1/*-
2 * Copyright 2013 John-Mark Gurney <jmg@FreeBSD.org>
3 * All rights reserved.
4 *
5 * Copyright 2015 Netflix, Inc.
6 *
7 * Redistribution and use in source and binary forms, with or without
8 * modification, are permitted provided that the following conditions
9 * are met:
10 * 1. Redistributions of source code must retain the above copyright
11 *    notice, this list of conditions and the following disclaimer.
12 * 2. Redistributions in binary form must reproduce the above copyright
13 *    notice, this list of conditions and the following disclaimer in the
14 *    documentation and/or other materials provided with the distribution.
15 *
16 * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
17 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
18 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
19 * ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
20 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
21 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
22 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
23 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
24 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
25 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
26 * SUCH DAMAGE.
27 *
28 */
29
30#ifndef _AESENCDEC_H_
31#define _AESENCDEC_H_
32
33#include <crypto/aesni/aesni_os.h>
34
35#include <wmmintrin.h>
36
37static inline void
38aesni_enc8(int rounds, const __m128i *keysched, __m128i a,
39    __m128i b, __m128i c, __m128i d, __m128i e, __m128i f, __m128i g,
40    __m128i h, __m128i out[8])
41{
42	int i;
43
44	a ^= keysched[0];
45	b ^= keysched[0];
46	c ^= keysched[0];
47	d ^= keysched[0];
48	e ^= keysched[0];
49	f ^= keysched[0];
50	g ^= keysched[0];
51	h ^= keysched[0];
52
53	for (i = 0; i < rounds; i++) {
54		a = _mm_aesenc_si128(a, keysched[i + 1]);
55		b = _mm_aesenc_si128(b, keysched[i + 1]);
56		c = _mm_aesenc_si128(c, keysched[i + 1]);
57		d = _mm_aesenc_si128(d, keysched[i + 1]);
58		e = _mm_aesenc_si128(e, keysched[i + 1]);
59		f = _mm_aesenc_si128(f, keysched[i + 1]);
60		g = _mm_aesenc_si128(g, keysched[i + 1]);
61		h = _mm_aesenc_si128(h, keysched[i + 1]);
62	}
63
64	out[0] = _mm_aesenclast_si128(a, keysched[i + 1]);
65	out[1] = _mm_aesenclast_si128(b, keysched[i + 1]);
66	out[2] = _mm_aesenclast_si128(c, keysched[i + 1]);
67	out[3] = _mm_aesenclast_si128(d, keysched[i + 1]);
68	out[4] = _mm_aesenclast_si128(e, keysched[i + 1]);
69	out[5] = _mm_aesenclast_si128(f, keysched[i + 1]);
70	out[6] = _mm_aesenclast_si128(g, keysched[i + 1]);
71	out[7] = _mm_aesenclast_si128(h, keysched[i + 1]);
72}
73
74static inline void
75aesni_dec8(int rounds, const __m128i *keysched, __m128i a,
76    __m128i b, __m128i c, __m128i d, __m128i e, __m128i f, __m128i g,
77    __m128i h, __m128i out[8])
78{
79	int i;
80
81	a ^= keysched[0];
82	b ^= keysched[0];
83	c ^= keysched[0];
84	d ^= keysched[0];
85	e ^= keysched[0];
86	f ^= keysched[0];
87	g ^= keysched[0];
88	h ^= keysched[0];
89
90	for (i = 0; i < rounds; i++) {
91		a = _mm_aesdec_si128(a, keysched[i + 1]);
92		b = _mm_aesdec_si128(b, keysched[i + 1]);
93		c = _mm_aesdec_si128(c, keysched[i + 1]);
94		d = _mm_aesdec_si128(d, keysched[i + 1]);
95		e = _mm_aesdec_si128(e, keysched[i + 1]);
96		f = _mm_aesdec_si128(f, keysched[i + 1]);
97		g = _mm_aesdec_si128(g, keysched[i + 1]);
98		h = _mm_aesdec_si128(h, keysched[i + 1]);
99	}
100
101	out[0] = _mm_aesdeclast_si128(a, keysched[i + 1]);
102	out[1] = _mm_aesdeclast_si128(b, keysched[i + 1]);
103	out[2] = _mm_aesdeclast_si128(c, keysched[i + 1]);
104	out[3] = _mm_aesdeclast_si128(d, keysched[i + 1]);
105	out[4] = _mm_aesdeclast_si128(e, keysched[i + 1]);
106	out[5] = _mm_aesdeclast_si128(f, keysched[i + 1]);
107	out[6] = _mm_aesdeclast_si128(g, keysched[i + 1]);
108	out[7] = _mm_aesdeclast_si128(h, keysched[i + 1]);
109}
110
111/* rounds is passed in as rounds - 1 */
112static inline __m128i
113aesni_enc(int rounds, const __m128i *keysched, const __m128i from)
114{
115	__m128i tmp;
116	int i;
117
118	tmp = from ^ keysched[0];
119	for (i = 1; i < rounds; i += 2) {
120		tmp = _mm_aesenc_si128(tmp, keysched[i]);
121		tmp = _mm_aesenc_si128(tmp, keysched[i + 1]);
122	}
123
124	tmp = _mm_aesenc_si128(tmp, keysched[rounds]);
125	return _mm_aesenclast_si128(tmp, keysched[rounds + 1]);
126}
127
128static inline __m128i
129aesni_dec(int rounds, const __m128i *keysched, const __m128i from)
130{
131	__m128i tmp;
132	int i;
133
134	tmp = from ^ keysched[0];
135
136	for (i = 1; i < rounds; i += 2) {
137		tmp = _mm_aesdec_si128(tmp, keysched[i]);
138		tmp = _mm_aesdec_si128(tmp, keysched[i + 1]);
139	}
140
141	tmp = _mm_aesdec_si128(tmp, keysched[rounds]);
142	return _mm_aesdeclast_si128(tmp, keysched[rounds + 1]);
143}
144
145#endif /* _AESENCDEC_H_ */
146