/* LibTomCrypt, modular cryptographic library -- Tom St Denis * * LibTomCrypt is a library that provides various cryptographic * algorithms in a highly modular and flexible manner. * * The library is free for all purposes without any express * gurantee it works. * * Tom St Denis, tomstdenis@iahu.ca, http://libtomcrypt.org *//* AES implementation by Tom St Denis * * Derived from the Public Domain source code by---  * rijndael-alg-fst.c  *  * @version 3.0 (December 2000)  *  * Optimised ANSI C code for the Rijndael cipher (now AES)  *  * @author Vincent Rijmen <vincent.rijmen@esat.kuleuven.ac.be>  * @author Antoon Bosselaers <antoon.bosselaers@esat.kuleuven.ac.be>  * @author Paulo Barreto <paulo.barreto@terra.com.br>---  modified by Stefan Scherrer to allow compilation under DDK */#include "tom.h"#pragma code_seg()#pragma data_seg()#define byte(x, n) ((unsigned char)((x) >> (8 * (n))))#if 0const struct _cipher_descriptor rijndael_desc ={    "rijndael",    6,    16, 32, 16, 10,    &rijndael_setup,    &rijndael_ecb_encrypt,    &rijndael_ecb_decrypt,    &rijndael_test,    &rijndael_keysize};const struct _cipher_descriptor aes_desc ={    "aes",    6,    16, 32, 16, 10,    &rijndael_setup,    &rijndael_ecb_encrypt,    &rijndael_ecb_decrypt,    &rijndael_test,    &rijndael_keysize};#endif#include "aes_tab.c"int rijndael_setup(const unsigned char *key, int keylen, int rounds, symmetric_key *skey){    int i, j;    ULONG32 temp, *rk, *rrk;    _ARGCHK(key != NULL);    _ARGCHK(skey != NULL);    if (keylen != 16 && keylen != 24 && keylen != 32) {       return CRYPT_INVALID_KEYSIZE;    }    if (rounds != 0 && rounds != (10 + ((keylen/8)-2)*2)) {       return CRYPT_INVALID_ROUNDS;    }    skey->rijndael.Nr = 10 + ((keylen/8)-2)*2;    /* setup the forward key */    i                 = 0;    rk                = skey->rijndael.eK;    LOAD32H(rk[0], key     );    LOAD32H(rk[1], key +  4);    LOAD32H(rk[2], key +  8);    LOAD32H(rk[3], key + 12);    if (keylen == 16) {        j = 44;        for (;;) {            temp  = rk[3];            rk[4] = rk[0] ^                (Te4_3[byte(temp, 2)]) ^                (Te4_2[byte(temp, 1)]) ^                (Te4_1[byte(temp, 0)]) ^                (Te4_0[byte(temp, 3)]) ^                rcon[i];            rk[5] = rk[1] ^ rk[4];            rk[6] = rk[2] ^ rk[5];            rk[7] = rk[3] ^ rk[6];            if (++i == 10) {               break;            }            rk += 4;        }    } else if (keylen == 24) {        j = 52;        LOAD32H(rk[4], key + 16);        LOAD32H(rk[5], key + 20);        for (;;) {        #ifdef _MSC_VER            temp = skey->rijndael.eK[rk - skey->rijndael.eK + 5];        #else            temp = rk[5];        #endif            rk[ 6] = rk[ 0] ^                (Te4_3[byte(temp, 2)]) ^                (Te4_2[byte(temp, 1)]) ^                (Te4_1[byte(temp, 0)]) ^                (Te4_0[byte(temp, 3)]) ^                rcon[i];            rk[ 7] = rk[ 1] ^ rk[ 6];            rk[ 8] = rk[ 2] ^ rk[ 7];            rk[ 9] = rk[ 3] ^ rk[ 8];            if (++i == 8) {                break;            }            rk[10] = rk[ 4] ^ rk[ 9];            rk[11] = rk[ 5] ^ rk[10];            rk += 6;        }    } else if (keylen == 32) {        j = 60;        LOAD32H(rk[4], key + 16);        LOAD32H(rk[5], key + 20);        LOAD32H(rk[6], key + 24);        LOAD32H(rk[7], key + 28);        for (;;) {        #ifdef _MSC_VER            temp = skey->rijndael.eK[rk - skey->rijndael.eK + 7];        #else            temp = rk[7];        #endif            rk[ 8] = rk[ 0] ^                (Te4_3[byte(temp, 2)]) ^                (Te4_2[byte(temp, 1)]) ^                (Te4_1[byte(temp, 0)]) ^                (Te4_0[byte(temp, 3)]) ^                rcon[i];            rk[ 9] = rk[ 1] ^ rk[ 8];            rk[10] = rk[ 2] ^ rk[ 9];            rk[11] = rk[ 3] ^ rk[10];            if (++i == 7) {                break;            }            temp = rk[11];            rk[12] = rk[ 4] ^                (Te4_3[byte(temp, 3)]) ^                (Te4_2[byte(temp, 2)]) ^                (Te4_1[byte(temp, 1)]) ^                (Te4_0[byte(temp, 0)]);            rk[13] = rk[ 5] ^ rk[12];            rk[14] = rk[ 6] ^ rk[13];            rk[15] = rk[ 7] ^ rk[14];            rk += 8;        }    } else {       /* this can't happen */       j = 4;    }    /* setup the inverse key now */    rk   = skey->rijndael.dK;    rrk  = skey->rijndael.eK + j - 4;    /* apply the inverse MixColumn transform to all round keys but the first and the last: */    /* copy first */    *rk++ = *rrk++;    *rk++ = *rrk++;    *rk++ = *rrk++;    *rk   = *rrk;    rk -= 3; rrk -= 3;    for (i = 1; i < skey->rijndael.Nr; i++) {        rrk -= 4;        rk  += 4;    #ifdef SMALL_CODE        temp = rrk[0];        rk[0] =            Td0[255 & Te4[byte(temp, 3)]] ^            Td1[255 & Te4[byte(temp, 2)]] ^            Td2[255 & Te4[byte(temp, 1)]] ^            Td3[255 & Te4[byte(temp, 0)]];        temp = rrk[1];        rk[1] =            Td0[255 & Te4[byte(temp, 3)]] ^            Td1[255 & Te4[byte(temp, 2)]] ^            Td2[255 & Te4[byte(temp, 1)]] ^            Td3[255 & Te4[byte(temp, 0)]];        temp = rrk[2];        rk[2] =            Td0[255 & Te4[byte(temp, 3)]] ^            Td1[255 & Te4[byte(temp, 2)]] ^            Td2[255 & Te4[byte(temp, 1)]] ^            Td3[255 & Te4[byte(temp, 0)]];        temp = rrk[3];        rk[3] =            Td0[255 & Te4[byte(temp, 3)]] ^            Td1[255 & Te4[byte(temp, 2)]] ^            Td2[255 & Te4[byte(temp, 1)]] ^            Td3[255 & Te4[byte(temp, 0)]];     #else        temp = rrk[0];        rk[0] =            Tks0[byte(temp, 3)] ^            Tks1[byte(temp, 2)] ^            Tks2[byte(temp, 1)] ^            Tks3[byte(temp, 0)];        temp = rrk[1];        rk[1] =            Tks0[byte(temp, 3)] ^            Tks1[byte(temp, 2)] ^            Tks2[byte(temp, 1)] ^            Tks3[byte(temp, 0)];        temp = rrk[2];        rk[2] =            Tks0[byte(temp, 3)] ^            Tks1[byte(temp, 2)] ^            Tks2[byte(temp, 1)] ^            Tks3[byte(temp, 0)];        temp = rrk[3];        rk[3] =            Tks0[byte(temp, 3)] ^            Tks1[byte(temp, 2)] ^            Tks2[byte(temp, 1)] ^            Tks3[byte(temp, 0)];      #endif    }    /* copy last */    rrk -= 4;    rk  += 4;    *rk++ = *rrk++;    *rk++ = *rrk++;    *rk++ = *rrk++;    *rk   = *rrk;    return CRYPT_OK;}#ifdef CLEAN_STACKstatic void _rijndael_ecb_encrypt(const unsigned char *pt, unsigned char *ct, symmetric_key *skey)#elsevoid rijndael_ecb_encrypt(const unsigned char *pt, unsigned char *ct, symmetric_key *skey)#endif{    ULONG32 s0, s1, s2, s3, t0, t1, t2, t3, *rk;    int Nr, r;    _ARGCHK(pt != NULL);    _ARGCHK(ct != NULL);    _ARGCHK(skey != NULL);    Nr = skey->rijndael.Nr;    rk = skey->rijndael.eK;    /*     * map byte array block to cipher state     * and add initial round key:     */    LOAD32H(s0, pt      ); s0 ^= rk[0];    LOAD32H(s1, pt  +  4); s1 ^= rk[1];    LOAD32H(s2, pt  +  8); s2 ^= rk[2];    LOAD32H(s3, pt  + 12); s3 ^= rk[3];    /*     * Nr - 1 full rounds:     */    r = Nr >> 1;    for (;;) {        t0 =            Te0[byte(s0, 3)] ^            Te1[byte(s1, 2)] ^            Te2[byte(s2, 1)] ^            Te3[byte(s3, 0)] ^            rk[4];        t1 =            Te0[byte(s1, 3)] ^            Te1[byte(s2, 2)] ^            Te2[byte(s3, 1)] ^            Te3[byte(s0, 0)] ^            rk[5];        t2 =            Te0[byte(s2, 3)] ^            Te1[byte(s3, 2)] ^            Te2[byte(s0, 1)] ^            Te3[byte(s1, 0)] ^            rk[6];        t3 =            Te0[byte(s3, 3)] ^            Te1[byte(s0, 2)] ^