1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
|
/*
---------------------------------------------------------------------------
Copyright (c) 2003, Dr Brian Gladman < >, Worcester, UK.
All rights reserved.
LICENSE TERMS
The free distribution and use of this software in both source and binary
form is allowed (with or without changes) provided that:
1. distributions of this source code include the above copyright
notice, this list of conditions and the following disclaimer;
2. distributions in binary form include the above copyright
notice, this list of conditions and the following disclaimer
in the documentation and/or other associated materials;
3. the copyright holder's name is not used to endorse products
built using this software without specific written permission.
ALTERNATIVELY, provided that this notice is retained in full, this product
may be distributed under the terms of the GNU General Public License (GPL),
in which case the provisions of the GPL apply INSTEAD OF those given above.
DISCLAIMER
This software is provided 'as is' with no explicit or implied warranties
in respect of its properties, including, but not limited to, correctness
and/or fitness for purpose.
---------------------------------------------------------------------------
Issue Date: 26/08/2003
This file contains the code for implementing encryption and decryption
for AES (Rijndael) for block and key sizes of 16, 24 and 32 bytes. It
can optionally be replaced by code written in assembler using NASM. For
further details see the file aesopt.h
*/
#include "aesopt.h"
#define si(y,x,k,c) (s(y,c) = word_in(x, c) ^ (k)[c])
#define so(y,x,c) word_out(y, c, s(x,c))
#if defined(ARRAYS)
#define locals(y,x) x[4],y[4]
#else
#define locals(y,x) x##0,x##1,x##2,x##3,y##0,y##1,y##2,y##3
#endif
#define l_copy(y, x) s(y,0) = s(x,0); s(y,1) = s(x,1); \
s(y,2) = s(x,2); s(y,3) = s(x,3);
#define state_in(y,x,k) si(y,x,k,0); si(y,x,k,1); si(y,x,k,2); si(y,x,k,3)
#define state_out(y,x) so(y,x,0); so(y,x,1); so(y,x,2); so(y,x,3)
#define round(rm,y,x,k) rm(y,x,k,0); rm(y,x,k,1); rm(y,x,k,2); rm(y,x,k,3)
#if defined(ENCRYPTION) && !defined(AES_ASM)
/* Visual C++ .Net v7.1 provides the fastest encryption code when using
Pentium optimization with small code but this is poor for decryption
so we need to control this with the following VC++ pragmas
*/
#if defined(_MSC_VER)
#pragma optimize( "s", on )
#endif
/* Given the column (c) of the output state variable, the following
macros give the input state variables which are needed in its
computation for each row (r) of the state. All the alternative
macros give the same end values but expand into different ways
of calculating these values. In particular the complex macro
used for dynamically variable block sizes is designed to expand
to a compile time constant whenever possible but will expand to
conditional clauses on some branches (I am grateful to Frank
Yellin for this construction)
*/
#define fwd_var(x,r,c)\
( r == 0 ? ( c == 0 ? s(x,0) : c == 1 ? s(x,1) : c == 2 ? s(x,2) : s(x,3))\
: r == 1 ? ( c == 0 ? s(x,1) : c == 1 ? s(x,2) : c == 2 ? s(x,3) : s(x,0))\
: r == 2 ? ( c == 0 ? s(x,2) : c == 1 ? s(x,3) : c == 2 ? s(x,0) : s(x,1))\
: ( c == 0 ? s(x,3) : c == 1 ? s(x,0) : c == 2 ? s(x,1) : s(x,2)))
#if defined(FT4_SET)
#undef dec_fmvars
#define fwd_rnd(y,x,k,c) (s(y,c) = (k)[c] ^ four_tables(x,t_use(f,n),fwd_var,rf1,c))
#elif defined(FT1_SET)
#undef dec_fmvars
#define fwd_rnd(y,x,k,c) (s(y,c) = (k)[c] ^ one_table(x,upr,t_use(f,n),fwd_var,rf1,c))
#else
#define fwd_rnd(y,x,k,c) (s(y,c) = (k)[c] ^ fwd_mcol(no_table(x,t_use(s,box),fwd_var,rf1,c)))
#endif
#if defined(FL4_SET)
#define fwd_lrnd(y,x,k,c) (s(y,c) = (k)[c] ^ four_tables(x,t_use(f,l),fwd_var,rf1,c))
#elif defined(FL1_SET)
#define fwd_lrnd(y,x,k,c) (s(y,c) = (k)[c] ^ one_table(x,ups,t_use(f,l),fwd_var,rf1,c))
#else
#define fwd_lrnd(y,x,k,c) (s(y,c) = (k)[c] ^ no_table(x,t_use(s,box),fwd_var,rf1,c))
#endif
aes_rval aes_encrypt(const void *in_blk, void *out_blk, const aes_encrypt_ctx cx[1])
{ aes_32t locals(b0, b1);
const aes_32t *kp = cx->ks;
#ifdef dec_fmvars
dec_fmvars; /* declare variables for fwd_mcol() if needed */
#endif
aes_32t nr = (kp[45] ^ kp[52] ^ kp[53] ? kp[52] : 14);
#ifdef AES_ERR_CHK
if( (nr != 10 || !(kp[0] | kp[3] | kp[4]))
&& (nr != 12 || !(kp[0] | kp[5] | kp[6]))
&& (nr != 14 || !(kp[0] | kp[7] | kp[8])) )
return aes_error;
#endif
state_in(b0, in_blk, kp);
#if (ENC_UNROLL == FULL)
switch(nr)
{
case 14:
round(fwd_rnd, b1, b0, kp + 1 * N_COLS);
round(fwd_rnd, b0, b1, kp + 2 * N_COLS);
kp += 2 * N_COLS;
case 12:
round(fwd_rnd, b1, b0, kp + 1 * N_COLS);
round(fwd_rnd, b0, b1, kp + 2 * N_COLS);
kp += 2 * N_COLS;
case 10:
round(fwd_rnd, b1, b0, kp + 1 * N_COLS);
round(fwd_rnd, b0, b1, kp + 2 * N_COLS);
round(fwd_rnd, b1, b0, kp + 3 * N_COLS);
round(fwd_rnd, b0, b1, kp + 4 * N_COLS);
round(fwd_rnd, b1, b0, kp + 5 * N_COLS);
round(fwd_rnd, b0, b1, kp + 6 * N_COLS);
round(fwd_rnd, b1, b0, kp + 7 * N_COLS);
round(fwd_rnd, b0, b1, kp + 8 * N_COLS);
round(fwd_rnd, b1, b0, kp + 9 * N_COLS);
round(fwd_lrnd, b0, b1, kp +10 * N_COLS);
}
#else
#if (ENC_UNROLL == PARTIAL)
{ aes_32t rnd;
for(rnd = 0; rnd < (nr >> 1) - 1; ++rnd)
{
kp += N_COLS;
round(fwd_rnd, b1, b0, kp);
kp += N_COLS;
round(fwd_rnd, b0, b1, kp);
}
kp += N_COLS;
round(fwd_rnd, b1, b0, kp);
#else
{ aes_32t rnd;
for(rnd = 0; rnd < nr - 1; ++rnd)
{
kp += N_COLS;
round(fwd_rnd, b1, b0, kp);
l_copy(b0, b1);
}
#endif
kp += N_COLS;
round(fwd_lrnd, b0, b1, kp);
}
#endif
state_out(out_blk, b0);
#ifdef AES_ERR_CHK
return aes_good;
#endif
}
#endif
#if defined(DECRYPTION) && !defined(AES_ASM)
/* Visual C++ .Net v7.1 provides the fastest encryption code when using
Pentium optimization with small code but this is poor for decryption
so we need to control this with the following VC++ pragmas
*/
#if defined(_MSC_VER)
#pragma optimize( "t", on )
#endif
/* Given the column (c) of the output state variable, the following
macros give the input state variables which are needed in its
computation for each row (r) of the state. All the alternative
macros give the same end values but expand into different ways
of calculating these values. In particular the complex macro
used for dynamically variable block sizes is designed to expand
to a compile time constant whenever possible but will expand to
conditional clauses on some branches (I am grateful to Frank
Yellin for this construction)
*/
#define inv_var(x,r,c)\
( r == 0 ? ( c == 0 ? s(x,0) : c == 1 ? s(x,1) : c == 2 ? s(x,2) : s(x,3))\
: r == 1 ? ( c == 0 ? s(x,3) : c == 1 ? s(x,0) : c == 2 ? s(x,1) : s(x,2))\
: r == 2 ? ( c == 0 ? s(x,2) : c == 1 ? s(x,3) : c == 2 ? s(x,0) : s(x,1))\
: ( c == 0 ? s(x,1) : c == 1 ? s(x,2) : c == 2 ? s(x,3) : s(x,0)))
#if defined(IT4_SET)
#undef dec_imvars
#define inv_rnd(y,x,k,c) (s(y,c) = (k)[c] ^ four_tables(x,t_use(i,n),inv_var,rf1,c))
#elif defined(IT1_SET)
#undef dec_imvars
#define inv_rnd(y,x,k,c) (s(y,c) = (k)[c] ^ one_table(x,upr,t_use(i,n),inv_var,rf1,c))
#else
#define inv_rnd(y,x,k,c) (s(y,c) = inv_mcol((k)[c] ^ no_table(x,t_use(i,box),inv_var,rf1,c)))
#endif
#if defined(IL4_SET)
#define inv_lrnd(y,x,k,c) (s(y,c) = (k)[c] ^ four_tables(x,t_use(i,l),inv_var,rf1,c))
#elif defined(IL1_SET)
#define inv_lrnd(y,x,k,c) (s(y,c) = (k)[c] ^ one_table(x,ups,t_use(i,l),inv_var,rf1,c))
#else
#define inv_lrnd(y,x,k,c) (s(y,c) = (k)[c] ^ no_table(x,t_use(i,box),inv_var,rf1,c))
#endif
aes_rval aes_decrypt(const void *in_blk, void *out_blk, const aes_decrypt_ctx cx[1])
{ aes_32t locals(b0, b1);
#ifdef dec_imvars
dec_imvars; /* declare variables for inv_mcol() if needed */
#endif
aes_32t nr = (cx->ks[45] ^ cx->ks[52] ^ cx->ks[53] ? cx->ks[52] : 14);
const aes_32t *kp = cx->ks + nr * N_COLS;
#ifdef AES_ERR_CHK
if( (nr != 10 || !(cx->ks[0] | cx->ks[3] | cx->ks[4]))
&& (nr != 12 || !(cx->ks[0] | cx->ks[5] | cx->ks[6]))
&& (nr != 14 || !(cx->ks[0] | cx->ks[7] | cx->ks[8])) )
return aes_error;
#endif
state_in(b0, in_blk, kp);
#if (DEC_UNROLL == FULL)
switch(nr)
{
case 14:
round(inv_rnd, b1, b0, kp - 1 * N_COLS);
round(inv_rnd, b0, b1, kp - 2 * N_COLS);
kp -= 2 * N_COLS;
case 12:
round(inv_rnd, b1, b0, kp - 1 * N_COLS);
round(inv_rnd, b0, b1, kp - 2 * N_COLS);
kp -= 2 * N_COLS;
case 10:
round(inv_rnd, b1, b0, kp - 1 * N_COLS);
round(inv_rnd, b0, b1, kp - 2 * N_COLS);
round(inv_rnd, b1, b0, kp - 3 * N_COLS);
round(inv_rnd, b0, b1, kp - 4 * N_COLS);
round(inv_rnd, b1, b0, kp - 5 * N_COLS);
round(inv_rnd, b0, b1, kp - 6 * N_COLS);
round(inv_rnd, b1, b0, kp - 7 * N_COLS);
round(inv_rnd, b0, b1, kp - 8 * N_COLS);
round(inv_rnd, b1, b0, kp - 9 * N_COLS);
round(inv_lrnd, b0, b1, kp - 10 * N_COLS);
}
#else
#if (DEC_UNROLL == PARTIAL)
{ aes_32t rnd;
for(rnd = 0; rnd < (nr >> 1) - 1; ++rnd)
{
kp -= N_COLS;
round(inv_rnd, b1, b0, kp);
kp -= N_COLS;
round(inv_rnd, b0, b1, kp);
}
kp -= N_COLS;
round(inv_rnd, b1, b0, kp);
#else
{ aes_32t rnd;
for(rnd = 0; rnd < nr - 1; ++rnd)
{
kp -= N_COLS;
round(inv_rnd, b1, b0, kp);
l_copy(b0, b1);
}
#endif
kp -= N_COLS;
round(inv_lrnd, b0, b1, kp);
}
#endif
state_out(out_blk, b0);
#ifdef AES_ERR_CHK
return aes_good;
#endif
}
#endif
|
