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
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
|
/* SSSE3 vector permutation AES for Libgcrypt
* Copyright (C) 2014-2017 Jussi Kivilinna <jussi.kivilinna@iki.fi>
*
* This file is part of Libgcrypt.
*
* Libgcrypt is free software; you can redistribute it and/or modify
* it under the terms of the GNU Lesser General Public License as
* published by the Free Software Foundation; either version 2.1 of
* the License, or (at your option) any later version.
*
* Libgcrypt is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU Lesser General Public License for more details.
*
* You should have received a copy of the GNU Lesser General Public
* License along with this program; if not, see <http://www.gnu.org/licenses/>.
*
*
* The code is based on the public domain library libvpaes version 0.5
* available at http://crypto.stanford.edu/vpaes/ and which carries
* this notice:
*
* libvpaes: constant-time SSSE3 AES encryption and decryption.
* version 0.5
*
* By Mike Hamburg, Stanford University, 2009. Public domain.
* I wrote essentially all of this code. I did not write the test
* vectors; they are the NIST known answer tests. I hereby release all
* the code and documentation here that I wrote into the public domain.
*
* This is an implementation of AES following my paper,
* "Accelerating AES with Vector Permute Instructions"
* CHES 2009; http://shiftleft.org/papers/vector_aes/
*/
#include <config.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h> /* for memcmp() */
#include "types.h" /* for byte and u32 typedefs */
#include "g10lib.h"
#include "cipher.h"
#include "bufhelp.h"
#include "rijndael-internal.h"
#include "./cipher-internal.h"
#ifdef USE_SSSE3
#if _GCRY_GCC_VERSION >= 40400 /* 4.4 */
/* Prevent compiler from issuing SSE instructions between asm blocks. */
# pragma GCC target("no-sse")
#endif
#if __clang__
# pragma clang attribute push (__attribute__((target("no-sse"))), apply_to = function)
#endif
#define ALWAYS_INLINE inline __attribute__((always_inline))
#define NO_INSTRUMENT_FUNCTION __attribute__((no_instrument_function))
#define ASM_FUNC_ATTR NO_INSTRUMENT_FUNCTION
#define ASM_FUNC_ATTR_INLINE ASM_FUNC_ATTR ALWAYS_INLINE
/* Copy of ocb_get_l needed here as GCC is unable to inline ocb_get_l
because of 'pragma target'. */
static ASM_FUNC_ATTR_INLINE const unsigned char *
aes_ocb_get_l (gcry_cipher_hd_t c, u64 n)
{
unsigned long ntz;
/* Assumes that N != 0. */
asm ("rep;bsfl %k[low], %k[ntz]\n\t"
: [ntz] "=r" (ntz)
: [low] "r" ((unsigned long)n)
: "cc");
return c->u_mode.ocb.L[ntz];
}
/* Assembly functions in rijndael-ssse3-amd64-asm.S. Note that these
have custom calling convention (additional XMM parameters). */
extern void _gcry_aes_ssse3_enc_preload(void);
extern void _gcry_aes_ssse3_dec_preload(void);
extern void _gcry_aes_ssse3_schedule_core(const void *key, u64 keybits,
void *buffer, u64 decrypt,
u64 rotoffs);
extern void _gcry_aes_ssse3_encrypt_core(const void *key, u64 nrounds);
extern void _gcry_aes_ssse3_decrypt_core(const void *key, u64 nrounds);
/* Two macros to be called prior and after the use of SSSE3
instructions. There should be no external function calls between
the use of these macros. There purpose is to make sure that the
SSE registers are cleared and won't reveal any information about
the key or the data. */
#ifdef HAVE_COMPATIBLE_GCC_WIN64_PLATFORM_AS
# define SSSE3_STATE_SIZE (16 * 10)
/* XMM6-XMM15 are callee-saved registers on WIN64. */
# define vpaes_ssse3_prepare() \
asm volatile ("movdqu %%xmm6, 0*16(%0)\n\t" \
"movdqu %%xmm7, 1*16(%0)\n\t" \
"movdqu %%xmm8, 2*16(%0)\n\t" \
"movdqu %%xmm9, 3*16(%0)\n\t" \
"movdqu %%xmm10, 4*16(%0)\n\t" \
"movdqu %%xmm11, 5*16(%0)\n\t" \
"movdqu %%xmm12, 6*16(%0)\n\t" \
"movdqu %%xmm13, 7*16(%0)\n\t" \
"movdqu %%xmm14, 8*16(%0)\n\t" \
"movdqu %%xmm15, 9*16(%0)\n\t" \
: \
: "r" (ssse3_state) \
: "memory" )
# define vpaes_ssse3_cleanup() \
asm volatile ("pxor %%xmm0, %%xmm0 \n\t" \
"pxor %%xmm1, %%xmm1 \n\t" \
"pxor %%xmm2, %%xmm2 \n\t" \
"pxor %%xmm3, %%xmm3 \n\t" \
"pxor %%xmm4, %%xmm4 \n\t" \
"pxor %%xmm5, %%xmm5 \n\t" \
"movdqu 0*16(%0), %%xmm6 \n\t" \
"movdqu 1*16(%0), %%xmm7 \n\t" \
"movdqu 2*16(%0), %%xmm8 \n\t" \
"movdqu 3*16(%0), %%xmm9 \n\t" \
"movdqu 4*16(%0), %%xmm10 \n\t" \
"movdqu 5*16(%0), %%xmm11 \n\t" \
"movdqu 6*16(%0), %%xmm12 \n\t" \
"movdqu 7*16(%0), %%xmm13 \n\t" \
"movdqu 8*16(%0), %%xmm14 \n\t" \
"movdqu 9*16(%0), %%xmm15 \n\t" \
: \
: "r" (ssse3_state) \
: "memory" )
#else
# define SSSE3_STATE_SIZE 1
# define vpaes_ssse3_prepare() (void)ssse3_state
# define vpaes_ssse3_cleanup() \
asm volatile ("pxor %%xmm0, %%xmm0 \n\t" \
"pxor %%xmm1, %%xmm1 \n\t" \
"pxor %%xmm2, %%xmm2 \n\t" \
"pxor %%xmm3, %%xmm3 \n\t" \
"pxor %%xmm4, %%xmm4 \n\t" \
"pxor %%xmm5, %%xmm5 \n\t" \
"pxor %%xmm6, %%xmm6 \n\t" \
"pxor %%xmm7, %%xmm7 \n\t" \
"pxor %%xmm8, %%xmm8 \n\t" \
::: "memory" )
#endif
#define vpaes_ssse3_prepare_enc() \
vpaes_ssse3_prepare(); \
_gcry_aes_ssse3_enc_preload();
#define vpaes_ssse3_prepare_dec() \
vpaes_ssse3_prepare(); \
_gcry_aes_ssse3_dec_preload();
void ASM_FUNC_ATTR
_gcry_aes_ssse3_do_setkey (RIJNDAEL_context *ctx, const byte *key)
{
unsigned int keybits = (ctx->rounds - 10) * 32 + 128;
byte ssse3_state[SSSE3_STATE_SIZE];
vpaes_ssse3_prepare();
_gcry_aes_ssse3_schedule_core(key, keybits, &ctx->keyschenc32[0][0], 0, 48);
/* Save key for setting up decryption. */
if (keybits > 192)
asm volatile ("movdqu (%[src]), %%xmm0\n\t"
"movdqu 16(%[src]), %%xmm1\n\t"
"movdqu %%xmm0, (%[dst])\n\t"
"movdqu %%xmm1, 16(%[dst])\n\t"
: /* No output */
: [dst] "r" (&ctx->keyschdec32[0][0]), [src] "r" (key)
: "memory" );
else if (keybits == 192)
asm volatile ("movdqu (%[src]), %%xmm0\n\t"
"movq 16(%[src]), %%xmm1\n\t"
"movdqu %%xmm0, (%[dst])\n\t"
"movq %%xmm1, 16(%[dst])\n\t"
: /* No output */
: [dst] "r" (&ctx->keyschdec32[0][0]), [src] "r" (key)
: "memory" );
else
asm volatile ("movdqu (%[src]), %%xmm0\n\t"
"movdqu %%xmm0, (%[dst])\n\t"
: /* No output */
: [dst] "r" (&ctx->keyschdec32[0][0]), [src] "r" (key)
: "memory" );
vpaes_ssse3_cleanup();
}
/* Make a decryption key from an encryption key. */
static ASM_FUNC_ATTR_INLINE void
do_ssse3_prepare_decryption (RIJNDAEL_context *ctx,
byte ssse3_state[SSSE3_STATE_SIZE])
{
unsigned int keybits = (ctx->rounds - 10) * 32 + 128;
vpaes_ssse3_prepare();
_gcry_aes_ssse3_schedule_core(&ctx->keyschdec32[0][0], keybits,
&ctx->keyschdec32[ctx->rounds][0], 1,
(keybits == 192) ? 0 : 32);
vpaes_ssse3_cleanup();
}
void ASM_FUNC_ATTR
_gcry_aes_ssse3_prepare_decryption (RIJNDAEL_context *ctx)
{
byte ssse3_state[SSSE3_STATE_SIZE];
do_ssse3_prepare_decryption(ctx, ssse3_state);
}
/* Encrypt one block using the Intel SSSE3 instructions. Block is input
* and output through SSE register xmm0. */
static ASM_FUNC_ATTR_INLINE void
do_vpaes_ssse3_enc (const RIJNDAEL_context *ctx, unsigned int nrounds)
{
_gcry_aes_ssse3_encrypt_core(ctx->keyschenc32, nrounds);
}
/* Decrypt one block using the Intel SSSE3 instructions. Block is input
* and output through SSE register xmm0. */
static ASM_FUNC_ATTR_INLINE void
do_vpaes_ssse3_dec (const RIJNDAEL_context *ctx, unsigned int nrounds)
{
_gcry_aes_ssse3_decrypt_core(ctx->keyschdec32, nrounds);
}
unsigned int ASM_FUNC_ATTR
_gcry_aes_ssse3_encrypt (const RIJNDAEL_context *ctx, unsigned char *dst,
const unsigned char *src)
{
unsigned int nrounds = ctx->rounds;
byte ssse3_state[SSSE3_STATE_SIZE];
vpaes_ssse3_prepare_enc ();
asm volatile ("movdqu %[src], %%xmm0\n\t"
:
: [src] "m" (*src)
: "memory" );
do_vpaes_ssse3_enc (ctx, nrounds);
asm volatile ("movdqu %%xmm0, %[dst]\n\t"
: [dst] "=m" (*dst)
:
: "memory" );
vpaes_ssse3_cleanup ();
return 0;
}
void ASM_FUNC_ATTR
_gcry_aes_ssse3_cfb_enc (RIJNDAEL_context *ctx, unsigned char *iv,
unsigned char *outbuf, const unsigned char *inbuf,
size_t nblocks)
{
unsigned int nrounds = ctx->rounds;
byte ssse3_state[SSSE3_STATE_SIZE];
vpaes_ssse3_prepare_enc ();
asm volatile ("movdqu %[iv], %%xmm0\n\t"
: /* No output */
: [iv] "m" (*iv)
: "memory" );
for ( ;nblocks; nblocks-- )
{
do_vpaes_ssse3_enc (ctx, nrounds);
asm volatile ("movdqu %[inbuf], %%xmm1\n\t"
"pxor %%xmm1, %%xmm0\n\t"
"movdqu %%xmm0, %[outbuf]\n\t"
: [outbuf] "=m" (*outbuf)
: [inbuf] "m" (*inbuf)
: "memory" );
outbuf += BLOCKSIZE;
inbuf += BLOCKSIZE;
}
asm volatile ("movdqu %%xmm0, %[iv]\n\t"
: [iv] "=m" (*iv)
:
: "memory" );
vpaes_ssse3_cleanup ();
}
void ASM_FUNC_ATTR
_gcry_aes_ssse3_cbc_enc (RIJNDAEL_context *ctx, unsigned char *iv,
unsigned char *outbuf, const unsigned char *inbuf,
size_t nblocks, int cbc_mac)
{
unsigned int nrounds = ctx->rounds;
byte ssse3_state[SSSE3_STATE_SIZE];
vpaes_ssse3_prepare_enc ();
asm volatile ("movdqu %[iv], %%xmm7\n\t"
: /* No output */
: [iv] "m" (*iv)
: "memory" );
for ( ;nblocks; nblocks-- )
{
asm volatile ("movdqu %[inbuf], %%xmm0\n\t"
"pxor %%xmm7, %%xmm0\n\t"
: /* No output */
: [inbuf] "m" (*inbuf)
: "memory" );
do_vpaes_ssse3_enc (ctx, nrounds);
asm volatile ("movdqa %%xmm0, %%xmm7\n\t"
"movdqu %%xmm0, %[outbuf]\n\t"
: [outbuf] "=m" (*outbuf)
:
: "memory" );
inbuf += BLOCKSIZE;
if (!cbc_mac)
outbuf += BLOCKSIZE;
}
asm volatile ("movdqu %%xmm7, %[iv]\n\t"
: [iv] "=m" (*iv)
:
: "memory" );
vpaes_ssse3_cleanup ();
}
void ASM_FUNC_ATTR
_gcry_aes_ssse3_ctr_enc (RIJNDAEL_context *ctx, unsigned char *ctr,
unsigned char *outbuf, const unsigned char *inbuf,
size_t nblocks)
{
static const unsigned char be_mask[16] __attribute__ ((aligned (16))) =
{ 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0 };
unsigned int nrounds = ctx->rounds;
byte ssse3_state[SSSE3_STATE_SIZE];
u64 ctrlow;
vpaes_ssse3_prepare_enc ();
asm volatile ("movdqa %[mask], %%xmm6\n\t" /* Preload mask */
"movdqa (%[ctr]), %%xmm7\n\t" /* Preload CTR */
"movq 8(%[ctr]), %q[ctrlow]\n\t"
"bswapq %q[ctrlow]\n\t"
: [ctrlow] "=r" (ctrlow)
: [mask] "m" (*be_mask),
[ctr] "r" (ctr)
: "memory", "cc");
for ( ;nblocks; nblocks-- )
{
asm volatile ("movdqa %%xmm7, %%xmm0\n\t" /* xmm0 := CTR (xmm7) */
"pcmpeqd %%xmm1, %%xmm1\n\t"
"psrldq $8, %%xmm1\n\t" /* xmm1 = -1 */
"pshufb %%xmm6, %%xmm7\n\t"
"psubq %%xmm1, %%xmm7\n\t" /* xmm7++ (big endian) */
/* detect if 64-bit carry handling is needed */
"incq %q[ctrlow]\n\t"
"jnz .Lno_carry%=\n\t"
"pslldq $8, %%xmm1\n\t" /* move lower 64-bit to high */
"psubq %%xmm1, %%xmm7\n\t" /* add carry to upper 64bits */
".Lno_carry%=:\n\t"
"pshufb %%xmm6, %%xmm7\n\t"
: [ctrlow] "+r" (ctrlow)
:
: "cc", "memory");
do_vpaes_ssse3_enc (ctx, nrounds);
asm volatile ("movdqu %[src], %%xmm1\n\t" /* xmm1 := input */
"pxor %%xmm1, %%xmm0\n\t" /* EncCTR ^= input */
"movdqu %%xmm0, %[dst]" /* Store EncCTR. */
: [dst] "=m" (*outbuf)
: [src] "m" (*inbuf)
: "memory");
outbuf += BLOCKSIZE;
inbuf += BLOCKSIZE;
}
asm volatile ("movdqu %%xmm7, %[ctr]\n\t" /* Update CTR (mem). */
: [ctr] "=m" (*ctr)
:
: "memory" );
vpaes_ssse3_cleanup ();
}
unsigned int ASM_FUNC_ATTR
_gcry_aes_ssse3_decrypt (const RIJNDAEL_context *ctx, unsigned char *dst,
const unsigned char *src)
{
unsigned int nrounds = ctx->rounds;
byte ssse3_state[SSSE3_STATE_SIZE];
vpaes_ssse3_prepare_dec ();
asm volatile ("movdqu %[src], %%xmm0\n\t"
:
: [src] "m" (*src)
: "memory" );
do_vpaes_ssse3_dec (ctx, nrounds);
asm volatile ("movdqu %%xmm0, %[dst]\n\t"
: [dst] "=m" (*dst)
:
: "memory" );
vpaes_ssse3_cleanup ();
return 0;
}
void ASM_FUNC_ATTR
_gcry_aes_ssse3_cfb_dec (RIJNDAEL_context *ctx, unsigned char *iv,
unsigned char *outbuf, const unsigned char *inbuf,
size_t nblocks)
{
unsigned int nrounds = ctx->rounds;
byte ssse3_state[SSSE3_STATE_SIZE];
vpaes_ssse3_prepare_enc ();
asm volatile ("movdqu %[iv], %%xmm0\n\t"
: /* No output */
: [iv] "m" (*iv)
: "memory" );
for ( ;nblocks; nblocks-- )
{
do_vpaes_ssse3_enc (ctx, nrounds);
asm volatile ("movdqa %%xmm0, %%xmm6\n\t"
"movdqu %[inbuf], %%xmm0\n\t"
"pxor %%xmm0, %%xmm6\n\t"
"movdqu %%xmm6, %[outbuf]\n\t"
: [outbuf] "=m" (*outbuf)
: [inbuf] "m" (*inbuf)
: "memory" );
outbuf += BLOCKSIZE;
inbuf += BLOCKSIZE;
}
asm volatile ("movdqu %%xmm0, %[iv]\n\t"
: [iv] "=m" (*iv)
:
: "memory" );
vpaes_ssse3_cleanup ();
}
void ASM_FUNC_ATTR
_gcry_aes_ssse3_cbc_dec (RIJNDAEL_context *ctx, unsigned char *iv,
unsigned char *outbuf, const unsigned char *inbuf,
size_t nblocks)
{
unsigned int nrounds = ctx->rounds;
byte ssse3_state[SSSE3_STATE_SIZE];
if ( !ctx->decryption_prepared )
{
do_ssse3_prepare_decryption ( ctx, ssse3_state );
ctx->decryption_prepared = 1;
}
vpaes_ssse3_prepare_dec ();
asm volatile ("movdqu %[iv], %%xmm7\n\t" /* use xmm7 as fast IV storage */
: /* No output */
: [iv] "m" (*iv)
: "memory");
for ( ;nblocks; nblocks-- )
{
asm volatile ("movdqu %[inbuf], %%xmm0\n\t"
"movdqa %%xmm0, %%xmm6\n\t" /* use xmm6 as savebuf */
: /* No output */
: [inbuf] "m" (*inbuf)
: "memory");
do_vpaes_ssse3_dec (ctx, nrounds);
asm volatile ("pxor %%xmm7, %%xmm0\n\t" /* xor IV with output */
"movdqu %%xmm0, %[outbuf]\n\t"
"movdqu %%xmm6, %%xmm7\n\t" /* store savebuf as new IV */
: [outbuf] "=m" (*outbuf)
:
: "memory");
outbuf += BLOCKSIZE;
inbuf += BLOCKSIZE;
}
asm volatile ("movdqu %%xmm7, %[iv]\n\t" /* store IV */
: /* No output */
: [iv] "m" (*iv)
: "memory");
vpaes_ssse3_cleanup ();
}
static void ASM_FUNC_ATTR
ssse3_ocb_enc (gcry_cipher_hd_t c, void *outbuf_arg,
const void *inbuf_arg, size_t nblocks)
{
RIJNDAEL_context *ctx = (void *)&c->context.c;
unsigned char *outbuf = outbuf_arg;
const unsigned char *inbuf = inbuf_arg;
u64 n = c->u_mode.ocb.data_nblocks;
unsigned int nrounds = ctx->rounds;
byte ssse3_state[SSSE3_STATE_SIZE];
vpaes_ssse3_prepare_enc ();
/* Preload Offset and Checksum */
asm volatile ("movdqu %[iv], %%xmm7\n\t"
"movdqu %[ctr], %%xmm6\n\t"
: /* No output */
: [iv] "m" (*c->u_iv.iv),
[ctr] "m" (*c->u_ctr.ctr)
: "memory" );
for ( ;nblocks; nblocks-- )
{
const unsigned char *l;
l = aes_ocb_get_l(c, ++n);
/* Offset_i = Offset_{i-1} xor L_{ntz(i)} */
/* Checksum_i = Checksum_{i-1} xor P_i */
/* C_i = Offset_i xor ENCIPHER(K, P_i xor Offset_i) */
asm volatile ("movdqu %[l], %%xmm1\n\t"
"movdqu %[inbuf], %%xmm0\n\t"
"pxor %%xmm1, %%xmm7\n\t"
"pxor %%xmm0, %%xmm6\n\t"
"pxor %%xmm7, %%xmm0\n\t"
:
: [l] "m" (*l),
[inbuf] "m" (*inbuf)
: "memory" );
do_vpaes_ssse3_enc (ctx, nrounds);
asm volatile ("pxor %%xmm7, %%xmm0\n\t"
"movdqu %%xmm0, %[outbuf]\n\t"
: [outbuf] "=m" (*outbuf)
:
: "memory" );
inbuf += BLOCKSIZE;
outbuf += BLOCKSIZE;
}
c->u_mode.ocb.data_nblocks = n;
asm volatile ("movdqu %%xmm7, %[iv]\n\t"
"movdqu %%xmm6, %[ctr]\n\t"
: [iv] "=m" (*c->u_iv.iv),
[ctr] "=m" (*c->u_ctr.ctr)
:
: "memory" );
vpaes_ssse3_cleanup ();
}
static void ASM_FUNC_ATTR
ssse3_ocb_dec (gcry_cipher_hd_t c, void *outbuf_arg,
const void *inbuf_arg, size_t nblocks)
{
RIJNDAEL_context *ctx = (void *)&c->context.c;
unsigned char *outbuf = outbuf_arg;
const unsigned char *inbuf = inbuf_arg;
u64 n = c->u_mode.ocb.data_nblocks;
unsigned int nrounds = ctx->rounds;
byte ssse3_state[SSSE3_STATE_SIZE];
if ( !ctx->decryption_prepared )
{
do_ssse3_prepare_decryption ( ctx, ssse3_state );
ctx->decryption_prepared = 1;
}
vpaes_ssse3_prepare_dec ();
/* Preload Offset and Checksum */
asm volatile ("movdqu %[iv], %%xmm7\n\t"
"movdqu %[ctr], %%xmm6\n\t"
: /* No output */
: [iv] "m" (*c->u_iv.iv),
[ctr] "m" (*c->u_ctr.ctr)
: "memory" );
for ( ;nblocks; nblocks-- )
{
const unsigned char *l;
l = aes_ocb_get_l(c, ++n);
/* Offset_i = Offset_{i-1} xor L_{ntz(i)} */
/* P_i = Offset_i xor DECIPHER(K, C_i xor Offset_i) */
/* Checksum_i = Checksum_{i-1} xor P_i */
asm volatile ("movdqu %[l], %%xmm1\n\t"
"movdqu %[inbuf], %%xmm0\n\t"
"pxor %%xmm1, %%xmm7\n\t"
"pxor %%xmm7, %%xmm0\n\t"
:
: [l] "m" (*l),
[inbuf] "m" (*inbuf)
: "memory" );
do_vpaes_ssse3_dec (ctx, nrounds);
asm volatile ("pxor %%xmm7, %%xmm0\n\t"
"pxor %%xmm0, %%xmm6\n\t"
"movdqu %%xmm0, %[outbuf]\n\t"
: [outbuf] "=m" (*outbuf)
:
: "memory" );
inbuf += BLOCKSIZE;
outbuf += BLOCKSIZE;
}
c->u_mode.ocb.data_nblocks = n;
asm volatile ("movdqu %%xmm7, %[iv]\n\t"
"movdqu %%xmm6, %[ctr]\n\t"
: [iv] "=m" (*c->u_iv.iv),
[ctr] "=m" (*c->u_ctr.ctr)
:
: "memory" );
vpaes_ssse3_cleanup ();
}
size_t ASM_FUNC_ATTR
_gcry_aes_ssse3_ocb_crypt(gcry_cipher_hd_t c, void *outbuf_arg,
const void *inbuf_arg, size_t nblocks, int encrypt)
{
if (encrypt)
ssse3_ocb_enc(c, outbuf_arg, inbuf_arg, nblocks);
else
ssse3_ocb_dec(c, outbuf_arg, inbuf_arg, nblocks);
return 0;
}
size_t ASM_FUNC_ATTR
_gcry_aes_ssse3_ocb_auth (gcry_cipher_hd_t c, const void *abuf_arg,
size_t nblocks)
{
RIJNDAEL_context *ctx = (void *)&c->context.c;
const unsigned char *abuf = abuf_arg;
u64 n = c->u_mode.ocb.aad_nblocks;
unsigned int nrounds = ctx->rounds;
byte ssse3_state[SSSE3_STATE_SIZE];
vpaes_ssse3_prepare_enc ();
/* Preload Offset and Sum */
asm volatile ("movdqu %[iv], %%xmm7\n\t"
"movdqu %[ctr], %%xmm6\n\t"
: /* No output */
: [iv] "m" (*c->u_mode.ocb.aad_offset),
[ctr] "m" (*c->u_mode.ocb.aad_sum)
: "memory" );
for ( ;nblocks; nblocks-- )
{
const unsigned char *l;
l = aes_ocb_get_l(c, ++n);
/* Offset_i = Offset_{i-1} xor L_{ntz(i)} */
/* Sum_i = Sum_{i-1} xor ENCIPHER(K, A_i xor Offset_i) */
asm volatile ("movdqu %[l], %%xmm1\n\t"
"movdqu %[abuf], %%xmm0\n\t"
"pxor %%xmm1, %%xmm7\n\t"
"pxor %%xmm7, %%xmm0\n\t"
:
: [l] "m" (*l),
[abuf] "m" (*abuf)
: "memory" );
do_vpaes_ssse3_enc (ctx, nrounds);
asm volatile ("pxor %%xmm0, %%xmm6\n\t"
:
:
: "memory" );
abuf += BLOCKSIZE;
}
c->u_mode.ocb.aad_nblocks = n;
asm volatile ("movdqu %%xmm7, %[iv]\n\t"
"movdqu %%xmm6, %[ctr]\n\t"
: [iv] "=m" (*c->u_mode.ocb.aad_offset),
[ctr] "=m" (*c->u_mode.ocb.aad_sum)
:
: "memory" );
vpaes_ssse3_cleanup ();
return 0;
}
#if __clang__
# pragma clang attribute pop
#endif
#endif /* USE_SSSE3 */
|