1 /*- 2 * Copyright 2013 John-Mark Gurney <jmg@FreeBSD.org> 3 * All rights reserved. 4 * 5 * Redistribution and use in source and binary forms, with or without 6 * modification, are permitted provided that the following conditions 7 * are met: 8 * 1. Redistributions of source code must retain the above copyright 9 * notice, this list of conditions and the following disclaimer. 10 * 2. Redistributions in binary form must reproduce the above copyright 11 * notice, this list of conditions and the following disclaimer in the 12 * documentation and/or other materials provided with the distribution. 13 * 14 * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND 15 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 16 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 17 * ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE 18 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 19 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 20 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 21 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 22 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 23 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 24 * SUCH DAMAGE. 25 * 26 * $FreeBSD$ 27 * 28 */ 29 30 #include <wmmintrin.h> 31 32 static inline void 33 aesni_enc8(int rounds, const uint8_t *key_schedule, __m128i a, 34 __m128i b, __m128i c, __m128i d, __m128i e, __m128i f, __m128i g, 35 __m128i h, __m128i out[8]) 36 { 37 const __m128i *keysched = (const __m128i *)key_schedule; 38 int i; 39 40 a ^= keysched[0]; 41 b ^= keysched[0]; 42 c ^= keysched[0]; 43 d ^= keysched[0]; 44 e ^= keysched[0]; 45 f ^= keysched[0]; 46 g ^= keysched[0]; 47 h ^= keysched[0]; 48 49 for (i = 0; i < rounds; i++) { 50 a = _mm_aesenc_si128(a, keysched[i + 1]); 51 b = _mm_aesenc_si128(b, keysched[i + 1]); 52 c = _mm_aesenc_si128(c, keysched[i + 1]); 53 d = _mm_aesenc_si128(d, keysched[i + 1]); 54 e = _mm_aesenc_si128(e, keysched[i + 1]); 55 f = _mm_aesenc_si128(f, keysched[i + 1]); 56 g = _mm_aesenc_si128(g, keysched[i + 1]); 57 h = _mm_aesenc_si128(h, keysched[i + 1]); 58 } 59 60 out[0] = _mm_aesenclast_si128(a, keysched[i + 1]); 61 out[1] = _mm_aesenclast_si128(b, keysched[i + 1]); 62 out[2] = _mm_aesenclast_si128(c, keysched[i + 1]); 63 out[3] = _mm_aesenclast_si128(d, keysched[i + 1]); 64 out[4] = _mm_aesenclast_si128(e, keysched[i + 1]); 65 out[5] = _mm_aesenclast_si128(f, keysched[i + 1]); 66 out[6] = _mm_aesenclast_si128(g, keysched[i + 1]); 67 out[7] = _mm_aesenclast_si128(h, keysched[i + 1]); 68 } 69 70 static inline void 71 aesni_dec8(int rounds, const uint8_t *key_schedule, __m128i a, 72 __m128i b, __m128i c, __m128i d, __m128i e, __m128i f, __m128i g, 73 __m128i h, __m128i out[8]) 74 { 75 const __m128i *keysched = (const __m128i *)key_schedule; 76 int i; 77 78 a ^= keysched[0]; 79 b ^= keysched[0]; 80 c ^= keysched[0]; 81 d ^= keysched[0]; 82 e ^= keysched[0]; 83 f ^= keysched[0]; 84 g ^= keysched[0]; 85 h ^= keysched[0]; 86 87 for (i = 0; i < rounds; i++) { 88 a = _mm_aesdec_si128(a, keysched[i + 1]); 89 b = _mm_aesdec_si128(b, keysched[i + 1]); 90 c = _mm_aesdec_si128(c, keysched[i + 1]); 91 d = _mm_aesdec_si128(d, keysched[i + 1]); 92 e = _mm_aesdec_si128(e, keysched[i + 1]); 93 f = _mm_aesdec_si128(f, keysched[i + 1]); 94 g = _mm_aesdec_si128(g, keysched[i + 1]); 95 h = _mm_aesdec_si128(h, keysched[i + 1]); 96 } 97 98 out[0] = _mm_aesdeclast_si128(a, keysched[i + 1]); 99 out[1] = _mm_aesdeclast_si128(b, keysched[i + 1]); 100 out[2] = _mm_aesdeclast_si128(c, keysched[i + 1]); 101 out[3] = _mm_aesdeclast_si128(d, keysched[i + 1]); 102 out[4] = _mm_aesdeclast_si128(e, keysched[i + 1]); 103 out[5] = _mm_aesdeclast_si128(f, keysched[i + 1]); 104 out[6] = _mm_aesdeclast_si128(g, keysched[i + 1]); 105 out[7] = _mm_aesdeclast_si128(h, keysched[i + 1]); 106 } 107 108 static inline __m128i 109 aesni_enc(int rounds, const uint8_t *key_schedule, const __m128i from) 110 { 111 __m128i tmp; 112 const __m128i *keysched = (const __m128i *)key_schedule; 113 int i; 114 115 tmp = from ^ keysched[0]; 116 117 for (i = 0; i < rounds; i++) 118 tmp = _mm_aesenc_si128(tmp, keysched[i + 1]); 119 120 return _mm_aesenclast_si128(tmp, keysched[i + 1]); 121 } 122 123 static inline __m128i 124 aesni_dec(int rounds, const uint8_t *key_schedule, const __m128i from) 125 { 126 __m128i tmp; 127 const __m128i *keysched = (const __m128i *)key_schedule; 128 int i; 129 130 tmp = from ^ keysched[0]; 131 132 for (i = 0; i < rounds; i++) 133 tmp = _mm_aesdec_si128(tmp, keysched[i + 1]); 134 135 return _mm_aesdeclast_si128(tmp, keysched[i + 1]); 136 } 137