]>
Commit | Line | Data |
---|---|---|
1 | .Dd September 6, 2001 | |
2 | .Dt RANDOM 4 | |
3 | .Os Darwin | |
4 | .Sh NAME | |
5 | .Nm random | |
6 | , | |
7 | .Nm urandom | |
8 | .Nd random data source devices. | |
9 | .Sh SYNOPSIS | |
10 | .Cd "pseudo-device random" | |
11 | .Sh DESCRIPTION | |
12 | The | |
13 | .Nm | |
14 | device produces uniformly distributed random byte values | |
15 | of potentially high quality. | |
16 | .Pp | |
17 | To obtain random bytes, open | |
18 | .Nm /dev/random | |
19 | for reading and read from it. | |
20 | .Pp | |
21 | The same random data is also available from | |
22 | .Xr getentropy 2 . | |
23 | Using the | |
24 | .Xr getentropy 2 | |
25 | system call interface will provide resiliency to file descriptor exhaustion, chroot, or sandboxing which can make | |
26 | .Nm /dev/random | |
27 | unavailable. Additionally, the | |
28 | .Xr arc4random 3 | |
29 | API provides a fast userspace random number generator built on the | |
30 | .Nm | |
31 | data source and is preferred over directly accessing the system's random device. | |
32 | .Pp | |
33 | .Nm /dev/urandom | |
34 | is a compatibility nod to Linux. On Linux, | |
35 | .Nm /dev/urandom | |
36 | will produce lower quality output if the entropy pool drains, while | |
37 | .Nm /dev/random | |
38 | will prefer to block and wait for additional entropy to be collected. | |
39 | With Yarrow, this choice and distinction is not necessary, and | |
40 | the two devices behave identically. You may use either. | |
41 | .Pp | |
42 | The | |
43 | .Nm | |
44 | device implements the | |
45 | .Nm Yarrow | |
46 | pseudo random number generator algorithm and maintains its entropy pool. | |
47 | The kernel automatically seeds the algorithm with additional entropy during normal execution. | |
48 | .Sh FILES | |
49 | .Bl -tag -width /dev/urandom -compact | |
50 | .It Pa /dev/random | |
51 | .It Pa /dev/urandom | |
52 | .El | |
53 | .Sh HISTORY | |
54 | A | |
55 | .Nm | |
56 | device appeared in the Linux operating system. |