|
cannam@127
|
1 <!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
|
|
cannam@127
|
2 <html>
|
|
cannam@127
|
3 <!-- This manual is for FFTW
|
|
cannam@127
|
4 (version 3.3.5, 30 July 2016).
|
|
cannam@127
|
5
|
|
cannam@127
|
6 Copyright (C) 2003 Matteo Frigo.
|
|
cannam@127
|
7
|
|
cannam@127
|
8 Copyright (C) 2003 Massachusetts Institute of Technology.
|
|
cannam@127
|
9
|
|
cannam@127
|
10 Permission is granted to make and distribute verbatim copies of this
|
|
cannam@127
|
11 manual provided the copyright notice and this permission notice are
|
|
cannam@127
|
12 preserved on all copies.
|
|
cannam@127
|
13
|
|
cannam@127
|
14 Permission is granted to copy and distribute modified versions of this
|
|
cannam@127
|
15 manual under the conditions for verbatim copying, provided that the
|
|
cannam@127
|
16 entire resulting derived work is distributed under the terms of a
|
|
cannam@127
|
17 permission notice identical to this one.
|
|
cannam@127
|
18
|
|
cannam@127
|
19 Permission is granted to copy and distribute translations of this manual
|
|
cannam@127
|
20 into another language, under the above conditions for modified versions,
|
|
cannam@127
|
21 except that this permission notice may be stated in a translation
|
|
cannam@127
|
22 approved by the Free Software Foundation. -->
|
|
cannam@127
|
23 <!-- Created by GNU Texinfo 5.2, http://www.gnu.org/software/texinfo/ -->
|
|
cannam@127
|
24 <head>
|
|
cannam@127
|
25 <title>FFTW 3.3.5: MPI Plan Creation</title>
|
|
cannam@127
|
26
|
|
cannam@127
|
27 <meta name="description" content="FFTW 3.3.5: MPI Plan Creation">
|
|
cannam@127
|
28 <meta name="keywords" content="FFTW 3.3.5: MPI Plan Creation">
|
|
cannam@127
|
29 <meta name="resource-type" content="document">
|
|
cannam@127
|
30 <meta name="distribution" content="global">
|
|
cannam@127
|
31 <meta name="Generator" content="makeinfo">
|
|
cannam@127
|
32 <meta http-equiv="Content-Type" content="text/html; charset=utf-8">
|
|
cannam@127
|
33 <link href="index.html#Top" rel="start" title="Top">
|
|
cannam@127
|
34 <link href="Concept-Index.html#Concept-Index" rel="index" title="Concept Index">
|
|
cannam@127
|
35 <link href="index.html#SEC_Contents" rel="contents" title="Table of Contents">
|
|
cannam@127
|
36 <link href="FFTW-MPI-Reference.html#FFTW-MPI-Reference" rel="up" title="FFTW MPI Reference">
|
|
cannam@127
|
37 <link href="MPI-Wisdom-Communication.html#MPI-Wisdom-Communication" rel="next" title="MPI Wisdom Communication">
|
|
cannam@127
|
38 <link href="MPI-Data-Distribution-Functions.html#MPI-Data-Distribution-Functions" rel="prev" title="MPI Data Distribution Functions">
|
|
cannam@127
|
39 <style type="text/css">
|
|
cannam@127
|
40 <!--
|
|
cannam@127
|
41 a.summary-letter {text-decoration: none}
|
|
cannam@127
|
42 blockquote.smallquotation {font-size: smaller}
|
|
cannam@127
|
43 div.display {margin-left: 3.2em}
|
|
cannam@127
|
44 div.example {margin-left: 3.2em}
|
|
cannam@127
|
45 div.indentedblock {margin-left: 3.2em}
|
|
cannam@127
|
46 div.lisp {margin-left: 3.2em}
|
|
cannam@127
|
47 div.smalldisplay {margin-left: 3.2em}
|
|
cannam@127
|
48 div.smallexample {margin-left: 3.2em}
|
|
cannam@127
|
49 div.smallindentedblock {margin-left: 3.2em; font-size: smaller}
|
|
cannam@127
|
50 div.smalllisp {margin-left: 3.2em}
|
|
cannam@127
|
51 kbd {font-style:oblique}
|
|
cannam@127
|
52 pre.display {font-family: inherit}
|
|
cannam@127
|
53 pre.format {font-family: inherit}
|
|
cannam@127
|
54 pre.menu-comment {font-family: serif}
|
|
cannam@127
|
55 pre.menu-preformatted {font-family: serif}
|
|
cannam@127
|
56 pre.smalldisplay {font-family: inherit; font-size: smaller}
|
|
cannam@127
|
57 pre.smallexample {font-size: smaller}
|
|
cannam@127
|
58 pre.smallformat {font-family: inherit; font-size: smaller}
|
|
cannam@127
|
59 pre.smalllisp {font-size: smaller}
|
|
cannam@127
|
60 span.nocodebreak {white-space:nowrap}
|
|
cannam@127
|
61 span.nolinebreak {white-space:nowrap}
|
|
cannam@127
|
62 span.roman {font-family:serif; font-weight:normal}
|
|
cannam@127
|
63 span.sansserif {font-family:sans-serif; font-weight:normal}
|
|
cannam@127
|
64 ul.no-bullet {list-style: none}
|
|
cannam@127
|
65 -->
|
|
cannam@127
|
66 </style>
|
|
cannam@127
|
67
|
|
cannam@127
|
68
|
|
cannam@127
|
69 </head>
|
|
cannam@127
|
70
|
|
cannam@127
|
71 <body lang="en" bgcolor="#FFFFFF" text="#000000" link="#0000FF" vlink="#800080" alink="#FF0000">
|
|
cannam@127
|
72 <a name="MPI-Plan-Creation"></a>
|
|
cannam@127
|
73 <div class="header">
|
|
cannam@127
|
74 <p>
|
|
cannam@127
|
75 Next: <a href="MPI-Wisdom-Communication.html#MPI-Wisdom-Communication" accesskey="n" rel="next">MPI Wisdom Communication</a>, Previous: <a href="MPI-Data-Distribution-Functions.html#MPI-Data-Distribution-Functions" accesskey="p" rel="prev">MPI Data Distribution Functions</a>, Up: <a href="FFTW-MPI-Reference.html#FFTW-MPI-Reference" accesskey="u" rel="up">FFTW MPI Reference</a> [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html#Concept-Index" title="Index" rel="index">Index</a>]</p>
|
|
cannam@127
|
76 </div>
|
|
cannam@127
|
77 <hr>
|
|
cannam@127
|
78 <a name="MPI-Plan-Creation-1"></a>
|
|
cannam@127
|
79 <h4 class="subsection">6.12.5 MPI Plan Creation</h4>
|
|
cannam@127
|
80
|
|
cannam@127
|
81 <a name="Complex_002ddata-MPI-DFTs"></a>
|
|
cannam@127
|
82 <h4 class="subsubheading">Complex-data MPI DFTs</h4>
|
|
cannam@127
|
83
|
|
cannam@127
|
84 <p>Plans for complex-data DFTs (see <a href="2d-MPI-example.html#g_t2d-MPI-example">2d MPI example</a>) are created by:
|
|
cannam@127
|
85 </p>
|
|
cannam@127
|
86 <a name="index-fftw_005fmpi_005fplan_005fdft_005f1d"></a>
|
|
cannam@127
|
87 <a name="index-fftw_005fmpi_005fplan_005fdft_005f2d-1"></a>
|
|
cannam@127
|
88 <a name="index-fftw_005fmpi_005fplan_005fdft_005f3d"></a>
|
|
cannam@127
|
89 <a name="index-fftw_005fmpi_005fplan_005fdft"></a>
|
|
cannam@127
|
90 <a name="index-fftw_005fmpi_005fplan_005fmany_005fdft"></a>
|
|
cannam@127
|
91 <div class="example">
|
|
cannam@127
|
92 <pre class="example">fftw_plan fftw_mpi_plan_dft_1d(ptrdiff_t n0, fftw_complex *in, fftw_complex *out,
|
|
cannam@127
|
93 MPI_Comm comm, int sign, unsigned flags);
|
|
cannam@127
|
94 fftw_plan fftw_mpi_plan_dft_2d(ptrdiff_t n0, ptrdiff_t n1,
|
|
cannam@127
|
95 fftw_complex *in, fftw_complex *out,
|
|
cannam@127
|
96 MPI_Comm comm, int sign, unsigned flags);
|
|
cannam@127
|
97 fftw_plan fftw_mpi_plan_dft_3d(ptrdiff_t n0, ptrdiff_t n1, ptrdiff_t n2,
|
|
cannam@127
|
98 fftw_complex *in, fftw_complex *out,
|
|
cannam@127
|
99 MPI_Comm comm, int sign, unsigned flags);
|
|
cannam@127
|
100 fftw_plan fftw_mpi_plan_dft(int rnk, const ptrdiff_t *n,
|
|
cannam@127
|
101 fftw_complex *in, fftw_complex *out,
|
|
cannam@127
|
102 MPI_Comm comm, int sign, unsigned flags);
|
|
cannam@127
|
103 fftw_plan fftw_mpi_plan_many_dft(int rnk, const ptrdiff_t *n,
|
|
cannam@127
|
104 ptrdiff_t howmany, ptrdiff_t block, ptrdiff_t tblock,
|
|
cannam@127
|
105 fftw_complex *in, fftw_complex *out,
|
|
cannam@127
|
106 MPI_Comm comm, int sign, unsigned flags);
|
|
cannam@127
|
107 </pre></div>
|
|
cannam@127
|
108
|
|
cannam@127
|
109 <a name="index-MPI-communicator-2"></a>
|
|
cannam@127
|
110 <a name="index-collective-function-4"></a>
|
|
cannam@127
|
111 <p>These are similar to their serial counterparts (see <a href="Complex-DFTs.html#Complex-DFTs">Complex DFTs</a>)
|
|
cannam@127
|
112 in specifying the dimensions, sign, and flags of the transform. The
|
|
cannam@127
|
113 <code>comm</code> argument gives an MPI communicator that specifies the set
|
|
cannam@127
|
114 of processes to participate in the transform; plan creation is a
|
|
cannam@127
|
115 collective function that must be called for all processes in the
|
|
cannam@127
|
116 communicator. The <code>in</code> and <code>out</code> pointers refer only to a
|
|
cannam@127
|
117 portion of the overall transform data (see <a href="MPI-Data-Distribution.html#MPI-Data-Distribution">MPI Data Distribution</a>)
|
|
cannam@127
|
118 as specified by the ‘<samp>local_size</samp>’ functions in the previous
|
|
cannam@127
|
119 section. Unless <code>flags</code> contains <code>FFTW_ESTIMATE</code>, these
|
|
cannam@127
|
120 arrays are overwritten during plan creation as for the serial
|
|
cannam@127
|
121 interface. For multi-dimensional transforms, any dimensions <code>>
|
|
cannam@127
|
122 1</code> are supported; for one-dimensional transforms, only composite
|
|
cannam@127
|
123 (non-prime) <code>n0</code> are currently supported (unlike the serial
|
|
cannam@127
|
124 FFTW). Requesting an unsupported transform size will yield a
|
|
cannam@127
|
125 <code>NULL</code> plan. (As in the serial interface, highly composite sizes
|
|
cannam@127
|
126 generally yield the best performance.)
|
|
cannam@127
|
127 </p>
|
|
cannam@127
|
128 <a name="index-advanced-interface-6"></a>
|
|
cannam@127
|
129 <a name="index-FFTW_005fMPI_005fDEFAULT_005fBLOCK-2"></a>
|
|
cannam@127
|
130 <a name="index-stride-3"></a>
|
|
cannam@127
|
131 <p>The advanced-interface <code>fftw_mpi_plan_many_dft</code> additionally
|
|
cannam@127
|
132 allows you to specify the block sizes for the first dimension
|
|
cannam@127
|
133 (<code>block</code>) of the n<sub>0</sub> × n<sub>1</sub> × n<sub>2</sub> × … × n<sub>d-1</sub> input data and the first dimension
|
|
cannam@127
|
134 (<code>tblock</code>) of the n<sub>1</sub> × n<sub>0</sub> × n<sub>2</sub> ×…× n<sub>d-1</sub> transposed data (at intermediate
|
|
cannam@127
|
135 steps of the transform, and for the output if
|
|
cannam@127
|
136 <code>FFTW_TRANSPOSED_OUT</code> is specified in <code>flags</code>). These must
|
|
cannam@127
|
137 be the same block sizes as were passed to the corresponding
|
|
cannam@127
|
138 ‘<samp>local_size</samp>’ function; you can pass <code>FFTW_MPI_DEFAULT_BLOCK</code>
|
|
cannam@127
|
139 to use FFTW’s default block size as in the basic interface. Also, the
|
|
cannam@127
|
140 <code>howmany</code> parameter specifies that the transform is of contiguous
|
|
cannam@127
|
141 <code>howmany</code>-tuples rather than individual complex numbers; this
|
|
cannam@127
|
142 corresponds to the same parameter in the serial advanced interface
|
|
cannam@127
|
143 (see <a href="Advanced-Complex-DFTs.html#Advanced-Complex-DFTs">Advanced Complex DFTs</a>) with <code>stride = howmany</code> and
|
|
cannam@127
|
144 <code>dist = 1</code>.
|
|
cannam@127
|
145 </p>
|
|
cannam@127
|
146 <a name="MPI-flags"></a>
|
|
cannam@127
|
147 <h4 class="subsubheading">MPI flags</h4>
|
|
cannam@127
|
148
|
|
cannam@127
|
149 <p>The <code>flags</code> can be any of those for the serial FFTW
|
|
cannam@127
|
150 (see <a href="Planner-Flags.html#Planner-Flags">Planner Flags</a>), and in addition may include one or more of
|
|
cannam@127
|
151 the following MPI-specific flags, which improve performance at the
|
|
cannam@127
|
152 cost of changing the output or input data formats.
|
|
cannam@127
|
153 </p>
|
|
cannam@127
|
154 <ul>
|
|
cannam@127
|
155 <li> <a name="index-FFTW_005fMPI_005fSCRAMBLED_005fOUT-2"></a>
|
|
cannam@127
|
156 <a name="index-FFTW_005fMPI_005fSCRAMBLED_005fIN-2"></a>
|
|
cannam@127
|
157 <code>FFTW_MPI_SCRAMBLED_OUT</code>, <code>FFTW_MPI_SCRAMBLED_IN</code>: valid for
|
|
cannam@127
|
158 1d transforms only, these flags indicate that the output/input of the
|
|
cannam@127
|
159 transform are in an undocumented “scrambled” order. A forward
|
|
cannam@127
|
160 <code>FFTW_MPI_SCRAMBLED_OUT</code> transform can be inverted by a backward
|
|
cannam@127
|
161 <code>FFTW_MPI_SCRAMBLED_IN</code> (times the usual 1/<i>N</i> normalization).
|
|
cannam@127
|
162 See <a href="One_002ddimensional-distributions.html#One_002ddimensional-distributions">One-dimensional distributions</a>.
|
|
cannam@127
|
163
|
|
cannam@127
|
164 </li><li> <a name="index-FFTW_005fMPI_005fTRANSPOSED_005fOUT-2"></a>
|
|
cannam@127
|
165 <a name="index-FFTW_005fMPI_005fTRANSPOSED_005fIN-2"></a>
|
|
cannam@127
|
166 <code>FFTW_MPI_TRANSPOSED_OUT</code>, <code>FFTW_MPI_TRANSPOSED_IN</code>: valid
|
|
cannam@127
|
167 for multidimensional (<code>rnk > 1</code>) transforms only, these flags
|
|
cannam@127
|
168 specify that the output or input of an n<sub>0</sub> × n<sub>1</sub> × n<sub>2</sub> × … × n<sub>d-1</sub> transform is
|
|
cannam@127
|
169 transposed to n<sub>1</sub> × n<sub>0</sub> × n<sub>2</sub> ×…× n<sub>d-1</sub>. See <a href="Transposed-distributions.html#Transposed-distributions">Transposed distributions</a>.
|
|
cannam@127
|
170
|
|
cannam@127
|
171 </li></ul>
|
|
cannam@127
|
172
|
|
cannam@127
|
173 <a name="Real_002ddata-MPI-DFTs"></a>
|
|
cannam@127
|
174 <h4 class="subsubheading">Real-data MPI DFTs</h4>
|
|
cannam@127
|
175
|
|
cannam@127
|
176 <a name="index-r2c-4"></a>
|
|
cannam@127
|
177 <p>Plans for real-input/output (r2c/c2r) DFTs (see <a href="Multi_002ddimensional-MPI-DFTs-of-Real-Data.html#Multi_002ddimensional-MPI-DFTs-of-Real-Data">Multi-dimensional MPI DFTs of Real Data</a>) are created by:
|
|
cannam@127
|
178 </p>
|
|
cannam@127
|
179 <a name="index-fftw_005fmpi_005fplan_005fdft_005fr2c_005f2d"></a>
|
|
cannam@127
|
180 <a name="index-fftw_005fmpi_005fplan_005fdft_005fr2c_005f2d-1"></a>
|
|
cannam@127
|
181 <a name="index-fftw_005fmpi_005fplan_005fdft_005fr2c_005f3d"></a>
|
|
cannam@127
|
182 <a name="index-fftw_005fmpi_005fplan_005fdft_005fr2c"></a>
|
|
cannam@127
|
183 <a name="index-fftw_005fmpi_005fplan_005fdft_005fc2r_005f2d"></a>
|
|
cannam@127
|
184 <a name="index-fftw_005fmpi_005fplan_005fdft_005fc2r_005f2d-1"></a>
|
|
cannam@127
|
185 <a name="index-fftw_005fmpi_005fplan_005fdft_005fc2r_005f3d"></a>
|
|
cannam@127
|
186 <a name="index-fftw_005fmpi_005fplan_005fdft_005fc2r"></a>
|
|
cannam@127
|
187 <div class="example">
|
|
cannam@127
|
188 <pre class="example">fftw_plan fftw_mpi_plan_dft_r2c_2d(ptrdiff_t n0, ptrdiff_t n1,
|
|
cannam@127
|
189 double *in, fftw_complex *out,
|
|
cannam@127
|
190 MPI_Comm comm, unsigned flags);
|
|
cannam@127
|
191 fftw_plan fftw_mpi_plan_dft_r2c_2d(ptrdiff_t n0, ptrdiff_t n1,
|
|
cannam@127
|
192 double *in, fftw_complex *out,
|
|
cannam@127
|
193 MPI_Comm comm, unsigned flags);
|
|
cannam@127
|
194 fftw_plan fftw_mpi_plan_dft_r2c_3d(ptrdiff_t n0, ptrdiff_t n1, ptrdiff_t n2,
|
|
cannam@127
|
195 double *in, fftw_complex *out,
|
|
cannam@127
|
196 MPI_Comm comm, unsigned flags);
|
|
cannam@127
|
197 fftw_plan fftw_mpi_plan_dft_r2c(int rnk, const ptrdiff_t *n,
|
|
cannam@127
|
198 double *in, fftw_complex *out,
|
|
cannam@127
|
199 MPI_Comm comm, unsigned flags);
|
|
cannam@127
|
200 fftw_plan fftw_mpi_plan_dft_c2r_2d(ptrdiff_t n0, ptrdiff_t n1,
|
|
cannam@127
|
201 fftw_complex *in, double *out,
|
|
cannam@127
|
202 MPI_Comm comm, unsigned flags);
|
|
cannam@127
|
203 fftw_plan fftw_mpi_plan_dft_c2r_2d(ptrdiff_t n0, ptrdiff_t n1,
|
|
cannam@127
|
204 fftw_complex *in, double *out,
|
|
cannam@127
|
205 MPI_Comm comm, unsigned flags);
|
|
cannam@127
|
206 fftw_plan fftw_mpi_plan_dft_c2r_3d(ptrdiff_t n0, ptrdiff_t n1, ptrdiff_t n2,
|
|
cannam@127
|
207 fftw_complex *in, double *out,
|
|
cannam@127
|
208 MPI_Comm comm, unsigned flags);
|
|
cannam@127
|
209 fftw_plan fftw_mpi_plan_dft_c2r(int rnk, const ptrdiff_t *n,
|
|
cannam@127
|
210 fftw_complex *in, double *out,
|
|
cannam@127
|
211 MPI_Comm comm, unsigned flags);
|
|
cannam@127
|
212 </pre></div>
|
|
cannam@127
|
213
|
|
cannam@127
|
214 <p>Similar to the serial interface (see <a href="Real_002ddata-DFTs.html#Real_002ddata-DFTs">Real-data DFTs</a>), these
|
|
cannam@127
|
215 transform logically n<sub>0</sub> × n<sub>1</sub> × n<sub>2</sub> × … × n<sub>d-1</sub> real data to/from n<sub>0</sub> × n<sub>1</sub> × n<sub>2</sub> × … × (n<sub>d-1</sub>/2 + 1) complex
|
|
cannam@127
|
216 data, representing the non-redundant half of the conjugate-symmetry
|
|
cannam@127
|
217 output of a real-input DFT (see <a href="Multi_002ddimensional-Transforms.html#Multi_002ddimensional-Transforms">Multi-dimensional Transforms</a>).
|
|
cannam@127
|
218 However, the real array must be stored within a padded n<sub>0</sub> × n<sub>1</sub> × n<sub>2</sub> × … × [2 (n<sub>d-1</sub>/2 + 1)]
|
|
cannam@127
|
219 array (much like the in-place serial r2c transforms, but here for
|
|
cannam@127
|
220 out-of-place transforms as well). Currently, only multi-dimensional
|
|
cannam@127
|
221 (<code>rnk > 1</code>) r2c/c2r transforms are supported (requesting a plan
|
|
cannam@127
|
222 for <code>rnk = 1</code> will yield <code>NULL</code>). As explained above
|
|
cannam@127
|
223 (see <a href="Multi_002ddimensional-MPI-DFTs-of-Real-Data.html#Multi_002ddimensional-MPI-DFTs-of-Real-Data">Multi-dimensional MPI DFTs of Real Data</a>), the data
|
|
cannam@127
|
224 distribution of both the real and complex arrays is given by the
|
|
cannam@127
|
225 ‘<samp>local_size</samp>’ function called for the dimensions of the
|
|
cannam@127
|
226 <em>complex</em> array. Similar to the other planning functions, the
|
|
cannam@127
|
227 input and output arrays are overwritten when the plan is created
|
|
cannam@127
|
228 except in <code>FFTW_ESTIMATE</code> mode.
|
|
cannam@127
|
229 </p>
|
|
cannam@127
|
230 <p>As for the complex DFTs above, there is an advance interface that
|
|
cannam@127
|
231 allows you to manually specify block sizes and to transform contiguous
|
|
cannam@127
|
232 <code>howmany</code>-tuples of real/complex numbers:
|
|
cannam@127
|
233 </p>
|
|
cannam@127
|
234 <a name="index-fftw_005fmpi_005fplan_005fmany_005fdft_005fr2c"></a>
|
|
cannam@127
|
235 <a name="index-fftw_005fmpi_005fplan_005fmany_005fdft_005fc2r"></a>
|
|
cannam@127
|
236 <div class="example">
|
|
cannam@127
|
237 <pre class="example">fftw_plan fftw_mpi_plan_many_dft_r2c
|
|
cannam@127
|
238 (int rnk, const ptrdiff_t *n, ptrdiff_t howmany,
|
|
cannam@127
|
239 ptrdiff_t iblock, ptrdiff_t oblock,
|
|
cannam@127
|
240 double *in, fftw_complex *out,
|
|
cannam@127
|
241 MPI_Comm comm, unsigned flags);
|
|
cannam@127
|
242 fftw_plan fftw_mpi_plan_many_dft_c2r
|
|
cannam@127
|
243 (int rnk, const ptrdiff_t *n, ptrdiff_t howmany,
|
|
cannam@127
|
244 ptrdiff_t iblock, ptrdiff_t oblock,
|
|
cannam@127
|
245 fftw_complex *in, double *out,
|
|
cannam@127
|
246 MPI_Comm comm, unsigned flags);
|
|
cannam@127
|
247 </pre></div>
|
|
cannam@127
|
248
|
|
cannam@127
|
249 <a name="MPI-r2r-transforms"></a>
|
|
cannam@127
|
250 <h4 class="subsubheading">MPI r2r transforms</h4>
|
|
cannam@127
|
251
|
|
cannam@127
|
252 <a name="index-r2r-4"></a>
|
|
cannam@127
|
253 <p>There are corresponding plan-creation routines for r2r
|
|
cannam@127
|
254 transforms (see <a href="More-DFTs-of-Real-Data.html#More-DFTs-of-Real-Data">More DFTs of Real Data</a>), currently supporting
|
|
cannam@127
|
255 multidimensional (<code>rnk > 1</code>) transforms only (<code>rnk = 1</code> will
|
|
cannam@127
|
256 yield a <code>NULL</code> plan):
|
|
cannam@127
|
257 </p>
|
|
cannam@127
|
258 <div class="example">
|
|
cannam@127
|
259 <pre class="example">fftw_plan fftw_mpi_plan_r2r_2d(ptrdiff_t n0, ptrdiff_t n1,
|
|
cannam@127
|
260 double *in, double *out,
|
|
cannam@127
|
261 MPI_Comm comm,
|
|
cannam@127
|
262 fftw_r2r_kind kind0, fftw_r2r_kind kind1,
|
|
cannam@127
|
263 unsigned flags);
|
|
cannam@127
|
264 fftw_plan fftw_mpi_plan_r2r_3d(ptrdiff_t n0, ptrdiff_t n1, ptrdiff_t n2,
|
|
cannam@127
|
265 double *in, double *out,
|
|
cannam@127
|
266 MPI_Comm comm,
|
|
cannam@127
|
267 fftw_r2r_kind kind0, fftw_r2r_kind kind1, fftw_r2r_kind kind2,
|
|
cannam@127
|
268 unsigned flags);
|
|
cannam@127
|
269 fftw_plan fftw_mpi_plan_r2r(int rnk, const ptrdiff_t *n,
|
|
cannam@127
|
270 double *in, double *out,
|
|
cannam@127
|
271 MPI_Comm comm, const fftw_r2r_kind *kind,
|
|
cannam@127
|
272 unsigned flags);
|
|
cannam@127
|
273 fftw_plan fftw_mpi_plan_many_r2r(int rnk, const ptrdiff_t *n,
|
|
cannam@127
|
274 ptrdiff_t iblock, ptrdiff_t oblock,
|
|
cannam@127
|
275 double *in, double *out,
|
|
cannam@127
|
276 MPI_Comm comm, const fftw_r2r_kind *kind,
|
|
cannam@127
|
277 unsigned flags);
|
|
cannam@127
|
278 </pre></div>
|
|
cannam@127
|
279
|
|
cannam@127
|
280 <p>The parameters are much the same as for the complex DFTs above, except
|
|
cannam@127
|
281 that the arrays are of real numbers (and hence the outputs of the
|
|
cannam@127
|
282 ‘<samp>local_size</samp>’ data-distribution functions should be interpreted as
|
|
cannam@127
|
283 counts of real rather than complex numbers). Also, the <code>kind</code>
|
|
cannam@127
|
284 parameters specify the r2r kinds along each dimension as for the
|
|
cannam@127
|
285 serial interface (see <a href="Real_002dto_002dReal-Transform-Kinds.html#Real_002dto_002dReal-Transform-Kinds">Real-to-Real Transform Kinds</a>). See <a href="Other-Multi_002ddimensional-Real_002ddata-MPI-Transforms.html#Other-Multi_002ddimensional-Real_002ddata-MPI-Transforms">Other Multi-dimensional Real-data MPI Transforms</a>.
|
|
cannam@127
|
286 </p>
|
|
cannam@127
|
287 <a name="MPI-transposition"></a>
|
|
cannam@127
|
288 <h4 class="subsubheading">MPI transposition</h4>
|
|
cannam@127
|
289 <a name="index-transpose-5"></a>
|
|
cannam@127
|
290
|
|
cannam@127
|
291 <p>FFTW also provides routines to plan a transpose of a distributed
|
|
cannam@127
|
292 <code>n0</code> by <code>n1</code> array of real numbers, or an array of
|
|
cannam@127
|
293 <code>howmany</code>-tuples of real numbers with specified block sizes
|
|
cannam@127
|
294 (see <a href="FFTW-MPI-Transposes.html#FFTW-MPI-Transposes">FFTW MPI Transposes</a>):
|
|
cannam@127
|
295 </p>
|
|
cannam@127
|
296 <a name="index-fftw_005fmpi_005fplan_005ftranspose-1"></a>
|
|
cannam@127
|
297 <a name="index-fftw_005fmpi_005fplan_005fmany_005ftranspose-1"></a>
|
|
cannam@127
|
298 <div class="example">
|
|
cannam@127
|
299 <pre class="example">fftw_plan fftw_mpi_plan_transpose(ptrdiff_t n0, ptrdiff_t n1,
|
|
cannam@127
|
300 double *in, double *out,
|
|
cannam@127
|
301 MPI_Comm comm, unsigned flags);
|
|
cannam@127
|
302 fftw_plan fftw_mpi_plan_many_transpose
|
|
cannam@127
|
303 (ptrdiff_t n0, ptrdiff_t n1, ptrdiff_t howmany,
|
|
cannam@127
|
304 ptrdiff_t block0, ptrdiff_t block1,
|
|
cannam@127
|
305 double *in, double *out, MPI_Comm comm, unsigned flags);
|
|
cannam@127
|
306 </pre></div>
|
|
cannam@127
|
307
|
|
cannam@127
|
308 <a name="index-new_002darray-execution-2"></a>
|
|
cannam@127
|
309 <a name="index-fftw_005fmpi_005fexecute_005fr2r-1"></a>
|
|
cannam@127
|
310 <p>These plans are used with the <code>fftw_mpi_execute_r2r</code> new-array
|
|
cannam@127
|
311 execute function (see <a href="Using-MPI-Plans.html#Using-MPI-Plans">Using MPI Plans</a>), since they count as (rank
|
|
cannam@127
|
312 zero) r2r plans from FFTW’s perspective.
|
|
cannam@127
|
313 </p>
|
|
cannam@127
|
314 <hr>
|
|
cannam@127
|
315 <div class="header">
|
|
cannam@127
|
316 <p>
|
|
cannam@127
|
317 Next: <a href="MPI-Wisdom-Communication.html#MPI-Wisdom-Communication" accesskey="n" rel="next">MPI Wisdom Communication</a>, Previous: <a href="MPI-Data-Distribution-Functions.html#MPI-Data-Distribution-Functions" accesskey="p" rel="prev">MPI Data Distribution Functions</a>, Up: <a href="FFTW-MPI-Reference.html#FFTW-MPI-Reference" accesskey="u" rel="up">FFTW MPI Reference</a> [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html#Concept-Index" title="Index" rel="index">Index</a>]</p>
|
|
cannam@127
|
318 </div>
|
|
cannam@127
|
319
|
|
cannam@127
|
320
|
|
cannam@127
|
321
|
|
cannam@127
|
322 </body>
|
|
cannam@127
|
323 </html>
|
