annotate src/fftw-3.3.3/doc/html/Basic-distributed_002dtranspose-interface.html @ 10:37bf6b4a2645

Add FFTW3
author Chris Cannam
date Wed, 20 Mar 2013 15:35:50 +0000
parents
children
rev   line source
Chris@10 1 <html lang="en">
Chris@10 2 <head>
Chris@10 3 <title>Basic distributed-transpose interface - FFTW 3.3.3</title>
Chris@10 4 <meta http-equiv="Content-Type" content="text/html">
Chris@10 5 <meta name="description" content="FFTW 3.3.3">
Chris@10 6 <meta name="generator" content="makeinfo 4.13">
Chris@10 7 <link title="Top" rel="start" href="index.html#Top">
Chris@10 8 <link rel="up" href="FFTW-MPI-Transposes.html#FFTW-MPI-Transposes" title="FFTW MPI Transposes">
Chris@10 9 <link rel="prev" href="FFTW-MPI-Transposes.html#FFTW-MPI-Transposes" title="FFTW MPI Transposes">
Chris@10 10 <link rel="next" href="Advanced-distributed_002dtranspose-interface.html#Advanced-distributed_002dtranspose-interface" title="Advanced distributed-transpose interface">
Chris@10 11 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
Chris@10 12 <!--
Chris@10 13 This manual is for FFTW
Chris@10 14 (version 3.3.3, 25 November 2012).
Chris@10 15
Chris@10 16 Copyright (C) 2003 Matteo Frigo.
Chris@10 17
Chris@10 18 Copyright (C) 2003 Massachusetts Institute of Technology.
Chris@10 19
Chris@10 20 Permission is granted to make and distribute verbatim copies of
Chris@10 21 this manual provided the copyright notice and this permission
Chris@10 22 notice are preserved on all copies.
Chris@10 23
Chris@10 24 Permission is granted to copy and distribute modified versions of
Chris@10 25 this manual under the conditions for verbatim copying, provided
Chris@10 26 that the entire resulting derived work is distributed under the
Chris@10 27 terms of a permission notice identical to this one.
Chris@10 28
Chris@10 29 Permission is granted to copy and distribute translations of this
Chris@10 30 manual into another language, under the above conditions for
Chris@10 31 modified versions, except that this permission notice may be
Chris@10 32 stated in a translation approved by the Free Software Foundation.
Chris@10 33 -->
Chris@10 34 <meta http-equiv="Content-Style-Type" content="text/css">
Chris@10 35 <style type="text/css"><!--
Chris@10 36 pre.display { font-family:inherit }
Chris@10 37 pre.format { font-family:inherit }
Chris@10 38 pre.smalldisplay { font-family:inherit; font-size:smaller }
Chris@10 39 pre.smallformat { font-family:inherit; font-size:smaller }
Chris@10 40 pre.smallexample { font-size:smaller }
Chris@10 41 pre.smalllisp { font-size:smaller }
Chris@10 42 span.sc { font-variant:small-caps }
Chris@10 43 span.roman { font-family:serif; font-weight:normal; }
Chris@10 44 span.sansserif { font-family:sans-serif; font-weight:normal; }
Chris@10 45 --></style>
Chris@10 46 </head>
Chris@10 47 <body>
Chris@10 48 <div class="node">
Chris@10 49 <a name="Basic-distributed-transpose-interface"></a>
Chris@10 50 <a name="Basic-distributed_002dtranspose-interface"></a>
Chris@10 51 <p>
Chris@10 52 Next:&nbsp;<a rel="next" accesskey="n" href="Advanced-distributed_002dtranspose-interface.html#Advanced-distributed_002dtranspose-interface">Advanced distributed-transpose interface</a>,
Chris@10 53 Previous:&nbsp;<a rel="previous" accesskey="p" href="FFTW-MPI-Transposes.html#FFTW-MPI-Transposes">FFTW MPI Transposes</a>,
Chris@10 54 Up:&nbsp;<a rel="up" accesskey="u" href="FFTW-MPI-Transposes.html#FFTW-MPI-Transposes">FFTW MPI Transposes</a>
Chris@10 55 <hr>
Chris@10 56 </div>
Chris@10 57
Chris@10 58 <h4 class="subsection">6.7.1 Basic distributed-transpose interface</h4>
Chris@10 59
Chris@10 60 <p>In particular, suppose that we have an <code>n0</code> by <code>n1</code> array in
Chris@10 61 row-major order, block-distributed across the <code>n0</code> dimension. To
Chris@10 62 transpose this into an <code>n1</code> by <code>n0</code> array block-distributed
Chris@10 63 across the <code>n1</code> dimension, we would create a plan by calling the
Chris@10 64 following function:
Chris@10 65
Chris@10 66 <pre class="example"> fftw_plan fftw_mpi_plan_transpose(ptrdiff_t n0, ptrdiff_t n1,
Chris@10 67 double *in, double *out,
Chris@10 68 MPI_Comm comm, unsigned flags);
Chris@10 69 </pre>
Chris@10 70 <p><a name="index-fftw_005fmpi_005fplan_005ftranspose-397"></a>
Chris@10 71 The input and output arrays (<code>in</code> and <code>out</code>) can be the
Chris@10 72 same. The transpose is actually executed by calling
Chris@10 73 <code>fftw_execute</code> on the plan, as usual.
Chris@10 74 <a name="index-fftw_005fexecute-398"></a>
Chris@10 75
Chris@10 76 <p>The <code>flags</code> are the usual FFTW planner flags, but support
Chris@10 77 two additional flags: <code>FFTW_MPI_TRANSPOSED_OUT</code> and/or
Chris@10 78 <code>FFTW_MPI_TRANSPOSED_IN</code>. What these flags indicate, for
Chris@10 79 transpose plans, is that the output and/or input, respectively, are
Chris@10 80 <em>locally</em> transposed. That is, on each process input data is
Chris@10 81 normally stored as a <code>local_n0</code> by <code>n1</code> array in row-major
Chris@10 82 order, but for an <code>FFTW_MPI_TRANSPOSED_IN</code> plan the input data is
Chris@10 83 stored as <code>n1</code> by <code>local_n0</code> in row-major order. Similarly,
Chris@10 84 <code>FFTW_MPI_TRANSPOSED_OUT</code> means that the output is <code>n0</code> by
Chris@10 85 <code>local_n1</code> instead of <code>local_n1</code> by <code>n0</code>.
Chris@10 86 <a name="index-FFTW_005fMPI_005fTRANSPOSED_005fOUT-399"></a><a name="index-FFTW_005fMPI_005fTRANSPOSED_005fIN-400"></a>
Chris@10 87
Chris@10 88 <p>To determine the local size of the array on each process before and
Chris@10 89 after the transpose, as well as the amount of storage that must be
Chris@10 90 allocated, one should call <code>fftw_mpi_local_size_2d_transposed</code>,
Chris@10 91 just as for a 2d DFT as described in the previous section:
Chris@10 92 <a name="index-data-distribution-401"></a>
Chris@10 93 <pre class="example"> ptrdiff_t fftw_mpi_local_size_2d_transposed
Chris@10 94 (ptrdiff_t n0, ptrdiff_t n1, MPI_Comm comm,
Chris@10 95 ptrdiff_t *local_n0, ptrdiff_t *local_0_start,
Chris@10 96 ptrdiff_t *local_n1, ptrdiff_t *local_1_start);
Chris@10 97 </pre>
Chris@10 98 <p><a name="index-fftw_005fmpi_005flocal_005fsize_005f2d_005ftransposed-402"></a>
Chris@10 99 Again, the return value is the local storage to allocate, which in
Chris@10 100 this case is the number of <em>real</em> (<code>double</code>) values rather
Chris@10 101 than complex numbers as in the previous examples.
Chris@10 102
Chris@10 103 </body></html>
Chris@10 104