scrypt-1.2.1/lib/crypto/crypto_scrypt_smix.c

   1 /*-
   2  * Copyright 2009 Colin Percival
   3  * All rights reserved.
   4  *
   5  * Redistribution and use in source and binary forms, with or without
   6  * modification, are permitted provided that the following conditions
   7  * are met:
   8  * 1. Redistributions of source code must retain the above copyright
   9  *    notice, this list of conditions and the following disclaimer.
  10  * 2. Redistributions in binary form must reproduce the above copyright
  11  *    notice, this list of conditions and the following disclaimer in the
  12  *    documentation and/or other materials provided with the distribution.
  13  *
  14  * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
  15  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
  16  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
  17  * ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
  18  * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
  19  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
  20  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
  21  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
  22  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
  23  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
  24  * SUCH DAMAGE.
  25  *
  26  * This file was originally written by Colin Percival as part of the Tarsnap
  27  * online backup system.
  28  */
  29 #include <stdint.h>
  30
  31 #include "sysendian.h"
  32
  33 #include "crypto_scrypt_smix.h"
  34
  35 static void blkcpy(void *, const void *, size_t);
  36 static void blkxor(void *, const void *, size_t);
  37 static void salsa20_8(uint32_t[16]);
  38 static void blockmix_salsa8(const uint32_t *, uint32_t *, uint32_t *, size_t);
  39 static uint64_t integerify(const void *, size_t);
  40
  41 static void
  42 blkcpy(void * dest, const void * src, size_t len)
  43 {
  44         size_t * D = dest;
  45         const size_t * S = src;
  46         size_t L = len / sizeof(size_t);
  47         size_t i;
  48
  49         for (i = 0; i < L; i++)
  50                 D[i] = S[i];
  51 }
  52
  53 static void
  54 blkxor(void * dest, const void * src, size_t len)
  55 {
  56         size_t * D = dest;
  57         const size_t * S = src;
  58         size_t L = len / sizeof(size_t);
  59         size_t i;
  60
  61         for (i = 0; i < L; i++)
  62                 D[i] ^= S[i];
  63 }
  64
  65 /**
  66  * salsa20_8(B):
  67  * Apply the salsa20/8 core to the provided block.
  68  */
  69 static void
  70 salsa20_8(uint32_t B[16])
  71 {
  72         uint32_t x[16];
  73         size_t i;
  74
  75         blkcpy(x, B, 64);
  76         for (i = 0; i < 8; i += 2) {
  77 #define R(a,b) (((a) << (b)) | ((a) >> (32 - (b))))
  78                 /* Operate on columns. */
  79                 x[ 4] ^= R(x[ 0]+x[12], 7);  x[ 8] ^= R(x[ 4]+x[ 0], 9);
  80                 x[12] ^= R(x[ 8]+x[ 4],13);  x[ 0] ^= R(x[12]+x[ 8],18);
  81
  82                 x[ 9] ^= R(x[ 5]+x[ 1], 7);  x[13] ^= R(x[ 9]+x[ 5], 9);
  83                 x[ 1] ^= R(x[13]+x[ 9],13);  x[ 5] ^= R(x[ 1]+x[13],18);
  84
  85                 x[14] ^= R(x[10]+x[ 6], 7);  x[ 2] ^= R(x[14]+x[10], 9);
  86                 x[ 6] ^= R(x[ 2]+x[14],13);  x[10] ^= R(x[ 6]+x[ 2],18);
  87
  88                 x[ 3] ^= R(x[15]+x[11], 7);  x[ 7] ^= R(x[ 3]+x[15], 9);
  89                 x[11] ^= R(x[ 7]+x[ 3],13);  x[15] ^= R(x[11]+x[ 7],18);
  90
  91                 /* Operate on rows. */
  92                 x[ 1] ^= R(x[ 0]+x[ 3], 7);  x[ 2] ^= R(x[ 1]+x[ 0], 9);
  93                 x[ 3] ^= R(x[ 2]+x[ 1],13);  x[ 0] ^= R(x[ 3]+x[ 2],18);
  94
  95                 x[ 6] ^= R(x[ 5]+x[ 4], 7);  x[ 7] ^= R(x[ 6]+x[ 5], 9);
  96                 x[ 4] ^= R(x[ 7]+x[ 6],13);  x[ 5] ^= R(x[ 4]+x[ 7],18);
  97
  98                 x[11] ^= R(x[10]+x[ 9], 7);  x[ 8] ^= R(x[11]+x[10], 9);
  99                 x[ 9] ^= R(x[ 8]+x[11],13);  x[10] ^= R(x[ 9]+x[ 8],18);
 100
 101                 x[12] ^= R(x[15]+x[14], 7);  x[13] ^= R(x[12]+x[15], 9);
 102                 x[14] ^= R(x[13]+x[12],13);  x[15] ^= R(x[14]+x[13],18);
 103 #undef R
 104         }
 105         for (i = 0; i < 16; i++)
 106                 B[i] += x[i];
 107 }
 108
 109 /**
 110  * blockmix_salsa8(Bin, Bout, X, r):
 111  * Compute Bout = BlockMix_{salsa20/8, r}(Bin).  The input Bin must be 128r
 112  * bytes in length; the output Bout must also be the same size.  The
 113  * temporary space X must be 64 bytes.
 114  */
 115 static void
 116 blockmix_salsa8(const uint32_t * Bin, uint32_t * Bout, uint32_t * X, size_t r)
 117 {
 118         size_t i;
 119
 120         /* 1: X <-- B_{2r - 1} */
 121         blkcpy(X, &Bin[(2 * r - 1) * 16], 64);
 122
 123         /* 2: for i = 0 to 2r - 1 do */
 124         for (i = 0; i < 2 * r; i += 2) {
 125                 /* 3: X <-- H(X \xor B_i) */
 126                 blkxor(X, &Bin[i * 16], 64);
 127                 salsa20_8(X);
 128
 129                 /* 4: Y_i <-- X */
 130                 /* 6: B' <-- (Y_0, Y_2 ... Y_{2r-2}, Y_1, Y_3 ... Y_{2r-1}) */
 131                 blkcpy(&Bout[i * 8], X, 64);
 132
 133                 /* 3: X <-- H(X \xor B_i) */
 134                 blkxor(X, &Bin[i * 16 + 16], 64);
 135                 salsa20_8(X);
 136
 137                 /* 4: Y_i <-- X */
 138                 /* 6: B' <-- (Y_0, Y_2 ... Y_{2r-2}, Y_1, Y_3 ... Y_{2r-1}) */
 139                 blkcpy(&Bout[i * 8 + r * 16], X, 64);
 140         }
 141 }
 142
 143 /**
 144  * integerify(B, r):
 145  * Return the result of parsing B_{2r-1} as a little-endian integer.
 146  */
 147 static uint64_t
 148 integerify(const void * B, size_t r)
 149 {
 150         const uint32_t * X = (const void *)((uintptr_t)(B) + (2 * r - 1) * 64);
 151
 152         return (((uint64_t)(X[1]) << 32) + X[0]);
 153 }
 154
 155 /**
 156  * crypto_scrypt_smix(B, r, N, V, XY):
 157  * Compute B = SMix_r(B, N).  The input B must be 128r bytes in length;
 158  * the temporary storage V must be 128rN bytes in length; the temporary
 159  * storage XY must be 256r + 64 bytes in length.  The value N must be a
 160  * power of 2 greater than 1.  The arrays B, V, and XY must be aligned to a
 161  * multiple of 64 bytes.
 162  */
 163 void
 164 crypto_scrypt_smix(uint8_t * B, size_t r, uint64_t N, void * _V, void * XY)
 165 {
 166         uint32_t * X = XY;
 167         uint32_t * Y = (void *)((uint8_t *)(XY) + 128 * r);
 168         uint32_t * Z = (void *)((uint8_t *)(XY) + 256 * r);
 169         uint32_t * V = _V;
 170         uint64_t i;
 171         uint64_t j;
 172         size_t k;
 173
 174         /* 1: X <-- B */
 175         for (k = 0; k < 32 * r; k++)
 176                 X[k] = le32dec(&B[4 * k]);
 177
 178         /* 2: for i = 0 to N - 1 do */
 179         for (i = 0; i < N; i += 2) {
 180                 /* 3: V_i <-- X */
 181                 blkcpy(&V[i * (32 * r)], X, 128 * r);
 182
 183                 /* 4: X <-- H(X) */
 184                 blockmix_salsa8(X, Y, Z, r);
 185
 186                 /* 3: V_i <-- X */
 187                 blkcpy(&V[(i + 1) * (32 * r)], Y, 128 * r);
 188
 189                 /* 4: X <-- H(X) */
 190                 blockmix_salsa8(Y, X, Z, r);
 191         }
 192
 193         /* 6: for i = 0 to N - 1 do */
 194         for (i = 0; i < N; i += 2) {
 195                 /* 7: j <-- Integerify(X) mod N */
 196                 j = integerify(X, r) & (N - 1);
 197
 198                 /* 8: X <-- H(X \xor V_j) */
 199                 blkxor(X, &V[j * (32 * r)], 128 * r);
 200                 blockmix_salsa8(X, Y, Z, r);
 201
 202                 /* 7: j <-- Integerify(X) mod N */
 203                 j = integerify(Y, r) & (N - 1);
 204
 205                 /* 8: X <-- H(X \xor V_j) */
 206                 blkxor(Y, &V[j * (32 * r)], 128 * r);
 207                 blockmix_salsa8(Y, X, Z, r);
 208         }
 209
 210         /* 10: B' <-- X */
 211         for (k = 0; k < 32 * r; k++)
 212                 le32enc(&B[4 * k], X[k]);
 213 }