ 1813 X##ba ^= trailingBits; \ 1814 } \ 1815 else { \ 1816 wrapOne(X, input, output, 0, ba) \ 1817 X##be ^= trailingBits; \ 1818 } \ 1819 } \ 1820 else { \ 1821 wrapOne(X, input, output, 0, ba) \ 1822 wrapOneInvert(X, input, output, 1, be) \ 1823 if (laneCount < 3) { \ 1824 X##bi ^= trailingBits; \ 1825 } \ 1826 else { \ 1827 wrapOneInvert(X, input, output, 2, bi) \ 1828 X##bo ^= trailingBits; \ 1829 } \ 1830 } \ 1831 } \ 1832 else { \ 1833 wrapOne(X, input, output, 0, ba) \ 1834 wrapOneInvert(X, input, output, 1, be) \ 1835 wrapOneInvert(X, input, output, 2, bi) \ 1836 wrapOne(X, input, output, 3, bo) \ 1837 if (laneCount < 6) { \ 1838 if (laneCount < 5) { \ 1839 X##bu ^= trailingBits; \ 1840 } \ 1841 else { \ 1842 wrapOne(X, input, output, 4, bu) \ 1843 X##ga ^= trailingBits; \ 1844 } \ 1845 } \ 1846 else { \ 1847 wrapOne(X, input, output, 4, bu) \ 1848 wrapOne(X, input, output, 5, ga) \ 1849 if (laneCount < 7) { \ 1850 X##ge ^= trailingBits; \ 1851 } \ 1852 else { \ 1853 wrapOne(X, input, output, 6, ge) \ 1854 X##gi ^= trailingBits; \ 1855 } \ 1856 } \ 1857 } \ 1858 } \ 1859 else { \ 1860 wrapOne(X, input, output, 0, ba) \ 1861 wrapOneInvert(X, input, output, 1, be) \ 1862 wrapOneInvert(X, input, output, 2, bi) \ 1863 wrapOne(X, input, output, 3, bo) \ 1864 wrapOne(X, input, output, 4, bu) \ 1865 wrapOne(X, input, output, 5, ga) \ 1866 wrapOne(X, input, output, 6, ge) \ 1867 wrapOne(X, input, output, 7, gi) \ 1868 if (laneCount < 12) { \ 1869 if (laneCount < 10) { \ 1870 if (laneCount < 9) { \ 1871 X##go ^= trailingBits; \ 1872 } \ 1873 else { \ 1874 wrapOneInvert(X, input, output, 8, go) \ 1875 X##gu ^= trailingBits; \ 1876 } \ 1877 } \ 1878 else { \ 1879 wrapOneInvert(X, input, output, 8, go) \ 1880 wrapOne(X, input, output, 9, gu) \ 1881 if (laneCount < 11) { \ 1882 X##ka ^= trailingBits; \ 1883 } \ 1884 else { \ 1885 wrapOne(X, input, output, 10, ka) \ 1886 X##ke ^= trailingBits; \ 1887 } \ 1888 } \ 1889 } \ 1890 else { \ 1891 wrapOneInvert(X, input, output, 8, go) \ 1892 wrapOne(X, input, output, 9, gu) \ 1893 wrapOne(X, input, output, 10, ka) \ 1894 wrapOne(X, input, output, 11, ke) \ 1895 if (laneCount < 14) { \ 1896 if (laneCount < 13) { \ 1897 X##ki ^= trailingBits; \ 1898 } \ 1899 else { \ 1900 wrapOneInvert(X, input, output, 12, ki) \ 1901 X##ko ^= trailingBits; \ 1902 } \ 1903 } \ 1904 else { \ 1905 wrapOneInvert(X, input, output, 12, ki) \ 1906 wrapOne(X, input, output, 13, ko) \ 1907 if (laneCount < 15) { \ 1908 X##ku ^= trailingBits; \ 1909 } \ 1910 else { \ 1911 wrapOne(X, input, output, 14, ku) \ 1912 X##ma ^= trailingBits; \ 1913 } \ 1914 } \ 1915 } \ 1916 } \ 1917 } \ 1918 else { \ 1919 wrapOne(X, input, output, 0, ba) \ 1920 wrapOneInvert(X, input, output, 1, be) \ 1921 wrapOneInvert(X, input, output, 2, bi) \ 1922 wrapOne(X, input, output, 3, bo) \ 1923 wrapOne(X, input, output, 4, bu) \ 1924 wrapOne(X, input, output, 5, ga) \ 1925 wrapOne(X, input, output, 6, ge) \ 1926 wrapOne(X, input, output, 7, gi) \ 1927 wrapOneInvert(X, input, output, 8, go) \ 1928 wrapOne(X, input, output, 9, gu) \ 1929 wrapOne(X, input, output, 10, ka) \ 1930 wrapOne(X, input, output, 11, ke) \ 1931 wrapOneInvert(X, input, output, 12, ki) \ 1932 wrapOne(X, input, output, 13, ko) \ 1933 wrapOne(X, input, output, 14, ku) \ 1934 wrapOne(X, input, output, 15, ma) \ 1935 if (laneCount < 24) { \ 1936 if (laneCount < 20) { \ 1937 if (laneCount < 18) { \ 1938 if (laneCount < 17) { \ 1939 X##me ^= trailingBits; \ 1940 } \ 1941 else { \ 1942 wrapOne(X, input, output, 16, me) \ 1943 X##mi ^= trailingBits; \ 1944 } \ 1945 } \ 1946 else { \ 1947 wrapOne(X, input, output, 16, me) \ 1948 wrapOneInvert(X, input, output, 17, mi) \ 1949 if (laneCount < 19) { \ 1950 X##mo ^= trailingBits; \ 1951 } \ 1952 else { \ 1953 wrapOne(X, input, output, 18, mo) \ 1954 X##mu ^= trailingBits; \ 1955 } \ 1956 } \ 1957 } \ 1958 else { \ 1959 wrapOne(X, input, output, 16, me) \ 1960 wrapOneInvert(X, input, output, 17, mi) \ 1961 wrapOne(X, input, output, 18, mo) \ 1962 wrapOne(X, input, output, 19, mu) \ 1963 if (laneCount < 22) { \ 1964 if (laneCount < 21) { \ 1965 X##sa ^= trailingBits; \ 1966 } \ 1967 else { \ 1968 wrapOneInvert(X, input, output, 20, sa) \ 1969 X##se ^= trailingBits; \ 1970 } \ 1971 } \ 1972 else { \ 1973 wrapOneInvert(X, input, output, 20, sa) \ 1974 wrapOne(X, input, output, 21, se) \ 1975 if (laneCount < 23) { \ 1976 X##si ^= trailingBits; \ 1977 } \ 1978 else { \ 1979 wrapOne(X, input, output, 22, si) \ 1980 X##so ^= trailingBits; \ 1981 } \ 1982 } \ 1983 } \ 1984 } \ 1985 else { \ 1986 wrapOne(X, input, output, 16, me) \ 1987 wrapOneInvert(X, input, output, 17, mi) \ 1988 wrapOne(X, input, output, 18, mo) \ 1989 wrapOne(X, input, output, 19, mu) \ 1990 wrapOneInvert(X, input, output, 20, sa) \ 1991 wrapOne(X, input, output, 21, se) \ 1992 wrapOne(X, input, output, 22, si) \ 1993 wrapOne(X, input, output, 23, so) \ 1994 if (laneCount < 25) { \ 1995 X##su ^= trailingBits; \ 1996 } \ 1997 else { \ 1998 wrapOne(X, input, output, 24, su) \ 1999 } \ 2000 } \ 2001 } 2002 2003#define unwrap(X, input, output, laneCount, trailingBits) \ 2004 if (laneCount < 16) { \ 2005 if (laneCount < 8) { \ 2006 if (laneCount < 4) { \ 2007 if (laneCount < 2) { \ 2008 if (laneCount < 1) { \ 2009 X##ba ^= trailingBits; \ 2010 } \ 2011 else { \ 2012 unwrapOne(X, input, output, 0, ba) \ 2013 X##be ^= trailingBits; \ 2014 } \ 2015 } \ 2016 else { \ 2017 unwrapOne(X, input, output, 0, ba) \ 2018 unwrapOneInvert(X, input, output, 1, be) \ 2019 if (laneCount < 3) { \ 2020 X##bi ^= trailingBits; \ 2021 } \ 2022 else { \ 2023 unwrapOneInvert(X, input, output, 2, bi) \ 2024 X##bo ^= trailingBits; \ 2025 } \ 2026 } \ 2027 } \ 2028 else { \ 2029 unwrapOne(X, input, output, 0, ba) \ 2030 unwrapOneInvert(X, input, output, 1, be) \ 2031 unwrapOneInvert(X, input, output, 2, bi) \ 2032 unwrapOne(X, input, output, 3, bo) \ 2033 if (laneCount < 6) { \ 2034 if (laneCount < 5) { \ 2035 X##bu ^= trailingBits; \ 2036 } \ 2037 else { \ 2038 unwrapOne(X, input, output, 4, bu) \ 2039 X##ga ^= trailingBits; \ 2040 } \ 2041 } \ 2042 else { \ 2043 unwrapOne(X, input, output, 4, bu) \ 2044 unwrapOne(X, input, output, 5, ga) \ 2045 if (laneCount < 7) { \ 2046 X##ge ^= trailingBits; \ 2047 } \ 2048 else { \ 2049 unwrapOne(X, input, output, 6, ge) \ 2050 X##gi ^= trailingBits; \ 2051 } \ 2052 } \ 2053 } \ 2054 } \ 2055 else { \ 2056 unwrapOne(X, input, output, 0, ba) \ 2057 unwrapOneInvert(X, input, output, 1, be) \ 2058 unwrapOneInvert(X, input, output, 2, bi) \ 2059 unwrapOne(X, input, output, 3, bo) \ 2060 unwrapOne(X, input, output, 4, bu) \ 2061 unwrapOne(X, input, output, 5, ga) \ 2062 unwrapOne(X, input, output, 6, ge) \ 2063 unwrapOne(X, input, output, 7, gi) \ 2064 if (laneCount < 12) { \ 2065 if (laneCount < 10) { \ 2066 if (laneCount < 9) { \ 2067 X##go ^= trailingBits; \ 2068 } \ 2069 else { \ 2070 unwrapOneInvert(X, input, output, 8, go) \ 2071 X##gu ^= trailingBits; \ 2072 } \ 2073 } \ 2074 else { \ 2075 unwrapOneInvert(X, input, output, 8, go) \ 2076 unwrapOne(X, input, output, 9, gu) \ 2077 if (laneCount < 11) { \ 2078 X##ka ^= trailingBits; \ 2079 } \ 2080 else { \ 2081 unwrapOne(X, input, output, 10, ka) \ 2082 X##ke ^= trailingBits; \ 2083 } \ 2084 } \ 2085 } \ 2086 else { \ 2087 unwrapOneInvert(X, input, output, 8, go) \ 2088 unwrapOne(X, input, output, 9, gu) \ 2089 unwrapOne(X, input, output, 10, ka) \ 2090 unwrapOne(X, input, output, 11, ke) \ 2091 if (laneCount < 14) { \ 2092 if (laneCount < 13) { \ 2093 X##ki ^= trailingBits; \ 2094 } \ 2095 else { \ 2096 unwrapOneInvert(X, input, output, 12, ki) \ 2097 X##ko ^= trailingBits; \ 2098 } \ 2099 } \ 2100 else { \ 2101 unwrapOneInvert(X, input, output, 12, ki) \ 2102 unwrapOne(X, input, output, 13, ko) \ 2103 if (laneCount < 15) { \ 2104 X##ku ^= trailingBits; \ 2105 } \ 2106 else { \ 2107 unwrapOne(X, input, output, 14, ku) \ 2108 X##ma ^= trailingBits; \ 2109 } \ 2110 } \ 2111 } \ 2112 } \ 2113 } \ 2114 else { \ 2115 unwrapOne(X, input, output, 0, ba) \ 2116 unwrapOneInvert(X, input, output, 1, be) \ 2117 unwrapOneInvert(X, input, output, 2, bi) \ 2118 unwrapOne(X, input, output, 3, bo) \ 2119 unwrapOne(X, input, output, 4, bu) \ 2120 unwrapOne(X, input, output, 5, ga) \ 2121 unwrapOne(X, input, output, 6, ge) \ 2122 unwrapOne(X, input, output, 7, gi) \ 2123 unwrapOneInvert(X, input, output, 8, go) \ 2124 unwrapOne(X, input, output, 9, gu) \ 2125 unwrapOne(X, input, output, 10, ka) \ 2126 unwrapOne(X, input, output, 11, ke) \ 2127 unwrapOneInvert(X, input, output, 12, ki) \ 2128 unwrapOne(X, input, output, 13, ko) \ 2129 unwrapOne(X, input, output, 14, ku) \ 2130 unwrapOne(X, input, output, 15, ma) \ 2131 if (laneCount < 24) { \ 2132 if (laneCount < 20) { \ 2133 if (laneCount < 18) { \ 2134 if (laneCount < 17) { \ 2135 X##me ^= trailingBits; \ 2136 } \ 2137 else { \ 2138 unwrapOne(X, input, output, 16, me) \ 2139 X##mi ^= trailingBits; \ 2140 } \ 2141 } \ 2142 else { \ 2143 unwrapOne(X, input, output, 16, me) \ 2144 unwrapOneInvert(X, input, output, 17, mi) \ 2145 if (laneCount < 19) { \ 2146 X##mo ^= trailingBits; \ 2147 } \ 2148 else { \ 2149 unwrapOne(X, input, output, 18, mo) \ 2150 X##mu ^= trailingBits; \ 2151 } \ 2152 } \ 2153 } \ 2154 else { \ 2155 unwrapOne(X, input, output, 16, me) \ 2156 unwrapOneInvert(X, input, output, 17, mi) \ 2157 unwrapOne(X, input, output, 18, mo) \ 2158 unwrapOne(X, input, output, 19, mu) \ 2159 if (laneCount < 22) { \ 2160 if (laneCount < 21) { \ 2161 X##sa ^= trailingBits; \ 2162 } \ 2163 else { \ 2164 unwrapOneInvert(X, input, output, 20, sa) \ 2165 X##se ^= trailingBits; \ 2166 } \ 2167 } \ 2168 else { \ 2169 unwrapOneInvert(X, input, output, 20, sa) \ 2170 unwrapOne(X, input, output, 21, se) \ 2171 if (laneCount < 23) { \ 2172 X##si ^= trailingBits; \ 2173 } \ 2174 else { \ 2175 unwrapOne(X, input, output, 22, si) \ 2176 X##so ^= trailingBits; \ 2177 } \ 2178 } \ 2179 } \ 2180 } \ 2181 else { \ 2182 unwrapOne(X, input, output, 16, me) \ 2183 unwrapOneInvert(X, input, output, 17, mi) \ 2184 unwrapOne(X, input, output, 18, mo) \ 2185 unwrapOne(X, input, output, 19, mu) \ 2186 unwrapOneInvert(X, input, output, 20, sa) \ 2187 unwrapOne(X, input, output, 21, se) \ 2188 unwrapOne(X, input, output, 22, si) \ 2189 unwrapOne(X, input, output, 23, so) \ 2190 if (laneCount < 25) { \ 2191 X##su ^= trailingBits; \ 2192 } \ 2193 else { \ 2194 unwrapOne(X, input, output, 24, su) \ 2195 } \ 2196 } \ 2197 } 2198
