0x64,0x69,0x7e,0x73,0x50,0x5d,0x4a,0x47,
	0xdc,0xd1,0xc6,0xcb,0xe8,0xe5,0xf2,0xff,
	0xb4,0xb9,0xae,0xa3,0x80,0x8d,0x9a,0x97
};

static const uint_8t gfmul_e[256] = {
	0x00,0x0e,0x1c,0x12,0x38,0x36,0x24,0x2a,
	0x70,0x7e,0x6c,0x62,0x48,0x46,0x54,0x5a,
	0xe0,0xee,0xfc,0xf2,0xd8,0xd6,0xc4,0xca,
	0x90,0x9e,0x8c,0x82,0xa8,0xa6,0xb4,0xba,
	0xdb,0xd5,0xc7,0xc9,0xe3,0xed,0xff,0xf1,
	0xab,0xa5,0xb7,0xb9,0x93,0x9d,0x8f,0x81,
	0x3b,0x35,0x27,0x29,0x03,0x0d,0x1f,0x11,
	0x4b,0x45,0x57,0x59,0x73,0x7d,0x6f,0x61,
	0xad,0xa3,0xb1,0xbf,0x95,0x9b,0x89,0x87,
	0xdd,0xd3,0xc1,0xcf,0xe5,0xeb,0xf9,0xf7,
	0x4d,0x43,0x51,0x5f,0x75,0x7b,0x69,0x67,
	0x3d,0x33,0x21,0x2f,0x05,0x0b,0x19,0x17,
	0x76,0x78,0x6a,0x64,0x4e,0x40,0x52,0x5c,
	0x06,0x08,0x1a,0x14,0x3e,0x30,0x22,0x2c,
	0x96,0x98,0x8a,0x84,0xae,0xa0,0xb2,0xbc,
	0xe6,0xe8,0xfa,0xf4,0xde,0xd0,0xc2,0xcc,
	0x41,0x4f,0x5d,0x53,0x79,0x77,0x65,0x6b,
	0x31,0x3f,0x2d,0x23,0x09,0x07,0x15,0x1b,
	0xa1,0xaf,0xbd,0xb3,0x99,0x97,0x85,0x8b,
	0xd1,0xdf,0xcd,0xc3,0xe9,0xe7,0xf5,0xfb,
	0x9a,0x94,0x86,0x88,0xa2,0xac,0xbe,0xb0,
	0xea,0xe4,0xf6,0xf8,0xd2,0xdc,0xce,0xc0,
	0x7a,0x74,0x66,0x68,0x42,0x4c,0x5e,0x50,
	0x0a,0x04,0x16,0x18,0x32,0x3c,0x2e,0x20,
	0xec,0xe2,0xf0,0xfe,0xd4,0xda,0xc8,0xc6,
	0x9c,0x92,0x80,0x8e,0xa4,0xaa,0xb8,0xb6,
	0x0c,0x02,0x10,0x1e,0x34,0x3a,0x28,0x26,
	0x7c,0x72,0x60,0x6e,0x44,0x4a,0x58,0x56,
	0x37,0x39,0x2b,0x25,0x0f,0x01,0x13,0x1d,
	0x47,0x49,0x5b,0x55,0x7f,0x71,0x63,0x6d,
	0xd7,0xd9,0xcb,0xc5,0xef,0xe1,0xf3,0xfd,
	0xa7,0xa9,0xbb,0xb5,0x9f,0x91,0x83,0x8d
};

#if defined( HAVE_UINT_32T )
  typedef unsigned long uint_32t;
#endif

#if defined( HAVE_MEMCPY )
#  define block_copy(d, s, l) memcpy(d, s, l)
#  define block16_copy(d, s)  memcpy(d, s, N_BLOCK)
#else
#  define block_copy(d, s, l) copy_block(d, s, l)
#  define block16_copy(d, s)  copy_block16(d, s)
#endif

/* block size 'nn' must be a multiple of four */

static void copy_block16( void *d, const void *s )
{
#if defined( HAVE_UINT_32T )
    ((uint_32t*)d)[ 0] = ((uint_32t*)s)[ 0];
    ((uint_32t*)d)[ 1] = ((uint_32t*)s)[ 1];
    ((uint_32t*)d)[ 2] = ((uint_32t*)s)[ 2];
    ((uint_32t*)d)[ 3] = ((uint_32t*)s)[ 3];
#else
    ((uint_8t*)d)[ 0] = ((uint_8t*)s)[ 0];
    ((uint_8t*)d)[ 1] = ((uint_8t*)s)[ 1];
    ((uint_8t*)d)[ 2] = ((uint_8t*)s)[ 2];
    ((uint_8t*)d)[ 3] = ((uint_8t*)s)[ 3];
    ((uint_8t*)d)[ 4] = ((uint_8t*)s)[ 4];
    ((uint_8t*)d)[ 5] = ((uint_8t*)s)[ 5];
    ((uint_8t*)d)[ 6] = ((uint_8t*)s)[ 6];
    ((uint_8t*)d)[ 7] = ((uint_8t*)s)[ 7];
    ((uint_8t*)d)[ 8] = ((uint_8t*)s)[ 8];
    ((uint_8t*)d)[ 9] = ((uint_8t*)s)[ 9];
    ((uint_8t*)d)[10] = ((uint_8t*)s)[10];
    ((uint_8t*)d)[11] = ((uint_8t*)s)[11];
    ((uint_8t*)d)[12] = ((uint_8t*)s)[12];
    ((uint_8t*)d)[13] = ((uint_8t*)s)[13];
    ((uint_8t*)d)[14] = ((uint_8t*)s)[14];
    ((uint_8t*)d)[15] = ((uint_8t*)s)[15];
#endif
}

static void copy_block( void * d, void *s, uint_8t nn )
{
    while( nn-- )
        *((uint_8t*)d)++ = *((uint_8t*)s)++;
}

static void xor_block( void *d, const void *s )
{
#if defined( HAVE_UINT_32T )
    ((uint_32t*)d)[ 0] ^= ((uint_32t*)s)[ 0];
    ((uint_32t*)d)[ 1] ^= ((uint_32t*)s)[ 1];
    ((uint_32t*)d)[ 2] ^= ((uint_32t*)s)[ 2];
    ((uint_32t*)d)[ 3] ^= ((uint_32t*)s)[ 3];
#else
    ((uint_8t*)d)[ 0] ^= ((uint_8t*)s)[ 0];
    ((uint_8t*)d)[ 1] ^= ((uint_8t*)s)[ 1];
    ((uint_8t*)d)[ 2] ^= ((uint_8t*)s)[ 2];
    ((uint_8t*)d)[ 3] ^= ((uint_8t*)s)[ 3];
    ((uint_8t*)d)[ 4] ^= ((uint_8t*)s)[ 4];
    ((uint_8t*)d)[ 5] ^= ((uint_8t*)s)[ 5];
    ((uint_8t*)d)[ 6] ^= ((uint_8t*)s)[ 6];
    ((uint_8t*)d)[ 7] ^= ((uint_8t*)s)[ 7];
    ((uint_8t*)d)[ 8] ^= ((uint_8t*)s)[ 8];
    ((uint_8t*)d)[ 9] ^= ((uint_8t*)s)[ 9];
    ((uint_8t*)d)[10] ^= ((uint_8t*)s)[10];
    ((uint_8t*)d)[11] ^= ((uint_8t*)s)[11];
    ((uint_8t*)d)[12] ^= ((uint_8t*)s)[12];
    ((uint_8t*)d)[13] ^= ((uint_8t*)s)[13];
    ((uint_8t*)d)[14] ^= ((uint_8t*)s)[14];
    ((uint_8t*)d)[15] ^= ((uint_8t*)s)[15];
#endif
}

static void copy_and_key( void *d, const void *s, const void *k )
{
#if defined( HAVE_UINT_32T )
    ((uint_32t*)d)[ 0] = ((uint_32t*)s)[ 0] ^ ((uint_32t*)k)[ 0];
    ((uint_32t*)d)[ 1] = ((uint_32t*)s)[ 1] ^ ((uint_32t*)k)[ 1];
    ((uint_32t*)d)[ 2] = ((uint_32t*)s)[ 2] ^ ((uint_32t*)k)[ 2];
    ((uint_32t*)d)[ 3] = ((uint_32t*)s)[ 3] ^ ((uint_32t*)k)[ 3];
#elif 1
    ((uint_8t*)d)[ 0] = ((uint_8t*)s)[ 0] ^ ((uint_8t*)k)[ 0];
    ((uint_8t*)d)[ 1] = ((uint_8t*)s)[ 1] ^ ((uint_8t*)k)[ 1];
    ((uint_8t*)d)[ 2] = ((uint_8t*)s)[ 2] ^ ((uint_8t*)k)[ 2];
    ((uint_8t*)d)[ 3] = ((uint_8t*)s)[ 3] ^ ((uint_8t*)k)[ 3];
    ((uint_8t*)d)[ 4] = ((uint_8t*)s)[ 4] ^ ((uint_8t*)k)[ 4];
    ((uint_8t*)d)[ 5] = ((uint_8t*)s)[ 5] ^ ((uint_8t*)k)[ 5];
    ((uint_8t*)d)[ 6] = ((uint_8t*)s)[ 6] ^ ((uint_8t*)k)[ 6];
    ((uint_8t*)d)[ 7] = ((uint_8t*)s)[ 7] ^ ((uint_8t*)k)[ 7];
    ((uint_8t*)d)[ 8] = ((uint_8t*)s)[ 8] ^ ((uint_8t*)k)[ 8];
    ((uint_8t*)d)[ 9] = ((uint_8t*)s)[ 9] ^ ((uint_8t*)k)[ 9];
    ((uint_8t*)d)[10] = ((uint_8t*)s)[10] ^ ((uint_8t*)k)[10];
    ((uint_8t*)d)[11] = ((uint_8t*)s)[11] ^ ((uint_8t*)k)[11];
    ((uint_8t*)d)[12] = ((uint_8t*)s)[12] ^ ((uint_8t*)k)[12];
    ((uint_8t*)d)[13] = ((uint_8t*)s)[13] ^ ((uint_8t*)k)[13];
    ((uint_8t*)d)[14] = ((uint_8t*)s)[14] ^ ((uint_8t*)k)[14];
    ((uint_8t*)d)[15] = ((uint_8t*)s)[15] ^ ((uint_8t*)k)[15];
#else
    block16_copy(d, s);
    xor_block(d, k);
#endif
}

static void add_round_key( uint_8t d[N_BLOCK], const uint_8t k[N_BLOCK] )
{
    xor_block(d, k);
}

static void shift_sub_rows( uint_8t st[N_BLOCK] )
{   uint_8t tt;

    st[ 0] = s_box[st[ 0]]; st[ 4] = s_box[st[ 4]];
    st[ 8] = s_box[st[ 8]]; st[12] = s_box[st[12]];

    tt = st[1]; st[ 1] = s_box[st[ 5]]; st[ 5] = s_box[st[ 9]];
    st[ 9] = s_box[st[13]]; st[13] = s_box[ tt ];

    tt = st[2]; st[ 2] = s_box[st[10]]; st[10] = s_box[ tt ];
    tt = st[6]; st[ 6] = s_box[st[14]]; st[14] = s_box[ tt ];

    tt = st[15]; st[15] = s_box[st[11]]; st[11] = s_box[st[ 7]];
    st[ 7] = s_box[st[ 3]]; st[ 3] = s_box[ tt ];
}

static void inv_shift_sub_rows( uint_8t st[N_BLOCK] )
{   uint_8t tt;

    st[ 0] = inv_s_box[st[ 0]]; st[ 4] = inv_s_box[st[ 4]];
    st[ 8] = inv_s_box[st[ 8]]; st[12] = inv_s_box[st[12]];

    tt = st[13]; st[13] = inv_s_box[st[9]]; st[ 9] = inv_s_box[st[5]];
    st[ 5] = inv_s_box[st[1]]; st[ 1] = inv_s_box[ tt ];

    tt = st[2]; st[ 2] = inv_s_box[st[10]]; st[10] = inv_s_box[ tt ];
    tt = st[6]; st[ 6] = inv_s_box[st[14]]; st[14] = inv_s_box[ tt ];

    tt = st[3]; st[ 3] = inv_s_box[st[ 7]]; st[ 7] = inv_s_box[st[11]];
    st[11] = inv_s_box[st[15]]; st[15] = inv_s_box[ tt ];
}

#if defined( VERSION_1 )
  static void mix_sub_columns( uint_8t dt[N_BLOCK] )
  { uint_8t st[N_BLOCK];
    block16_copy(st, dt);
#else
  static void mix_sub_columns( uint_8t dt[N_BLOCK], uint_8t st[N_BLOCK] )
  {
#endif
    dt[ 0] = gfm2_s_box[st[0]] ^ gfm3_s_box[st[5]] ^ s_box[st[10]] ^ s_box[st[15]];
    dt[ 1] = s_box[st[0]] ^ gfm2_s_box[st[5]] ^ gfm3_s_box[st[10]] ^ s_box[st[15]];
    dt[ 2] = s_box[st[0]] ^ s_box[st[5]] ^ gfm2_s_box[st[10]] ^ gfm3_s_box[st[15]];
    dt[ 3] = gfm3_s_box[st[0]] ^ s_box[st[5]] ^ s_box[st[10]] ^ gfm2_s_box[st[15]];

    dt[ 4] = gfm2_s_box[st[4]] ^ gfm3_s_box[st[9]] ^ s_box[st[14]] ^ s_box[st[3]];
    dt[ 5] = s_box[st[4]] ^ gfm2_s_box[st[9]] ^ gfm3_s_box[st[14]] ^ s_box[st[3]];
    dt[ 6] = s_box[st[4]] ^ s_box[st[9]] ^ gfm2_s_box[st[14]] ^ gfm3_s_box[st[3]];
    dt[ 7] = gfm3_s_box[st[4]] ^ s_box[st[9]] ^ s_box[st[14]] ^ gfm2_s_box[st[3]];

    dt[ 8] = gfm2_s_box[st[8]] ^ gfm3_s_box[st[13]] ^ s_box[st[2]] ^ s_box[st[7]];
    dt[ 9] = s_box[st[8]] ^ gfm2_s_box[st[13]] ^ gfm3_s_box[st[2]] ^ s_box[st[7]];
    dt[10] = s_box[st[8]] ^ s_box[st[13]] ^ gfm2_s_box[st[2]] ^ gfm3_s_box[st[7]];
    dt[11] = gfm3_s_box[st[8]] ^ s_box[st[13]] ^ s_box[st[2]] ^ gfm2_s_box[st[7]];

    dt[12] = gfm2_s_box[st[12]] ^ gfm3_s_box[st[1]] ^ s_box[st[6]] ^ s_box[st[11]];
    dt[13] = s_box[st[12]] ^ gfm2_s_box[st[1]] ^ gfm3_s_box[st[6]] ^ s_box[st[11]];
    dt[14] = s_box[st[12]] ^ s_box[st[1]] ^ gfm2_s_box[st[6]] ^ gfm3_s_box[st[11]];
    dt[15] = gfm3_s_box[st[12]] ^ s_box[st[1]] ^ s_box[st[6]] ^ gfm2_s_box[st[11]];
  }

#if defined( VERSION_1 )
  static void inv_mix_sub_columns( uint_8t dt[N_BLOCK] )
  { uint_8t st[N_BLOCK];
    block16_copy(st, dt);
#else
  static void inv_mix_sub_columns( uint_8t dt[N_BLOCK], uint_8t st[N_BLOCK] )
  {
#endif
    dt[ 0] = inv_s_box[gfmul_e[st[ 0]] ^ gfmul_b[st[ 1]] ^ gfmul_d[st[ 2]] ^ gfmul_9[st[ 3]]];
    dt[ 5] = inv_s_box[gfmul_9[st[ 0]] ^ gfmul_e[st[ 1]] ^ gfmul_b[st[ 2]] ^ gfmul_d[st[ 3]]];
    dt[10] = inv_s_box[gfmul_d[st[ 0]] ^ gfmul_9[st[ 1]] ^ gfmul_e[st[ 2]] ^ gfmul_b[st[ 3]]];
    dt[15] = inv_s_box[gfmul_b[st[ 0]] ^ gfmul_d[st[ 1]] ^ gfmul_9[st[ 2]] ^ gfmul_e[st[ 3]]];

    dt[ 4] = inv_s_box[gfmul_e[st[ 4]] ^ gfmul_b[st[ 5]] ^ gfmul_d[st[ 6]] ^ gfmul_9[st[ 7]]];
    dt[ 9] = inv_s_box[gfmul_9[st[ 4]] ^ gfmul_e[st[ 5]] ^ gfmul_b[st[ 6]] ^ gfmul_d[st[ 7]]];
    dt[14] = inv_s_box[gfmul_d[st[ 4]] ^ gfmul_9[st[ 5]] ^ gfmul_e[st[ 6]] ^ gfmul_b[st[ 7]]];
    dt[ 3] = inv_s_box[gfmul_b[st[ 4]] ^ gfmul_d[st[ 5]] ^ gfmul_9[st[ 6]] ^ gfmul_e[st[ 7]]];

    dt[ 8] = inv_s_box[gfmul_e[st[ 8]] ^ gfmul_b[st[ 9]] ^ gfmul_d[st[10]] ^ gfmul_9[st[11]]];
    dt[13] = inv_s_box[gfmul_9[st[ 8]] ^ gfmul_e[st[ 9]] ^ gfmul_b[st[10]] ^ gfmul_d[st[11]]];
    dt[ 2] = inv_s_box[gfmul_d[st[ 8]] ^ gfmul_9[st[ 9]] ^ gfmul_e[st[10]] ^ gfmul_b[st[11]]];
    dt[ 7] = inv_s_box[gfmul_b[st[ 8]] ^ gfmul_d[st[ 9]] ^ gfmul_9[st[10]] ^ gfmul_e[st[11]]];

    dt[12] = inv_s_box[gfmul_e[st[12]] ^ gfmul_b[st[13]] ^ gfmul_d[st[14]] ^ gfmul_9[st[15]]];
    dt[ 1] = inv_s_box[gfmul_9[st[12]] ^ gfmul_e[st[13]] ^ gfmul_b[st[14]] ^ gfmul_d[st[15]]];
    dt[ 6] = inv_s_box[gfmul_d[st[12]] ^ gfmul_9[st[13]] ^ gfmul_e[st[14]] ^ gfmul_b[st[15]]];
    dt[11] = inv_s_box[gfmul_b[st[12]] ^ gfmul_d[st[13]] ^ gfmul_9[st[14]] ^ gfmul_e[st[15]]];
  }

#if defined( AES_ENC_PREKEYED ) || defined( AES_DEC_PREKEYED )

/*  Set the cipher key for the pre-keyed version */

return_type aes_set_key( const unsigned char key[], length_type keylen, aes_context ctx[1] )
{
    uint_8t cc, rc, hi;

    switch( keylen )
    {
    case 16:
    case 128:
        keylen = 16;
        break;
    case 24:
    case 192:
        keylen = 24;
        break;
    case 32:
    case 256:
        keylen = 32;
        break;
    default:
        ctx->rnd = 0;
        return -1;
    }
    block_copy(ctx->ksch, key, keylen);
    hi = (keylen + 28) << 2;
    ctx->rnd = (hi >> 4) - 1;
    for( cc = keylen, rc = 1; cc < hi; cc += 4 )
    {   uint_8t tt, t0, t1, t2, t3;

        t0 = ctx->ksch[cc - 4];
        t1 = ctx->ksch[cc - 3];
        t2 = ctx->ksch[cc - 2];
        t3 = ctx->ksch[cc - 1];
        if( cc % keylen == 0 )
        {
            tt = t0;
            t0 = s_box[t1] ^ rc;
            t1 = s_box[t2];
            t2 = s_box[t3];
            t3 = s_box[tt];
            rc = f2(rc);
        }
        else if( keylen > 24 && cc % keylen == 16 )
        {
            t0 = s_box[t0];
            t1 = s_box[t1];
            t2 = s_box[t2];
            t3 = s_box[t3];
        }
        tt = cc - keylen;
        ctx->ksch[cc + 0] = ctx->ksch[tt + 0] ^ t0;
        ctx->ksch[cc + 1] = ctx->ksch[tt + 1] ^ t1;
        ctx->ksch[cc + 2] = ctx->ksch[tt + 2] ^ t2;
        ctx->ksch[cc + 3] = ctx->ksch[tt + 3] ^ t3;
    }
    return 0;
}

#endif

#if defined( AES_ENC_PREKEYED )

/*  Encrypt a single block of 16 bytes */

return_type aes_encrypt( const unsigned char in[N_BLOCK], unsigned char  out[N_BLOCK], const aes_context ctx[1] )
{
    if( ctx->rnd )
    {
        uint_8t s1[N_BLOCK], r;
        copy_and_key( s1, in, ctx->ksch );

        for( r = 1 ; r < ctx->rnd ; ++r )
#if defined( VERSION_1 )
        {
            mix_sub_columns( s1 );
            add_round_key( s1, ctx->ksch + r * N_BLOCK);