Mercurial > hg > sv-dependency-builds
comparison src/fftw-3.3.3/doc/html/FFTW-MPI-Fortran-Interface.html @ 10:37bf6b4a2645
Add FFTW3
| author | Chris Cannam |
|---|---|
| date | Wed, 20 Mar 2013 15:35:50 +0000 |
| parents | |
| children |
comparison
equal
deleted
inserted
replaced
| 9:c0fb53affa76 | 10:37bf6b4a2645 |
|---|---|
| 1 <html lang="en"> | |
| 2 <head> | |
| 3 <title>FFTW MPI Fortran Interface - FFTW 3.3.3</title> | |
| 4 <meta http-equiv="Content-Type" content="text/html"> | |
| 5 <meta name="description" content="FFTW 3.3.3"> | |
| 6 <meta name="generator" content="makeinfo 4.13"> | |
| 7 <link title="Top" rel="start" href="index.html#Top"> | |
| 8 <link rel="up" href="Distributed_002dmemory-FFTW-with-MPI.html#Distributed_002dmemory-FFTW-with-MPI" title="Distributed-memory FFTW with MPI"> | |
| 9 <link rel="prev" href="FFTW-MPI-Reference.html#FFTW-MPI-Reference" title="FFTW MPI Reference"> | |
| 10 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage"> | |
| 11 <!-- | |
| 12 This manual is for FFTW | |
| 13 (version 3.3.3, 25 November 2012). | |
| 14 | |
| 15 Copyright (C) 2003 Matteo Frigo. | |
| 16 | |
| 17 Copyright (C) 2003 Massachusetts Institute of Technology. | |
| 18 | |
| 19 Permission is granted to make and distribute verbatim copies of | |
| 20 this manual provided the copyright notice and this permission | |
| 21 notice are preserved on all copies. | |
| 22 | |
| 23 Permission is granted to copy and distribute modified versions of | |
| 24 this manual under the conditions for verbatim copying, provided | |
| 25 that the entire resulting derived work is distributed under the | |
| 26 terms of a permission notice identical to this one. | |
| 27 | |
| 28 Permission is granted to copy and distribute translations of this | |
| 29 manual into another language, under the above conditions for | |
| 30 modified versions, except that this permission notice may be | |
| 31 stated in a translation approved by the Free Software Foundation. | |
| 32 --> | |
| 33 <meta http-equiv="Content-Style-Type" content="text/css"> | |
| 34 <style type="text/css"><!-- | |
| 35 pre.display { font-family:inherit } | |
| 36 pre.format { font-family:inherit } | |
| 37 pre.smalldisplay { font-family:inherit; font-size:smaller } | |
| 38 pre.smallformat { font-family:inherit; font-size:smaller } | |
| 39 pre.smallexample { font-size:smaller } | |
| 40 pre.smalllisp { font-size:smaller } | |
| 41 span.sc { font-variant:small-caps } | |
| 42 span.roman { font-family:serif; font-weight:normal; } | |
| 43 span.sansserif { font-family:sans-serif; font-weight:normal; } | |
| 44 --></style> | |
| 45 </head> | |
| 46 <body> | |
| 47 <div class="node"> | |
| 48 <a name="FFTW-MPI-Fortran-Interface"></a> | |
| 49 <p> | |
| 50 Previous: <a rel="previous" accesskey="p" href="FFTW-MPI-Reference.html#FFTW-MPI-Reference">FFTW MPI Reference</a>, | |
| 51 Up: <a rel="up" accesskey="u" href="Distributed_002dmemory-FFTW-with-MPI.html#Distributed_002dmemory-FFTW-with-MPI">Distributed-memory FFTW with MPI</a> | |
| 52 <hr> | |
| 53 </div> | |
| 54 | |
| 55 <h3 class="section">6.13 FFTW MPI Fortran Interface</h3> | |
| 56 | |
| 57 <p><a name="index-Fortran-interface-494"></a> | |
| 58 <a name="index-iso_005fc_005fbinding-495"></a>The FFTW MPI interface is callable from modern Fortran compilers | |
| 59 supporting the Fortran 2003 <code>iso_c_binding</code> standard for calling | |
| 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>, | |
| 61 this means that you can directly call FFTW's C interface from Fortran | |
| 62 with only minor changes in syntax. There are, however, a few things | |
| 63 specific to the MPI interface to keep in mind: | |
| 64 | |
| 65 <ul> | |
| 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 | |
| 67 <code>use, intrinsic :: iso_c_binding</code> as before). The | |
| 68 <code>fftw3-mpi.f03</code> file includes <code>fftw3.f03</code>, so you should | |
| 69 <em>not</em> <code>include</code> them both yourself. (You will also want to | |
| 70 include the MPI header file, usually via <code>include 'mpif.h'</code> or | |
| 71 similar, although though this is not needed by <code>fftw3-mpi.f03</code> | |
| 72 <i>per se</i>.) (To use the ‘<samp><span class="samp">fftwl_</span></samp>’ <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>.) | |
| 73 | |
| 74 <li>Because of the different storage conventions between C and Fortran, | |
| 75 you reverse the order of your array dimensions when passing them to | |
| 76 FFTW (see <a href="Reversing-array-dimensions.html#Reversing-array-dimensions">Reversing array dimensions</a>). This is merely a | |
| 77 difference in notation and incurs no performance overhead. However, | |
| 78 it means that, whereas in C the <em>first</em> dimension is distributed, | |
| 79 in Fortran the <em>last</em> dimension of your array is distributed. | |
| 80 | |
| 81 <li><a name="index-MPI-communicator-496"></a>In Fortran, communicators are stored as <code>integer</code> types; there is | |
| 82 no <code>MPI_Comm</code> type, nor is there any way to access a C | |
| 83 <code>MPI_Comm</code>. Fortunately, this is taken care of for you by the | |
| 84 FFTW Fortran interface: whenever the C interface expects an | |
| 85 <code>MPI_Comm</code> type, you should pass the Fortran communicator as an | |
| 86 <code>integer</code>.<a rel="footnote" href="#fn-1" name="fnd-1"><sup>1</sup></a> | |
| 87 | |
| 88 <li>Because you need to call the ‘<samp><span class="samp">local_size</span></samp>’ function to find out | |
| 89 how much space to allocate, and this may be <em>larger</em> than the | |
| 90 local portion of the array (see <a href="MPI-Data-Distribution.html#MPI-Data-Distribution">MPI Data Distribution</a>), you should | |
| 91 <em>always</em> allocate your arrays dynamically using FFTW's allocation | |
| 92 routines as described in <a href="Allocating-aligned-memory-in-Fortran.html#Allocating-aligned-memory-in-Fortran">Allocating aligned memory in Fortran</a>. | |
| 93 (Coincidentally, this also provides the best performance by | |
| 94 guaranteeding proper data alignment.) | |
| 95 | |
| 96 <li>Because all sizes in the MPI FFTW interface are declared as | |
| 97 <code>ptrdiff_t</code> in C, you should use <code>integer(C_INTPTR_T)</code> in | |
| 98 Fortran (see <a href="FFTW-Fortran-type-reference.html#FFTW-Fortran-type-reference">FFTW Fortran type reference</a>). | |
| 99 | |
| 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 | |
| 101 using the new-array execute functions for all plans, even in the | |
| 102 common case where you are executing the plan on the same arrays for | |
| 103 which the plan was created (see <a href="Plan-execution-in-Fortran.html#Plan-execution-in-Fortran">Plan execution in Fortran</a>). | |
| 104 However, note that in the MPI interface these functions are changed: | |
| 105 <code>fftw_execute_dft</code> becomes <code>fftw_mpi_execute_dft</code>, | |
| 106 etcetera. See <a href="Using-MPI-Plans.html#Using-MPI-Plans">Using MPI Plans</a>. | |
| 107 | |
| 108 </ul> | |
| 109 | |
| 110 <p>For example, here is a Fortran code snippet to perform a distributed | |
| 111 L × M complex DFT in-place. (This assumes you have already | |
| 112 initialized MPI with <code>MPI_init</code> and have also performed | |
| 113 <code>call fftw_mpi_init</code>.) | |
| 114 | |
| 115 <pre class="example"> use, intrinsic :: iso_c_binding | |
| 116 include 'fftw3-mpi.f03' | |
| 117 integer(C_INTPTR_T), parameter :: L = ... | |
| 118 integer(C_INTPTR_T), parameter :: M = ... | |
| 119 type(C_PTR) :: plan, cdata | |
| 120 complex(C_DOUBLE_COMPLEX), pointer :: data(:,:) | |
| 121 integer(C_INTPTR_T) :: i, j, alloc_local, local_M, local_j_offset | |
| 122 | |
| 123 ! <span class="roman">get local data size and allocate (note dimension reversal)</span> | |
| 124 alloc_local = fftw_mpi_local_size_2d(M, L, MPI_COMM_WORLD, & | |
| 125 local_M, local_j_offset) | |
| 126 cdata = fftw_alloc_complex(alloc_local) | |
| 127 call c_f_pointer(cdata, data, [L,local_M]) | |
| 128 | |
| 129 ! <span class="roman">create MPI plan for in-place forward DFT (note dimension reversal)</span> | |
| 130 plan = fftw_mpi_plan_dft_2d(M, L, data, data, MPI_COMM_WORLD, & | |
| 131 FFTW_FORWARD, FFTW_MEASURE) | |
| 132 | |
| 133 ! <span class="roman">initialize data to some function</span> my_function(i,j) | |
| 134 do j = 1, local_M | |
| 135 do i = 1, L | |
| 136 data(i, j) = my_function(i, j + local_j_offset) | |
| 137 end do | |
| 138 end do | |
| 139 | |
| 140 ! <span class="roman">compute transform (as many times as desired)</span> | |
| 141 call fftw_mpi_execute_dft(plan, data, data) | |
| 142 | |
| 143 call fftw_destroy_plan(plan) | |
| 144 call fftw_free(cdata) | |
| 145 </pre> | |
| 146 <p>Note that when we called <code>fftw_mpi_local_size_2d</code> and | |
| 147 <code>fftw_mpi_plan_dft_2d</code> with the dimensions in reversed order, | |
| 148 since a L × M Fortran array is viewed by FFTW in C as a | |
| 149 M × L array. This means that the array was distributed over | |
| 150 the <code>M</code> dimension, the local portion of which is a | |
| 151 L × local_M array in Fortran. (You must <em>not</em> use an | |
| 152 <code>allocate</code> statement to allocate an L × local_M array, | |
| 153 however; you must allocate <code>alloc_local</code> complex numbers, which | |
| 154 may be greater than <code>L * local_M</code>, in order to reserve space for | |
| 155 intermediate steps of the transform.) Finally, we mention that | |
| 156 because C's array indices are zero-based, the <code>local_j_offset</code> | |
| 157 argument can conveniently be interpreted as an offset in the 1-based | |
| 158 <code>j</code> index (rather than as a starting index as in C). | |
| 159 | |
| 160 <p>If instead you had used the <code>ior(FFTW_MEASURE, | |
| 161 FFTW_MPI_TRANSPOSED_OUT)</code> flag, the output of the transform would be a | |
| 162 transposed M × local_L array, associated with the <em>same</em> | |
| 163 <code>cdata</code> allocation (since the transform is in-place), and which | |
| 164 you could declare with: | |
| 165 | |
| 166 <pre class="example"> complex(C_DOUBLE_COMPLEX), pointer :: tdata(:,:) | |
| 167 ... | |
| 168 call c_f_pointer(cdata, tdata, [M,local_L]) | |
| 169 </pre> | |
| 170 <p>where <code>local_L</code> would have been obtained by changing the | |
| 171 <code>fftw_mpi_local_size_2d</code> call to: | |
| 172 | |
| 173 <pre class="example"> alloc_local = fftw_mpi_local_size_2d_transposed(M, L, MPI_COMM_WORLD, & | |
| 174 local_M, local_j_offset, local_L, local_i_offset) | |
| 175 </pre> | |
| 176 <div class="footnote"> | |
| 177 <hr> | |
| 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 | |
| 179 actually calling the C functions directly. You are calling wrapper | |
| 180 functions that translate the communicator with <code>MPI_Comm_f2c</code> | |
| 181 before calling the ordinary C interface. This is all done | |
| 182 transparently, however, since the <code>fftw3-mpi.f03</code> interface file | |
| 183 renames the wrappers so that they are called in Fortran with the same | |
| 184 names as the C interface functions.</p> | |
| 185 | |
| 186 <hr></div> | |
| 187 | |
| 188 </body></html> | |
| 189 |