|
cannam@95
|
1 <html lang="en">
|
|
cannam@95
|
2 <head>
|
|
cannam@95
|
3 <title>New-array Execute Functions - 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="FFTW-Reference.html#FFTW-Reference" title="FFTW Reference">
|
|
cannam@95
|
9 <link rel="prev" href="Guru-Interface.html#Guru-Interface" title="Guru Interface">
|
|
cannam@95
|
10 <link rel="next" href="Wisdom.html#Wisdom" title="Wisdom">
|
|
cannam@95
|
11 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
|
|
cannam@95
|
12 <!--
|
|
cannam@95
|
13 This manual is for FFTW
|
|
cannam@95
|
14 (version 3.3.3, 25 November 2012).
|
|
cannam@95
|
15
|
|
cannam@95
|
16 Copyright (C) 2003 Matteo Frigo.
|
|
cannam@95
|
17
|
|
cannam@95
|
18 Copyright (C) 2003 Massachusetts Institute of Technology.
|
|
cannam@95
|
19
|
|
cannam@95
|
20 Permission is granted to make and distribute verbatim copies of
|
|
cannam@95
|
21 this manual provided the copyright notice and this permission
|
|
cannam@95
|
22 notice are preserved on all copies.
|
|
cannam@95
|
23
|
|
cannam@95
|
24 Permission is granted to copy and distribute modified versions of
|
|
cannam@95
|
25 this manual under the conditions for verbatim copying, provided
|
|
cannam@95
|
26 that the entire resulting derived work is distributed under the
|
|
cannam@95
|
27 terms of a permission notice identical to this one.
|
|
cannam@95
|
28
|
|
cannam@95
|
29 Permission is granted to copy and distribute translations of this
|
|
cannam@95
|
30 manual into another language, under the above conditions for
|
|
cannam@95
|
31 modified versions, except that this permission notice may be
|
|
cannam@95
|
32 stated in a translation approved by the Free Software Foundation.
|
|
cannam@95
|
33 -->
|
|
cannam@95
|
34 <meta http-equiv="Content-Style-Type" content="text/css">
|
|
cannam@95
|
35 <style type="text/css"><!--
|
|
cannam@95
|
36 pre.display { font-family:inherit }
|
|
cannam@95
|
37 pre.format { font-family:inherit }
|
|
cannam@95
|
38 pre.smalldisplay { font-family:inherit; font-size:smaller }
|
|
cannam@95
|
39 pre.smallformat { font-family:inherit; font-size:smaller }
|
|
cannam@95
|
40 pre.smallexample { font-size:smaller }
|
|
cannam@95
|
41 pre.smalllisp { font-size:smaller }
|
|
cannam@95
|
42 span.sc { font-variant:small-caps }
|
|
cannam@95
|
43 span.roman { font-family:serif; font-weight:normal; }
|
|
cannam@95
|
44 span.sansserif { font-family:sans-serif; font-weight:normal; }
|
|
cannam@95
|
45 --></style>
|
|
cannam@95
|
46 </head>
|
|
cannam@95
|
47 <body>
|
|
cannam@95
|
48 <div class="node">
|
|
cannam@95
|
49 <a name="New-array-Execute-Functions"></a>
|
|
cannam@95
|
50 <a name="New_002darray-Execute-Functions"></a>
|
|
cannam@95
|
51 <p>
|
|
cannam@95
|
52 Next: <a rel="next" accesskey="n" href="Wisdom.html#Wisdom">Wisdom</a>,
|
|
cannam@95
|
53 Previous: <a rel="previous" accesskey="p" href="Guru-Interface.html#Guru-Interface">Guru Interface</a>,
|
|
cannam@95
|
54 Up: <a rel="up" accesskey="u" href="FFTW-Reference.html#FFTW-Reference">FFTW Reference</a>
|
|
cannam@95
|
55 <hr>
|
|
cannam@95
|
56 </div>
|
|
cannam@95
|
57
|
|
cannam@95
|
58 <h3 class="section">4.6 New-array Execute Functions</h3>
|
|
cannam@95
|
59
|
|
cannam@95
|
60 <p><a name="index-execute-264"></a><a name="index-new_002darray-execution-265"></a>
|
|
cannam@95
|
61 Normally, one executes a plan for the arrays with which the plan was
|
|
cannam@95
|
62 created, by calling <code>fftw_execute(plan)</code> as described in <a href="Using-Plans.html#Using-Plans">Using Plans</a>.
|
|
cannam@95
|
63 <a name="index-fftw_005fexecute-266"></a>However, it is possible for sophisticated users to apply a given plan
|
|
cannam@95
|
64 to a <em>different</em> array using the “new-array execute” functions
|
|
cannam@95
|
65 detailed below, provided that the following conditions are met:
|
|
cannam@95
|
66
|
|
cannam@95
|
67 <ul>
|
|
cannam@95
|
68 <li>The array size, strides, etcetera are the same (since those are set by
|
|
cannam@95
|
69 the plan).
|
|
cannam@95
|
70
|
|
cannam@95
|
71 <li>The input and output arrays are the same (in-place) or different
|
|
cannam@95
|
72 (out-of-place) if the plan was originally created to be in-place or
|
|
cannam@95
|
73 out-of-place, respectively.
|
|
cannam@95
|
74
|
|
cannam@95
|
75 <li>For split arrays, the separations between the real and imaginary
|
|
cannam@95
|
76 parts, <code>ii-ri</code> and <code>io-ro</code>, are the same as they were for
|
|
cannam@95
|
77 the input and output arrays when the plan was created. (This
|
|
cannam@95
|
78 condition is automatically satisfied for interleaved arrays.)
|
|
cannam@95
|
79
|
|
cannam@95
|
80 <li>The <dfn>alignment</dfn> of the new input/output arrays is the same as that
|
|
cannam@95
|
81 of the input/output arrays when the plan was created, unless the plan
|
|
cannam@95
|
82 was created with the <code>FFTW_UNALIGNED</code> flag.
|
|
cannam@95
|
83 <a name="index-FFTW_005fUNALIGNED-267"></a>Here, the alignment is a platform-dependent quantity (for example, it is
|
|
cannam@95
|
84 the address modulo 16 if SSE SIMD instructions are used, but the address
|
|
cannam@95
|
85 modulo 4 for non-SIMD single-precision FFTW on the same machine). In
|
|
cannam@95
|
86 general, only arrays allocated with <code>fftw_malloc</code> are guaranteed to
|
|
cannam@95
|
87 be equally aligned (see <a href="SIMD-alignment-and-fftw_005fmalloc.html#SIMD-alignment-and-fftw_005fmalloc">SIMD alignment and fftw_malloc</a>).
|
|
cannam@95
|
88
|
|
cannam@95
|
89 </ul>
|
|
cannam@95
|
90
|
|
cannam@95
|
91 <p><a name="index-alignment-268"></a>The alignment issue is especially critical, because if you don't use
|
|
cannam@95
|
92 <code>fftw_malloc</code> then you may have little control over the alignment
|
|
cannam@95
|
93 of arrays in memory. For example, neither the C++ <code>new</code> function
|
|
cannam@95
|
94 nor the Fortran <code>allocate</code> statement provide strong enough
|
|
cannam@95
|
95 guarantees about data alignment. If you don't use <code>fftw_malloc</code>,
|
|
cannam@95
|
96 therefore, you probably have to use <code>FFTW_UNALIGNED</code> (which
|
|
cannam@95
|
97 disables most SIMD support). If possible, it is probably better for
|
|
cannam@95
|
98 you to simply create multiple plans (creating a new plan is quick once
|
|
cannam@95
|
99 one exists for a given size), or better yet re-use the same array for
|
|
cannam@95
|
100 your transforms.
|
|
cannam@95
|
101
|
|
cannam@95
|
102 <p>If you are tempted to use the new-array execute interface because you
|
|
cannam@95
|
103 want to transform a known bunch of arrays of the same size, you should
|
|
cannam@95
|
104 probably go use the advanced interface instead (see <a href="Advanced-Interface.html#Advanced-Interface">Advanced Interface</a>)).
|
|
cannam@95
|
105
|
|
cannam@95
|
106 <p>The new-array execute functions are:
|
|
cannam@95
|
107
|
|
cannam@95
|
108 <pre class="example"> void fftw_execute_dft(
|
|
cannam@95
|
109 const fftw_plan p,
|
|
cannam@95
|
110 fftw_complex *in, fftw_complex *out);
|
|
cannam@95
|
111
|
|
cannam@95
|
112 void fftw_execute_split_dft(
|
|
cannam@95
|
113 const fftw_plan p,
|
|
cannam@95
|
114 double *ri, double *ii, double *ro, double *io);
|
|
cannam@95
|
115
|
|
cannam@95
|
116 void fftw_execute_dft_r2c(
|
|
cannam@95
|
117 const fftw_plan p,
|
|
cannam@95
|
118 double *in, fftw_complex *out);
|
|
cannam@95
|
119
|
|
cannam@95
|
120 void fftw_execute_split_dft_r2c(
|
|
cannam@95
|
121 const fftw_plan p,
|
|
cannam@95
|
122 double *in, double *ro, double *io);
|
|
cannam@95
|
123
|
|
cannam@95
|
124 void fftw_execute_dft_c2r(
|
|
cannam@95
|
125 const fftw_plan p,
|
|
cannam@95
|
126 fftw_complex *in, double *out);
|
|
cannam@95
|
127
|
|
cannam@95
|
128 void fftw_execute_split_dft_c2r(
|
|
cannam@95
|
129 const fftw_plan p,
|
|
cannam@95
|
130 double *ri, double *ii, double *out);
|
|
cannam@95
|
131
|
|
cannam@95
|
132 void fftw_execute_r2r(
|
|
cannam@95
|
133 const fftw_plan p,
|
|
cannam@95
|
134 double *in, double *out);
|
|
cannam@95
|
135 </pre>
|
|
cannam@95
|
136 <p><a name="index-fftw_005fexecute_005fdft-269"></a><a name="index-fftw_005fexecute_005fsplit_005fdft-270"></a><a name="index-fftw_005fexecute_005fdft_005fr2c-271"></a><a name="index-fftw_005fexecute_005fsplit_005fdft_005fr2c-272"></a><a name="index-fftw_005fexecute_005fdft_005fc2r-273"></a><a name="index-fftw_005fexecute_005fsplit_005fdft_005fc2r-274"></a><a name="index-fftw_005fexecute_005fr2r-275"></a>
|
|
cannam@95
|
137 These execute the <code>plan</code> to compute the corresponding transform on
|
|
cannam@95
|
138 the input/output arrays specified by the subsequent arguments. The
|
|
cannam@95
|
139 input/output array arguments have the same meanings as the ones passed
|
|
cannam@95
|
140 to the guru planner routines in the preceding sections. The <code>plan</code>
|
|
cannam@95
|
141 is not modified, and these routines can be called as many times as
|
|
cannam@95
|
142 desired, or intermixed with calls to the ordinary <code>fftw_execute</code>.
|
|
cannam@95
|
143
|
|
cannam@95
|
144 <p>The <code>plan</code> <em>must</em> have been created for the transform type
|
|
cannam@95
|
145 corresponding to the execute function, e.g. it must be a complex-DFT
|
|
cannam@95
|
146 plan for <code>fftw_execute_dft</code>. Any of the planner routines for that
|
|
cannam@95
|
147 transform type, from the basic to the guru interface, could have been
|
|
cannam@95
|
148 used to create the plan, however.
|
|
cannam@95
|
149
|
|
cannam@95
|
150 <!-- -->
|
|
cannam@95
|
151 </body></html>
|
|
cannam@95
|
152
|
