📄 drand48.html
字号:
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2 Final//EN"><html><head><!-- Copyright 1997 The Open Group, All Rights Reserved --><title>drand48</title></head><body bgcolor=white><center><font size=2>The Single UNIX ® Specification, Version 2<br>Copyright © 1997 The Open Group</font></center><hr size=2 noshade><h4><a name = "tag_000_003_160"> </a>NAME</h4><blockquote>drand48, erand48, jrand48, lcong48, lrand48, mrand48, nrand48, seed48, srand48- generate uniformly distributed pseudo-random numbers</blockquote><h4><a name = "tag_000_003_161"> </a>SYNOPSIS</h4><blockquote><pre><code>#include <<a href="stdlib.h.html">stdlib.h</a>>double drand48(void);double erand48(unsigned short int <i>xsubi</i>[3]);long int jrand48(unsigned short int <i>xsubi</i>[3]);void lcong48(unsigned short int <i>param</i>[7]);long int lrand48(void);long int mrand48(void);long int nrand48(unsigned short int <i>xsubi</i>[3]);unsigned short int *seed48(unsigned short int <i>seed16v</i>[3]);void srand48(long int <i>seedval</i>);</code></pre></blockquote><h4><a name = "tag_000_003_162"> </a>DESCRIPTION</h4><blockquote>This family of functions generates pseudo-random numbers usinga linear congruential algorithm and 48-bit integer arithmetic.<p>The<i>drand48()</i>and<i>erand48()</i>functionsreturn non-negative, double-precision, floating-point values,uniformly distributed over the interval [0.0 , 1.0].<p>The<i>lrand48()</i>and<i>nrand48()</i>functionsreturn non-negative, long integers, uniformly distributed over theinterval [0,2<small><sup>31</sup></small>].<p>The<i>mrand48()</i>and<i>jrand48()</i>functionsreturn signed long integers uniformly distributed over theinterval [-2<sup><small>31</small></sup>,2<sup><small>31</small></sup>].<p>The<i>srand48()</i>,<i>seed48()</i>and<i>lcong48()</i>are initialisation entry points, one of which should be invoked beforeeither<i>drand48()</i>,<i>lrand48()</i>or<i>mrand48()</i>is called.(Although it is not recommended practice,constant default initialiser values will be supplied automatically if<i>drand48()</i>,<i>lrand48()</i>or<i>mrand48()</i>is called without a prior call to an initialisation entry point.)The<i>erand48()</i>,<i>nrand48()</i>and<i>jrand48()</i>functions do not require an initialisation entry point to be called first.<p>All the routines work by generating a sequence of 48-bit integervalues, X<small><sub>i</sub></small> , according to the linear congruential formula:<pre><dl compact><dt> <dd>X<small><sub>n+1</sub></small> = (aX<small><sub>n</sub></small> + c)<small><sub>mod m</sub></small> n>= 0</dl></pre><p>The parameter <i>m</i> = 2<small><sup>48</sup></small>; hence 48-bit integer arithmetic isperformed.Unless<i>lcong48()</i>is invoked, the multiplier value <i>a</i> and the addend value <i>c</i>are given by:<pre><dl compact><dt> <dd><i>a</i> = 5DEECE66D<small><sub>16</sub></small> = 273673163155<small><sub>8</sub></small><i>c</i> = B<small><sub>16</sub></small> = 13<small><sub>8</sub></small></dl></pre><p>The value returned by any of the<i>drand48()</i>,<i>erand48()</i>,<i>jrand48()</i>,<i>lrand48()</i>,<i>mrand48()</i>or<i>nrand48()</i>functionsis computed by first generating the next 48-bit X<small><sub>i</sub></small> in the sequence.Then the appropriate number of bits, according to the type of data itemto be returned, are copied from the high-order (leftmost) bits of X<small><sub>i</sub></small>and transformed into the returned value.<p>The<i>drand48()</i>,<i>lrand48()</i>and<i>mrand48()</i>functionsstore the last 48-bit X<small><sub>i</sub></small> generated in an internal buffer;that is why they must be initialised prior to being invoked.The<i>erand48()</i>,<i>nrand48()</i>and<i>jrand48()</i>functions require the calling program to provide storage for thesuccessive X<small><sub>i</sub></small> values in the arrayspecified as an argument when the functions are invoked.That is why these routines do not have to be initialised; the callingprogram merely has to place the desired initialvalue of X<small><sub>i</sub></small> into the array and pass it as an argument.By using differentarguments,<i>erand48()</i>,<i>nrand48()</i>and<i>jrand48()</i>allow separate modules of a large program to generate several<i>independent</i>streams of pseudo-random numbers, that is the sequence of numbersin each stream will<i>not</i>depend upon how many times the routines are called to generatenumbers for the other streams.<p>The initialiser function<i>srand48()</i>sets the high-order 32 bits of X<small><sub>i</sub></small> to the low-order 32bits contained in its argument.The low-order 16 bits of X<small><sub>i</sub></small> are set to the arbitrary value330E<small><sub>16</sub></small> .<p>The initialiser function<i>seed48()</i>sets the value of X<small><sub>i</sub></small> to the 48-bit value specified in theargument array.The low-order 16 bits of X<small><sub>i</sub></small> are set to the low-order 16 bits of<i>seed16v[0]</i>.The mid-order 16 bits of X<small><sub>i</sub></small> are set to the low-order 16 bits of<i>seed16v[1]</i>.The high-order 16 bits of X<small><sub>i</sub></small> are set to the low-order 16 bits of<i>seed16v[2]</i>.In addition, the previous value of X<small><sub>i</sub></small> is copied into a 48-bitinternal buffer, used only by<i>seed48()</i>,and a pointer to this buffer is the value returned by<i>seed48()</i>.This returned pointer, which can just be ignored if not needed, is usefulif a program is to be restarted from a given point at some future time- use the pointer to get at and store the last X<small><sub>i</sub></small> value, andthen use this value to re-initialise via<i>seed48()</i>when the program is restarted.<p>The initialiser function<i>lcong48()</i>allows the user to specify the initial X<small><sub>i</sub></small>, the multipliervalue <i>a</i>, and the addend value <i>c</i>.Argument array elements<i>param</i>[<i>0-2</i>]specify X<small><sub>i</sub></small>,<i>param</i>[<i>3-5</i>]specify the multiplier <i>a</i>, and<i>param</i>[<i>6</i>]specifies the 16-bit addend <i>c</i>.After<i>lcong48()</i>is called, a subsequent call to either<i>srand48()</i>or<i>seed48()</i>will restore the standard multiplier and addend values, <i>a</i> and <i>c</i>,specified above.<p>The<i>drand48()</i>,<i>lrand48()</i>and<i>mrand48()</i>interfaces need not be reentrant.</blockquote><h4><a name = "tag_000_003_163"> </a>RETURN VALUE</h4><blockquote>As described in the DESCRIPTION above.</blockquote><h4><a name = "tag_000_003_164"> </a>ERRORS</h4><blockquote>No errors are defined.</blockquote><h4><a name = "tag_000_003_165"> </a>EXAMPLES</h4><blockquote>None.</blockquote><h4><a name = "tag_000_003_166"> </a>APPLICATION USAGE</h4><blockquote>None.</blockquote><h4><a name = "tag_000_003_167"> </a>FUTURE DIRECTIONS</h4><blockquote>None.</blockquote><h4><a name = "tag_000_003_168"> </a>SEE ALSO</h4><blockquote><i><a href="rand.html">rand()</a></i>,<i><a href="stdlib.h.html"><stdlib.h></a></i>.</blockquote><h4>DERIVATION</h4><blockquote>Derived from Issue 1 of the SVID.</blockquote><hr size=2 noshade><center><font size=2>UNIX ® is a registered Trademark of The Open Group.<br>Copyright © 1997 The Open Group<br> [ <a href="../index.html">Main Index</a> | <a href="../xshix.html">XSH</a> | <a href="../xcuix.html">XCU</a> | <a href="../xbdix.html">XBD</a> | <a href="../cursesix.html">XCURSES</a> | <a href="../xnsix.html">XNS</a> ]</font></center><hr size=2 noshade></body></html>⌨️ 快捷键说明
复制代码
Ctrl + C
搜索代码
Ctrl + F
全屏模式
F11
切换主题
Ctrl + Shift + D
显示快捷键
?
增大字号
Ctrl + =
减小字号
Ctrl + -