|
cannam@95
|
1 <html lang="en">
|
|
cannam@95
|
2 <head>
|
|
cannam@95
|
3 <title>64-bit Guru Interface - FFTW 3.3.3</title>
|
|
cannam@95
|
4 <meta http-equiv="Content-Type" content="text/html">
|
|
cannam@95
|
5 <meta name="description" content="FFTW 3.3.3">
|
|
cannam@95
|
6 <meta name="generator" content="makeinfo 4.13">
|
|
cannam@95
|
7 <link title="Top" rel="start" href="index.html#Top">
|
|
cannam@95
|
8 <link rel="up" href="Guru-Interface.html#Guru-Interface" title="Guru Interface">
|
|
cannam@95
|
9 <link rel="prev" href="Guru-Real_002dto_002dreal-Transforms.html#Guru-Real_002dto_002dreal-Transforms" title="Guru Real-to-real Transforms">
|
|
cannam@95
|
10 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
|
|
cannam@95
|
11 <!--
|
|
cannam@95
|
12 This manual is for FFTW
|
|
cannam@95
|
13 (version 3.3.3, 25 November 2012).
|
|
cannam@95
|
14
|
|
cannam@95
|
15 Copyright (C) 2003 Matteo Frigo.
|
|
cannam@95
|
16
|
|
cannam@95
|
17 Copyright (C) 2003 Massachusetts Institute of Technology.
|
|
cannam@95
|
18
|
|
cannam@95
|
19 Permission is granted to make and distribute verbatim copies of
|
|
cannam@95
|
20 this manual provided the copyright notice and this permission
|
|
cannam@95
|
21 notice are preserved on all copies.
|
|
cannam@95
|
22
|
|
cannam@95
|
23 Permission is granted to copy and distribute modified versions of
|
|
cannam@95
|
24 this manual under the conditions for verbatim copying, provided
|
|
cannam@95
|
25 that the entire resulting derived work is distributed under the
|
|
cannam@95
|
26 terms of a permission notice identical to this one.
|
|
cannam@95
|
27
|
|
cannam@95
|
28 Permission is granted to copy and distribute translations of this
|
|
cannam@95
|
29 manual into another language, under the above conditions for
|
|
cannam@95
|
30 modified versions, except that this permission notice may be
|
|
cannam@95
|
31 stated in a translation approved by the Free Software Foundation.
|
|
cannam@95
|
32 -->
|
|
cannam@95
|
33 <meta http-equiv="Content-Style-Type" content="text/css">
|
|
cannam@95
|
34 <style type="text/css"><!--
|
|
cannam@95
|
35 pre.display { font-family:inherit }
|
|
cannam@95
|
36 pre.format { font-family:inherit }
|
|
cannam@95
|
37 pre.smalldisplay { font-family:inherit; font-size:smaller }
|
|
cannam@95
|
38 pre.smallformat { font-family:inherit; font-size:smaller }
|
|
cannam@95
|
39 pre.smallexample { font-size:smaller }
|
|
cannam@95
|
40 pre.smalllisp { font-size:smaller }
|
|
cannam@95
|
41 span.sc { font-variant:small-caps }
|
|
cannam@95
|
42 span.roman { font-family:serif; font-weight:normal; }
|
|
cannam@95
|
43 span.sansserif { font-family:sans-serif; font-weight:normal; }
|
|
cannam@95
|
44 --></style>
|
|
cannam@95
|
45 </head>
|
|
cannam@95
|
46 <body>
|
|
cannam@95
|
47 <div class="node">
|
|
cannam@95
|
48 <a name="64-bit-Guru-Interface"></a>
|
|
cannam@95
|
49 <a name="g_t64_002dbit-Guru-Interface"></a>
|
|
cannam@95
|
50 <p>
|
|
cannam@95
|
51 Previous: <a rel="previous" accesskey="p" href="Guru-Real_002dto_002dreal-Transforms.html#Guru-Real_002dto_002dreal-Transforms">Guru Real-to-real Transforms</a>,
|
|
cannam@95
|
52 Up: <a rel="up" accesskey="u" href="Guru-Interface.html#Guru-Interface">Guru Interface</a>
|
|
cannam@95
|
53 <hr>
|
|
cannam@95
|
54 </div>
|
|
cannam@95
|
55
|
|
cannam@95
|
56 <h4 class="subsection">4.5.6 64-bit Guru Interface</h4>
|
|
cannam@95
|
57
|
|
cannam@95
|
58 <p><a name="index-g_t64_002dbit-architecture-260"></a>
|
|
cannam@95
|
59 When compiled in 64-bit mode on a 64-bit architecture (where addresses
|
|
cannam@95
|
60 are 64 bits wide), FFTW uses 64-bit quantities internally for all
|
|
cannam@95
|
61 transform sizes, strides, and so on—you don't have to do anything
|
|
cannam@95
|
62 special to exploit this. However, in the ordinary FFTW interfaces,
|
|
cannam@95
|
63 you specify the transform size by an <code>int</code> quantity, which is
|
|
cannam@95
|
64 normally only 32 bits wide. This means that, even though FFTW is
|
|
cannam@95
|
65 using 64-bit sizes internally, you cannot specify a single transform
|
|
cannam@95
|
66 dimension larger than
|
|
cannam@95
|
67 2<sup><small>31</small></sup>−1numbers.
|
|
cannam@95
|
68
|
|
cannam@95
|
69 <p>We expect that few users will require transforms larger than this, but,
|
|
cannam@95
|
70 for those who do, we provide a 64-bit version of the guru interface in
|
|
cannam@95
|
71 which all sizes are specified as integers of type <code>ptrdiff_t</code>
|
|
cannam@95
|
72 instead of <code>int</code>. (<code>ptrdiff_t</code> is a signed integer type
|
|
cannam@95
|
73 defined by the C standard to be wide enough to represent address
|
|
cannam@95
|
74 differences, and thus must be at least 64 bits wide on a 64-bit
|
|
cannam@95
|
75 machine.) We stress that there is <em>no performance advantage</em> to
|
|
cannam@95
|
76 using this interface—the same internal FFTW code is employed
|
|
cannam@95
|
77 regardless—and it is only necessary if you want to specify very
|
|
cannam@95
|
78 large transform sizes.
|
|
cannam@95
|
79 <a name="index-ptrdiff_005ft-261"></a>
|
|
cannam@95
|
80
|
|
cannam@95
|
81 <p>In particular, the 64-bit guru interface is a set of planner routines
|
|
cannam@95
|
82 that are exactly the same as the guru planner routines, except that
|
|
cannam@95
|
83 they are named with ‘<samp><span class="samp">guru64</span></samp>’ instead of ‘<samp><span class="samp">guru</span></samp>’ and they take
|
|
cannam@95
|
84 arguments of type <code>fftw_iodim64</code> instead of <code>fftw_iodim</code>.
|
|
cannam@95
|
85 For example, instead of <code>fftw_plan_guru_dft</code>, we have
|
|
cannam@95
|
86 <code>fftw_plan_guru64_dft</code>.
|
|
cannam@95
|
87
|
|
cannam@95
|
88 <pre class="example"> fftw_plan fftw_plan_guru64_dft(
|
|
cannam@95
|
89 int rank, const fftw_iodim64 *dims,
|
|
cannam@95
|
90 int howmany_rank, const fftw_iodim64 *howmany_dims,
|
|
cannam@95
|
91 fftw_complex *in, fftw_complex *out,
|
|
cannam@95
|
92 int sign, unsigned flags);
|
|
cannam@95
|
93 </pre>
|
|
cannam@95
|
94 <p><a name="index-fftw_005fplan_005fguru64_005fdft-262"></a>
|
|
cannam@95
|
95 The <code>fftw_iodim64</code> type is similar to <code>fftw_iodim</code>, with the
|
|
cannam@95
|
96 same interpretation, except that it uses type <code>ptrdiff_t</code> instead
|
|
cannam@95
|
97 of type <code>int</code>.
|
|
cannam@95
|
98
|
|
cannam@95
|
99 <pre class="example"> typedef struct {
|
|
cannam@95
|
100 ptrdiff_t n;
|
|
cannam@95
|
101 ptrdiff_t is;
|
|
cannam@95
|
102 ptrdiff_t os;
|
|
cannam@95
|
103 } fftw_iodim64;
|
|
cannam@95
|
104 </pre>
|
|
cannam@95
|
105 <p><a name="index-fftw_005fiodim64-263"></a>
|
|
cannam@95
|
106 Every other ‘<samp><span class="samp">fftw_plan_guru</span></samp>’ function also has a
|
|
cannam@95
|
107 ‘<samp><span class="samp">fftw_plan_guru64</span></samp>’ equivalent, but we do not repeat their
|
|
cannam@95
|
108 documentation here since they are identical to the 32-bit versions
|
|
cannam@95
|
109 except as noted above.
|
|
cannam@95
|
110
|
|
cannam@95
|
111 <!-- -->
|
|
cannam@95
|
112 </body></html>
|
|
cannam@95
|
113
|
