1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
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
|
/*
* CAST-256
* (C) 1999-2007 Jack Lloyd
*
* Distributed under the terms of the Botan license
*/
#include <botan/cast256.h>
#include <botan/loadstor.h>
#include <botan/rotate.h>
namespace Botan {
namespace {
/*
* CAST-256 Round Type 1
*/
void round1(u32bit& out, u32bit in, u32bit mask, u32bit rot)
{
u32bit temp = rotate_left(mask + in, rot);
out ^= (CAST_SBOX1[get_byte(0, temp)] ^ CAST_SBOX2[get_byte(1, temp)]) -
CAST_SBOX3[get_byte(2, temp)] + CAST_SBOX4[get_byte(3, temp)];
}
/*
* CAST-256 Round Type 2
*/
void round2(u32bit& out, u32bit in, u32bit mask, u32bit rot)
{
u32bit temp = rotate_left(mask ^ in, rot);
out ^= (CAST_SBOX1[get_byte(0, temp)] - CAST_SBOX2[get_byte(1, temp)] +
CAST_SBOX3[get_byte(2, temp)]) ^ CAST_SBOX4[get_byte(3, temp)];
}
/*
* CAST-256 Round Type 3
*/
void round3(u32bit& out, u32bit in, u32bit mask, u32bit rot)
{
u32bit temp = rotate_left(mask - in, rot);
out ^= ((CAST_SBOX1[get_byte(0, temp)] + CAST_SBOX2[get_byte(1, temp)]) ^
CAST_SBOX3[get_byte(2, temp)]) - CAST_SBOX4[get_byte(3, temp)];
}
}
/*
* CAST-256 Encryption
*/
void CAST_256::encrypt_n(const byte in[], byte out[], size_t blocks) const
{
for(size_t i = 0; i != blocks; ++i)
{
u32bit A = load_be<u32bit>(in, 0);
u32bit B = load_be<u32bit>(in, 1);
u32bit C = load_be<u32bit>(in, 2);
u32bit D = load_be<u32bit>(in, 3);
round1(C, D, MK[ 0], RK[ 0]); round2(B, C, MK[ 1], RK[ 1]);
round3(A, B, MK[ 2], RK[ 2]); round1(D, A, MK[ 3], RK[ 3]);
round1(C, D, MK[ 4], RK[ 4]); round2(B, C, MK[ 5], RK[ 5]);
round3(A, B, MK[ 6], RK[ 6]); round1(D, A, MK[ 7], RK[ 7]);
round1(C, D, MK[ 8], RK[ 8]); round2(B, C, MK[ 9], RK[ 9]);
round3(A, B, MK[10], RK[10]); round1(D, A, MK[11], RK[11]);
round1(C, D, MK[12], RK[12]); round2(B, C, MK[13], RK[13]);
round3(A, B, MK[14], RK[14]); round1(D, A, MK[15], RK[15]);
round1(C, D, MK[16], RK[16]); round2(B, C, MK[17], RK[17]);
round3(A, B, MK[18], RK[18]); round1(D, A, MK[19], RK[19]);
round1(C, D, MK[20], RK[20]); round2(B, C, MK[21], RK[21]);
round3(A, B, MK[22], RK[22]); round1(D, A, MK[23], RK[23]);
round1(D, A, MK[27], RK[27]); round3(A, B, MK[26], RK[26]);
round2(B, C, MK[25], RK[25]); round1(C, D, MK[24], RK[24]);
round1(D, A, MK[31], RK[31]); round3(A, B, MK[30], RK[30]);
round2(B, C, MK[29], RK[29]); round1(C, D, MK[28], RK[28]);
round1(D, A, MK[35], RK[35]); round3(A, B, MK[34], RK[34]);
round2(B, C, MK[33], RK[33]); round1(C, D, MK[32], RK[32]);
round1(D, A, MK[39], RK[39]); round3(A, B, MK[38], RK[38]);
round2(B, C, MK[37], RK[37]); round1(C, D, MK[36], RK[36]);
round1(D, A, MK[43], RK[43]); round3(A, B, MK[42], RK[42]);
round2(B, C, MK[41], RK[41]); round1(C, D, MK[40], RK[40]);
round1(D, A, MK[47], RK[47]); round3(A, B, MK[46], RK[46]);
round2(B, C, MK[45], RK[45]); round1(C, D, MK[44], RK[44]);
store_be(out, A, B, C, D);
in += BLOCK_SIZE;
out += BLOCK_SIZE;
}
}
/*
* CAST-256 Decryption
*/
void CAST_256::decrypt_n(const byte in[], byte out[], size_t blocks) const
{
for(size_t i = 0; i != blocks; ++i)
{
u32bit A = load_be<u32bit>(in, 0);
u32bit B = load_be<u32bit>(in, 1);
u32bit C = load_be<u32bit>(in, 2);
u32bit D = load_be<u32bit>(in, 3);
round1(C, D, MK[44], RK[44]); round2(B, C, MK[45], RK[45]);
round3(A, B, MK[46], RK[46]); round1(D, A, MK[47], RK[47]);
round1(C, D, MK[40], RK[40]); round2(B, C, MK[41], RK[41]);
round3(A, B, MK[42], RK[42]); round1(D, A, MK[43], RK[43]);
round1(C, D, MK[36], RK[36]); round2(B, C, MK[37], RK[37]);
round3(A, B, MK[38], RK[38]); round1(D, A, MK[39], RK[39]);
round1(C, D, MK[32], RK[32]); round2(B, C, MK[33], RK[33]);
round3(A, B, MK[34], RK[34]); round1(D, A, MK[35], RK[35]);
round1(C, D, MK[28], RK[28]); round2(B, C, MK[29], RK[29]);
round3(A, B, MK[30], RK[30]); round1(D, A, MK[31], RK[31]);
round1(C, D, MK[24], RK[24]); round2(B, C, MK[25], RK[25]);
round3(A, B, MK[26], RK[26]); round1(D, A, MK[27], RK[27]);
round1(D, A, MK[23], RK[23]); round3(A, B, MK[22], RK[22]);
round2(B, C, MK[21], RK[21]); round1(C, D, MK[20], RK[20]);
round1(D, A, MK[19], RK[19]); round3(A, B, MK[18], RK[18]);
round2(B, C, MK[17], RK[17]); round1(C, D, MK[16], RK[16]);
round1(D, A, MK[15], RK[15]); round3(A, B, MK[14], RK[14]);
round2(B, C, MK[13], RK[13]); round1(C, D, MK[12], RK[12]);
round1(D, A, MK[11], RK[11]); round3(A, B, MK[10], RK[10]);
round2(B, C, MK[ 9], RK[ 9]); round1(C, D, MK[ 8], RK[ 8]);
round1(D, A, MK[ 7], RK[ 7]); round3(A, B, MK[ 6], RK[ 6]);
round2(B, C, MK[ 5], RK[ 5]); round1(C, D, MK[ 4], RK[ 4]);
round1(D, A, MK[ 3], RK[ 3]); round3(A, B, MK[ 2], RK[ 2]);
round2(B, C, MK[ 1], RK[ 1]); round1(C, D, MK[ 0], RK[ 0]);
store_be(out, A, B, C, D);
in += BLOCK_SIZE;
out += BLOCK_SIZE;
}
}
/*
* CAST-256 Key Schedule
*/
void CAST_256::key_schedule(const byte key[], size_t length)
{
MK.resize(48);
RK.resize(48);
secure_vector<u32bit> K(8);
for(size_t i = 0; i != length; ++i)
K[i/4] = (K[i/4] << 8) + key[i];
u32bit A = K[0], B = K[1], C = K[2], D = K[3],
E = K[4], F = K[5], G = K[6], H = K[7];
for(size_t i = 0; i != 48; i += 4)
{
round1(G, H, KEY_MASK[4*i+ 0], KEY_ROT[(4*i+ 0) % 32]);
round2(F, G, KEY_MASK[4*i+ 1], KEY_ROT[(4*i+ 1) % 32]);
round3(E, F, KEY_MASK[4*i+ 2], KEY_ROT[(4*i+ 2) % 32]);
round1(D, E, KEY_MASK[4*i+ 3], KEY_ROT[(4*i+ 3) % 32]);
round2(C, D, KEY_MASK[4*i+ 4], KEY_ROT[(4*i+ 4) % 32]);
round3(B, C, KEY_MASK[4*i+ 5], KEY_ROT[(4*i+ 5) % 32]);
round1(A, B, KEY_MASK[4*i+ 6], KEY_ROT[(4*i+ 6) % 32]);
round2(H, A, KEY_MASK[4*i+ 7], KEY_ROT[(4*i+ 7) % 32]);
round1(G, H, KEY_MASK[4*i+ 8], KEY_ROT[(4*i+ 8) % 32]);
round2(F, G, KEY_MASK[4*i+ 9], KEY_ROT[(4*i+ 9) % 32]);
round3(E, F, KEY_MASK[4*i+10], KEY_ROT[(4*i+10) % 32]);
round1(D, E, KEY_MASK[4*i+11], KEY_ROT[(4*i+11) % 32]);
round2(C, D, KEY_MASK[4*i+12], KEY_ROT[(4*i+12) % 32]);
round3(B, C, KEY_MASK[4*i+13], KEY_ROT[(4*i+13) % 32]);
round1(A, B, KEY_MASK[4*i+14], KEY_ROT[(4*i+14) % 32]);
round2(H, A, KEY_MASK[4*i+15], KEY_ROT[(4*i+15) % 32]);
RK[i ] = (A % 32);
RK[i+1] = (C % 32);
RK[i+2] = (E % 32);
RK[i+3] = (G % 32);
MK[i ] = H;
MK[i+1] = F;
MK[i+2] = D;
MK[i+3] = B;
}
}
}
|
