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
|
/*
* XTEA in SIMD
* (C) 2009 Jack Lloyd
*
* Distributed under the terms of the Botan license
*/
#include <botan/xtea_sse2.h>
#include <botan/loadstor.h>
#include <botan/simd_32.h>
namespace Botan {
namespace {
void xtea_encrypt_8(const byte in[64], byte out[64], const u32bit EK[64])
{
SIMD_32 L0 = SIMD_32::load_be(in );
SIMD_32 R0 = SIMD_32::load_be(in + 16);
SIMD_32 L1 = SIMD_32::load_be(in + 32);
SIMD_32 R1 = SIMD_32::load_be(in + 48);
SIMD_32::transpose(L0, R0, L1, R1);
for(u32bit i = 0; i != 32; i += 2)
{
SIMD_32 K0(EK[2*i ]);
SIMD_32 K1(EK[2*i+1]);
SIMD_32 K2(EK[2*i+2]);
SIMD_32 K3(EK[2*i+3]);
L0 += (((R0 << 4) ^ (R0 >> 5)) + R0) ^ K0;
L1 += (((R1 << 4) ^ (R1 >> 5)) + R1) ^ K0;
R0 += (((L0 << 4) ^ (L0 >> 5)) + L0) ^ K1;
R1 += (((L1 << 4) ^ (L1 >> 5)) + L1) ^ K1;
L0 += (((R0 << 4) ^ (R0 >> 5)) + R0) ^ K2;
L1 += (((R1 << 4) ^ (R1 >> 5)) + R1) ^ K2;
R0 += (((L0 << 4) ^ (L0 >> 5)) + L0) ^ K3;
R1 += (((L1 << 4) ^ (L1 >> 5)) + L1) ^ K3;
}
SIMD_32::transpose(L0, R0, L1, R1);
L0.store_be(out);
R0.store_be(out + 16);
L1.store_be(out + 32);
R1.store_be(out + 48);
}
void xtea_decrypt_8(const byte in[64], byte out[64], const u32bit EK[64])
{
SIMD_32 L0 = SIMD_32::load_be(in );
SIMD_32 R0 = SIMD_32::load_be(in + 16);
SIMD_32 L1 = SIMD_32::load_be(in + 32);
SIMD_32 R1 = SIMD_32::load_be(in + 48);
SIMD_32::transpose(L0, R0, L1, R1);
for(u32bit i = 0; i != 32; i += 2)
{
SIMD_32 K0(EK[63 - 2*i]);
SIMD_32 K1(EK[62 - 2*i]);
SIMD_32 K2(EK[61 - 2*i]);
SIMD_32 K3(EK[60 - 2*i]);
R0 -= (((L0 << 4) ^ (L0 >> 5)) + L0) ^ K0;
R1 -= (((L1 << 4) ^ (L1 >> 5)) + L1) ^ K0;
L0 -= (((R0 << 4) ^ (R0 >> 5)) + R0) ^ K1;
L1 -= (((R1 << 4) ^ (R1 >> 5)) + R1) ^ K1;
R0 -= (((L0 << 4) ^ (L0 >> 5)) + L0) ^ K2;
R1 -= (((L1 << 4) ^ (L1 >> 5)) + L1) ^ K2;
L0 -= (((R0 << 4) ^ (R0 >> 5)) + R0) ^ K3;
L1 -= (((R1 << 4) ^ (R1 >> 5)) + R1) ^ K3;
}
SIMD_32::transpose(L0, R0, L1, R1);
L0.store_be(out);
R0.store_be(out + 16);
L1.store_be(out + 32);
R1.store_be(out + 48);
}
}
/*
* XTEA Encryption
*/
void XTEA_SSE2::encrypt_n(const byte in[], byte out[], u32bit blocks) const
{
while(blocks >= 8)
{
xtea_encrypt_8(in, out, this->EK);
in += 8 * BLOCK_SIZE;
out += 8 * BLOCK_SIZE;
blocks -= 8;
}
XTEA::encrypt_n(in, out, blocks);
}
/*
* XTEA Decryption
*/
void XTEA_SSE2::decrypt_n(const byte in[], byte out[], u32bit blocks) const
{
while(blocks >= 8)
{
xtea_decrypt_8(in, out, this->EK);
in += 8 * BLOCK_SIZE;
out += 8 * BLOCK_SIZE;
blocks -= 8;
}
XTEA::decrypt_n(in, out, blocks);
}
}
|
