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
|
/*
* Copyright (c) 2016 Thomas Pornin <pornin@bolet.org>
*
* 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 "inner.h"
/* see inner.h */
void
br_aes_ct_bitslice_invSbox(uint32_t *q)
{
/*
* AES S-box is:
* S(x) = A(I(x)) ^ 0x63
* where I() is inversion in GF(256), and A() is a linear
* transform (0 is formally defined to be its own inverse).
* Since inversion is an involution, the inverse S-box can be
* computed from the S-box as:
* iS(x) = B(S(B(x ^ 0x63)) ^ 0x63)
* where B() is the inverse of A(). Indeed, for any y in GF(256):
* iS(S(y)) = B(A(I(B(A(I(y)) ^ 0x63 ^ 0x63))) ^ 0x63 ^ 0x63) = y
*
* Note: we reuse the implementation of the forward S-box,
* instead of duplicating it here, so that total code size is
* lower. By merging the B() transforms into the S-box circuit
* we could make faster CBC decryption, but CBC decryption is
* already quite faster than CBC encryption because we can
* process two blocks in parallel.
*/
uint32_t q0, q1, q2, q3, q4, q5, q6, q7;
q0 = ~q[0];
q1 = ~q[1];
q2 = q[2];
q3 = q[3];
q4 = q[4];
q5 = ~q[5];
q6 = ~q[6];
q7 = q[7];
q[7] = q1 ^ q4 ^ q6;
q[6] = q0 ^ q3 ^ q5;
q[5] = q7 ^ q2 ^ q4;
q[4] = q6 ^ q1 ^ q3;
q[3] = q5 ^ q0 ^ q2;
q[2] = q4 ^ q7 ^ q1;
q[1] = q3 ^ q6 ^ q0;
q[0] = q2 ^ q5 ^ q7;
br_aes_ct_bitslice_Sbox(q);
q0 = ~q[0];
q1 = ~q[1];
q2 = q[2];
q3 = q[3];
q4 = q[4];
q5 = ~q[5];
q6 = ~q[6];
q7 = q[7];
q[7] = q1 ^ q4 ^ q6;
q[6] = q0 ^ q3 ^ q5;
q[5] = q7 ^ q2 ^ q4;
q[4] = q6 ^ q1 ^ q3;
q[3] = q5 ^ q0 ^ q2;
q[2] = q4 ^ q7 ^ q1;
q[1] = q3 ^ q6 ^ q0;
q[0] = q2 ^ q5 ^ q7;
}
static void
add_round_key(uint32_t *q, const uint32_t *sk)
{
int i;
for (i = 0; i < 8; i ++) {
q[i] ^= sk[i];
}
}
static void
inv_shift_rows(uint32_t *q)
{
int i;
for (i = 0; i < 8; i ++) {
uint32_t x;
x = q[i];
q[i] = (x & 0x000000FF)
| ((x & 0x00003F00) << 2) | ((x & 0x0000C000) >> 6)
| ((x & 0x000F0000) << 4) | ((x & 0x00F00000) >> 4)
| ((x & 0x03000000) << 6) | ((x & 0xFC000000) >> 2);
}
}
static inline uint32_t
rotr16(uint32_t x)
{
return (x << 16) | (x >> 16);
}
static void
inv_mix_columns(uint32_t *q)
{
uint32_t q0, q1, q2, q3, q4, q5, q6, q7;
uint32_t r0, r1, r2, r3, r4, r5, r6, r7;
q0 = q[0];
q1 = q[1];
q2 = q[2];
q3 = q[3];
q4 = q[4];
q5 = q[5];
q6 = q[6];
q7 = q[7];
r0 = (q0 >> 8) | (q0 << 24);
r1 = (q1 >> 8) | (q1 << 24);
r2 = (q2 >> 8) | (q2 << 24);
r3 = (q3 >> 8) | (q3 << 24);
r4 = (q4 >> 8) | (q4 << 24);
r5 = (q5 >> 8) | (q5 << 24);
r6 = (q6 >> 8) | (q6 << 24);
r7 = (q7 >> 8) | (q7 << 24);
q[0] = q5 ^ q6 ^ q7 ^ r0 ^ r5 ^ r7 ^ rotr16(q0 ^ q5 ^ q6 ^ r0 ^ r5);
q[1] = q0 ^ q5 ^ r0 ^ r1 ^ r5 ^ r6 ^ r7 ^ rotr16(q1 ^ q5 ^ q7 ^ r1 ^ r5 ^ r6);
q[2] = q0 ^ q1 ^ q6 ^ r1 ^ r2 ^ r6 ^ r7 ^ rotr16(q0 ^ q2 ^ q6 ^ r2 ^ r6 ^ r7);
q[3] = q0 ^ q1 ^ q2 ^ q5 ^ q6 ^ r0 ^ r2 ^ r3 ^ r5 ^ rotr16(q0 ^ q1 ^ q3 ^ q5 ^ q6 ^ q7 ^ r0 ^ r3 ^ r5 ^ r7);
q[4] = q1 ^ q2 ^ q3 ^ q5 ^ r1 ^ r3 ^ r4 ^ r5 ^ r6 ^ r7 ^ rotr16(q1 ^ q2 ^ q4 ^ q5 ^ q7 ^ r1 ^ r4 ^ r5 ^ r6);
q[5] = q2 ^ q3 ^ q4 ^ q6 ^ r2 ^ r4 ^ r5 ^ r6 ^ r7 ^ rotr16(q2 ^ q3 ^ q5 ^ q6 ^ r2 ^ r5 ^ r6 ^ r7);
q[6] = q3 ^ q4 ^ q5 ^ q7 ^ r3 ^ r5 ^ r6 ^ r7 ^ rotr16(q3 ^ q4 ^ q6 ^ q7 ^ r3 ^ r6 ^ r7);
q[7] = q4 ^ q5 ^ q6 ^ r4 ^ r6 ^ r7 ^ rotr16(q4 ^ q5 ^ q7 ^ r4 ^ r7);
}
/* see inner.h */
void
br_aes_ct_bitslice_decrypt(unsigned num_rounds,
const uint32_t *skey, uint32_t *q)
{
unsigned u;
add_round_key(q, skey + (num_rounds << 3));
for (u = num_rounds - 1; u > 0; u --) {
inv_shift_rows(q);
br_aes_ct_bitslice_invSbox(q);
add_round_key(q, skey + (u << 3));
inv_mix_columns(q);
}
inv_shift_rows(q);
br_aes_ct_bitslice_invSbox(q);
add_round_key(q, skey);
}
|
