diff options
author | sam <sam@FreeBSD.org> | 2006-03-07 05:47:04 +0000 |
---|---|---|
committer | sam <sam@FreeBSD.org> | 2006-03-07 05:47:04 +0000 |
commit | 8d55057fb42bf9070fd379acbcb6fc4ef793d2a7 (patch) | |
tree | dacc7977efdefefb8b105113edeb5215c0e01234 /contrib/hostapd/aes.c | |
parent | d1a1fd4aa94cd9c5cb443c4c1337f91c8c46fde0 (diff) | |
download | FreeBSD-src-8d55057fb42bf9070fd379acbcb6fc4ef793d2a7.zip FreeBSD-src-8d55057fb42bf9070fd379acbcb6fc4ef793d2a7.tar.gz |
Import of hostapd 0.4.8
Diffstat (limited to 'contrib/hostapd/aes.c')
-rw-r--r-- | contrib/hostapd/aes.c | 555 |
1 files changed, 264 insertions, 291 deletions
diff --git a/contrib/hostapd/aes.c b/contrib/hostapd/aes.c index eabebd0..ce94778 100644 --- a/contrib/hostapd/aes.c +++ b/contrib/hostapd/aes.c @@ -1,8 +1,15 @@ /* + * AES (Rijndael) cipher + * * Modifications to public domain implementation: * - support only 128-bit keys * - cleanup - * Copyright (c) 2003-2004, Jouni Malinen <jkmaline@cc.hut.fi> + * - use C pre-processor to make it easier to change S table access + * - added option (AES_SMALL_TABLES) for reducing code size by about 8 kB at + * cost of reduced throughput (quite small difference on Pentium 4, + * 10-25% when using -O1 or -O2 optimization) + * + * Copyright (c) 2003-2005, Jouni Malinen <jkmaline@cc.hut.fi> * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License version 2 as @@ -14,7 +21,7 @@ * See README and COPYING for more details. */ -/** +/* * rijndael-alg-fst.c * * @version 3.0 (December 2000) @@ -40,7 +47,8 @@ * EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. */ -#define FULL_UNROLL +/* #define FULL_UNROLL */ +#define AES_SMALL_TABLES /* @@ -123,6 +131,7 @@ static const u32 Te0[256] = { 0x824141c3U, 0x299999b0U, 0x5a2d2d77U, 0x1e0f0f11U, 0x7bb0b0cbU, 0xa85454fcU, 0x6dbbbbd6U, 0x2c16163aU, }; +#ifndef AES_SMALL_TABLES static const u32 Te1[256] = { 0xa5c66363U, 0x84f87c7cU, 0x99ee7777U, 0x8df67b7bU, 0x0dfff2f2U, 0xbdd66b6bU, 0xb1de6f6fU, 0x5491c5c5U, @@ -388,6 +397,7 @@ static const u32 Te4[256] = { 0x41414141U, 0x99999999U, 0x2d2d2d2dU, 0x0f0f0f0fU, 0xb0b0b0b0U, 0x54545454U, 0xbbbbbbbbU, 0x16161616U, }; +#endif /* AES_SMALL_TABLES */ static const u32 Td0[256] = { 0x51f4a750U, 0x7e416553U, 0x1a17a4c3U, 0x3a275e96U, 0x3bab6bcbU, 0x1f9d45f1U, 0xacfa58abU, 0x4be30393U, @@ -454,6 +464,7 @@ static const u32 Td0[256] = { 0x39a80171U, 0x080cb3deU, 0xd8b4e49cU, 0x6456c190U, 0x7bcb8461U, 0xd532b670U, 0x486c5c74U, 0xd0b85742U, }; +#ifndef AES_SMALL_TABLES static const u32 Td1[256] = { 0x5051f4a7U, 0x537e4165U, 0xc31a17a4U, 0x963a275eU, 0xcb3bab6bU, 0xf11f9d45U, 0xabacfa58U, 0x934be303U, @@ -724,6 +735,116 @@ static const u32 rcon[] = { 0x10000000, 0x20000000, 0x40000000, 0x80000000, 0x1B000000, 0x36000000, /* for 128-bit blocks, Rijndael never uses more than 10 rcon values */ }; +#else /* AES_SMALL_TABLES */ +static const u8 Td4s[256] = { + 0x52U, 0x09U, 0x6aU, 0xd5U, 0x30U, 0x36U, 0xa5U, 0x38U, + 0xbfU, 0x40U, 0xa3U, 0x9eU, 0x81U, 0xf3U, 0xd7U, 0xfbU, + 0x7cU, 0xe3U, 0x39U, 0x82U, 0x9bU, 0x2fU, 0xffU, 0x87U, + 0x34U, 0x8eU, 0x43U, 0x44U, 0xc4U, 0xdeU, 0xe9U, 0xcbU, + 0x54U, 0x7bU, 0x94U, 0x32U, 0xa6U, 0xc2U, 0x23U, 0x3dU, + 0xeeU, 0x4cU, 0x95U, 0x0bU, 0x42U, 0xfaU, 0xc3U, 0x4eU, + 0x08U, 0x2eU, 0xa1U, 0x66U, 0x28U, 0xd9U, 0x24U, 0xb2U, + 0x76U, 0x5bU, 0xa2U, 0x49U, 0x6dU, 0x8bU, 0xd1U, 0x25U, + 0x72U, 0xf8U, 0xf6U, 0x64U, 0x86U, 0x68U, 0x98U, 0x16U, + 0xd4U, 0xa4U, 0x5cU, 0xccU, 0x5dU, 0x65U, 0xb6U, 0x92U, + 0x6cU, 0x70U, 0x48U, 0x50U, 0xfdU, 0xedU, 0xb9U, 0xdaU, + 0x5eU, 0x15U, 0x46U, 0x57U, 0xa7U, 0x8dU, 0x9dU, 0x84U, + 0x90U, 0xd8U, 0xabU, 0x00U, 0x8cU, 0xbcU, 0xd3U, 0x0aU, + 0xf7U, 0xe4U, 0x58U, 0x05U, 0xb8U, 0xb3U, 0x45U, 0x06U, + 0xd0U, 0x2cU, 0x1eU, 0x8fU, 0xcaU, 0x3fU, 0x0fU, 0x02U, + 0xc1U, 0xafU, 0xbdU, 0x03U, 0x01U, 0x13U, 0x8aU, 0x6bU, + 0x3aU, 0x91U, 0x11U, 0x41U, 0x4fU, 0x67U, 0xdcU, 0xeaU, + 0x97U, 0xf2U, 0xcfU, 0xceU, 0xf0U, 0xb4U, 0xe6U, 0x73U, + 0x96U, 0xacU, 0x74U, 0x22U, 0xe7U, 0xadU, 0x35U, 0x85U, + 0xe2U, 0xf9U, 0x37U, 0xe8U, 0x1cU, 0x75U, 0xdfU, 0x6eU, + 0x47U, 0xf1U, 0x1aU, 0x71U, 0x1dU, 0x29U, 0xc5U, 0x89U, + 0x6fU, 0xb7U, 0x62U, 0x0eU, 0xaaU, 0x18U, 0xbeU, 0x1bU, + 0xfcU, 0x56U, 0x3eU, 0x4bU, 0xc6U, 0xd2U, 0x79U, 0x20U, + 0x9aU, 0xdbU, 0xc0U, 0xfeU, 0x78U, 0xcdU, 0x5aU, 0xf4U, + 0x1fU, 0xddU, 0xa8U, 0x33U, 0x88U, 0x07U, 0xc7U, 0x31U, + 0xb1U, 0x12U, 0x10U, 0x59U, 0x27U, 0x80U, 0xecU, 0x5fU, + 0x60U, 0x51U, 0x7fU, 0xa9U, 0x19U, 0xb5U, 0x4aU, 0x0dU, + 0x2dU, 0xe5U, 0x7aU, 0x9fU, 0x93U, 0xc9U, 0x9cU, 0xefU, + 0xa0U, 0xe0U, 0x3bU, 0x4dU, 0xaeU, 0x2aU, 0xf5U, 0xb0U, + 0xc8U, 0xebU, 0xbbU, 0x3cU, 0x83U, 0x53U, 0x99U, 0x61U, + 0x17U, 0x2bU, 0x04U, 0x7eU, 0xbaU, 0x77U, 0xd6U, 0x26U, + 0xe1U, 0x69U, 0x14U, 0x63U, 0x55U, 0x21U, 0x0cU, 0x7dU, +}; +static const u8 rcons[] = { + 0x01, 0x02, 0x04, 0x08, 0x10, 0x20, 0x40, 0x80, 0x1B, 0x36 + /* for 128-bit blocks, Rijndael never uses more than 10 rcon values */ +}; +#endif /* AES_SMALL_TABLES */ + + +#ifndef AES_SMALL_TABLES + +#define RCON(i) rcon[(i)] + +#define TE0(i) Te0[((i) >> 24) & 0xff] +#define TE1(i) Te1[((i) >> 16) & 0xff] +#define TE2(i) Te2[((i) >> 8) & 0xff] +#define TE3(i) Te3[(i) & 0xff] +#define TE41(i) (Te4[((i) >> 24) & 0xff] & 0xff000000) +#define TE42(i) (Te4[((i) >> 16) & 0xff] & 0x00ff0000) +#define TE43(i) (Te4[((i) >> 8) & 0xff] & 0x0000ff00) +#define TE44(i) (Te4[(i) & 0xff] & 0x000000ff) +#define TE421(i) (Te4[((i) >> 16) & 0xff] & 0xff000000) +#define TE432(i) (Te4[((i) >> 8) & 0xff] & 0x00ff0000) +#define TE443(i) (Te4[(i) & 0xff] & 0x0000ff00) +#define TE414(i) (Te4[((i) >> 24) & 0xff] & 0x000000ff) +#define TE4(i) (Te4[(i)] & 0x000000ff) + +#define TD0(i) Td0[((i) >> 24) & 0xff] +#define TD1(i) Td1[((i) >> 16) & 0xff] +#define TD2(i) Td2[((i) >> 8) & 0xff] +#define TD3(i) Td3[(i) & 0xff] +#define TD41(i) (Td4[((i) >> 24) & 0xff] & 0xff000000) +#define TD42(i) (Td4[((i) >> 16) & 0xff] & 0x00ff0000) +#define TD43(i) (Td4[((i) >> 8) & 0xff] & 0x0000ff00) +#define TD44(i) (Td4[(i) & 0xff] & 0x000000ff) +#define TD0_(i) Td0[(i) & 0xff] +#define TD1_(i) Td1[(i) & 0xff] +#define TD2_(i) Td2[(i) & 0xff] +#define TD3_(i) Td3[(i) & 0xff] + +#else /* AES_SMALL_TABLES */ + +#define RCON(i) (rcons[(i)] << 24) + +static inline u32 rotr(u32 val, int bits) +{ + return (val >> bits) | (val << (32 - bits)); +} + +#define TE0(i) Te0[((i) >> 24) & 0xff] +#define TE1(i) rotr(Te0[((i) >> 16) & 0xff], 8) +#define TE2(i) rotr(Te0[((i) >> 8) & 0xff], 16) +#define TE3(i) rotr(Te0[(i) & 0xff], 24) +#define TE41(i) ((Te0[((i) >> 24) & 0xff] << 8) & 0xff000000) +#define TE42(i) (Te0[((i) >> 16) & 0xff] & 0x00ff0000) +#define TE43(i) (Te0[((i) >> 8) & 0xff] & 0x0000ff00) +#define TE44(i) ((Te0[(i) & 0xff] >> 8) & 0x000000ff) +#define TE421(i) ((Te0[((i) >> 16) & 0xff] << 8) & 0xff000000) +#define TE432(i) (Te0[((i) >> 8) & 0xff] & 0x00ff0000) +#define TE443(i) (Te0[(i) & 0xff] & 0x0000ff00) +#define TE414(i) ((Te0[((i) >> 24) & 0xff] >> 8) & 0x000000ff) +#define TE4(i) ((Te0[(i)] >> 8) & 0x000000ff) + +#define TD0(i) Td0[((i) >> 24) & 0xff] +#define TD1(i) rotr(Td0[((i) >> 16) & 0xff], 8) +#define TD2(i) rotr(Td0[((i) >> 8) & 0xff], 16) +#define TD3(i) rotr(Td0[(i) & 0xff], 24) +#define TD41(i) (Td4s[((i) >> 24) & 0xff] << 24) +#define TD42(i) (Td4s[((i) >> 16) & 0xff] << 16) +#define TD43(i) (Td4s[((i) >> 8) & 0xff] << 8) +#define TD44(i) (Td4s[(i) & 0xff]) +#define TD0_(i) Td0[(i) & 0xff] +#define TD1_(i) rotr(Td0[(i) & 0xff], 8) +#define TD2_(i) rotr(Td0[(i) & 0xff], 16) +#define TD3_(i) rotr(Td0[(i) & 0xff], 24) + +#endif /* AES_SMALL_TABLES */ #define SWAP(x) (_lrotl(x, 8) & 0x00ff00ff | _lrotr(x, 8) & 0xff00ff00) @@ -755,11 +876,8 @@ void rijndaelKeySetupEnc(u32 rk[/*44*/], const u8 cipherKey[]) for (i = 0; i < 10; i++) { temp = rk[3]; rk[4] = rk[0] ^ - (Te4[(temp >> 16) & 0xff] & 0xff000000) ^ - (Te4[(temp >> 8) & 0xff] & 0x00ff0000) ^ - (Te4[(temp ) & 0xff] & 0x0000ff00) ^ - (Te4[(temp >> 24) ] & 0x000000ff) ^ - rcon[i]; + TE421(temp) ^ TE432(temp) ^ TE443(temp) ^ TE414(temp) ^ + RCON(i); rk[5] = rk[1] ^ rk[4]; rk[6] = rk[2] ^ rk[5]; rk[7] = rk[3] ^ rk[6]; @@ -790,33 +908,19 @@ void rijndaelKeySetupDec(u32 rk[/*44*/], const u8 cipherKey[]) * first and the last: */ for (i = 1; i < Nr; i++) { rk += 4; - rk[0] = - Td0[Te4[(rk[0] >> 24) ] & 0xff] ^ - Td1[Te4[(rk[0] >> 16) & 0xff] & 0xff] ^ - Td2[Te4[(rk[0] >> 8) & 0xff] & 0xff] ^ - Td3[Te4[(rk[0] ) & 0xff] & 0xff]; - rk[1] = - Td0[Te4[(rk[1] >> 24) ] & 0xff] ^ - Td1[Te4[(rk[1] >> 16) & 0xff] & 0xff] ^ - Td2[Te4[(rk[1] >> 8) & 0xff] & 0xff] ^ - Td3[Te4[(rk[1] ) & 0xff] & 0xff]; - rk[2] = - Td0[Te4[(rk[2] >> 24) ] & 0xff] ^ - Td1[Te4[(rk[2] >> 16) & 0xff] & 0xff] ^ - Td2[Te4[(rk[2] >> 8) & 0xff] & 0xff] ^ - Td3[Te4[(rk[2] ) & 0xff] & 0xff]; - rk[3] = - Td0[Te4[(rk[3] >> 24) ] & 0xff] ^ - Td1[Te4[(rk[3] >> 16) & 0xff] & 0xff] ^ - Td2[Te4[(rk[3] >> 8) & 0xff] & 0xff] ^ - Td3[Te4[(rk[3] ) & 0xff] & 0xff]; + for (j = 0; j < 4; j++) { + rk[j] = TD0_(TE4((rk[j] >> 24) )) ^ + TD1_(TE4((rk[j] >> 16) & 0xff)) ^ + TD2_(TE4((rk[j] >> 8) & 0xff)) ^ + TD3_(TE4((rk[j] ) & 0xff)); + } } } void rijndaelEncrypt(const u32 rk[/*44*/], const u8 pt[16], u8 ct[16]) { u32 s0, s1, s2, s3, t0, t1, t2, t3; - int Nr = 10; + const int Nr = 10; #ifndef FULL_UNROLL int r; #endif /* ?FULL_UNROLL */ @@ -829,153 +933,61 @@ void rijndaelEncrypt(const u32 rk[/*44*/], const u8 pt[16], u8 ct[16]) s1 = GETU32(pt + 4) ^ rk[1]; s2 = GETU32(pt + 8) ^ rk[2]; s3 = GETU32(pt + 12) ^ rk[3]; + +#define ROUND(i,d,s) \ +d##0 = TE0(s##0) ^ TE1(s##1) ^ TE2(s##2) ^ TE3(s##3) ^ rk[4 * i]; \ +d##1 = TE0(s##1) ^ TE1(s##2) ^ TE2(s##3) ^ TE3(s##0) ^ rk[4 * i + 1]; \ +d##2 = TE0(s##2) ^ TE1(s##3) ^ TE2(s##0) ^ TE3(s##1) ^ rk[4 * i + 2]; \ +d##3 = TE0(s##3) ^ TE1(s##0) ^ TE2(s##1) ^ TE3(s##2) ^ rk[4 * i + 3] + #ifdef FULL_UNROLL - /* round 1: */ - t0 = Te0[s0 >> 24] ^ Te1[(s1 >> 16) & 0xff] ^ Te2[(s2 >> 8) & 0xff] ^ Te3[s3 & 0xff] ^ rk[ 4]; - t1 = Te0[s1 >> 24] ^ Te1[(s2 >> 16) & 0xff] ^ Te2[(s3 >> 8) & 0xff] ^ Te3[s0 & 0xff] ^ rk[ 5]; - t2 = Te0[s2 >> 24] ^ Te1[(s3 >> 16) & 0xff] ^ Te2[(s0 >> 8) & 0xff] ^ Te3[s1 & 0xff] ^ rk[ 6]; - t3 = Te0[s3 >> 24] ^ Te1[(s0 >> 16) & 0xff] ^ Te2[(s1 >> 8) & 0xff] ^ Te3[s2 & 0xff] ^ rk[ 7]; - /* round 2: */ - s0 = Te0[t0 >> 24] ^ Te1[(t1 >> 16) & 0xff] ^ Te2[(t2 >> 8) & 0xff] ^ Te3[t3 & 0xff] ^ rk[ 8]; - s1 = Te0[t1 >> 24] ^ Te1[(t2 >> 16) & 0xff] ^ Te2[(t3 >> 8) & 0xff] ^ Te3[t0 & 0xff] ^ rk[ 9]; - s2 = Te0[t2 >> 24] ^ Te1[(t3 >> 16) & 0xff] ^ Te2[(t0 >> 8) & 0xff] ^ Te3[t1 & 0xff] ^ rk[10]; - s3 = Te0[t3 >> 24] ^ Te1[(t0 >> 16) & 0xff] ^ Te2[(t1 >> 8) & 0xff] ^ Te3[t2 & 0xff] ^ rk[11]; - /* round 3: */ - t0 = Te0[s0 >> 24] ^ Te1[(s1 >> 16) & 0xff] ^ Te2[(s2 >> 8) & 0xff] ^ Te3[s3 & 0xff] ^ rk[12]; - t1 = Te0[s1 >> 24] ^ Te1[(s2 >> 16) & 0xff] ^ Te2[(s3 >> 8) & 0xff] ^ Te3[s0 & 0xff] ^ rk[13]; - t2 = Te0[s2 >> 24] ^ Te1[(s3 >> 16) & 0xff] ^ Te2[(s0 >> 8) & 0xff] ^ Te3[s1 & 0xff] ^ rk[14]; - t3 = Te0[s3 >> 24] ^ Te1[(s0 >> 16) & 0xff] ^ Te2[(s1 >> 8) & 0xff] ^ Te3[s2 & 0xff] ^ rk[15]; - /* round 4: */ - s0 = Te0[t0 >> 24] ^ Te1[(t1 >> 16) & 0xff] ^ Te2[(t2 >> 8) & 0xff] ^ Te3[t3 & 0xff] ^ rk[16]; - s1 = Te0[t1 >> 24] ^ Te1[(t2 >> 16) & 0xff] ^ Te2[(t3 >> 8) & 0xff] ^ Te3[t0 & 0xff] ^ rk[17]; - s2 = Te0[t2 >> 24] ^ Te1[(t3 >> 16) & 0xff] ^ Te2[(t0 >> 8) & 0xff] ^ Te3[t1 & 0xff] ^ rk[18]; - s3 = Te0[t3 >> 24] ^ Te1[(t0 >> 16) & 0xff] ^ Te2[(t1 >> 8) & 0xff] ^ Te3[t2 & 0xff] ^ rk[19]; - /* round 5: */ - t0 = Te0[s0 >> 24] ^ Te1[(s1 >> 16) & 0xff] ^ Te2[(s2 >> 8) & 0xff] ^ Te3[s3 & 0xff] ^ rk[20]; - t1 = Te0[s1 >> 24] ^ Te1[(s2 >> 16) & 0xff] ^ Te2[(s3 >> 8) & 0xff] ^ Te3[s0 & 0xff] ^ rk[21]; - t2 = Te0[s2 >> 24] ^ Te1[(s3 >> 16) & 0xff] ^ Te2[(s0 >> 8) & 0xff] ^ Te3[s1 & 0xff] ^ rk[22]; - t3 = Te0[s3 >> 24] ^ Te1[(s0 >> 16) & 0xff] ^ Te2[(s1 >> 8) & 0xff] ^ Te3[s2 & 0xff] ^ rk[23]; - /* round 6: */ - s0 = Te0[t0 >> 24] ^ Te1[(t1 >> 16) & 0xff] ^ Te2[(t2 >> 8) & 0xff] ^ Te3[t3 & 0xff] ^ rk[24]; - s1 = Te0[t1 >> 24] ^ Te1[(t2 >> 16) & 0xff] ^ Te2[(t3 >> 8) & 0xff] ^ Te3[t0 & 0xff] ^ rk[25]; - s2 = Te0[t2 >> 24] ^ Te1[(t3 >> 16) & 0xff] ^ Te2[(t0 >> 8) & 0xff] ^ Te3[t1 & 0xff] ^ rk[26]; - s3 = Te0[t3 >> 24] ^ Te1[(t0 >> 16) & 0xff] ^ Te2[(t1 >> 8) & 0xff] ^ Te3[t2 & 0xff] ^ rk[27]; - /* round 7: */ - t0 = Te0[s0 >> 24] ^ Te1[(s1 >> 16) & 0xff] ^ Te2[(s2 >> 8) & 0xff] ^ Te3[s3 & 0xff] ^ rk[28]; - t1 = Te0[s1 >> 24] ^ Te1[(s2 >> 16) & 0xff] ^ Te2[(s3 >> 8) & 0xff] ^ Te3[s0 & 0xff] ^ rk[29]; - t2 = Te0[s2 >> 24] ^ Te1[(s3 >> 16) & 0xff] ^ Te2[(s0 >> 8) & 0xff] ^ Te3[s1 & 0xff] ^ rk[30]; - t3 = Te0[s3 >> 24] ^ Te1[(s0 >> 16) & 0xff] ^ Te2[(s1 >> 8) & 0xff] ^ Te3[s2 & 0xff] ^ rk[31]; - /* round 8: */ - s0 = Te0[t0 >> 24] ^ Te1[(t1 >> 16) & 0xff] ^ Te2[(t2 >> 8) & 0xff] ^ Te3[t3 & 0xff] ^ rk[32]; - s1 = Te0[t1 >> 24] ^ Te1[(t2 >> 16) & 0xff] ^ Te2[(t3 >> 8) & 0xff] ^ Te3[t0 & 0xff] ^ rk[33]; - s2 = Te0[t2 >> 24] ^ Te1[(t3 >> 16) & 0xff] ^ Te2[(t0 >> 8) & 0xff] ^ Te3[t1 & 0xff] ^ rk[34]; - s3 = Te0[t3 >> 24] ^ Te1[(t0 >> 16) & 0xff] ^ Te2[(t1 >> 8) & 0xff] ^ Te3[t2 & 0xff] ^ rk[35]; - /* round 9: */ - t0 = Te0[s0 >> 24] ^ Te1[(s1 >> 16) & 0xff] ^ Te2[(s2 >> 8) & 0xff] ^ Te3[s3 & 0xff] ^ rk[36]; - t1 = Te0[s1 >> 24] ^ Te1[(s2 >> 16) & 0xff] ^ Te2[(s3 >> 8) & 0xff] ^ Te3[s0 & 0xff] ^ rk[37]; - t2 = Te0[s2 >> 24] ^ Te1[(s3 >> 16) & 0xff] ^ Te2[(s0 >> 8) & 0xff] ^ Te3[s1 & 0xff] ^ rk[38]; - t3 = Te0[s3 >> 24] ^ Te1[(s0 >> 16) & 0xff] ^ Te2[(s1 >> 8) & 0xff] ^ Te3[s2 & 0xff] ^ rk[39]; - rk += Nr << 2; + + ROUND(1,t,s); + ROUND(2,s,t); + ROUND(3,t,s); + ROUND(4,s,t); + ROUND(5,t,s); + ROUND(6,s,t); + ROUND(7,t,s); + ROUND(8,s,t); + ROUND(9,t,s); + + rk += Nr << 2; + #else /* !FULL_UNROLL */ - /* - * Nr - 1 full rounds: - */ - r = Nr >> 1; - for (;;) { - t0 = - Te0[(s0 >> 24) ] ^ - Te1[(s1 >> 16) & 0xff] ^ - Te2[(s2 >> 8) & 0xff] ^ - Te3[(s3 ) & 0xff] ^ - rk[4]; - t1 = - Te0[(s1 >> 24) ] ^ - Te1[(s2 >> 16) & 0xff] ^ - Te2[(s3 >> 8) & 0xff] ^ - Te3[(s0 ) & 0xff] ^ - rk[5]; - t2 = - Te0[(s2 >> 24) ] ^ - Te1[(s3 >> 16) & 0xff] ^ - Te2[(s0 >> 8) & 0xff] ^ - Te3[(s1 ) & 0xff] ^ - rk[6]; - t3 = - Te0[(s3 >> 24) ] ^ - Te1[(s0 >> 16) & 0xff] ^ - Te2[(s1 >> 8) & 0xff] ^ - Te3[(s2 ) & 0xff] ^ - rk[7]; - rk += 8; - if (--r == 0) { - break; - } + /* Nr - 1 full rounds: */ + r = Nr >> 1; + for (;;) { + ROUND(1,t,s); + rk += 8; + if (--r == 0) + break; + ROUND(0,s,t); + } - s0 = - Te0[(t0 >> 24) ] ^ - Te1[(t1 >> 16) & 0xff] ^ - Te2[(t2 >> 8) & 0xff] ^ - Te3[(t3 ) & 0xff] ^ - rk[0]; - s1 = - Te0[(t1 >> 24) ] ^ - Te1[(t2 >> 16) & 0xff] ^ - Te2[(t3 >> 8) & 0xff] ^ - Te3[(t0 ) & 0xff] ^ - rk[1]; - s2 = - Te0[(t2 >> 24) ] ^ - Te1[(t3 >> 16) & 0xff] ^ - Te2[(t0 >> 8) & 0xff] ^ - Te3[(t1 ) & 0xff] ^ - rk[2]; - s3 = - Te0[(t3 >> 24) ] ^ - Te1[(t0 >> 16) & 0xff] ^ - Te2[(t1 >> 8) & 0xff] ^ - Te3[(t2 ) & 0xff] ^ - rk[3]; - } #endif /* ?FULL_UNROLL */ - /* + +#undef ROUND + + /* * apply last round and * map cipher state to byte array block: */ - s0 = - (Te4[(t0 >> 24) ] & 0xff000000) ^ - (Te4[(t1 >> 16) & 0xff] & 0x00ff0000) ^ - (Te4[(t2 >> 8) & 0xff] & 0x0000ff00) ^ - (Te4[(t3 ) & 0xff] & 0x000000ff) ^ - rk[0]; + s0 = TE41(t0) ^ TE42(t1) ^ TE43(t2) ^ TE44(t3) ^ rk[0]; PUTU32(ct , s0); - s1 = - (Te4[(t1 >> 24) ] & 0xff000000) ^ - (Te4[(t2 >> 16) & 0xff] & 0x00ff0000) ^ - (Te4[(t3 >> 8) & 0xff] & 0x0000ff00) ^ - (Te4[(t0 ) & 0xff] & 0x000000ff) ^ - rk[1]; + s1 = TE41(t1) ^ TE42(t2) ^ TE43(t3) ^ TE44(t0) ^ rk[1]; PUTU32(ct + 4, s1); - s2 = - (Te4[(t2 >> 24) ] & 0xff000000) ^ - (Te4[(t3 >> 16) & 0xff] & 0x00ff0000) ^ - (Te4[(t0 >> 8) & 0xff] & 0x0000ff00) ^ - (Te4[(t1 ) & 0xff] & 0x000000ff) ^ - rk[2]; + s2 = TE41(t2) ^ TE42(t3) ^ TE43(t0) ^ TE44(t1) ^ rk[2]; PUTU32(ct + 8, s2); - s3 = - (Te4[(t3 >> 24) ] & 0xff000000) ^ - (Te4[(t0 >> 16) & 0xff] & 0x00ff0000) ^ - (Te4[(t1 >> 8) & 0xff] & 0x0000ff00) ^ - (Te4[(t2 ) & 0xff] & 0x000000ff) ^ - rk[3]; + s3 = TE41(t3) ^ TE42(t0) ^ TE43(t1) ^ TE44(t2) ^ rk[3]; PUTU32(ct + 12, s3); } void rijndaelDecrypt(const u32 rk[/*44*/], const u8 ct[16], u8 pt[16]) { u32 s0, s1, s2, s3, t0, t1, t2, t3; - int Nr = 10; + const int Nr = 10; #ifndef FULL_UNROLL int r; #endif /* ?FULL_UNROLL */ @@ -984,149 +996,110 @@ void rijndaelDecrypt(const u32 rk[/*44*/], const u8 ct[16], u8 pt[16]) * map byte array block to cipher state * and add initial round key: */ - s0 = GETU32(ct ) ^ rk[0]; - s1 = GETU32(ct + 4) ^ rk[1]; - s2 = GETU32(ct + 8) ^ rk[2]; - s3 = GETU32(ct + 12) ^ rk[3]; + s0 = GETU32(ct ) ^ rk[0]; + s1 = GETU32(ct + 4) ^ rk[1]; + s2 = GETU32(ct + 8) ^ rk[2]; + s3 = GETU32(ct + 12) ^ rk[3]; + +#define ROUND(i,d,s) \ +d##0 = TD0(s##0) ^ TD1(s##3) ^ TD2(s##2) ^ TD3(s##1) ^ rk[4 * i]; \ +d##1 = TD0(s##1) ^ TD1(s##0) ^ TD2(s##3) ^ TD3(s##2) ^ rk[4 * i + 1]; \ +d##2 = TD0(s##2) ^ TD1(s##1) ^ TD2(s##0) ^ TD3(s##3) ^ rk[4 * i + 2]; \ +d##3 = TD0(s##3) ^ TD1(s##2) ^ TD2(s##1) ^ TD3(s##0) ^ rk[4 * i + 3] + #ifdef FULL_UNROLL - /* round 1: */ - t0 = Td0[s0 >> 24] ^ Td1[(s3 >> 16) & 0xff] ^ Td2[(s2 >> 8) & 0xff] ^ Td3[s1 & 0xff] ^ rk[ 4]; - t1 = Td0[s1 >> 24] ^ Td1[(s0 >> 16) & 0xff] ^ Td2[(s3 >> 8) & 0xff] ^ Td3[s2 & 0xff] ^ rk[ 5]; - t2 = Td0[s2 >> 24] ^ Td1[(s1 >> 16) & 0xff] ^ Td2[(s0 >> 8) & 0xff] ^ Td3[s3 & 0xff] ^ rk[ 6]; - t3 = Td0[s3 >> 24] ^ Td1[(s2 >> 16) & 0xff] ^ Td2[(s1 >> 8) & 0xff] ^ Td3[s0 & 0xff] ^ rk[ 7]; - /* round 2: */ - s0 = Td0[t0 >> 24] ^ Td1[(t3 >> 16) & 0xff] ^ Td2[(t2 >> 8) & 0xff] ^ Td3[t1 & 0xff] ^ rk[ 8]; - s1 = Td0[t1 >> 24] ^ Td1[(t0 >> 16) & 0xff] ^ Td2[(t3 >> 8) & 0xff] ^ Td3[t2 & 0xff] ^ rk[ 9]; - s2 = Td0[t2 >> 24] ^ Td1[(t1 >> 16) & 0xff] ^ Td2[(t0 >> 8) & 0xff] ^ Td3[t3 & 0xff] ^ rk[10]; - s3 = Td0[t3 >> 24] ^ Td1[(t2 >> 16) & 0xff] ^ Td2[(t1 >> 8) & 0xff] ^ Td3[t0 & 0xff] ^ rk[11]; - /* round 3: */ - t0 = Td0[s0 >> 24] ^ Td1[(s3 >> 16) & 0xff] ^ Td2[(s2 >> 8) & 0xff] ^ Td3[s1 & 0xff] ^ rk[12]; - t1 = Td0[s1 >> 24] ^ Td1[(s0 >> 16) & 0xff] ^ Td2[(s3 >> 8) & 0xff] ^ Td3[s2 & 0xff] ^ rk[13]; - t2 = Td0[s2 >> 24] ^ Td1[(s1 >> 16) & 0xff] ^ Td2[(s0 >> 8) & 0xff] ^ Td3[s3 & 0xff] ^ rk[14]; - t3 = Td0[s3 >> 24] ^ Td1[(s2 >> 16) & 0xff] ^ Td2[(s1 >> 8) & 0xff] ^ Td3[s0 & 0xff] ^ rk[15]; - /* round 4: */ - s0 = Td0[t0 >> 24] ^ Td1[(t3 >> 16) & 0xff] ^ Td2[(t2 >> 8) & 0xff] ^ Td3[t1 & 0xff] ^ rk[16]; - s1 = Td0[t1 >> 24] ^ Td1[(t0 >> 16) & 0xff] ^ Td2[(t3 >> 8) & 0xff] ^ Td3[t2 & 0xff] ^ rk[17]; - s2 = Td0[t2 >> 24] ^ Td1[(t1 >> 16) & 0xff] ^ Td2[(t0 >> 8) & 0xff] ^ Td3[t3 & 0xff] ^ rk[18]; - s3 = Td0[t3 >> 24] ^ Td1[(t2 >> 16) & 0xff] ^ Td2[(t1 >> 8) & 0xff] ^ Td3[t0 & 0xff] ^ rk[19]; - /* round 5: */ - t0 = Td0[s0 >> 24] ^ Td1[(s3 >> 16) & 0xff] ^ Td2[(s2 >> 8) & 0xff] ^ Td3[s1 & 0xff] ^ rk[20]; - t1 = Td0[s1 >> 24] ^ Td1[(s0 >> 16) & 0xff] ^ Td2[(s3 >> 8) & 0xff] ^ Td3[s2 & 0xff] ^ rk[21]; - t2 = Td0[s2 >> 24] ^ Td1[(s1 >> 16) & 0xff] ^ Td2[(s0 >> 8) & 0xff] ^ Td3[s3 & 0xff] ^ rk[22]; - t3 = Td0[s3 >> 24] ^ Td1[(s2 >> 16) & 0xff] ^ Td2[(s1 >> 8) & 0xff] ^ Td3[s0 & 0xff] ^ rk[23]; - /* round 6: */ - s0 = Td0[t0 >> 24] ^ Td1[(t3 >> 16) & 0xff] ^ Td2[(t2 >> 8) & 0xff] ^ Td3[t1 & 0xff] ^ rk[24]; - s1 = Td0[t1 >> 24] ^ Td1[(t0 >> 16) & 0xff] ^ Td2[(t3 >> 8) & 0xff] ^ Td3[t2 & 0xff] ^ rk[25]; - s2 = Td0[t2 >> 24] ^ Td1[(t1 >> 16) & 0xff] ^ Td2[(t0 >> 8) & 0xff] ^ Td3[t3 & 0xff] ^ rk[26]; - s3 = Td0[t3 >> 24] ^ Td1[(t2 >> 16) & 0xff] ^ Td2[(t1 >> 8) & 0xff] ^ Td3[t0 & 0xff] ^ rk[27]; - /* round 7: */ - t0 = Td0[s0 >> 24] ^ Td1[(s3 >> 16) & 0xff] ^ Td2[(s2 >> 8) & 0xff] ^ Td3[s1 & 0xff] ^ rk[28]; - t1 = Td0[s1 >> 24] ^ Td1[(s0 >> 16) & 0xff] ^ Td2[(s3 >> 8) & 0xff] ^ Td3[s2 & 0xff] ^ rk[29]; - t2 = Td0[s2 >> 24] ^ Td1[(s1 >> 16) & 0xff] ^ Td2[(s0 >> 8) & 0xff] ^ Td3[s3 & 0xff] ^ rk[30]; - t3 = Td0[s3 >> 24] ^ Td1[(s2 >> 16) & 0xff] ^ Td2[(s1 >> 8) & 0xff] ^ Td3[s0 & 0xff] ^ rk[31]; - /* round 8: */ - s0 = Td0[t0 >> 24] ^ Td1[(t3 >> 16) & 0xff] ^ Td2[(t2 >> 8) & 0xff] ^ Td3[t1 & 0xff] ^ rk[32]; - s1 = Td0[t1 >> 24] ^ Td1[(t0 >> 16) & 0xff] ^ Td2[(t3 >> 8) & 0xff] ^ Td3[t2 & 0xff] ^ rk[33]; - s2 = Td0[t2 >> 24] ^ Td1[(t1 >> 16) & 0xff] ^ Td2[(t0 >> 8) & 0xff] ^ Td3[t3 & 0xff] ^ rk[34]; - s3 = Td0[t3 >> 24] ^ Td1[(t2 >> 16) & 0xff] ^ Td2[(t1 >> 8) & 0xff] ^ Td3[t0 & 0xff] ^ rk[35]; - /* round 9: */ - t0 = Td0[s0 >> 24] ^ Td1[(s3 >> 16) & 0xff] ^ Td2[(s2 >> 8) & 0xff] ^ Td3[s1 & 0xff] ^ rk[36]; - t1 = Td0[s1 >> 24] ^ Td1[(s0 >> 16) & 0xff] ^ Td2[(s3 >> 8) & 0xff] ^ Td3[s2 & 0xff] ^ rk[37]; - t2 = Td0[s2 >> 24] ^ Td1[(s1 >> 16) & 0xff] ^ Td2[(s0 >> 8) & 0xff] ^ Td3[s3 & 0xff] ^ rk[38]; - t3 = Td0[s3 >> 24] ^ Td1[(s2 >> 16) & 0xff] ^ Td2[(s1 >> 8) & 0xff] ^ Td3[s0 & 0xff] ^ rk[39]; + + ROUND(1,t,s); + ROUND(2,s,t); + ROUND(3,t,s); + ROUND(4,s,t); + ROUND(5,t,s); + ROUND(6,s,t); + ROUND(7,t,s); + ROUND(8,s,t); + ROUND(9,t,s); + rk += Nr << 2; + #else /* !FULL_UNROLL */ - /* - * Nr - 1 full rounds: - */ - r = Nr >> 1; - for (;;) { - t0 = - Td0[(s0 >> 24) ] ^ - Td1[(s3 >> 16) & 0xff] ^ - Td2[(s2 >> 8) & 0xff] ^ - Td3[(s1 ) & 0xff] ^ - rk[4]; - t1 = - Td0[(s1 >> 24) ] ^ - Td1[(s0 >> 16) & 0xff] ^ - Td2[(s3 >> 8) & 0xff] ^ - Td3[(s2 ) & 0xff] ^ - rk[5]; - t2 = - Td0[(s2 >> 24) ] ^ - Td1[(s1 >> 16) & 0xff] ^ - Td2[(s0 >> 8) & 0xff] ^ - Td3[(s3 ) & 0xff] ^ - rk[6]; - t3 = - Td0[(s3 >> 24) ] ^ - Td1[(s2 >> 16) & 0xff] ^ - Td2[(s1 >> 8) & 0xff] ^ - Td3[(s0 ) & 0xff] ^ - rk[7]; - rk += 8; - if (--r == 0) { - break; - } + /* Nr - 1 full rounds: */ + r = Nr >> 1; + for (;;) { + ROUND(1,t,s); + rk += 8; + if (--r == 0) + break; + ROUND(0,s,t); + } - s0 = - Td0[(t0 >> 24) ] ^ - Td1[(t3 >> 16) & 0xff] ^ - Td2[(t2 >> 8) & 0xff] ^ - Td3[(t1 ) & 0xff] ^ - rk[0]; - s1 = - Td0[(t1 >> 24) ] ^ - Td1[(t0 >> 16) & 0xff] ^ - Td2[(t3 >> 8) & 0xff] ^ - Td3[(t2 ) & 0xff] ^ - rk[1]; - s2 = - Td0[(t2 >> 24) ] ^ - Td1[(t1 >> 16) & 0xff] ^ - Td2[(t0 >> 8) & 0xff] ^ - Td3[(t3 ) & 0xff] ^ - rk[2]; - s3 = - Td0[(t3 >> 24) ] ^ - Td1[(t2 >> 16) & 0xff] ^ - Td2[(t1 >> 8) & 0xff] ^ - Td3[(t0 ) & 0xff] ^ - rk[3]; - } #endif /* ?FULL_UNROLL */ - /* + +#undef ROUND + + /* * apply last round and * map cipher state to byte array block: */ - s0 = - (Td4[(t0 >> 24) ] & 0xff000000) ^ - (Td4[(t3 >> 16) & 0xff] & 0x00ff0000) ^ - (Td4[(t2 >> 8) & 0xff] & 0x0000ff00) ^ - (Td4[(t1 ) & 0xff] & 0x000000ff) ^ - rk[0]; + s0 = TD41(t0) ^ TD42(t3) ^ TD43(t2) ^ TD44(t1) ^ rk[0]; PUTU32(pt , s0); - s1 = - (Td4[(t1 >> 24) ] & 0xff000000) ^ - (Td4[(t0 >> 16) & 0xff] & 0x00ff0000) ^ - (Td4[(t3 >> 8) & 0xff] & 0x0000ff00) ^ - (Td4[(t2 ) & 0xff] & 0x000000ff) ^ - rk[1]; + s1 = TD41(t1) ^ TD42(t0) ^ TD43(t3) ^ TD44(t2) ^ rk[1]; PUTU32(pt + 4, s1); - s2 = - (Td4[(t2 >> 24) ] & 0xff000000) ^ - (Td4[(t1 >> 16) & 0xff] & 0x00ff0000) ^ - (Td4[(t0 >> 8) & 0xff] & 0x0000ff00) ^ - (Td4[(t3 ) & 0xff] & 0x000000ff) ^ - rk[2]; + s2 = TD41(t2) ^ TD42(t1) ^ TD43(t0) ^ TD44(t3) ^ rk[2]; PUTU32(pt + 8, s2); - s3 = - (Td4[(t3 >> 24) ] & 0xff000000) ^ - (Td4[(t2 >> 16) & 0xff] & 0x00ff0000) ^ - (Td4[(t1 >> 8) & 0xff] & 0x0000ff00) ^ - (Td4[(t0 ) & 0xff] & 0x000000ff) ^ - rk[3]; + s3 = TD41(t3) ^ TD42(t2) ^ TD43(t1) ^ TD44(t0) ^ rk[3]; PUTU32(pt + 12, s3); } + + + +/* Generic wrapper functions for AES functions */ + +void * aes_encrypt_init(const u8 *key, size_t len) +{ + u32 *rk; + if (len != 16) + return NULL; + rk = malloc(4 * 44); + if (rk == NULL) + return NULL; + rijndaelKeySetupEnc(rk, key); + return rk; +} + + +void aes_encrypt(void *ctx, const u8 *plain, u8 *crypt) +{ + rijndaelEncrypt(ctx, plain, crypt); +} + + +void aes_encrypt_deinit(void *ctx) +{ + free(ctx); +} + + +void * aes_decrypt_init(const u8 *key, size_t len) +{ + u32 *rk; + if (len != 16) + return NULL; + rk = malloc(4 * 44); + if (rk == NULL) + return NULL; + rijndaelKeySetupDec(rk, key); + return rk; +} + + +void aes_decrypt(void *ctx, const u8 *crypt, u8 *plain) +{ + rijndaelDecrypt(ctx, crypt, plain); +} + + +void aes_decrypt_deinit(void *ctx) +{ + free(ctx); +} |