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 * $FreeBSD$ 29 * 30 */ 31 32 #ifndef _AESENCDEC_H_ 33 #define _AESENCDEC_H_ 34 35 #include <crypto/aesni/aesni_os.h> 36 37 #include <wmmintrin.h> 38 39 static inline void 40 aesni_enc8(int rounds, const __m128i *keysched, __m128i a, 41 __m128i b, __m128i c, __m128i d, __m128i e, __m128i f, __m128i g, 42 __m128i h, __m128i out[8]) 43 { 44 int i; 45 46 a ^= keysched[0]; 47 b ^= keysched[0]; 48 c ^= keysched[0]; 49 d ^= keysched[0]; 50 e ^= keysched[0]; 51 f ^= keysched[0]; 52 g ^= keysched[0]; 53 h ^= keysched[0]; 54 55 for (i = 0; i < rounds; i++) { 56 a = _mm_aesenc_si128(a, keysched[i + 1]); 57 b = _mm_aesenc_si128(b, keysched[i + 1]); 58 c = _mm_aesenc_si128(c, keysched[i + 1]); 59 d = _mm_aesenc_si128(d, keysched[i + 1]); 60 e = _mm_aesenc_si128(e, keysched[i + 1]); 61 f = _mm_aesenc_si128(f, keysched[i + 1]); 62 g = _mm_aesenc_si128(g, keysched[i + 1]); 63 h = _mm_aesenc_si128(h, keysched[i + 1]); 64 } 65 66 out[0] = _mm_aesenclast_si128(a, keysched[i + 1]); 67 out[1] = _mm_aesenclast_si128(b, keysched[i + 1]); 68 out[2] = _mm_aesenclast_si128(c, keysched[i + 1]); 69 out[3] = _mm_aesenclast_si128(d, keysched[i + 1]); 70 out[4] = _mm_aesenclast_si128(e, keysched[i + 1]); 71 out[5] = _mm_aesenclast_si128(f, keysched[i + 1]); 72 out[6] = _mm_aesenclast_si128(g, keysched[i + 1]); 73 out[7] = _mm_aesenclast_si128(h, keysched[i + 1]); 74 } 75 76 static inline void 77 aesni_dec8(int rounds, const __m128i *keysched, __m128i a, 78 __m128i b, __m128i c, __m128i d, __m128i e, __m128i f, __m128i g, 79 __m128i h, __m128i out[8]) 80 { 81 int i; 82 83 a ^= keysched[0]; 84 b ^= keysched[0]; 85 c ^= keysched[0]; 86 d ^= keysched[0]; 87 e ^= keysched[0]; 88 f ^= keysched[0]; 89 g ^= keysched[0]; 90 h ^= keysched[0]; 91 92 for (i = 0; i < rounds; i++) { 93 a = _mm_aesdec_si128(a, keysched[i + 1]); 94 b = _mm_aesdec_si128(b, keysched[i + 1]); 95 c = _mm_aesdec_si128(c, keysched[i + 1]); 96 d = _mm_aesdec_si128(d, keysched[i + 1]); 97 e = _mm_aesdec_si128(e, keysched[i + 1]); 98 f = _mm_aesdec_si128(f, keysched[i + 1]); 99 g = _mm_aesdec_si128(g, keysched[i + 1]); 100 h = _mm_aesdec_si128(h, keysched[i + 1]); 101 } 102 103 out[0] = _mm_aesdeclast_si128(a, keysched[i + 1]); 104 out[1] = _mm_aesdeclast_si128(b, keysched[i + 1]); 105 out[2] = _mm_aesdeclast_si128(c, keysched[i + 1]); 106 out[3] = _mm_aesdeclast_si128(d, keysched[i + 1]); 107 out[4] = _mm_aesdeclast_si128(e, keysched[i + 1]); 108 out[5] = _mm_aesdeclast_si128(f, keysched[i + 1]); 109 out[6] = _mm_aesdeclast_si128(g, keysched[i + 1]); 110 out[7] = _mm_aesdeclast_si128(h, keysched[i + 1]); 111 } 112 113 /* rounds is passed in as rounds - 1 */ 114 static inline __m128i 115 aesni_enc(int rounds, const __m128i *keysched, const __m128i from) 116 { 117 __m128i tmp; 118 int i; 119 120 tmp = from ^ keysched[0]; 121 for (i = 1; i < rounds; i += 2) { 122 tmp = _mm_aesenc_si128(tmp, keysched[i]); 123 tmp = _mm_aesenc_si128(tmp, keysched[i + 1]); 124 } 125 126 tmp = _mm_aesenc_si128(tmp, keysched[rounds]); 127 return _mm_aesenclast_si128(tmp, keysched[rounds + 1]); 128 } 129 130 static inline __m128i 131 aesni_dec(int rounds, const __m128i *keysched, const __m128i from) 132 { 133 __m128i tmp; 134 int i; 135 136 tmp = from ^ keysched[0]; 137 138 for (i = 1; i < rounds; i += 2) { 139 tmp = _mm_aesdec_si128(tmp, keysched[i]); 140 tmp = _mm_aesdec_si128(tmp, keysched[i + 1]); 141 } 142 143 tmp = _mm_aesdec_si128(tmp, keysched[rounds]); 144 return _mm_aesdeclast_si128(tmp, keysched[rounds + 1]); 145 } 146 147 #endif /* _AESENCDEC_H_ */ 148