Mercurial > hg > sv-dependency-builds

annotate src/fftw-3.3.3/doc/html/FFTW-MPI-Fortran-Interface.html @ 10:37bf6b4a2645

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
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>FFTW MPI Fortran 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="Distributed_002dmemory-FFTW-with-MPI.html#Distributed_002dmemory-FFTW-with-MPI" title="Distributed-memory FFTW with MPI">
Chris@10 9 <link rel="prev" href="FFTW-MPI-Reference.html#FFTW-MPI-Reference" title="FFTW MPI Reference">
Chris@10 10 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
Chris@10 11 <!--
Chris@10 12 This manual is for FFTW
Chris@10 13 (version 3.3.3, 25 November 2012).
Chris@10 14
Chris@10 15 Copyright (C) 2003 Matteo Frigo.
Chris@10 16
Chris@10 17 Copyright (C) 2003 Massachusetts Institute of Technology.
Chris@10 18
Chris@10 19 Permission is granted to make and distribute verbatim copies of
Chris@10 20 this manual provided the copyright notice and this permission
Chris@10 21 notice are preserved on all copies.
Chris@10 22
Chris@10 23 Permission is granted to copy and distribute modified versions of
Chris@10 24 this manual under the conditions for verbatim copying, provided
Chris@10 25 that the entire resulting derived work is distributed under the
Chris@10 26 terms of a permission notice identical to this one.
Chris@10 27
Chris@10 28 Permission is granted to copy and distribute translations of this
Chris@10 29 manual into another language, under the above conditions for
Chris@10 30 modified versions, except that this permission notice may be
Chris@10 31 stated in a translation approved by the Free Software Foundation.
Chris@10 32 -->
Chris@10 33 <meta http-equiv="Content-Style-Type" content="text/css">
Chris@10 34 <style type="text/css"><!--
Chris@10 35 pre.display { font-family:inherit }
Chris@10 36 pre.format { font-family:inherit }
Chris@10 37 pre.smalldisplay { font-family:inherit; font-size:smaller }
Chris@10 38 pre.smallformat { font-family:inherit; font-size:smaller }
Chris@10 39 pre.smallexample { font-size:smaller }
Chris@10 40 pre.smalllisp { font-size:smaller }
Chris@10 41 span.sc { font-variant:small-caps }
Chris@10 42 span.roman { font-family:serif; font-weight:normal; }
Chris@10 43 span.sansserif { font-family:sans-serif; font-weight:normal; }
Chris@10 44 --></style>
Chris@10 45 </head>
Chris@10 46 <body>
Chris@10 47 <div class="node">
Chris@10 48 <a name="FFTW-MPI-Fortran-Interface"></a>
Chris@10 49 <p>
Chris@10 50 Previous:&nbsp;<a rel="previous" accesskey="p" href="FFTW-MPI-Reference.html#FFTW-MPI-Reference">FFTW MPI Reference</a>,
Chris@10 51 Up:&nbsp;<a rel="up" accesskey="u" href="Distributed_002dmemory-FFTW-with-MPI.html#Distributed_002dmemory-FFTW-with-MPI">Distributed-memory FFTW with MPI</a>
Chris@10 52 <hr>
Chris@10 53 </div>
Chris@10 54
Chris@10 55 <h3 class="section">6.13 FFTW MPI Fortran Interface</h3>
Chris@10 56
Chris@10 57 <p><a name="index-Fortran-interface-494"></a>
Chris@10 58 <a name="index-iso_005fc_005fbinding-495"></a>The FFTW MPI interface is callable from modern Fortran compilers
Chris@10 59 supporting the Fortran 2003 <code>iso_c_binding</code> standard for calling
Chris@10 60 C functions. As described in <a href="Calling-FFTW-from-Modern-Fortran.html#Calling-FFTW-from-Modern-Fortran">Calling FFTW from Modern Fortran</a>,
Chris@10 61 this means that you can directly call FFTW's C interface from Fortran
Chris@10 62 with only minor changes in syntax. There are, however, a few things
Chris@10 63 specific to the MPI interface to keep in mind:
Chris@10 64
Chris@10 65 <ul>
Chris@10 66 <li>Instead of including <code>fftw3.f03</code> as in <a href="Overview-of-Fortran-interface.html#Overview-of-Fortran-interface">Overview of Fortran interface</a>, you should <code>include 'fftw3-mpi.f03'</code> (after
Chris@10 67 <code>use, intrinsic :: iso_c_binding</code> as before). The
Chris@10 68 <code>fftw3-mpi.f03</code> file includes <code>fftw3.f03</code>, so you should
Chris@10 69 <em>not</em> <code>include</code> them both yourself. (You will also want to
Chris@10 70 include the MPI header file, usually via <code>include 'mpif.h'</code> or
Chris@10 71 similar, although though this is not needed by <code>fftw3-mpi.f03</code>
Chris@10 72 <i>per se</i>.) (To use the &lsquo;<samp><span class="samp">fftwl_</span></samp>&rsquo; <code>long double</code> extended-precision routines in supporting compilers, you should include <code>fftw3f-mpi.f03</code> in <em>addition</em> to <code>fftw3-mpi.f03</code>. See <a href="Extended-and-quadruple-precision-in-Fortran.html#Extended-and-quadruple-precision-in-Fortran">Extended and quadruple precision in Fortran</a>.)
Chris@10 73
Chris@10 74 <li>Because of the different storage conventions between C and Fortran,
Chris@10 75 you reverse the order of your array dimensions when passing them to
Chris@10 76 FFTW (see <a href="Reversing-array-dimensions.html#Reversing-array-dimensions">Reversing array dimensions</a>). This is merely a
Chris@10 77 difference in notation and incurs no performance overhead. However,
Chris@10 78 it means that, whereas in C the <em>first</em> dimension is distributed,
Chris@10 79 in Fortran the <em>last</em> dimension of your array is distributed.
Chris@10 80
Chris@10 81 <li><a name="index-MPI-communicator-496"></a>In Fortran, communicators are stored as <code>integer</code> types; there is
Chris@10 82 no <code>MPI_Comm</code> type, nor is there any way to access a C
Chris@10 83 <code>MPI_Comm</code>. Fortunately, this is taken care of for you by the
Chris@10 84 FFTW Fortran interface: whenever the C interface expects an
Chris@10 85 <code>MPI_Comm</code> type, you should pass the Fortran communicator as an
Chris@10 86 <code>integer</code>.<a rel="footnote" href="#fn-1" name="fnd-1"><sup>1</sup></a>
Chris@10 87
Chris@10 88 <li>Because you need to call the &lsquo;<samp><span class="samp">local_size</span></samp>&rsquo; function to find out
Chris@10 89 how much space to allocate, and this may be <em>larger</em> than the
Chris@10 90 local portion of the array (see <a href="MPI-Data-Distribution.html#MPI-Data-Distribution">MPI Data Distribution</a>), you should
Chris@10 91 <em>always</em> allocate your arrays dynamically using FFTW's allocation
Chris@10 92 routines as described in <a href="Allocating-aligned-memory-in-Fortran.html#Allocating-aligned-memory-in-Fortran">Allocating aligned memory in Fortran</a>.
Chris@10 93 (Coincidentally, this also provides the best performance by
Chris@10 94 guaranteeding proper data alignment.)
Chris@10 95
Chris@10 96 <li>Because all sizes in the MPI FFTW interface are declared as
Chris@10 97 <code>ptrdiff_t</code> in C, you should use <code>integer(C_INTPTR_T)</code> in
Chris@10 98 Fortran (see <a href="FFTW-Fortran-type-reference.html#FFTW-Fortran-type-reference">FFTW Fortran type reference</a>).
Chris@10 99
Chris@10 100 <li><a name="index-fftw_005fexecute_005fdft-497"></a><a name="index-fftw_005fmpi_005fexecute_005fdft-498"></a><a name="index-new_002darray-execution-499"></a>In Fortran, because of the language semantics, we generally recommend
Chris@10 101 using the new-array execute functions for all plans, even in the
Chris@10 102 common case where you are executing the plan on the same arrays for
Chris@10 103 which the plan was created (see <a href="Plan-execution-in-Fortran.html#Plan-execution-in-Fortran">Plan execution in Fortran</a>).
Chris@10 104 However, note that in the MPI interface these functions are changed:
Chris@10 105 <code>fftw_execute_dft</code> becomes <code>fftw_mpi_execute_dft</code>,
Chris@10 106 etcetera. See <a href="Using-MPI-Plans.html#Using-MPI-Plans">Using MPI Plans</a>.
Chris@10 107
Chris@10 108 </ul>
Chris@10 109
Chris@10 110 <p>For example, here is a Fortran code snippet to perform a distributed
Chris@10 111 L&nbsp;&times;&nbsp;M complex DFT in-place. (This assumes you have already
Chris@10 112 initialized MPI with <code>MPI_init</code> and have also performed
Chris@10 113 <code>call fftw_mpi_init</code>.)
Chris@10 114
Chris@10 115 <pre class="example"> use, intrinsic :: iso_c_binding
Chris@10 116 include 'fftw3-mpi.f03'
Chris@10 117 integer(C_INTPTR_T), parameter :: L = ...
Chris@10 118 integer(C_INTPTR_T), parameter :: M = ...
Chris@10 119 type(C_PTR) :: plan, cdata
Chris@10 120 complex(C_DOUBLE_COMPLEX), pointer :: data(:,:)
Chris@10 121 integer(C_INTPTR_T) :: i, j, alloc_local, local_M, local_j_offset
Chris@10 122
Chris@10 123 ! <span class="roman">get local data size and allocate (note dimension reversal)</span>
Chris@10 124 alloc_local = fftw_mpi_local_size_2d(M, L, MPI_COMM_WORLD, &amp;
Chris@10 125 local_M, local_j_offset)
Chris@10 126 cdata = fftw_alloc_complex(alloc_local)
Chris@10 127 call c_f_pointer(cdata, data, [L,local_M])
Chris@10 128
Chris@10 129 ! <span class="roman">create MPI plan for in-place forward DFT (note dimension reversal)</span>
Chris@10 130 plan = fftw_mpi_plan_dft_2d(M, L, data, data, MPI_COMM_WORLD, &amp;
Chris@10 131 FFTW_FORWARD, FFTW_MEASURE)
Chris@10 132
Chris@10 133 ! <span class="roman">initialize data to some function</span> my_function(i,j)
Chris@10 134 do j = 1, local_M
Chris@10 135 do i = 1, L
Chris@10 136 data(i, j) = my_function(i, j + local_j_offset)
Chris@10 137 end do
Chris@10 138 end do
Chris@10 139
Chris@10 140 ! <span class="roman">compute transform (as many times as desired)</span>
Chris@10 141 call fftw_mpi_execute_dft(plan, data, data)
Chris@10 142
Chris@10 143 call fftw_destroy_plan(plan)
Chris@10 144 call fftw_free(cdata)
Chris@10 145 </pre>
Chris@10 146 <p>Note that when we called <code>fftw_mpi_local_size_2d</code> and
Chris@10 147 <code>fftw_mpi_plan_dft_2d</code> with the dimensions in reversed order,
Chris@10 148 since a L&nbsp;&times;&nbsp;M Fortran array is viewed by FFTW in C as a
Chris@10 149 M&nbsp;&times;&nbsp;L array. This means that the array was distributed over
Chris@10 150 the <code>M</code> dimension, the local portion of which is a
Chris@10 151 L&nbsp;&times;&nbsp;local_M array in Fortran. (You must <em>not</em> use an
Chris@10 152 <code>allocate</code> statement to allocate an L&nbsp;&times;&nbsp;local_M array,
Chris@10 153 however; you must allocate <code>alloc_local</code> complex numbers, which
Chris@10 154 may be greater than <code>L * local_M</code>, in order to reserve space for
Chris@10 155 intermediate steps of the transform.) Finally, we mention that
Chris@10 156 because C's array indices are zero-based, the <code>local_j_offset</code>
Chris@10 157 argument can conveniently be interpreted as an offset in the 1-based
Chris@10 158 <code>j</code> index (rather than as a starting index as in C).
Chris@10 159
Chris@10 160 <p>If instead you had used the <code>ior(FFTW_MEASURE,
Chris@10 161 FFTW_MPI_TRANSPOSED_OUT)</code> flag, the output of the transform would be a
Chris@10 162 transposed M&nbsp;&times;&nbsp;local_L array, associated with the <em>same</em>
Chris@10 163 <code>cdata</code> allocation (since the transform is in-place), and which
Chris@10 164 you could declare with:
Chris@10 165
Chris@10 166 <pre class="example"> complex(C_DOUBLE_COMPLEX), pointer :: tdata(:,:)
Chris@10 167 ...
Chris@10 168 call c_f_pointer(cdata, tdata, [M,local_L])
Chris@10 169 </pre>
Chris@10 170 <p>where <code>local_L</code> would have been obtained by changing the
Chris@10 171 <code>fftw_mpi_local_size_2d</code> call to:
Chris@10 172
Chris@10 173 <pre class="example"> alloc_local = fftw_mpi_local_size_2d_transposed(M, L, MPI_COMM_WORLD, &amp;
Chris@10 174 local_M, local_j_offset, local_L, local_i_offset)
Chris@10 175 </pre>
Chris@10 176 <div class="footnote">
Chris@10 177 <hr>
Chris@10 178 <h4>Footnotes</h4><p class="footnote"><small>[<a name="fn-1" href="#fnd-1">1</a>]</small> Technically, this is because you aren't
Chris@10 179 actually calling the C functions directly. You are calling wrapper
Chris@10 180 functions that translate the communicator with <code>MPI_Comm_f2c</code>
Chris@10 181 before calling the ordinary C interface. This is all done
Chris@10 182 transparently, however, since the <code>fftw3-mpi.f03</code> interface file
Chris@10 183 renames the wrappers so that they are called in Fortran with the same
Chris@10 184 names as the C interface functions.</p>
Chris@10 185
Chris@10 186 <hr></div>
Chris@10 187
Chris@10 188 </body></html>
Chris@10 189