summaryrefslogtreecommitdiffstats
path: root/crypto/ecc.c
blob: ad739255951f01da1bf42ce68904a070084f97e2 (plain)
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
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
/*
 * Copyright (c) 2013, Kenneth MacKay
 * All rights reserved.
 *
 * Redistribution and use in source and binary forms, with or without
 * modification, are permitted provided that the following conditions are
 * met:
 *  * Redistributions of source code must retain the above copyright
 *   notice, this list of conditions and the following disclaimer.
 *  * Redistributions in binary form must reproduce the above copyright
 *    notice, this list of conditions and the following disclaimer in the
 *    documentation and/or other materials provided with the distribution.
 *
 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
 * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
 * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
 * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
 * HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
 * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
 * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
 * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
 * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
 * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
 * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
 */

#include <linux/random.h>
#include <linux/slab.h>
#include <linux/swab.h>
#include <linux/fips.h>
#include <crypto/ecdh.h>
#include <crypto/rng.h>

#include "ecc.h"
#include "ecc_curve_defs.h"

typedef struct {
	u64 m_low;
	u64 m_high;
} uint128_t;

static inline const struct ecc_curve *ecc_get_curve(unsigned int curve_id)
{
	switch (curve_id) {
	/* In FIPS mode only allow P256 and higher */
	case ECC_CURVE_NIST_P192:
		return fips_enabled ? NULL : &nist_p192;
	case ECC_CURVE_NIST_P256:
		return &nist_p256;
	default:
		return NULL;
	}
}

static u64 *ecc_alloc_digits_space(unsigned int ndigits)
{
	size_t len = ndigits * sizeof(u64);

	if (!len)
		return NULL;

	return kmalloc(len, GFP_KERNEL);
}

static void ecc_free_digits_space(u64 *space)
{
	kzfree(space);
}

static struct ecc_point *ecc_alloc_point(unsigned int ndigits)
{
	struct ecc_point *p = kmalloc(sizeof(*p), GFP_KERNEL);

	if (!p)
		return NULL;

	p->x = ecc_alloc_digits_space(ndigits);
	if (!p->x)
		goto err_alloc_x;

	p->y = ecc_alloc_digits_space(ndigits);
	if (!p->y)
		goto err_alloc_y;

	p->ndigits = ndigits;

	return p;

err_alloc_y:
	ecc_free_digits_space(p->x);
err_alloc_x:
	kfree(p);
	return NULL;
}

static void ecc_free_point(struct ecc_point *p)
{
	if (!p)
		return;

	kzfree(p->x);
	kzfree(p->y);
	kzfree(p);
}

static void vli_clear(u64 *vli, unsigned int ndigits)
{
	int i;

	for (i = 0; i < ndigits; i++)
		vli[i] = 0;
}

/* Returns true if vli == 0, false otherwise. */
static bool vli_is_zero(const u64 *vli, unsigned int ndigits)
{
	int i;

	for (i = 0; i < ndigits; i++) {
		if (vli[i])
			return false;
	}

	return true;
}

/* Returns nonzero if bit bit of vli is set. */
static u64 vli_test_bit(const u64 *vli, unsigned int bit)
{
	return (vli[bit / 64] & ((u64)1 << (bit % 64)));
}

/* Counts the number of 64-bit "digits" in vli. */
static unsigned int vli_num_digits(const u64 *vli, unsigned int ndigits)
{
	int i;

	/* Search from the end until we find a non-zero digit.
	 * We do it in reverse because we expect that most digits will
	 * be nonzero.
	 */
	for (i = ndigits - 1; i >= 0 && vli[i] == 0; i--);

	return (i + 1);
}

/* Counts the number of bits required for vli. */
static unsigned int vli_num_bits(const u64 *vli, unsigned int ndigits)
{
	unsigned int i, num_digits;
	u64 digit;

	num_digits = vli_num_digits(vli, ndigits);
	if (num_digits == 0)
		return 0;

	digit = vli[num_digits - 1];
	for (i = 0; digit; i++)
		digit >>= 1;

	return ((num_digits - 1) * 64 + i);
}

/* Sets dest = src. */
static void vli_set(u64 *dest, const u64 *src, unsigned int ndigits)
{
	int i;

	for (i = 0; i < ndigits; i++)
		dest[i] = src[i];
}

/* Returns sign of left - right. */
static int vli_cmp(const u64 *left, const u64 *right, unsigned int ndigits)
{
	int i;

	for (i = ndigits - 1; i >= 0; i--) {
		if (left[i] > right[i])
			return 1;
		else if (left[i] < right[i])
			return -1;
	}

	return 0;
}

/* Computes result = in << c, returning carry. Can modify in place
 * (if result == in). 0 < shift < 64.
 */
static u64 vli_lshift(u64 *result, const u64 *in, unsigned int shift,
		      unsigned int ndigits)
{
	u64 carry = 0;
	int i;

	for (i = 0; i < ndigits; i++) {
		u64 temp = in[i];

		result[i] = (temp << shift) | carry;
		carry = temp >> (64 - shift);
	}

	return carry;
}

/* Computes vli = vli >> 1. */
static void vli_rshift1(u64 *vli, unsigned int ndigits)
{
	u64 *end = vli;
	u64 carry = 0;

	vli += ndigits;

	while (vli-- > end) {
		u64 temp = *vli;
		*vli = (temp >> 1) | carry;
		carry = temp << 63;
	}
}

/* Computes result = left + right, returning carry. Can modify in place. */
static u64 vli_add(u64 *result, const u64 *left, const u64 *right,
		   unsigned int ndigits)
{
	u64 carry = 0;
	int i;

	for (i = 0; i < ndigits; i++) {
		u64 sum;

		sum = left[i] + right[i] + carry;
		if (sum != left[i])
			carry = (sum < left[i]);

		result[i] = sum;
	}

	return carry;
}

/* Computes result = left - right, returning borrow. Can modify in place. */
static u64 vli_sub(u64 *result, const u64 *left, const u64 *right,
		   unsigned int ndigits)
{
	u64 borrow = 0;
	int i;

	for (i = 0; i < ndigits; i++) {
		u64 diff;

		diff = left[i] - right[i] - borrow;
		if (diff != left[i])
			borrow = (diff > left[i]);

		result[i] = diff;
	}

	return borrow;
}

static uint128_t mul_64_64(u64 left, u64 right)
{
	u64 a0 = left & 0xffffffffull;
	u64 a1 = left >> 32;
	u64 b0 = right & 0xffffffffull;
	u64 b1 = right >> 32;
	u64 m0 = a0 * b0;
	u64 m1 = a0 * b1;
	u64 m2 = a1 * b0;
	u64 m3 = a1 * b1;
	uint128_t result;

	m2 += (m0 >> 32);
	m2 += m1;

	/* Overflow */
	if (m2 < m1)
		m3 += 0x100000000ull;

	result.m_low = (m0 & 0xffffffffull) | (m2 << 32);
	result.m_high = m3 + (m2 >> 32);

	return result;
}

static uint128_t add_128_128(uint128_t a, uint128_t b)
{
	uint128_t result;

	result.m_low = a.m_low + b.m_low;
	result.m_high = a.m_high + b.m_high + (result.m_low < a.m_low);

	return result;
}

static void vli_mult(u64 *result, const u64 *left, const u64 *right,
		     unsigned int ndigits)
{
	uint128_t r01 = { 0, 0 };
	u64 r2 = 0;
	unsigned int i, k;

	/* Compute each digit of result in sequence, maintaining the
	 * carries.
	 */
	for (k = 0; k < ndigits * 2 - 1; k++) {
		unsigned int min;

		if (k < ndigits)
			min = 0;
		else
			min = (k + 1) - ndigits;

		for (i = min; i <= k && i < ndigits; i++) {
			uint128_t product;

			product = mul_64_64(left[i], right[k - i]);

			r01 = add_128_128(r01, product);
			r2 += (r01.m_high < product.m_high);
		}

		result[k] = r01.m_low;
		r01.m_low = r01.m_high;
		r01.m_high = r2;
		r2 = 0;
	}

	result[ndigits * 2 - 1] = r01.m_low;
}

static void vli_square(u64 *result, const u64 *left, unsigned int ndigits)
{
	uint128_t r01 = { 0, 0 };
	u64 r2 = 0;
	int i, k;

	for (k = 0; k < ndigits * 2 - 1; k++) {
		unsigned int min;

		if (k < ndigits)
			min = 0;
		else
			min = (k + 1) - ndigits;

		for (i = min; i <= k && i <= k - i; i++) {
			uint128_t product;

			product = mul_64_64(left[i], left[k - i]);

			if (i < k - i) {
				r2 += product.m_high >> 63;
				product.m_high = (product.m_high << 1) |
						 (product.m_low >> 63);
				product.m_low <<= 1;
			}

			r01 = add_128_128(r01, product);
			r2 += (r01.m_high < product.m_high);
		}

		result[k] = r01.m_low;
		r01.m_low = r01.m_high;
		r01.m_high = r2;
		r2 = 0;
	}

	result[ndigits * 2 - 1] = r01.m_low;
}

/* Computes result = (left + right) % mod.
 * Assumes that left < mod and right < mod, result != mod.
 */
static void vli_mod_add(u64 *result, const u64 *left, const u64 *right,
			const u64 *mod, unsigned int ndigits)
{
	u64 carry;

	carry = vli_add(result, left, right, ndigits);

	/* result > mod (result = mod + remainder), so subtract mod to
	 * get remainder.
	 */
	if (carry || vli_cmp(result, mod, ndigits) >= 0)
		vli_sub(result, result, mod, ndigits);
}

/* Computes result = (left - right) % mod.
 * Assumes that left < mod and right < mod, result != mod.
 */
static void vli_mod_sub(u64 *result, const u64 *left, const u64 *right,
			const u64 *mod, unsigned int ndigits)
{
	u64 borrow = vli_sub(result, left, right, ndigits);

	/* In this case, p_result == -diff == (max int) - diff.
	 * Since -x % d == d - x, we can get the correct result from
	 * result + mod (with overflow).
	 */
	if (borrow)
		vli_add(result, result, mod, ndigits);
}

/* Computes p_result = p_product % curve_p.
 * See algorithm 5 and 6 from
 * http://www.isys.uni-klu.ac.at/PDF/2001-0126-MT.pdf
 */
static void vli_mmod_fast_192(u64 *result, const u64 *product,
			      const u64 *curve_prime, u64 *tmp)
{
	const unsigned int ndigits = 3;
	int carry;

	vli_set(result, product, ndigits);

	vli_set(tmp, &product[3], ndigits);
	carry = vli_add(result, result, tmp, ndigits);

	tmp[0] = 0;
	tmp[1] = product[3];
	tmp[2] = product[4];
	carry += vli_add(result, result, tmp, ndigits);

	tmp[0] = tmp[1] = product[5];
	tmp[2] = 0;
	carry += vli_add(result, result, tmp, ndigits);

	while (carry || vli_cmp(curve_prime, result, ndigits) != 1)
		carry -= vli_sub(result, result, curve_prime, ndigits);
}

/* Computes result = product % curve_prime
 * from http://www.nsa.gov/ia/_files/nist-routines.pdf
 */
static void vli_mmod_fast_256(u64 *result, const u64 *product,
			      const u64 *curve_prime, u64 *tmp)
{
	int carry;
	const unsigned int ndigits = 4;

	/* t */
	vli_set(result, product, ndigits);

	/* s1 */
	tmp[0] = 0;
	tmp[1] = product[5] & 0xffffffff00000000ull;
	tmp[2] = product[6];
	tmp[3] = product[7];
	carry = vli_lshift(tmp, tmp, 1, ndigits);
	carry += vli_add(result, result, tmp, ndigits);

	/* s2 */
	tmp[1] = product[6] << 32;
	tmp[2] = (product[6] >> 32) | (product[7] << 32);
	tmp[3] = product[7] >> 32;
	carry += vli_lshift(tmp, tmp, 1, ndigits);
	carry += vli_add(result, result, tmp, ndigits);

	/* s3 */
	tmp[0] = product[4];
	tmp[1] = product[5] & 0xffffffff;
	tmp[2] = 0;
	tmp[3] = product[7];
	carry += vli_add(result, result, tmp, ndigits);

	/* s4 */
	tmp[0] = (product[4] >> 32) | (product[5] << 32);
	tmp[1] = (product[5] >> 32) | (product[6] & 0xffffffff00000000ull);
	tmp[2] = product[7];
	tmp[3] = (product[6] >> 32) | (product[4] << 32);
	carry += vli_add(result, result, tmp, ndigits);

	/* d1 */
	tmp[0] = (product[5] >> 32) | (product[6] << 32);
	tmp[1] = (product[6] >> 32);
	tmp[2] = 0;
	tmp[3] = (product[4] & 0xffffffff) | (product[5] << 32);
	carry -= vli_sub(result, result, tmp, ndigits);

	/* d2 */
	tmp[0] = product[6];
	tmp[1] = product[7];
	tmp[2] = 0;
	tmp[3] = (product[4] >> 32) | (product[5] & 0xffffffff00000000ull);
	carry -= vli_sub(result, result, tmp, ndigits);

	/* d3 */
	tmp[0] = (product[6] >> 32) | (product[7] << 32);
	tmp[1] = (product[7] >> 32) | (product[4] << 32);
	tmp[2] = (product[4] >> 32) | (product[5] << 32);
	tmp[3] = (product[6] << 32);
	carry -= vli_sub(result, result, tmp, ndigits);

	/* d4 */
	tmp[0] = product[7];
	tmp[1] = product[4] & 0xffffffff00000000ull;
	tmp[2] = product[5];
	tmp[3] = product[6] & 0xffffffff00000000ull;
	carry -= vli_sub(result, result, tmp, ndigits);

	if (carry < 0) {
		do {
			carry += vli_add(result, result, curve_prime, ndigits);
		} while (carry < 0);
	} else {
		while (carry || vli_cmp(curve_prime, result, ndigits) != 1)
			carry -= vli_sub(result, result, curve_prime, ndigits);
	}
}

/* Computes result = product % curve_prime
 *  from http://www.nsa.gov/ia/_files/nist-routines.pdf
*/
static bool vli_mmod_fast(u64 *result, u64 *product,
			  const u64 *curve_prime, unsigned int ndigits)
{
	u64 tmp[2 * ECC_MAX_DIGITS];

	switch (ndigits) {
	case 3:
		vli_mmod_fast_192(result, product, curve_prime, tmp);
		break;
	case 4:
		vli_mmod_fast_256(result, product, curve_prime, tmp);
		break;
	default:
		pr_err("unsupports digits size!\n");
		return false;
	}

	return true;
}

/* Computes result = (left * right) % curve_prime. */
static void vli_mod_mult_fast(u64 *result, const u64 *left, const u64 *right,
			      const u64 *curve_prime, unsigned int ndigits)
{
	u64 product[2 * ECC_MAX_DIGITS];

	vli_mult(product, left, right, ndigits);
	vli_mmod_fast(result, product, curve_prime, ndigits);
}

/* Computes result = left^2 % curve_prime. */
static void vli_mod_square_fast(u64 *result, const u64 *left,
				const u64 *curve_prime, unsigned int ndigits)
{
	u64 product[2 * ECC_MAX_DIGITS];

	vli_square(product, left, ndigits);
	vli_mmod_fast(result, product, curve_prime, ndigits);
}

#define EVEN(vli) (!(vli[0] & 1))
/* Computes result = (1 / p_input) % mod. All VLIs are the same size.
 * See "From Euclid's GCD to Montgomery Multiplication to the Great Divide"
 * https://labs.oracle.com/techrep/2001/smli_tr-2001-95.pdf
 */
static void vli_mod_inv(u64 *result, const u64 *input, const u64 *mod,
			unsigned int ndigits)
{
	u64 a[ECC_MAX_DIGITS], b[ECC_MAX_DIGITS];
	u64 u[ECC_MAX_DIGITS], v[ECC_MAX_DIGITS];
	u64 carry;
	int cmp_result;

	if (vli_is_zero(input, ndigits)) {
		vli_clear(result, ndigits);
		return;
	}

	vli_set(a, input, ndigits);
	vli_set(b, mod, ndigits);
	vli_clear(u, ndigits);
	u[0] = 1;
	vli_clear(v, ndigits);

	while ((cmp_result = vli_cmp(a, b, ndigits)) != 0) {
		carry = 0;

		if (EVEN(a)) {
			vli_rshift1(a, ndigits);

			if (!EVEN(u))
				carry = vli_add(u, u, mod, ndigits);

			vli_rshift1(u, ndigits);
			if (carry)
				u[ndigits - 1] |= 0x8000000000000000ull;
		} else if (EVEN(b)) {
			vli_rshift1(b, ndigits);

			if (!EVEN(v))
				carry = vli_add(v, v, mod, ndigits);

			vli_rshift1(v, ndigits);
			if (carry)
				v[ndigits - 1] |= 0x8000000000000000ull;
		} else if (cmp_result > 0) {
			vli_sub(a, a, b, ndigits);
			vli_rshift1(a, ndigits);

			if (vli_cmp(u, v, ndigits) < 0)
				vli_add(u, u, mod, ndigits);

			vli_sub(u, u, v, ndigits);
			if (!EVEN(u))
				carry = vli_add(u, u, mod, ndigits);

			vli_rshift1(u, ndigits);
			if (carry)
				u[ndigits - 1] |= 0x8000000000000000ull;
		} else {
			vli_sub(b, b, a, ndigits);
			vli_rshift1(b, ndigits);

			if (vli_cmp(v, u, ndigits) < 0)
				vli_add(v, v, mod, ndigits);

			vli_sub(v, v, u, ndigits);
			if (!EVEN(v))
				carry = vli_add(v, v, mod, ndigits);

			vli_rshift1(v, ndigits);
			if (carry)
				v[ndigits - 1] |= 0x8000000000000000ull;
		}
	}

	vli_set(result, u, ndigits);
}

/* ------ Point operations ------ */

/* Returns true if p_point is the point at infinity, false otherwise. */
static bool ecc_point_is_zero(const struct ecc_point *point)
{
	return (vli_is_zero(point->x, point->ndigits) &&
		vli_is_zero(point->y, point->ndigits));
}

/* Point multiplication algorithm using Montgomery's ladder with co-Z
 * coordinates. From http://eprint.iacr.org/2011/338.pdf
 */

/* Double in place */
static void ecc_point_double_jacobian(u64 *x1, u64 *y1, u64 *z1,
				      u64 *curve_prime, unsigned int ndigits)
{
	/* t1 = x, t2 = y, t3 = z */
	u64 t4[ECC_MAX_DIGITS];
	u64 t5[ECC_MAX_DIGITS];

	if (vli_is_zero(z1, ndigits))
		return;

	/* t4 = y1^2 */
	vli_mod_square_fast(t4, y1, curve_prime, ndigits);
	/* t5 = x1*y1^2 = A */
	vli_mod_mult_fast(t5, x1, t4, curve_prime, ndigits);
	/* t4 = y1^4 */
	vli_mod_square_fast(t4, t4, curve_prime, ndigits);
	/* t2 = y1*z1 = z3 */
	vli_mod_mult_fast(y1, y1, z1, curve_prime, ndigits);
	/* t3 = z1^2 */
	vli_mod_square_fast(z1, z1, curve_prime, ndigits);

	/* t1 = x1 + z1^2 */
	vli_mod_add(x1, x1, z1, curve_prime, ndigits);
	/* t3 = 2*z1^2 */
	vli_mod_add(z1, z1, z1, curve_prime, ndigits);
	/* t3 = x1 - z1^2 */
	vli_mod_sub(z1, x1, z1, curve_prime, ndigits);
	/* t1 = x1^2 - z1^4 */
	vli_mod_mult_fast(x1, x1, z1, curve_prime, ndigits);

	/* t3 = 2*(x1^2 - z1^4) */
	vli_mod_add(z1, x1, x1, curve_prime, ndigits);
	/* t1 = 3*(x1^2 - z1^4) */
	vli_mod_add(x1, x1, z1, curve_prime, ndigits);
	if (vli_test_bit(x1, 0)) {
		u64 carry = vli_add(x1, x1, curve_prime, ndigits);

		vli_rshift1(x1, ndigits);
		x1[ndigits - 1] |= carry << 63;
	} else {
		vli_rshift1(x1, ndigits);
	}
	/* t1 = 3/2*(x1^2 - z1^4) = B */

	/* t3 = B^2 */
	vli_mod_square_fast(z1, x1, curve_prime, ndigits);
	/* t3 = B^2 - A */
	vli_mod_sub(z1, z1, t5, curve_prime, ndigits);
	/* t3 = B^2 - 2A = x3 */
	vli_mod_sub(z1, z1, t5, curve_prime, ndigits);
	/* t5 = A - x3 */
	vli_mod_sub(t5, t5, z1, curve_prime, ndigits);
	/* t1 = B * (A - x3) */
	vli_mod_mult_fast(x1, x1, t5, curve_prime, ndigits);
	/* t4 = B * (A - x3) - y1^4 = y3 */
	vli_mod_sub(t4, x1, t4, curve_prime, ndigits);

	vli_set(x1, z1, ndigits);
	vli_set(z1, y1, ndigits);
	vli_set(y1, t4, ndigits);
}

/* Modify (x1, y1) => (x1 * z^2, y1 * z^3) */
static void apply_z(u64 *x1, u64 *y1, u64 *z, u64 *curve_prime,
		    unsigned int ndigits)
{
	u64 t1[ECC_MAX_DIGITS];

	vli_mod_square_fast(t1, z, curve_prime, ndigits);    /* z^2 */
	vli_mod_mult_fast(x1, x1, t1, curve_prime, ndigits); /* x1 * z^2 */
	vli_mod_mult_fast(t1, t1, z, curve_prime, ndigits);  /* z^3 */
	vli_mod_mult_fast(y1, y1, t1, curve_prime, ndigits); /* y1 * z^3 */
}

/* P = (x1, y1) => 2P, (x2, y2) => P' */
static void xycz_initial_double(u64 *x1, u64 *y1, u64 *x2, u64 *y2,
				u64 *p_initial_z, u64 *curve_prime,
				unsigned int ndigits)
{
	u64 z[ECC_MAX_DIGITS];

	vli_set(x2, x1, ndigits);
	vli_set(y2, y1, ndigits);

	vli_clear(z, ndigits);
	z[0] = 1;

	if (p_initial_z)
		vli_set(z, p_initial_z, ndigits);

	apply_z(x1, y1, z, curve_prime, ndigits);

	ecc_point_double_jacobian(x1, y1, z, curve_prime, ndigits);

	apply_z(x2, y2, z, curve_prime, ndigits);
}

/* Input P = (x1, y1, Z), Q = (x2, y2, Z)
 * Output P' = (x1', y1', Z3), P + Q = (x3, y3, Z3)
 * or P => P', Q => P + Q
 */
static void xycz_add(u64 *x1, u64 *y1, u64 *x2, u64 *y2, u64 *curve_prime,
		     unsigned int ndigits)
{
	/* t1 = X1, t2 = Y1, t3 = X2, t4 = Y2 */
	u64 t5[ECC_MAX_DIGITS];

	/* t5 = x2 - x1 */
	vli_mod_sub(t5, x2, x1, curve_prime, ndigits);
	/* t5 = (x2 - x1)^2 = A */
	vli_mod_square_fast(t5, t5, curve_prime, ndigits);
	/* t1 = x1*A = B */
	vli_mod_mult_fast(x1, x1, t5, curve_prime, ndigits);
	/* t3 = x2*A = C */
	vli_mod_mult_fast(x2, x2, t5, curve_prime, ndigits);
	/* t4 = y2 - y1 */
	vli_mod_sub(y2, y2, y1, curve_prime, ndigits);
	/* t5 = (y2 - y1)^2 = D */
	vli_mod_square_fast(t5, y2, curve_prime, ndigits);

	/* t5 = D - B */
	vli_mod_sub(t5, t5, x1, curve_prime, ndigits);
	/* t5 = D - B - C = x3 */
	vli_mod_sub(t5, t5, x2, curve_prime, ndigits);
	/* t3 = C - B */
	vli_mod_sub(x2, x2, x1, curve_prime, ndigits);
	/* t2 = y1*(C - B) */
	vli_mod_mult_fast(y1, y1, x2, curve_prime, ndigits);
	/* t3 = B - x3 */
	vli_mod_sub(x2, x1, t5, curve_prime, ndigits);
	/* t4 = (y2 - y1)*(B - x3) */
	vli_mod_mult_fast(y2, y2, x2, curve_prime, ndigits);
	/* t4 = y3 */
	vli_mod_sub(y2, y2, y1, curve_prime, ndigits);

	vli_set(x2, t5, ndigits);
}

/* Input P = (x1, y1, Z), Q = (x2, y2, Z)
 * Output P + Q = (x3, y3, Z3), P - Q = (x3', y3', Z3)
 * or P => P - Q, Q => P + Q
 */
static void xycz_add_c(u64 *x1, u64 *y1, u64 *x2, u64 *y2, u64 *curve_prime,
		       unsigned int ndigits)
{
	/* t1 = X1, t2 = Y1, t3 = X2, t4 = Y2 */
	u64 t5[ECC_MAX_DIGITS];
	u64 t6[ECC_MAX_DIGITS];
	u64 t7[ECC_MAX_DIGITS];

	/* t5 = x2 - x1 */
	vli_mod_sub(t5, x2, x1, curve_prime, ndigits);
	/* t5 = (x2 - x1)^2 = A */
	vli_mod_square_fast(t5, t5, curve_prime, ndigits);
	/* t1 = x1*A = B */
	vli_mod_mult_fast(x1, x1, t5, curve_prime, ndigits);
	/* t3 = x2*A = C */
	vli_mod_mult_fast(x2, x2, t5, curve_prime, ndigits);
	/* t4 = y2 + y1 */
	vli_mod_add(t5, y2, y1, curve_prime, ndigits);
	/* t4 = y2 - y1 */
	vli_mod_sub(y2, y2, y1, curve_prime, ndigits);

	/* t6 = C - B */
	vli_mod_sub(t6, x2, x1, curve_prime, ndigits);
	/* t2 = y1 * (C - B) */
	vli_mod_mult_fast(y1, y1, t6, curve_prime, ndigits);
	/* t6 = B + C */
	vli_mod_add(t6, x1, x2, curve_prime, ndigits);
	/* t3 = (y2 - y1)^2 */
	vli_mod_square_fast(x2, y2, curve_prime, ndigits);
	/* t3 = x3 */
	vli_mod_sub(x2, x2, t6, curve_prime, ndigits);

	/* t7 = B - x3 */
	vli_mod_sub(t7, x1, x2, curve_prime, ndigits);
	/* t4 = (y2 - y1)*(B - x3) */
	vli_mod_mult_fast(y2, y2, t7, curve_prime, ndigits);
	/* t4 = y3 */
	vli_mod_sub(y2, y2, y1, curve_prime, ndigits);

	/* t7 = (y2 + y1)^2 = F */
	vli_mod_square_fast(t7, t5, curve_prime, ndigits);
	/* t7 = x3' */
	vli_mod_sub(t7, t7, t6, curve_prime, ndigits);
	/* t6 = x3' - B */
	vli_mod_sub(t6, t7, x1, curve_prime, ndigits);
	/* t6 = (y2 + y1)*(x3' - B) */
	vli_mod_mult_fast(t6, t6, t5, curve_prime, ndigits);
	/* t2 = y3' */
	vli_mod_sub(y1, t6, y1, curve_prime, ndigits);

	vli_set(x1, t7, ndigits);
}

static void ecc_point_mult(struct ecc_point *result,
			   const struct ecc_point *point, const u64 *scalar,
			   u64 *initial_z, const struct ecc_curve *curve,
			   unsigned int ndigits)
{
	/* R0 and R1 */
	u64 rx[2][ECC_MAX_DIGITS];
	u64 ry[2][ECC_MAX_DIGITS];
	u64 z[ECC_MAX_DIGITS];
	u64 sk[2][ECC_MAX_DIGITS];
	u64 *curve_prime = curve->p;
	int i, nb;
	int num_bits;
	int carry;

	carry = vli_add(sk[0], scalar, curve->n, ndigits);
	vli_add(sk[1], sk[0], curve->n, ndigits);
	scalar = sk[!carry];
	num_bits = sizeof(u64) * ndigits * 8 + 1;

	vli_set(rx[1], point->x, ndigits);
	vli_set(ry[1], point->y, ndigits);

	xycz_initial_double(rx[1], ry[1], rx[0], ry[0], initial_z, curve_prime,
			    ndigits);

	for (i = num_bits - 2; i > 0; i--) {
		nb = !vli_test_bit(scalar, i);
		xycz_add_c(rx[1 - nb], ry[1 - nb], rx[nb], ry[nb], curve_prime,
			   ndigits);
		xycz_add(rx[nb], ry[nb], rx[1 - nb], ry[1 - nb], curve_prime,
			 ndigits);
	}

	nb = !vli_test_bit(scalar, 0);
	xycz_add_c(rx[1 - nb], ry[1 - nb], rx[nb], ry[nb], curve_prime,
		   ndigits);

	/* Find final 1/Z value. */
	/* X1 - X0 */
	vli_mod_sub(z, rx[1], rx[0], curve_prime, ndigits);
	/* Yb * (X1 - X0) */
	vli_mod_mult_fast(z, z, ry[1 - nb], curve_prime, ndigits);
	/* xP * Yb * (X1 - X0) */
	vli_mod_mult_fast(z, z, point->x, curve_prime, ndigits);

	/* 1 / (xP * Yb * (X1 - X0)) */
	vli_mod_inv(z, z, curve_prime, point->ndigits);

	/* yP / (xP * Yb * (X1 - X0)) */
	vli_mod_mult_fast(z, z, point->y, curve_prime, ndigits);
	/* Xb * yP / (xP * Yb * (X1 - X0)) */
	vli_mod_mult_fast(z, z, rx[1 - nb], curve_prime, ndigits);
	/* End 1/Z calculation */

	xycz_add(rx[nb], ry[nb], rx[1 - nb], ry[1 - nb], curve_prime, ndigits);

	apply_z(rx[0], ry[0], z, curve_prime, ndigits);

	vli_set(result->x, rx[0], ndigits);
	vli_set(result->y, ry[0], ndigits);
}

static inline void ecc_swap_digits(const u64 *in, u64 *out,
				   unsigned int ndigits)
{
	const __be64 *src = (__force __be64 *)in;
	int i;

	for (i = 0; i < ndigits; i++)
		out[i] = be64_to_cpu(src[ndigits - 1 - i]);
}

static int __ecc_is_key_valid(const struct ecc_curve *curve,
			      const u64 *private_key, unsigned int ndigits)
{
	u64 one[ECC_MAX_DIGITS] = { 1, };
	u64 res[ECC_MAX_DIGITS];

	if (!private_key)
		return -EINVAL;

	if (curve->g.ndigits != ndigits)
		return -EINVAL;

	/* Make sure the private key is in the range [2, n-3]. */
	if (vli_cmp(one, private_key, ndigits) != -1)
		return -EINVAL;
	vli_sub(res, curve->n, one, ndigits);
	vli_sub(res, res, one, ndigits);
	if (vli_cmp(res, private_key, ndigits) != 1)
		return -EINVAL;

	return 0;
}

int ecc_is_key_valid(unsigned int curve_id, unsigned int ndigits,
		     const u64 *private_key, unsigned int private_key_len)
{
	int nbytes;
	const struct ecc_curve *curve = ecc_get_curve(curve_id);

	nbytes = ndigits << ECC_DIGITS_TO_BYTES_SHIFT;

	if (private_key_len != nbytes)
		return -EINVAL;

	return __ecc_is_key_valid(curve, private_key, ndigits);
}

/*
 * ECC private keys are generated using the method of extra random bits,
 * equivalent to that described in FIPS 186-4, Appendix B.4.1.
 *
 * d = (c mod(n–1)) + 1    where c is a string of random bits, 64 bits longer
 *                         than requested
 * 0 <= c mod(n-1) <= n-2  and implies that
 * 1 <= d <= n-1
 *
 * This method generates a private key uniformly distributed in the range
 * [1, n-1].
 */
int ecc_gen_privkey(unsigned int curve_id, unsigned int ndigits, u64 *privkey)
{
	const struct ecc_curve *curve = ecc_get_curve(curve_id);
	u64 priv[ECC_MAX_DIGITS];
	unsigned int nbytes = ndigits << ECC_DIGITS_TO_BYTES_SHIFT;
	unsigned int nbits = vli_num_bits(curve->n, ndigits);
	int err;

	/* Check that N is included in Table 1 of FIPS 186-4, section 6.1.1 */
	if (nbits < 160 || ndigits > ARRAY_SIZE(priv))
		return -EINVAL;

	/*
	 * FIPS 186-4 recommends that the private key should be obtained from a
	 * RBG with a security strength equal to or greater than the security
	 * strength associated with N.
	 *
	 * The maximum security strength identified by NIST SP800-57pt1r4 for
	 * ECC is 256 (N >= 512).
	 *
	 * This condition is met by the default RNG because it selects a favored
	 * DRBG with a security strength of 256.
	 */
	if (crypto_get_default_rng())
		return -EFAULT;

	err = crypto_rng_get_bytes(crypto_default_rng, (u8 *)priv, nbytes);
	crypto_put_default_rng();
	if (err)
		return err;

	/* Make sure the private key is in the valid range. */
	if (__ecc_is_key_valid(curve, priv, ndigits))
		return -EINVAL;

	ecc_swap_digits(priv, privkey, ndigits);

	return 0;
}

int ecc_make_pub_key(unsigned int curve_id, unsigned int ndigits,
		     const u64 *private_key, u64 *public_key)
{
	int ret = 0;
	struct ecc_point *pk;
	u64 priv[ECC_MAX_DIGITS];
	const struct ecc_curve *curve = ecc_get_curve(curve_id);

	if (!private_key || !curve || ndigits > ARRAY_SIZE(priv)) {
		ret = -EINVAL;
		goto out;
	}

	ecc_swap_digits(private_key, priv, ndigits);

	pk = ecc_alloc_point(ndigits);
	if (!pk) {
		ret = -ENOMEM;
		goto out;
	}

	ecc_point_mult(pk, &curve->g, priv, NULL, curve, ndigits);
	if (ecc_point_is_zero(pk)) {
		ret = -EAGAIN;
		goto err_free_point;
	}

	ecc_swap_digits(pk->x, public_key, ndigits);
	ecc_swap_digits(pk->y, &public_key[ndigits], ndigits);

err_free_point:
	ecc_free_point(pk);
out:
	return ret;
}

/* SP800-56A section 5.6.2.3.4 partial verification: ephemeral keys only */
static int ecc_is_pubkey_valid_partial(const struct ecc_curve *curve,
				       struct ecc_point *pk)
{
	u64 yy[ECC_MAX_DIGITS], xxx[ECC_MAX_DIGITS], w[ECC_MAX_DIGITS];

	/* Check 1: Verify key is not the zero point. */
	if (ecc_point_is_zero(pk))
		return -EINVAL;

	/* Check 2: Verify key is in the range [1, p-1]. */
	if (vli_cmp(curve->p, pk->x, pk->ndigits) != 1)
		return -EINVAL;
	if (vli_cmp(curve->p, pk->y, pk->ndigits) != 1)
		return -EINVAL;

	/* Check 3: Verify that y^2 == (x^3 + a·x + b) mod p */
	vli_mod_square_fast(yy, pk->y, curve->p, pk->ndigits); /* y^2 */
	vli_mod_square_fast(xxx, pk->x, curve->p, pk->ndigits); /* x^2 */
	vli_mod_mult_fast(xxx, xxx, pk->x, curve->p, pk->ndigits); /* x^3 */
	vli_mod_mult_fast(w, curve->a, pk->x, curve->p, pk->ndigits); /* a·x */
	vli_mod_add(w, w, curve->b, curve->p, pk->ndigits); /* a·x + b */
	vli_mod_add(w, w, xxx, curve->p, pk->ndigits); /* x^3 + a·x + b */
	if (vli_cmp(yy, w, pk->ndigits) != 0) /* Equation */
		return -EINVAL;

	return 0;

}

int crypto_ecdh_shared_secret(unsigned int curve_id, unsigned int ndigits,
			      const u64 *private_key, const u64 *public_key,
			      u64 *secret)
{
	int ret = 0;
	struct ecc_point *product, *pk;
	u64 priv[ECC_MAX_DIGITS];
	u64 rand_z[ECC_MAX_DIGITS];
	unsigned int nbytes;
	const struct ecc_curve *curve = ecc_get_curve(curve_id);

	if (!private_key || !public_key || !curve ||
	    ndigits > ARRAY_SIZE(priv) || ndigits > ARRAY_SIZE(rand_z)) {
		ret = -EINVAL;
		goto out;
	}

	nbytes = ndigits << ECC_DIGITS_TO_BYTES_SHIFT;

	get_random_bytes(rand_z, nbytes);

	pk = ecc_alloc_point(ndigits);
	if (!pk) {
		ret = -ENOMEM;
		goto out;
	}

	ecc_swap_digits(public_key, pk->x, ndigits);
	ecc_swap_digits(&public_key[ndigits], pk->y, ndigits);
	ret = ecc_is_pubkey_valid_partial(curve, pk);
	if (ret)
		goto err_alloc_product;

	ecc_swap_digits(private_key, priv, ndigits);

	product = ecc_alloc_point(ndigits);
	if (!product) {
		ret = -ENOMEM;
		goto err_alloc_product;
	}

	ecc_point_mult(product, pk, priv, rand_z, curve, ndigits);

	ecc_swap_digits(product->x, secret, ndigits);

	if (ecc_point_is_zero(product))
		ret = -EFAULT;

	ecc_free_point(product);
err_alloc_product:
	ecc_free_point(pk);
out:
	return ret;
}