internal-spongent.c 16 KB
Newer Older
Rhys Weatherley committed
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
/*
 * Copyright (C) 2020 Southern Storm Software, Pty Ltd.
 *
 * Permission is hereby granted, free of charge, to any person obtaining a
 * copy of this software and associated documentation files (the "Software"),
 * to deal in the Software without restriction, including without limitation
 * the rights to use, copy, modify, merge, publish, distribute, sublicense,
 * and/or sell copies of the Software, and to permit persons to whom the
 * Software is furnished to do so, subject to the following conditions:
 *
 * The above copyright notice and this permission notice shall be included
 * in all copies or substantial portions of the Software.
 *
 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS
 * OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
 * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
 * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
 * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
 * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
 * DEALINGS IN THE SOFTWARE.
 */

#include "internal-spongent.h"

25 26
#if !defined(__AVR__)

Rhys Weatherley committed
27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348
/**
 * \brief Applies the Spongent-pi S-box in parallel to the 8 nibbles
 * of a 32-bit word.
 *
 * \param x3 The input values to the parallel S-boxes.
 *
 * \return The output values from the parallel S-boxes.
 *
 * Based on the bit-sliced S-box implementation from here:
 * https://github.com/DadaIsCrazy/usuba/blob/master/data/sboxes/spongent.ua
 *
 * Note that spongent.ua numbers bits from highest to lowest, so x0 is the
 * high bit of each nibble and x3 is the low bit.
 */
static uint32_t spongent_sbox(uint32_t x3)
{
    uint32_t q0, q1, q2, q3, t0, t1, t2, t3;
    uint32_t x2 = (x3 >> 1);
    uint32_t x1 = (x2 >> 1);
    uint32_t x0 = (x1 >> 1);
    q0 = x0 ^ x2;
    q1 = x1 ^ x2;
    t0 = q0 & q1;
    q2 = ~(x0 ^ x1 ^ x3 ^ t0);
    t1 = q2 & ~x0;
    q3 = x1 ^ t1;
    t2 = q3 & (q3 ^ x2 ^ x3 ^ t0);
    t3 = (x2 ^ t0) & ~(x1 ^ t0);
    q0 = x1 ^ x2 ^ x3 ^ t2;
    q1 = x0 ^ x2 ^ x3 ^ t0 ^ t1;
    q2 = x0 ^ x1 ^ x2 ^ t1;
    q3 = x0 ^ x3 ^ t0 ^ t3;
    return ((q0 << 3) & 0x88888888U) | ((q1 << 2) & 0x44444444U) |
           ((q2 << 1) & 0x22222222U) |  (q3       & 0x11111111U);
}

void spongent160_permute(spongent160_state_t *state)
{
    static uint8_t const RC[] = {
        /* Round constants for Spongent-pi[160] */
        0x75, 0xae, 0x6a, 0x56, 0x54, 0x2a, 0x29, 0x94,
        0x53, 0xca, 0x27, 0xe4, 0x4f, 0xf2, 0x1f, 0xf8,
        0x3e, 0x7c, 0x7d, 0xbe, 0x7a, 0x5e, 0x74, 0x2e,
        0x68, 0x16, 0x50, 0x0a, 0x21, 0x84, 0x43, 0xc2,
        0x07, 0xe0, 0x0e, 0x70, 0x1c, 0x38, 0x38, 0x1c,
        0x71, 0x8e, 0x62, 0x46, 0x44, 0x22, 0x09, 0x90,
        0x12, 0x48, 0x24, 0x24, 0x49, 0x92, 0x13, 0xc8,
        0x26, 0x64, 0x4d, 0xb2, 0x1b, 0xd8, 0x36, 0x6c,
        0x6d, 0xb6, 0x5a, 0x5a, 0x35, 0xac, 0x6b, 0xd6,
        0x56, 0x6a, 0x2d, 0xb4, 0x5b, 0xda, 0x37, 0xec,
        0x6f, 0xf6, 0x5e, 0x7a, 0x3d, 0xbc, 0x7b, 0xde,
        0x76, 0x6e, 0x6c, 0x36, 0x58, 0x1a, 0x31, 0x8c,
        0x63, 0xc6, 0x46, 0x62, 0x0d, 0xb0, 0x1a, 0x58,
        0x34, 0x2c, 0x69, 0x96, 0x52, 0x4a, 0x25, 0xa4,
        0x4b, 0xd2, 0x17, 0xe8, 0x2e, 0x74, 0x5d, 0xba,
        0x3b, 0xdc, 0x77, 0xee, 0x6e, 0x76, 0x5c, 0x3a,
        0x39, 0x9c, 0x73, 0xce, 0x66, 0x66, 0x4c, 0x32,
        0x19, 0x98, 0x32, 0x4c, 0x65, 0xa6, 0x4a, 0x52,
        0x15, 0xa8, 0x2a, 0x54, 0x55, 0xaa, 0x2b, 0xd4,
        0x57, 0xea, 0x2f, 0xf4, 0x5f, 0xfa, 0x3f, 0xfc
    };
    const uint8_t *rc = RC;
    uint32_t x0, x1, x2, x3, x4;
    uint32_t t0, t1, t2, t3, t4;
    uint8_t round;

    /* Load the state into local variables and convert from little-endian */
#if defined(LW_UTIL_LITTLE_ENDIAN)
    x0 = state->W[0];
    x1 = state->W[1];
    x2 = state->W[2];
    x3 = state->W[3];
    x4 = state->W[4];
#else
    x0 = le_load_word32(state->B);
    x1 = le_load_word32(state->B + 4);
    x2 = le_load_word32(state->B + 8);
    x3 = le_load_word32(state->B + 12);
    x4 = le_load_word32(state->B + 16);
#endif

    /* Perform the 80 rounds of Spongent-pi[160] */
    for (round = 0; round < 80; ++round, rc += 2) {
        /* Add the round constant to front and back of the state */
        x0 ^= rc[0];
        x4 ^= ((uint32_t)(rc[1])) << 24;

        /* Apply the S-box to all 4-bit groups in the state */
        t0 = spongent_sbox(x0);
        t1 = spongent_sbox(x1);
        t2 = spongent_sbox(x2);
        t3 = spongent_sbox(x3);
        t4 = spongent_sbox(x4);

        /* Permute the bits of the state.  Bit i is moved to (40 * i) % 159
         * for all bits except the last which is left where it is.
         * BCP = bit copy, BUP = move bit up, BDN = move bit down */
        #define BCP(x, bit) ((x) & (((uint32_t)1) << (bit)))
        #define BUP(x, from, to) \
            (((x) << ((to) - (from))) & (((uint32_t)1) << (to)))
        #define BDN(x, from, to) \
            (((x) >> ((from) - (to))) & (((uint32_t)1) << (to)))
        x0 = BCP(t0,  0)     ^ BDN(t0,  4,  1) ^ BDN(t0,  8,  2) ^
             BDN(t0, 12,  3) ^ BDN(t0, 16,  4) ^ BDN(t0, 20,  5) ^
             BDN(t0, 24,  6) ^ BDN(t0, 28,  7) ^ BUP(t1,  0,  8) ^
             BUP(t1,  4,  9) ^ BUP(t1,  8, 10) ^ BDN(t1, 12, 11) ^
             BDN(t1, 16, 12) ^ BDN(t1, 20, 13) ^ BDN(t1, 24, 14) ^
             BDN(t1, 28, 15) ^ BUP(t2,  0, 16) ^ BUP(t2,  4, 17) ^
             BUP(t2,  8, 18) ^ BUP(t2, 12, 19) ^ BUP(t2, 16, 20) ^
             BUP(t2, 20, 21) ^ BDN(t2, 24, 22) ^ BDN(t2, 28, 23) ^
             BUP(t3,  0, 24) ^ BUP(t3,  4, 25) ^ BUP(t3,  8, 26) ^
             BUP(t3, 12, 27) ^ BUP(t3, 16, 28) ^ BUP(t3, 20, 29) ^
             BUP(t3, 24, 30) ^ BUP(t3, 28, 31);
        x1 = BUP(t0,  1,  8) ^ BUP(t0,  5,  9) ^ BUP(t0,  9, 10) ^
             BDN(t0, 13, 11) ^ BDN(t0, 17, 12) ^ BDN(t0, 21, 13) ^
             BDN(t0, 25, 14) ^ BDN(t0, 29, 15) ^ BUP(t1,  1, 16) ^
             BUP(t1,  5, 17) ^ BUP(t1,  9, 18) ^ BUP(t1, 13, 19) ^
             BUP(t1, 17, 20) ^ BCP(t1, 21)     ^ BDN(t1, 25, 22) ^
             BDN(t1, 29, 23) ^ BUP(t2,  1, 24) ^ BUP(t2,  5, 25) ^
             BUP(t2,  9, 26) ^ BUP(t2, 13, 27) ^ BUP(t2, 17, 28) ^
             BUP(t2, 21, 29) ^ BUP(t2, 25, 30) ^ BUP(t2, 29, 31) ^
             BCP(t4,  0)     ^ BDN(t4,  4,  1) ^ BDN(t4,  8,  2) ^
             BDN(t4, 12,  3) ^ BDN(t4, 16,  4) ^ BDN(t4, 20,  5) ^
             BDN(t4, 24,  6) ^ BDN(t4, 28,  7);
        x2 = BUP(t0,  2, 16) ^ BUP(t0,  6, 17) ^ BUP(t0, 10, 18) ^
             BUP(t0, 14, 19) ^ BUP(t0, 18, 20) ^ BDN(t0, 22, 21) ^
             BDN(t0, 26, 22) ^ BDN(t0, 30, 23) ^ BUP(t1,  2, 24) ^
             BUP(t1,  6, 25) ^ BUP(t1, 10, 26) ^ BUP(t1, 14, 27) ^
             BUP(t1, 18, 28) ^ BUP(t1, 22, 29) ^ BUP(t1, 26, 30) ^
             BUP(t1, 30, 31) ^ BDN(t3,  1,  0) ^ BDN(t3,  5,  1) ^
             BDN(t3,  9,  2) ^ BDN(t3, 13,  3) ^ BDN(t3, 17,  4) ^
             BDN(t3, 21,  5) ^ BDN(t3, 25,  6) ^ BDN(t3, 29,  7) ^
             BUP(t4,  1,  8) ^ BUP(t4,  5,  9) ^ BUP(t4,  9, 10) ^
             BDN(t4, 13, 11) ^ BDN(t4, 17, 12) ^ BDN(t4, 21, 13) ^
             BDN(t4, 25, 14) ^ BDN(t4, 29, 15);
        x3 = BUP(t0,  3, 24) ^ BUP(t0,  7, 25) ^ BUP(t0, 11, 26) ^
             BUP(t0, 15, 27) ^ BUP(t0, 19, 28) ^ BUP(t0, 23, 29) ^
             BUP(t0, 27, 30) ^ BCP(t0, 31)     ^ BDN(t2,  2,  0) ^
             BDN(t2,  6,  1) ^ BDN(t2, 10,  2) ^ BDN(t2, 14,  3) ^
             BDN(t2, 18,  4) ^ BDN(t2, 22,  5) ^ BDN(t2, 26,  6) ^
             BDN(t2, 30,  7) ^ BUP(t3,  2,  8) ^ BUP(t3,  6,  9) ^
             BCP(t3, 10)     ^ BDN(t3, 14, 11) ^ BDN(t3, 18, 12) ^
             BDN(t3, 22, 13) ^ BDN(t3, 26, 14) ^ BDN(t3, 30, 15) ^
             BUP(t4,  2, 16) ^ BUP(t4,  6, 17) ^ BUP(t4, 10, 18) ^
             BUP(t4, 14, 19) ^ BUP(t4, 18, 20) ^ BDN(t4, 22, 21) ^
             BDN(t4, 26, 22) ^ BDN(t4, 30, 23);
        x4 = BDN(t1,  3,  0) ^ BDN(t1,  7,  1) ^ BDN(t1, 11,  2) ^
             BDN(t1, 15,  3) ^ BDN(t1, 19,  4) ^ BDN(t1, 23,  5) ^
             BDN(t1, 27,  6) ^ BDN(t1, 31,  7) ^ BUP(t2,  3,  8) ^
             BUP(t2,  7,  9) ^ BDN(t2, 11, 10) ^ BDN(t2, 15, 11) ^
             BDN(t2, 19, 12) ^ BDN(t2, 23, 13) ^ BDN(t2, 27, 14) ^
             BDN(t2, 31, 15) ^ BUP(t3,  3, 16) ^ BUP(t3,  7, 17) ^
             BUP(t3, 11, 18) ^ BUP(t3, 15, 19) ^ BUP(t3, 19, 20) ^
             BDN(t3, 23, 21) ^ BDN(t3, 27, 22) ^ BDN(t3, 31, 23) ^
             BUP(t4,  3, 24) ^ BUP(t4,  7, 25) ^ BUP(t4, 11, 26) ^
             BUP(t4, 15, 27) ^ BUP(t4, 19, 28) ^ BUP(t4, 23, 29) ^
             BUP(t4, 27, 30) ^ BCP(t4, 31);
    }

    /* Store the local variables back to the state in little-endian order */
#if defined(LW_UTIL_LITTLE_ENDIAN)
    state->W[0] = x0;
    state->W[1] = x1;
    state->W[2] = x2;
    state->W[3] = x3;
    state->W[4] = x4;
#else
    le_store_word32(state->B,      x0);
    le_store_word32(state->B +  4, x1);
    le_store_word32(state->B +  8, x2);
    le_store_word32(state->B + 12, x3);
    le_store_word32(state->B + 16, x4);
#endif
}

void spongent176_permute(spongent176_state_t *state)
{
    static uint8_t const RC[] = {
        /* Round constants for Spongent-pi[176] */
        0x45, 0xa2, 0x0b, 0xd0, 0x16, 0x68, 0x2c, 0x34,
        0x59, 0x9a, 0x33, 0xcc, 0x67, 0xe6, 0x4e, 0x72,
        0x1d, 0xb8, 0x3a, 0x5c, 0x75, 0xae, 0x6a, 0x56,
        0x54, 0x2a, 0x29, 0x94, 0x53, 0xca, 0x27, 0xe4,
        0x4f, 0xf2, 0x1f, 0xf8, 0x3e, 0x7c, 0x7d, 0xbe,
        0x7a, 0x5e, 0x74, 0x2e, 0x68, 0x16, 0x50, 0x0a,
        0x21, 0x84, 0x43, 0xc2, 0x07, 0xe0, 0x0e, 0x70,
        0x1c, 0x38, 0x38, 0x1c, 0x71, 0x8e, 0x62, 0x46,
        0x44, 0x22, 0x09, 0x90, 0x12, 0x48, 0x24, 0x24,
        0x49, 0x92, 0x13, 0xc8, 0x26, 0x64, 0x4d, 0xb2,
        0x1b, 0xd8, 0x36, 0x6c, 0x6d, 0xb6, 0x5a, 0x5a,
        0x35, 0xac, 0x6b, 0xd6, 0x56, 0x6a, 0x2d, 0xb4,
        0x5b, 0xda, 0x37, 0xec, 0x6f, 0xf6, 0x5e, 0x7a,
        0x3d, 0xbc, 0x7b, 0xde, 0x76, 0x6e, 0x6c, 0x36,
        0x58, 0x1a, 0x31, 0x8c, 0x63, 0xc6, 0x46, 0x62,
        0x0d, 0xb0, 0x1a, 0x58, 0x34, 0x2c, 0x69, 0x96,
        0x52, 0x4a, 0x25, 0xa4, 0x4b, 0xd2, 0x17, 0xe8,
        0x2e, 0x74, 0x5d, 0xba, 0x3b, 0xdc, 0x77, 0xee,
        0x6e, 0x76, 0x5c, 0x3a, 0x39, 0x9c, 0x73, 0xce,
        0x66, 0x66, 0x4c, 0x32, 0x19, 0x98, 0x32, 0x4c,
        0x65, 0xa6, 0x4a, 0x52, 0x15, 0xa8, 0x2a, 0x54,
        0x55, 0xaa, 0x2b, 0xd4, 0x57, 0xea, 0x2f, 0xf4,
        0x5f, 0xfa, 0x3f, 0xfc
    };
    const uint8_t *rc = RC;
    uint32_t x0, x1, x2, x3, x4, x5;
    uint32_t t0, t1, t2, t3, t4, t5;
    uint8_t round;

    /* Load the state into local variables and convert from little-endian */
#if defined(LW_UTIL_LITTLE_ENDIAN)
    x0 = state->W[0];
    x1 = state->W[1];
    x2 = state->W[2];
    x3 = state->W[3];
    x4 = state->W[4];
    x5 = state->W[5];
#else
    x0 = le_load_word32(state->B);
    x1 = le_load_word32(state->B + 4);
    x2 = le_load_word32(state->B + 8);
    x3 = le_load_word32(state->B + 12);
    x4 = le_load_word32(state->B + 16);
    x5 = le_load_word16(state->B + 20); /* Last word is only 16 bits */
#endif

    /* Perform the 90 rounds of Spongent-pi[176] */
    for (round = 0; round < 90; ++round, rc += 2) {
        /* Add the round constant to front and back of the state */
        x0 ^= rc[0];
        x5 ^= ((uint32_t)(rc[1])) << 8;

        /* Apply the S-box to all 4-bit groups in the state */
        t0 = spongent_sbox(x0);
        t1 = spongent_sbox(x1);
        t2 = spongent_sbox(x2);
        t3 = spongent_sbox(x3);
        t4 = spongent_sbox(x4);
        t5 = spongent_sbox(x5);

        /* Permute the bits of the state.  Bit i is moved to (44 * i) % 175
         * for all bits except the last which is left where it is.
         * BCP = bit copy, BUP = move bit up, BDN = move bit down */
        x0 = BCP(t0,  0)     ^ BDN(t0,  4,  1) ^ BDN(t0,  8,  2) ^
             BDN(t0, 12,  3) ^ BDN(t0, 16,  4) ^ BDN(t0, 20,  5) ^
             BDN(t0, 24,  6) ^ BDN(t0, 28,  7) ^ BUP(t1,  0,  8) ^
             BUP(t1,  4,  9) ^ BUP(t1,  8, 10) ^ BDN(t1, 12, 11) ^
             BDN(t1, 16, 12) ^ BDN(t1, 20, 13) ^ BDN(t1, 24, 14) ^
             BDN(t1, 28, 15) ^ BUP(t2,  0, 16) ^ BUP(t2,  4, 17) ^
             BUP(t2,  8, 18) ^ BUP(t2, 12, 19) ^ BUP(t2, 16, 20) ^
             BUP(t2, 20, 21) ^ BDN(t2, 24, 22) ^ BDN(t2, 28, 23) ^
             BUP(t3,  0, 24) ^ BUP(t3,  4, 25) ^ BUP(t3,  8, 26) ^
             BUP(t3, 12, 27) ^ BUP(t3, 16, 28) ^ BUP(t3, 20, 29) ^
             BUP(t3, 24, 30) ^ BUP(t3, 28, 31);
        x1 = BUP(t0,  1, 12) ^ BUP(t0,  5, 13) ^ BUP(t0,  9, 14) ^
             BUP(t0, 13, 15) ^ BDN(t0, 17, 16) ^ BDN(t0, 21, 17) ^
             BDN(t0, 25, 18) ^ BDN(t0, 29, 19) ^ BUP(t1,  1, 20) ^
             BUP(t1,  5, 21) ^ BUP(t1,  9, 22) ^ BUP(t1, 13, 23) ^
             BUP(t1, 17, 24) ^ BUP(t1, 21, 25) ^ BUP(t1, 25, 26) ^
             BDN(t1, 29, 27) ^ BUP(t2,  1, 28) ^ BUP(t2,  5, 29) ^
             BUP(t2,  9, 30) ^ BUP(t2, 13, 31) ^ BCP(t4,  0)     ^
             BDN(t4,  4,  1) ^ BDN(t4,  8,  2) ^ BDN(t4, 12,  3) ^
             BDN(t4, 16,  4) ^ BDN(t4, 20,  5) ^ BDN(t4, 24,  6) ^
             BDN(t4, 28,  7) ^ BUP(t5,  0,  8) ^ BUP(t5,  4,  9) ^
             BUP(t5,  8, 10) ^ BDN(t5, 12, 11);
        x2 = BUP(t0,  2, 24) ^ BUP(t0,  6, 25) ^ BUP(t0, 10, 26) ^
             BUP(t0, 14, 27) ^ BUP(t0, 18, 28) ^ BUP(t0, 22, 29) ^
             BUP(t0, 26, 30) ^ BUP(t0, 30, 31) ^ BDN(t2, 17,  0) ^
             BDN(t2, 21,  1) ^ BDN(t2, 25,  2) ^ BDN(t2, 29,  3) ^
             BUP(t3,  1,  4) ^ BCP(t3,  5)     ^ BDN(t3,  9,  6) ^
             BDN(t3, 13,  7) ^ BDN(t3, 17,  8) ^ BDN(t3, 21,  9) ^
             BDN(t3, 25, 10) ^ BDN(t3, 29, 11) ^ BUP(t4,  1, 12) ^
             BUP(t4,  5, 13) ^ BUP(t4,  9, 14) ^ BUP(t4, 13, 15) ^
             BDN(t4, 17, 16) ^ BDN(t4, 21, 17) ^ BDN(t4, 25, 18) ^
             BDN(t4, 29, 19) ^ BUP(t5,  1, 20) ^ BUP(t5,  5, 21) ^
             BUP(t5,  9, 22) ^ BUP(t5, 13, 23);
        x3 = BDN(t1,  2,  0) ^ BDN(t1,  6,  1) ^ BDN(t1, 10,  2) ^
             BDN(t1, 14,  3) ^ BDN(t1, 18,  4) ^ BDN(t1, 22,  5) ^
             BDN(t1, 26,  6) ^ BDN(t1, 30,  7) ^ BUP(t2,  2,  8) ^
             BUP(t2,  6,  9) ^ BCP(t2, 10)     ^ BDN(t2, 14, 11) ^
             BDN(t2, 18, 12) ^ BDN(t2, 22, 13) ^ BDN(t2, 26, 14) ^
             BDN(t2, 30, 15) ^ BUP(t3,  2, 16) ^ BUP(t3,  6, 17) ^
             BUP(t3, 10, 18) ^ BUP(t3, 14, 19) ^ BUP(t3, 18, 20) ^
             BDN(t3, 22, 21) ^ BDN(t3, 26, 22) ^ BDN(t3, 30, 23) ^
             BUP(t4,  2, 24) ^ BUP(t4,  6, 25) ^ BUP(t4, 10, 26) ^
             BUP(t4, 14, 27) ^ BUP(t4, 18, 28) ^ BUP(t4, 22, 29) ^
             BUP(t4, 26, 30) ^ BUP(t4, 30, 31);
        x4 = BUP(t0,  3,  4) ^ BDN(t0,  7,  5) ^ BDN(t0, 11,  6) ^
             BDN(t0, 15,  7) ^ BDN(t0, 19,  8) ^ BDN(t0, 23,  9) ^
             BDN(t0, 27, 10) ^ BDN(t0, 31, 11) ^ BUP(t1,  3, 12) ^
             BUP(t1,  7, 13) ^ BUP(t1, 11, 14) ^ BCP(t1, 15)     ^
             BDN(t1, 19, 16) ^ BDN(t1, 23, 17) ^ BDN(t1, 27, 18) ^
             BDN(t1, 31, 19) ^ BUP(t2,  3, 20) ^ BUP(t2,  7, 21) ^
             BUP(t2, 11, 22) ^ BUP(t2, 15, 23) ^ BUP(t2, 19, 24) ^
             BUP(t2, 23, 25) ^ BDN(t2, 27, 26) ^ BDN(t2, 31, 27) ^
             BUP(t3,  3, 28) ^ BUP(t3,  7, 29) ^ BUP(t3, 11, 30) ^
             BUP(t3, 15, 31) ^ BDN(t5,  2,  0) ^ BDN(t5,  6,  1) ^
             BDN(t5, 10,  2) ^ BDN(t5, 14,  3);
        x5 = BDN(t3, 19,  0) ^ BDN(t3, 23,  1) ^ BDN(t3, 27,  2) ^
             BDN(t3, 31,  3) ^ BUP(t4,  3,  4) ^ BDN(t4,  7,  5) ^
             BDN(t4, 11,  6) ^ BDN(t4, 15,  7) ^ BDN(t4, 19,  8) ^
             BDN(t4, 23,  9) ^ BDN(t4, 27, 10) ^ BDN(t4, 31, 11) ^
             BUP(t5,  3, 12) ^ BUP(t5,  7, 13) ^ BUP(t5, 11, 14) ^
             BCP(t5, 15);
    }

    /* Store the local variables back to the state in little-endian order */
#if defined(LW_UTIL_LITTLE_ENDIAN)
    state->W[0] = x0;
    state->W[1] = x1;
    state->W[2] = x2;
    state->W[3] = x3;
    state->W[4] = x4;
    state->W[5] = x5;
#else
    le_store_word32(state->B,      x0);
    le_store_word32(state->B +  4, x1);
    le_store_word32(state->B +  8, x2);
    le_store_word32(state->B + 12, x3);
    le_store_word32(state->B + 16, x4);
    le_store_word16(state->B + 20, x5); /* Last word is only 16 bits */
#endif
}
349 350

#endif /* !__AVR__ */