random.3: Some minor improvements to wording/clarity

This commit is contained in:
Conrad Meyer 2020-01-20 23:44:10 +00:00
parent 6ee287b9c2
commit 825a73c36f
Notes: svn2git 2020-12-20 02:59:44 +00:00
svn path=/head/; revision=356935

View File

@ -28,7 +28,7 @@
.\" @(#)random.3 8.1 (Berkeley) 6/4/93
.\" $FreeBSD$
.\"
.Dd April 22, 2019
.Dd January 20, 2020
.Dt RANDOM 3
.Os
.Sh NAME
@ -60,7 +60,7 @@ Applications which require unpredictable random numbers should use
instead.
.Ef
.Pp
The
Unless initialized with less than 32 bytes of state, the
.Fn random
function
uses a non-linear additive feedback random number generator employing a
@ -72,53 +72,51 @@ The period of this random number generator is very large, approximately
.if t 16\(mu(2\u\s731\s10\d\(mi1).
.if n 16*((2**31)\(mi1).
.Pp
If initialized with less than 32 bytes of state,
.Fn random
uses the same poor-quality Park-Miller LCG as
.Xr rand 3 .
.Pp
The
.Fn random
and
.Fn srandom
functions have (almost) the same calling sequence and initialization properties as the
functions are analagous to
.Xr rand 3
and
.Xr srand 3
functions.
.Xr srand 3 .
The difference is that
.Xr rand 3
produces a much less random sequence \(em in fact, the low dozen bits
generated by rand go through a cyclic pattern.
All the bits generated by
.Fn random
are usable.
For example,
.Sq Li random()&01
will produce a random binary
value.
is a worse pseudo-random number generator.
.Pp
Like
.Xr rand 3 ,
.Fn random
will by default produce a sequence of numbers that can be duplicated
by calling
.Fn srandom
with
.Ql 1
as the seed.
is implicitly initialized as if
.Fn srandom "1"
had been invoked explicitly.
.Pp
The
.Fn srandomdev
routine initializes a state array using
pseudo-random numbers obtained from the kernel.
Note that this particular seeding
procedure can generate states which are impossible to reproduce by
calling
.Fn srandom
with any value, since the succeeding terms in the
state buffer are no longer derived from the LC algorithm applied to
a fixed seed.
routine initializes the state array using random numbers obtained from the
kernel.
This can generate states which are impossible to reproduce by calling
.Fn srandom ,
because the succeeding terms in the state buffer are no longer derived from the
Park-Miller LCG algorithm applied to a fixed seed.
.Pp
The
.Fn initstate
routine allows a state array, passed in as an argument, to be initialized
for future use.
routine initializes the provided state array of
.Vt uint32_t
values and uses it in future
.Fn random
invocations.
(Despite the
.Vt char *
type of
.Fa state ,
the underlying object must be a naturally aligned array of 32-bit values.)
The size of the state array (in bytes) is used by
.Fn initstate
to decide how sophisticated a random number generator it should use \(em the
@ -127,26 +125,21 @@ more state, the better the random numbers will be.
8, 32, 64, 128, and 256 bytes; other amounts will be rounded down to
the nearest known amount.
Using less than 8 bytes will cause an error.)
The seed for the initialization (which specifies a starting point for
the random number sequence, and provides for restarting at the same
point) is also an argument.
The
.Fa seed
is used as in
.Fn srandom .
The
.Fn initstate
function
returns a pointer to the previous state information array.
.Pp
Once a state has been initialized, the
.Fn setstate
routine provides for rapid switching between states.
The
.Fn setstate
function
returns a pointer to the previous state array; its
argument state array is used for further random number generation
until the next call to
.Fn initstate
or
.Fn setstate .
routine switches
.Fn random
to using the provided state.
It returns a pointer to the previous state.
.Pp
Once a state array has been initialized, it may be restarted at a
different point either by calling