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
|
/* SPDX-License-Identifier: GPL-2.0-or-later */
/*
* Shared glue code for 128bit block ciphers, AVX2 assembler macros
*
* Copyright © 2012-2013 Jussi Kivilinna <jussi.kivilinna@mbnet.fi>
*/
#define load_16way(src, x0, x1, x2, x3, x4, x5, x6, x7) \
vmovdqu (0*32)(src), x0; \
vmovdqu (1*32)(src), x1; \
vmovdqu (2*32)(src), x2; \
vmovdqu (3*32)(src), x3; \
vmovdqu (4*32)(src), x4; \
vmovdqu (5*32)(src), x5; \
vmovdqu (6*32)(src), x6; \
vmovdqu (7*32)(src), x7;
#define store_16way(dst, x0, x1, x2, x3, x4, x5, x6, x7) \
vmovdqu x0, (0*32)(dst); \
vmovdqu x1, (1*32)(dst); \
vmovdqu x2, (2*32)(dst); \
vmovdqu x3, (3*32)(dst); \
vmovdqu x4, (4*32)(dst); \
vmovdqu x5, (5*32)(dst); \
vmovdqu x6, (6*32)(dst); \
vmovdqu x7, (7*32)(dst);
#define store_cbc_16way(src, dst, x0, x1, x2, x3, x4, x5, x6, x7, t0) \
vpxor t0, t0, t0; \
vinserti128 $1, (src), t0, t0; \
vpxor t0, x0, x0; \
vpxor (0*32+16)(src), x1, x1; \
vpxor (1*32+16)(src), x2, x2; \
vpxor (2*32+16)(src), x3, x3; \
vpxor (3*32+16)(src), x4, x4; \
vpxor (4*32+16)(src), x5, x5; \
vpxor (5*32+16)(src), x6, x6; \
vpxor (6*32+16)(src), x7, x7; \
store_16way(dst, x0, x1, x2, x3, x4, x5, x6, x7);
#define inc_le128(x, minus_one, tmp) \
vpcmpeqq minus_one, x, tmp; \
vpsubq minus_one, x, x; \
vpslldq $8, tmp, tmp; \
vpsubq tmp, x, x;
#define add2_le128(x, minus_one, minus_two, tmp1, tmp2) \
vpcmpeqq minus_one, x, tmp1; \
vpcmpeqq minus_two, x, tmp2; \
vpsubq minus_two, x, x; \
vpor tmp2, tmp1, tmp1; \
vpslldq $8, tmp1, tmp1; \
vpsubq tmp1, x, x;
#define load_ctr_16way(iv, bswap, x0, x1, x2, x3, x4, x5, x6, x7, t0, t0x, t1, \
t1x, t2, t2x, t3, t3x, t4, t5) \
vpcmpeqd t0, t0, t0; \
vpsrldq $8, t0, t0; /* ab: -1:0 ; cd: -1:0 */ \
vpaddq t0, t0, t4; /* ab: -2:0 ; cd: -2:0 */\
\
/* load IV and byteswap */ \
vmovdqu (iv), t2x; \
vmovdqa t2x, t3x; \
inc_le128(t2x, t0x, t1x); \
vbroadcasti128 bswap, t1; \
vinserti128 $1, t2x, t3, t2; /* ab: le0 ; cd: le1 */ \
vpshufb t1, t2, x0; \
\
/* construct IVs */ \
add2_le128(t2, t0, t4, t3, t5); /* ab: le2 ; cd: le3 */ \
vpshufb t1, t2, x1; \
add2_le128(t2, t0, t4, t3, t5); \
vpshufb t1, t2, x2; \
add2_le128(t2, t0, t4, t3, t5); \
vpshufb t1, t2, x3; \
add2_le128(t2, t0, t4, t3, t5); \
vpshufb t1, t2, x4; \
add2_le128(t2, t0, t4, t3, t5); \
vpshufb t1, t2, x5; \
add2_le128(t2, t0, t4, t3, t5); \
vpshufb t1, t2, x6; \
add2_le128(t2, t0, t4, t3, t5); \
vpshufb t1, t2, x7; \
vextracti128 $1, t2, t2x; \
inc_le128(t2x, t0x, t3x); \
vmovdqu t2x, (iv);
#define store_ctr_16way(src, dst, x0, x1, x2, x3, x4, x5, x6, x7) \
vpxor (0*32)(src), x0, x0; \
vpxor (1*32)(src), x1, x1; \
vpxor (2*32)(src), x2, x2; \
vpxor (3*32)(src), x3, x3; \
vpxor (4*32)(src), x4, x4; \
vpxor (5*32)(src), x5, x5; \
vpxor (6*32)(src), x6, x6; \
vpxor (7*32)(src), x7, x7; \
store_16way(dst, x0, x1, x2, x3, x4, x5, x6, x7);
|
