| 1 | /* Copyright (C) 1995-1998 Eric Young (eay@cryptsoft.com) |
| 2 | * All rights reserved. |
| 3 | * |
| 4 | * This package is an SSL implementation written |
| 5 | * by Eric Young (eay@cryptsoft.com). |
| 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 (tjh@cryptsoft.com). |
| 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 (eay@cryptsoft.com)" |
| 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 (tjh@cryptsoft.com)" |
| 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/sha.h> |
| 58 | |
| 59 | #include <string.h> |
| 60 | |
| 61 | #include <openssl/mem.h> |
| 62 | |
| 63 | #include "internal.h" |
| 64 | #include "../../internal.h" |
| 65 | |
| 66 | |
| 67 | // The 32-bit hash algorithms share a common byte-order neutral collector and |
| 68 | // padding function implementations that operate on unaligned data, |
| 69 | // ../digest/md32_common.h. SHA-512 is the only 64-bit hash algorithm, as of |
| 70 | // this writing, so there is no need for a common collector/padding |
| 71 | // implementation yet. |
| 72 | |
| 73 | int SHA384_Init(SHA512_CTX *sha) { |
| 74 | sha->h[0] = UINT64_C(0xcbbb9d5dc1059ed8); |
| 75 | sha->h[1] = UINT64_C(0x629a292a367cd507); |
| 76 | sha->h[2] = UINT64_C(0x9159015a3070dd17); |
| 77 | sha->h[3] = UINT64_C(0x152fecd8f70e5939); |
| 78 | sha->h[4] = UINT64_C(0x67332667ffc00b31); |
| 79 | sha->h[5] = UINT64_C(0x8eb44a8768581511); |
| 80 | sha->h[6] = UINT64_C(0xdb0c2e0d64f98fa7); |
| 81 | sha->h[7] = UINT64_C(0x47b5481dbefa4fa4); |
| 82 | |
| 83 | sha->Nl = 0; |
| 84 | sha->Nh = 0; |
| 85 | sha->num = 0; |
| 86 | sha->md_len = SHA384_DIGEST_LENGTH; |
| 87 | return 1; |
| 88 | } |
| 89 | |
| 90 | |
| 91 | int SHA512_Init(SHA512_CTX *sha) { |
| 92 | sha->h[0] = UINT64_C(0x6a09e667f3bcc908); |
| 93 | sha->h[1] = UINT64_C(0xbb67ae8584caa73b); |
| 94 | sha->h[2] = UINT64_C(0x3c6ef372fe94f82b); |
| 95 | sha->h[3] = UINT64_C(0xa54ff53a5f1d36f1); |
| 96 | sha->h[4] = UINT64_C(0x510e527fade682d1); |
| 97 | sha->h[5] = UINT64_C(0x9b05688c2b3e6c1f); |
| 98 | sha->h[6] = UINT64_C(0x1f83d9abfb41bd6b); |
| 99 | sha->h[7] = UINT64_C(0x5be0cd19137e2179); |
| 100 | |
| 101 | sha->Nl = 0; |
| 102 | sha->Nh = 0; |
| 103 | sha->num = 0; |
| 104 | sha->md_len = SHA512_DIGEST_LENGTH; |
| 105 | return 1; |
| 106 | } |
| 107 | |
| 108 | uint8_t *SHA384(const uint8_t *data, size_t len, |
| 109 | uint8_t out[SHA384_DIGEST_LENGTH]) { |
| 110 | SHA512_CTX ctx; |
| 111 | SHA384_Init(&ctx); |
| 112 | SHA384_Update(&ctx, data, len); |
| 113 | SHA384_Final(out, &ctx); |
| 114 | OPENSSL_cleanse(&ctx, sizeof(ctx)); |
| 115 | return out; |
| 116 | } |
| 117 | |
| 118 | uint8_t *SHA512(const uint8_t *data, size_t len, |
| 119 | uint8_t out[SHA512_DIGEST_LENGTH]) { |
| 120 | SHA512_CTX ctx; |
| 121 | SHA512_Init(&ctx); |
| 122 | SHA512_Update(&ctx, data, len); |
| 123 | SHA512_Final(out, &ctx); |
| 124 | OPENSSL_cleanse(&ctx, sizeof(ctx)); |
| 125 | return out; |
| 126 | } |
| 127 | |
| 128 | #if !defined(SHA512_ASM) |
| 129 | static void sha512_block_data_order(uint64_t *state, const uint8_t *in, |
| 130 | size_t num_blocks); |
| 131 | #endif |
| 132 | |
| 133 | |
| 134 | int SHA384_Final(uint8_t out[SHA384_DIGEST_LENGTH], SHA512_CTX *sha) { |
| 135 | // |SHA384_Init| sets |sha->md_len| to |SHA384_DIGEST_LENGTH|, so this has a |
| 136 | // |smaller output. |
| 137 | return SHA512_Final(out, sha); |
| 138 | } |
| 139 | |
| 140 | int SHA384_Update(SHA512_CTX *sha, const void *data, size_t len) { |
| 141 | return SHA512_Update(sha, data, len); |
| 142 | } |
| 143 | |
| 144 | void SHA512_Transform(SHA512_CTX *c, const uint8_t block[SHA512_CBLOCK]) { |
| 145 | sha512_block_data_order(c->h, block, 1); |
| 146 | } |
| 147 | |
| 148 | int SHA512_Update(SHA512_CTX *c, const void *in_data, size_t len) { |
| 149 | uint64_t l; |
| 150 | uint8_t *p = c->p; |
| 151 | const uint8_t *data = in_data; |
| 152 | |
| 153 | if (len == 0) { |
| 154 | return 1; |
| 155 | } |
| 156 | |
| 157 | l = (c->Nl + (((uint64_t)len) << 3)) & UINT64_C(0xffffffffffffffff); |
| 158 | if (l < c->Nl) { |
| 159 | c->Nh++; |
| 160 | } |
| 161 | if (sizeof(len) >= 8) { |
| 162 | c->Nh += (((uint64_t)len) >> 61); |
| 163 | } |
| 164 | c->Nl = l; |
| 165 | |
| 166 | if (c->num != 0) { |
| 167 | size_t n = sizeof(c->p) - c->num; |
| 168 | |
| 169 | if (len < n) { |
| 170 | OPENSSL_memcpy(p + c->num, data, len); |
| 171 | c->num += (unsigned int)len; |
| 172 | return 1; |
| 173 | } else { |
| 174 | OPENSSL_memcpy(p + c->num, data, n), c->num = 0; |
| 175 | len -= n; |
| 176 | data += n; |
| 177 | sha512_block_data_order(c->h, p, 1); |
| 178 | } |
| 179 | } |
| 180 | |
| 181 | if (len >= sizeof(c->p)) { |
| 182 | sha512_block_data_order(c->h, data, len / sizeof(c->p)); |
| 183 | data += len; |
| 184 | len %= sizeof(c->p); |
| 185 | data -= len; |
| 186 | } |
| 187 | |
| 188 | if (len != 0) { |
| 189 | OPENSSL_memcpy(p, data, len); |
| 190 | c->num = (int)len; |
| 191 | } |
| 192 | |
| 193 | return 1; |
| 194 | } |
| 195 | |
| 196 | int SHA512_Final(uint8_t out[SHA512_DIGEST_LENGTH], SHA512_CTX *sha) { |
| 197 | uint8_t *p = sha->p; |
| 198 | size_t n = sha->num; |
| 199 | |
| 200 | p[n] = 0x80; // There always is a room for one |
| 201 | n++; |
| 202 | if (n > (sizeof(sha->p) - 16)) { |
| 203 | OPENSSL_memset(p + n, 0, sizeof(sha->p) - n); |
| 204 | n = 0; |
| 205 | sha512_block_data_order(sha->h, p, 1); |
| 206 | } |
| 207 | |
| 208 | OPENSSL_memset(p + n, 0, sizeof(sha->p) - 16 - n); |
| 209 | p[sizeof(sha->p) - 1] = (uint8_t)(sha->Nl); |
| 210 | p[sizeof(sha->p) - 2] = (uint8_t)(sha->Nl >> 8); |
| 211 | p[sizeof(sha->p) - 3] = (uint8_t)(sha->Nl >> 16); |
| 212 | p[sizeof(sha->p) - 4] = (uint8_t)(sha->Nl >> 24); |
| 213 | p[sizeof(sha->p) - 5] = (uint8_t)(sha->Nl >> 32); |
| 214 | p[sizeof(sha->p) - 6] = (uint8_t)(sha->Nl >> 40); |
| 215 | p[sizeof(sha->p) - 7] = (uint8_t)(sha->Nl >> 48); |
| 216 | p[sizeof(sha->p) - 8] = (uint8_t)(sha->Nl >> 56); |
| 217 | p[sizeof(sha->p) - 9] = (uint8_t)(sha->Nh); |
| 218 | p[sizeof(sha->p) - 10] = (uint8_t)(sha->Nh >> 8); |
| 219 | p[sizeof(sha->p) - 11] = (uint8_t)(sha->Nh >> 16); |
| 220 | p[sizeof(sha->p) - 12] = (uint8_t)(sha->Nh >> 24); |
| 221 | p[sizeof(sha->p) - 13] = (uint8_t)(sha->Nh >> 32); |
| 222 | p[sizeof(sha->p) - 14] = (uint8_t)(sha->Nh >> 40); |
| 223 | p[sizeof(sha->p) - 15] = (uint8_t)(sha->Nh >> 48); |
| 224 | p[sizeof(sha->p) - 16] = (uint8_t)(sha->Nh >> 56); |
| 225 | |
| 226 | sha512_block_data_order(sha->h, p, 1); |
| 227 | |
| 228 | if (out == NULL) { |
| 229 | // TODO(davidben): This NULL check is absent in other low-level hash 'final' |
| 230 | // functions and is one of the few places one can fail. |
| 231 | return 0; |
| 232 | } |
| 233 | |
| 234 | switch (sha->md_len) { |
| 235 | // Let compiler decide if it's appropriate to unroll... |
| 236 | case SHA384_DIGEST_LENGTH: |
| 237 | for (n = 0; n < SHA384_DIGEST_LENGTH / 8; n++) { |
| 238 | uint64_t t = sha->h[n]; |
| 239 | |
| 240 | *(out++) = (uint8_t)(t >> 56); |
| 241 | *(out++) = (uint8_t)(t >> 48); |
| 242 | *(out++) = (uint8_t)(t >> 40); |
| 243 | *(out++) = (uint8_t)(t >> 32); |
| 244 | *(out++) = (uint8_t)(t >> 24); |
| 245 | *(out++) = (uint8_t)(t >> 16); |
| 246 | *(out++) = (uint8_t)(t >> 8); |
| 247 | *(out++) = (uint8_t)(t); |
| 248 | } |
| 249 | break; |
| 250 | case SHA512_DIGEST_LENGTH: |
| 251 | for (n = 0; n < SHA512_DIGEST_LENGTH / 8; n++) { |
| 252 | uint64_t t = sha->h[n]; |
| 253 | |
| 254 | *(out++) = (uint8_t)(t >> 56); |
| 255 | *(out++) = (uint8_t)(t >> 48); |
| 256 | *(out++) = (uint8_t)(t >> 40); |
| 257 | *(out++) = (uint8_t)(t >> 32); |
| 258 | *(out++) = (uint8_t)(t >> 24); |
| 259 | *(out++) = (uint8_t)(t >> 16); |
| 260 | *(out++) = (uint8_t)(t >> 8); |
| 261 | *(out++) = (uint8_t)(t); |
| 262 | } |
| 263 | break; |
| 264 | // ... as well as make sure md_len is not abused. |
| 265 | default: |
| 266 | // TODO(davidben): This bad |md_len| case is one of the few places a |
| 267 | // low-level hash 'final' function can fail. This should never happen. |
| 268 | return 0; |
| 269 | } |
| 270 | |
| 271 | return 1; |
| 272 | } |
| 273 | |
| 274 | #ifndef SHA512_ASM |
| 275 | static const uint64_t K512[80] = { |
| 276 | UINT64_C(0x428a2f98d728ae22), UINT64_C(0x7137449123ef65cd), |
| 277 | UINT64_C(0xb5c0fbcfec4d3b2f), UINT64_C(0xe9b5dba58189dbbc), |
| 278 | UINT64_C(0x3956c25bf348b538), UINT64_C(0x59f111f1b605d019), |
| 279 | UINT64_C(0x923f82a4af194f9b), UINT64_C(0xab1c5ed5da6d8118), |
| 280 | UINT64_C(0xd807aa98a3030242), UINT64_C(0x12835b0145706fbe), |
| 281 | UINT64_C(0x243185be4ee4b28c), UINT64_C(0x550c7dc3d5ffb4e2), |
| 282 | UINT64_C(0x72be5d74f27b896f), UINT64_C(0x80deb1fe3b1696b1), |
| 283 | UINT64_C(0x9bdc06a725c71235), UINT64_C(0xc19bf174cf692694), |
| 284 | UINT64_C(0xe49b69c19ef14ad2), UINT64_C(0xefbe4786384f25e3), |
| 285 | UINT64_C(0x0fc19dc68b8cd5b5), UINT64_C(0x240ca1cc77ac9c65), |
| 286 | UINT64_C(0x2de92c6f592b0275), UINT64_C(0x4a7484aa6ea6e483), |
| 287 | UINT64_C(0x5cb0a9dcbd41fbd4), UINT64_C(0x76f988da831153b5), |
| 288 | UINT64_C(0x983e5152ee66dfab), UINT64_C(0xa831c66d2db43210), |
| 289 | UINT64_C(0xb00327c898fb213f), UINT64_C(0xbf597fc7beef0ee4), |
| 290 | UINT64_C(0xc6e00bf33da88fc2), UINT64_C(0xd5a79147930aa725), |
| 291 | UINT64_C(0x06ca6351e003826f), UINT64_C(0x142929670a0e6e70), |
| 292 | UINT64_C(0x27b70a8546d22ffc), UINT64_C(0x2e1b21385c26c926), |
| 293 | UINT64_C(0x4d2c6dfc5ac42aed), UINT64_C(0x53380d139d95b3df), |
| 294 | UINT64_C(0x650a73548baf63de), UINT64_C(0x766a0abb3c77b2a8), |
| 295 | UINT64_C(0x81c2c92e47edaee6), UINT64_C(0x92722c851482353b), |
| 296 | UINT64_C(0xa2bfe8a14cf10364), UINT64_C(0xa81a664bbc423001), |
| 297 | UINT64_C(0xc24b8b70d0f89791), UINT64_C(0xc76c51a30654be30), |
| 298 | UINT64_C(0xd192e819d6ef5218), UINT64_C(0xd69906245565a910), |
| 299 | UINT64_C(0xf40e35855771202a), UINT64_C(0x106aa07032bbd1b8), |
| 300 | UINT64_C(0x19a4c116b8d2d0c8), UINT64_C(0x1e376c085141ab53), |
| 301 | UINT64_C(0x2748774cdf8eeb99), UINT64_C(0x34b0bcb5e19b48a8), |
| 302 | UINT64_C(0x391c0cb3c5c95a63), UINT64_C(0x4ed8aa4ae3418acb), |
| 303 | UINT64_C(0x5b9cca4f7763e373), UINT64_C(0x682e6ff3d6b2b8a3), |
| 304 | UINT64_C(0x748f82ee5defb2fc), UINT64_C(0x78a5636f43172f60), |
| 305 | UINT64_C(0x84c87814a1f0ab72), UINT64_C(0x8cc702081a6439ec), |
| 306 | UINT64_C(0x90befffa23631e28), UINT64_C(0xa4506cebde82bde9), |
| 307 | UINT64_C(0xbef9a3f7b2c67915), UINT64_C(0xc67178f2e372532b), |
| 308 | UINT64_C(0xca273eceea26619c), UINT64_C(0xd186b8c721c0c207), |
| 309 | UINT64_C(0xeada7dd6cde0eb1e), UINT64_C(0xf57d4f7fee6ed178), |
| 310 | UINT64_C(0x06f067aa72176fba), UINT64_C(0x0a637dc5a2c898a6), |
| 311 | UINT64_C(0x113f9804bef90dae), UINT64_C(0x1b710b35131c471b), |
| 312 | UINT64_C(0x28db77f523047d84), UINT64_C(0x32caab7b40c72493), |
| 313 | UINT64_C(0x3c9ebe0a15c9bebc), UINT64_C(0x431d67c49c100d4c), |
| 314 | UINT64_C(0x4cc5d4becb3e42b6), UINT64_C(0x597f299cfc657e2a), |
| 315 | UINT64_C(0x5fcb6fab3ad6faec), UINT64_C(0x6c44198c4a475817), |
| 316 | }; |
| 317 | |
| 318 | #if defined(__GNUC__) && __GNUC__ >= 2 && !defined(OPENSSL_NO_ASM) |
| 319 | #if defined(__x86_64) || defined(__x86_64__) |
| 320 | #define ROTR(a, n) \ |
| 321 | ({ \ |
| 322 | uint64_t ret; \ |
| 323 | __asm__("rorq %1, %0" : "=r"(ret) : "J"(n), "0"(a) : "cc"); \ |
| 324 | ret; \ |
| 325 | }) |
| 326 | #elif(defined(_ARCH_PPC) && defined(__64BIT__)) || defined(_ARCH_PPC64) |
| 327 | #define ROTR(a, n) \ |
| 328 | ({ \ |
| 329 | uint64_t ret; \ |
| 330 | __asm__("rotrdi %0, %1, %2" : "=r"(ret) : "r"(a), "K"(n)); \ |
| 331 | ret; \ |
| 332 | }) |
| 333 | #elif defined(__aarch64__) |
| 334 | #define ROTR(a, n) \ |
| 335 | ({ \ |
| 336 | uint64_t ret; \ |
| 337 | __asm__("ror %0, %1, %2" : "=r"(ret) : "r"(a), "I"(n)); \ |
| 338 | ret; \ |
| 339 | }) |
| 340 | #endif |
| 341 | #elif defined(_MSC_VER) && defined(_WIN64) |
| 342 | #pragma intrinsic(_rotr64) |
| 343 | #define ROTR(a, n) _rotr64((a), n) |
| 344 | #endif |
| 345 | |
| 346 | #ifndef ROTR |
| 347 | #define ROTR(x, s) (((x) >> s) | (x) << (64 - s)) |
| 348 | #endif |
| 349 | |
| 350 | static inline uint64_t load_u64_be(const void *ptr) { |
| 351 | uint64_t ret; |
| 352 | OPENSSL_memcpy(&ret, ptr, sizeof(ret)); |
| 353 | return CRYPTO_bswap8(ret); |
| 354 | } |
| 355 | |
| 356 | #define Sigma0(x) (ROTR((x), 28) ^ ROTR((x), 34) ^ ROTR((x), 39)) |
| 357 | #define Sigma1(x) (ROTR((x), 14) ^ ROTR((x), 18) ^ ROTR((x), 41)) |
| 358 | #define sigma0(x) (ROTR((x), 1) ^ ROTR((x), 8) ^ ((x) >> 7)) |
| 359 | #define sigma1(x) (ROTR((x), 19) ^ ROTR((x), 61) ^ ((x) >> 6)) |
| 360 | |
| 361 | #define Ch(x, y, z) (((x) & (y)) ^ ((~(x)) & (z))) |
| 362 | #define Maj(x, y, z) (((x) & (y)) ^ ((x) & (z)) ^ ((y) & (z))) |
| 363 | |
| 364 | |
| 365 | #if defined(__i386) || defined(__i386__) || defined(_M_IX86) |
| 366 | // This code should give better results on 32-bit CPU with less than |
| 367 | // ~24 registers, both size and performance wise... |
| 368 | static void sha512_block_data_order(uint64_t *state, const uint8_t *in, |
| 369 | size_t num) { |
| 370 | uint64_t A, E, T; |
| 371 | uint64_t X[9 + 80], *F; |
| 372 | int i; |
| 373 | |
| 374 | while (num--) { |
| 375 | F = X + 80; |
| 376 | A = state[0]; |
| 377 | F[1] = state[1]; |
| 378 | F[2] = state[2]; |
| 379 | F[3] = state[3]; |
| 380 | E = state[4]; |
| 381 | F[5] = state[5]; |
| 382 | F[6] = state[6]; |
| 383 | F[7] = state[7]; |
| 384 | |
| 385 | for (i = 0; i < 16; i++, F--) { |
| 386 | T = load_u64_be(in + i * 8); |
| 387 | F[0] = A; |
| 388 | F[4] = E; |
| 389 | F[8] = T; |
| 390 | T += F[7] + Sigma1(E) + Ch(E, F[5], F[6]) + K512[i]; |
| 391 | E = F[3] + T; |
| 392 | A = T + Sigma0(A) + Maj(A, F[1], F[2]); |
| 393 | } |
| 394 | |
| 395 | for (; i < 80; i++, F--) { |
| 396 | T = sigma0(F[8 + 16 - 1]); |
| 397 | T += sigma1(F[8 + 16 - 14]); |
| 398 | T += F[8 + 16] + F[8 + 16 - 9]; |
| 399 | |
| 400 | F[0] = A; |
| 401 | F[4] = E; |
| 402 | F[8] = T; |
| 403 | T += F[7] + Sigma1(E) + Ch(E, F[5], F[6]) + K512[i]; |
| 404 | E = F[3] + T; |
| 405 | A = T + Sigma0(A) + Maj(A, F[1], F[2]); |
| 406 | } |
| 407 | |
| 408 | state[0] += A; |
| 409 | state[1] += F[1]; |
| 410 | state[2] += F[2]; |
| 411 | state[3] += F[3]; |
| 412 | state[4] += E; |
| 413 | state[5] += F[5]; |
| 414 | state[6] += F[6]; |
| 415 | state[7] += F[7]; |
| 416 | |
| 417 | in += 16 * 8; |
| 418 | } |
| 419 | } |
| 420 | |
| 421 | #else |
| 422 | |
| 423 | #define ROUND_00_15(i, a, b, c, d, e, f, g, h) \ |
| 424 | do { \ |
| 425 | T1 += h + Sigma1(e) + Ch(e, f, g) + K512[i]; \ |
| 426 | h = Sigma0(a) + Maj(a, b, c); \ |
| 427 | d += T1; \ |
| 428 | h += T1; \ |
| 429 | } while (0) |
| 430 | |
| 431 | #define ROUND_16_80(i, j, a, b, c, d, e, f, g, h, X) \ |
| 432 | do { \ |
| 433 | s0 = X[(j + 1) & 0x0f]; \ |
| 434 | s0 = sigma0(s0); \ |
| 435 | s1 = X[(j + 14) & 0x0f]; \ |
| 436 | s1 = sigma1(s1); \ |
| 437 | T1 = X[(j) & 0x0f] += s0 + s1 + X[(j + 9) & 0x0f]; \ |
| 438 | ROUND_00_15(i + j, a, b, c, d, e, f, g, h); \ |
| 439 | } while (0) |
| 440 | |
| 441 | static void sha512_block_data_order(uint64_t *state, const uint8_t *in, |
| 442 | size_t num) { |
| 443 | uint64_t a, b, c, d, e, f, g, h, s0, s1, T1; |
| 444 | uint64_t X[16]; |
| 445 | int i; |
| 446 | |
| 447 | while (num--) { |
| 448 | |
| 449 | a = state[0]; |
| 450 | b = state[1]; |
| 451 | c = state[2]; |
| 452 | d = state[3]; |
| 453 | e = state[4]; |
| 454 | f = state[5]; |
| 455 | g = state[6]; |
| 456 | h = state[7]; |
| 457 | |
| 458 | T1 = X[0] = load_u64_be(in); |
| 459 | ROUND_00_15(0, a, b, c, d, e, f, g, h); |
| 460 | T1 = X[1] = load_u64_be(in + 8); |
| 461 | ROUND_00_15(1, h, a, b, c, d, e, f, g); |
| 462 | T1 = X[2] = load_u64_be(in + 2 * 8); |
| 463 | ROUND_00_15(2, g, h, a, b, c, d, e, f); |
| 464 | T1 = X[3] = load_u64_be(in + 3 * 8); |
| 465 | ROUND_00_15(3, f, g, h, a, b, c, d, e); |
| 466 | T1 = X[4] = load_u64_be(in + 4 * 8); |
| 467 | ROUND_00_15(4, e, f, g, h, a, b, c, d); |
| 468 | T1 = X[5] = load_u64_be(in + 5 * 8); |
| 469 | ROUND_00_15(5, d, e, f, g, h, a, b, c); |
| 470 | T1 = X[6] = load_u64_be(in + 6 * 8); |
| 471 | ROUND_00_15(6, c, d, e, f, g, h, a, b); |
| 472 | T1 = X[7] = load_u64_be(in + 7 * 8); |
| 473 | ROUND_00_15(7, b, c, d, e, f, g, h, a); |
| 474 | T1 = X[8] = load_u64_be(in + 8 * 8); |
| 475 | ROUND_00_15(8, a, b, c, d, e, f, g, h); |
| 476 | T1 = X[9] = load_u64_be(in + 9 * 8); |
| 477 | ROUND_00_15(9, h, a, b, c, d, e, f, g); |
| 478 | T1 = X[10] = load_u64_be(in + 10 * 8); |
| 479 | ROUND_00_15(10, g, h, a, b, c, d, e, f); |
| 480 | T1 = X[11] = load_u64_be(in + 11 * 8); |
| 481 | ROUND_00_15(11, f, g, h, a, b, c, d, e); |
| 482 | T1 = X[12] = load_u64_be(in + 12 * 8); |
| 483 | ROUND_00_15(12, e, f, g, h, a, b, c, d); |
| 484 | T1 = X[13] = load_u64_be(in + 13 * 8); |
| 485 | ROUND_00_15(13, d, e, f, g, h, a, b, c); |
| 486 | T1 = X[14] = load_u64_be(in + 14 * 8); |
| 487 | ROUND_00_15(14, c, d, e, f, g, h, a, b); |
| 488 | T1 = X[15] = load_u64_be(in + 15 * 8); |
| 489 | ROUND_00_15(15, b, c, d, e, f, g, h, a); |
| 490 | |
| 491 | for (i = 16; i < 80; i += 16) { |
| 492 | ROUND_16_80(i, 0, a, b, c, d, e, f, g, h, X); |
| 493 | ROUND_16_80(i, 1, h, a, b, c, d, e, f, g, X); |
| 494 | ROUND_16_80(i, 2, g, h, a, b, c, d, e, f, X); |
| 495 | ROUND_16_80(i, 3, f, g, h, a, b, c, d, e, X); |
| 496 | ROUND_16_80(i, 4, e, f, g, h, a, b, c, d, X); |
| 497 | ROUND_16_80(i, 5, d, e, f, g, h, a, b, c, X); |
| 498 | ROUND_16_80(i, 6, c, d, e, f, g, h, a, b, X); |
| 499 | ROUND_16_80(i, 7, b, c, d, e, f, g, h, a, X); |
| 500 | ROUND_16_80(i, 8, a, b, c, d, e, f, g, h, X); |
| 501 | ROUND_16_80(i, 9, h, a, b, c, d, e, f, g, X); |
| 502 | ROUND_16_80(i, 10, g, h, a, b, c, d, e, f, X); |
| 503 | ROUND_16_80(i, 11, f, g, h, a, b, c, d, e, X); |
| 504 | ROUND_16_80(i, 12, e, f, g, h, a, b, c, d, X); |
| 505 | ROUND_16_80(i, 13, d, e, f, g, h, a, b, c, X); |
| 506 | ROUND_16_80(i, 14, c, d, e, f, g, h, a, b, X); |
| 507 | ROUND_16_80(i, 15, b, c, d, e, f, g, h, a, X); |
| 508 | } |
| 509 | |
| 510 | state[0] += a; |
| 511 | state[1] += b; |
| 512 | state[2] += c; |
| 513 | state[3] += d; |
| 514 | state[4] += e; |
| 515 | state[5] += f; |
| 516 | state[6] += g; |
| 517 | state[7] += h; |
| 518 | |
| 519 | in += 16 * 8; |
| 520 | } |
| 521 | } |
| 522 | |
| 523 | #endif |
| 524 | |
| 525 | #endif // !SHA512_ASM |
| 526 | |
| 527 | #undef ROTR |
| 528 | #undef Sigma0 |
| 529 | #undef Sigma1 |
| 530 | #undef sigma0 |
| 531 | #undef sigma1 |
| 532 | #undef Ch |
| 533 | #undef Maj |
| 534 | #undef ROUND_00_15 |
| 535 | #undef ROUND_16_80 |
| 536 | |