-
Notifications
You must be signed in to change notification settings - Fork 0
/
gost3411-2012-avx.h
153 lines (142 loc) · 4.92 KB
/
gost3411-2012-avx.h
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
// Copyright (c) 2020, Dmitry Dulesov <[email protected]>
//
// Redistribution and use in source and binary forms, with or without
// modification, are permitted provided that the following conditions are met:
//
// 1. Redistributions of source code must retain the above copyright notice, this
// list of conditions and the following disclaimer.
//
// 2. Redistributions in binary form must reproduce the above copyright notice,
// this list of conditions and the following disclaimer in the documentation
// and/or other materials provided with the distribution.
//
// THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND
// ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
// WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
// DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR
// ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
// (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
// LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
// ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
// (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
// SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
#ifndef _AVX_H
#define _AVX_H
#ifndef __AVX2__
#error "require avx2 compiler support. gcc -mavx2 -dM -E - </dev/null"
#endif
#include "gost3411-2012-precalc.h"
#include <mmintrin.h>
#include <emmintrin.h>
#include <smmintrin.h>
#include <immintrin.h>
#define STORE_YMM(M, ymm0, ymm1 ) \
_mm256_store_si256((__m256i *) &(M->QWORD[0]), ymm0 ); \
_mm256_store_si256((__m256i *) &(M->QWORD[4]), ymm1 );
#define LOAD_YMM(MP, ymm0, ymm1) \
ymm0 = _mm256_load_si256((const __m256i *) &(MP->QWORD[0])); \
ymm1 = _mm256_load_si256((const __m256i *) &(MP->QWORD[4]));
#define LOADU_YMM(M, ymm0, ymm1) \
ymm0 = _mm256_loadu_si256((const __m256i *) (M )); \
ymm1 = _mm256_loadu_si256((const __m256i *) (M+32));
#define XOR_YMM(MP, ymm0, ymm1) { \
ymm0 = _mm256_xor_si256(ymm0, _mm256_load_si256((const __m256i *) &(MP.QWORD[0])));\
ymm1 = _mm256_xor_si256(ymm1, _mm256_load_si256((const __m256i *) &(MP.QWORD[4])));\
}
#define XOR_YMM2( ymm0, ymm1, ymm2, ymm3) {\
ymm0 = _mm256_xor_si256(ymm0, ymm2); \
ymm1 = _mm256_xor_si256(ymm1, ymm3); \
}
/*
#define YMM_EXTRACT(shift, rr, ymm0, ymm1 ) {\
rr = Ax[0] [ (_mm256_extract_epi64(ymm0, 0) >> shift ) & 0xFF ];\
rr ^= Ax[1] [ (_mm256_extract_epi64(ymm0, 1) >> shift ) & 0xFF ];\
rr ^= Ax[2] [ (_mm256_extract_epi64(ymm0, 2) >> shift ) & 0xFF ];\
rr ^= Ax[3] [ (_mm256_extract_epi64(ymm0, 3) >> shift ) & 0xFF ];\
rr ^= Ax[4] [ (_mm256_extract_epi64(ymm1, 0) >> shift ) & 0xFF ];\
rr ^= Ax[5] [ (_mm256_extract_epi64(ymm1, 1) >> shift ) & 0xFF ];\
rr ^= Ax[6] [ (_mm256_extract_epi64(ymm1, 2) >> shift ) & 0xFF ];\
rr ^= Ax[7] [ (_mm256_extract_epi64(ymm1, 3) >> shift ) & 0xFF ];\
}
*/
#define YMM_EXTRACT2(shift, outymm, ymm0, ymm1 ) {\
register unsigned long long i = _mm256_extract_epi64(ymm0, 0) >> shift; \
r0 = Ax[0][ i & 0xFF ]; \
i=i>>8; \
r1 = Ax[0][ i & 0xFF ]; \
i=i>>8; \
r2 = Ax[0][ i & 0xFF ]; \
i=i>>8; \
r3 = Ax[0][ i & 0xFF ]; \
i = _mm256_extract_epi64(ymm0, 1) >> shift; \
r0 ^= Ax[1][ i & 0xFF ]; \
i=i>>8; \
r1 ^= Ax[1][ i & 0xFF ]; \
i=i>>8; \
r2 ^= Ax[1][ i & 0xFF ]; \
i=i>>8; \
r3 ^= Ax[1][ i & 0xFF ]; \
\
i = _mm256_extract_epi64(ymm0, 2) >> shift; \
r0 ^= Ax[2][ i & 0xFF ]; \
i=i>>8; \
r1 ^= Ax[2][ i & 0xFF ]; \
i=i>>8; \
r2 ^= Ax[2][ i & 0xFF ]; \
i=i>>8; \
r3 ^= Ax[2][ i & 0xFF ]; \
\
i = _mm256_extract_epi64(ymm0, 3) >> shift; \
r0 ^= Ax[3][ i & 0xFF ]; \
i=i>>8; \
r1 ^= Ax[3][ i & 0xFF ]; \
i=i>>8; \
r2 ^= Ax[3][ i & 0xFF ]; \
i=i>>8; \
r3 ^= Ax[3][ i & 0xFF ]; \
\
i = _mm256_extract_epi64(ymm1, 0) >> shift; \
r0 ^= Ax[4][ i & 0xFF ]; \
i=i>>8; \
r1 ^= Ax[4][ i & 0xFF ]; \
i=i>>8; \
r2 ^= Ax[4][ i & 0xFF ]; \
i=i>>8; \
r3 ^= Ax[4][ i & 0xFF ]; \
\
i = _mm256_extract_epi64(ymm1, 1) >> shift; \
r0 ^= Ax[5][ i & 0xFF ]; \
i=i>>8; \
r1 ^= Ax[5][ i & 0xFF ]; \
i=i>>8; \
r2 ^= Ax[5][ i & 0xFF ]; \
i=i>>8; \
r3 ^= Ax[5][ i & 0xFF ]; \
\
i = _mm256_extract_epi64(ymm1, 2) >> shift; \
r0 ^= Ax[6][ i & 0xFF ]; \
i=i>>8; \
r1 ^= Ax[6][ i & 0xFF ]; \
i=i>>8; \
r2 ^= Ax[6][ i & 0xFF ]; \
i=i>>8; \
r3 ^= Ax[6][ i & 0xFF ]; \
\
i = _mm256_extract_epi64(ymm1, 3) >> shift; \
r0 ^= Ax[7][ i & 0xFF ]; \
i=i>>8; \
r1 ^= Ax[7][ i & 0xFF ]; \
i=i>>8; \
r2 ^= Ax[7][ i & 0xFF ]; \
i=i>>8; \
r3 ^= Ax[7][ i & 0xFF ]; \
\
outymm = _mm256_set_epi64x(r3, r2, r1, r0); \
}
#define YMM_LPS(ymm0, ymm1) { \
__m256i tx0, tx1; \
YMM_EXTRACT2(0, tx0, ymm0, ymm1); \
YMM_EXTRACT2(32, tx1, ymm0, ymm1); \
ymm0 = tx0; ymm1 = tx1; \
}
#endif