1 /*
2 * Core bignum functions
3 *
4 * Copyright The Mbed TLS Contributors
5 * SPDX-License-Identifier: Apache-2.0 OR GPL-2.0-or-later
6 */
7
8 #include "common.h"
9
10 #if defined(MBEDTLS_BIGNUM_C)
11
12 #include <string.h>
13
14 #include "mbedtls/error.h"
15 #include "mbedtls/platform_util.h"
16 #include "constant_time_internal.h"
17
18 #include "mbedtls/platform.h"
19
20 #include "bignum_core.h"
21 #include "bn_mul.h"
22 #include "constant_time_internal.h"
23
mbedtls_mpi_core_clz(mbedtls_mpi_uint a)24 size_t mbedtls_mpi_core_clz(mbedtls_mpi_uint a)
25 {
26 #if defined(__has_builtin)
27 #if (MBEDTLS_MPI_UINT_MAX == UINT_MAX) && __has_builtin(__builtin_clz)
28 #define core_clz __builtin_clz
29 #elif (MBEDTLS_MPI_UINT_MAX == ULONG_MAX) && __has_builtin(__builtin_clzl)
30 #define core_clz __builtin_clzl
31 #elif (MBEDTLS_MPI_UINT_MAX == ULLONG_MAX) && __has_builtin(__builtin_clzll)
32 #define core_clz __builtin_clzll
33 #endif
34 #endif
35 #if defined(core_clz)
36 return (size_t) core_clz(a);
37 #else
38 size_t j;
39 mbedtls_mpi_uint mask = (mbedtls_mpi_uint) 1 << (biL - 1);
40
41 for (j = 0; j < biL; j++) {
42 if (a & mask) {
43 break;
44 }
45
46 mask >>= 1;
47 }
48
49 return j;
50 #endif
51 }
52
mbedtls_mpi_core_bitlen(const mbedtls_mpi_uint * A,size_t A_limbs)53 size_t mbedtls_mpi_core_bitlen(const mbedtls_mpi_uint *A, size_t A_limbs)
54 {
55 int i;
56 size_t j;
57
58 for (i = ((int) A_limbs) - 1; i >= 0; i--) {
59 if (A[i] != 0) {
60 j = biL - mbedtls_mpi_core_clz(A[i]);
61 return (i * biL) + j;
62 }
63 }
64
65 return 0;
66 }
67
mpi_bigendian_to_host(mbedtls_mpi_uint a)68 static mbedtls_mpi_uint mpi_bigendian_to_host(mbedtls_mpi_uint a)
69 {
70 if (MBEDTLS_IS_BIG_ENDIAN) {
71 /* Nothing to do on bigendian systems. */
72 return a;
73 } else {
74 #if defined(MBEDTLS_HAVE_INT32)
75 return (mbedtls_mpi_uint) MBEDTLS_BSWAP32(a);
76 #elif defined(MBEDTLS_HAVE_INT64)
77 return (mbedtls_mpi_uint) MBEDTLS_BSWAP64(a);
78 #endif
79 }
80 }
81
mbedtls_mpi_core_bigendian_to_host(mbedtls_mpi_uint * A,size_t A_limbs)82 void mbedtls_mpi_core_bigendian_to_host(mbedtls_mpi_uint *A,
83 size_t A_limbs)
84 {
85 mbedtls_mpi_uint *cur_limb_left;
86 mbedtls_mpi_uint *cur_limb_right;
87 if (A_limbs == 0) {
88 return;
89 }
90
91 /*
92 * Traverse limbs and
93 * - adapt byte-order in each limb
94 * - swap the limbs themselves.
95 * For that, simultaneously traverse the limbs from left to right
96 * and from right to left, as long as the left index is not bigger
97 * than the right index (it's not a problem if limbs is odd and the
98 * indices coincide in the last iteration).
99 */
100 for (cur_limb_left = A, cur_limb_right = A + (A_limbs - 1);
101 cur_limb_left <= cur_limb_right;
102 cur_limb_left++, cur_limb_right--) {
103 mbedtls_mpi_uint tmp;
104 /* Note that if cur_limb_left == cur_limb_right,
105 * this code effectively swaps the bytes only once. */
106 tmp = mpi_bigendian_to_host(*cur_limb_left);
107 *cur_limb_left = mpi_bigendian_to_host(*cur_limb_right);
108 *cur_limb_right = tmp;
109 }
110 }
111
112 /* Whether min <= A, in constant time.
113 * A_limbs must be at least 1. */
mbedtls_mpi_core_uint_le_mpi(mbedtls_mpi_uint min,const mbedtls_mpi_uint * A,size_t A_limbs)114 mbedtls_ct_condition_t mbedtls_mpi_core_uint_le_mpi(mbedtls_mpi_uint min,
115 const mbedtls_mpi_uint *A,
116 size_t A_limbs)
117 {
118 /* min <= least significant limb? */
119 mbedtls_ct_condition_t min_le_lsl = mbedtls_ct_uint_ge(A[0], min);
120
121 /* limbs other than the least significant one are all zero? */
122 mbedtls_ct_condition_t msll_mask = MBEDTLS_CT_FALSE;
123 for (size_t i = 1; i < A_limbs; i++) {
124 msll_mask = mbedtls_ct_bool_or(msll_mask, mbedtls_ct_bool(A[i]));
125 }
126
127 /* min <= A iff the lowest limb of A is >= min or the other limbs
128 * are not all zero. */
129 return mbedtls_ct_bool_or(msll_mask, min_le_lsl);
130 }
131
mbedtls_mpi_core_lt_ct(const mbedtls_mpi_uint * A,const mbedtls_mpi_uint * B,size_t limbs)132 mbedtls_ct_condition_t mbedtls_mpi_core_lt_ct(const mbedtls_mpi_uint *A,
133 const mbedtls_mpi_uint *B,
134 size_t limbs)
135 {
136 mbedtls_ct_condition_t ret = MBEDTLS_CT_FALSE, cond = MBEDTLS_CT_FALSE, done = MBEDTLS_CT_FALSE;
137
138 for (size_t i = limbs; i > 0; i--) {
139 /*
140 * If B[i - 1] < A[i - 1] then A < B is false and the result must
141 * remain 0.
142 *
143 * Again even if we can make a decision, we just mark the result and
144 * the fact that we are done and continue looping.
145 */
146 cond = mbedtls_ct_uint_lt(B[i - 1], A[i - 1]);
147 done = mbedtls_ct_bool_or(done, cond);
148
149 /*
150 * If A[i - 1] < B[i - 1] then A < B is true.
151 *
152 * Again even if we can make a decision, we just mark the result and
153 * the fact that we are done and continue looping.
154 */
155 cond = mbedtls_ct_uint_lt(A[i - 1], B[i - 1]);
156 ret = mbedtls_ct_bool_or(ret, mbedtls_ct_bool_and(cond, mbedtls_ct_bool_not(done)));
157 done = mbedtls_ct_bool_or(done, cond);
158 }
159
160 /*
161 * If all the limbs were equal, then the numbers are equal, A < B is false
162 * and leaving the result 0 is correct.
163 */
164
165 return ret;
166 }
167
mbedtls_mpi_core_cond_assign(mbedtls_mpi_uint * X,const mbedtls_mpi_uint * A,size_t limbs,mbedtls_ct_condition_t assign)168 void mbedtls_mpi_core_cond_assign(mbedtls_mpi_uint *X,
169 const mbedtls_mpi_uint *A,
170 size_t limbs,
171 mbedtls_ct_condition_t assign)
172 {
173 if (X == A) {
174 return;
175 }
176
177 /* This function is very performance-sensitive for RSA. For this reason
178 * we have the loop below, instead of calling mbedtls_ct_memcpy_if
179 * (this is more optimal since here we don't have to handle the case where
180 * we copy awkwardly sized data).
181 */
182 for (size_t i = 0; i < limbs; i++) {
183 X[i] = mbedtls_ct_mpi_uint_if(assign, A[i], X[i]);
184 }
185 }
186
mbedtls_mpi_core_cond_swap(mbedtls_mpi_uint * X,mbedtls_mpi_uint * Y,size_t limbs,mbedtls_ct_condition_t swap)187 void mbedtls_mpi_core_cond_swap(mbedtls_mpi_uint *X,
188 mbedtls_mpi_uint *Y,
189 size_t limbs,
190 mbedtls_ct_condition_t swap)
191 {
192 if (X == Y) {
193 return;
194 }
195
196 for (size_t i = 0; i < limbs; i++) {
197 mbedtls_mpi_uint tmp = X[i];
198 X[i] = mbedtls_ct_mpi_uint_if(swap, Y[i], X[i]);
199 Y[i] = mbedtls_ct_mpi_uint_if(swap, tmp, Y[i]);
200 }
201 }
202
mbedtls_mpi_core_read_le(mbedtls_mpi_uint * X,size_t X_limbs,const unsigned char * input,size_t input_length)203 int mbedtls_mpi_core_read_le(mbedtls_mpi_uint *X,
204 size_t X_limbs,
205 const unsigned char *input,
206 size_t input_length)
207 {
208 const size_t limbs = CHARS_TO_LIMBS(input_length);
209
210 if (X_limbs < limbs) {
211 return MBEDTLS_ERR_MPI_BUFFER_TOO_SMALL;
212 }
213
214 if (X != NULL) {
215 memset(X, 0, X_limbs * ciL);
216
217 for (size_t i = 0; i < input_length; i++) {
218 size_t offset = ((i % ciL) << 3);
219 X[i / ciL] |= ((mbedtls_mpi_uint) input[i]) << offset;
220 }
221 }
222
223 return 0;
224 }
225
mbedtls_mpi_core_read_be(mbedtls_mpi_uint * X,size_t X_limbs,const unsigned char * input,size_t input_length)226 int mbedtls_mpi_core_read_be(mbedtls_mpi_uint *X,
227 size_t X_limbs,
228 const unsigned char *input,
229 size_t input_length)
230 {
231 const size_t limbs = CHARS_TO_LIMBS(input_length);
232
233 if (X_limbs < limbs) {
234 return MBEDTLS_ERR_MPI_BUFFER_TOO_SMALL;
235 }
236
237 /* If X_limbs is 0, input_length must also be 0 (from previous test).
238 * Nothing to do. */
239 if (X_limbs == 0) {
240 return 0;
241 }
242
243 memset(X, 0, X_limbs * ciL);
244
245 /* memcpy() with (NULL, 0) is undefined behaviour */
246 if (input_length != 0) {
247 size_t overhead = (X_limbs * ciL) - input_length;
248 unsigned char *Xp = (unsigned char *) X;
249 memcpy(Xp + overhead, input, input_length);
250 }
251
252 mbedtls_mpi_core_bigendian_to_host(X, X_limbs);
253
254 return 0;
255 }
256
mbedtls_mpi_core_write_le(const mbedtls_mpi_uint * A,size_t A_limbs,unsigned char * output,size_t output_length)257 int mbedtls_mpi_core_write_le(const mbedtls_mpi_uint *A,
258 size_t A_limbs,
259 unsigned char *output,
260 size_t output_length)
261 {
262 size_t stored_bytes = A_limbs * ciL;
263 size_t bytes_to_copy;
264
265 if (stored_bytes < output_length) {
266 bytes_to_copy = stored_bytes;
267 } else {
268 bytes_to_copy = output_length;
269
270 /* The output buffer is smaller than the allocated size of A.
271 * However A may fit if its leading bytes are zero. */
272 for (size_t i = bytes_to_copy; i < stored_bytes; i++) {
273 if (GET_BYTE(A, i) != 0) {
274 return MBEDTLS_ERR_MPI_BUFFER_TOO_SMALL;
275 }
276 }
277 }
278
279 for (size_t i = 0; i < bytes_to_copy; i++) {
280 output[i] = GET_BYTE(A, i);
281 }
282
283 if (stored_bytes < output_length) {
284 /* Write trailing 0 bytes */
285 memset(output + stored_bytes, 0, output_length - stored_bytes);
286 }
287
288 return 0;
289 }
290
mbedtls_mpi_core_write_be(const mbedtls_mpi_uint * X,size_t X_limbs,unsigned char * output,size_t output_length)291 int mbedtls_mpi_core_write_be(const mbedtls_mpi_uint *X,
292 size_t X_limbs,
293 unsigned char *output,
294 size_t output_length)
295 {
296 size_t stored_bytes;
297 size_t bytes_to_copy;
298 unsigned char *p;
299
300 stored_bytes = X_limbs * ciL;
301
302 if (stored_bytes < output_length) {
303 /* There is enough space in the output buffer. Write initial
304 * null bytes and record the position at which to start
305 * writing the significant bytes. In this case, the execution
306 * trace of this function does not depend on the value of the
307 * number. */
308 bytes_to_copy = stored_bytes;
309 p = output + output_length - stored_bytes;
310 memset(output, 0, output_length - stored_bytes);
311 } else {
312 /* The output buffer is smaller than the allocated size of X.
313 * However X may fit if its leading bytes are zero. */
314 bytes_to_copy = output_length;
315 p = output;
316 for (size_t i = bytes_to_copy; i < stored_bytes; i++) {
317 if (GET_BYTE(X, i) != 0) {
318 return MBEDTLS_ERR_MPI_BUFFER_TOO_SMALL;
319 }
320 }
321 }
322
323 for (size_t i = 0; i < bytes_to_copy; i++) {
324 p[bytes_to_copy - i - 1] = GET_BYTE(X, i);
325 }
326
327 return 0;
328 }
329
mbedtls_mpi_core_shift_r(mbedtls_mpi_uint * X,size_t limbs,size_t count)330 void mbedtls_mpi_core_shift_r(mbedtls_mpi_uint *X, size_t limbs,
331 size_t count)
332 {
333 size_t i, v0, v1;
334 mbedtls_mpi_uint r0 = 0, r1;
335
336 v0 = count / biL;
337 v1 = count & (biL - 1);
338
339 if (v0 > limbs || (v0 == limbs && v1 > 0)) {
340 memset(X, 0, limbs * ciL);
341 return;
342 }
343
344 /*
345 * shift by count / limb_size
346 */
347 if (v0 > 0) {
348 for (i = 0; i < limbs - v0; i++) {
349 X[i] = X[i + v0];
350 }
351
352 for (; i < limbs; i++) {
353 X[i] = 0;
354 }
355 }
356
357 /*
358 * shift by count % limb_size
359 */
360 if (v1 > 0) {
361 for (i = limbs; i > 0; i--) {
362 r1 = X[i - 1] << (biL - v1);
363 X[i - 1] >>= v1;
364 X[i - 1] |= r0;
365 r0 = r1;
366 }
367 }
368 }
369
mbedtls_mpi_core_shift_l(mbedtls_mpi_uint * X,size_t limbs,size_t count)370 void mbedtls_mpi_core_shift_l(mbedtls_mpi_uint *X, size_t limbs,
371 size_t count)
372 {
373 size_t i, v0, v1;
374 mbedtls_mpi_uint r0 = 0, r1;
375
376 v0 = count / (biL);
377 v1 = count & (biL - 1);
378
379 /*
380 * shift by count / limb_size
381 */
382 if (v0 > 0) {
383 for (i = limbs; i > v0; i--) {
384 X[i - 1] = X[i - v0 - 1];
385 }
386
387 for (; i > 0; i--) {
388 X[i - 1] = 0;
389 }
390 }
391
392 /*
393 * shift by count % limb_size
394 */
395 if (v1 > 0) {
396 for (i = v0; i < limbs; i++) {
397 r1 = X[i] >> (biL - v1);
398 X[i] <<= v1;
399 X[i] |= r0;
400 r0 = r1;
401 }
402 }
403 }
404
mbedtls_mpi_core_add(mbedtls_mpi_uint * X,const mbedtls_mpi_uint * A,const mbedtls_mpi_uint * B,size_t limbs)405 mbedtls_mpi_uint mbedtls_mpi_core_add(mbedtls_mpi_uint *X,
406 const mbedtls_mpi_uint *A,
407 const mbedtls_mpi_uint *B,
408 size_t limbs)
409 {
410 mbedtls_mpi_uint c = 0;
411
412 for (size_t i = 0; i < limbs; i++) {
413 mbedtls_mpi_uint t = c + A[i];
414 c = (t < A[i]);
415 t += B[i];
416 c += (t < B[i]);
417 X[i] = t;
418 }
419
420 return c;
421 }
422
mbedtls_mpi_core_add_if(mbedtls_mpi_uint * X,const mbedtls_mpi_uint * A,size_t limbs,unsigned cond)423 mbedtls_mpi_uint mbedtls_mpi_core_add_if(mbedtls_mpi_uint *X,
424 const mbedtls_mpi_uint *A,
425 size_t limbs,
426 unsigned cond)
427 {
428 mbedtls_mpi_uint c = 0;
429
430 mbedtls_ct_condition_t do_add = mbedtls_ct_bool(cond);
431
432 for (size_t i = 0; i < limbs; i++) {
433 mbedtls_mpi_uint add = mbedtls_ct_mpi_uint_if_else_0(do_add, A[i]);
434 mbedtls_mpi_uint t = c + X[i];
435 c = (t < X[i]);
436 t += add;
437 c += (t < add);
438 X[i] = t;
439 }
440
441 return c;
442 }
443
mbedtls_mpi_core_sub(mbedtls_mpi_uint * X,const mbedtls_mpi_uint * A,const mbedtls_mpi_uint * B,size_t limbs)444 mbedtls_mpi_uint mbedtls_mpi_core_sub(mbedtls_mpi_uint *X,
445 const mbedtls_mpi_uint *A,
446 const mbedtls_mpi_uint *B,
447 size_t limbs)
448 {
449 mbedtls_mpi_uint c = 0;
450
451 for (size_t i = 0; i < limbs; i++) {
452 mbedtls_mpi_uint z = (A[i] < c);
453 mbedtls_mpi_uint t = A[i] - c;
454 c = (t < B[i]) + z;
455 X[i] = t - B[i];
456 }
457
458 return c;
459 }
460
mbedtls_mpi_core_mla(mbedtls_mpi_uint * d,size_t d_len,const mbedtls_mpi_uint * s,size_t s_len,mbedtls_mpi_uint b)461 mbedtls_mpi_uint mbedtls_mpi_core_mla(mbedtls_mpi_uint *d, size_t d_len,
462 const mbedtls_mpi_uint *s, size_t s_len,
463 mbedtls_mpi_uint b)
464 {
465 mbedtls_mpi_uint c = 0; /* carry */
466 /*
467 * It is a documented precondition of this function that d_len >= s_len.
468 * If that's not the case, we swap these round: this turns what would be
469 * a buffer overflow into an incorrect result.
470 */
471 if (d_len < s_len) {
472 s_len = d_len;
473 }
474 size_t excess_len = d_len - s_len;
475 size_t steps_x8 = s_len / 8;
476 size_t steps_x1 = s_len & 7;
477
478 while (steps_x8--) {
479 MULADDC_X8_INIT
480 MULADDC_X8_CORE
481 MULADDC_X8_STOP
482 }
483
484 while (steps_x1--) {
485 MULADDC_X1_INIT
486 MULADDC_X1_CORE
487 MULADDC_X1_STOP
488 }
489
490 while (excess_len--) {
491 *d += c;
492 c = (*d < c);
493 d++;
494 }
495
496 return c;
497 }
498
mbedtls_mpi_core_mul(mbedtls_mpi_uint * X,const mbedtls_mpi_uint * A,size_t A_limbs,const mbedtls_mpi_uint * B,size_t B_limbs)499 void mbedtls_mpi_core_mul(mbedtls_mpi_uint *X,
500 const mbedtls_mpi_uint *A, size_t A_limbs,
501 const mbedtls_mpi_uint *B, size_t B_limbs)
502 {
503 memset(X, 0, (A_limbs + B_limbs) * ciL);
504
505 for (size_t i = 0; i < B_limbs; i++) {
506 (void) mbedtls_mpi_core_mla(X + i, A_limbs + 1, A, A_limbs, B[i]);
507 }
508 }
509
510 /*
511 * Fast Montgomery initialization (thanks to Tom St Denis).
512 */
mbedtls_mpi_core_montmul_init(const mbedtls_mpi_uint * N)513 mbedtls_mpi_uint mbedtls_mpi_core_montmul_init(const mbedtls_mpi_uint *N)
514 {
515 mbedtls_mpi_uint x = N[0];
516
517 x += ((N[0] + 2) & 4) << 1;
518
519 for (unsigned int i = biL; i >= 8; i /= 2) {
520 x *= (2 - (N[0] * x));
521 }
522
523 return ~x + 1;
524 }
525
mbedtls_mpi_core_montmul(mbedtls_mpi_uint * X,const mbedtls_mpi_uint * A,const mbedtls_mpi_uint * B,size_t B_limbs,const mbedtls_mpi_uint * N,size_t AN_limbs,mbedtls_mpi_uint mm,mbedtls_mpi_uint * T)526 void mbedtls_mpi_core_montmul(mbedtls_mpi_uint *X,
527 const mbedtls_mpi_uint *A,
528 const mbedtls_mpi_uint *B,
529 size_t B_limbs,
530 const mbedtls_mpi_uint *N,
531 size_t AN_limbs,
532 mbedtls_mpi_uint mm,
533 mbedtls_mpi_uint *T)
534 {
535 memset(T, 0, (2 * AN_limbs + 1) * ciL);
536
537 for (size_t i = 0; i < AN_limbs; i++) {
538 /* T = (T + u0*B + u1*N) / 2^biL */
539 mbedtls_mpi_uint u0 = A[i];
540 mbedtls_mpi_uint u1 = (T[0] + u0 * B[0]) * mm;
541
542 (void) mbedtls_mpi_core_mla(T, AN_limbs + 2, B, B_limbs, u0);
543 (void) mbedtls_mpi_core_mla(T, AN_limbs + 2, N, AN_limbs, u1);
544
545 T++;
546 }
547
548 /*
549 * The result we want is (T >= N) ? T - N : T.
550 *
551 * For better constant-time properties in this function, we always do the
552 * subtraction, with the result in X.
553 *
554 * We also look to see if there was any carry in the final additions in the
555 * loop above.
556 */
557
558 mbedtls_mpi_uint carry = T[AN_limbs];
559 mbedtls_mpi_uint borrow = mbedtls_mpi_core_sub(X, T, N, AN_limbs);
560
561 /*
562 * Using R as the Montgomery radix (auxiliary modulus) i.e. 2^(biL*AN_limbs):
563 *
564 * T can be in one of 3 ranges:
565 *
566 * 1) T < N : (carry, borrow) = (0, 1): we want T
567 * 2) N <= T < R : (carry, borrow) = (0, 0): we want X
568 * 3) T >= R : (carry, borrow) = (1, 1): we want X
569 *
570 * and (carry, borrow) = (1, 0) can't happen.
571 *
572 * So the correct return value is already in X if (carry ^ borrow) = 0,
573 * but is in (the lower AN_limbs limbs of) T if (carry ^ borrow) = 1.
574 */
575 mbedtls_ct_memcpy_if(mbedtls_ct_bool(carry ^ borrow),
576 (unsigned char *) X,
577 (unsigned char *) T,
578 NULL,
579 AN_limbs * sizeof(mbedtls_mpi_uint));
580 }
581
mbedtls_mpi_core_get_mont_r2_unsafe(mbedtls_mpi * X,const mbedtls_mpi * N)582 int mbedtls_mpi_core_get_mont_r2_unsafe(mbedtls_mpi *X,
583 const mbedtls_mpi *N)
584 {
585 int ret = MBEDTLS_ERR_ERROR_CORRUPTION_DETECTED;
586
587 MBEDTLS_MPI_CHK(mbedtls_mpi_lset(X, 1));
588 MBEDTLS_MPI_CHK(mbedtls_mpi_shift_l(X, N->n * 2 * biL));
589 MBEDTLS_MPI_CHK(mbedtls_mpi_mod_mpi(X, X, N));
590 MBEDTLS_MPI_CHK(mbedtls_mpi_shrink(X, N->n));
591
592 cleanup:
593 return ret;
594 }
595
596 MBEDTLS_STATIC_TESTABLE
mbedtls_mpi_core_ct_uint_table_lookup(mbedtls_mpi_uint * dest,const mbedtls_mpi_uint * table,size_t limbs,size_t count,size_t index)597 void mbedtls_mpi_core_ct_uint_table_lookup(mbedtls_mpi_uint *dest,
598 const mbedtls_mpi_uint *table,
599 size_t limbs,
600 size_t count,
601 size_t index)
602 {
603 for (size_t i = 0; i < count; i++, table += limbs) {
604 mbedtls_ct_condition_t assign = mbedtls_ct_uint_eq(i, index);
605 mbedtls_mpi_core_cond_assign(dest, table, limbs, assign);
606 }
607 }
608
609 /* Fill X with n_bytes random bytes.
610 * X must already have room for those bytes.
611 * The ordering of the bytes returned from the RNG is suitable for
612 * deterministic ECDSA (see RFC 6979 §3.3 and the specification of
613 * mbedtls_mpi_core_random()).
614 */
mbedtls_mpi_core_fill_random(mbedtls_mpi_uint * X,size_t X_limbs,size_t n_bytes,int (* f_rng)(void *,unsigned char *,size_t),void * p_rng)615 int mbedtls_mpi_core_fill_random(
616 mbedtls_mpi_uint *X, size_t X_limbs,
617 size_t n_bytes,
618 int (*f_rng)(void *, unsigned char *, size_t), void *p_rng)
619 {
620 int ret = MBEDTLS_ERR_ERROR_CORRUPTION_DETECTED;
621 const size_t limbs = CHARS_TO_LIMBS(n_bytes);
622 const size_t overhead = (limbs * ciL) - n_bytes;
623
624 if (X_limbs < limbs) {
625 return MBEDTLS_ERR_MPI_BAD_INPUT_DATA;
626 }
627
628 memset(X, 0, overhead);
629 memset((unsigned char *) X + limbs * ciL, 0, (X_limbs - limbs) * ciL);
630 MBEDTLS_MPI_CHK(f_rng(p_rng, (unsigned char *) X + overhead, n_bytes));
631 mbedtls_mpi_core_bigendian_to_host(X, limbs);
632
633 cleanup:
634 return ret;
635 }
636
mbedtls_mpi_core_random(mbedtls_mpi_uint * X,mbedtls_mpi_uint min,const mbedtls_mpi_uint * N,size_t limbs,int (* f_rng)(void *,unsigned char *,size_t),void * p_rng)637 int mbedtls_mpi_core_random(mbedtls_mpi_uint *X,
638 mbedtls_mpi_uint min,
639 const mbedtls_mpi_uint *N,
640 size_t limbs,
641 int (*f_rng)(void *, unsigned char *, size_t),
642 void *p_rng)
643 {
644 mbedtls_ct_condition_t ge_lower = MBEDTLS_CT_TRUE, lt_upper = MBEDTLS_CT_FALSE;
645 size_t n_bits = mbedtls_mpi_core_bitlen(N, limbs);
646 size_t n_bytes = (n_bits + 7) / 8;
647 int ret = MBEDTLS_ERR_ERROR_CORRUPTION_DETECTED;
648
649 /*
650 * When min == 0, each try has at worst a probability 1/2 of failing
651 * (the msb has a probability 1/2 of being 0, and then the result will
652 * be < N), so after 30 tries failure probability is a most 2**(-30).
653 *
654 * When N is just below a power of 2, as is the case when generating
655 * a random scalar on most elliptic curves, 1 try is enough with
656 * overwhelming probability. When N is just above a power of 2,
657 * as when generating a random scalar on secp224k1, each try has
658 * a probability of failing that is almost 1/2.
659 *
660 * The probabilities are almost the same if min is nonzero but negligible
661 * compared to N. This is always the case when N is crypto-sized, but
662 * it's convenient to support small N for testing purposes. When N
663 * is small, use a higher repeat count, otherwise the probability of
664 * failure is macroscopic.
665 */
666 int count = (n_bytes > 4 ? 30 : 250);
667
668 /*
669 * Match the procedure given in RFC 6979 §3.3 (deterministic ECDSA)
670 * when f_rng is a suitably parametrized instance of HMAC_DRBG:
671 * - use the same byte ordering;
672 * - keep the leftmost n_bits bits of the generated octet string;
673 * - try until result is in the desired range.
674 * This also avoids any bias, which is especially important for ECDSA.
675 */
676 do {
677 MBEDTLS_MPI_CHK(mbedtls_mpi_core_fill_random(X, limbs,
678 n_bytes,
679 f_rng, p_rng));
680 mbedtls_mpi_core_shift_r(X, limbs, 8 * n_bytes - n_bits);
681
682 if (--count == 0) {
683 ret = MBEDTLS_ERR_MPI_NOT_ACCEPTABLE;
684 goto cleanup;
685 }
686
687 ge_lower = mbedtls_mpi_core_uint_le_mpi(min, X, limbs);
688 lt_upper = mbedtls_mpi_core_lt_ct(X, N, limbs);
689 } while (mbedtls_ct_bool_and(ge_lower, lt_upper) == MBEDTLS_CT_FALSE);
690
691 cleanup:
692 return ret;
693 }
694
exp_mod_get_window_size(size_t Ebits)695 static size_t exp_mod_get_window_size(size_t Ebits)
696 {
697 #if MBEDTLS_MPI_WINDOW_SIZE >= 6
698 return (Ebits > 671) ? 6 : (Ebits > 239) ? 5 : (Ebits > 79) ? 4 : 1;
699 #elif MBEDTLS_MPI_WINDOW_SIZE == 5
700 return (Ebits > 239) ? 5 : (Ebits > 79) ? 4 : 1;
701 #elif MBEDTLS_MPI_WINDOW_SIZE > 1
702 return (Ebits > 79) ? MBEDTLS_MPI_WINDOW_SIZE : 1;
703 #else
704 (void) Ebits;
705 return 1;
706 #endif
707 }
708
mbedtls_mpi_core_exp_mod_working_limbs(size_t AN_limbs,size_t E_limbs)709 size_t mbedtls_mpi_core_exp_mod_working_limbs(size_t AN_limbs, size_t E_limbs)
710 {
711 const size_t wsize = exp_mod_get_window_size(E_limbs * biL);
712 const size_t welem = ((size_t) 1) << wsize;
713
714 /* How big does each part of the working memory pool need to be? */
715 const size_t table_limbs = welem * AN_limbs;
716 const size_t select_limbs = AN_limbs;
717 const size_t temp_limbs = 2 * AN_limbs + 1;
718
719 return table_limbs + select_limbs + temp_limbs;
720 }
721
exp_mod_precompute_window(const mbedtls_mpi_uint * A,const mbedtls_mpi_uint * N,size_t AN_limbs,mbedtls_mpi_uint mm,const mbedtls_mpi_uint * RR,size_t welem,mbedtls_mpi_uint * Wtable,mbedtls_mpi_uint * temp)722 static void exp_mod_precompute_window(const mbedtls_mpi_uint *A,
723 const mbedtls_mpi_uint *N,
724 size_t AN_limbs,
725 mbedtls_mpi_uint mm,
726 const mbedtls_mpi_uint *RR,
727 size_t welem,
728 mbedtls_mpi_uint *Wtable,
729 mbedtls_mpi_uint *temp)
730 {
731 /* W[0] = 1 (in Montgomery presentation) */
732 memset(Wtable, 0, AN_limbs * ciL);
733 Wtable[0] = 1;
734 mbedtls_mpi_core_montmul(Wtable, Wtable, RR, AN_limbs, N, AN_limbs, mm, temp);
735
736 /* W[1] = A (already in Montgomery presentation) */
737 mbedtls_mpi_uint *W1 = Wtable + AN_limbs;
738 memcpy(W1, A, AN_limbs * ciL);
739
740 /* W[i+1] = W[i] * W[1], i >= 2 */
741 mbedtls_mpi_uint *Wprev = W1;
742 for (size_t i = 2; i < welem; i++) {
743 mbedtls_mpi_uint *Wcur = Wprev + AN_limbs;
744 mbedtls_mpi_core_montmul(Wcur, Wprev, W1, AN_limbs, N, AN_limbs, mm, temp);
745 Wprev = Wcur;
746 }
747 }
748
749 /* Exponentiation: X := A^E mod N.
750 *
751 * A must already be in Montgomery form.
752 *
753 * As in other bignum functions, assume that AN_limbs and E_limbs are nonzero.
754 *
755 * RR must contain 2^{2*biL} mod N.
756 *
757 * The algorithm is a variant of Left-to-right k-ary exponentiation: HAC 14.82
758 * (The difference is that the body in our loop processes a single bit instead
759 * of a full window.)
760 */
mbedtls_mpi_core_exp_mod(mbedtls_mpi_uint * X,const mbedtls_mpi_uint * A,const mbedtls_mpi_uint * N,size_t AN_limbs,const mbedtls_mpi_uint * E,size_t E_limbs,const mbedtls_mpi_uint * RR,mbedtls_mpi_uint * T)761 void mbedtls_mpi_core_exp_mod(mbedtls_mpi_uint *X,
762 const mbedtls_mpi_uint *A,
763 const mbedtls_mpi_uint *N,
764 size_t AN_limbs,
765 const mbedtls_mpi_uint *E,
766 size_t E_limbs,
767 const mbedtls_mpi_uint *RR,
768 mbedtls_mpi_uint *T)
769 {
770 const size_t wsize = exp_mod_get_window_size(E_limbs * biL);
771 const size_t welem = ((size_t) 1) << wsize;
772
773 /* This is how we will use the temporary storage T, which must have space
774 * for table_limbs, select_limbs and (2 * AN_limbs + 1) for montmul. */
775 const size_t table_limbs = welem * AN_limbs;
776 const size_t select_limbs = AN_limbs;
777
778 /* Pointers to specific parts of the temporary working memory pool */
779 mbedtls_mpi_uint *const Wtable = T;
780 mbedtls_mpi_uint *const Wselect = Wtable + table_limbs;
781 mbedtls_mpi_uint *const temp = Wselect + select_limbs;
782
783 /*
784 * Window precomputation
785 */
786
787 const mbedtls_mpi_uint mm = mbedtls_mpi_core_montmul_init(N);
788
789 /* Set Wtable[i] = A^(2^i) (in Montgomery representation) */
790 exp_mod_precompute_window(A, N, AN_limbs,
791 mm, RR,
792 welem, Wtable, temp);
793
794 /*
795 * Fixed window exponentiation
796 */
797
798 /* X = 1 (in Montgomery presentation) initially */
799 memcpy(X, Wtable, AN_limbs * ciL);
800
801 /* We'll process the bits of E from most significant
802 * (limb_index=E_limbs-1, E_bit_index=biL-1) to least significant
803 * (limb_index=0, E_bit_index=0). */
804 size_t E_limb_index = E_limbs;
805 size_t E_bit_index = 0;
806 /* At any given time, window contains window_bits bits from E.
807 * window_bits can go up to wsize. */
808 size_t window_bits = 0;
809 mbedtls_mpi_uint window = 0;
810
811 do {
812 /* Square */
813 mbedtls_mpi_core_montmul(X, X, X, AN_limbs, N, AN_limbs, mm, temp);
814
815 /* Move to the next bit of the exponent */
816 if (E_bit_index == 0) {
817 --E_limb_index;
818 E_bit_index = biL - 1;
819 } else {
820 --E_bit_index;
821 }
822 /* Insert next exponent bit into window */
823 ++window_bits;
824 window <<= 1;
825 window |= (E[E_limb_index] >> E_bit_index) & 1;
826
827 /* Clear window if it's full. Also clear the window at the end,
828 * when we've finished processing the exponent. */
829 if (window_bits == wsize ||
830 (E_bit_index == 0 && E_limb_index == 0)) {
831 /* Select Wtable[window] without leaking window through
832 * memory access patterns. */
833 mbedtls_mpi_core_ct_uint_table_lookup(Wselect, Wtable,
834 AN_limbs, welem, window);
835 /* Multiply X by the selected element. */
836 mbedtls_mpi_core_montmul(X, X, Wselect, AN_limbs, N, AN_limbs, mm,
837 temp);
838 window = 0;
839 window_bits = 0;
840 }
841 } while (!(E_bit_index == 0 && E_limb_index == 0));
842 }
843
mbedtls_mpi_core_sub_int(mbedtls_mpi_uint * X,const mbedtls_mpi_uint * A,mbedtls_mpi_uint c,size_t limbs)844 mbedtls_mpi_uint mbedtls_mpi_core_sub_int(mbedtls_mpi_uint *X,
845 const mbedtls_mpi_uint *A,
846 mbedtls_mpi_uint c, /* doubles as carry */
847 size_t limbs)
848 {
849 for (size_t i = 0; i < limbs; i++) {
850 mbedtls_mpi_uint s = A[i];
851 mbedtls_mpi_uint t = s - c;
852 c = (t > s);
853 X[i] = t;
854 }
855
856 return c;
857 }
858
mbedtls_mpi_core_check_zero_ct(const mbedtls_mpi_uint * A,size_t limbs)859 mbedtls_mpi_uint mbedtls_mpi_core_check_zero_ct(const mbedtls_mpi_uint *A,
860 size_t limbs)
861 {
862 mbedtls_mpi_uint bits = 0;
863
864 for (size_t i = 0; i < limbs; i++) {
865 bits |= A[i];
866 }
867
868 return bits;
869 }
870
mbedtls_mpi_core_to_mont_rep(mbedtls_mpi_uint * X,const mbedtls_mpi_uint * A,const mbedtls_mpi_uint * N,size_t AN_limbs,mbedtls_mpi_uint mm,const mbedtls_mpi_uint * rr,mbedtls_mpi_uint * T)871 void mbedtls_mpi_core_to_mont_rep(mbedtls_mpi_uint *X,
872 const mbedtls_mpi_uint *A,
873 const mbedtls_mpi_uint *N,
874 size_t AN_limbs,
875 mbedtls_mpi_uint mm,
876 const mbedtls_mpi_uint *rr,
877 mbedtls_mpi_uint *T)
878 {
879 mbedtls_mpi_core_montmul(X, A, rr, AN_limbs, N, AN_limbs, mm, T);
880 }
881
mbedtls_mpi_core_from_mont_rep(mbedtls_mpi_uint * X,const mbedtls_mpi_uint * A,const mbedtls_mpi_uint * N,size_t AN_limbs,mbedtls_mpi_uint mm,mbedtls_mpi_uint * T)882 void mbedtls_mpi_core_from_mont_rep(mbedtls_mpi_uint *X,
883 const mbedtls_mpi_uint *A,
884 const mbedtls_mpi_uint *N,
885 size_t AN_limbs,
886 mbedtls_mpi_uint mm,
887 mbedtls_mpi_uint *T)
888 {
889 const mbedtls_mpi_uint Rinv = 1; /* 1/R in Mont. rep => 1 */
890
891 mbedtls_mpi_core_montmul(X, A, &Rinv, 1, N, AN_limbs, mm, T);
892 }
893
894 #endif /* MBEDTLS_BIGNUM_C */
895