1rand(3)                             OpenSSL                            rand(3)
2
3
4

NAME

6       rand - pseudo-random number generator
7

SYNOPSIS

9        #include <openssl/rand.h>
10
11        int  RAND_set_rand_engine(ENGINE *engine);
12
13        int  RAND_bytes(unsigned char *buf, int num);
14        int  RAND_pseudo_bytes(unsigned char *buf, int num);
15
16        void RAND_seed(const void *buf, int num);
17        void RAND_add(const void *buf, int num, double entropy);
18        int  RAND_status(void);
19
20        int  RAND_load_file(const char *file, long max_bytes);
21        int  RAND_write_file(const char *file);
22        const char *RAND_file_name(char *file, size_t num);
23
24        int  RAND_egd(const char *path);
25
26        void RAND_set_rand_method(const RAND_METHOD *meth);
27        const RAND_METHOD *RAND_get_rand_method(void);
28        RAND_METHOD *RAND_SSLeay(void);
29
30        void RAND_cleanup(void);
31
32        /* For Win32 only */
33        void RAND_screen(void);
34        int RAND_event(UINT, WPARAM, LPARAM);
35

DESCRIPTION

37       Since the introduction of the ENGINE API, the recommended way of
38       controlling default implementations is by using the ENGINE API
39       functions. The default RAND_METHOD, as set by RAND_set_rand_method()
40       and returned by RAND_get_rand_method(), is only used if no ENGINE has
41       been set as the default "rand" implementation. Hence, these two
42       functions are no longer the recommended way to control defaults.
43
44       If an alternative RAND_METHOD implementation is being used (either set
45       directly or as provided by an ENGINE module), then it is entirely
46       responsible for the generation and management of a cryptographically
47       secure PRNG stream. The mechanisms described below relate solely to the
48       software PRNG implementation built in to OpenSSL and used by default.
49
50       These functions implement a cryptographically secure pseudo-random
51       number generator (PRNG). It is used by other library functions for
52       example to generate random keys, and applications can use it when they
53       need randomness.
54
55       A cryptographic PRNG must be seeded with unpredictable data such as
56       mouse movements or keys pressed at random by the user. This is
57       described in RAND_add(3). Its state can be saved in a seed file (see
58       RAND_load_file(3)) to avoid having to go through the seeding process
59       whenever the application is started.
60
61       RAND_bytes(3) describes how to obtain random data from the PRNG.
62

INTERNALS

64       The RAND_SSLeay() method implements a PRNG based on a cryptographic
65       hash function.
66
67       The following description of its design is based on the SSLeay
68       documentation:
69
70       First up I will state the things I believe I need for a good RNG.
71
72       1.  A good hashing algorithm to mix things up and to convert the RNG
73           'state' to random numbers.
74
75       2.  An initial source of random 'state'.
76
77       3.  The state should be very large.  If the RNG is being used to
78           generate 4096 bit RSA keys, 2 2048 bit random strings are required
79           (at a minimum).  If your RNG state only has 128 bits, you are
80           obviously limiting the search space to 128 bits, not 2048.  I'm
81           probably getting a little carried away on this last point but it
82           does indicate that it may not be a bad idea to keep quite a lot of
83           RNG state.  It should be easier to break a cipher than guess the
84           RNG seed data.
85
86       4.  Any RNG seed data should influence all subsequent random numbers
87           generated.  This implies that any random seed data entered will
88           have an influence on all subsequent random numbers generated.
89
90       5.  When using data to seed the RNG state, the data used should not be
91           extractable from the RNG state.  I believe this should be a
92           requirement because one possible source of 'secret' semi random
93           data would be a private key or a password.  This data must not be
94           disclosed by either subsequent random numbers or a 'core' dump left
95           by a program crash.
96
97       6.  Given the same initial 'state', 2 systems should deviate in their
98           RNG state (and hence the random numbers generated) over time if at
99           all possible.
100
101       7.  Given the random number output stream, it should not be possible to
102           determine the RNG state or the next random number.
103
104       The algorithm is as follows.
105
106       There is global state made up of a 1023 byte buffer (the 'state'), a
107       working hash value ('md'), and a counter ('count').
108
109       Whenever seed data is added, it is inserted into the 'state' as
110       follows.
111
112       The input is chopped up into units of 20 bytes (or less for the last
113       block).  Each of these blocks is run through the hash function as
114       follows:  The data passed to the hash function is the current 'md', the
115       same number of bytes from the 'state' (the location determined by in
116       incremented looping index) as the current 'block', the new key data
117       'block', and 'count' (which is incremented after each use).  The result
118       of this is kept in 'md' and also xored into the 'state' at the same
119       locations that were used as input into the hash function. I believe
120       this system addresses points 1 (hash function; currently SHA-1), 3 (the
121       'state'), 4 (via the 'md'), 5 (by the use of a hash function and xor).
122
123       When bytes are extracted from the RNG, the following process is used.
124       For each group of 10 bytes (or less), we do the following:
125
126       Input into the hash function the local 'md' (which is initialized from
127       the global 'md' before any bytes are generated), the bytes that are to
128       be overwritten by the random bytes, and bytes from the 'state'
129       (incrementing looping index). From this digest output (which is kept in
130       'md'), the top (up to) 10 bytes are returned to the caller and the
131       bottom 10 bytes are xored into the 'state'.
132
133       Finally, after we have finished 'num' random bytes for the caller,
134       'count' (which is incremented) and the local and global 'md' are fed
135       into the hash function and the results are kept in the global 'md'.
136
137       I believe the above addressed points 1 (use of SHA-1), 6 (by hashing
138       into the 'state' the 'old' data from the caller that is about to be
139       overwritten) and 7 (by not using the 10 bytes given to the caller to
140       update the 'state', but they are used to update 'md').
141
142       So of the points raised, only 2 is not addressed (but see RAND_add(3)).
143

SEE ALSO

145       BN_rand(3), RAND_add(3), RAND_load_file(3), RAND_egd(3), RAND_bytes(3),
146       RAND_set_rand_method(3), RAND_cleanup(3)
147
148
149
1501.0.2o                            2019-09-10                           rand(3)
Impressum