3 * Generating big random numbers
6 /* nettle, low-level cryptographics library
8 * Copyright (C) 2002 Niels Möller
10 * The nettle library is free software; you can redistribute it and/or modify
11 * it under the terms of the GNU Lesser General Public License as published by
12 * the Free Software Foundation; either version 2.1 of the License, or (at your
13 * option) any later version.
15 * The nettle library is distributed in the hope that it will be useful, but
16 * WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
17 * or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public
18 * License for more details.
20 * You should have received a copy of the GNU Lesser General Public License
21 * along with the nettle library; see the file COPYING.LIB. If not, write to
22 * the Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston,
33 #include "nettle-internal.h"
36 nettle_mpz_random_size(mpz_t x
,
37 void *ctx
, nettle_random_func
*random
,
40 unsigned length
= (bits
+ 7) / 8;
41 TMP_DECL(data
, uint8_t, NETTLE_MAX_BIGNUM_SIZE
);
42 TMP_ALLOC(data
, length
);
44 random(ctx
, length
, data
);
46 nettle_mpz_set_str_256_u(x
, length
, data
);
49 mpz_fdiv_r_2exp(x
, x
, bits
);
52 /* Returns a random number x, 0 <= x < n */
54 nettle_mpz_random(mpz_t x
,
55 void *ctx
, nettle_random_func
*random
,
58 /* NOTE: This leaves some bias, which may be bad for DSA. A better
59 * way might be to generate a random number of mpz_sizeinbase(n, 2)
60 * bits, and loop until one smaller than n is found. */
62 /* From Daniel Bleichenbacher (via coderpunks):
64 * There is still a theoretical attack possible with 8 extra bits.
65 * But, the attack would need about 2^66 signatures 2^66 memory and
66 * 2^66 time (if I remember that correctly). Compare that to DSA,
67 * where the attack requires 2^22 signatures 2^40 memory and 2^64
68 * time. And of course, the numbers above are not a real threat for
69 * PGP. Using 16 extra bits (i.e. generating a 176 bit random number
70 * and reducing it modulo q) will defeat even this theoretical
73 * More generally log_2(q)/8 extra bits are enough to defeat my
74 * attack. NIST also plans to update the standard.
77 /* Add a few bits extra, to decrease the bias from the final modulo
78 * operation. NIST FIPS 186-3 specifies 64 extra bits, for use with
81 nettle_mpz_random_size(x
,
83 mpz_sizeinbase(n
, 2) + 64);