-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathaesni.c
1790 lines (1701 loc) · 58.3 KB
/
aesni.c
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
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
/** aesni.c
* Implementation of AES en/decryption
* using intel's AES-NI instruction set.
* From
* https://software.intel.com/sites/default/files/article/165683/aes-wp-2012-09-22-v01.pdf
* adapted by
* Kurt Garloff <[email protected]>, 7/2014
*/
#include "aes.h"
#include "aesni.h"
#include "secmem.h"
#include <string.h>
#include "archdep.h"
#if defined(__x86_64__) || defined(__i386__) || defined(__i486__) || defined(__i586__) || defined(__i686__)
#include <wmmintrin.h>
#if defined(__AVX2__) && !defined(NO_AVX2)
#include <immintrin.h>
static int probe_aes_ni()
{
return !have_aesni || !have_avx2;
}
#else /* AVX2 */
#include <smmintrin.h>
static int probe_aes_ni()
{
return !have_aesni;
}
#endif /* AVX2 */
#endif /* x86 */
/* Unaligned 128bit integer type, missing from gcc < 5 emmintrin.h */
typedef long long __m128i_u __attribute__ ((__vector_size__ (16), __may_alias__, __aligned__ (1)));
#define MMCLEAR(xmmreg) xmmreg = _mm_setzero_si128()
#define MM256CLEAR(xmmreg) xmmreg = _mm256_setzero_si256()
#define SIZE128 (ssize_t)sizeof(__m128i)
#define SIZE256 (ssize_t)sizeof(__m256i)
#define MMCLEAR3 \
asm volatile(" pxor %%xmm2, %%xmm2 \n" \
" pxor %%xmm1, %%xmm1 \n" \
" pxor %%xmm0, %%xmm0 \n" \
: : : "xmm2", "xmm1", "xmm0")
#define MMCLEAR4 \
asm volatile(" pxor %%xmm3, %%xmm3 \n" \
" pxor %%xmm2, %%xmm2 \n" \
" pxor %%xmm1, %%xmm1 \n" \
" pxor %%xmm0, %%xmm0 \n" \
: : : "xmm3", "xmm2", "xmm1", "xmm0")
#define MMCLEAR5 \
asm volatile(" pxor %%xmm4, %%xmm4 \n" \
" pxor %%xmm3, %%xmm3 \n" \
" pxor %%xmm2, %%xmm2 \n" \
" pxor %%xmm1, %%xmm1 \n" \
" pxor %%xmm0, %%xmm0 \n" \
: : : "xmm4", "xmm3", "xmm2", "xmm1", "xmm0")
#define MMCLEARALL_MAN \
asm volatile(" pxor %%xmm7, %%xmm7 \n" \
" pxor %%xmm6, %%xmm6 \n" \
" pxor %%xmm5, %%xmm5 \n" \
" pxor %%xmm4, %%xmm4 \n" \
" pxor %%xmm3, %%xmm3 \n" \
" pxor %%xmm2, %%xmm2 \n" \
" pxor %%xmm1, %%xmm1 \n" \
" pxor %%xmm0, %%xmm0 \n" \
: : : "xmm7", "xmm6", "xmm5", "xmm4", "xmm3", "xmm2", "xmm1", "xmm0")
#if defined(__x86_64__)
#define MMCLEARALL_MAN2 \
asm volatile(" pxor %%xmm15, %%xmm15 \n" \
" pxor %%xmm14, %%xmm14 \n" \
" pxor %%xmm13, %%xmm13 \n" \
" pxor %%xmm12, %%xmm12 \n" \
" pxor %%xmm11, %%xmm11 \n" \
" pxor %%xmm10, %%xmm10 \n" \
" pxor %%xmm9, %%xmm9 \n" \
" pxor %%xmm8, %%xmm8 \n" \
: : : "xmm15", "xmm14", "xmm13", "xmm12", "xmm11", "xmm10", "xmm9", "xmm8");
#else
#define MMCLEARALL_MAN2 do {} while(0)
#endif
#if defined(__AVX2__) && !defined(NO_AVX2)
#define MMCLEARALL _mm256_zeroall()
#define MM256CLEAR5 \
asm volatile(" vpxor %%ymm4, %%ymm4, %%ymm4 \n" \
" vpxor %%ymm3, %%ymm3, %%ymm3 \n" \
" vpxor %%ymm2, %%ymm2, %%ymm2 \n" \
" vpxor %%ymm1, %%ymm1, %%ymm1 \n" \
" vpxor %%ymm0, %%ymm0, %%ymm0 \n" \
: : : "ymm4", "ymm3", "ymm2", "ymm1", "ymm0")
#else
#define MMCLEARALL do { MMCLEARALL_MAN; MMCLEARALL_MAN2; } while(0)
#endif
static inline __m128i KEY_128_ASSIST(__m128i temp1, __m128i temp2)
{
register __m128i temp3;
temp2 = _mm_shuffle_epi32(temp2 ,0xff);
temp3 = _mm_slli_si128(temp1, 0x04);
temp1 = _mm_xor_si128(temp1, temp3);
temp3 = _mm_slli_si128(temp3, 0x04);
temp1 = _mm_xor_si128(temp1, temp3);
temp3 = _mm_slli_si128(temp3, 0x04);
temp1 = _mm_xor_si128(temp1, temp3);
temp1 = _mm_xor_si128(temp1, temp2);
//MMCLEAR(temp3);
return temp1;
}
static
void AESNI_128_EKey_Expansion_r(const unsigned char *userkey,
unsigned char *ekey,
unsigned int rounds)
{
register __m128i temp1, temp2;
__m128i *Key_Schedule = (__m128i*)ekey;
temp1 = _mm_loadu_si128(( __m128i*)userkey);
Key_Schedule[0] = temp1;
temp2 = _mm_aeskeygenassist_si128(temp1, 0x01);
temp1 = KEY_128_ASSIST(temp1, temp2);
Key_Schedule[1] = temp1;
temp2 = _mm_aeskeygenassist_si128(temp1, 0x02);
temp1 = KEY_128_ASSIST(temp1, temp2);
Key_Schedule[2] = temp1;
temp2 = _mm_aeskeygenassist_si128(temp1, 0x04);
temp1 = KEY_128_ASSIST(temp1, temp2);
Key_Schedule[3] = temp1;
temp2 = _mm_aeskeygenassist_si128(temp1, 0x08);
temp1 = KEY_128_ASSIST(temp1, temp2);
Key_Schedule[4] = temp1;
temp2 = _mm_aeskeygenassist_si128(temp1, 0x10);
temp1 = KEY_128_ASSIST(temp1, temp2);
Key_Schedule[5] = temp1;
temp2 = _mm_aeskeygenassist_si128(temp1, 0x20);
temp1 = KEY_128_ASSIST(temp1, temp2);
Key_Schedule[6] = temp1;
temp2 = _mm_aeskeygenassist_si128(temp1, 0x40);
temp1 = KEY_128_ASSIST(temp1, temp2);
Key_Schedule[7] = temp1;
temp2 = _mm_aeskeygenassist_si128(temp1, 0x80);
temp1 = KEY_128_ASSIST(temp1, temp2);
Key_Schedule[8] = temp1;
temp2 = _mm_aeskeygenassist_si128(temp1, 0x1b);
temp1 = KEY_128_ASSIST(temp1, temp2);
Key_Schedule[9] = temp1;
temp2 = _mm_aeskeygenassist_si128(temp1, 0x36);
temp1 = KEY_128_ASSIST(temp1, temp2);
Key_Schedule[10] = temp1;
if (rounds > 10) {
temp2 = _mm_aeskeygenassist_si128(temp1, 0x6c);
temp1 = KEY_128_ASSIST(temp1, temp2);
Key_Schedule[11] = temp1;
temp2 = _mm_aeskeygenassist_si128(temp1, 0xd8);
temp1 = KEY_128_ASSIST(temp1, temp2);
Key_Schedule[12] = temp1;
}
MMCLEAR(temp2);
MMCLEAR(temp1);
}
static inline void KEY_192_ASSIST(__m128i *temp1, __m128i *temp2, __m128i *temp3)
{
register __m128i temp4;
*temp2 = _mm_shuffle_epi32(*temp2, 0x55);
temp4 = _mm_slli_si128(*temp1, 0x04);
*temp1 = _mm_xor_si128(*temp1, temp4);
temp4 = _mm_slli_si128(temp4, 0x04);
*temp1 = _mm_xor_si128(*temp1, temp4);
temp4 = _mm_slli_si128(temp4, 0x04);
*temp1 = _mm_xor_si128(*temp1, temp4);
*temp1 = _mm_xor_si128(*temp1, *temp2);
*temp2 = _mm_shuffle_epi32(*temp1, 0xff);
temp4 = _mm_slli_si128(*temp3, 0x04);
*temp3 = _mm_xor_si128(*temp3, temp4);
*temp3 = _mm_xor_si128(*temp3, *temp2);
MMCLEAR(temp4);
}
static
void AESNI_192_EKey_Expansion_r(const unsigned char *userkey,
unsigned char *key,
unsigned int rounds)
{
__m128i temp1, temp2, temp3;
__m128i *Key_Schedule = ( __m128i*)key;
temp1 = _mm_loadu_si128(( __m128i*)userkey);
temp3 = _mm_loadu_si128(( __m128i*)(userkey+16));
Key_Schedule[0] = temp1;
Key_Schedule[1] = temp3;
temp2 = _mm_aeskeygenassist_si128(temp3, 0x01);
KEY_192_ASSIST(&temp1, &temp2, &temp3);
Key_Schedule[1] = ( __m128i) _mm_shuffle_pd((__m128d)Key_Schedule[1],
(__m128d)temp1, 0);
Key_Schedule[2] = ( __m128i) _mm_shuffle_pd((__m128d)temp1,
(__m128d)temp3, 1);
temp2 = _mm_aeskeygenassist_si128(temp3, 0x02);
KEY_192_ASSIST(&temp1, &temp2, &temp3);
Key_Schedule[3] = temp1;
Key_Schedule[4] = temp3;
temp2 = _mm_aeskeygenassist_si128(temp3, 0x04);
KEY_192_ASSIST(&temp1, &temp2, &temp3);
Key_Schedule[4] = ( __m128i) _mm_shuffle_pd((__m128d)Key_Schedule[4],
(__m128d)temp1, 0);
Key_Schedule[5] = ( __m128i) _mm_shuffle_pd((__m128d)temp1,
(__m128d)temp3, 1);
temp2 = _mm_aeskeygenassist_si128(temp3, 0x08);
KEY_192_ASSIST(&temp1, &temp2, &temp3);
Key_Schedule[6] = temp1;
Key_Schedule[7] = temp3;
temp2 = _mm_aeskeygenassist_si128(temp3, 0x10);
KEY_192_ASSIST(&temp1, &temp2, &temp3);
Key_Schedule[7] = ( __m128i) _mm_shuffle_pd((__m128d)Key_Schedule[7],
(__m128d)temp1, 0);
Key_Schedule[8] = ( __m128i) _mm_shuffle_pd((__m128d)temp1,
(__m128d)temp3, 1);
temp2 = _mm_aeskeygenassist_si128(temp3, 0x20);
KEY_192_ASSIST(&temp1, &temp2, &temp3);
Key_Schedule[9] = temp1;
Key_Schedule[10] = temp3;
temp2 = _mm_aeskeygenassist_si128(temp3, 0x40);
KEY_192_ASSIST(&temp1, &temp2, &temp3);
Key_Schedule[10] = ( __m128i) _mm_shuffle_pd((__m128d)Key_Schedule[10],
(__m128d)temp1, 0);
Key_Schedule[11] = ( __m128i) _mm_shuffle_pd((__m128d)temp1,
(__m128d)temp3, 1);
temp2 = _mm_aeskeygenassist_si128(temp3, 0x80);
KEY_192_ASSIST(&temp1, &temp2, &temp3);
Key_Schedule[12] = temp1;
if (rounds > 12) {
Key_Schedule[13] = temp3;
temp2 = _mm_aeskeygenassist_si128(temp3, 0x1b);
KEY_192_ASSIST(&temp1, &temp2, &temp3);
Key_Schedule[13] = ( __m128i) _mm_shuffle_pd((__m128d)Key_Schedule[13],
(__m128d)temp1, 0);
Key_Schedule[14] = ( __m128i) _mm_shuffle_pd((__m128d)temp1,
(__m128d)temp3, 1);
temp2 = _mm_aeskeygenassist_si128(temp3, 0x36);
KEY_192_ASSIST(&temp1, &temp2, &temp3);
Key_Schedule[15] = temp1;
}
MMCLEAR(temp3);
MMCLEAR(temp2);
MMCLEAR(temp1);
}
static inline void KEY_256_ASSIST_1(__m128i *temp1, __m128i *temp2)
{
register __m128i temp4;
*temp2 = _mm_shuffle_epi32(*temp2, 0xff);
temp4 = _mm_slli_si128(*temp1, 0x04);
*temp1 = _mm_xor_si128(*temp1, temp4);
temp4 = _mm_slli_si128(temp4, 0x04);
*temp1 = _mm_xor_si128(*temp1, temp4);
temp4 = _mm_slli_si128(temp4, 0x04);
*temp1 = _mm_xor_si128(*temp1, temp4);
*temp1 = _mm_xor_si128(*temp1, *temp2);
}
static inline void KEY_256_ASSIST_2(__m128i *temp1, __m128i *temp3)
{
register __m128i temp2, temp4;
temp4 = _mm_aeskeygenassist_si128(*temp1, 0x00);
temp2 = _mm_shuffle_epi32(temp4, 0xaa);
temp4 = _mm_slli_si128(*temp3, 0x04);
*temp3 = _mm_xor_si128(*temp3, temp4);
temp4 = _mm_slli_si128(temp4, 0x04);
*temp3 = _mm_xor_si128(*temp3, temp4);
temp4 = _mm_slli_si128(temp4, 0x04);
*temp3 = _mm_xor_si128(*temp3, temp4);
*temp3 = _mm_xor_si128(*temp3, temp2);
}
static
void AESNI_256_EKey_Expansion_r(const unsigned char *userkey,
unsigned char *key,
unsigned int rounds)
{
__m128i temp1, temp2, temp3;
__m128i *Key_Schedule = (__m128i*)key;
temp1 = _mm_loadu_si128((__m128i*)userkey);
temp3 = _mm_loadu_si128((__m128i*)(userkey+16));
Key_Schedule[0] = temp1;
Key_Schedule[1] = temp3;
temp2 = _mm_aeskeygenassist_si128(temp3, 0x01);
KEY_256_ASSIST_1(&temp1, &temp2);
Key_Schedule[2] = temp1;
KEY_256_ASSIST_2(&temp1, &temp3);
Key_Schedule[3] = temp3;
temp2 = _mm_aeskeygenassist_si128(temp3, 0x02);
KEY_256_ASSIST_1(&temp1, &temp2);
Key_Schedule[4] = temp1;
KEY_256_ASSIST_2(&temp1, &temp3);
Key_Schedule[5] = temp3;
temp2 = _mm_aeskeygenassist_si128(temp3, 0x04);
KEY_256_ASSIST_1(&temp1, &temp2);
Key_Schedule[6] = temp1;
KEY_256_ASSIST_2(&temp1, &temp3);
Key_Schedule[7] = temp3;
temp2 = _mm_aeskeygenassist_si128(temp3, 0x08);
KEY_256_ASSIST_1(&temp1, &temp2);
Key_Schedule[8] = temp1;
KEY_256_ASSIST_2(&temp1, &temp3);
Key_Schedule[9] = temp3;
temp2 = _mm_aeskeygenassist_si128(temp3, 0x10);
KEY_256_ASSIST_1(&temp1, &temp2);
Key_Schedule[10] = temp1;
KEY_256_ASSIST_2(&temp1, &temp3);
Key_Schedule[11] = temp3;
temp2 = _mm_aeskeygenassist_si128(temp3, 0x20);
KEY_256_ASSIST_1(&temp1, &temp2);
Key_Schedule[12] = temp1;
KEY_256_ASSIST_2(&temp1, &temp3);
Key_Schedule[13] = temp3;
temp2 = _mm_aeskeygenassist_si128(temp3, 0x40);
KEY_256_ASSIST_1(&temp1, &temp2);
Key_Schedule[14] = temp1;
if (rounds > 14) {
KEY_256_ASSIST_2(&temp1, &temp3);
Key_Schedule[15] = temp3;
temp2 = _mm_aeskeygenassist_si128(temp3, 0x80);
KEY_256_ASSIST_1(&temp1, &temp2);
Key_Schedule[16] = temp1;
if (rounds > 16) {
KEY_256_ASSIST_2(&temp1, &temp3);
Key_Schedule[17] = temp3;
temp2 = _mm_aeskeygenassist_si128(temp3, 0x1b);
KEY_256_ASSIST_1(&temp1, &temp2);
Key_Schedule[18] = temp1;
}
}
MMCLEAR(temp3);
MMCLEAR(temp2);
MMCLEAR(temp1);
}
static
inline void AESNI_EKey_DKey(const unsigned char* ekey,
unsigned char* dkey,
int rounds)
{
const __m128i *EKeys = (const __m128i*)ekey;
__m128i *DKeys = (__m128i*)dkey;
int r;
DKeys[rounds] = EKeys[0];
for (r = 1; r < rounds; ++r)
DKeys[rounds-r] = _mm_aesimc_si128(EKeys[r]);
DKeys[0] = EKeys[rounds];
}
static
void AESNI_128_DKey_Expansion_r(const unsigned char *userkey,
unsigned char *dkey,
unsigned int rounds)
{
AESNI_128_EKey_Expansion_r(userkey, (unsigned char*)crypto->xkeys, rounds);
AESNI_EKey_DKey((unsigned char*)crypto->xkeys, dkey, rounds);
}
static
void AESNI_192_DKey_Expansion_r(const unsigned char *userkey,
unsigned char *dkey,
unsigned int rounds)
{
AESNI_192_EKey_Expansion_r(userkey, (unsigned char*)crypto->xkeys, rounds);
AESNI_EKey_DKey((unsigned char*)crypto->xkeys, dkey, rounds);
}
static
void AESNI_256_DKey_Expansion_r(const unsigned char *userkey,
unsigned char *dkey,
unsigned int rounds)
{
AESNI_256_EKey_Expansion_r(userkey, (unsigned char*)crypto->xkeys, rounds);
AESNI_EKey_DKey((unsigned char*)crypto->xkeys, dkey, rounds);
}
#ifdef DEBUG
#include <stdio.h>
void static _debug_print(const __m128i m, const char* msg)
{
union {
unsigned char a[16];
unsigned int b[4];
} val;
_mm_storeu_si128((__m128i*)&val, m);
int i;
printf("%s 0x", msg);
for (i = 15; i >= 0; --i)
printf("%02x ", val.a[i]);
printf(" %08x %08x %08x %08x ", val.b[3], val.b[2], val.b[1], val.b[0]);
for (i = 0; i < 16; ++i)
printf(" %02x", val.a[i]);
printf("\n");
}
#endif
typedef __m128i (crypt_blk_fn)(const __m128i in, const unsigned char *rkeys, unsigned int rounds);
typedef void (crypt_4blks_fn)(__m128i *i0, __m128i *i1, __m128i *i2, __m128i *i3,
const unsigned char *rkeys, unsigned int rounds);
typedef void (crypt_8blks_fn)(__m128i *i0, __m128i *i1, __m128i *i2, __m128i *i3,
__m128i *i4, __m128i *i5, __m128i *i6, __m128i *i7,
const unsigned char *rkeys, unsigned int rounds);
static inline
__m128i Encrypt_Block(const __m128i in, const unsigned char *ekey, unsigned int rounds)
{
uint r;
const __m128i *rkeys = (__m128i*)ekey;
//register __m128i rk = _mm_loadu_si128(rkeys++);
register __m128i tmp = _mm_xor_si128(in, *rkeys++);
//rk = _mm_loadu_si128(rkeys++);
for (r = 1; r < rounds; ++r) {
tmp = _mm_aesenc_si128(tmp, *rkeys++);
//rk = _mm_loadu_si128(rkeys++);
}
tmp = _mm_aesenclast_si128(tmp, *rkeys);
//MMCLEAR(rk);
return tmp;
}
static inline
__m128i Decrypt_Block(const __m128i in, const unsigned char *dkey, unsigned int rounds)
{
uint r;
const __m128i *rkeys = (__m128i*)dkey;
//register __m128i rk = _mm_loadu_si128(rkeys++);
register __m128i tmp = _mm_xor_si128(in, *rkeys++);
//rk = _mm_loadu_si128(rkeys++);
for (r = 1; r < rounds; ++r) {
tmp = _mm_aesdec_si128(tmp, *rkeys++);
//rk = _mm_loadu_si128(rkeys++);
}
tmp = _mm_aesdeclast_si128(tmp, *rkeys);
//MMCLEAR(rk);
return tmp;
}
#ifdef UNUSED
static inline
void Encrypt_4Blocks(__m128i *i0, __m128i *i1, __m128i *i2, __m128i *i3,
const unsigned char *ekey, unsigned int rounds)
{
uint r;
const __m128i *rkeys = (__m128i*)ekey;
register __m128i rk asm("xmm0") = _mm_loadu_si128(rkeys++);
*i0 = _mm_xor_si128(*i0, rk);
*i1 = _mm_xor_si128(*i1, rk);
*i2 = _mm_xor_si128(*i2, rk);
*i3 = _mm_xor_si128(*i3, rk);
rk = _mm_loadu_si128(rkeys++);
for (r = 1; r < rounds; ++r) {
*i0 = _mm_aesenc_si128(*i0, rk);
*i1 = _mm_aesenc_si128(*i1, rk);
*i2 = _mm_aesenc_si128(*i2, rk);
*i3 = _mm_aesenc_si128(*i3, rk);
rk = _mm_loadu_si128(rkeys++);
}
/* Last round ... */
*i0 = _mm_aesenclast_si128(*i0, rk);
*i1 = _mm_aesenclast_si128(*i1, rk);
*i2 = _mm_aesenclast_si128(*i2, rk);
*i3 = _mm_aesenclast_si128(*i3, rk);
MMCLEAR(rk);
//asm volatile("pxor %%xmm5, %%xmm5\n" :::"xmm5");
}
#endif
static inline
void Decrypt_4Blocks(__m128i *i0, __m128i *i1, __m128i *i2, __m128i *i3,
const unsigned char *dkey, unsigned int rounds)
{
uint r;
const __m128i *rkeys = (__m128i*)dkey;
register __m128i rk asm("xmm0") = _mm_loadu_si128(rkeys++);
*i0 = _mm_xor_si128(*i0, rk);
*i1 = _mm_xor_si128(*i1, rk);
*i2 = _mm_xor_si128(*i2, rk);
*i3 = _mm_xor_si128(*i3, rk);
rk = _mm_loadu_si128(rkeys++);
for (r = 1; r < rounds; ++r) {
*i0 = _mm_aesdec_si128(*i0, rk);
*i1 = _mm_aesdec_si128(*i1, rk);
*i2 = _mm_aesdec_si128(*i2, rk);
*i3 = _mm_aesdec_si128(*i3, rk);
rk = _mm_loadu_si128(rkeys++);
}
/* Last round ... */
*i0 = _mm_aesdeclast_si128(*i0, rk);
*i1 = _mm_aesdeclast_si128(*i1, rk);
*i2 = _mm_aesdeclast_si128(*i2, rk);
*i3 = _mm_aesdeclast_si128(*i3, rk);
MMCLEAR(rk);
}
static inline
void Encrypt_8Blocks(__m128i *i0, __m128i *i1, __m128i *i2, __m128i *i3,
__m128i *i4, __m128i *i5, __m128i *i6, __m128i *i7,
const unsigned char *ekey, unsigned int rounds)
{
int r;
const __m128i *rkeys = (__m128i*)ekey;
register __m128i rk /*asm("xmm0")*/ = _mm_loadu_si128(rkeys++);
*i0 = _mm_xor_si128(*i0, rk);
*i1 = _mm_xor_si128(*i1, rk);
*i2 = _mm_xor_si128(*i2, rk);
*i3 = _mm_xor_si128(*i3, rk);
*i4 = _mm_xor_si128(*i4, rk);
*i5 = _mm_xor_si128(*i5, rk);
*i6 = _mm_xor_si128(*i6, rk);
*i7 = _mm_xor_si128(*i7, rk);
rk = _mm_loadu_si128(rkeys++);
for (r = rounds-1; r > 0; --r) {
*i0 = _mm_aesenc_si128(*i0, rk);
*i1 = _mm_aesenc_si128(*i1, rk);
*i2 = _mm_aesenc_si128(*i2, rk);
*i3 = _mm_aesenc_si128(*i3, rk);
*i4 = _mm_aesenc_si128(*i4, rk);
*i5 = _mm_aesenc_si128(*i5, rk);
*i6 = _mm_aesenc_si128(*i6, rk);
*i7 = _mm_aesenc_si128(*i7, rk);
rk = _mm_loadu_si128(rkeys++);
}
/* Last round ... */
*i0 = _mm_aesenclast_si128(*i0, rk);
*i1 = _mm_aesenclast_si128(*i1, rk);
*i2 = _mm_aesenclast_si128(*i2, rk);
*i3 = _mm_aesenclast_si128(*i3, rk);
*i4 = _mm_aesenclast_si128(*i4, rk);
*i5 = _mm_aesenclast_si128(*i5, rk);
*i6 = _mm_aesenclast_si128(*i6, rk);
*i7 = _mm_aesenclast_si128(*i7, rk);
MMCLEAR(rk);
}
static inline
void Decrypt_8Blocks(__m128i *i0, __m128i *i1, __m128i *i2, __m128i *i3,
__m128i *i4, __m128i *i5, __m128i *i6, __m128i *i7,
const unsigned char *dkey, unsigned int rounds)
{
int r;
const __m128i *rkeys = (__m128i*)dkey;
register __m128i rk /*asm("xmm0")*/ = _mm_loadu_si128(rkeys++);
*i0 = _mm_xor_si128(*i0, rk);
*i1 = _mm_xor_si128(*i1, rk);
*i2 = _mm_xor_si128(*i2, rk);
*i3 = _mm_xor_si128(*i3, rk);
*i4 = _mm_xor_si128(*i4, rk);
*i5 = _mm_xor_si128(*i5, rk);
*i6 = _mm_xor_si128(*i6, rk);
*i7 = _mm_xor_si128(*i7, rk);
rk = _mm_loadu_si128(rkeys++);
for (r = rounds-1; r >0; --r) {
*i0 = _mm_aesdec_si128(*i0, rk);
*i1 = _mm_aesdec_si128(*i1, rk);
*i2 = _mm_aesdec_si128(*i2, rk);
*i3 = _mm_aesdec_si128(*i3, rk);
*i4 = _mm_aesdec_si128(*i4, rk);
*i5 = _mm_aesdec_si128(*i5, rk);
*i6 = _mm_aesdec_si128(*i6, rk);
*i7 = _mm_aesdec_si128(*i7, rk);
rk = _mm_loadu_si128(rkeys++);
}
/* Last round ... */
*i0 = _mm_aesdeclast_si128(*i0, rk);
*i1 = _mm_aesdeclast_si128(*i1, rk);
*i2 = _mm_aesdeclast_si128(*i2, rk);
*i3 = _mm_aesdeclast_si128(*i3, rk);
*i4 = _mm_aesdeclast_si128(*i4, rk);
*i5 = _mm_aesdeclast_si128(*i5, rk);
*i6 = _mm_aesdeclast_si128(*i6, rk);
*i7 = _mm_aesdeclast_si128(*i7, rk);
MMCLEAR(rk);
}
static inline
__m128i _mkmask(char ln)
{
ln &= 0x0f;
return (ln >= 8?
_mm_set_epi64x((1ULL<<(8*ln-64))-1, 0xffffffffffffffffULL):
_mm_set_epi64x(0ULL, (1ULL<<(8*ln))-1)
);
}
static inline
__m128i _mkpad(char ln)
{
uchar by = 16 - (ln & 0x0f);
return _mm_set_epi8(by, by, by, by, by, by, by, by,
by, by, by, by, by, by, by, by);
}
#if 0
static
void AESNI_ECB_Encrypt_old(const unsigned char* in, unsigned char* out,
ssize_t len, const unsigned char* key, unsigned int rounds)
{
while (len >= SIZE128) {
register __m128i blk = _mm_loadu_si128((const __m128i*)in);
blk = Encrypt_Block(blk, key, rounds);
_mm_storeu_si128((__m128i*)out, blk);
len -= SIZE128;
in += SIZE128;
out += SIZE128;
}
if (len) {
register __m128i blk = _mm_loadu_si128((const __m128i*)in);
__m128i mask = _mkmask(len);
__m128i pad = _mkpad(len);
__m128i imask = _mm_set_epi64x(0xffffffffffffffffULL, 0xffffffffffffffffULL);
imask = _mm_xor_si128(imask, mask);
pad = _mm_and_si128(pad, imask);
blk = _mm_and_si128(blk, mask);
blk = _mm_or_si128(blk, pad);
blk = Encrypt_Block(blk, key, rounds);
_mm_storeu_si128((__m128i*)out, blk);
}
}
static
void AESNI_ECB_Decrypt_old(const unsigned char* in, unsigned char* out,
ssize_t len, const unsigned char* key, unsigned int rounds)
{
while (len > 0) {
register __m128i blk = _mm_loadu_si128((const __m128i*)in);
blk = Decrypt_Block(blk, key, rounds);
_mm_storeu_si128((__m128i*)out, blk);
len -= SIZE128;
in += SIZE128;
out += SIZE128;
}
}
#endif
static inline
int AESNI_ECB_Crypt_Tmpl(crypt_8blks_fn *crypt8, crypt_blk_fn *crypt, int enc,
const unsigned char* key, unsigned int rounds,
unsigned int pad,
const unsigned char* in, unsigned char* out,
ssize_t len, ssize_t *olen)
{
*olen = len;
while (len >= 8*SIZE128) {
__m128i blk0 = _mm_loadu_si128((const __m128i*)in);
__m128i blk1 = _mm_loadu_si128((const __m128i*)(in+SIZE128));
__m128i blk2 = _mm_loadu_si128((const __m128i*)(in+2*SIZE128));
__m128i blk3 = _mm_loadu_si128((const __m128i*)(in+3*SIZE128));
__m128i blk4 = _mm_loadu_si128((const __m128i*)(in+4*SIZE128));
__m128i blk5 = _mm_loadu_si128((const __m128i*)(in+5*SIZE128));
__m128i blk6 = _mm_loadu_si128((const __m128i*)(in+6*SIZE128));
__m128i blk7 = _mm_loadu_si128((const __m128i*)(in+7*SIZE128));
crypt8(&blk0, &blk1, &blk2, &blk3, &blk4, &blk5, &blk6, &blk7, key, rounds);
_mm_storeu_si128((__m128i*)out, blk0);
_mm_storeu_si128((__m128i*)(out+SIZE128), blk1);
_mm_storeu_si128((__m128i*)(out+2*SIZE128), blk2);
_mm_storeu_si128((__m128i*)(out+3*SIZE128), blk3);
_mm_storeu_si128((__m128i*)(out+4*SIZE128), blk4);
_mm_storeu_si128((__m128i*)(out+5*SIZE128), blk5);
_mm_storeu_si128((__m128i*)(out+6*SIZE128), blk6);
_mm_storeu_si128((__m128i*)(out+7*SIZE128), blk7);
len -= 8*SIZE128;
in += 8*SIZE128;
out += 8*SIZE128;
}
while (len > 0 || (enc && len == 0 && pad == PAD_ALWAYS)) {
register __m128i blk;
if (len)
blk = _mm_loadu_si128((const __m128i*)in);
else
MMCLEAR(blk);
if (enc && len < SIZE128) {
__m128i mask = _mkmask(len);
blk = _mm_and_si128(blk, mask);
if (pad) {
__m128i padv = _mkpad(len);
__m128i imask = _mm_set_epi64x(0xffffffffffffffffULL, 0xffffffffffffffffULL);
imask = _mm_xor_si128(imask, mask);
padv = _mm_and_si128(padv, imask);
blk = _mm_or_si128(blk, padv);
#ifdef DEBUG
if (!len) {
_debug_print(mask, "mask");
_debug_print(imask, "imask");
_debug_print(padv, "padv");
}
#endif
}
*olen += 16-(len&15);
}
blk = crypt(blk, key, rounds);
_mm_storeu_si128((__m128i*)out, blk);
len -= SIZE128;
in += SIZE128;
out += SIZE128;
}
MMCLEARALL;
if (enc)
return (pad == PAD_ALWAYS || (len&15))? 16-(len&15): 0;
if (pad)
return dec_fix_olen_pad(olen, pad, out);
else
return 0;
}
#if defined(__AVX2__) && !defined(NO_AVX2) && defined(__VAES__)
#ifdef __GNUC__
#define ALWAYS_INLINE __attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
#else
#define ALWAYS_INLINE
#endif
/* Why is this intrinsic missing ??? */
static inline __m256i ALWAYS_INLINE
_mm256_broadcast_si128 (const __m128i *__X)
{
register __m256i ret;
asm volatile(" vbroadcasti128 (%1), %0"
: "=x"(ret)
: "r"(__X), "m"(*__X)
);
return ret;
//return (__m256i) __builtin_ia32_vbroadcasti128 (*(__v2di)__X);
}
typedef void (crypt_4x2blks_fn)(__m256i *i0, __m256i *i1, __m256i *i2, __m256i *i3,
const unsigned char *rkeys, unsigned int rounds);
static inline
void Encrypt_4x2Blocks(__m256i *i0, __m256i *i1, __m256i *i2, __m256i *i3,
const unsigned char *ekey, unsigned int rounds)
{
int r;
const __m128i *rkeys = (__m128i*)ekey;
register __m256i rk = _mm256_broadcast_si128(rkeys++);
*i0 = _mm256_xor_si256(*i0, rk);
*i1 = _mm256_xor_si256(*i1, rk);
*i2 = _mm256_xor_si256(*i2, rk);
*i3 = _mm256_xor_si256(*i3, rk);
rk = _mm256_broadcast_si128(rkeys++);
for (r = rounds-1; r > 0; --r) {
*i0 = _mm256_aesenc_epi128(*i0, rk);
*i1 = _mm256_aesenc_epi128(*i1, rk);
*i2 = _mm256_aesenc_epi128(*i2, rk);
*i3 = _mm256_aesenc_epi128(*i3, rk);
rk = _mm256_broadcast_si128(rkeys++);
}
/* Last round ... */
*i0 = _mm256_aesenclast_epi128(*i0, rk);
*i1 = _mm256_aesenclast_epi128(*i1, rk);
*i2 = _mm256_aesenclast_epi128(*i2, rk);
*i3 = _mm256_aesenclast_epi128(*i3, rk);
MM256CLEAR(rk);
}
static inline
void Decrypt_4x2Blocks(__m256i *i0, __m256i *i1, __m256i *i2, __m256i *i3,
const unsigned char *dkey, unsigned int rounds)
{
int r;
const __m128i *rkeys = (__m128i*)dkey;
register __m256i rk = _mm256_broadcast_si128(rkeys++);
*i0 = _mm256_xor_si256(*i0, rk);
*i1 = _mm256_xor_si256(*i1, rk);
*i2 = _mm256_xor_si256(*i2, rk);
*i3 = _mm256_xor_si256(*i3, rk);
rk = _mm256_broadcast_si128(rkeys++);
for (r = rounds-1; r > 0; --r) {
*i0 = _mm256_aesdec_epi128(*i0, rk);
*i1 = _mm256_aesdec_epi128(*i1, rk);
*i2 = _mm256_aesdec_epi128(*i2, rk);
*i3 = _mm256_aesdec_epi128(*i3, rk);
rk = _mm256_broadcast_si128(rkeys++);
}
/* Last round ... */
*i0 = _mm256_aesdeclast_epi128(*i0, rk);
*i1 = _mm256_aesdeclast_epi128(*i1, rk);
*i2 = _mm256_aesdeclast_epi128(*i2, rk);
*i3 = _mm256_aesdeclast_epi128(*i3, rk);
MM256CLEAR(rk);
}
static inline
int AESNI_ECB_Crypt_Tmpl2(crypt_4x2blks_fn *crypt4x2, crypt_blk_fn *crypt, int enc,
const unsigned char* key, unsigned int rounds,
unsigned int pad,
const unsigned char* in, unsigned char* out,
ssize_t len, ssize_t *olen)
{
*olen = len;
__m256i blk0, blk1, blk2, blk3;
while (len >= 4*SIZE256) {
blk0 = _mm256_loadu_si256((const __m256i*)in);
blk1 = _mm256_loadu_si256((const __m256i*)(in+SIZE256));
blk2 = _mm256_loadu_si256((const __m256i*)(in+2*SIZE256));
blk3 = _mm256_loadu_si256((const __m256i*)(in+3*SIZE256));
crypt4x2(&blk0, &blk1, &blk2, &blk3, key, rounds);
_mm256_storeu_si256((__m256i*)out, blk0);
_mm256_storeu_si256((__m256i*)(out+SIZE256), blk1);
_mm256_storeu_si256((__m256i*)(out+2*SIZE256), blk2);
_mm256_storeu_si256((__m256i*)(out+3*SIZE256), blk3);
len -= 4*SIZE256;
in += 4*SIZE256;
out += 4*SIZE256;
}
MM256CLEAR(blk0); MM256CLEAR(blk1); MM256CLEAR(blk2); MM256CLEAR(blk3);
__m128i blk;
while (len > 0 || (enc && len == 0 && pad == PAD_ALWAYS)) {
blk = _mm_loadu_si128((const __m128i*)in);
if (enc && len < SIZE128) {
__m128i mask = _mkmask(len);
blk = _mm_and_si128(blk, mask);
if (pad) {
__m128i padv = _mkpad(len);
__m128i imask = _mm_set_epi64x(0xffffffffffffffffULL, 0xffffffffffffffffULL);
imask = _mm_xor_si128(imask, mask);
padv = _mm_and_si128(padv, imask);
blk = _mm_or_si128(blk, padv);
#ifdef DEBUG
if (!len) {
_debug_print(mask, "mask");
_debug_print(imask, "imask");
_debug_print(padv, "padv");
}
#endif
}
*olen += 16-(len&15);
}
blk = crypt(blk, key, rounds);
_mm_storeu_si128((__m128i*)out, blk);
len -= SIZE128;
in += SIZE128;
out += SIZE128;
}
MMCLEAR(blk);
if (enc)
return (pad == PAD_ALWAYS || (len&15))? 16-(len&15): 0;
if (pad)
return dec_fix_olen_pad(olen, pad, out);
else
return 0;
}
static
int AESNI_ECB_Encrypt(const unsigned char* key, unsigned int rounds,
unsigned char *iv, unsigned int pad,
const unsigned char* in, unsigned char* out,
ssize_t len, ssize_t *olen)
{
if (have_vaes)
return AESNI_ECB_Crypt_Tmpl2(Encrypt_4x2Blocks, Encrypt_Block, 1,
key, rounds, pad, in, out, len, olen);
else
return AESNI_ECB_Crypt_Tmpl(Encrypt_8Blocks, Encrypt_Block, 1,
key, rounds, pad, in, out, len, olen);
}
static
int AESNI_ECB_Decrypt(const unsigned char* key, unsigned int rounds,
unsigned char *iv, unsigned int pad,
const unsigned char* in, unsigned char* out,
ssize_t len, ssize_t *olen)
{
if (have_vaes)
return AESNI_ECB_Crypt_Tmpl2(Decrypt_4x2Blocks, Decrypt_Block, 0,
key, rounds, pad, in, out, len, olen);
else
return AESNI_ECB_Crypt_Tmpl(Decrypt_8Blocks, Decrypt_Block, 0,
key, rounds, pad, in, out, len, olen);
}
#else
static
int AESNI_ECB_Encrypt(const unsigned char* key, unsigned int rounds,
unsigned char *iv, unsigned int pad,
const unsigned char* in, unsigned char* out,
ssize_t len, ssize_t *olen)
{
return AESNI_ECB_Crypt_Tmpl(Encrypt_8Blocks, Encrypt_Block, 1,
key, rounds, pad, in, out, len, olen);
}
static
int AESNI_ECB_Decrypt(const unsigned char* key, unsigned int rounds,
unsigned char *iv, unsigned int pad,
const unsigned char* in, unsigned char* out,
ssize_t len, ssize_t *olen)
{
return AESNI_ECB_Crypt_Tmpl(Decrypt_8Blocks, Decrypt_Block, 0,
key, rounds, pad, in, out, len, olen);
}
#endif
#if 0
static
void AESNI_ECB_Decrypt(const unsigned char* key, unsigned int rounds,
unsigned char *iv,
const unsigned char* in, unsigned char* out,
ssize_t len)
{
while (len >= 4*SIZE128) {
__m128i blk0 = _mm_loadu_si128((const __m128i*)in);
__m128i blk1 = _mm_loadu_si128((const __m128i*)(in+SIZE128));
__m128i blk2 = _mm_loadu_si128((const __m128i*)(in+2*SIZE128));
__m128i blk3 = _mm_loadu_si128((const __m128i*)(in+3*SIZE128));
Decrypt_4Blocks(&blk0, &blk1, &blk2, &blk3, key, rounds);
_mm_storeu_si128((__m128i*)out, blk0);
_mm_storeu_si128((__m128i*)(out+SIZE128), blk1);
_mm_storeu_si128((__m128i*)(out+2*SIZE128), blk2);
_mm_storeu_si128((__m128i*)(out+3*SIZE128), blk3);
len -= 4*SIZE128;
in += 4*SIZE128;
out += 4*SIZE128;
}
while (len > 0) {
register __m128i blk = _mm_loadu_si128((const __m128i*)in);
blk = Decrypt_Block(blk, key, rounds);
_mm_storeu_si128((__m128i*)out, blk);
len -= SIZE128;
in += SIZE128;
out += SIZE128;
}
}
#endif
static inline
int AESNI_CBC_Encrypt_Tmpl(crypt_blk_fn *encrypt,
const unsigned char* key, unsigned int rounds,
unsigned char* iv, unsigned int pad,
const unsigned char* in, unsigned char* out,
ssize_t len, ssize_t *olen)
{
register __m128i ivb = _mm_loadu_si128((const __m128i*)iv);
register __m128i dat;
*olen = len;
while (len >= SIZE128) {
//dat = _mm_loadu_si128((const __m128i*)in);
ivb = _mm_xor_si128(ivb, *(const __m128i*)in);
ivb = encrypt(ivb, key, rounds);
_mm_storeu_si128((__m128i*)out, ivb);
len -= SIZE128;
in += SIZE128;
out += SIZE128;
}
//_mm_storeu_si128((__m128i*)iv, ivb);
if (len || pad == PAD_ALWAYS) {
//dat = _mm_loadu_si128((const __m128i*)in);
__m128i mask = _mkmask(len);
dat = _mm_and_si128(*(const __m128i*)in, mask);
if (pad) {
__m128i padv = _mkpad(len);
__m128i imask = _mm_set_epi64x(0xffffffffffffffffULL, 0xffffffffffffffffULL);
imask = _mm_xor_si128(imask, mask);
padv = _mm_and_si128(padv, imask);
dat = _mm_or_si128(dat, padv);
}
ivb = _mm_xor_si128(ivb, dat);
ivb = encrypt(ivb, key, rounds);
_mm_storeu_si128((__m128i*)out, ivb);
*olen += 16-(*olen&15);
}
_mm_storeu_si128((__m128i*)iv, ivb);
/* FIXME: Clear the right registers */
MMCLEAR(dat); MMCLEAR(ivb);
return (pad == PAD_ALWAYS || (len&15))? 16-(len&15): 0;
}
static
int AESNI_CBC_Encrypt( const unsigned char* key, unsigned int rounds,
unsigned char* iv, unsigned int pad,
const unsigned char* in, unsigned char* out,
ssize_t len, ssize_t *olen)
{
return AESNI_CBC_Encrypt_Tmpl(Encrypt_Block, key, rounds, iv, pad, in, out, len, olen);
}
static inline
int AESNI_CBC_Decrypt_Tmpl(crypt_4blks_fn *decrypt4, crypt_blk_fn *decrypt,
const unsigned char* key, unsigned int rounds,
unsigned char* iv, unsigned int pad,
const unsigned char* in, unsigned char* out,
ssize_t len, ssize_t *olen)
{
register __m128i ivb = _mm_loadu_si128((const __m128i*)iv);
*olen = len;