This commit is contained in:
2025-07-12 12:17:44 +03:00
parent c759f60ff7
commit 792e1b937a
3507 changed files with 492613 additions and 0 deletions

View File

@@ -0,0 +1,26 @@
BFNNCONV_SRC = bfnnconv.pl m-ascii.pl m-html.pl m-info.pl m-lout.pl m-post.pl
FAQ = fftw-faq.ascii fftw-faq.html
EXTRA_DIST = fftw-faq.bfnn $(FAQ) $(BFNNCONV_SRC) html.refs
html.refs2: html.refs
cp -f ${srcdir}/html.refs html.refs2
fftw-faq.ascii: fftw-faq.html
# generates both fftw-faq.html and fftw-faq.ascii
fftw-faq.html: $(BFNNCONV_SRC) fftw-faq.bfnn html.refs2
@echo converting...
perl -I${srcdir} ${srcdir}/bfnnconv.pl < ${srcdir}/fftw-faq.bfnn
@echo converting again...
perl -I${srcdir} ${srcdir}/bfnnconv.pl < ${srcdir}/fftw-faq.bfnn
rm -f fftw-faq.ascii
mv stdin.ascii fftw-faq.ascii
rm -rf fftw-faq.html
mv -f stdin.html fftw-faq.html
faq: $(FAQ)
clean-local:
rm -f *~ core a.out *.lout *.ps *.info *.ascii *.xrefdb *.post
rm -rf *.html html.refs2

View File

@@ -0,0 +1,506 @@
# Makefile.in generated by automake 1.16.3 from Makefile.am.
# @configure_input@
# Copyright (C) 1994-2020 Free Software Foundation, Inc.
# This Makefile.in is free software; the Free Software Foundation
# gives unlimited permission to copy and/or distribute it,
# with or without modifications, as long as this notice is preserved.
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY, to the extent permitted by law; without
# even the implied warranty of MERCHANTABILITY or FITNESS FOR A
# PARTICULAR PURPOSE.
@SET_MAKE@
VPATH = @srcdir@
am__is_gnu_make = { \
if test -z '$(MAKELEVEL)'; then \
false; \
elif test -n '$(MAKE_HOST)'; then \
true; \
elif test -n '$(MAKE_VERSION)' && test -n '$(CURDIR)'; then \
true; \
else \
false; \
fi; \
}
am__make_running_with_option = \
case $${target_option-} in \
?) ;; \
*) echo "am__make_running_with_option: internal error: invalid" \
"target option '$${target_option-}' specified" >&2; \
exit 1;; \
esac; \
has_opt=no; \
sane_makeflags=$$MAKEFLAGS; \
if $(am__is_gnu_make); then \
sane_makeflags=$$MFLAGS; \
else \
case $$MAKEFLAGS in \
*\\[\ \ ]*) \
bs=\\; \
sane_makeflags=`printf '%s\n' "$$MAKEFLAGS" \
| sed "s/$$bs$$bs[$$bs $$bs ]*//g"`;; \
esac; \
fi; \
skip_next=no; \
strip_trailopt () \
{ \
flg=`printf '%s\n' "$$flg" | sed "s/$$1.*$$//"`; \
}; \
for flg in $$sane_makeflags; do \
test $$skip_next = yes && { skip_next=no; continue; }; \
case $$flg in \
*=*|--*) continue;; \
-*I) strip_trailopt 'I'; skip_next=yes;; \
-*I?*) strip_trailopt 'I';; \
-*O) strip_trailopt 'O'; skip_next=yes;; \
-*O?*) strip_trailopt 'O';; \
-*l) strip_trailopt 'l'; skip_next=yes;; \
-*l?*) strip_trailopt 'l';; \
-[dEDm]) skip_next=yes;; \
-[JT]) skip_next=yes;; \
esac; \
case $$flg in \
*$$target_option*) has_opt=yes; break;; \
esac; \
done; \
test $$has_opt = yes
am__make_dryrun = (target_option=n; $(am__make_running_with_option))
am__make_keepgoing = (target_option=k; $(am__make_running_with_option))
pkgdatadir = $(datadir)/@PACKAGE@
pkgincludedir = $(includedir)/@PACKAGE@
pkglibdir = $(libdir)/@PACKAGE@
pkglibexecdir = $(libexecdir)/@PACKAGE@
am__cd = CDPATH="$${ZSH_VERSION+.}$(PATH_SEPARATOR)" && cd
install_sh_DATA = $(install_sh) -c -m 644
install_sh_PROGRAM = $(install_sh) -c
install_sh_SCRIPT = $(install_sh) -c
INSTALL_HEADER = $(INSTALL_DATA)
transform = $(program_transform_name)
NORMAL_INSTALL = :
PRE_INSTALL = :
POST_INSTALL = :
NORMAL_UNINSTALL = :
PRE_UNINSTALL = :
POST_UNINSTALL = :
build_triplet = @build@
host_triplet = @host@
subdir = doc/FAQ
ACLOCAL_M4 = $(top_srcdir)/aclocal.m4
am__aclocal_m4_deps = $(top_srcdir)/m4/acx_mpi.m4 \
$(top_srcdir)/m4/acx_pthread.m4 \
$(top_srcdir)/m4/ax_cc_maxopt.m4 \
$(top_srcdir)/m4/ax_check_compiler_flags.m4 \
$(top_srcdir)/m4/ax_compiler_vendor.m4 \
$(top_srcdir)/m4/ax_gcc_aligns_stack.m4 \
$(top_srcdir)/m4/ax_gcc_version.m4 \
$(top_srcdir)/m4/ax_openmp.m4 $(top_srcdir)/m4/libtool.m4 \
$(top_srcdir)/m4/ltoptions.m4 $(top_srcdir)/m4/ltsugar.m4 \
$(top_srcdir)/m4/ltversion.m4 $(top_srcdir)/m4/lt~obsolete.m4 \
$(top_srcdir)/configure.ac
am__configure_deps = $(am__aclocal_m4_deps) $(CONFIGURE_DEPENDENCIES) \
$(ACLOCAL_M4)
DIST_COMMON = $(srcdir)/Makefile.am $(am__DIST_COMMON)
mkinstalldirs = $(install_sh) -d
CONFIG_HEADER = $(top_builddir)/config.h
CONFIG_CLEAN_FILES =
CONFIG_CLEAN_VPATH_FILES =
AM_V_P = $(am__v_P_@AM_V@)
am__v_P_ = $(am__v_P_@AM_DEFAULT_V@)
am__v_P_0 = false
am__v_P_1 = :
AM_V_GEN = $(am__v_GEN_@AM_V@)
am__v_GEN_ = $(am__v_GEN_@AM_DEFAULT_V@)
am__v_GEN_0 = @echo " GEN " $@;
am__v_GEN_1 =
AM_V_at = $(am__v_at_@AM_V@)
am__v_at_ = $(am__v_at_@AM_DEFAULT_V@)
am__v_at_0 = @
am__v_at_1 =
SOURCES =
DIST_SOURCES =
am__can_run_installinfo = \
case $$AM_UPDATE_INFO_DIR in \
n|no|NO) false;; \
*) (install-info --version) >/dev/null 2>&1;; \
esac
am__tagged_files = $(HEADERS) $(SOURCES) $(TAGS_FILES) $(LISP)
am__DIST_COMMON = $(srcdir)/Makefile.in
DISTFILES = $(DIST_COMMON) $(DIST_SOURCES) $(TEXINFOS) $(EXTRA_DIST)
ACLOCAL = @ACLOCAL@
ALLOCA = @ALLOCA@
ALTIVEC_CFLAGS = @ALTIVEC_CFLAGS@
AMTAR = @AMTAR@
AM_DEFAULT_VERBOSITY = @AM_DEFAULT_VERBOSITY@
AR = @AR@
AS = @AS@
AUTOCONF = @AUTOCONF@
AUTOHEADER = @AUTOHEADER@
AUTOMAKE = @AUTOMAKE@
AVX2_CFLAGS = @AVX2_CFLAGS@
AVX512_CFLAGS = @AVX512_CFLAGS@
AVX_128_FMA_CFLAGS = @AVX_128_FMA_CFLAGS@
AVX_CFLAGS = @AVX_CFLAGS@
AWK = @AWK@
CC = @CC@
CCDEPMODE = @CCDEPMODE@
CFLAGS = @CFLAGS@
CHECK_PL_OPTS = @CHECK_PL_OPTS@
CPP = @CPP@
CPPFLAGS = @CPPFLAGS@
CYGPATH_W = @CYGPATH_W@
C_FFTW_R2R_KIND = @C_FFTW_R2R_KIND@
C_MPI_FINT = @C_MPI_FINT@
DEFS = @DEFS@
DEPDIR = @DEPDIR@
DLLTOOL = @DLLTOOL@
DSYMUTIL = @DSYMUTIL@
DUMPBIN = @DUMPBIN@
ECHO_C = @ECHO_C@
ECHO_N = @ECHO_N@
ECHO_T = @ECHO_T@
EGREP = @EGREP@
EXEEXT = @EXEEXT@
F77 = @F77@
FFLAGS = @FFLAGS@
FGREP = @FGREP@
FLIBS = @FLIBS@
GREP = @GREP@
INDENT = @INDENT@
INSTALL = @INSTALL@
INSTALL_DATA = @INSTALL_DATA@
INSTALL_PROGRAM = @INSTALL_PROGRAM@
INSTALL_SCRIPT = @INSTALL_SCRIPT@
INSTALL_STRIP_PROGRAM = @INSTALL_STRIP_PROGRAM@
KCVI_CFLAGS = @KCVI_CFLAGS@
LD = @LD@
LDFLAGS = @LDFLAGS@
LIBOBJS = @LIBOBJS@
LIBQUADMATH = @LIBQUADMATH@
LIBS = @LIBS@
LIBTOOL = @LIBTOOL@
LIPO = @LIPO@
LN_S = @LN_S@
LTLIBOBJS = @LTLIBOBJS@
LT_SYS_LIBRARY_PATH = @LT_SYS_LIBRARY_PATH@
MAINT = @MAINT@
MAKEINFO = @MAKEINFO@
MANIFEST_TOOL = @MANIFEST_TOOL@
MKDIR_P = @MKDIR_P@
MPICC = @MPICC@
MPILIBS = @MPILIBS@
MPIRUN = @MPIRUN@
NEON_CFLAGS = @NEON_CFLAGS@
NM = @NM@
NMEDIT = @NMEDIT@
OBJDUMP = @OBJDUMP@
OBJEXT = @OBJEXT@
OCAMLBUILD = @OCAMLBUILD@
OPENMP_CFLAGS = @OPENMP_CFLAGS@
OTOOL = @OTOOL@
OTOOL64 = @OTOOL64@
PACKAGE = @PACKAGE@
PACKAGE_BUGREPORT = @PACKAGE_BUGREPORT@
PACKAGE_NAME = @PACKAGE_NAME@
PACKAGE_STRING = @PACKAGE_STRING@
PACKAGE_TARNAME = @PACKAGE_TARNAME@
PACKAGE_URL = @PACKAGE_URL@
PACKAGE_VERSION = @PACKAGE_VERSION@
PATH_SEPARATOR = @PATH_SEPARATOR@
POW_LIB = @POW_LIB@
PRECISION = @PRECISION@
PREC_SUFFIX = @PREC_SUFFIX@
PTHREAD_CC = @PTHREAD_CC@
PTHREAD_CFLAGS = @PTHREAD_CFLAGS@
PTHREAD_LIBS = @PTHREAD_LIBS@
RANLIB = @RANLIB@
SED = @SED@
SET_MAKE = @SET_MAKE@
SHARED_VERSION_INFO = @SHARED_VERSION_INFO@
SHELL = @SHELL@
SSE2_CFLAGS = @SSE2_CFLAGS@
STACK_ALIGN_CFLAGS = @STACK_ALIGN_CFLAGS@
STRIP = @STRIP@
THREADLIBS = @THREADLIBS@
VERSION = @VERSION@
VSX_CFLAGS = @VSX_CFLAGS@
abs_builddir = @abs_builddir@
abs_srcdir = @abs_srcdir@
abs_top_builddir = @abs_top_builddir@
abs_top_srcdir = @abs_top_srcdir@
ac_ct_AR = @ac_ct_AR@
ac_ct_CC = @ac_ct_CC@
ac_ct_DUMPBIN = @ac_ct_DUMPBIN@
ac_ct_F77 = @ac_ct_F77@
acx_pthread_config = @acx_pthread_config@
am__include = @am__include@
am__leading_dot = @am__leading_dot@
am__quote = @am__quote@
am__tar = @am__tar@
am__untar = @am__untar@
bindir = @bindir@
build = @build@
build_alias = @build_alias@
build_cpu = @build_cpu@
build_os = @build_os@
build_vendor = @build_vendor@
builddir = @builddir@
datadir = @datadir@
datarootdir = @datarootdir@
docdir = @docdir@
dvidir = @dvidir@
exec_prefix = @exec_prefix@
host = @host@
host_alias = @host_alias@
host_cpu = @host_cpu@
host_os = @host_os@
host_vendor = @host_vendor@
htmldir = @htmldir@
includedir = @includedir@
infodir = @infodir@
install_sh = @install_sh@
libdir = @libdir@
libexecdir = @libexecdir@
localedir = @localedir@
localstatedir = @localstatedir@
mandir = @mandir@
mkdir_p = @mkdir_p@
oldincludedir = @oldincludedir@
pdfdir = @pdfdir@
prefix = @prefix@
program_transform_name = @program_transform_name@
psdir = @psdir@
runstatedir = @runstatedir@
sbindir = @sbindir@
sharedstatedir = @sharedstatedir@
srcdir = @srcdir@
sysconfdir = @sysconfdir@
target_alias = @target_alias@
top_build_prefix = @top_build_prefix@
top_builddir = @top_builddir@
top_srcdir = @top_srcdir@
BFNNCONV_SRC = bfnnconv.pl m-ascii.pl m-html.pl m-info.pl m-lout.pl m-post.pl
FAQ = fftw-faq.ascii fftw-faq.html
EXTRA_DIST = fftw-faq.bfnn $(FAQ) $(BFNNCONV_SRC) html.refs
all: all-am
.SUFFIXES:
$(srcdir)/Makefile.in: @MAINTAINER_MODE_TRUE@ $(srcdir)/Makefile.am $(am__configure_deps)
@for dep in $?; do \
case '$(am__configure_deps)' in \
*$$dep*) \
( cd $(top_builddir) && $(MAKE) $(AM_MAKEFLAGS) am--refresh ) \
&& { if test -f $@; then exit 0; else break; fi; }; \
exit 1;; \
esac; \
done; \
echo ' cd $(top_srcdir) && $(AUTOMAKE) --gnu doc/FAQ/Makefile'; \
$(am__cd) $(top_srcdir) && \
$(AUTOMAKE) --gnu doc/FAQ/Makefile
Makefile: $(srcdir)/Makefile.in $(top_builddir)/config.status
@case '$?' in \
*config.status*) \
cd $(top_builddir) && $(MAKE) $(AM_MAKEFLAGS) am--refresh;; \
*) \
echo ' cd $(top_builddir) && $(SHELL) ./config.status $(subdir)/$@ $(am__maybe_remake_depfiles)'; \
cd $(top_builddir) && $(SHELL) ./config.status $(subdir)/$@ $(am__maybe_remake_depfiles);; \
esac;
$(top_builddir)/config.status: $(top_srcdir)/configure $(CONFIG_STATUS_DEPENDENCIES)
cd $(top_builddir) && $(MAKE) $(AM_MAKEFLAGS) am--refresh
$(top_srcdir)/configure: @MAINTAINER_MODE_TRUE@ $(am__configure_deps)
cd $(top_builddir) && $(MAKE) $(AM_MAKEFLAGS) am--refresh
$(ACLOCAL_M4): @MAINTAINER_MODE_TRUE@ $(am__aclocal_m4_deps)
cd $(top_builddir) && $(MAKE) $(AM_MAKEFLAGS) am--refresh
$(am__aclocal_m4_deps):
mostlyclean-libtool:
-rm -f *.lo
clean-libtool:
-rm -rf .libs _libs
tags TAGS:
ctags CTAGS:
cscope cscopelist:
distdir: $(BUILT_SOURCES)
$(MAKE) $(AM_MAKEFLAGS) distdir-am
distdir-am: $(DISTFILES)
@srcdirstrip=`echo "$(srcdir)" | sed 's/[].[^$$\\*]/\\\\&/g'`; \
topsrcdirstrip=`echo "$(top_srcdir)" | sed 's/[].[^$$\\*]/\\\\&/g'`; \
list='$(DISTFILES)'; \
dist_files=`for file in $$list; do echo $$file; done | \
sed -e "s|^$$srcdirstrip/||;t" \
-e "s|^$$topsrcdirstrip/|$(top_builddir)/|;t"`; \
case $$dist_files in \
*/*) $(MKDIR_P) `echo "$$dist_files" | \
sed '/\//!d;s|^|$(distdir)/|;s,/[^/]*$$,,' | \
sort -u` ;; \
esac; \
for file in $$dist_files; do \
if test -f $$file || test -d $$file; then d=.; else d=$(srcdir); fi; \
if test -d $$d/$$file; then \
dir=`echo "/$$file" | sed -e 's,/[^/]*$$,,'`; \
if test -d "$(distdir)/$$file"; then \
find "$(distdir)/$$file" -type d ! -perm -700 -exec chmod u+rwx {} \;; \
fi; \
if test -d $(srcdir)/$$file && test $$d != $(srcdir); then \
cp -fpR $(srcdir)/$$file "$(distdir)$$dir" || exit 1; \
find "$(distdir)/$$file" -type d ! -perm -700 -exec chmod u+rwx {} \;; \
fi; \
cp -fpR $$d/$$file "$(distdir)$$dir" || exit 1; \
else \
test -f "$(distdir)/$$file" \
|| cp -p $$d/$$file "$(distdir)/$$file" \
|| exit 1; \
fi; \
done
check-am: all-am
check: check-am
all-am: Makefile
installdirs:
install: install-am
install-exec: install-exec-am
install-data: install-data-am
uninstall: uninstall-am
install-am: all-am
@$(MAKE) $(AM_MAKEFLAGS) install-exec-am install-data-am
installcheck: installcheck-am
install-strip:
if test -z '$(STRIP)'; then \
$(MAKE) $(AM_MAKEFLAGS) INSTALL_PROGRAM="$(INSTALL_STRIP_PROGRAM)" \
install_sh_PROGRAM="$(INSTALL_STRIP_PROGRAM)" INSTALL_STRIP_FLAG=-s \
install; \
else \
$(MAKE) $(AM_MAKEFLAGS) INSTALL_PROGRAM="$(INSTALL_STRIP_PROGRAM)" \
install_sh_PROGRAM="$(INSTALL_STRIP_PROGRAM)" INSTALL_STRIP_FLAG=-s \
"INSTALL_PROGRAM_ENV=STRIPPROG='$(STRIP)'" install; \
fi
mostlyclean-generic:
clean-generic:
distclean-generic:
-test -z "$(CONFIG_CLEAN_FILES)" || rm -f $(CONFIG_CLEAN_FILES)
-test . = "$(srcdir)" || test -z "$(CONFIG_CLEAN_VPATH_FILES)" || rm -f $(CONFIG_CLEAN_VPATH_FILES)
maintainer-clean-generic:
@echo "This command is intended for maintainers to use"
@echo "it deletes files that may require special tools to rebuild."
clean: clean-am
clean-am: clean-generic clean-libtool clean-local mostlyclean-am
distclean: distclean-am
-rm -f Makefile
distclean-am: clean-am distclean-generic
dvi: dvi-am
dvi-am:
html: html-am
html-am:
info: info-am
info-am:
install-data-am:
install-dvi: install-dvi-am
install-dvi-am:
install-exec-am:
install-html: install-html-am
install-html-am:
install-info: install-info-am
install-info-am:
install-man:
install-pdf: install-pdf-am
install-pdf-am:
install-ps: install-ps-am
install-ps-am:
installcheck-am:
maintainer-clean: maintainer-clean-am
-rm -f Makefile
maintainer-clean-am: distclean-am maintainer-clean-generic
mostlyclean: mostlyclean-am
mostlyclean-am: mostlyclean-generic mostlyclean-libtool
pdf: pdf-am
pdf-am:
ps: ps-am
ps-am:
uninstall-am:
.MAKE: install-am install-strip
.PHONY: all all-am check check-am clean clean-generic clean-libtool \
clean-local cscopelist-am ctags-am distclean distclean-generic \
distclean-libtool distdir dvi dvi-am html html-am info info-am \
install install-am install-data install-data-am install-dvi \
install-dvi-am install-exec install-exec-am install-html \
install-html-am install-info install-info-am install-man \
install-pdf install-pdf-am install-ps install-ps-am \
install-strip installcheck installcheck-am installdirs \
maintainer-clean maintainer-clean-generic mostlyclean \
mostlyclean-generic mostlyclean-libtool pdf pdf-am ps ps-am \
tags-am uninstall uninstall-am
.PRECIOUS: Makefile
html.refs2: html.refs
cp -f ${srcdir}/html.refs html.refs2
fftw-faq.ascii: fftw-faq.html
# generates both fftw-faq.html and fftw-faq.ascii
fftw-faq.html: $(BFNNCONV_SRC) fftw-faq.bfnn html.refs2
@echo converting...
perl -I${srcdir} ${srcdir}/bfnnconv.pl < ${srcdir}/fftw-faq.bfnn
@echo converting again...
perl -I${srcdir} ${srcdir}/bfnnconv.pl < ${srcdir}/fftw-faq.bfnn
rm -f fftw-faq.ascii
mv stdin.ascii fftw-faq.ascii
rm -rf fftw-faq.html
mv -f stdin.html fftw-faq.html
faq: $(FAQ)
clean-local:
rm -f *~ core a.out *.lout *.ps *.info *.ascii *.xrefdb *.post
rm -rf *.html html.refs2
# Tell versions [3.59,3.63) of GNU make to not export all variables.
# Otherwise a system limit (for SysV at least) may be exceeded.
.NOEXPORT:

298
fftw-3.3.10/doc/FAQ/bfnnconv.pl Executable file
View File

@@ -0,0 +1,298 @@
#!/usr/bin/perl --
# Copyright (C) 1993-1995 Ian Jackson.
# This file is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation; either version 2, or (at your option)
# any later version.
# It is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
# You should have received a copy of the GNU General Public License
# along with GNU Emacs; see the file COPYING. If not, write to
# the Free Software Foundation, Inc., 59 Temple Place - Suite 330,
# Boston, MA 02111-1307, USA.
# (Note: I do not consider works produced using these BFNN processing
# tools to be derivative works of the tools, so they are NOT covered
# by the GPL. However, I would appreciate it if you credited me if
# appropriate in any documents you format using BFNN.)
@outputs=('ascii','info','html');
while ($ARGV[0] =~ m/^\-/) {
$_= shift(@ARGV);
if (m/^-only/) {
@outputs= (shift(@ARGV));
} else {
warn "unknown option `$_' ignored";
}
}
$prefix= $ARGV[0];
$prefix= 'stdin' unless length($prefix);
$prefix =~ s/\.bfnn$//;
if (open(O,"$prefix.xrefdb")) {
@xrefdb= <O>;
close(O);
} else {
warn "no $prefix.xrefdb ($!)";
}
$section= -1;
for $thisxr (@xrefdb) {
$_= $thisxr;
chop;
if (m/^Q (\w+) ((\d+)\.(\d+)) (.*)$/) {
$qrefn{$1}= $2;
$qreft{$1}= $5;
$qn2ref{$3,$4}= $1;
$maxsection= $3;
$maxquestion[$3]= $4;
} elsif (m/^S (\d+) /) {
$maxsection= $1;
$sn2title{$1}=$';
}
}
open(U,">$prefix.xrefdb-new");
for $x (@outputs) { require("m-$x.pl"); }
&call('init');
while (<>) {
chop;
next if m/^\\comment\b/;
if (!m/\S/) {
&call('endpara');
next;
}
if (s/^\\section +//) {
$line= $_;
$section++; $question=0;
print U "S $section $line\n";
$|=1; print "S$section",' 'x10,"\r"; $|=0;
&call('endpara');
&call('startmajorheading',"$section",
"Section $section",
$section<$maxsection ? "Section ".($section+1) : '',
$section>1 ? 'Section '.($section-1) : 'Top');
&text($line);
&call('endmajorheading');
if ($section) {
&call('endpara');
&call('startindex');
for $thisxr (@xrefdb) {
$_= $thisxr;
chop;
if (m/^Q (\w+) (\d+)\.(\d+) (.*)$/) {
$ref= $1; $num1= $2; $num2= $3; $text= $4;
next unless $num1 == $section;
&call('startindexitem',$ref,"Q$num1.$num2","Question $num1.$num2");
&text($text);
&call('endindexitem');
}
}
&call('endindex');
}
} elsif (s/^\\question \d{2}[a-z]{3}((:\w+)?) +//) {
$line= $_;
$question++;
$qrefstring= $1;
$qrefstring= "q_${section}_$question" unless $qrefstring =~ s/^://;
print U "Q $qrefstring $section.$question $line\n";
$|=1; print "Q$section.$question",' 'x10,"\r"; $|=0;
&call('endpara');
&call('startminorheading',$qrefstring,
"Question $section.$question",
$question < $maxquestion[$section] ? "Question $section.".($question+1) :
$section < $maxsection ? "Question ".($section+1).".1" : '',
$question > 1 ? "Question $section.".($question-1) :
$section > 1 ? "Question ".($section-1).'.'.($maxquestion[$section-1]) :
'Top',
"Section $section");
&text("Question $section.$question. $line");
&call('endminorheading');
} elsif (s/^\\only +//) {
@saveoutputs= @outputs;
@outputs=();
for $x (split(/\s+/,$_)) {
push(@outputs,$x) if grep($x eq $_, @saveoutputs);
}
} elsif (s/^\\endonly$//) {
@outputs= @saveoutputs;
} elsif (s/^\\copyto +//) {
$fh= $';
while(<>) {
last if m/^\\endcopy$/;
while (s/^([^\`]*)\`//) {
print $fh $1;
m/([^\\])\`/ || warn "`$_'";
$_= $';
$cmd= $`.$1;
$it= `$cmd`; chop $it;
print $fh $it;
}
print $fh $_;
}
} elsif (m/\\index$/) {
&call('startindex');
for $thisxr (@xrefdb) {
$_= $thisxr;
chop;
if (m/^Q (\w+) (\d+\.\d+) (.*)$/) {
$ref= $1; $num= $2; $text= $3;
&call('startindexitem',$ref,"Q$num","Question $num");
&text($text);
&call('endindexitem');
} elsif (m/^S (\d+) (.*)$/) {
$num= $1; $text= $2;
next unless $num;
&call('startindexmainitem',"s_$num",
"Section $num.","Section $num");
&text($text);
&call('endindexitem');
} else {
warn $_;
}
}
&call('endindex');
} elsif (m/^\\call-(\w+) +(\w+)\s*(.*)$/) {
$fn= $1.'_'.$2;
eval { &$fn($3); };
warn $@ if length($@);
} elsif (m/^\\call +(\w+)\s*(.*)$/) {
eval { &call($1,$2); };
warn $@ if length($@);
} elsif (s/^\\set +(\w+)\s*//) {
$svalue= $'; $svari= $1;
eval("\$user_$svari=\$svalue"); $@ && warn "setting $svalue failed: $@\n";
} elsif (m/^\\verbatim$/) {
&call('startverbatim');
while (<>) {
chop;
last if m/^\\endverbatim$/;
&call('verbatim',$_);
}
&call('endverbatim');
} else {
s/\.$/\. /;
&text($_." ");
}
}
print ' 'x25,"\r";
&call('finish');
rename("$prefix.xrefdb-new","$prefix.xrefdb") || warn "rename xrefdb: $!";
exit 0;
sub text {
local($in,$rhs,$word,$refn,$reft,$fn,$style);
$in= "$holdover$_[0]";
$holdover= '';
while ($in =~ m/\\/) {
#print STDERR ">$`##$'\n";
$rhs=$';
&call('text',$`);
$_= $rhs;
if (m/^\w+ $/) {
$holdover= "\\$&";
$in= '';
} elsif (s/^fn\s+([^\s\\]*\w)//) {
$in= $_;
$word= $1;
&call('courier');
&call('text',$word);
&call('endcourier');
} elsif (s/^tab\s+(\d+)\s+//) {
$in= $_; &call('tab',$1);
} elsif (s/^nl\s+//) {
$in= $_; &call('newline');
} elsif (s/^qref\s+(\w+)//) {
$refn= $qrefn{$1};
$reft= $qreft{$1};
if (!length($refn)) {
warn "unknown question `$1'";
}
$in= "$`\\pageref:$1:$refn:$reft\\endpageref.$_";
} elsif (s/^pageref:(\w+):([^:\n]+)://) {
$in= $_;
&call('pageref',$1,$2);
} elsif (s/^endpageref\.//) {
$in= $_; &call('endpageref');
} elsif (s/^(\w+)\{//) {
$in= $_; $fn= $1;
eval { &call("$fn"); };
if (length($@)) { warn $@; $fn= 'x'; }
push(@styles,$fn);
} elsif (s/^\}//) {
$in= $_;
$fn= pop(@styles);
if ($fn ne 'x') { &call("end$fn"); }
} elsif (s/^\\//) {
$in= $_;
&call('text',"\\");
} elsif (s,^(\w+)\s+([-A-Za-z0-9.\@:/]*\w),,) {
#print STDERR "**$&**$_\n";
$in= $_;
$style=$1; $word= $2;
&call($style);
&call('text',$word);
&call("end$style");
} else {
warn "unknown control `\\$_'";
$in= $_;
}
}
&call('text',$in);
}
sub call {
local ($fnbase, @callargs) = @_;
local ($coutput);
for $coutput (@outputs) {
if ($fnbase eq 'text' && eval("\@${coutput}_cmds")) {
#print STDERR "special handling text (@callargs) for $coutput\n";
$evstrg= "\$${coutput}_args[\$#${coutput}_args].=\"\@callargs\"";
eval($evstrg);
length($@) && warn "call adding for $coutput (($evstrg)): $@";
} else {
$fntc= $coutput.'_'.$fnbase;
&$fntc(@callargs);
}
}
}
sub recurse {
local (@outputs) = $coutput;
local ($holdover);
&text($_[0]);
}
sub arg {
#print STDERR "arg($_[0]) from $coutput\n";
$cmd= $_[0];
eval("push(\@${coutput}_cmds,\$cmd); push(\@${coutput}_args,'')");
length($@) && warn "arg setting up for $coutput: $@";
}
sub endarg {
#print STDERR "endarg($_[0]) from $coutput\n";
$evstrg= "\$${coutput}_cmd= \$cmd= pop(\@${coutput}_cmds); ".
"\$${coutput}_arg= \$arg= pop(\@${coutput}_args); ";
eval($evstrg);
length($@) && warn "endarg extracting for $coutput (($evstrg)): $@";
#print STDERR ">call $coutput $cmd $arg< (($evstrg))\n";
$evstrg= "&${coutput}_do_${cmd}(\$arg)";
eval($evstrg);
length($@) && warn "endarg running ${coutput}_do_${cmd} (($evstrg)): $@";
}

View File

@@ -0,0 +1,842 @@
FFTW FREQUENTLY ASKED QUESTIONS WITH ANSWERS
14 Sep 2021
Matteo Frigo
Steven G. Johnson
<fftw@fftw.org>
This is the list of Frequently Asked Questions about FFTW, a collection of
fast C routines for computing the Discrete Fourier Transform in one or
more dimensions.
===============================================================================
Index
Section 1. Introduction and General Information
Q1.1 What is FFTW?
Q1.2 How do I obtain FFTW?
Q1.3 Is FFTW free software?
Q1.4 What is this about non-free licenses?
Q1.5 In the West? I thought MIT was in the East?
Section 2. Installing FFTW
Q2.1 Which systems does FFTW run on?
Q2.2 Does FFTW run on Windows?
Q2.3 My compiler has trouble with FFTW.
Q2.4 FFTW does not compile on Solaris, complaining about const.
Q2.5 What's the difference between --enable-3dnow and --enable-k7?
Q2.6 What's the difference between the fma and the non-fma versions?
Q2.7 Which language is FFTW written in?
Q2.8 Can I call FFTW from Fortran?
Q2.9 Can I call FFTW from C++?
Q2.10 Why isn't FFTW written in Fortran/C++?
Q2.11 How do I compile FFTW to run in single precision?
Q2.12 --enable-k7 does not work on x86-64
Section 3. Using FFTW
Q3.1 Why not support the FFTW 2 interface in FFTW 3?
Q3.2 Why do FFTW 3 plans encapsulate the input/output arrays and not ju
Q3.3 FFTW seems really slow.
Q3.4 FFTW slows down after repeated calls.
Q3.5 An FFTW routine is crashing when I call it.
Q3.6 My Fortran program crashes when calling FFTW.
Q3.7 FFTW gives results different from my old FFT.
Q3.8 FFTW gives different results between runs
Q3.9 Can I save FFTW's plans?
Q3.10 Why does your inverse transform return a scaled result?
Q3.11 How can I make FFTW put the origin (zero frequency) at the center
Q3.12 How do I FFT an image/audio file in *foobar* format?
Q3.13 My program does not link (on Unix).
Q3.14 I included your header, but linking still fails.
Q3.15 My program crashes, complaining about stack space.
Q3.16 FFTW seems to have a memory leak.
Q3.17 The output of FFTW's transform is all zeros.
Q3.18 How do I call FFTW from the Microsoft language du jour?
Q3.19 Can I compute only a subset of the DFT outputs?
Q3.20 Can I use FFTW's routines for in-place and out-of-place matrix tra
Section 4. Internals of FFTW
Q4.1 How does FFTW work?
Q4.2 Why is FFTW so fast?
Section 5. Known bugs
Q5.1 FFTW 1.1 crashes in rfftwnd on Linux.
Q5.2 The MPI transforms in FFTW 1.2 give incorrect results/leak memory.
Q5.3 The test programs in FFTW 1.2.1 fail when I change FFTW to use sin
Q5.4 The test program in FFTW 1.2.1 fails for n > 46340.
Q5.5 The threaded code fails on Linux Redhat 5.0
Q5.6 FFTW 2.0's rfftwnd fails for rank > 1 transforms with a final dime
Q5.7 FFTW 2.0's complex transforms give the wrong results with prime fa
Q5.8 FFTW 2.1.1's MPI test programs crash with MPICH.
Q5.9 FFTW 2.1.2's multi-threaded transforms don't work on AIX.
Q5.10 FFTW 2.1.2's complex transforms give incorrect results for large p
Q5.11 FFTW 2.1.3's multi-threaded transforms don't give any speedup on S
Q5.12 FFTW 2.1.3 crashes on AIX.
===============================================================================
Section 1. Introduction and General Information
Q1.1 What is FFTW?
Q1.2 How do I obtain FFTW?
Q1.3 Is FFTW free software?
Q1.4 What is this about non-free licenses?
Q1.5 In the West? I thought MIT was in the East?
-------------------------------------------------------------------------------
Question 1.1. What is FFTW?
FFTW is a free collection of fast C routines for computing the Discrete
Fourier Transform in one or more dimensions. It includes complex, real,
symmetric, and parallel transforms, and can handle arbitrary array sizes
efficiently. FFTW is typically faster than other publically-available FFT
implementations, and is even competitive with vendor-tuned libraries.
(See our web page for extensive benchmarks.) To achieve this performance,
FFTW uses novel code-generation and runtime self-optimization techniques
(along with many other tricks).
-------------------------------------------------------------------------------
Question 1.2. How do I obtain FFTW?
FFTW can be found at the FFTW web page. You can also retrieve it from
ftp.fftw.org in /pub/fftw.
-------------------------------------------------------------------------------
Question 1.3. Is FFTW free software?
Starting with version 1.3, FFTW is Free Software in the technical sense
defined by the Free Software Foundation (see Categories of Free and
Non-Free Software), and is distributed under the terms of the GNU General
Public License. Previous versions of FFTW were distributed without fee
for noncommercial use, but were not technically ``free.''
Non-free licenses for FFTW are also available that permit different terms
of use than the GPL.
-------------------------------------------------------------------------------
Question 1.4. What is this about non-free licenses?
The non-free licenses are for companies that wish to use FFTW in their
products but are unwilling to release their software under the GPL (which
would require them to release source code and allow free redistribution).
Such users can purchase an unlimited-use license from MIT. Contact us for
more details.
We could instead have released FFTW under the LGPL, or even disallowed
non-Free usage. Suffice it to say, however, that MIT owns the copyright
to FFTW and they only let us GPL it because we convinced them that it
would neither affect their licensing revenue nor irritate existing
licensees.
-------------------------------------------------------------------------------
Question 1.5. In the West? I thought MIT was in the East?
Not to an Italian. You could say that we're a Spaghetti Western (with
apologies to Sergio Leone).
===============================================================================
Section 2. Installing FFTW
Q2.1 Which systems does FFTW run on?
Q2.2 Does FFTW run on Windows?
Q2.3 My compiler has trouble with FFTW.
Q2.4 FFTW does not compile on Solaris, complaining about const.
Q2.5 What's the difference between --enable-3dnow and --enable-k7?
Q2.6 What's the difference between the fma and the non-fma versions?
Q2.7 Which language is FFTW written in?
Q2.8 Can I call FFTW from Fortran?
Q2.9 Can I call FFTW from C++?
Q2.10 Why isn't FFTW written in Fortran/C++?
Q2.11 How do I compile FFTW to run in single precision?
Q2.12 --enable-k7 does not work on x86-64
-------------------------------------------------------------------------------
Question 2.1. Which systems does FFTW run on?
FFTW is written in ANSI C, and should work on any system with a decent C
compiler. (See also Q2.2 `Does FFTW run on Windows?', Q2.3 `My compiler
has trouble with FFTW.'.) FFTW can also take advantage of certain
hardware-specific features, such as cycle counters and SIMD instructions,
but this is optional.
-------------------------------------------------------------------------------
Question 2.2. Does FFTW run on Windows?
Yes, many people have reported successfully using FFTW on Windows with
various compilers. FFTW was not developed on Windows, but the source code
is essentially straight ANSI C. See also the FFTW Windows installation
notes, Q2.3 `My compiler has trouble with FFTW.', and Q3.18 `How do I call
FFTW from the Microsoft language du jour?'.
-------------------------------------------------------------------------------
Question 2.3. My compiler has trouble with FFTW.
Complain fiercely to the vendor of the compiler.
We have successfully used gcc 3.2.x on x86 and PPC, a recent Compaq C
compiler for Alpha, version 6 of IBM's xlc compiler for AIX, Intel's icc
versions 5-7, and Sun WorkShop cc version 6.
FFTW is likely to push compilers to their limits, however, and several
compiler bugs have been exposed by FFTW. A partial list follows.
gcc 2.95.x for Solaris/SPARC produces incorrect code for the test program
(workaround: recompile the libbench2 directory with -O2).
NetBSD/macppc 1.6 comes with a gcc version that also miscompiles the test
program. (Please report a workaround if you know one.)
gcc 3.2.3 for ARM reportedly crashes during compilation. This bug is
reportedly fixed in later versions of gcc.
Versions 8.0 and 8.1 of Intel's icc falsely claim to be gcc, so you should
specify CC="icc -no-gcc"; this is automatic in FFTW 3.1. icc-8.0.066
reportely produces incorrect code for FFTW 2.1.5, but is fixed in version
8.1. icc-7.1 compiler build 20030402Z appears to produce incorrect
dependencies, causing the compilation to fail. icc-7.1 build 20030307Z
appears to work fine. (Use icc -V to check which build you have.) As of
2003/04/18, build 20030402Z appears not to be available any longer on
Intel's website, whereas the older build 20030307Z is available.
ranlib of GNU binutils 2.9.1 on Irix has been observed to corrupt the FFTW
libraries, causing a link failure when FFTW is compiled. Since ranlib is
completely superfluous on Irix, we suggest deleting it from your system
and replacing it with a symbolic link to /bin/echo.
If support for SIMD instructions is enabled in FFTW, further compiler
problems may appear:
gcc 3.4.[0123] for x86 produces incorrect SSE2 code for FFTW when -O2 (the
best choice for FFTW) is used, causing FFTW to crash (make check crashes).
This bug is fixed in gcc 3.4.4. On x86_64 (amd64/em64t), gcc 3.4.4
reportedly still has a similar problem, but this is fixed as of gcc 3.4.6.
gcc-3.2 for x86 produces incorrect SIMD code if -O3 is used. The same
compiler produces incorrect SIMD code if no optimization is used, too.
When using gcc-3.2, it is a good idea not to change the default CFLAGS
selected by the configure script.
Some 3.0.x and 3.1.x versions of gcc on x86 may crash. gcc so-called 2.96
shipping with RedHat 7.3 crashes when compiling SIMD code. In both cases,
please upgrade to gcc-3.2 or later.
Intel's icc 6.0 misaligns SSE constants, but FFTW has a workaround. icc
8.x fails to compile FFTW 3.0.x because it falsely claims to be gcc; we
believe this to be a bug in icc, but FFTW 3.1 has a workaround.
Visual C++ 2003 reportedly produces incorrect code for SSE/SSE2 when
compiling FFTW. This bug was reportedly fixed in VC++ 2005;
alternatively, you could switch to the Intel compiler. VC++ 6.0 also
reportedly produces incorrect code for the file reodft11e-r2hc-odd.c
unless optimizations are disabled for that file.
gcc 2.95 on MacOS X miscompiles AltiVec code (fixed in later versions).
gcc 3.2.x miscompiles AltiVec permutations, but FFTW has a workaround.
gcc 4.0.1 on MacOS for Intel crashes when compiling FFTW; a workaround is
to compile one file without optimization: cd kernel; make CFLAGS=" "
trig.lo.
gcc 4.1.1 reportedly crashes when compiling FFTW for MIPS; the workaround
is to compile the file it crashes on (t2_64.c) with a lower optimization
level.
gcc versions 4.1.2 to 4.2.0 for x86 reportedly miscompile FFTW 3.1's test
program, causing make check to crash (gcc bug #26528). The bug was
reportedly fixed in gcc version 4.2.1 and later. A workaround is to
compile libbench2/verify-lib.c without optimization.
-------------------------------------------------------------------------------
Question 2.4. FFTW does not compile on Solaris, complaining about const.
We know that at least on Solaris 2.5.x with Sun's compilers 4.2 you might
get error messages from make such as
"./fftw.h", line 88: warning: const is a keyword in ANSI C
This is the case when the configure script reports that const does not
work:
checking for working const... (cached) no
You should be aware that Solaris comes with two compilers, namely,
/opt/SUNWspro/SC4.2/bin/cc and /usr/ucb/cc. The latter compiler is
non-ANSI. Indeed, it is a perverse shell script that calls the real
compiler in non-ANSI mode. In order to compile FFTW, change your path so
that the right cc is used.
To know whether your compiler is the right one, type cc -V. If the
compiler prints ``ucbcc'', as in
ucbcc: WorkShop Compilers 4.2 30 Oct 1996 C 4.2
then the compiler is wrong. The right message is something like
cc: WorkShop Compilers 4.2 30 Oct 1996 C 4.2
-------------------------------------------------------------------------------
Question 2.5. What's the difference between --enable-3dnow and --enable-k7?
--enable-k7 enables 3DNow! instructions on K7 processors (AMD Athlon and
its variants). K7 support is provided by assembly routines generated by a
special purpose compiler. As of fftw-3.2, --enable-k7 is no longer
supported.
--enable-3dnow enables generic 3DNow! support using gcc builtin functions.
This works on earlier AMD processors, but it is not as fast as our special
assembly routines. As of fftw-3.1, --enable-3dnow is no longer supported.
-------------------------------------------------------------------------------
Question 2.6. What's the difference between the fma and the non-fma versions?
The fma version tries to exploit the fused multiply-add instructions
implemented in many processors such as PowerPC, ia-64, and MIPS. The two
FFTW packages are otherwise identical. In FFTW 3.1, the fma and non-fma
versions were merged together into a single package, and the configure
script attempts to automatically guess which version to use.
The FFTW 3.1 configure script enables fma by default on PowerPC, Itanium,
and PA-RISC, and disables it otherwise. You can force one or the other by
using the --enable-fma or --disable-fma flag for configure.
Definitely use fma if you have a PowerPC-based system with gcc (or IBM
xlc). This includes all GNU/Linux systems for PowerPC and the older
PowerPC-based MacOS systems. Also use it on PA-RISC and Itanium with the
HP/UX compiler.
Definitely do not use the fma version if you have an ia-32 processor
(Intel, AMD, MacOS on Intel, etcetera).
For other architectures/compilers, the situation is not so clear. For
example, ia-64 has the fma instruction, but gcc-3.2 appears not to exploit
it correctly. Other compilers may do the right thing, but we have not
tried them. Please send us your feedback so that we can update this FAQ
entry.
-------------------------------------------------------------------------------
Question 2.7. Which language is FFTW written in?
FFTW is written in ANSI C. Most of the code, however, was automatically
generated by a program called genfft, written in the Objective Caml
dialect of ML. You do not need to know ML or to have an Objective Caml
compiler in order to use FFTW.
genfft is provided with the FFTW sources, which means that you can play
with the code generator if you want. In this case, you need a working
Objective Caml system. Objective Caml is available from the Caml web
page.
-------------------------------------------------------------------------------
Question 2.8. Can I call FFTW from Fortran?
Yes, FFTW (versions 1.3 and higher) contains a Fortran-callable interface,
documented in the FFTW manual.
By default, FFTW configures its Fortran interface to work with the first
compiler it finds, e.g. g77. To configure for a different, incompatible
Fortran compiler foobar, use ./configure F77=foobar when installing FFTW.
(In the case of g77, however, FFTW 3.x also includes an extra set of
Fortran-callable routines with one less underscore at the end of
identifiers, which should cover most other Fortran compilers on Linux at
least.)
-------------------------------------------------------------------------------
Question 2.9. Can I call FFTW from C++?
Most definitely. FFTW should compile and/or link under any C++ compiler.
Moreover, it is likely that the C++ <complex> template class is
bit-compatible with FFTW's complex-number format (see the FFTW manual for
more details).
-------------------------------------------------------------------------------
Question 2.10. Why isn't FFTW written in Fortran/C++?
Because we don't like those languages, and neither approaches the
portability of C.
-------------------------------------------------------------------------------
Question 2.11. How do I compile FFTW to run in single precision?
On a Unix system: configure --enable-float. On a non-Unix system: edit
config.h to #define the symbol FFTW_SINGLE (for FFTW 3.x). In both cases,
you must then recompile FFTW. In FFTW 3, all FFTW identifiers will then
begin with fftwf_ instead of fftw_.
-------------------------------------------------------------------------------
Question 2.12. --enable-k7 does not work on x86-64
Support for --enable-k7 was discontinued in fftw-3.2.
The fftw-3.1 release supports --enable-k7. This option only works on
32-bit x86 machines that implement 3DNow!, including the AMD Athlon and
the AMD Opteron in 32-bit mode. --enable-k7 does not work on AMD Opteron
in 64-bit mode. Use --enable-sse for x86-64 machines.
FFTW supports 3DNow! by means of assembly code generated by a
special-purpose compiler. It is hard to produce assembly code that works
in both 32-bit and 64-bit mode.
===============================================================================
Section 3. Using FFTW
Q3.1 Why not support the FFTW 2 interface in FFTW 3?
Q3.2 Why do FFTW 3 plans encapsulate the input/output arrays and not ju
Q3.3 FFTW seems really slow.
Q3.4 FFTW slows down after repeated calls.
Q3.5 An FFTW routine is crashing when I call it.
Q3.6 My Fortran program crashes when calling FFTW.
Q3.7 FFTW gives results different from my old FFT.
Q3.8 FFTW gives different results between runs
Q3.9 Can I save FFTW's plans?
Q3.10 Why does your inverse transform return a scaled result?
Q3.11 How can I make FFTW put the origin (zero frequency) at the center
Q3.12 How do I FFT an image/audio file in *foobar* format?
Q3.13 My program does not link (on Unix).
Q3.14 I included your header, but linking still fails.
Q3.15 My program crashes, complaining about stack space.
Q3.16 FFTW seems to have a memory leak.
Q3.17 The output of FFTW's transform is all zeros.
Q3.18 How do I call FFTW from the Microsoft language du jour?
Q3.19 Can I compute only a subset of the DFT outputs?
Q3.20 Can I use FFTW's routines for in-place and out-of-place matrix tra
-------------------------------------------------------------------------------
Question 3.1. Why not support the FFTW 2 interface in FFTW 3?
FFTW 3 has semantics incompatible with earlier versions: its plans can
only be used for a given stride, multiplicity, and other characteristics
of the input and output arrays; these stronger semantics are necessary for
performance reasons. Thus, it is impossible to efficiently emulate the
older interface (whose plans can be used for any transform of the same
size). We believe that it should be possible to upgrade most programs
without any difficulty, however.
-------------------------------------------------------------------------------
Question 3.2. Why do FFTW 3 plans encapsulate the input/output arrays and not just the algorithm?
There are several reasons:
* It was important for performance reasons that the plan be specific to
array characteristics like the stride (and alignment, for SIMD), and
requiring that the user maintain these invariants is error prone.
* In most high-performance applications, as far as we can tell, you are
usually transforming the same array over and over, so FFTW's semantics
should not be a burden.
* If you need to transform another array of the same size, creating a new
plan once the first exists is a cheap operation.
* If you need to transform many arrays of the same size at once, you
should really use the plan_many routines in FFTW's "advanced" interface.
* If the abovementioned array characteristics are the same, you are
willing to pay close attention to the documentation, and you really need
to, we provide a "new-array execution" interface to apply a plan to a
new array.
-------------------------------------------------------------------------------
Question 3.3. FFTW seems really slow.
You are probably recreating the plan before every transform, rather than
creating it once and reusing it for all transforms of the same size. FFTW
is designed to be used in the following way:
* First, you create a plan. This will take several seconds.
* Then, you reuse the plan many times to perform FFTs. These are fast.
If you don't need to compute many transforms and the time for the planner
is significant, you have two options. First, you can use the
FFTW_ESTIMATE option in the planner, which uses heuristics instead of
runtime measurements and produces a good plan in a short time. Second,
you can use the wisdom feature to precompute the plan; see Q3.9 `Can I
save FFTW's plans?'
-------------------------------------------------------------------------------
Question 3.4. FFTW slows down after repeated calls.
Probably, NaNs or similar are creeping into your data, and the slowdown is
due to the resulting floating-point exceptions. For example, be aware
that repeatedly FFTing the same array is a diverging process (because FFTW
computes the unnormalized transform).
-------------------------------------------------------------------------------
Question 3.5. An FFTW routine is crashing when I call it.
Did the FFTW test programs pass (make check, or cd tests; make bigcheck if
you want to be paranoid)? If so, you almost certainly have a bug in your
own code. For example, you could be passing invalid arguments (such as
wrongly-sized arrays) to FFTW, or you could simply have memory corruption
elsewhere in your program that causes random crashes later on. Please
don't complain to us unless you can come up with a minimal self-contained
program (preferably under 30 lines) that illustrates the problem.
-------------------------------------------------------------------------------
Question 3.6. My Fortran program crashes when calling FFTW.
As described in the manual, on 64-bit machines you must store the plans in
variables large enough to hold a pointer, for example integer*8. We
recommend using integer*8 on 32-bit machines as well, to simplify porting.
-------------------------------------------------------------------------------
Question 3.7. FFTW gives results different from my old FFT.
People follow many different conventions for the DFT, and you should be
sure to know the ones that we use (described in the FFTW manual). In
particular, you should be aware that the FFTW_FORWARD/FFTW_BACKWARD
directions correspond to signs of -1/+1 in the exponent of the DFT
definition. (*Numerical Recipes* uses the opposite convention.)
You should also know that we compute an unnormalized transform. In
contrast, Matlab is an example of program that computes a normalized
transform. See Q3.10 `Why does your inverse transform return a scaled
result?'.
Finally, note that floating-point arithmetic is not exact, so different
FFT algorithms will give slightly different results (on the order of the
numerical accuracy; typically a fractional difference of 1e-15 or so in
double precision).
-------------------------------------------------------------------------------
Question 3.8. FFTW gives different results between runs
If you use FFTW_MEASURE or FFTW_PATIENT mode, then the algorithm FFTW
employs is not deterministic: it depends on runtime performance
measurements. This will cause the results to vary slightly from run to
run. However, the differences should be slight, on the order of the
floating-point precision, and therefore should have no practical impact on
most applications.
If you use saved plans (wisdom) or FFTW_ESTIMATE mode, however, then the
algorithm is deterministic and the results should be identical between
runs.
-------------------------------------------------------------------------------
Question 3.9. Can I save FFTW's plans?
Yes. Starting with version 1.2, FFTW provides the wisdom mechanism for
saving plans; see the FFTW manual.
-------------------------------------------------------------------------------
Question 3.10. Why does your inverse transform return a scaled result?
Computing the forward transform followed by the backward transform (or
vice versa) yields the original array scaled by the size of the array.
(For multi-dimensional transforms, the size of the array is the product of
the dimensions.) We could, instead, have chosen a normalization that
would have returned the unscaled array. Or, to accomodate the many
conventions in this matter, the transform routines could have accepted a
"scale factor" parameter. We did not do this, however, for two reasons.
First, we didn't want to sacrifice performance in the common case where
the scale factor is 1. Second, in real applications the FFT is followed or
preceded by some computation on the data, into which the scale factor can
typically be absorbed at little or no cost.
-------------------------------------------------------------------------------
Question 3.11. How can I make FFTW put the origin (zero frequency) at the center of its output?
For human viewing of a spectrum, it is often convenient to put the origin
in frequency space at the center of the output array, rather than in the
zero-th element (the default in FFTW). If all of the dimensions of your
array are even, you can accomplish this by simply multiplying each element
of the input array by (-1)^(i + j + ...), where i, j, etcetera are the
indices of the element. (This trick is a general property of the DFT, and
is not specific to FFTW.)
-------------------------------------------------------------------------------
Question 3.12. How do I FFT an image/audio file in *foobar* format?
FFTW performs an FFT on an array of floating-point values. You can
certainly use it to compute the transform of an image or audio stream, but
you are responsible for figuring out your data format and converting it to
the form FFTW requires.
-------------------------------------------------------------------------------
Question 3.13. My program does not link (on Unix).
The libraries must be listed in the correct order (-lfftw3 -lm for FFTW
3.x) and *after* your program sources/objects. (The general rule is that
if *A* uses *B*, then *A* must be listed before *B* in the link command.).
-------------------------------------------------------------------------------
Question 3.14. I included your header, but linking still fails.
You're a C++ programmer, aren't you? You have to compile the FFTW library
and link it into your program, not just #include <fftw3.h>. (Yes, this is
really a FAQ.)
-------------------------------------------------------------------------------
Question 3.15. My program crashes, complaining about stack space.
You cannot declare large arrays with automatic storage (e.g. via
fftw_complex array[N]); you should use fftw_malloc (or equivalent) to
allocate the arrays you want to transform if they are larger than a few
hundred elements.
-------------------------------------------------------------------------------
Question 3.16. FFTW seems to have a memory leak.
After you create a plan, FFTW caches the information required to quickly
recreate the plan. (See Q3.9 `Can I save FFTW's plans?') It also
maintains a small amount of other persistent memory. You can deallocate
all of FFTW's internally allocated memory, if you wish, by calling
fftw_cleanup(), as documented in the manual.
-------------------------------------------------------------------------------
Question 3.17. The output of FFTW's transform is all zeros.
You should initialize your input array *after* creating the plan, unless
you use FFTW_ESTIMATE: planning with FFTW_MEASURE or FFTW_PATIENT
overwrites the input/output arrays, as described in the manual.
-------------------------------------------------------------------------------
Question 3.18. How do I call FFTW from the Microsoft language du jour?
Please *do not* ask us Windows-specific questions. We do not use Windows.
We know nothing about Visual Basic, Visual C++, or .NET. Please find the
appropriate Usenet discussion group and ask your question there. See also
Q2.2 `Does FFTW run on Windows?'.
-------------------------------------------------------------------------------
Question 3.19. Can I compute only a subset of the DFT outputs?
In general, no, an FFT intrinsically computes all outputs from all inputs.
In principle, there is something called a *pruned FFT* that can do what
you want, but to compute K outputs out of N the complexity is in general
O(N log K) instead of O(N log N), thus saving only a small additive factor
in the log. (The same argument holds if you instead have only K nonzero
inputs.)
There are some specific cases in which you can get the O(N log K)
performance benefits easily, however, by combining a few ordinary FFTs.
In particular, the case where you want the first K outputs, where K
divides N, can be handled by performing N/K transforms of size K and then
summing the outputs multiplied by appropriate phase factors. For more
details, see pruned FFTs with FFTW.
There are also some algorithms that compute pruned transforms
*approximately*, but they are beyond the scope of this FAQ.
-------------------------------------------------------------------------------
Question 3.20. Can I use FFTW's routines for in-place and out-of-place matrix transposition?
You can use the FFTW guru interface to create a rank-0 transform of vector
rank 2 where the vector strides are transposed. (A rank-0 transform is
equivalent to a 1D transform of size 1, which. just copies the input into
the output.) Specifying the same location for the input and output makes
the transpose in-place.
For double-valued data stored in row-major format, plan creation looks
like this:
fftw_plan plan_transpose(int rows, int cols, double *in, double *out)
{
const unsigned flags = FFTW_ESTIMATE; /* other flags are possible */
fftw_iodim howmany_dims[2];
howmany_dims[0].n = rows;
howmany_dims[0].is = cols;
howmany_dims[0].os = 1;
howmany_dims[1].n = cols;
howmany_dims[1].is = 1;
howmany_dims[1].os = rows;
return fftw_plan_guru_r2r(/*rank=*/ 0, /*dims=*/ NULL,
/*howmany_rank=*/ 2, howmany_dims,
in, out, /*kind=*/ NULL, flags);
}
(This entry was written by Rhys Ulerich.)
===============================================================================
Section 4. Internals of FFTW
Q4.1 How does FFTW work?
Q4.2 Why is FFTW so fast?
-------------------------------------------------------------------------------
Question 4.1. How does FFTW work?
The innovation (if it can be so called) in FFTW consists in having a
variety of composable *solvers*, representing different FFT algorithms and
implementation strategies, whose combination into a particular *plan* for
a given size can be determined at runtime according to the characteristics
of your machine/compiler. This peculiar software architecture allows FFTW
to adapt itself to almost any machine.
For more details (albeit somewhat outdated), see the paper "FFTW: An
Adaptive Software Architecture for the FFT", by M. Frigo and S. G.
Johnson, *Proc. ICASSP* 3, 1381 (1998), also available at the FFTW web
page.
-------------------------------------------------------------------------------
Question 4.2. Why is FFTW so fast?
This is a complex question, and there is no simple answer. In fact, the
authors do not fully know the answer, either. In addition to many small
performance hacks throughout FFTW, there are three general reasons for
FFTW's speed.
* FFTW uses a variety of FFT algorithms and implementation styles that
can be arbitrarily composed to adapt itself to a machine. See Q4.1 `How
does FFTW work?'.
* FFTW uses a code generator to produce highly-optimized routines for
computing small transforms.
* FFTW uses explicit divide-and-conquer to take advantage of the memory
hierarchy.
For more details (albeit somewhat outdated), see the paper "FFTW: An
Adaptive Software Architecture for the FFT", by M. Frigo and S. G.
Johnson, *Proc. ICASSP* 3, 1381 (1998), available along with other
references at the FFTW web page.
===============================================================================
Section 5. Known bugs
Q5.1 FFTW 1.1 crashes in rfftwnd on Linux.
Q5.2 The MPI transforms in FFTW 1.2 give incorrect results/leak memory.
Q5.3 The test programs in FFTW 1.2.1 fail when I change FFTW to use sin
Q5.4 The test program in FFTW 1.2.1 fails for n > 46340.
Q5.5 The threaded code fails on Linux Redhat 5.0
Q5.6 FFTW 2.0's rfftwnd fails for rank > 1 transforms with a final dime
Q5.7 FFTW 2.0's complex transforms give the wrong results with prime fa
Q5.8 FFTW 2.1.1's MPI test programs crash with MPICH.
Q5.9 FFTW 2.1.2's multi-threaded transforms don't work on AIX.
Q5.10 FFTW 2.1.2's complex transforms give incorrect results for large p
Q5.11 FFTW 2.1.3's multi-threaded transforms don't give any speedup on S
Q5.12 FFTW 2.1.3 crashes on AIX.
-------------------------------------------------------------------------------
Question 5.1. FFTW 1.1 crashes in rfftwnd on Linux.
This bug was fixed in FFTW 1.2. There was a bug in rfftwnd causing an
incorrect amount of memory to be allocated. The bug showed up in Linux
with libc-5.3.12 (and nowhere else that we know of).
-------------------------------------------------------------------------------
Question 5.2. The MPI transforms in FFTW 1.2 give incorrect results/leak memory.
These bugs were corrected in FFTW 1.2.1. The MPI transforms (really, just
the transpose routines) in FFTW 1.2 had bugs that could cause errors in
some situations.
-------------------------------------------------------------------------------
Question 5.3. The test programs in FFTW 1.2.1 fail when I change FFTW to use single precision.
This bug was fixed in FFTW 1.3. (Older versions of FFTW did work in
single precision, but the test programs didn't--the error tolerances in
the tests were set for double precision.)
-------------------------------------------------------------------------------
Question 5.4. The test program in FFTW 1.2.1 fails for n > 46340.
This bug was fixed in FFTW 1.3. FFTW 1.2.1 produced the right answer, but
the test program was wrong. For large n, n*n in the naive transform that
we used for comparison overflows 32 bit integer precision, breaking the
test.
-------------------------------------------------------------------------------
Question 5.5. The threaded code fails on Linux Redhat 5.0
We had problems with glibc-2.0.5. The code should work with glibc-2.0.7.
-------------------------------------------------------------------------------
Question 5.6. FFTW 2.0's rfftwnd fails for rank > 1 transforms with a final dimension >= 65536.
This bug was fixed in FFTW 2.0.1. (There was a 32-bit integer overflow
due to a poorly-parenthesized expression.)
-------------------------------------------------------------------------------
Question 5.7. FFTW 2.0's complex transforms give the wrong results with prime factors 17 to 97.
There was a bug in the complex transforms that could cause incorrect
results under (hopefully rare) circumstances for lengths with
intermediate-size prime factors (17-97). This bug was fixed in FFTW
2.1.1.
-------------------------------------------------------------------------------
Question 5.8. FFTW 2.1.1's MPI test programs crash with MPICH.
This bug was fixed in FFTW 2.1.2. The 2.1/2.1.1 MPI test programs crashed
when using the MPICH implementation of MPI with the ch_p4 device (TCP/IP);
the transforms themselves worked fine.
-------------------------------------------------------------------------------
Question 5.9. FFTW 2.1.2's multi-threaded transforms don't work on AIX.
This bug was fixed in FFTW 2.1.3. The multi-threaded transforms in
previous versions didn't work with AIX's pthreads implementation, which
idiosyncratically creates threads in detached (non-joinable) mode by
default.
-------------------------------------------------------------------------------
Question 5.10. FFTW 2.1.2's complex transforms give incorrect results for large prime sizes.
This bug was fixed in FFTW 2.1.3. FFTW's complex-transform algorithm for
prime sizes (in versions 2.0 to 2.1.2) had an integer overflow problem
that caused incorrect results for many primes greater than 32768 (on
32-bit machines). (Sizes without large prime factors are not affected.)
-------------------------------------------------------------------------------
Question 5.11. FFTW 2.1.3's multi-threaded transforms don't give any speedup on Solaris.
This bug was fixed in FFTW 2.1.4. (By default, Solaris creates threads
that do not parallelize over multiple processors, so one has to request
the proper behavior specifically.)
-------------------------------------------------------------------------------
Question 5.12. FFTW 2.1.3 crashes on AIX.
The FFTW 2.1.3 configure script picked incorrect compiler flags for the
xlc compiler on newer IBM processors. This is fixed in FFTW 2.1.4.

View File

@@ -0,0 +1,708 @@
\comment This is the source for the FFTW FAQ list, in
\comment the Bizarre Format With No Name. It is turned into Lout
\comment input, HTML, plain ASCII and an Info document by a Perl script.
\comment
\comment The format and scripts come from the Linux FAQ, by
\comment Ian Jackson.
\set brieftitle FFTW FAQ
\set author <A href="http://www.fftw.org">Matteo Frigo and Steven G. Johnson</A> / <A href="mailto:fftw@fftw.org">fftw@fftw.org</A>
\set authormail fftw@fftw.org
\set title FFTW Frequently Asked Questions with Answers
\set copyholder Matteo Frigo and Massachusetts Institute of Technology
\call-html startup html.refs2
\copyto ASCII
FFTW FREQUENTLY ASKED QUESTIONS WITH ANSWERS
`date '+%d %h %Y'`
Matteo Frigo
Steven G. Johnson
<fftw@fftw.org>
\endcopy
\copyto INFO
INFO-DIR-SECTION Development
START-INFO-DIR-ENTRY
* FFTW FAQ: (fftw-faq). FFTW Frequently Asked Questions with Answers.
END-INFO-DIR-ENTRY

File: $prefix.info, Node: Top, Next: Question 1.1, Up: (dir)
FFTW FREQUENTLY ASKED QUESTIONS WITH ANSWERS
`date '+%d %h %Y'`
Matteo Frigo
Steven G. Johnson
<fftw@fftw.org>
\endcopy
This is the list of Frequently Asked Questions about FFTW, a
collection of fast C routines for computing the Discrete Fourier
Transform in one or more dimensions.
\section Index
\index
\comment ######################################################################
\section Introduction and General Information
\question 26aug:whatisfftw What is FFTW?
FFTW is a free collection of fast C routines for computing the
Discrete Fourier Transform in one or more dimensions. It includes
complex, real, symmetric, and parallel transforms, and can handle
arbitrary array sizes efficiently. FFTW is typically faster than
other publically-available FFT implementations, and is even
competitive with vendor-tuned libraries. (See our web page for
extensive benchmarks.) To achieve this performance, FFTW uses novel
code-generation and runtime self-optimization techniques (along with
many other tricks).
\question 26aug:whereisfftw How do I obtain FFTW?
FFTW can be found at \docref{the FFTW web page\}. You can also
retrieve it from \ftpon ftp.fftw.org in \ftpin /pub/fftw.
\question 26aug:isfftwfree Is FFTW free software?
Starting with version 1.3, FFTW is Free Software in the technical
sense defined by the Free Software Foundation (see \docref{Categories
of Free and Non-Free Software\}), and is distributed under the terms
of the GNU General Public License. Previous versions of FFTW were
distributed without fee for noncommercial use, but were not
technically ``free.''
Non-free licenses for FFTW are also available that permit different
terms of use than the GPL.
\question 10apr:nonfree What is this about non-free licenses?
The non-free licenses are for companies that wish to use FFTW in their
products but are unwilling to release their software under the GPL
(which would require them to release source code and allow free
redistribution). Such users can purchase an unlimited-use license
from MIT. Contact us for more details.
We could instead have released FFTW under the LGPL, or even disallowed
non-Free usage. Suffice it to say, however, that MIT owns the
copyright to FFTW and they only let us GPL it because we convinced
them that it would neither affect their licensing revenue nor irritate
existing licensees.
\question 24oct:west In the West? I thought MIT was in the East?
Not to an Italian. You could say that we're a Spaghetti Western
(with apologies to Sergio Leone).
\comment ######################################################################
\section Installing FFTW
\question 26aug:systems Which systems does FFTW run on?
FFTW is written in ANSI C, and should work on any system with a decent
C compiler. (See also \qref runOnWindows, \qref compilerCrashes.)
FFTW can also take advantage of certain hardware-specific features,
such as cycle counters and SIMD instructions, but this is optional.
\question 26aug:runOnWindows Does FFTW run on Windows?
Yes, many people have reported successfully using FFTW on Windows with
various compilers. FFTW was not developed on Windows, but the source
code is essentially straight ANSI C. See also the \docref{FFTW
Windows installation notes\}, \qref compilerCrashes, and \qref
vbetalia.
\question 26aug:compilerCrashes My compiler has trouble with FFTW.
Complain fiercely to the vendor of the compiler.
We have successfully used \courier{gcc\} 3.2.x on x86 and PPC, a
recent Compaq C compiler for Alpha, version 6 of IBM's \courier{xlc\}
compiler for AIX, Intel's \courier{icc\} versions 5-7, and Sun
WorkShop \courier{cc\} version 6.
FFTW is likely to push compilers to their limits, however, and several
compiler bugs have been exposed by FFTW. A partial list follows.
\courier{gcc\} 2.95.x for Solaris/SPARC produces incorrect code for
the test program (workaround: recompile the \courier{libbench2\}
directory with \courier{-O2\}).
NetBSD/macppc 1.6 comes with a \courier{gcc\} version that also
miscompiles the test program. (Please report a workaround if you know
one.)
\courier{gcc\} 3.2.3 for ARM reportedly crashes during compilation.
This bug is reportedly fixed in later versions of \courier{gcc\}.
Versions 8.0 and 8.1 of Intel's \courier{icc\} falsely claim to be
\courier{gcc\}, so you should specify \courier{CC="icc -no-gcc"\};
this is automatic in FFTW 3.1. \courier{icc-8.0.066\} reportely
produces incorrect code for FFTW 2.1.5, but is fixed in version 8.1.
\courier{icc-7.1\} compiler build 20030402Z appears to produce
incorrect dependencies, causing the compilation to fail.
\courier{icc-7.1\} build 20030307Z appears to work fine. (Use
\courier{icc -V\} to check which build you have.) As of 2003/04/18,
build 20030402Z appears not to be available any longer on Intel's
website, whereas the older build 20030307Z is available.
\courier{ranlib\} of GNU \courier{binutils\} 2.9.1 on Irix has been
observed to corrupt the FFTW libraries, causing a link failure when
FFTW is compiled. Since \courier{ranlib\} is completely superfluous
on Irix, we suggest deleting it from your system and replacing it with
a symbolic link to \courier{/bin/echo\}.
If support for SIMD instructions is enabled in FFTW, further compiler
problems may appear:
\courier{gcc\} 3.4.[0123] for x86 produces incorrect SSE2 code for
FFTW when \courier{-O2\} (the best choice for FFTW) is used, causing
FFTW to crash (\courier{make check\} crashes). This bug is fixed in
\courier{gcc\} 3.4.4. On x86_64 (amd64/em64t), \courier{gcc\} 3.4.4
reportedly still has a similar problem, but this is fixed as of
\courier{gcc\} 3.4.6.
\courier{gcc-3.2\} for x86 produces incorrect SIMD code if
\courier{-O3\} is used. The same compiler produces incorrect SIMD
code if no optimization is used, too. When using \courier{gcc-3.2\},
it is a good idea not to change the default \courier{CFLAGS\} selected
by the \courier{configure\} script.
Some 3.0.x and 3.1.x versions of \courier{gcc\} on \courier{x86\} may
crash. \courier{gcc\} so-called 2.96 shipping with RedHat 7.3 crashes
when compiling SIMD code. In both cases, please upgrade to
\courier{gcc-3.2\} or later.
Intel's \courier{icc\} 6.0 misaligns SSE constants, but FFTW has a
workaround. \courier{icc\} 8.x fails to compile FFTW 3.0.x because it
falsely claims to be \courier{gcc\}; we believe this to be a bug in
\courier{icc\}, but FFTW 3.1 has a workaround.
Visual C++ 2003 reportedly produces incorrect code for SSE/SSE2 when
compiling FFTW. This bug was reportedly fixed in VC++ 2005;
alternatively, you could switch to the Intel compiler. VC++ 6.0 also
reportedly produces incorrect code for the file
\courier{reodft11e-r2hc-odd.c\} unless optimizations are disabled for
that file.
\courier{gcc\} 2.95 on MacOS X miscompiles AltiVec code (fixed in
later versions). \courier{gcc\} 3.2.x miscompiles AltiVec
permutations, but FFTW has a workaround. \courier{gcc\} 4.0.1 on
MacOS for Intel crashes when compiling FFTW; a workaround is to
compile one file without optimization: \courier{cd kernel; make
CFLAGS=" " trig.lo\}.
\courier{gcc\} 4.1.1 reportedly crashes when compiling FFTW for MIPS;
the workaround is to compile the file it crashes on
(\courier{t2_64.c\}) with a lower optimization level.
\courier{gcc\} versions 4.1.2 to 4.2.0 for x86 reportedly miscompile
FFTW 3.1's test program, causing \courier{make check\} to crash
(\courier{gcc\} bug #26528). The bug was reportedly fixed in
\courier{gcc\} version 4.2.1 and later. A workaround is to compile
\courier{libbench2/verify-lib.c\} without optimization.
\question 26aug:solarisSucks FFTW does not compile on Solaris, complaining about \courier{const\}.
We know that at least on Solaris 2.5.x with Sun's compilers 4.2 you
might get error messages from \courier{make\} such as
\courier{"./fftw.h", line 88: warning: const is a keyword in ANSI C\}
This is the case when the \courier{configure\} script reports that
\courier{const\} does not work:
\courier{checking for working const... (cached) no\}
You should be aware that Solaris comes with two compilers, namely,
\courier{/opt/SUNWspro/SC4.2/bin/cc\} and \courier{/usr/ucb/cc\}. The
latter compiler is non-ANSI. Indeed, it is a perverse shell script
that calls the real compiler in non-ANSI mode. In order
to compile FFTW, change your path so that the right \courier{cc\}
is used.
To know whether your compiler is the right one, type
\courier{cc -V\}. If the compiler prints ``\courier{ucbcc\}'',
as in
\courier{ucbcc: WorkShop Compilers 4.2 30 Oct 1996 C 4.2\}
then the compiler is wrong. The right message is something like
\courier{cc: WorkShop Compilers 4.2 30 Oct 1996 C 4.2\}
\question 19mar:3dnow What's the difference between \courier{--enable-3dnow\} and \courier{--enable-k7\}?
\courier{--enable-k7\} enables 3DNow! instructions on K7 processors
(AMD Athlon and its variants). K7 support is provided by assembly
routines generated by a special purpose compiler.
As of fftw-3.2, --enable-k7 is no longer supported.
\courier{--enable-3dnow\} enables generic 3DNow! support using
\courier{gcc\} builtin functions. This works on earlier AMD
processors, but it is not as fast as our special assembly routines.
As of fftw-3.1, --enable-3dnow is no longer supported.
\question 18apr:fma What's the difference between the fma and the non-fma versions?
The fma version tries to exploit the fused multiply-add instructions
implemented in many processors such as PowerPC, ia-64, and MIPS. The
two FFTW packages are otherwise identical. In FFTW 3.1, the fma and
non-fma versions were merged together into a single package, and the
\courier{configure\} script attempts to automatically guess which
version to use.
The FFTW 3.1 \courier{configure\} script enables fma by default on
PowerPC, Itanium, and PA-RISC, and disables it otherwise. You can
force one or the other by using the \courier{--enable-fma\} or
\courier{--disable-fma\} flag for \courier{configure\}.
Definitely use fma if you have a PowerPC-based system with
\courier{gcc\} (or IBM \courier{xlc\}). This includes all GNU/Linux
systems for PowerPC and the older PowerPC-based MacOS systems. Also
use it on PA-RISC and Itanium with the HP/UX compiler.
Definitely do not use the fma version if you have an ia-32 processor
(Intel, AMD, MacOS on Intel, etcetera).
For other architectures/compilers, the situation is not so clear. For
example, ia-64 has the fma instruction, but \courier{gcc-3.2\} appears
not to exploit it correctly. Other compilers may do the right thing,
but we have not tried them. Please send us your feedback so that we
can update this FAQ entry.
\question 26aug:languages Which language is FFTW written in?
FFTW is written in ANSI C. Most of the code, however, was
automatically generated by a program called \courier{genfft\}, written
in the Objective Caml dialect of ML. You do not need to know ML or to
have an Objective Caml compiler in order to use FFTW.
\courier{genfft\} is provided with the FFTW sources, which means that
you can play with the code generator if you want. In this case, you
need a working Objective Caml system. Objective Caml is available
from \docref{the Caml web page\}.
\question 26aug:fortran Can I call FFTW from Fortran?
Yes, FFTW (versions 1.3 and higher) contains a Fortran-callable
interface, documented in the FFTW manual.
By default, FFTW configures its Fortran interface to work with the
first compiler it finds, e.g. \courier{g77\}. To configure for a
different, incompatible Fortran compiler \courier{foobar\}, use
\courier{./configure F77=foobar\} when installing FFTW. (In the case
of \courier{g77\}, however, FFTW 3.x also includes an extra set of
Fortran-callable routines with one less underscore at the end of
identifiers, which should cover most other Fortran compilers on Linux
at least.)
\question 26aug:cplusplus Can I call FFTW from C++?
Most definitely. FFTW should compile and/or link under any C++
compiler. Moreover, it is likely that the C++ \courier{<complex>\}
template class is bit-compatible with FFTW's complex-number format
(see the FFTW manual for more details).
\question 26aug:whynotfortran Why isn't FFTW written in Fortran/C++?
Because we don't like those languages, and neither approaches the
portability of C.
\question 29mar:singleprec How do I compile FFTW to run in single precision?
On a Unix system: \courier{configure --enable-float\}. On a non-Unix
system: edit \courier{config.h\} to \courier{#define\} the symbol
\courier{FFTW_SINGLE\} (for FFTW 3.x). In both cases, you must then
recompile FFTW. In FFTW 3, all FFTW identifiers will then begin with
\courier{fftwf_\} instead of \courier{fftw_\}.
\question 28mar:64bitk7 --enable-k7 does not work on x86-64
Support for --enable-k7 was discontinued in fftw-3.2.
The fftw-3.1 release supports --enable-k7. This option only works on
32-bit x86 machines that implement 3DNow!, including the AMD Athlon
and the AMD Opteron in 32-bit mode. --enable-k7 does not work on AMD
Opteron in 64-bit mode. Use --enable-sse for x86-64 machines.
FFTW supports 3DNow! by means of assembly code generated by a
special-purpose compiler. It is hard to produce assembly code that
works in both 32-bit and 64-bit mode.
\comment ######################################################################
\section Using FFTW
\question 15mar:fftw2to3 Why not support the FFTW 2 interface in FFTW 3?
FFTW 3 has semantics incompatible with earlier versions: its plans can
only be used for a given stride, multiplicity, and other
characteristics of the input and output arrays; these stronger
semantics are necessary for performance reasons. Thus, it is
impossible to efficiently emulate the older interface (whose plans can
be used for any transform of the same size). We believe that it
should be possible to upgrade most programs without any difficulty,
however.
\question 20mar:planperarray Why do FFTW 3 plans encapsulate the input/output arrays and not just the algorithm?
There are several reasons:
\call startlist
\call item
It was important for performance reasons that the plan be specific to
array characteristics like the stride (and alignment, for SIMD), and
requiring that the user maintain these invariants is error prone.
\call item
In most high-performance applications, as far as we can tell, you are
usually transforming the same array over and over, so FFTW's semantics
should not be a burden.
\call item
If you need to transform another array of the same size, creating a
new plan once the first exists is a cheap operation.
\call item
If you need to transform many arrays of the same size at once, you
should really use the \courier{plan_many\} routines in FFTW's "advanced"
interface.
\call item
If the abovementioned array characteristics are the same, you are
willing to pay close attention to the documentation, and you really
need to, we provide a "new-array execution" interface to apply a plan
to a new array.
\call endlist
\question 25may:slow FFTW seems really slow.
You are probably recreating the plan before every transform, rather
than creating it once and reusing it for all transforms of the same
size. FFTW is designed to be used in the following way:
\call startlist
\call item
First, you create a plan. This will take several seconds.
\call item
Then, you reuse the plan many times to perform FFTs. These are fast.
\call endlist
If you don't need to compute many transforms and the time for the
planner is significant, you have two options. First, you can use the
\courier{FFTW_ESTIMATE\} option in the planner, which uses heuristics
instead of runtime measurements and produces a good plan in a short
time. Second, you can use the wisdom feature to precompute the plan;
see \qref savePlans
\question 22oct:slows FFTW slows down after repeated calls.
Probably, NaNs or similar are creeping into your data, and the
slowdown is due to the resulting floating-point exceptions. For
example, be aware that repeatedly FFTing the same array is a diverging
process (because FFTW computes the unnormalized transform).
\question 22oct:segfault An FFTW routine is crashing when I call it.
Did the FFTW test programs pass (\courier{make check\}, or \courier{cd
tests; make bigcheck\} if you want to be paranoid)? If so, you almost
certainly have a bug in your own code. For example, you could be
passing invalid arguments (such as wrongly-sized arrays) to FFTW, or
you could simply have memory corruption elsewhere in your program that
causes random crashes later on. Please don't complain to us unless
you can come up with a minimal self-contained program (preferably
under 30 lines) that illustrates the problem.
\question 22oct:fortran64 My Fortran program crashes when calling FFTW.
As described in the manual, on 64-bit machines you must store the
plans in variables large enough to hold a pointer, for example
\courier{integer*8\}. We recommend using \courier{integer*8\} on
32-bit machines as well, to simplify porting.
\question 24mar:conventions FFTW gives results different from my old FFT.
People follow many different conventions for the DFT, and you should
be sure to know the ones that we use (described in the FFTW manual).
In particular, you should be aware that the
\courier{FFTW_FORWARD\}/\courier{FFTW_BACKWARD\} directions correspond
to signs of -1/+1 in the exponent of the DFT definition.
(\italic{Numerical Recipes\} uses the opposite convention.)
You should also know that we compute an unnormalized transform. In
contrast, Matlab is an example of program that computes a normalized
transform. See \qref whyscaled.
Finally, note that floating-point arithmetic is not exact, so
different FFT algorithms will give slightly different results (on the
order of the numerical accuracy; typically a fractional difference of
1e-15 or so in double precision).
\question 31aug:nondeterministic FFTW gives different results between runs
If you use \courier{FFTW_MEASURE\} or \courier{FFTW_PATIENT\} mode,
then the algorithm FFTW employs is not deterministic: it depends on
runtime performance measurements. This will cause the results to vary
slightly from run to run. However, the differences should be slight,
on the order of the floating-point precision, and therefore should
have no practical impact on most applications.
If you use saved plans (wisdom) or \courier{FFTW_ESTIMATE\} mode,
however, then the algorithm is deterministic and the results should be
identical between runs.
\question 26aug:savePlans Can I save FFTW's plans?
Yes. Starting with version 1.2, FFTW provides the \courier{wisdom\}
mechanism for saving plans; see the FFTW manual.
\question 14sep:whyscaled Why does your inverse transform return a scaled result?
Computing the forward transform followed by the backward transform (or
vice versa) yields the original array scaled by the size of the array.
(For multi-dimensional transforms, the size of the array is the
product of the dimensions.) We could, instead, have chosen a
normalization that would have returned the unscaled array. Or, to
accomodate the many conventions in this matter, the transform routines
could have accepted a "scale factor" parameter. We did not do this,
however, for two reasons. First, we didn't want to sacrifice
performance in the common case where the scale factor is 1. Second, in
real applications the FFT is followed or preceded by some computation
on the data, into which the scale factor can typically be absorbed at
little or no cost.
\question 02dec:centerorigin How can I make FFTW put the origin (zero frequency) at the center of its output?
For human viewing of a spectrum, it is often convenient to put the
origin in frequency space at the center of the output array, rather
than in the zero-th element (the default in FFTW). If all of the
dimensions of your array are even, you can accomplish this by simply
multiplying each element of the input array by (-1)^(i + j + ...),
where i, j, etcetera are the indices of the element. (This trick is a
general property of the DFT, and is not specific to FFTW.)
\question 08may:imageaudio How do I FFT an image/audio file in \italic{foobar\} format?
FFTW performs an FFT on an array of floating-point values. You can
certainly use it to compute the transform of an image or audio stream,
but you are responsible for figuring out your data format and
converting it to the form FFTW requires.
\question 09apr:linkfails My program does not link (on Unix).
The libraries must be listed in the correct order (\courier{-lfftw3
-lm\} for FFTW 3.x) and \italic{after\} your program sources/objects.
(The general rule is that if \italic{A\} uses \italic{B\}, then
\italic{A\} must be listed before \italic{B\} in the link command.).
\question 15mar:linkheader I included your header, but linking still fails.
You're a C++ programmer, aren't you? You have to compile the FFTW
library and link it into your program, not just \courier{#include
<fftw3.h>\}. (Yes, this is really a FAQ.)
\question 22oct:nostack My program crashes, complaining about stack space.
You cannot declare large arrays with automatic storage (e.g. via
\courier{fftw_complex array[N]\}); you should use
\courier{fftw_malloc\} (or equivalent) to allocate the arrays you want
to transform if they are larger than a few hundred elements.
\question 13may:leaks FFTW seems to have a memory leak.
After you create a plan, FFTW caches the information required to
quickly recreate the plan. (See \qref savePlans) It also maintains a
small amount of other persistent memory. You can deallocate all of
FFTW's internally allocated memory, if you wish, by calling
\courier{fftw_cleanup()\}, as documented in the manual.
\question 16may:allzero The output of FFTW's transform is all zeros.
You should initialize your input array \italic{after\} creating the
plan, unless you use \courier{FFTW_ESTIMATE\}: planning with
\courier{FFTW_MEASURE\} or \courier{FFTW_PATIENT\} overwrites the
input/output arrays, as described in the manual.
\question 05sep:vbetalia How do I call FFTW from the Microsoft language du jour?
Please \italic{do not\} ask us Windows-specific questions. We do not
use Windows. We know nothing about Visual Basic, Visual C++, or .NET.
Please find the appropriate Usenet discussion group and ask your
question there. See also \qref runOnWindows.
\question 15oct:pruned Can I compute only a subset of the DFT outputs?
In general, no, an FFT intrinsically computes all outputs from all
inputs. In principle, there is something called a \italic{pruned
FFT\} that can do what you want, but to compute K outputs out of N the
complexity is in general O(N log K) instead of O(N log N), thus saving
only a small additive factor in the log. (The same argument holds if
you instead have only K nonzero inputs.)
There are some specific cases in which you can get the O(N log K)
performance benefits easily, however, by combining a few ordinary
FFTs. In particular, the case where you want the first K outputs,
where K divides N, can be handled by performing N/K transforms of size
K and then summing the outputs multiplied by appropriate phase
factors. For more details, see \docref{pruned FFTs with FFTW\}.
There are also some algorithms that compute pruned transforms
\italic{approximately\}, but they are beyond the scope of this FAQ.
\question 21jan:transpose Can I use FFTW's routines for in-place and out-of-place matrix transposition?
You can use the FFTW guru interface to create a rank-0 transform of
vector rank 2 where the vector strides are transposed. (A rank-0
transform is equivalent to a 1D transform of size 1, which. just
copies the input into the output.) Specifying the same location for
the input and output makes the transpose in-place.
For double-valued data stored in row-major format, plan creation looks like
this:
\verbatim
fftw_plan plan_transpose(int rows, int cols, double *in, double *out)
{
const unsigned flags = FFTW_ESTIMATE; /* other flags are possible */
fftw_iodim howmany_dims[2];
howmany_dims[0].n = rows;
howmany_dims[0].is = cols;
howmany_dims[0].os = 1;
howmany_dims[1].n = cols;
howmany_dims[1].is = 1;
howmany_dims[1].os = rows;
return fftw_plan_guru_r2r(/*rank=*/ 0, /*dims=*/ NULL,
/*howmany_rank=*/ 2, howmany_dims,
in, out, /*kind=*/ NULL, flags);
}
\endverbatim
(This entry was written by Rhys Ulerich.)
\comment ######################################################################
\section Internals of FFTW
\question 26aug:howworks How does FFTW work?
The innovation (if it can be so called) in FFTW consists in having a
variety of composable \italic{solvers\}, representing different FFT
algorithms and implementation strategies, whose combination into a
particular \italic{plan\} for a given size can be determined at
runtime according to the characteristics of your machine/compiler.
This peculiar software architecture allows FFTW to adapt itself to
almost any machine.
For more details (albeit somewhat outdated), see the paper "FFTW: An
Adaptive Software Architecture for the FFT", by M. Frigo and
S. G. Johnson, \italic{Proc. ICASSP\} 3, 1381 (1998), also
available at \docref{the FFTW web page\}.
\question 26aug:whyfast Why is FFTW so fast?
This is a complex question, and there is no simple answer. In fact,
the authors do not fully know the answer, either. In addition to many
small performance hacks throughout FFTW, there are three general
reasons for FFTW's speed.
\call startlist
\call item
FFTW uses a variety of FFT algorithms and implementation styles
that can be arbitrarily composed to adapt itself to
a machine. See \qref howworks.
\call item
FFTW uses a code generator to produce highly-optimized
routines for computing small transforms.
\call item
FFTW uses explicit divide-and-conquer to take advantage
of the memory hierarchy.
\call endlist
For more details (albeit somewhat outdated), see the paper "FFTW: An
Adaptive Software Architecture for the FFT", by M. Frigo and
S. G. Johnson, \italic{Proc. ICASSP\} 3, 1381 (1998),
available along with other references at \docref{the FFTW web page\}.
\comment ######################################################################
\section Known bugs
\question 27aug:rfftwndbug FFTW 1.1 crashes in rfftwnd on Linux.
This bug was fixed in FFTW 1.2. There was a bug in \courier{rfftwnd\}
causing an incorrect amount of memory to be allocated. The bug showed
up in Linux with libc-5.3.12 (and nowhere else that we know of).
\question 15oct:fftwmpibug The MPI transforms in FFTW 1.2 give incorrect results/leak memory.
These bugs were corrected in FFTW 1.2.1. The MPI transforms (really,
just the transpose routines) in FFTW 1.2 had bugs that could cause
errors in some situations.
\question 05nov:testsingbug The test programs in FFTW 1.2.1 fail when I change FFTW to use single precision.
This bug was fixed in FFTW 1.3. (Older versions of FFTW did
work in single precision, but the test programs didn't--the error
tolerances in the tests were set for double precision.)
\question 24mar:teststoobig The test program in FFTW 1.2.1 fails for n > 46340.
This bug was fixed in FFTW 1.3. FFTW 1.2.1 produced the right answer,
but the test program was wrong. For large n, n*n in the naive
transform that we used for comparison overflows 32 bit integer
precision, breaking the test.
\question 24aug:linuxthreads The threaded code fails on Linux Redhat 5.0
We had problems with glibc-2.0.5. The code should work with
glibc-2.0.7.
\question 26sep:bigrfftwnd FFTW 2.0's rfftwnd fails for rank > 1 transforms with a final dimension >= 65536.
This bug was fixed in FFTW 2.0.1. (There was a 32-bit integer overflow due
to a poorly-parenthesized expression.)
\question 26mar:primebug FFTW 2.0's complex transforms give the wrong results with prime factors 17 to 97.
There was a bug in the complex transforms that could cause incorrect
results under (hopefully rare) circumstances for lengths with
intermediate-size prime factors (17-97). This bug was fixed in FFTW
2.1.1.
\question 05apr:mpichbug FFTW 2.1.1's MPI test programs crash with MPICH.
This bug was fixed in FFTW 2.1.2. The 2.1/2.1.1 MPI test programs crashed
when using the MPICH implementation of MPI with the \courier{ch_p4\}
device (TCP/IP); the transforms themselves worked fine.
\question 25may:aixthreadbug FFTW 2.1.2's multi-threaded transforms don't work on AIX.
This bug was fixed in FFTW 2.1.3. The multi-threaded transforms in
previous versions didn't work with AIX's \courier{pthreads\}
implementation, which idiosyncratically creates threads in detached
(non-joinable) mode by default.
\question 27sep:bigprimebug FFTW 2.1.2's complex transforms give incorrect results for large prime sizes.
This bug was fixed in FFTW 2.1.3. FFTW's complex-transform algorithm
for prime sizes (in versions 2.0 to 2.1.2) had an integer overflow
problem that caused incorrect results for many primes greater than
32768 (on 32-bit machines). (Sizes without large prime factors are
not affected.)
\question 25may:solaristhreadbug FFTW 2.1.3's multi-threaded transforms don't give any speedup on Solaris.
This bug was fixed in FFTW 2.1.4. (By default, Solaris creates
threads that do not parallelize over multiple processors, so one has
to request the proper behavior specifically.)
\question 03may:aixflags FFTW 2.1.3 crashes on AIX.
The FFTW 2.1.3 \courier{configure\} script picked incorrect compiler
flags for the \courier{xlc\} compiler on newer IBM processors. This
is fixed in FFTW 2.1.4.
\comment Here it ends!

View File

@@ -0,0 +1,110 @@
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN">
<html>
<head><title>
FFTW Frequently Asked Questions with Answers
</title>
<link rev="made" href="mailto:fftw@fftw.org">
<link rel="Contents" href="index.html">
<link rel="Start" href="index.html">
<META name="description"
content="Frequently asked questions and answers (FAQ) for FFTW.">
<link rel="Bookmark" title="FFTW FAQ" href="index.html">
<LINK rel="Bookmark" title="FFTW Home Page"
href="http://www.fftw.org">
<LINK rel="Bookmark" title="FFTW Manual"
href="http://www.fftw.org/doc/">
</head><body text="#000000" bgcolor="#FFFFFF"><h1>
FFTW Frequently Asked Questions with Answers
</h1>
This is the list of Frequently Asked Questions about FFTW, a
collection of fast C routines for computing the Discrete Fourier
Transform in one or more dimensions.
<h1>
Index
</h1>
<ul>
<li><b><font size="+2"><a href="section1.html" rel=subdocument>Section 1. Introduction and General Information</a></font></b>
<li><a href="section1.html#whatisfftw" rel=subdocument>Q1.1. What is FFTW?</a>
<li><a href="section1.html#whereisfftw" rel=subdocument>Q1.2. How do I obtain FFTW?</a>
<li><a href="section1.html#isfftwfree" rel=subdocument>Q1.3. Is FFTW free software?</a>
<li><a href="section1.html#nonfree" rel=subdocument>Q1.4. What is this about non-free licenses?</a>
<li><a href="section1.html#west" rel=subdocument>Q1.5. In the West? I thought MIT was in the East?</a>
<br><br><li><b><font size="+2"><a href="section2.html" rel=subdocument>Section 2. Installing FFTW</a></font></b>
<li><a href="section2.html#systems" rel=subdocument>Q2.1. Which systems does FFTW run on?</a>
<li><a href="section2.html#runOnWindows" rel=subdocument>Q2.2. Does FFTW run on Windows?</a>
<li><a href="section2.html#compilerCrashes" rel=subdocument>Q2.3. My compiler has trouble with FFTW.</a>
<li><a href="section2.html#solarisSucks" rel=subdocument>Q2.4. FFTW does not compile on Solaris, complaining about
<code>const</code>.</a>
<li><a href="section2.html#3dnow" rel=subdocument>Q2.5. What's the difference between <code>--enable-3dnow</code> and <code>--enable-k7</code>?</a>
<li><a href="section2.html#fma" rel=subdocument>Q2.6. What's the difference between the fma and the non-fma
versions?</a>
<li><a href="section2.html#languages" rel=subdocument>Q2.7. Which language is FFTW written in?</a>
<li><a href="section2.html#fortran" rel=subdocument>Q2.8. Can I call FFTW from Fortran?</a>
<li><a href="section2.html#cplusplus" rel=subdocument>Q2.9. Can I call FFTW from C++?</a>
<li><a href="section2.html#whynotfortran" rel=subdocument>Q2.10. Why isn't FFTW written in Fortran/C++?</a>
<li><a href="section2.html#singleprec" rel=subdocument>Q2.11. How do I compile FFTW to run in single precision?</a>
<li><a href="section2.html#64bitk7" rel=subdocument>Q2.12. --enable-k7 does not work on x86-64</a>
<br><br><li><b><font size="+2"><a href="section3.html" rel=subdocument>Section 3. Using FFTW</a></font></b>
<li><a href="section3.html#fftw2to3" rel=subdocument>Q3.1. Why not support the FFTW 2 interface in FFTW
3?</a>
<li><a href="section3.html#planperarray" rel=subdocument>Q3.2. Why do FFTW 3 plans encapsulate the input/output arrays and not just
the algorithm?</a>
<li><a href="section3.html#slow" rel=subdocument>Q3.3. FFTW seems really slow.</a>
<li><a href="section3.html#slows" rel=subdocument>Q3.4. FFTW slows down after repeated calls.</a>
<li><a href="section3.html#segfault" rel=subdocument>Q3.5. An FFTW routine is crashing when I call it.</a>
<li><a href="section3.html#fortran64" rel=subdocument>Q3.6. My Fortran program crashes when calling FFTW.</a>
<li><a href="section3.html#conventions" rel=subdocument>Q3.7. FFTW gives results different from my old
FFT.</a>
<li><a href="section3.html#nondeterministic" rel=subdocument>Q3.8. FFTW gives different results between runs</a>
<li><a href="section3.html#savePlans" rel=subdocument>Q3.9. Can I save FFTW's plans?</a>
<li><a href="section3.html#whyscaled" rel=subdocument>Q3.10. Why does your inverse transform return a scaled
result?</a>
<li><a href="section3.html#centerorigin" rel=subdocument>Q3.11. How can I make FFTW put the origin (zero frequency) at the center of
its output?</a>
<li><a href="section3.html#imageaudio" rel=subdocument>Q3.12. How do I FFT an image/audio file in <i>foobar</i> format?</a>
<li><a href="section3.html#linkfails" rel=subdocument>Q3.13. My program does not link (on Unix).</a>
<li><a href="section3.html#linkheader" rel=subdocument>Q3.14. I included your header, but linking still
fails.</a>
<li><a href="section3.html#nostack" rel=subdocument>Q3.15. My program crashes, complaining about stack
space.</a>
<li><a href="section3.html#leaks" rel=subdocument>Q3.16. FFTW seems to have a memory leak.</a>
<li><a href="section3.html#allzero" rel=subdocument>Q3.17. The output of FFTW's transform is all zeros.</a>
<li><a href="section3.html#vbetalia" rel=subdocument>Q3.18. How do I call FFTW from the Microsoft language du
jour?</a>
<li><a href="section3.html#pruned" rel=subdocument>Q3.19. Can I compute only a subset of the DFT outputs?</a>
<li><a href="section3.html#transpose" rel=subdocument>Q3.20. Can I use FFTW's routines for in-place and out-of-place matrix
transposition?</a>
<br><br><li><b><font size="+2"><a href="section4.html" rel=subdocument>Section 4. Internals of FFTW</a></font></b>
<li><a href="section4.html#howworks" rel=subdocument>Q4.1. How does FFTW work?</a>
<li><a href="section4.html#whyfast" rel=subdocument>Q4.2. Why is FFTW so fast?</a>
<br><br><li><b><font size="+2"><a href="section5.html" rel=subdocument>Section 5. Known bugs</a></font></b>
<li><a href="section5.html#rfftwndbug" rel=subdocument>Q5.1. FFTW 1.1 crashes in rfftwnd on Linux.</a>
<li><a href="section5.html#fftwmpibug" rel=subdocument>Q5.2. The MPI transforms in FFTW 1.2 give incorrect results/leak
memory.</a>
<li><a href="section5.html#testsingbug" rel=subdocument>Q5.3. The test programs in FFTW 1.2.1 fail when I change FFTW to use single
precision.</a>
<li><a href="section5.html#teststoobig" rel=subdocument>Q5.4. The test program in FFTW 1.2.1 fails for n &gt;
46340.</a>
<li><a href="section5.html#linuxthreads" rel=subdocument>Q5.5. The threaded code fails on Linux Redhat 5.0</a>
<li><a href="section5.html#bigrfftwnd" rel=subdocument>Q5.6. FFTW 2.0's rfftwnd fails for rank &gt; 1 transforms with a final
dimension &gt;= 65536.</a>
<li><a href="section5.html#primebug" rel=subdocument>Q5.7. FFTW 2.0's complex transforms give the wrong results with prime
factors 17 to 97.</a>
<li><a href="section5.html#mpichbug" rel=subdocument>Q5.8. FFTW 2.1.1's MPI test programs crash with
MPICH.</a>
<li><a href="section5.html#aixthreadbug" rel=subdocument>Q5.9. FFTW 2.1.2's multi-threaded transforms don't work on
AIX.</a>
<li><a href="section5.html#bigprimebug" rel=subdocument>Q5.10. FFTW 2.1.2's complex transforms give incorrect results for large prime
sizes.</a>
<li><a href="section5.html#solaristhreadbug" rel=subdocument>Q5.11. FFTW 2.1.3's multi-threaded transforms don't give any speedup on
Solaris.</a>
<li><a href="section5.html#aixflags" rel=subdocument>Q5.12. FFTW 2.1.3 crashes on AIX.</a>
</ul><hr>
<address>
<A href="http://www.fftw.org">Matteo Frigo and Steven G. Johnson</A> / <A href="mailto:fftw@fftw.org">fftw@fftw.org</A>
- 14 September 2021
</address><br>
Extracted from FFTW Frequently Asked Questions with Answers,
Copyright &copy; 2021 Matteo Frigo and Massachusetts Institute of Technology.
</body></html>

View File

@@ -0,0 +1,85 @@
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN">
<html>
<head><title>
FFTW FAQ - Section 1
</title>
<link rev="made" href="mailto:fftw@fftw.org">
<link rel="Contents" href="index.html">
<link rel="Start" href="index.html">
<link rel="Next" href="section2.html"><link rel="Bookmark" title="FFTW FAQ" href="index.html">
</head><body text="#000000" bgcolor="#FFFFFF"><h1>
FFTW FAQ - Section 1 <br>
Introduction and General Information
</h1>
<ul>
<li><a href="#whatisfftw" rel=subdocument>Q1.1. What is FFTW?</a>
<li><a href="#whereisfftw" rel=subdocument>Q1.2. How do I obtain FFTW?</a>
<li><a href="#isfftwfree" rel=subdocument>Q1.3. Is FFTW free software?</a>
<li><a href="#nonfree" rel=subdocument>Q1.4. What is this about non-free licenses?</a>
<li><a href="#west" rel=subdocument>Q1.5. In the West? I thought MIT was in the East?</a>
</ul><hr>
<h2><A name="whatisfftw">
Question 1.1. What is FFTW?
</A></h2>
FFTW is a free collection of fast C routines for computing the
Discrete Fourier Transform in one or more dimensions. It includes
complex, real, symmetric, and parallel transforms, and can handle
arbitrary array sizes efficiently. FFTW is typically faster than
other publically-available FFT implementations, and is even
competitive with vendor-tuned libraries. (See our web page for
extensive benchmarks.) To achieve this performance, FFTW uses novel
code-generation and runtime self-optimization techniques (along with
many other tricks).
<h2><A name="whereisfftw">
Question 1.2. How do I obtain FFTW?
</A></h2>
FFTW can be found at <A href="http://www.fftw.org">the FFTW web page</A>. You can also retrieve it from <code>ftp.fftw.org</code> in <A href="ftp://ftp.fftw.org/pub/fftw"><code>/pub/fftw</code></A>.
<h2><A name="isfftwfree">
Question 1.3. Is FFTW free software?
</A></h2>
Starting with version 1.3, FFTW is Free Software in the technical
sense defined by the Free Software Foundation (see
<A href="http://www.gnu.org/philosophy/categories.html">Categories of Free and Non-Free Software</A>), and is distributed under the terms of the GNU General Public License. Previous versions of FFTW were
distributed without fee for noncommercial use, but were not
technically ``free.''
<p>
Non-free licenses for FFTW are also available that permit different
terms of use than the GPL.
<h2><A name="nonfree">
Question 1.4. What is this about non-free
licenses?
</A></h2>
The non-free licenses are for companies that wish to use FFTW in their
products but are unwilling to release their software under the GPL
(which would require them to release source code and allow free
redistribution). Such users can purchase an unlimited-use license
from MIT. Contact us for more details.
<p>
We could instead have released FFTW under the LGPL, or even disallowed
non-Free usage. Suffice it to say, however, that MIT owns the
copyright to FFTW and they only let us GPL it because we convinced
them that it would neither affect their licensing revenue nor irritate
existing licensees.
<h2><A name="west">
Question 1.5. In the West? I thought MIT was in the
East?
</A></h2>
Not to an Italian. You could say that we're a Spaghetti Western
(with apologies to Sergio Leone). <hr>
Next: <a href="section2.html" rel=precedes>Installing FFTW</a>.<br>
<a href="index.html" rev=subdocument>Return to contents</a>.<p>
<address>
<A href="http://www.fftw.org">Matteo Frigo and Steven G. Johnson</A> / <A href="mailto:fftw@fftw.org">fftw@fftw.org</A>
- 14 September 2021
</address><br>
Extracted from FFTW Frequently Asked Questions with Answers,
Copyright &copy; 2021 Matteo Frigo and Massachusetts Institute of Technology.
</body></html>

View File

@@ -0,0 +1,285 @@
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN">
<html>
<head><title>
FFTW FAQ - Section 2
</title>
<link rev="made" href="mailto:fftw@fftw.org">
<link rel="Contents" href="index.html">
<link rel="Start" href="index.html">
<link rel="Next" href="section3.html"><link rel="Previous" href="section1.html"><link rel="Bookmark" title="FFTW FAQ" href="index.html">
</head><body text="#000000" bgcolor="#FFFFFF"><h1>
FFTW FAQ - Section 2 <br>
Installing FFTW
</h1>
<ul>
<li><a href="#systems" rel=subdocument>Q2.1. Which systems does FFTW run on?</a>
<li><a href="#runOnWindows" rel=subdocument>Q2.2. Does FFTW run on Windows?</a>
<li><a href="#compilerCrashes" rel=subdocument>Q2.3. My compiler has trouble with FFTW.</a>
<li><a href="#solarisSucks" rel=subdocument>Q2.4. FFTW does not compile on Solaris, complaining about
<code>const</code>.</a>
<li><a href="#3dnow" rel=subdocument>Q2.5. What's the difference between <code>--enable-3dnow</code> and <code>--enable-k7</code>?</a>
<li><a href="#fma" rel=subdocument>Q2.6. What's the difference between the fma and the non-fma
versions?</a>
<li><a href="#languages" rel=subdocument>Q2.7. Which language is FFTW written in?</a>
<li><a href="#fortran" rel=subdocument>Q2.8. Can I call FFTW from Fortran?</a>
<li><a href="#cplusplus" rel=subdocument>Q2.9. Can I call FFTW from C++?</a>
<li><a href="#whynotfortran" rel=subdocument>Q2.10. Why isn't FFTW written in Fortran/C++?</a>
<li><a href="#singleprec" rel=subdocument>Q2.11. How do I compile FFTW to run in single precision?</a>
<li><a href="#64bitk7" rel=subdocument>Q2.12. --enable-k7 does not work on x86-64</a>
</ul><hr>
<h2><A name="systems">
Question 2.1. Which systems does FFTW run
on?
</A></h2>
FFTW is written in ANSI C, and should work on any system with a decent
C compiler. (See also <A href="#runOnWindows">Q2.2 `Does FFTW run on Windows?'</A>, <A href="#compilerCrashes">Q2.3 `My compiler has trouble with FFTW.'</A>.) FFTW can also take advantage of certain hardware-specific features,
such as cycle counters and SIMD instructions, but this is optional.
<h2><A name="runOnWindows">
Question 2.2. Does FFTW run on Windows?
</A></h2>
Yes, many people have reported successfully using FFTW on Windows with
various compilers. FFTW was not developed on Windows, but the source
code is essentially straight ANSI C. See also the
<A href="http://www.fftw.org/install/windows.html">FFTW Windows installation notes</A>, <A href="#compilerCrashes">Q2.3 `My compiler has trouble with FFTW.'</A>, and <A href="section3.html#vbetalia">Q3.18 `How do I call FFTW from the Microsoft language du
jour?'</A>.
<h2><A name="compilerCrashes">
Question 2.3. My compiler has trouble with
FFTW.
</A></h2>
Complain fiercely to the vendor of the compiler.
<p>
We have successfully used <code>gcc</code> 3.2.x on x86 and PPC, a recent Compaq C compiler for Alpha, version 6 of IBM's
<code>xlc</code> compiler for AIX, Intel's <code>icc</code> versions 5-7, and Sun WorkShop <code>cc</code> version 6.
<p>
FFTW is likely to push compilers to their limits, however, and several
compiler bugs have been exposed by FFTW. A partial list follows.
<p>
<code>gcc</code> 2.95.x for Solaris/SPARC produces incorrect code for
the test program (workaround: recompile the
<code>libbench2</code> directory with <code>-O2</code>).
<p>
NetBSD/macppc 1.6 comes with a <code>gcc</code> version that also miscompiles the test program. (Please report a workaround if you know
one.)
<p>
<code>gcc</code> 3.2.3 for ARM reportedly crashes during compilation.
This bug is reportedly fixed in later versions of
<code>gcc</code>.
<p>
Versions 8.0 and 8.1 of Intel's <code>icc</code> falsely claim to be <code>gcc</code>, so you should specify <code>CC=&quot;icc -no-gcc&quot;</code>; this is automatic in FFTW 3.1. <code>icc-8.0.066</code> reportely produces incorrect code for FFTW 2.1.5, but is fixed in version 8.1.
<code>icc-7.1</code> compiler build 20030402Z appears to produce
incorrect dependencies, causing the compilation to fail.
<code>icc-7.1</code> build 20030307Z appears to work fine. (Use
<code>icc -V</code> to check which build you have.) As of 2003/04/18,
build 20030402Z appears not to be available any longer on Intel's
website, whereas the older build 20030307Z is available.
<p>
<code>ranlib</code> of GNU <code>binutils</code> 2.9.1 on Irix has been observed to corrupt the FFTW libraries, causing a link failure when
FFTW is compiled. Since <code>ranlib</code> is completely superfluous on Irix, we suggest deleting it from your system and replacing it with
a symbolic link to <code>/bin/echo</code>.
<p>
If support for SIMD instructions is enabled in FFTW, further compiler
problems may appear:
<p>
<code>gcc</code> 3.4.[0123] for x86 produces incorrect SSE2 code for
FFTW when <code>-O2</code> (the best choice for FFTW) is used, causing
FFTW to crash (<code>make check</code> crashes). This bug is fixed in <code>gcc</code> 3.4.4. On x86_64 (amd64/em64t), <code>gcc</code> 3.4.4 reportedly still has a similar problem, but this is fixed as of
<code>gcc</code> 3.4.6.
<p>
<code>gcc-3.2</code> for x86 produces incorrect SIMD code if
<code>-O3</code> is used. The same compiler produces incorrect SIMD
code if no optimization is used, too. When using
<code>gcc-3.2</code>, it is a good idea not to change the default
<code>CFLAGS</code> selected by the <code>configure</code> script.
<p>
Some 3.0.x and 3.1.x versions of <code>gcc</code> on <code>x86</code> may crash. <code>gcc</code> so-called 2.96 shipping with RedHat 7.3 crashes
when compiling SIMD code. In both cases, please upgrade to
<code>gcc-3.2</code> or later.
<p>
Intel's <code>icc</code> 6.0 misaligns SSE constants, but FFTW has a
workaround. <code>icc</code> 8.x fails to compile FFTW 3.0.x because it
falsely claims to be <code>gcc</code>; we believe this to be a bug in <code>icc</code>, but FFTW 3.1 has a workaround.
<p>
Visual C++ 2003 reportedly produces incorrect code for SSE/SSE2 when
compiling FFTW. This bug was reportedly fixed in VC++ 2005;
alternatively, you could switch to the Intel compiler. VC++ 6.0 also
reportedly produces incorrect code for the file
<code>reodft11e-r2hc-odd.c</code> unless optimizations are disabled for that file.
<p>
<code>gcc</code> 2.95 on MacOS X miscompiles AltiVec code (fixed in
later versions). <code>gcc</code> 3.2.x miscompiles AltiVec permutations, but FFTW has a workaround.
<code>gcc</code> 4.0.1 on MacOS for Intel crashes when compiling FFTW; a workaround is to
compile one file without optimization: <code>cd kernel; make CFLAGS=&quot; &quot; trig.lo</code>.
<p>
<code>gcc</code> 4.1.1 reportedly crashes when compiling FFTW for MIPS;
the workaround is to compile the file it crashes on
(<code>t2_64.c</code>) with a lower optimization level.
<p>
<code>gcc</code> versions 4.1.2 to 4.2.0 for x86 reportedly miscompile
FFTW 3.1's test program, causing <code>make check</code> to crash (<code>gcc</code> bug #26528). The bug was reportedly fixed in
<code>gcc</code> version 4.2.1 and later. A workaround is to compile
<code>libbench2/verify-lib.c</code> without optimization.
<h2><A name="solarisSucks">
Question 2.4. FFTW does not compile on Solaris, complaining about
<code>const</code>.
</A></h2>
We know that at least on Solaris 2.5.x with Sun's compilers 4.2 you
might get error messages from <code>make</code> such as
<p>
<code>&quot;./fftw.h&quot;, line 88: warning: const is a keyword in ANSI
C</code>
<p>
This is the case when the <code>configure</code> script reports that <code>const</code> does not work:
<p>
<code>checking for working const... (cached) no</code>
<p>
You should be aware that Solaris comes with two compilers, namely,
<code>/opt/SUNWspro/SC4.2/bin/cc</code> and <code>/usr/ucb/cc</code>. The latter compiler is non-ANSI. Indeed, it is a perverse shell script
that calls the real compiler in non-ANSI mode. In order
to compile FFTW, change your path so that the right
<code>cc</code> is used.
<p>
To know whether your compiler is the right one, type
<code>cc -V</code>. If the compiler prints ``<code>ucbcc</code>'', as in
<p>
<code>ucbcc: WorkShop Compilers 4.2 30 Oct 1996 C
4.2</code>
<p>
then the compiler is wrong. The right message is something like
<p>
<code>cc: WorkShop Compilers 4.2 30 Oct 1996 C
4.2</code>
<h2><A name="3dnow">
Question 2.5. What's the difference between
<code>--enable-3dnow</code> and <code>--enable-k7</code>?
</A></h2>
<code>--enable-k7</code> enables 3DNow! instructions on K7 processors
(AMD Athlon and its variants). K7 support is provided by assembly
routines generated by a special purpose compiler.
As of fftw-3.2, --enable-k7 is no longer supported.
<p>
<code>--enable-3dnow</code> enables generic 3DNow! support using <code>gcc</code> builtin functions. This works on earlier AMD
processors, but it is not as fast as our special assembly routines.
As of fftw-3.1, --enable-3dnow is no longer supported.
<h2><A name="fma">
Question 2.6. What's the difference between the fma and the non-fma
versions?
</A></h2>
The fma version tries to exploit the fused multiply-add instructions
implemented in many processors such as PowerPC, ia-64, and MIPS. The
two FFTW packages are otherwise identical. In FFTW 3.1, the fma and
non-fma versions were merged together into a single package, and the
<code>configure</code> script attempts to automatically guess which
version to use.
<p>
The FFTW 3.1 <code>configure</code> script enables fma by default on PowerPC, Itanium, and PA-RISC, and disables it otherwise. You can
force one or the other by using the <code>--enable-fma</code> or <code>--disable-fma</code> flag for <code>configure</code>.
<p>
Definitely use fma if you have a PowerPC-based system with
<code>gcc</code> (or IBM <code>xlc</code>). This includes all GNU/Linux systems for PowerPC and the older PowerPC-based MacOS systems. Also
use it on PA-RISC and Itanium with the HP/UX compiler.
<p>
Definitely do not use the fma version if you have an ia-32 processor
(Intel, AMD, MacOS on Intel, etcetera).
<p>
For other architectures/compilers, the situation is not so clear. For
example, ia-64 has the fma instruction, but
<code>gcc-3.2</code> appears not to exploit it correctly. Other compilers may do the right thing,
but we have not tried them. Please send us your feedback so that we
can update this FAQ entry.
<h2><A name="languages">
Question 2.7. Which language is FFTW written
in?
</A></h2>
FFTW is written in ANSI C. Most of the code, however, was
automatically generated by a program called
<code>genfft</code>, written in the Objective Caml dialect of ML. You do not need to know ML or to
have an Objective Caml compiler in order to use FFTW.
<p>
<code>genfft</code> is provided with the FFTW sources, which means that
you can play with the code generator if you want. In this case, you
need a working Objective Caml system. Objective Caml is available
from <A href="http://caml.inria.fr">the Caml web page</A>.
<h2><A name="fortran">
Question 2.8. Can I call FFTW from Fortran?
</A></h2>
Yes, FFTW (versions 1.3 and higher) contains a Fortran-callable
interface, documented in the FFTW manual.
<p>
By default, FFTW configures its Fortran interface to work with the
first compiler it finds, e.g. <code>g77</code>. To configure for a different, incompatible Fortran compiler
<code>foobar</code>, use <code>./configure F77=foobar</code> when installing FFTW. (In the case of <code>g77</code>, however, FFTW 3.x also includes an extra set of
Fortran-callable routines with one less underscore at the end of
identifiers, which should cover most other Fortran compilers on Linux
at least.)
<h2><A name="cplusplus">
Question 2.9. Can I call FFTW from C++?
</A></h2>
Most definitely. FFTW should compile and/or link under any C++
compiler. Moreover, it is likely that the C++
<code>&lt;complex&gt;</code> template class is bit-compatible with FFTW's complex-number format
(see the FFTW manual for more details).
<h2><A name="whynotfortran">
Question 2.10. Why isn't FFTW written in
Fortran/C++?
</A></h2>
Because we don't like those languages, and neither approaches the
portability of C.
<h2><A name="singleprec">
Question 2.11. How do I compile FFTW to run in single
precision?
</A></h2>
On a Unix system: <code>configure --enable-float</code>. On a non-Unix system: edit <code>config.h</code> to <code>#define</code> the symbol <code>FFTW_SINGLE</code> (for FFTW 3.x). In both cases, you must then
recompile FFTW. In FFTW 3, all FFTW identifiers will then begin with
<code>fftwf_</code> instead of <code>fftw_</code>.
<h2><A name="64bitk7">
Question 2.12. --enable-k7 does not work on
x86-64
</A></h2>
Support for --enable-k7 was discontinued in fftw-3.2.
<p>
The fftw-3.1 release supports --enable-k7. This option only works on
32-bit x86 machines that implement 3DNow!, including the AMD Athlon
and the AMD Opteron in 32-bit mode. --enable-k7 does not work on AMD
Opteron in 64-bit mode. Use --enable-sse for x86-64 machines.
<p>
FFTW supports 3DNow! by means of assembly code generated by a
special-purpose compiler. It is hard to produce assembly code that
works in both 32-bit and 64-bit mode. <hr>
Next: <a href="section3.html" rel=precedes>Using FFTW</a>.<br>
Back: <a href="section1.html" rev=precedes>Introduction and General Information</a>.<br>
<a href="index.html" rev=subdocument>Return to contents</a>.<p>
<address>
<A href="http://www.fftw.org">Matteo Frigo and Steven G. Johnson</A> / <A href="mailto:fftw@fftw.org">fftw@fftw.org</A>
- 14 September 2021
</address><br>
Extracted from FFTW Frequently Asked Questions with Answers,
Copyright &copy; 2021 Matteo Frigo and Massachusetts Institute of Technology.
</body></html>

View File

@@ -0,0 +1,334 @@
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN">
<html>
<head><title>
FFTW FAQ - Section 3
</title>
<link rev="made" href="mailto:fftw@fftw.org">
<link rel="Contents" href="index.html">
<link rel="Start" href="index.html">
<link rel="Next" href="section4.html"><link rel="Previous" href="section2.html"><link rel="Bookmark" title="FFTW FAQ" href="index.html">
</head><body text="#000000" bgcolor="#FFFFFF"><h1>
FFTW FAQ - Section 3 <br>
Using FFTW
</h1>
<ul>
<li><a href="#fftw2to3" rel=subdocument>Q3.1. Why not support the FFTW 2 interface in FFTW
3?</a>
<li><a href="#planperarray" rel=subdocument>Q3.2. Why do FFTW 3 plans encapsulate the input/output arrays and not just
the algorithm?</a>
<li><a href="#slow" rel=subdocument>Q3.3. FFTW seems really slow.</a>
<li><a href="#slows" rel=subdocument>Q3.4. FFTW slows down after repeated calls.</a>
<li><a href="#segfault" rel=subdocument>Q3.5. An FFTW routine is crashing when I call it.</a>
<li><a href="#fortran64" rel=subdocument>Q3.6. My Fortran program crashes when calling FFTW.</a>
<li><a href="#conventions" rel=subdocument>Q3.7. FFTW gives results different from my old
FFT.</a>
<li><a href="#nondeterministic" rel=subdocument>Q3.8. FFTW gives different results between runs</a>
<li><a href="#savePlans" rel=subdocument>Q3.9. Can I save FFTW's plans?</a>
<li><a href="#whyscaled" rel=subdocument>Q3.10. Why does your inverse transform return a scaled
result?</a>
<li><a href="#centerorigin" rel=subdocument>Q3.11. How can I make FFTW put the origin (zero frequency) at the center of
its output?</a>
<li><a href="#imageaudio" rel=subdocument>Q3.12. How do I FFT an image/audio file in <i>foobar</i> format?</a>
<li><a href="#linkfails" rel=subdocument>Q3.13. My program does not link (on Unix).</a>
<li><a href="#linkheader" rel=subdocument>Q3.14. I included your header, but linking still
fails.</a>
<li><a href="#nostack" rel=subdocument>Q3.15. My program crashes, complaining about stack
space.</a>
<li><a href="#leaks" rel=subdocument>Q3.16. FFTW seems to have a memory leak.</a>
<li><a href="#allzero" rel=subdocument>Q3.17. The output of FFTW's transform is all zeros.</a>
<li><a href="#vbetalia" rel=subdocument>Q3.18. How do I call FFTW from the Microsoft language du
jour?</a>
<li><a href="#pruned" rel=subdocument>Q3.19. Can I compute only a subset of the DFT outputs?</a>
<li><a href="#transpose" rel=subdocument>Q3.20. Can I use FFTW's routines for in-place and out-of-place matrix
transposition?</a>
</ul><hr>
<h2><A name="fftw2to3">
Question 3.1. Why not support the FFTW 2 interface in FFTW
3?
</A></h2>
FFTW 3 has semantics incompatible with earlier versions: its plans can
only be used for a given stride, multiplicity, and other
characteristics of the input and output arrays; these stronger
semantics are necessary for performance reasons. Thus, it is
impossible to efficiently emulate the older interface (whose plans can
be used for any transform of the same size). We believe that it
should be possible to upgrade most programs without any difficulty,
however.
<h2><A name="planperarray">
Question 3.2. Why do FFTW 3 plans encapsulate the input/output arrays
and not just the algorithm?
</A></h2>
There are several reasons:
<ul>
<li>It was important for performance reasons that the plan be specific to
array characteristics like the stride (and alignment, for SIMD), and
requiring that the user maintain these invariants is error prone.
<li>In most high-performance applications, as far as we can tell, you are
usually transforming the same array over and over, so FFTW's semantics
should not be a burden.
<li>If you need to transform another array of the same size, creating a
new plan once the first exists is a cheap operation.
<li>If you need to transform many arrays of the same size at once, you
should really use the <code>plan_many</code> routines in FFTW's &quot;advanced&quot;
interface.
<li>If the abovementioned array characteristics are the same, you are
willing to pay close attention to the documentation, and you really
need to, we provide a &quot;new-array execution&quot; interface to
apply a plan to a new array.
</ul>
<h2><A name="slow">
Question 3.3. FFTW seems really slow.
</A></h2>
You are probably recreating the plan before every transform, rather
than creating it once and reusing it for all transforms of the same
size. FFTW is designed to be used in the following way:
<ul>
<li>First, you create a plan. This will take several seconds.
<li>Then, you reuse the plan many times to perform FFTs. These are fast.
</ul>
If you don't need to compute many transforms and the time for the
planner is significant, you have two options. First, you can use the
<code>FFTW_ESTIMATE</code> option in the planner, which uses heuristics
instead of runtime measurements and produces a good plan in a short
time. Second, you can use the wisdom feature to precompute the plan;
see <A href="#savePlans">Q3.9 `Can I save FFTW's plans?'</A>
<h2><A name="slows">
Question 3.4. FFTW slows down after repeated
calls.
</A></h2>
Probably, NaNs or similar are creeping into your data, and the
slowdown is due to the resulting floating-point exceptions. For
example, be aware that repeatedly FFTing the same array is a diverging
process (because FFTW computes the unnormalized transform).
<h2><A name="segfault">
Question 3.5. An FFTW routine is crashing when I call
it.
</A></h2>
Did the FFTW test programs pass (<code>make check</code>, or <code>cd tests; make bigcheck</code> if you want to be paranoid)? If so, you almost
certainly have a bug in your own code. For example, you could be
passing invalid arguments (such as wrongly-sized arrays) to FFTW, or
you could simply have memory corruption elsewhere in your program that
causes random crashes later on. Please don't complain to us unless
you can come up with a minimal self-contained program (preferably
under 30 lines) that illustrates the problem.
<h2><A name="fortran64">
Question 3.6. My Fortran program crashes when calling
FFTW.
</A></h2>
As described in the manual, on 64-bit machines you must store the
plans in variables large enough to hold a pointer, for example
<code>integer*8</code>. We recommend using <code>integer*8</code> on 32-bit machines as well, to simplify porting.
<h2><A name="conventions">
Question 3.7. FFTW gives results different from my old
FFT.
</A></h2>
People follow many different conventions for the DFT, and you should
be sure to know the ones that we use (described in the FFTW manual).
In particular, you should be aware that the
<code>FFTW_FORWARD</code>/<code>FFTW_BACKWARD</code> directions correspond to signs of -1/+1 in the exponent of the DFT definition.
(<i>Numerical Recipes</i> uses the opposite convention.)
<p>
You should also know that we compute an unnormalized transform. In
contrast, Matlab is an example of program that computes a normalized
transform. See <A href="#whyscaled">Q3.10 `Why does your inverse transform return a scaled
result?'</A>.
<p>
Finally, note that floating-point arithmetic is not exact, so
different FFT algorithms will give slightly different results (on the
order of the numerical accuracy; typically a fractional difference of
1e-15 or so in double precision).
<h2><A name="nondeterministic">
Question 3.8. FFTW gives different results between
runs
</A></h2>
If you use <code>FFTW_MEASURE</code> or <code>FFTW_PATIENT</code> mode, then the algorithm FFTW employs is not deterministic: it depends on
runtime performance measurements. This will cause the results to vary
slightly from run to run. However, the differences should be slight,
on the order of the floating-point precision, and therefore should
have no practical impact on most applications.
<p>
If you use saved plans (wisdom) or <code>FFTW_ESTIMATE</code> mode, however, then the algorithm is deterministic and the results should be
identical between runs.
<h2><A name="savePlans">
Question 3.9. Can I save FFTW's plans?
</A></h2>
Yes. Starting with version 1.2, FFTW provides the
<code>wisdom</code> mechanism for saving plans; see the FFTW manual.
<h2><A name="whyscaled">
Question 3.10. Why does your inverse transform return a scaled
result?
</A></h2>
Computing the forward transform followed by the backward transform (or
vice versa) yields the original array scaled by the size of the array.
(For multi-dimensional transforms, the size of the array is the
product of the dimensions.) We could, instead, have chosen a
normalization that would have returned the unscaled array. Or, to
accomodate the many conventions in this matter, the transform routines
could have accepted a &quot;scale factor&quot; parameter. We did not
do this, however, for two reasons. First, we didn't want to sacrifice
performance in the common case where the scale factor is 1. Second, in
real applications the FFT is followed or preceded by some computation
on the data, into which the scale factor can typically be absorbed at
little or no cost.
<h2><A name="centerorigin">
Question 3.11. How can I make FFTW put the origin (zero frequency) at
the center of its output?
</A></h2>
For human viewing of a spectrum, it is often convenient to put the
origin in frequency space at the center of the output array, rather
than in the zero-th element (the default in FFTW). If all of the
dimensions of your array are even, you can accomplish this by simply
multiplying each element of the input array by (-1)^(i + j + ...),
where i, j, etcetera are the indices of the element. (This trick is a
general property of the DFT, and is not specific to FFTW.)
<h2><A name="imageaudio">
Question 3.12. How do I FFT an image/audio file in
<i>foobar</i> format?
</A></h2>
FFTW performs an FFT on an array of floating-point values. You can
certainly use it to compute the transform of an image or audio stream,
but you are responsible for figuring out your data format and
converting it to the form FFTW requires.
<h2><A name="linkfails">
Question 3.13. My program does not link (on
Unix).
</A></h2>
The libraries must be listed in the correct order
(<code>-lfftw3 -lm</code> for FFTW 3.x) and <i>after</i> your program sources/objects. (The general rule is that if <i>A</i> uses <i>B</i>, then <i>A</i> must be listed before <i>B</i> in the link command.).
<h2><A name="linkheader">
Question 3.14. I included your header, but linking still
fails.
</A></h2>
You're a C++ programmer, aren't you? You have to compile the FFTW
library and link it into your program, not just
<code>#include &lt;fftw3.h&gt;</code>. (Yes, this is really a FAQ.)
<h2><A name="nostack">
Question 3.15. My program crashes, complaining about stack
space.
</A></h2>
You cannot declare large arrays with automatic storage (e.g. via
<code>fftw_complex array[N]</code>); you should use <code>fftw_malloc</code> (or equivalent) to allocate the arrays you want
to transform if they are larger than a few hundred elements.
<h2><A name="leaks">
Question 3.16. FFTW seems to have a memory
leak.
</A></h2>
After you create a plan, FFTW caches the information required to
quickly recreate the plan. (See <A href="#savePlans">Q3.9 `Can I save FFTW's plans?'</A>) It also maintains a small amount of other persistent memory. You can deallocate all of
FFTW's internally allocated memory, if you wish, by calling
<code>fftw_cleanup()</code>, as documented in the manual.
<h2><A name="allzero">
Question 3.17. The output of FFTW's transform is all
zeros.
</A></h2>
You should initialize your input array <i>after</i> creating the plan, unless you use <code>FFTW_ESTIMATE</code>: planning with <code>FFTW_MEASURE</code> or <code>FFTW_PATIENT</code> overwrites the input/output arrays, as described in the manual.
<h2><A name="vbetalia">
Question 3.18. How do I call FFTW from the Microsoft language du
jour?
</A></h2>
Please <i>do not</i> ask us Windows-specific questions. We do not
use Windows. We know nothing about Visual Basic, Visual C++, or .NET.
Please find the appropriate Usenet discussion group and ask your
question there. See also <A href="section2.html#runOnWindows">Q2.2 `Does FFTW run on Windows?'</A>.
<h2><A name="pruned">
Question 3.19. Can I compute only a subset of the DFT
outputs?
</A></h2>
In general, no, an FFT intrinsically computes all outputs from all
inputs. In principle, there is something called a
<i>pruned FFT</i> that can do what you want, but to compute K outputs out of N the
complexity is in general O(N log K) instead of O(N log N), thus saving
only a small additive factor in the log. (The same argument holds if
you instead have only K nonzero inputs.)
<p>
There are some specific cases in which you can get the O(N log K)
performance benefits easily, however, by combining a few ordinary
FFTs. In particular, the case where you want the first K outputs,
where K divides N, can be handled by performing N/K transforms of size
K and then summing the outputs multiplied by appropriate phase
factors. For more details, see <A href="http://www.fftw.org/pruned.html">pruned FFTs with FFTW</A>.
<p>
There are also some algorithms that compute pruned transforms
<i>approximately</i>, but they are beyond the scope of this FAQ.
<h2><A name="transpose">
Question 3.20. Can I use FFTW's routines for in-place and
out-of-place matrix transposition?
</A></h2>
You can use the FFTW guru interface to create a rank-0 transform of
vector rank 2 where the vector strides are transposed. (A rank-0
transform is equivalent to a 1D transform of size 1, which. just
copies the input into the output.) Specifying the same location for
the input and output makes the transpose in-place.
<p>
For double-valued data stored in row-major format, plan creation looks
like this: <pre>
fftw_plan plan_transpose(int rows, int cols, double *in, double *out)
{
const unsigned flags = FFTW_ESTIMATE; /* other flags are possible */
fftw_iodim howmany_dims[2];
howmany_dims[0].n = rows;
howmany_dims[0].is = cols;
howmany_dims[0].os = 1;
howmany_dims[1].n = cols;
howmany_dims[1].is = 1;
howmany_dims[1].os = rows;
return fftw_plan_guru_r2r(/*rank=*/ 0, /*dims=*/ NULL,
/*howmany_rank=*/ 2, howmany_dims,
in, out, /*kind=*/ NULL, flags);
}
</pre>
(This entry was written by Rhys Ulerich.)
<hr>
Next: <a href="section4.html" rel=precedes>Internals of FFTW</a>.<br>
Back: <a href="section2.html" rev=precedes>Installing FFTW</a>.<br>
<a href="index.html" rev=subdocument>Return to contents</a>.<p>
<address>
<A href="http://www.fftw.org">Matteo Frigo and Steven G. Johnson</A> / <A href="mailto:fftw@fftw.org">fftw@fftw.org</A>
- 14 September 2021
</address><br>
Extracted from FFTW Frequently Asked Questions with Answers,
Copyright &copy; 2021 Matteo Frigo and Massachusetts Institute of Technology.
</body></html>

View File

@@ -0,0 +1,64 @@
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN">
<html>
<head><title>
FFTW FAQ - Section 4
</title>
<link rev="made" href="mailto:fftw@fftw.org">
<link rel="Contents" href="index.html">
<link rel="Start" href="index.html">
<link rel="Next" href="section5.html"><link rel="Previous" href="section3.html"><link rel="Bookmark" title="FFTW FAQ" href="index.html">
</head><body text="#000000" bgcolor="#FFFFFF"><h1>
FFTW FAQ - Section 4 <br>
Internals of FFTW
</h1>
<ul>
<li><a href="#howworks" rel=subdocument>Q4.1. How does FFTW work?</a>
<li><a href="#whyfast" rel=subdocument>Q4.2. Why is FFTW so fast?</a>
</ul><hr>
<h2><A name="howworks">
Question 4.1. How does FFTW work?
</A></h2>
The innovation (if it can be so called) in FFTW consists in having a
variety of composable <i>solvers</i>, representing different FFT algorithms and implementation strategies, whose combination into a
particular <i>plan</i> for a given size can be determined at runtime according to the characteristics of your machine/compiler.
This peculiar software architecture allows FFTW to adapt itself to
almost any machine.
<p>
For more details (albeit somewhat outdated), see the paper &quot;FFTW:
An Adaptive Software Architecture for the FFT&quot;, by M. Frigo and
S. G. Johnson, <i>Proc. ICASSP</i> 3, 1381 (1998), also available at <A href="http://www.fftw.org">the FFTW web page</A>.
<h2><A name="whyfast">
Question 4.2. Why is FFTW so fast?
</A></h2>
This is a complex question, and there is no simple answer. In fact,
the authors do not fully know the answer, either. In addition to many
small performance hacks throughout FFTW, there are three general
reasons for FFTW's speed.
<ul>
<li> FFTW uses a variety of FFT algorithms and implementation styles
that can be arbitrarily composed to adapt itself to
a machine. See <A href="#howworks">Q4.1 `How does FFTW work?'</A>.
<li> FFTW uses a code generator to produce highly-optimized
routines for computing small transforms.
<li> FFTW uses explicit divide-and-conquer to take advantage
of the memory hierarchy.
</ul>
For more details (albeit somewhat outdated), see the paper &quot;FFTW:
An Adaptive Software Architecture for the FFT&quot;, by M. Frigo and
S. G. Johnson, <i>Proc. ICASSP</i> 3, 1381 (1998), available along with other references at
<A href="http://www.fftw.org">the FFTW web page</A>. <hr>
Next: <a href="section5.html" rel=precedes>Known bugs</a>.<br>
Back: <a href="section3.html" rev=precedes>Using FFTW</a>.<br>
<a href="index.html" rev=subdocument>Return to contents</a>.<p>
<address>
<A href="http://www.fftw.org">Matteo Frigo and Steven G. Johnson</A> / <A href="mailto:fftw@fftw.org">fftw@fftw.org</A>
- 14 September 2021
</address><br>
Extracted from FFTW Frequently Asked Questions with Answers,
Copyright &copy; 2021 Matteo Frigo and Massachusetts Institute of Technology.
</body></html>

View File

@@ -0,0 +1,148 @@
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN">
<html>
<head><title>
FFTW FAQ - Section 5
</title>
<link rev="made" href="mailto:fftw@fftw.org">
<link rel="Contents" href="index.html">
<link rel="Start" href="index.html">
<link rel="Previous" href="section4.html"><link rel="Bookmark" title="FFTW FAQ" href="index.html">
</head><body text="#000000" bgcolor="#FFFFFF"><h1>
FFTW FAQ - Section 5 <br>
Known bugs
</h1>
<ul>
<li><a href="#rfftwndbug" rel=subdocument>Q5.1. FFTW 1.1 crashes in rfftwnd on Linux.</a>
<li><a href="#fftwmpibug" rel=subdocument>Q5.2. The MPI transforms in FFTW 1.2 give incorrect results/leak
memory.</a>
<li><a href="#testsingbug" rel=subdocument>Q5.3. The test programs in FFTW 1.2.1 fail when I change FFTW to use single
precision.</a>
<li><a href="#teststoobig" rel=subdocument>Q5.4. The test program in FFTW 1.2.1 fails for n &gt;
46340.</a>
<li><a href="#linuxthreads" rel=subdocument>Q5.5. The threaded code fails on Linux Redhat 5.0</a>
<li><a href="#bigrfftwnd" rel=subdocument>Q5.6. FFTW 2.0's rfftwnd fails for rank &gt; 1 transforms with a final
dimension &gt;= 65536.</a>
<li><a href="#primebug" rel=subdocument>Q5.7. FFTW 2.0's complex transforms give the wrong results with prime
factors 17 to 97.</a>
<li><a href="#mpichbug" rel=subdocument>Q5.8. FFTW 2.1.1's MPI test programs crash with
MPICH.</a>
<li><a href="#aixthreadbug" rel=subdocument>Q5.9. FFTW 2.1.2's multi-threaded transforms don't work on
AIX.</a>
<li><a href="#bigprimebug" rel=subdocument>Q5.10. FFTW 2.1.2's complex transforms give incorrect results for large prime
sizes.</a>
<li><a href="#solaristhreadbug" rel=subdocument>Q5.11. FFTW 2.1.3's multi-threaded transforms don't give any speedup on
Solaris.</a>
<li><a href="#aixflags" rel=subdocument>Q5.12. FFTW 2.1.3 crashes on AIX.</a>
</ul><hr>
<h2><A name="rfftwndbug">
Question 5.1. FFTW 1.1 crashes in rfftwnd on
Linux.
</A></h2>
This bug was fixed in FFTW 1.2. There was a bug in
<code>rfftwnd</code> causing an incorrect amount of memory to be allocated. The bug showed
up in Linux with libc-5.3.12 (and nowhere else that we know of).
<h2><A name="fftwmpibug">
Question 5.2. The MPI transforms in FFTW 1.2 give incorrect
results/leak memory.
</A></h2>
These bugs were corrected in FFTW 1.2.1. The MPI transforms (really,
just the transpose routines) in FFTW 1.2 had bugs that could cause
errors in some situations.
<h2><A name="testsingbug">
Question 5.3. The test programs in FFTW 1.2.1 fail when I change FFTW
to use single precision.
</A></h2>
This bug was fixed in FFTW 1.3. (Older versions of FFTW did
work in single precision, but the test programs didn't--the error
tolerances in the tests were set for double precision.)
<h2><A name="teststoobig">
Question 5.4. The test program in FFTW 1.2.1 fails for n &gt;
46340.
</A></h2>
This bug was fixed in FFTW 1.3. FFTW 1.2.1 produced the right answer,
but the test program was wrong. For large n, n*n in the naive
transform that we used for comparison overflows 32 bit integer
precision, breaking the test.
<h2><A name="linuxthreads">
Question 5.5. The threaded code fails on Linux Redhat
5.0
</A></h2>
We had problems with glibc-2.0.5. The code should work with
glibc-2.0.7.
<h2><A name="bigrfftwnd">
Question 5.6. FFTW 2.0's rfftwnd fails for rank &gt; 1 transforms
with a final dimension &gt;= 65536.
</A></h2>
This bug was fixed in FFTW 2.0.1. (There was a 32-bit integer
overflow due to a poorly-parenthesized expression.)
<h2><A name="primebug">
Question 5.7. FFTW 2.0's complex transforms give the wrong results
with prime factors 17 to 97.
</A></h2>
There was a bug in the complex transforms that could cause incorrect
results under (hopefully rare) circumstances for lengths with
intermediate-size prime factors (17-97). This bug was fixed in FFTW
2.1.1.
<h2><A name="mpichbug">
Question 5.8. FFTW 2.1.1's MPI test programs crash with
MPICH.
</A></h2>
This bug was fixed in FFTW 2.1.2. The 2.1/2.1.1 MPI test programs
crashed when using the MPICH implementation of MPI with the
<code>ch_p4</code> device (TCP/IP); the transforms themselves worked fine.
<h2><A name="aixthreadbug">
Question 5.9. FFTW 2.1.2's multi-threaded transforms don't work on
AIX.
</A></h2>
This bug was fixed in FFTW 2.1.3. The multi-threaded transforms in
previous versions didn't work with AIX's
<code>pthreads</code> implementation, which idiosyncratically creates threads in detached
(non-joinable) mode by default.
<h2><A name="bigprimebug">
Question 5.10. FFTW 2.1.2's complex transforms give incorrect results
for large prime sizes.
</A></h2>
This bug was fixed in FFTW 2.1.3. FFTW's complex-transform algorithm
for prime sizes (in versions 2.0 to 2.1.2) had an integer overflow
problem that caused incorrect results for many primes greater than
32768 (on 32-bit machines). (Sizes without large prime factors are
not affected.)
<h2><A name="solaristhreadbug">
Question 5.11. FFTW 2.1.3's multi-threaded transforms don't give any
speedup on Solaris.
</A></h2>
This bug was fixed in FFTW 2.1.4. (By default, Solaris creates
threads that do not parallelize over multiple processors, so one has
to request the proper behavior specifically.)
<h2><A name="aixflags">
Question 5.12. FFTW 2.1.3 crashes on AIX.
</A></h2>
The FFTW 2.1.3 <code>configure</code> script picked incorrect compiler flags for the <code>xlc</code> compiler on newer IBM processors. This
is fixed in FFTW 2.1.4. <hr>
Back: <a href="section4.html" rev=precedes>Internals of FFTW</a>.<br>
<a href="index.html" rev=subdocument>Return to contents</a>.<p>
<address>
<A href="http://www.fftw.org">Matteo Frigo and Steven G. Johnson</A> / <A href="mailto:fftw@fftw.org">fftw@fftw.org</A>
- 14 September 2021
</address><br>
Extracted from FFTW Frequently Asked Questions with Answers,
Copyright &copy; 2021 Matteo Frigo and Massachusetts Institute of Technology.
</body></html>

View File

@@ -0,0 +1,7 @@
\ References for the FFTW FAQ
\
the FFTW web page \ http://www.fftw.org
FFTW Windows installation notes \ http://www.fftw.org/install/windows.html
Categories of Free and Non-Free Software \ http://www.gnu.org/philosophy/categories.html
the Caml web page \ http://caml.inria.fr
pruned FFTs with FFTW \ http://www.fftw.org/pruned.html

View File

@@ -0,0 +1,189 @@
## ASCII output
# Copyright (C) 1993-1995 Ian Jackson.
# This file is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation; either version 2, or (at your option)
# any later version.
# It is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
# You should have received a copy of the GNU General Public License
# along with GNU Emacs; see the file COPYING. If not, write to
# the Free Software Foundation, Inc., 59 Temple Place - Suite 330,
# Boston, MA 02111-1307, USA.
# (Note: I do not consider works produced using these BFNN processing
# tools to be derivative works of the tools, so they are NOT covered
# by the GPL. However, I would appreciate it if you credited me if
# appropriate in any documents you format using BFNN.)
sub ascii_init {
open(ASCII,">$prefix.ascii");
}
sub ascii_startmajorheading {
print ASCII '='x79,"\n\n";
$ascii_status= 'h';
&ascii_text($_[0] ? "Section $_[0]. " : '');
}
sub ascii_startminorheading {
print ASCII '-'x79,"\n\n";
$ascii_status= 'h';
}
sub ascii_italic { &ascii_text('*'); }
sub ascii_enditalic { $ascii_para .= '*'; }
sub ascii_email { &ascii_text('<'); } sub ascii_endemail { &ascii_text('>'); }
sub ascii_ftpon { } sub ascii_endftpon { }
sub ascii_ftpin { } sub ascii_endftpin { }
sub ascii_docref { } sub ascii_enddocref { }
sub ascii_courier { } sub ascii_endcourier { }
sub ascii_newsgroup { } sub ascii_endnewsgroup { }
sub ascii_ftpsilent { $ascii_ignore++; }
sub ascii_endftpsilent { $ascii_ignore--; }
sub ascii_text {
return if $ascii_ignore;
if ($ascii_status eq '') {
$ascii_status= 'p';
}
$ascii_para .= $_[0];
}
sub ascii_tab {
local ($n) = $_[0]-length($ascii_para);
$ascii_para .= ' 'x$n if $n>0;
}
sub ascii_newline {
return unless $ascii_status eq 'p';
&ascii_writepara;
}
sub ascii_writepara {
local ($thisline, $thisword, $rest);
for (;;) {
last unless $ascii_para =~ m/\S/;
$thisline= $ascii_indentstring;
for (;;) {
last unless $ascii_para =~ m/^(\s*\S+)/;
unless (length($1) + length($thisline) < 75 ||
length($thisline) == length($ascii_indentstring)) {
last;
}
$thisline .= $1;
$ascii_para= $';
}
$ascii_para =~ s/^\s*//;
print ASCII $thisline,"\n";
$ascii_indentstring= $ascii_nextindent;
last unless length($ascii_para);
}
$ascii_status= ''; $ascii_para= '';
}
sub ascii_endpara {
return unless $ascii_status eq 'p';
&ascii_writepara;
print ASCII "\n";
}
sub ascii_endheading {
$ascii_para =~ s/\s*$//;
print ASCII "$ascii_para\n\n";
$ascii_status= '';
$ascii_para= '';
}
sub ascii_endmajorheading { &ascii_endheading(@_); }
sub ascii_endminorheading { &ascii_endheading(@_); }
sub ascii_startverbatim {
$ascii_vstatus= $ascii_status;
&ascii_writepara;
}
sub ascii_verbatim {
print ASCII $_[0],"\n";
}
sub ascii_endverbatim {
$ascii_status= $ascii_vstatus;
}
sub ascii_finish {
close(ASCII);
}
sub ascii_startindex { $ascii_status= ''; }
sub ascii_endindex { $ascii_status= 'p'; }
sub ascii_endindexitem {
printf ASCII " %-11s %-.66s\n",$ascii_left,$ascii_para;
$ascii_status= 'p';
$ascii_para= '';
}
sub ascii_startindexitem {
$ascii_left= $_[1];
}
sub ascii_startindexmainitem {
$ascii_left= $_[1];
print ASCII "\n" if $ascii_status eq 'p';
}
sub ascii_startindent {
$ascii_istatus= $ascii_status;
&ascii_writepara;
$ascii_indentstring= " $ascii_indentstring";
$ascii_nextindent= " $ascii_nextindent";
}
sub ascii_endindent {
$ascii_indentstring =~ s/^ //;
$ascii_nextindent =~ s/^ //;
$ascii_status= $ascii_istatus;
}
sub ascii_startpackedlist { $ascii_plc=0; }
sub ascii_endpackedlist { &ascii_newline if !$ascii_plc; }
sub ascii_packeditem {
&ascii_newline if !$ascii_plc;
&ascii_tab($ascii_plc*40+5);
$ascii_plc= !$ascii_plc;
}
sub ascii_startlist {
&ascii_endpara;
$ascii_indentstring= " $ascii_indentstring";
$ascii_nextindent= " $ascii_nextindent";
}
sub ascii_endlist {
&ascii_endpara;
$ascii_indentstring =~ s/^ //;
$ascii_nextindent =~ s/^ //;
}
sub ascii_item {
&ascii_newline;
$ascii_indentstring =~ s/ $/* /;
}
sub ascii_pageref {
&ascii_text("Q$_[1] \`");
}
sub ascii_endpageref {
&ascii_text("'");
}
1;

View File

@@ -0,0 +1,337 @@
## HTML output
# Copyright (C) 1993-1995 Ian Jackson.
# This file is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation; either version 2, or (at your option)
# any later version.
# It is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
# You should have received a copy of the GNU General Public License
# along with GNU Emacs; see the file COPYING. If not, write to
# the Free Software Foundation, Inc., 59 Temple Place - Suite 330,
# Boston, MA 02111-1307, USA.
# (Note: I do not consider works produced using these BFNN processing
# tools to be derivative works of the tools, so they are NOT covered
# by the GPL. However, I would appreciate it if you credited me if
# appropriate in any documents you format using BFNN.)
%saniarray= ('<','lt', '>','gt', '&','amp', '"','quot');
sub html_init {
$html_prefix = './'.$prefix;
$html_prefix =~ s:^\.//:/:;
system('rm','-r',"$html_prefix.html");
system('mkdir',"$html_prefix.html");
open(HTML,">$html_prefix.html/index.html");
print HTML "<!DOCTYPE HTML PUBLIC \"-//W3C//DTD HTML 3.2//EN\">\n";
print HTML "<html>\n";
$html_needpara= -1;
$html_end='';
chop($html_date=`date '+%d %B %Y'`);
chop($html_year=`date '+%Y'`);
}
sub html_startup {
print HTML <<END;
<head><title>
$user_title
</title>
<link rev="made" href="mailto:$user_authormail">
<link rel="Contents" href="index.html">
<link rel="Start" href="index.html">
<META name="description"
content="Frequently asked questions and answers (FAQ) for FFTW.">
<link rel="Bookmark" title="FFTW FAQ" href="index.html">
<LINK rel="Bookmark" title="FFTW Home Page"
href="http://www.fftw.org">
<LINK rel="Bookmark" title="FFTW Manual"
href="http://www.fftw.org/doc/">
</head><body text="#000000" bgcolor="#FFFFFF"><h1>
$user_title
</h1>
END
&html_readrefs($_[0]);
if (length($user_copyrightref)) {
local ($refn) = $qrefn{$user_copyrightref};
if (!length($refn)) {
warn "unknown question (copyright) `$user_copyrightref'";
}
$refn =~ m/(\d+)\.(\d+)/;
local ($s,$n) = ($1,$2);
$html_copyrighthref= ($s == $html_sectionn)?'':"section$s.html";
$html_copyrighthref.= "#$qn2ref{$s,$n}";
}
}
sub html_close {
print HTML $html_end,"<address>\n$user_author\n";
print HTML "- $html_date\n</address><br>\n";
print HTML "Extracted from $user_title,\n";
print HTML "<A href=\"$html_copyrighthref\">" if length($html_copyrighthref);
print HTML "Copyright &copy; $html_year $user_copyholder.";
print HTML "</A>" if length($html_copyrighthref);
print HTML "\n</body></html>\n";
close(HTML);
}
sub html_startmajorheading {
local ($ref, $this,$next,$back) = @_;
local ($nextt,$backt);
$this =~ s/^Section /section/; $html_sectionn= $ref;
$next =~ s/^Section /section/ && ($nextt= $sn2title{$'});
$back =~ s/^Section /section/ ? ($backt= $sn2title{$'}) : ($back='');
if ($html_sectionn) {
&html_close;
open(HTML,">$html_prefix.html/$this.html");
print HTML "<!DOCTYPE HTML PUBLIC \"-//W3C//DTD HTML 3.2//EN\">\n";
print HTML "<html>\n";
$html_end= "<hr>\n";
$html_end.= "Next: <a href=\"$next.html\" rel=precedes>$nextt</a>.<br>\n"
if $next;
$html_end.= "Back: <a href=\"$back.html\" rev=precedes>$backt</a>.<br>\n"
if $back;
$html_end.= "<a href=\"index.html\" rev=subdocument>";
$html_end.= "Return to contents</a>.<p>\n";
print HTML <<END;
<head><title>
$user_brieftitle - Section $html_sectionn
</title>
<link rev="made" href="mailto:$user_authormail">
<link rel="Contents" href="index.html">
<link rel="Start" href="index.html">
END
print HTML "<link rel=\"Next\" href=\"$next.html\">" if $next;
print HTML "<link rel=\"Previous\" href=\"$back.html\">" if $back;
print HTML <<END;
<link rel="Bookmark" title="FFTW FAQ" href="index.html">
</head><body text="#000000" bgcolor="#FFFFFF"><h1>
$user_brieftitle - Section $html_sectionn <br>
END
$html_needpara= -1;
}
else {
print HTML "\n<h1>\n";
$html_needpara=-1;
}
}
sub html_endmajorheading {
print HTML "\n</h1>\n\n";
$html_needpara=-1;
}
sub html_startminorheading {
local ($ref, $this) = @_;
$html_needpara=0;
$this =~ m/^Question (\d+)\.(\d+)/;
local ($s,$n) = ($1,$2);
print HTML "\n<h2><A name=\"$qn2ref{$s,$n}\">\n";
}
sub html_endminorheading {
print HTML "\n</A></h2>\n\n";
$html_needpara=-1;
}
sub html_newsgroup { &arg('newsgroup'); }
sub html_endnewsgroup { &endarg('newsgroup'); }
sub html_do_newsgroup {
print HTML "<A href=\"news:$_[0]\"><code>$_[0]</code></A>";
}
sub html_email { &arg('email'); }
sub html_endemail { &endarg('email'); }
sub html_do_email {
print HTML "<A href=\"mailto:$_[0]\"><code>$_[0]</code></A>";
}
sub html_courier { print HTML "<code>" ; }
sub html_endcourier { print HTML "</code>"; }
sub html_italic { print HTML "<i>" ; }
sub html_enditalic { print HTML "</i>" ; }
sub html_docref { &arg('docref'); }
sub html_enddocref { &endarg('docref'); }
sub html_do_docref {
if (!defined($html_refval{$_[0]})) {
warn "undefined HTML reference $_[0]";
$html_refval{$n}='UNDEFINED';
}
print HTML "<A href=\"$html_refval{$_[0]}\">";
&recurse($_[0]);
print HTML "</A>";
}
sub html_readrefs {
local ($p);
open(HTMLREFS,"<$_[0]") || (warn("failed to open HTML refs $_[0]: $!"),return);
while(<HTMLREFS>) {
next if m/^\\\s/;
s/\s*\n$//;
if (s/^\\prefix\s*//) {
$p= $'; next;
} elsif (s/^\s*(\S.*\S)\s*\\\s*//) {
$_=$1; $v=$';
s/\\\\/\\/g;
$html_refval{$_}= $p.$v;
} else {
warn("ununderstood line in HTML refs >$_<");
}
}
close(HTMLREFS);
}
sub html_ftpsilent { &arg('ftpsilent'); }
sub html_endftpsilent { &endarg('ftpsilent'); }
sub html_do_ftpsilent {
if ($_[0] =~ m/:/) {
$html_ftpsite= $`;
$html_ftpdir= $'.'/';
} else {
$html_ftpsite= $_[0];
$html_ftpdir= '';
}
}
sub html_ftpon { &arg('ftpon'); }
sub html_endftpon { &endarg('ftpon'); }
sub html_do_ftpon {
#print STDERR "ftpon($_[0])\n";
$html_ftpsite= $_[0]; $html_ftpdir= '';
print HTML "<code>";
&recurse($_[0]);
print HTML "</code>";
}
sub html_ftpin { &arg('ftpin'); }
sub html_endftpin { &endarg('ftpin'); }
sub html_do_ftpin {
#print STDERR "ftpin($_[0])\n";
print HTML "<A href=\"ftp://$html_ftpsite$html_ftpdir$_[0]\"><code>";
&recurse($_[0]);
print HTML "</code></A>";
}
sub html_text {
print HTML "\n<p>\n" if $html_needpara > 0;
$html_needpara=0;
$html_stuff= &html_sanitise($_[0]);
while ($html_stuff =~ s/^(.{40,70}) //) {
print HTML "$1\n";
}
print HTML $html_stuff;
}
sub html_tab {
$htmltabignore++ || warn "html tab ignored";
}
sub html_newline { print HTML "<br>\n" ; }
sub html_startverbatim { print HTML "<pre>\n" ; }
sub html_verbatim { print HTML &html_sanitise($_[0]),"\n"; }
sub html_endverbatim { print HTML "</pre>\n" ; $html_needpara= -1; }
sub html_endpara {
$html_needpara || $html_needpara++;
}
sub html_finish {
&html_close;
}
sub html_startindex {
print HTML "<ul>\n";
}
sub html_endindex {
print HTML "</ul><hr>\n";
}
sub html_startindexitem {
local ($ref,$qval) = @_;
$qval =~ m/Q(\d+)\.(\d+)/;
local ($s,$n) = ($1,$2);
print HTML "<li><a href=\"";
print HTML ($s == $html_sectionn)?'':"section$s.html";
print HTML "#$qn2ref{$s,$n}\" rel=subdocument>Q$s.$n. ";
$html_indexunhead='';
}
sub html_startindexmainitem {
local ($ref,$s) = @_;
$s =~ m/\d+/; $s= $&;
print HTML "<br><br>" if ($s > 1);
print HTML "<li><b><font size=\"+2\"><a href=\"section$s.html\" rel=subdocument>Section $s. ";
$html_indexunhead='</font></b>';
}
sub html_endindexitem {
print HTML "</a>$html_indexunhead\n";
}
sub html_startlist {
print HTML "\n";
$html_itemend="<ul>";
}
sub html_endlist {
print HTML "$html_itemend\n</ul>\n";
$html_needpara=-1
}
sub html_item {
print HTML "$html_itemend\n<li>";
$html_itemend="";
$html_needpara=-1;
}
sub html_startpackedlist {
print HTML "\n";
$html_itemend="<dir>";
}
sub html_endpackedlist {
print HTML "$html_itemend\n</dir>\n";
$html_needpara=-1;
}
sub html_packeditem {
print HTML "$html_itemend\n<li>";
$html_itemend="";
$html_needpara=-1;
}
sub html_startindent { print HTML "<blockquote>\n"; }
sub html_endindent { print HTML "</blockquote>\n"; }
sub html_pageref {
local ($ref,$sq) = @_;
$sq =~ m/(\d+)\.(\d+)/;
local ($s,$n) = ($1,$2);
print HTML "<A href=\"";
print HTML ($s == $html_sectionn)?'':"section$s.html";
print HTML "#$qn2ref{$s,$n}\">Q$sq \`";
}
sub html_endpageref {
print HTML "'</A>";
}
sub html_sanitise {
local ($in) = @_;
local ($out);
while ($in =~ m/[<>&"]/) {
$out.= $`. '&'. $saniarray{$&}. ';';
$in=$';
}
$out.= $in;
$out;
}
1;

View File

@@ -0,0 +1,226 @@
## Info output
# Copyright (C) 1993-1995 Ian Jackson.
# This file is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation; either version 2, or (at your option)
# any later version.
# It is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
# You should have received a copy of the GNU General Public License
# along with GNU Emacs; see the file COPYING. If not, write to
# the Free Software Foundation, Inc., 59 Temple Place - Suite 330,
# Boston, MA 02111-1307, USA.
# (Note: I do not consider works produced using these BFNN processing
# tools to be derivative works of the tools, so they are NOT covered
# by the GPL. However, I would appreciate it if you credited me if
# appropriate in any documents you format using BFNN.)
sub info_init {
open(INFO,">$prefix.info");
print INFO <<END;
Info file: $prefix.info, -*-Text-*-
produced by bfnnconv.pl from the Bizarre Format With No Name.
END
}
sub info_heading {
# refstring Node Next Previous Up
print INFO "\nFile: $prefix.info, Node: $_[1]";
print INFO ", Next: $_[2]" if length($_[2]);
print INFO ", Previous: $_[3]" if length($_[3]);
print INFO ", Up: $_[4]" if length($_[4]);
print INFO "\n\n";
$info_status= '';
}
sub info_startmajorheading {
return if $_[0] eq '0';
&info_heading('s_'.$_[0],@_[1..$#_],'Top');
}
sub info_startminorheading {
&info_heading(@_);
}
sub info_italic { &info_text('*'); }
sub info_enditalic { $info_para .= '*'; }
sub info_email { &info_text('<'); } sub info_endemail { &info_text('>'); }
sub info_ftpon { } sub info_endftpon { }
sub info_ftpin { } sub info_endftpin { }
sub info_docref { } sub info_enddocref { }
sub info_courier { } sub info_endcourier { }
sub info_newsgroup { } sub info_endnewsgroup { }
sub info_ftpsilent { $info_ignore++; }
sub info_endftpsilent { $info_ignore--; }
sub info_text {
return if $info_ignore;
if ($info_status eq '') {
$info_status= 'p';
}
$info_para .= $_[0];
}
sub info_tab {
local ($n) = $_[0]-length($info_para);
$info_para .= ' 'x$n if $n>0;
}
sub info_newline {
return unless $info_status eq 'p';
print INFO &info_writepara;
}
sub info_writepara {
local ($thisline, $thisword, $rest, $output);
for (;;) {
last unless $info_para =~ m/\S/;
$thisline= $info_indentstring;
for (;;) {
last unless $info_para =~ m/^(\s*\S+)/;
unless (length($1) + length($thisline) < 75 ||
length($thisline) == length($info_indentstring)) {
last;
}
$thisline .= $1;
$info_para= $';
}
$info_para =~ s/^\s*//;
$output.= $thisline."\n";
$info_indentstring= $info_nextindent;
last unless length($info_para);
}
$info_status= ''; $info_para= '';
return $output;
}
sub info_endpara {
return unless $info_status eq 'p';
print INFO &info_writepara;
print INFO "\n";
}
sub info_endheading {
$info_para =~ s/\s*$//;
print INFO "$info_para\n\n";
$info_status= '';
$info_para= '';
}
sub info_endmajorheading { &info_endheading(@_); }
sub info_endminorheading { &info_endheading(@_); }
sub info_startverbatim {
print INFO &info_writepara;
}
sub info_verbatim {
print INFO $_[0],"\n";
}
sub info_endverbatim {
$info_status= $info_vstatus;
}
sub info_finish {
close(INFO);
}
sub info_startindex {
&info_endpara;
$info_moredetail= '';
$info_status= '';
}
sub info_endindex {
print INFO "$info_moredetail\n" if length($info_moredetail);
}
sub info_endindexitem {
$info_indentstring= sprintf("* %-17s ",$info_label.'::');
$info_nextindent= ' 'x20;
local ($txt);
$txt= &info_writepara;
if ($info_main) {
print INFO $label.$txt;
$txt =~ s/^.{20}//;
$info_moredetail.= $txt;
} else {
$info_moredetail.= $label.$txt;
}
$info_indentstring= $info_nextindent= '';
$info_status='p';
}
sub info_startindexitem {
print INFO "* Menu:\n" if $info_status eq '';
$info_status= '';
$info_label= $_[2];
$info_main= 0;
}
sub info_startindexmainitem {
print INFO "* Menu:\n" if $info_status eq '';
$info_label= $_[2];
$info_main= 1;
$info_moredetail .= "\n$_[2], ";
$info_status= '';
}
sub info_startindent {
$info_istatus= $info_status;
print INFO &info_writepara;
$info_indentstring= " $info_indentstring";
$info_nextindent= " $info_nextindent";
}
sub info_endindent {
$info_indentstring =~ s/^ //;
$info_nextindent =~ s/^ //;
$info_status= $info_istatus;
}
sub info_startpackedlist { $info_plc=0; }
sub info_endpackedlist { &info_newline if !$info_plc; }
sub info_packeditem {
&info_newline if !$info_plc;
&info_tab($info_plc*40+5);
$info_plc= !$info_plc;
}
sub info_startlist {
$info_istatus= $info_status;
print INFO &info_writepara;
$info_indentstring= " $info_indentstring";
$info_nextindent= " $info_nextindent";
}
sub info_endlist {
$info_indentstring =~ s/^ //;
$info_nextindent =~ s/^ //;
$info_status= $info_lstatus;
}
sub info_item {
&info_newline;
$info_indentstring =~ s/ $/* /;
}
sub info_pageref {
&info_text("*Note Question $_[1]:: \`");
}
sub info_endpageref {
&info_text("'");
}
1;

View File

@@ -0,0 +1,242 @@
## Lout output
# Copyright (C) 1993-1995 Ian Jackson.
# This file is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation; either version 2, or (at your option)
# any later version.
# It is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
# You should have received a copy of the GNU General Public License
# along with GNU Emacs; see the file COPYING. If not, write to
# the Free Software Foundation, Inc., 59 Temple Place - Suite 330,
# Boston, MA 02111-1307, USA.
# (Note: I do not consider works produced using these BFNN processing
# tools to be derivative works of the tools, so they are NOT covered
# by the GPL. However, I would appreciate it if you credited me if
# appropriate in any documents you format using BFNN.)
sub lout_init {
open(LOUT,">$prefix.lout");
chop($dprint= `date '+%d %B %Y'`);
$dprint =~ s/^0//;
}
sub lout_startup {
local ($lbs) = &lout_sanitise($user_brieftitle);
print LOUT <<END;
\@SysInclude{ fontdefs }
\@SysInclude{ langdefs }
\@SysInclude{ dl }
\@SysInclude{ docf }
\@Use { \@DocumentLayout
\@OddTop { \@Null }
\@EvenTop { \@Null }
\@StartOddTop { \@Null }
\@StartEvenTop { \@Null }
\@OddFoot { { $lbs } \@Centre{ - \@PageNum - } \@Right{ $dprint } }
\@EvenFoot { { $lbs } \@Centre{ - \@PageNum - } \@Right{ $dprint } }
\@StartOddFoot { { $lbs } \@Centre{ - \@PageNum - } \@Right{ $dprint } }
\@StartEvenFoot { { $lbs } \@Centre{ - \@PageNum - } \@Right{ $dprint } }
\@ParaGap { 1.70vx }
\@InitialBreak { 1.0fx ragged hyphen }
}
\@Use { \@OrdinaryLayout }
END
$lout_textstatus= 'p';
}
sub lout_pageref {
print LOUT "Q$_[1] (page {\@PageOf{$_[0]}}) ";
&lout_text("\`");
}
sub lout_endpageref {
&lout_text("'");
}
sub lout_finish {
print LOUT "\@End \@Text\n";
close(L);
}
sub lout_startmajorheading {
$lout_styles .= 'h';
print LOUT <<END
\@CNP
{
newpath 0 ysize 0.3 ft sub moveto
xsize 0 rlineto
0 0.2 ft rlineto
xsize neg 0 rlineto
closepath fill
} \@Graphic { //1.6f \@HAdjust \@Heading{
END
;
$endh= "}\n{\@PageMark s_$_[0]}\n/1.0fo\n";
&lout_text($_[0] ? "Section $_[0]. " : '');
}
sub lout_startminorheading {
$lout_styles .= 'h';
print LOUT "//0.2f \@CNP {\@PageMark $_[0]} \@Heading{\n";
$endh= '';
}
sub lout_endheading {
$lout_styles =~ s/.$//; print LOUT "}\n$endh";
$lout_status= 'p';
}
sub lout_endmajorheading { &lout_endheading(@_); }
sub lout_endminorheading { &lout_endheading(@_); }
sub lout_courier {
$lout_styles .= 'f';
print LOUT "{{0.7 1.0} \@Scale {Courier Bold} \@Font {";
}
sub lout_endcourier {
$lout_styles =~ s/.$//; print LOUT "}}";
}
sub lout_italic { $lout_styles .= 'f'; print LOUT "{Slope \@Font {"; }
sub lout_enditalic { $lout_styles =~ s/.$//; print LOUT "}}"; }
sub lout_startindent { $lout_styles .= 'i'; print LOUT "\@IndentedDisplay {\n"; }
sub lout_endindent {
&lout_endpara;
$lout_styles =~ s/.$//; print LOUT "}\n\@LP\n";
}
sub lout_startpackedlist { $lout_plc=-1; }
sub lout_endpackedlist { &lout_newline if !$lout_plc; }
sub lout_packeditem {
&lout_newline if !$lout_plc;
&lout_tab(($lout_plc>0)*40+5);
$lout_plc= !$lout_plc;
}
sub lout_startlist {
&lout_endpara;
print LOUT "\@RawIndentedList style {\@Bullet} indent {0.5i} gap {1.1vx}\n";
$lout_styles .= 'l';
$lout_status= '';
}
sub lout_endlist {
&lout_endpara;
print LOUT "\@EndList\n\n";
$lout_styles =~ s/.$//;
}
sub lout_item {
&lout_endpara;
print LOUT "\@ListItem{";
$lout_styles.= 'I';
}
sub lout_startindex {
print LOUT "//0.0fe\n";
}
sub lout_endindex {
$lout_status='p';
}
sub lout_startindexmainitem {
$lout_marker= $_[0];
$lout_status= '';
print LOUT "//0.3vx Bold \@Font \@HAdjust { \@HContract { { $_[1] } |3cx {";
$lout_iiendheight= '1.00';
$lout_styles .= 'X';
}
sub lout_startindexitem {
$lout_marker= $_[0];
print LOUT "\@HAdjust { \@HContract { { $_[1] } |3cx {";
$lout_iiendheight= '0.95';
$lout_styles .= 'X';
}
sub lout_endindexitem {
print LOUT "} } |0c \@PageOf { $lout_marker } } //${lout_iiendheight}vx\n";
$lout_styles =~ s/.$//;
}
sub lout_email { &lout_courier; &lout_text('<'); }
sub lout_endemail { &lout_text('>'); &lout_endcourier; }
sub lout_ftpon { &lout_courier; } sub lout_endftpon { &lout_endcourier; }
sub lout_ftpin { &lout_courier; } sub lout_endftpin { &lout_endcourier; }
sub lout_docref { } sub lout_enddocref { }
sub lout_ftpsilent { $lout_ignore++; }
sub lout_endftpsilent { $lout_ignore--; }
sub lout_newsgroup { &lout_courier; }
sub lout_endnewsgroup { &lout_endcourier; }
sub lout_text {
return if $lout_ignore;
$lout_status= 'p';
$_= &lout_sanitise($_[0]);
s/ $/\n/ unless $lout_styles =~ m/[fhX]/;
print LOUT $_;
}
sub lout_tab {
local ($size) = $_[0]*0.5;
print LOUT " |${size}ft ";
}
sub lout_newline {
print LOUT " //1.0vx\n";
}
sub lout_sanitise {
local ($in) = @_;
local ($out);
$in= ' '.$in.' ';
$out='';
while ($in =~ m/(\s)(\S*[\@\/|\\\"\^\&\{\}\#]\S*)(\s)/) {
$out .= $`.$1;
$in = $3.$';
$_= $2;
s/[\\\"]/\\$&/g;
$out .= '"'.$_.'"';
}
$out .= $in;
$out =~ s/^ //; $out =~ s/ $//;
$out;
}
sub lout_endpara {
return if $lout_status eq '';
if ($lout_styles eq '') {
print LOUT "\@LP\n\n";
} elsif ($lout_styles =~ s/I$//) {
print LOUT "}\n";
}
$lout_status= '';
}
sub lout_startverbatim {
print LOUT "//0.4f\n\@RawIndentedDisplay lines \@Break".
" { {0.7 1.0} \@Scale {Courier Bold} \@Font {\n";
}
sub lout_verbatim {
$_= $_[0];
s/^\s*//;
print LOUT &lout_sanitise($_),"\n";
}
sub lout_endverbatim { print LOUT "}\n}\n//0.4f\n"; }
1;

View File

@@ -0,0 +1,189 @@
## POST output
# Copyright (C) 1993-1995 Ian Jackson.
# This file is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation; either version 2, or (at your option)
# any later version.
# It is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
# You should have received a copy of the GNU General Public License
# along with GNU Emacs; see the file COPYING. If not, write to
# the Free Software Foundation, Inc., 59 Temple Place - Suite 330,
# Boston, MA 02111-1307, USA.
# (Note: I do not consider works produced using these BFNN processing
# tools to be derivative works of the tools, so they are NOT covered
# by the GPL. However, I would appreciate it if you credited me if
# appropriate in any documents you format using BFNN.)
sub post_init {
open(POST,">$prefix.post");
}
sub post_startmajorheading {
print POST '='x79,"\n\n";
$post_status= 'h';
&post_text($_[0] ? "Section $_[0]. " : '');
}
sub post_startminorheading {
print POST '-'x77,"\n\n";
$post_status= 'h';
}
sub post_italic { &post_text('*'); }
sub post_enditalic { $post_para .= '*'; }
sub post_email { &post_text('<'); } sub post_endemail { &post_text('>'); }
sub post_ftpon { } sub post_endftpon { }
sub post_ftpin { } sub post_endftpin { }
sub post_docref { } sub post_enddocref { }
sub post_courier { } sub post_endcourier { }
sub post_newsgroup { } sub post_endnewsgroup { }
sub post_ftpsilent { $post_ignore++; }
sub post_endftpsilent { $post_ignore--; }
sub post_text {
return if $post_ignore;
if ($post_status eq '') {
$post_status= 'p';
}
$post_para .= $_[0];
}
sub post_tab {
local ($n) = $_[0]-length($post_para);
$post_para .= ' 'x$n if $n>0;
}
sub post_newline {
return unless $post_status eq 'p';
&post_writepara;
}
sub post_writepara {
local ($thisline, $thisword, $rest);
for (;;) {
last unless $post_para =~ m/\S/;
$thisline= $post_indentstring;
for (;;) {
last unless $post_para =~ m/^(\s*\S+)/;
unless (length($1) + length($thisline) < 75 ||
length($thisline) == length($post_indentstring)) {
last;
}
$thisline .= $1;
$post_para= $';
}
$post_para =~ s/^\s*//;
print POST $thisline,"\n";
$post_indentstring= $post_nextindent;
last unless length($post_para);
}
$post_status= ''; $post_para= '';
}
sub post_endpara {
return unless $post_status eq 'p';
&post_writepara;
print POST "\n";
}
sub post_endheading {
$post_para =~ s/\s*$//;
print POST "$post_para\n\n";
$post_status= '';
$post_para= '';
}
sub post_endmajorheading { &post_endheading(@_); }
sub post_endminorheading { &post_endheading(@_); }
sub post_startverbatim {
$post_vstatus= $post_status;
&post_writepara;
}
sub post_verbatim {
print POST $_[0],"\n";
}
sub post_endverbatim {
$post_status= $post_vstatus;
}
sub post_finish {
close(POST);
}
sub post_startindex { $post_status= ''; }
sub post_endindex { $post_status= 'p'; }
sub post_endindexitem {
printf POST " %-11s %-.66s\n",$post_left,$post_para;
$post_status= 'p';
$post_para= '';
}
sub post_startindexitem {
$post_left= $_[1];
}
sub post_startindexmainitem {
$post_left= $_[1];
print POST "\n" if $post_status eq 'p';
}
sub post_startindent {
$post_istatus= $post_status;
&post_writepara;
$post_indentstring= " $post_indentstring";
$post_nextindent= " $post_nextindent";
}
sub post_endindent {
$post_indentstring =~ s/^ //;
$post_nextindent =~ s/^ //;
$post_status= $post_istatus;
}
sub post_startpackedlist { $post_plc=0; }
sub post_endpackedlist { &post_newline if !$post_plc; }
sub post_packeditem {
&post_newline if !$post_plc;
&post_tab($post_plc*40+5);
$post_plc= !$post_plc;
}
sub post_startlist {
&post_endpara;
$post_indentstring= " $post_indentstring";
$post_nextindent= " $post_nextindent";
}
sub post_endlist {
&post_endpara;
$post_indentstring =~ s/^ //;
$post_nextindent =~ s/^ //;
}
sub post_item {
&post_newline;
$post_indentstring =~ s/ $/* /;
}
sub post_pageref {
&post_text("Q$_[1] \`");
}
sub post_endpageref {
&post_text("'");
}
1;

View File

@@ -0,0 +1,37 @@
SUBDIRS = FAQ
info_TEXINFOS = fftw3.texi
fftw3_TEXINFOS = acknowledgements.texi cindex.texi fftw3.texi findex.texi install.texi intro.texi legacy-fortran.texi license.texi modern-fortran.texi mpi.texi other.texi reference.texi threads.texi tutorial.texi upgrading.texi version.texi rfftwnd.pdf rfftwnd.eps
DVIPS = dvips -Pwww
EQN_IMAGES = equation-dft.png equation-dht.png equation-idft.png \
equation-redft00.png equation-redft01.png equation-redft10.png \
equation-redft11.png equation-rodft00.png equation-rodft01.png \
equation-rodft10.png equation-rodft11.png
EXTRA_DIST = f77_wisdom.f fftw3.pdf html rfftwnd.fig rfftwnd.eps \
rfftwnd.pdf rfftwnd-for-html.png $(EQN_IMAGES)
html: $(fftw3_TEXINFOS) $(EQN_IMAGES) rfftwnd-for-html.png
$(MAKEINFO) $(AM_MAKEINFOFLAGS) $(MAKEINFOFLAGS) -I $(srcdir) \
--html --number-sections -o html fftw3.texi
for i in $(EQN_IMAGES); do cp -f ${srcdir}/$$i html; done
cp -f ${srcdir}/rfftwnd-for-html.png html
maintainer-clean-local:
rm -rf html
if MAINTAINER_MODE
# generate the figure for the manual and distribute the binaries, so that
# people don't need to have fig2dev installed.
rfftwnd.eps: rfftwnd.fig
fig2dev -L eps -m .7 ${srcdir}/rfftwnd.fig rfftwnd.eps
rfftwnd-for-html.png: rfftwnd.fig
fig2dev -L png -m 1 ${srcdir}/rfftwnd.fig rfftwnd-for-html.png
rfftwnd.pdf: rfftwnd.fig
fig2dev -L pdf -m .7 ${srcdir}/rfftwnd.fig rfftwnd.pdf
endif

1028
fftw-3.3.10/doc/Makefile.in Normal file

File diff suppressed because it is too large Load Diff

View File

@@ -0,0 +1,89 @@
@node Acknowledgments, License and Copyright, Installation and Customization, Top
@chapter Acknowledgments
Matteo Frigo was supported in part by the Special Research Program SFB
F011 ``AURORA'' of the Austrian Science Fund FWF and by MIT Lincoln
Laboratory. For previous versions of FFTW, he was supported in part by the
Defense Advanced Research Projects Agency (DARPA), under Grants
N00014-94-1-0985 and F30602-97-1-0270, and by a Digital Equipment
Corporation Fellowship.
Steven G. Johnson was supported in part by a Dept.@ of Defense NDSEG
Fellowship, an MIT Karl Taylor Compton Fellowship, and by the Materials
Research Science and Engineering Center program of the National Science
Foundation under award DMR-9400334.
Code for the Cell Broadband Engine was graciously donated to the FFTW
project by the IBM Austin Research Lab and included in fftw-3.2. (This
code was removed in fftw-3.3.)
Code for the MIPS paired-single SIMD support was graciously donated to
the FFTW project by CodeSourcery, Inc.
We are grateful to Sun Microsystems Inc.@ for its donation of a
cluster of 9 8-processor Ultra HPC 5000 SMPs (24 Gflops peak). These
machines served as the primary platform for the development of early
versions of FFTW.
We thank Intel Corporation for donating a four-processor Pentium Pro
machine. We thank the GNU/Linux community for giving us a decent OS to
run on that machine.
We are thankful to the AMD corporation for donating an AMD Athlon XP 1700+
computer to the FFTW project.
We thank the Compaq/HP testdrive program and VA Software Corporation
(SourceForge.net) for providing remote access to machines that were used
to test FFTW.
The @code{genfft} suite of code generators was written using Objective
Caml, a dialect of ML. Objective Caml is a small and elegant language
developed by Xavier Leroy. The implementation is available from
@uref{http://caml.inria.fr/, @code{http://caml.inria.fr/}}. In previous
releases of FFTW, @code{genfft} was written in Caml Light, by the same
authors. An even earlier implementation of @code{genfft} was written in
Scheme, but Caml is definitely better for this kind of application.
@cindex Caml
@cindex LISP
FFTW uses many tools from the GNU project, including @code{automake},
@code{texinfo}, and @code{libtool}.
Prof.@ Charles E.@ Leiserson of MIT provided continuous support and
encouragement. This program would not exist without him. Charles also
proposed the name ``codelets'' for the basic FFT blocks.
@cindex codelet
Prof.@ John D.@ Joannopoulos of MIT demonstrated continuing tolerance of
Steven's ``extra-curricular'' computer-science activities, as well as
remarkable creativity in working them into his grant proposals.
Steven's physics degree would not exist without him.
Franz Franchetti wrote SIMD extensions to FFTW 2, which eventually
led to the SIMD support in FFTW 3.
Stefan Kral wrote most of the K7 code generator distributed with FFTW
3.0.x and 3.1.x.
Andrew Sterian contributed the Windows timing code in FFTW 2.
Didier Miras reported a bug in the test procedure used in FFTW 1.2. We
now use a completely different test algorithm by Funda Ergun that does
not require a separate FFT program to compare against.
Wolfgang Reimer contributed the Pentium cycle counter and a few fixes
that help portability.
Ming-Chang Liu uncovered a well-hidden bug in the complex transforms of
FFTW 2.0 and supplied a patch to correct it.
The FFTW FAQ was written in @code{bfnn} (Bizarre Format With No Name)
and formatted using the tools developed by Ian Jackson for the Linux
FAQ.
@emph{We are especially thankful to all of our users for their
continuing support, feedback, and interest during our development of
FFTW.}

View File

@@ -0,0 +1,3 @@
@node Concept Index, Library Index, License and Copyright, Top
@chapter Concept Index
@printindex cp

Binary file not shown.

After

Width:  |  Height:  |  Size: 1.1 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 1.7 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 1.1 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 1.8 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 1.5 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 1.4 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 1.5 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 1.6 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 1.9 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 1.5 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 1.6 KiB

View File

@@ -0,0 +1,79 @@
c Copyright (c) 2003, 2007-14 Matteo Frigo
c Copyright (c) 2003, 2007-14 Massachusetts Institute of Technology
c
c This program is free software; you can redistribute it and/or modify
c it under the terms of the GNU General Public License as published by
c the Free Software Foundation; either version 2 of the License, or
c (at your option) any later version.
c
c This program is distributed in the hope that it will be useful,
c but WITHOUT ANY WARRANTY; without even the implied warranty of
c MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
c GNU General Public License for more details.
c
c You should have received a copy of the GNU General Public License
c along with this program; if not, write to the Free Software
c Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
c
cccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccc
c
c This is an example implementation of Fortran wisdom export/import
c to/from a Fortran unit (file), exploiting the generic
c dfftw_export_wisdom/dfftw_import_wisdom functions.
c
c We cannot compile this file into the FFTW library itself, lest all
c FFTW-calling programs be required to link to the Fortran I/O
c libraries.
c
cccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccc
c Strictly speaking, the '$' format specifier, which allows us to
c write a character without a trailing newline, is not standard F77.
c However, it seems to be a nearly universal extension.
subroutine write_char(c, iunit)
character c
integer iunit
write(iunit,321) c
321 format(a,$)
end
subroutine export_wisdom_to_file(iunit)
integer iunit
external write_char
call dfftw_export_wisdom(write_char, iunit)
end
c Fortran 77 does not have any portable way to read an arbitrary
c file one character at a time. The best alternative seems to be to
c read a whole line into a buffer, since for fftw-exported wisdom we
c can bound the line length. (If the file contains longer lines,
c then the lines will be truncated and the wisdom import should
c simply fail.) Ugh.
subroutine read_char(ic, iunit)
integer ic
integer iunit
character*256 buf
save buf
integer ibuf
data ibuf/257/
save ibuf
if (ibuf .lt. 257) then
ic = ichar(buf(ibuf:ibuf))
ibuf = ibuf + 1
return
endif
read(iunit,123,end=666) buf
ic = ichar(buf(1:1))
ibuf = 2
return
666 ic = -1
ibuf = 257
123 format(a256)
end
subroutine import_wisdom_from_file(isuccess, iunit)
integer isuccess
integer iunit
external read_char
call dfftw_import_wisdom(isuccess, read_char, iunit)
end

169
fftw-3.3.10/doc/fftw3.info Normal file
View File

@@ -0,0 +1,169 @@
This is fftw3.info, produced by makeinfo version 6.7 from fftw3.texi.
This manual is for FFTW (version 3.3.10, 10 December 2020).
Copyright (C) 2003 Matteo Frigo.
Copyright (C) 2003 Massachusetts Institute of Technology.
Permission is granted to make and distribute verbatim copies of
this manual provided the copyright notice and this permission
notice are preserved on all copies.
Permission is granted to copy and distribute modified versions of
this manual under the conditions for verbatim copying, provided
that the entire resulting derived work is distributed under the
terms of a permission notice identical to this one.
Permission is granted to copy and distribute translations of this
manual into another language, under the above conditions for
modified versions, except that this permission notice may be stated
in a translation approved by the Free Software Foundation.
INFO-DIR-SECTION Development
START-INFO-DIR-ENTRY
* fftw3: (fftw3). FFTW User's Manual.
END-INFO-DIR-ENTRY

Indirect:
fftw3.info-1: 1058
fftw3.info-2: 301112

Tag Table:
(Indirect)
Node: Top1058
Node: Introduction1732
Node: Tutorial8032
Ref: Tutorial-Footnote-19274
Node: Complex One-Dimensional DFTs9368
Node: Complex Multi-Dimensional DFTs15102
Ref: Complex Multi-Dimensional DFTs-Footnote-118528
Node: One-Dimensional DFTs of Real Data18663
Node: Multi-Dimensional DFTs of Real Data23103
Node: More DFTs of Real Data27029
Node: The Halfcomplex-format DFT30526
Node: Real even/odd DFTs (cosine/sine transforms)33135
Ref: Real even/odd DFTs (cosine/sine transforms)-Footnote-138724
Ref: Real even/odd DFTs (cosine/sine transforms)-Footnote-238913
Node: The Discrete Hartley Transform39847
Ref: The Discrete Hartley Transform-Footnote-142029
Node: Other Important Topics42279
Node: SIMD alignment and fftw_malloc42572
Node: Multi-dimensional Array Format44761
Node: Row-major Format45381
Node: Column-major Format47077
Node: Fixed-size Arrays in C48158
Node: Dynamic Arrays in C49594
Node: Dynamic Arrays in C-The Wrong Way51228
Node: Words of Wisdom-Saving Plans52976
Node: Caveats in Using Wisdom55644
Node: FFTW Reference57727
Node: Data Types and Files58215
Node: Complex numbers58647
Node: Precision60385
Node: Memory Allocation61945
Node: Using Plans63510
Node: Basic Interface67535
Ref: Basic Interface-Footnote-168275
Node: Complex DFTs68339
Node: Planner Flags72304
Node: Real-data DFTs77740
Node: Real-data DFT Array Format82729
Node: Real-to-Real Transforms84985
Node: Real-to-Real Transform Kinds88948
Node: Advanced Interface91413
Node: Advanced Complex DFTs92153
Node: Advanced Real-data DFTs96421
Node: Advanced Real-to-real Transforms98749
Node: Guru Interface99854
Node: Interleaved and split arrays100778
Node: Guru vector and transform sizes101817
Node: Guru Complex DFTs104514
Node: Guru Real-data DFTs107350
Node: Guru Real-to-real Transforms110269
Node: 64-bit Guru Interface111588
Node: New-array Execute Functions113903
Node: Wisdom118399
Node: Wisdom Export118758
Node: Wisdom Import120736
Node: Forgetting Wisdom122763
Node: Wisdom Utilities123136
Node: What FFTW Really Computes124498
Node: The 1d Discrete Fourier Transform (DFT)125323
Node: The 1d Real-data DFT126681
Node: 1d Real-even DFTs (DCTs)128345
Node: 1d Real-odd DFTs (DSTs)131546
Node: 1d Discrete Hartley Transforms (DHTs)134482
Node: Multi-dimensional Transforms135158
Node: Multi-threaded FFTW137762
Node: Installation and Supported Hardware/Software139228
Node: Usage of Multi-threaded FFTW141052
Node: How Many Threads to Use?145843
Node: Thread safety146865
Node: Distributed-memory FFTW with MPI150165
Node: FFTW MPI Installation152740
Node: Linking and Initializing MPI FFTW154527
Node: 2d MPI example155750
Node: MPI Data Distribution159975
Node: Basic and advanced distribution interfaces162849
Node: Load balancing167270
Node: Transposed distributions168955
Node: One-dimensional distributions172722
Node: Multi-dimensional MPI DFTs of Real Data175287
Node: Other Multi-dimensional Real-data MPI Transforms179928
Node: FFTW MPI Transposes182101
Node: Basic distributed-transpose interface182941
Node: Advanced distributed-transpose interface185114
Node: An improved replacement for MPI_Alltoall186398
Node: FFTW MPI Wisdom188367
Ref: FFTW MPI Wisdom-Footnote-1191105
Node: Avoiding MPI Deadlocks192019
Node: FFTW MPI Performance Tips193044
Node: Combining MPI and Threads194509
Node: FFTW MPI Reference197972
Node: MPI Files and Data Types198551
Node: MPI Initialization199547
Node: Using MPI Plans200646
Node: MPI Data Distribution Functions202472
Node: MPI Plan Creation207929
Node: MPI Wisdom Communication218605
Node: FFTW MPI Fortran Interface219531
Ref: FFTW MPI Fortran Interface-Footnote-1225554
Node: Calling FFTW from Modern Fortran225962
Node: Overview of Fortran interface227312
Node: Extended and quadruple precision in Fortran230767
Node: Reversing array dimensions232147
Node: FFTW Fortran type reference235678
Node: Plan execution in Fortran240172
Node: Allocating aligned memory in Fortran243055
Node: Accessing the wisdom API from Fortran246417
Node: Wisdom File Export/Import from Fortran247194
Node: Wisdom String Export/Import from Fortran248856
Node: Wisdom Generic Export/Import from Fortran250841
Node: Defining an FFTW module253071
Node: Calling FFTW from Legacy Fortran254142
Node: Fortran-interface routines255699
Ref: Fortran-interface routines-Footnote-1259354
Ref: Fortran-interface routines-Footnote-2259557
Node: FFTW Constants in Fortran259690
Node: FFTW Execution in Fortran260844
Node: Fortran Examples263590
Node: Wisdom of Fortran?267163
Node: Upgrading from FFTW version 2268842
Ref: Upgrading from FFTW version 2-Footnote-1278565
Node: Installation and Customization278748
Node: Installation on Unix280389
Node: Installation on non-Unix systems288466
Node: Cycle Counters290683
Node: Generating your own code292434
Node: Acknowledgments294468
Node: License and Copyright298184
Node: Concept Index301112
Node: Library Index338972

End Tag Table

Local Variables:
coding: utf-8
End:

6304
fftw-3.3.10/doc/fftw3.info-1 Normal file

File diff suppressed because it is too large Load Diff

Binary file not shown.

BIN
fftw-3.3.10/doc/fftw3.pdf Normal file

Binary file not shown.

279
fftw-3.3.10/doc/fftw3.texi Normal file
View File

@@ -0,0 +1,279 @@
\input texinfo @c -*-texinfo-*-
@c Update by C-x C-e on: (texinfo-multiple-files-update "fftw3.texi" nil t)
@setfilename fftw3.info
@include version.texi
@finalout
@settitle FFTW @value{VERSION}
@setchapternewpage odd
@c define constant index (ct)
@defcodeindex ct
@syncodeindex ct fn
@syncodeindex vr fn
@syncodeindex pg fn
@syncodeindex tp fn
@c define foreign function index (ff)
@defcodeindex ff
@syncodeindex ff cp
@c define foreign constant index (fc)
@defcodeindex fc
@syncodeindex fc cp
@c define foreign program index (fp)
@defcodeindex fp
@syncodeindex fp cp
@comment %**end of header
@iftex
@paragraphindent 0
@parskip=@medskipamount
@end iftex
@c
@c The following macros are coded in a weird way:
@c @macro FOO
@c @noindent
@c <STUFF>
@c @refill
@c @end macro
@c The @noindent/@refill stuff is not necessary in texinfo up to version
@c 4, but it is a hack necessary to make texinfo-5 work.
@c Texinfo has been stable for the first 15 years of FFTW's history.
@c Then some genius, with too much time in his hands and on a mission to
@c deliver the world from the evil of the C language, decided to rewrite
@c makeinfo in Perl, the old C version of makeinfo being, as I said,
@c evil. The official excuse for the rewrite was that now I can have my
@c manual in XML format, as if XML were a feature.
@c The result of this stroke of genius is that texinfo-5 has different
@c rules for macro expansion than texinfo-4 does, specifically regarding
@c whether or not spaces after a macro are ignored. Texinfo-4 had weird
@c rules, but at least they were constant and internally more or less
@c consistent. Texinfo-5 has different rules, and even worse the rules
@c in texinfo-5 are inconsistent between the TeX and HTML output
@c processors. This situation makes it almost impossible for us to
@c produce a manual that works with both texinfo 4 and 5 in all modes
@c (TeX, info, and html). The @noindent/@refill hack is my best shot at
@c patching this situation.
@c "@noindent" has two effects: First, it makes texinfo-5 believe that
@c the next "@ifinfo" is on a new line, otherwise texinfo-5 complains
@c that it is not (even though it obviously is). Second, "@noindent" is
@c a macro that eats extra space, and we want this effect because somehow
@c macro expansion in texinfo-5 inserts extra spaces that were not there
@c in texinfo-4.
@c "@refill" stops texinfo-5 from interpreting the rest of the line after
@c a macro invocation as an argument to "@end tex". For example, in
@c "FFTW uses @Onlogn algorithms", somehow texinfo-5 thinks that
@c "algorithms" is an argument to "@end tex". "@noindent" would have the
@c same effect (as would any other macro invocation, I think), but,
@c unlike "@noindent", "@refill" does not eat spaces and does not scan
@c the rest of the input file for macro arguments. However, "@refill" is
@c deemed "obsolete" in the texinfo-5 source code, so expect this to
@c break at some point.
@c This situation is wholly unsatisfactory, and the GNU project is
@c obviously out of control. If this nonsense persists, we will abandon
@c texinfo and produce a latex-only version of the manual.
@macro Onlogn
@noindent
@ifinfo
O(n log n)
@end ifinfo
@html
<i>O</i>(<i>n</i>&nbsp;log&nbsp;<i>n</i>)
@end html
@tex
$O(n \\log n)$
@end tex
@refill
@end macro
@macro ndims
@noindent
@ifinfo
n[0] x n[1] x n[2] x ... x n[d-1]
@end ifinfo
@html
n<sub>0</sub>&nbsp;&times;&nbsp;n<sub>1</sub>&nbsp;&times;&nbsp;n<sub>2</sub>&nbsp;&times;&nbsp;&hellip;&nbsp;&times;&nbsp;n<sub>d-1</sub>
@end html
@tex
$n_0 \\times n_1 \\times n_2 \\times \\cdots \\times n_{d-1}$
@end tex
@refill
@end macro
@macro ndimshalf
@noindent
@ifinfo
n[0] x n[1] x n[2] x ... x (n[d-1]/2 + 1)
@end ifinfo
@html
n<sub>0</sub>&nbsp;&times;&nbsp;n<sub>1</sub>&nbsp;&times;&nbsp;n<sub>2</sub>&nbsp;&times;&nbsp;&hellip;&nbsp;&times;&nbsp;(n<sub>d-1</sub>/2 + 1)
@end html
@tex
$n_0 \\times n_1 \\times n_2 \\times \\cdots \\times (n_{d-1}/2 + 1)$
@end tex
@refill
@end macro
@macro ndimspad
@noindent
@ifinfo
n[0] x n[1] x n[2] x ... x [2 (n[d-1]/2 + 1)]
@end ifinfo
@html
n<sub>0</sub>&nbsp;&times;&nbsp;n<sub>1</sub>&nbsp;&times;&nbsp;n<sub>2</sub>&nbsp;&times;&nbsp;&hellip;&nbsp;&times;&nbsp;[2&nbsp;(n<sub>d-1</sub>/2 + 1)]
@end html
@tex
$n_0 \\times n_1 \\times n_2 \\times \\cdots \\times [2(n_{d-1}/2 + 1)]$
@end tex
@refill
@end macro
@macro twodims{d1, d2}
@noindent
@ifinfo
\d1\ x \d2\
@end ifinfo
@html
\d1\&nbsp;&times;&nbsp;\d2\
@end html
@tex
$\d1\ \\times \d2\$
@end tex
@refill
@end macro
@macro threedims{d1, d2, d3}
@noindent
@ifinfo
\d1\ x \d2\ x \d3\
@end ifinfo
@html
\d1\&nbsp;&times;&nbsp;\d2\&nbsp;&times;&nbsp;\d3\
@end html
@tex
$\d1\ \\times \d2\ \\times \d3\$
@end tex
@refill
@end macro
@macro dimk{k}
@noindent
@ifinfo
n[\k\]
@end ifinfo
@html
n<sub>\k\</sub>
@end html
@tex
$n_\k\$
@end tex
@refill
@end macro
@macro ndimstrans
@noindent
@ifinfo
n[1] x n[0] x n[2] x ... x n[d-1]
@end ifinfo
@html
n<sub>1</sub>&nbsp;&times;&nbsp;n<sub>0</sub>&nbsp;&times;&nbsp;n<sub>2</sub>&nbsp;&times;&hellip;&times;&nbsp;n<sub>d-1</sub>
@end html
@tex
$n_1 \\times n_0 \\times n_2 \\times \\cdots \\times n_{d-1}$
@end tex
@refill
@end macro
@copying
This manual is for FFTW
(version @value{VERSION}, @value{UPDATED}).
Copyright @copyright{} 2003 Matteo Frigo.
Copyright @copyright{} 2003 Massachusetts Institute of Technology.
@quotation
Permission is granted to make and distribute verbatim copies of this
manual provided the copyright notice and this permission notice are
preserved on all copies.
Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided that the
entire resulting derived work is distributed under the terms of a
permission notice identical to this one.
Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions,
except that this permission notice may be stated in a translation
approved by the Free Software Foundation.
@end quotation
@end copying
@dircategory Development
@direntry
* fftw3: (fftw3). FFTW User's Manual.
@end direntry
@titlepage
@title FFTW
@subtitle for version @value{VERSION}, @value{UPDATED}
@author Matteo Frigo
@author Steven G. Johnson
@page
@vskip 0pt plus 1filll
@insertcopying
@end titlepage
@contents
@ifnottex
@node Top, Introduction, (dir), (dir)
@top FFTW User Manual
Welcome to FFTW, the Fastest Fourier Transform in the West. FFTW is a
collection of fast C routines to compute the discrete Fourier transform.
This manual documents FFTW version @value{VERSION}.
@end ifnottex
@menu
* Introduction::
* Tutorial::
* Other Important Topics::
* FFTW Reference::
* Multi-threaded FFTW::
* Distributed-memory FFTW with MPI::
* Calling FFTW from Modern Fortran::
* Calling FFTW from Legacy Fortran::
* Upgrading from FFTW version 2::
* Installation and Customization::
* Acknowledgments::
* License and Copyright::
* Concept Index::
* Library Index::
@end menu
@c ************************************************************
@include intro.texi
@include tutorial.texi
@include other.texi
@include reference.texi
@include threads.texi
@include mpi.texi
@include modern-fortran.texi
@include legacy-fortran.texi
@include upgrading.texi
@include install.texi
@include acknowledgements.texi
@include license.texi
@include cindex.texi
@include findex.texi
@c ************************************************************
@bye

View File

@@ -0,0 +1,3 @@
@node Library Index, , Concept Index, Top
@chapter Library Index
@printindex fn

View File

@@ -0,0 +1,85 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html>
<!-- This manual is for FFTW
(version 3.3.10, 10 December 2020).
Copyright (C) 2003 Matteo Frigo.
Copyright (C) 2003 Massachusetts Institute of Technology.
Permission is granted to make and distribute verbatim copies of this
manual provided the copyright notice and this permission notice are
preserved on all copies.
Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided that the
entire resulting derived work is distributed under the terms of a
permission notice identical to this one.
Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions,
except that this permission notice may be stated in a translation
approved by the Free Software Foundation. -->
<!-- Created by GNU Texinfo 6.7, http://www.gnu.org/software/texinfo/ -->
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<title>1d Discrete Hartley Transforms (DHTs) (FFTW 3.3.10)</title>
<meta name="description" content="1d Discrete Hartley Transforms (DHTs) (FFTW 3.3.10)">
<meta name="keywords" content="1d Discrete Hartley Transforms (DHTs) (FFTW 3.3.10)">
<meta name="resource-type" content="document">
<meta name="distribution" content="global">
<meta name="Generator" content="makeinfo">
<link href="index.html" rel="start" title="Top">
<link href="Concept-Index.html" rel="index" title="Concept Index">
<link href="index.html#SEC_Contents" rel="contents" title="Table of Contents">
<link href="What-FFTW-Really-Computes.html" rel="up" title="What FFTW Really Computes">
<link href="Multi_002ddimensional-Transforms.html" rel="next" title="Multi-dimensional Transforms">
<link href="1d-Real_002dodd-DFTs-_0028DSTs_0029.html" rel="prev" title="1d Real-odd DFTs (DSTs)">
<style type="text/css">
<!--
a.summary-letter {text-decoration: none}
blockquote.indentedblock {margin-right: 0em}
div.display {margin-left: 3.2em}
div.example {margin-left: 3.2em}
div.lisp {margin-left: 3.2em}
kbd {font-style: oblique}
pre.display {font-family: inherit}
pre.format {font-family: inherit}
pre.menu-comment {font-family: serif}
pre.menu-preformatted {font-family: serif}
span.nolinebreak {white-space: nowrap}
span.roman {font-family: initial; font-weight: normal}
span.sansserif {font-family: sans-serif; font-weight: normal}
ul.no-bullet {list-style: none}
-->
</style>
</head>
<body lang="en">
<span id="g_t1d-Discrete-Hartley-Transforms-_0028DHTs_0029"></span><div class="header">
<p>
Next: <a href="Multi_002ddimensional-Transforms.html" accesskey="n" rel="next">Multi-dimensional Transforms</a>, Previous: <a href="1d-Real_002dodd-DFTs-_0028DSTs_0029.html" accesskey="p" rel="prev">1d Real-odd DFTs (DSTs)</a>, Up: <a href="What-FFTW-Really-Computes.html" accesskey="u" rel="up">What FFTW Really Computes</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
<hr>
<span id="g_t1d-Discrete-Hartley-Transforms-_0028DHTs_0029-1"></span><h4 class="subsection">4.8.5 1d Discrete Hartley Transforms (DHTs)</h4>
<span id="index-discrete-Hartley-transform-2"></span>
<span id="index-DHT-1"></span>
<p>The discrete Hartley transform (DHT) of a 1d real array <em>X</em> of size
<em>n</em> computes a real array <em>Y</em> of the same size, where:
<center><img src="equation-dht.png" align="top">.</center>
</p>
<span id="index-normalization-12"></span>
<p>FFTW computes an unnormalized transform, in that there is no coefficient
in front of the summation in the DHT. In other words, applying the
transform twice (the DHT is its own inverse) will multiply the input by
<em>n</em>.
</p>
</body>
</html>

View File

@@ -0,0 +1,155 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html>
<!-- This manual is for FFTW
(version 3.3.10, 10 December 2020).
Copyright (C) 2003 Matteo Frigo.
Copyright (C) 2003 Massachusetts Institute of Technology.
Permission is granted to make and distribute verbatim copies of this
manual provided the copyright notice and this permission notice are
preserved on all copies.
Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided that the
entire resulting derived work is distributed under the terms of a
permission notice identical to this one.
Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions,
except that this permission notice may be stated in a translation
approved by the Free Software Foundation. -->
<!-- Created by GNU Texinfo 6.7, http://www.gnu.org/software/texinfo/ -->
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<title>1d Real-even DFTs (DCTs) (FFTW 3.3.10)</title>
<meta name="description" content="1d Real-even DFTs (DCTs) (FFTW 3.3.10)">
<meta name="keywords" content="1d Real-even DFTs (DCTs) (FFTW 3.3.10)">
<meta name="resource-type" content="document">
<meta name="distribution" content="global">
<meta name="Generator" content="makeinfo">
<link href="index.html" rel="start" title="Top">
<link href="Concept-Index.html" rel="index" title="Concept Index">
<link href="index.html#SEC_Contents" rel="contents" title="Table of Contents">
<link href="What-FFTW-Really-Computes.html" rel="up" title="What FFTW Really Computes">
<link href="1d-Real_002dodd-DFTs-_0028DSTs_0029.html" rel="next" title="1d Real-odd DFTs (DSTs)">
<link href="The-1d-Real_002ddata-DFT.html" rel="prev" title="The 1d Real-data DFT">
<style type="text/css">
<!--
a.summary-letter {text-decoration: none}
blockquote.indentedblock {margin-right: 0em}
div.display {margin-left: 3.2em}
div.example {margin-left: 3.2em}
div.lisp {margin-left: 3.2em}
kbd {font-style: oblique}
pre.display {font-family: inherit}
pre.format {font-family: inherit}
pre.menu-comment {font-family: serif}
pre.menu-preformatted {font-family: serif}
span.nolinebreak {white-space: nowrap}
span.roman {font-family: initial; font-weight: normal}
span.sansserif {font-family: sans-serif; font-weight: normal}
ul.no-bullet {list-style: none}
-->
</style>
</head>
<body lang="en">
<span id="g_t1d-Real_002deven-DFTs-_0028DCTs_0029"></span><div class="header">
<p>
Next: <a href="1d-Real_002dodd-DFTs-_0028DSTs_0029.html" accesskey="n" rel="next">1d Real-odd DFTs (DSTs)</a>, Previous: <a href="The-1d-Real_002ddata-DFT.html" accesskey="p" rel="prev">The 1d Real-data DFT</a>, Up: <a href="What-FFTW-Really-Computes.html" accesskey="u" rel="up">What FFTW Really Computes</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
<hr>
<span id="g_t1d-Real_002deven-DFTs-_0028DCTs_0029-1"></span><h4 class="subsection">4.8.3 1d Real-even DFTs (DCTs)</h4>
<p>The Real-even symmetry DFTs in FFTW are exactly equivalent to the unnormalized
forward (and backward) DFTs as defined above, where the input array
<em>X</em> of length <em>N</em> is purely real and is also <em>even</em> symmetry. In
this case, the output array is likewise real and even symmetry.
<span id="index-real_002deven-DFT-1"></span>
<span id="index-REDFT-1"></span>
</p>
<span id="index-REDFT00"></span>
<p>For the case of <code>REDFT00</code>, this even symmetry means that
<i>X<sub>j</sub> = X<sub>N-j</sub></i>,
where we take <em>X</em> to be periodic so that
<i>X<sub>N</sub> = X</i><sub>0</sub>.
Because of this redundancy, only the first <em>n</em> real numbers are
actually stored, where <em>N = 2(n-1)</em>.
</p>
<p>The proper definition of even symmetry for <code>REDFT10</code>,
<code>REDFT01</code>, and <code>REDFT11</code> transforms is somewhat more intricate
because of the shifts by <em>1/2</em> of the input and/or output, although
the corresponding boundary conditions are given in <a href="Real-even_002fodd-DFTs-_0028cosine_002fsine-transforms_0029.html">Real even/odd DFTs (cosine/sine transforms)</a>. Because of the even symmetry, however,
the sine terms in the DFT all cancel and the remaining cosine terms are
written explicitly below. This formulation often leads people to call
such a transform a <em>discrete cosine transform</em> (DCT), although it is
really just a special case of the DFT.
<span id="index-discrete-cosine-transform-2"></span>
<span id="index-DCT-2"></span>
</p>
<p>In each of the definitions below, we transform a real array <em>X</em> of
length <em>n</em> to a real array <em>Y</em> of length <em>n</em>:
</p>
<span id="REDFT00-_0028DCT_002dI_0029"></span><h4 class="subsubheading">REDFT00 (DCT-I)</h4>
<span id="index-REDFT00-1"></span>
<p>An <code>REDFT00</code> transform (type-I DCT) in FFTW is defined by:
<center><img src="equation-redft00.png" align="top">.</center>
Note that this transform is not defined for <em>n=1</em>. For <em>n=2</em>,
the summation term above is dropped as you might expect.
</p>
<span id="REDFT10-_0028DCT_002dII_0029"></span><h4 class="subsubheading">REDFT10 (DCT-II)</h4>
<span id="index-REDFT10"></span>
<p>An <code>REDFT10</code> transform (type-II DCT, sometimes called &ldquo;the&rdquo; DCT) in FFTW is defined by:
<center><img src="equation-redft10.png" align="top">.</center>
</p>
<span id="REDFT01-_0028DCT_002dIII_0029"></span><h4 class="subsubheading">REDFT01 (DCT-III)</h4>
<span id="index-REDFT01"></span>
<p>An <code>REDFT01</code> transform (type-III DCT) in FFTW is defined by:
<center><img src="equation-redft01.png" align="top">.</center>
In the case of <em>n=1</em>, this reduces to
<i>Y</i><sub>0</sub> = <i>X</i><sub>0</sub>.
Up to a scale factor (see below), this is the inverse of <code>REDFT10</code> (&ldquo;the&rdquo; DCT), and so the <code>REDFT01</code> (DCT-III) is sometimes called the &ldquo;IDCT&rdquo;.
<span id="index-IDCT-3"></span>
</p>
<span id="REDFT11-_0028DCT_002dIV_0029"></span><h4 class="subsubheading">REDFT11 (DCT-IV)</h4>
<span id="index-REDFT11"></span>
<p>An <code>REDFT11</code> transform (type-IV DCT) in FFTW is defined by:
<center><img src="equation-redft11.png" align="top">.</center>
</p>
<span id="Inverses-and-Normalization"></span><h4 class="subsubheading">Inverses and Normalization</h4>
<p>These definitions correspond directly to the unnormalized DFTs used
elsewhere in FFTW (hence the factors of <em>2</em> in front of the
summations). The unnormalized inverse of <code>REDFT00</code> is
<code>REDFT00</code>, of <code>REDFT10</code> is <code>REDFT01</code> and vice versa, and
of <code>REDFT11</code> is <code>REDFT11</code>. Each unnormalized inverse results
in the original array multiplied by <em>N</em>, where <em>N</em> is the
<em>logical</em> DFT size. For <code>REDFT00</code>, <em>N=2(n-1)</em> (note that
<em>n=1</em> is not defined); otherwise, <em>N=2n</em>.
<span id="index-normalization-10"></span>
</p>
<p>In defining the discrete cosine transform, some authors also include
additional factors of
&radic;2
(or its inverse) multiplying selected inputs and/or outputs. This is a
mostly cosmetic change that makes the transform orthogonal, but
sacrifices the direct equivalence to a symmetric DFT.
</p>
<hr>
<div class="header">
<p>
Next: <a href="1d-Real_002dodd-DFTs-_0028DSTs_0029.html" accesskey="n" rel="next">1d Real-odd DFTs (DSTs)</a>, Previous: <a href="The-1d-Real_002ddata-DFT.html" accesskey="p" rel="prev">The 1d Real-data DFT</a>, Up: <a href="What-FFTW-Really-Computes.html" accesskey="u" rel="up">What FFTW Really Computes</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
</body>
</html>

View File

@@ -0,0 +1,152 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html>
<!-- This manual is for FFTW
(version 3.3.10, 10 December 2020).
Copyright (C) 2003 Matteo Frigo.
Copyright (C) 2003 Massachusetts Institute of Technology.
Permission is granted to make and distribute verbatim copies of this
manual provided the copyright notice and this permission notice are
preserved on all copies.
Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided that the
entire resulting derived work is distributed under the terms of a
permission notice identical to this one.
Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions,
except that this permission notice may be stated in a translation
approved by the Free Software Foundation. -->
<!-- Created by GNU Texinfo 6.7, http://www.gnu.org/software/texinfo/ -->
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<title>1d Real-odd DFTs (DSTs) (FFTW 3.3.10)</title>
<meta name="description" content="1d Real-odd DFTs (DSTs) (FFTW 3.3.10)">
<meta name="keywords" content="1d Real-odd DFTs (DSTs) (FFTW 3.3.10)">
<meta name="resource-type" content="document">
<meta name="distribution" content="global">
<meta name="Generator" content="makeinfo">
<link href="index.html" rel="start" title="Top">
<link href="Concept-Index.html" rel="index" title="Concept Index">
<link href="index.html#SEC_Contents" rel="contents" title="Table of Contents">
<link href="What-FFTW-Really-Computes.html" rel="up" title="What FFTW Really Computes">
<link href="1d-Discrete-Hartley-Transforms-_0028DHTs_0029.html" rel="next" title="1d Discrete Hartley Transforms (DHTs)">
<link href="1d-Real_002deven-DFTs-_0028DCTs_0029.html" rel="prev" title="1d Real-even DFTs (DCTs)">
<style type="text/css">
<!--
a.summary-letter {text-decoration: none}
blockquote.indentedblock {margin-right: 0em}
div.display {margin-left: 3.2em}
div.example {margin-left: 3.2em}
div.lisp {margin-left: 3.2em}
kbd {font-style: oblique}
pre.display {font-family: inherit}
pre.format {font-family: inherit}
pre.menu-comment {font-family: serif}
pre.menu-preformatted {font-family: serif}
span.nolinebreak {white-space: nowrap}
span.roman {font-family: initial; font-weight: normal}
span.sansserif {font-family: sans-serif; font-weight: normal}
ul.no-bullet {list-style: none}
-->
</style>
</head>
<body lang="en">
<span id="g_t1d-Real_002dodd-DFTs-_0028DSTs_0029"></span><div class="header">
<p>
Next: <a href="1d-Discrete-Hartley-Transforms-_0028DHTs_0029.html" accesskey="n" rel="next">1d Discrete Hartley Transforms (DHTs)</a>, Previous: <a href="1d-Real_002deven-DFTs-_0028DCTs_0029.html" accesskey="p" rel="prev">1d Real-even DFTs (DCTs)</a>, Up: <a href="What-FFTW-Really-Computes.html" accesskey="u" rel="up">What FFTW Really Computes</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
<hr>
<span id="g_t1d-Real_002dodd-DFTs-_0028DSTs_0029-1"></span><h4 class="subsection">4.8.4 1d Real-odd DFTs (DSTs)</h4>
<p>The Real-odd symmetry DFTs in FFTW are exactly equivalent to the unnormalized
forward (and backward) DFTs as defined above, where the input array
<em>X</em> of length <em>N</em> is purely real and is also <em>odd</em> symmetry. In
this case, the output is odd symmetry and purely imaginary.
<span id="index-real_002dodd-DFT-1"></span>
<span id="index-RODFT-1"></span>
</p>
<span id="index-RODFT00"></span>
<p>For the case of <code>RODFT00</code>, this odd symmetry means that
<i>X<sub>j</sub> = -X<sub>N-j</sub></i>,
where we take <em>X</em> to be periodic so that
<i>X<sub>N</sub> = X</i><sub>0</sub>.
Because of this redundancy, only the first <em>n</em> real numbers
starting at <em>j=1</em> are actually stored (the <em>j=0</em> element is
zero), where <em>N = 2(n+1)</em>.
</p>
<p>The proper definition of odd symmetry for <code>RODFT10</code>,
<code>RODFT01</code>, and <code>RODFT11</code> transforms is somewhat more intricate
because of the shifts by <em>1/2</em> of the input and/or output, although
the corresponding boundary conditions are given in <a href="Real-even_002fodd-DFTs-_0028cosine_002fsine-transforms_0029.html">Real even/odd DFTs (cosine/sine transforms)</a>. Because of the odd symmetry, however,
the cosine terms in the DFT all cancel and the remaining sine terms are
written explicitly below. This formulation often leads people to call
such a transform a <em>discrete sine transform</em> (DST), although it is
really just a special case of the DFT.
<span id="index-discrete-sine-transform-2"></span>
<span id="index-DST-2"></span>
</p>
<p>In each of the definitions below, we transform a real array <em>X</em> of
length <em>n</em> to a real array <em>Y</em> of length <em>n</em>:
</p>
<span id="RODFT00-_0028DST_002dI_0029"></span><h4 class="subsubheading">RODFT00 (DST-I)</h4>
<span id="index-RODFT00-1"></span>
<p>An <code>RODFT00</code> transform (type-I DST) in FFTW is defined by:
<center><img src="equation-rodft00.png" align="top">.</center>
</p>
<span id="RODFT10-_0028DST_002dII_0029"></span><h4 class="subsubheading">RODFT10 (DST-II)</h4>
<span id="index-RODFT10"></span>
<p>An <code>RODFT10</code> transform (type-II DST) in FFTW is defined by:
<center><img src="equation-rodft10.png" align="top">.</center>
</p>
<span id="RODFT01-_0028DST_002dIII_0029"></span><h4 class="subsubheading">RODFT01 (DST-III)</h4>
<span id="index-RODFT01"></span>
<p>An <code>RODFT01</code> transform (type-III DST) in FFTW is defined by:
<center><img src="equation-rodft01.png" align="top">.</center>
In the case of <em>n=1</em>, this reduces to
<i>Y</i><sub>0</sub> = <i>X</i><sub>0</sub>.
</p>
<span id="RODFT11-_0028DST_002dIV_0029"></span><h4 class="subsubheading">RODFT11 (DST-IV)</h4>
<span id="index-RODFT11"></span>
<p>An <code>RODFT11</code> transform (type-IV DST) in FFTW is defined by:
<center><img src="equation-rodft11.png" align="top">.</center>
</p>
<span id="Inverses-and-Normalization-1"></span><h4 class="subsubheading">Inverses and Normalization</h4>
<p>These definitions correspond directly to the unnormalized DFTs used
elsewhere in FFTW (hence the factors of <em>2</em> in front of the
summations). The unnormalized inverse of <code>RODFT00</code> is
<code>RODFT00</code>, of <code>RODFT10</code> is <code>RODFT01</code> and vice versa, and
of <code>RODFT11</code> is <code>RODFT11</code>. Each unnormalized inverse results
in the original array multiplied by <em>N</em>, where <em>N</em> is the
<em>logical</em> DFT size. For <code>RODFT00</code>, <em>N=2(n+1)</em>;
otherwise, <em>N=2n</em>.
<span id="index-normalization-11"></span>
</p>
<p>In defining the discrete sine transform, some authors also include
additional factors of
&radic;2
(or its inverse) multiplying selected inputs and/or outputs. This is a
mostly cosmetic change that makes the transform orthogonal, but
sacrifices the direct equivalence to an antisymmetric DFT.
</p>
<hr>
<div class="header">
<p>
Next: <a href="1d-Discrete-Hartley-Transforms-_0028DHTs_0029.html" accesskey="n" rel="next">1d Discrete Hartley Transforms (DHTs)</a>, Previous: <a href="1d-Real_002deven-DFTs-_0028DCTs_0029.html" accesskey="p" rel="prev">1d Real-even DFTs (DCTs)</a>, Up: <a href="What-FFTW-Really-Computes.html" accesskey="u" rel="up">What FFTW Really Computes</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
</body>
</html>

View File

@@ -0,0 +1,183 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html>
<!-- This manual is for FFTW
(version 3.3.10, 10 December 2020).
Copyright (C) 2003 Matteo Frigo.
Copyright (C) 2003 Massachusetts Institute of Technology.
Permission is granted to make and distribute verbatim copies of this
manual provided the copyright notice and this permission notice are
preserved on all copies.
Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided that the
entire resulting derived work is distributed under the terms of a
permission notice identical to this one.
Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions,
except that this permission notice may be stated in a translation
approved by the Free Software Foundation. -->
<!-- Created by GNU Texinfo 6.7, http://www.gnu.org/software/texinfo/ -->
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<title>2d MPI example (FFTW 3.3.10)</title>
<meta name="description" content="2d MPI example (FFTW 3.3.10)">
<meta name="keywords" content="2d MPI example (FFTW 3.3.10)">
<meta name="resource-type" content="document">
<meta name="distribution" content="global">
<meta name="Generator" content="makeinfo">
<link href="index.html" rel="start" title="Top">
<link href="Concept-Index.html" rel="index" title="Concept Index">
<link href="index.html#SEC_Contents" rel="contents" title="Table of Contents">
<link href="Distributed_002dmemory-FFTW-with-MPI.html" rel="up" title="Distributed-memory FFTW with MPI">
<link href="MPI-Data-Distribution.html" rel="next" title="MPI Data Distribution">
<link href="Linking-and-Initializing-MPI-FFTW.html" rel="prev" title="Linking and Initializing MPI FFTW">
<style type="text/css">
<!--
a.summary-letter {text-decoration: none}
blockquote.indentedblock {margin-right: 0em}
div.display {margin-left: 3.2em}
div.example {margin-left: 3.2em}
div.lisp {margin-left: 3.2em}
kbd {font-style: oblique}
pre.display {font-family: inherit}
pre.format {font-family: inherit}
pre.menu-comment {font-family: serif}
pre.menu-preformatted {font-family: serif}
span.nolinebreak {white-space: nowrap}
span.roman {font-family: initial; font-weight: normal}
span.sansserif {font-family: sans-serif; font-weight: normal}
ul.no-bullet {list-style: none}
-->
</style>
</head>
<body lang="en">
<span id="g_t2d-MPI-example"></span><div class="header">
<p>
Next: <a href="MPI-Data-Distribution.html" accesskey="n" rel="next">MPI Data Distribution</a>, Previous: <a href="Linking-and-Initializing-MPI-FFTW.html" accesskey="p" rel="prev">Linking and Initializing MPI FFTW</a>, Up: <a href="Distributed_002dmemory-FFTW-with-MPI.html" accesskey="u" rel="up">Distributed-memory FFTW with MPI</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
<hr>
<span id="g_t2d-MPI-example-1"></span><h3 class="section">6.3 2d MPI example</h3>
<p>Before we document the FFTW MPI interface in detail, we begin with a
simple example outlining how one would perform a two-dimensional
<code>N0</code> by <code>N1</code> complex DFT.
</p>
<div class="example">
<pre class="example">#include &lt;fftw3-mpi.h&gt;
int main(int argc, char **argv)
{
const ptrdiff_t N0 = ..., N1 = ...;
fftw_plan plan;
fftw_complex *data;
ptrdiff_t alloc_local, local_n0, local_0_start, i, j;
MPI_Init(&amp;argc, &amp;argv);
fftw_mpi_init();
/* <span class="roman">get local data size and allocate</span> */
alloc_local = fftw_mpi_local_size_2d(N0, N1, MPI_COMM_WORLD,
&amp;local_n0, &amp;local_0_start);
data = fftw_alloc_complex(alloc_local);
/* <span class="roman">create plan for in-place forward DFT</span> */
plan = fftw_mpi_plan_dft_2d(N0, N1, data, data, MPI_COMM_WORLD,
FFTW_FORWARD, FFTW_ESTIMATE);
/* <span class="roman">initialize data to some function</span> my_function(x,y) */
for (i = 0; i &lt; local_n0; ++i) for (j = 0; j &lt; N1; ++j)
data[i*N1 + j] = my_function(local_0_start + i, j);
/* <span class="roman">compute transforms, in-place, as many times as desired</span> */
fftw_execute(plan);
fftw_destroy_plan(plan);
MPI_Finalize();
}
</pre></div>
<p>As can be seen above, the MPI interface follows the same basic style
of allocate/plan/execute/destroy as the serial FFTW routines. All of
the MPI-specific routines are prefixed with &lsquo;<samp>fftw_mpi_</samp>&rsquo; instead
of &lsquo;<samp>fftw_</samp>&rsquo;. There are a few important differences, however:
</p>
<p>First, we must call <code>fftw_mpi_init()</code> after calling
<code>MPI_Init</code> (required in all MPI programs) and before calling any
other &lsquo;<samp>fftw_mpi_</samp>&rsquo; routine.
<span id="index-MPI_005fInit"></span>
<span id="index-fftw_005fmpi_005finit-1"></span>
</p>
<p>Second, when we create the plan with <code>fftw_mpi_plan_dft_2d</code>,
analogous to <code>fftw_plan_dft_2d</code>, we pass an additional argument:
the communicator, indicating which processes will participate in the
transform (here <code>MPI_COMM_WORLD</code>, indicating all processes).
Whenever you create, execute, or destroy a plan for an MPI transform,
you must call the corresponding FFTW routine on <em>all</em> processes
in the communicator for that transform. (That is, these are
<em>collective</em> calls.) Note that the plan for the MPI transform
uses the standard <code>fftw_execute</code> and <code>fftw_destroy</code> routines
(on the other hand, there are MPI-specific new-array execute functions
documented below).
<span id="index-collective-function"></span>
<span id="index-fftw_005fmpi_005fplan_005fdft_005f2d"></span>
<span id="index-MPI_005fCOMM_005fWORLD-1"></span>
</p>
<p>Third, all of the FFTW MPI routines take <code>ptrdiff_t</code> arguments
instead of <code>int</code> as for the serial FFTW. <code>ptrdiff_t</code> is a
standard C integer type which is (at least) 32 bits wide on a 32-bit
machine and 64 bits wide on a 64-bit machine. This is to make it easy
to specify very large parallel transforms on a 64-bit machine. (You
can specify 64-bit transform sizes in the serial FFTW, too, but only
by using the &lsquo;<samp>guru64</samp>&rsquo; planner interface. See <a href="64_002dbit-Guru-Interface.html">64-bit Guru Interface</a>.)
<span id="index-ptrdiff_005ft-1"></span>
<span id="index-64_002dbit-architecture-1"></span>
</p>
<p>Fourth, and most importantly, you don&rsquo;t allocate the entire
two-dimensional array on each process. Instead, you call
<code>fftw_mpi_local_size_2d</code> to find out what <em>portion</em> of the
array resides on each processor, and how much space to allocate.
Here, the portion of the array on each process is a <code>local_n0</code> by
<code>N1</code> slice of the total array, starting at index
<code>local_0_start</code>. The total number of <code>fftw_complex</code> numbers
to allocate is given by the <code>alloc_local</code> return value, which
<em>may</em> be greater than <code>local_n0 * N1</code> (in case some
intermediate calculations require additional storage). The data
distribution in FFTW&rsquo;s MPI interface is described in more detail by
the next section.
<span id="index-fftw_005fmpi_005flocal_005fsize_005f2d"></span>
<span id="index-data-distribution-1"></span>
</p>
<p>Given the portion of the array that resides on the local process, it
is straightforward to initialize the data (here to a function
<code>myfunction</code>) and otherwise manipulate it. Of course, at the end
of the program you may want to output the data somehow, but
synchronizing this output is up to you and is beyond the scope of this
manual. (One good way to output a large multi-dimensional distributed
array in MPI to a portable binary file is to use the free HDF5
library; see the <a href="http://www.hdfgroup.org/">HDF home page</a>.)
<span id="index-HDF5"></span>
<span id="index-MPI-I_002fO"></span>
</p>
<hr>
<div class="header">
<p>
Next: <a href="MPI-Data-Distribution.html" accesskey="n" rel="next">MPI Data Distribution</a>, Previous: <a href="Linking-and-Initializing-MPI-FFTW.html" accesskey="p" rel="prev">Linking and Initializing MPI FFTW</a>, Up: <a href="Distributed_002dmemory-FFTW-with-MPI.html" accesskey="u" rel="up">Distributed-memory FFTW with MPI</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
</body>
</html>

View File

@@ -0,0 +1,137 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html>
<!-- This manual is for FFTW
(version 3.3.10, 10 December 2020).
Copyright (C) 2003 Matteo Frigo.
Copyright (C) 2003 Massachusetts Institute of Technology.
Permission is granted to make and distribute verbatim copies of this
manual provided the copyright notice and this permission notice are
preserved on all copies.
Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided that the
entire resulting derived work is distributed under the terms of a
permission notice identical to this one.
Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions,
except that this permission notice may be stated in a translation
approved by the Free Software Foundation. -->
<!-- Created by GNU Texinfo 6.7, http://www.gnu.org/software/texinfo/ -->
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<title>64-bit Guru Interface (FFTW 3.3.10)</title>
<meta name="description" content="64-bit Guru Interface (FFTW 3.3.10)">
<meta name="keywords" content="64-bit Guru Interface (FFTW 3.3.10)">
<meta name="resource-type" content="document">
<meta name="distribution" content="global">
<meta name="Generator" content="makeinfo">
<link href="index.html" rel="start" title="Top">
<link href="Concept-Index.html" rel="index" title="Concept Index">
<link href="index.html#SEC_Contents" rel="contents" title="Table of Contents">
<link href="Guru-Interface.html" rel="up" title="Guru Interface">
<link href="New_002darray-Execute-Functions.html" rel="next" title="New-array Execute Functions">
<link href="Guru-Real_002dto_002dreal-Transforms.html" rel="prev" title="Guru Real-to-real Transforms">
<style type="text/css">
<!--
a.summary-letter {text-decoration: none}
blockquote.indentedblock {margin-right: 0em}
div.display {margin-left: 3.2em}
div.example {margin-left: 3.2em}
div.lisp {margin-left: 3.2em}
kbd {font-style: oblique}
pre.display {font-family: inherit}
pre.format {font-family: inherit}
pre.menu-comment {font-family: serif}
pre.menu-preformatted {font-family: serif}
span.nolinebreak {white-space: nowrap}
span.roman {font-family: initial; font-weight: normal}
span.sansserif {font-family: sans-serif; font-weight: normal}
ul.no-bullet {list-style: none}
-->
</style>
</head>
<body lang="en">
<span id="g_t64_002dbit-Guru-Interface"></span><div class="header">
<p>
Previous: <a href="Guru-Real_002dto_002dreal-Transforms.html" accesskey="p" rel="prev">Guru Real-to-real Transforms</a>, Up: <a href="Guru-Interface.html" accesskey="u" rel="up">Guru Interface</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
<hr>
<span id="g_t64_002dbit-Guru-Interface-1"></span><h4 class="subsection">4.5.6 64-bit Guru Interface</h4>
<span id="index-64_002dbit-architecture"></span>
<p>When compiled in 64-bit mode on a 64-bit architecture (where addresses
are 64 bits wide), FFTW uses 64-bit quantities internally for all
transform sizes, strides, and so on&mdash;you don&rsquo;t have to do anything
special to exploit this. However, in the ordinary FFTW interfaces,
you specify the transform size by an <code>int</code> quantity, which is
normally only 32 bits wide. This means that, even though FFTW is
using 64-bit sizes internally, you cannot specify a single transform
dimension larger than
2<sup><small>31</small></sup>&minus;1
numbers.
</p>
<p>We expect that few users will require transforms larger than this, but,
for those who do, we provide a 64-bit version of the guru interface in
which all sizes are specified as integers of type <code>ptrdiff_t</code>
instead of <code>int</code>. (<code>ptrdiff_t</code> is a signed integer type
defined by the C standard to be wide enough to represent address
differences, and thus must be at least 64 bits wide on a 64-bit
machine.) We stress that there is <em>no performance advantage</em> to
using this interface&mdash;the same internal FFTW code is employed
regardless&mdash;and it is only necessary if you want to specify very
large transform sizes.
<span id="index-ptrdiff_005ft"></span>
</p>
<p>In particular, the 64-bit guru interface is a set of planner routines
that are exactly the same as the guru planner routines, except that
they are named with &lsquo;<samp>guru64</samp>&rsquo; instead of &lsquo;<samp>guru</samp>&rsquo; and they take
arguments of type <code>fftw_iodim64</code> instead of <code>fftw_iodim</code>.
For example, instead of <code>fftw_plan_guru_dft</code>, we have
<code>fftw_plan_guru64_dft</code>.
</p>
<div class="example">
<pre class="example">fftw_plan fftw_plan_guru64_dft(
int rank, const fftw_iodim64 *dims,
int howmany_rank, const fftw_iodim64 *howmany_dims,
fftw_complex *in, fftw_complex *out,
int sign, unsigned flags);
</pre></div>
<span id="index-fftw_005fplan_005fguru64_005fdft"></span>
<p>The <code>fftw_iodim64</code> type is similar to <code>fftw_iodim</code>, with the
same interpretation, except that it uses type <code>ptrdiff_t</code> instead
of type <code>int</code>.
</p>
<div class="example">
<pre class="example">typedef struct {
ptrdiff_t n;
ptrdiff_t is;
ptrdiff_t os;
} fftw_iodim64;
</pre></div>
<span id="index-fftw_005fiodim64"></span>
<p>Every other &lsquo;<samp>fftw_plan_guru</samp>&rsquo; function also has a
&lsquo;<samp>fftw_plan_guru64</samp>&rsquo; equivalent, but we do not repeat their
documentation here since they are identical to the 32-bit versions
except as noted above.
</p>
<hr>
<div class="header">
<p>
Previous: <a href="Guru-Real_002dto_002dreal-Transforms.html" accesskey="p" rel="prev">Guru Real-to-real Transforms</a>, Up: <a href="Guru-Interface.html" accesskey="u" rel="up">Guru Interface</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
</body>
</html>

View File

@@ -0,0 +1,91 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html>
<!-- This manual is for FFTW
(version 3.3.10, 10 December 2020).
Copyright (C) 2003 Matteo Frigo.
Copyright (C) 2003 Massachusetts Institute of Technology.
Permission is granted to make and distribute verbatim copies of this
manual provided the copyright notice and this permission notice are
preserved on all copies.
Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided that the
entire resulting derived work is distributed under the terms of a
permission notice identical to this one.
Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions,
except that this permission notice may be stated in a translation
approved by the Free Software Foundation. -->
<!-- Created by GNU Texinfo 6.7, http://www.gnu.org/software/texinfo/ -->
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<title>Accessing the wisdom API from Fortran (FFTW 3.3.10)</title>
<meta name="description" content="Accessing the wisdom API from Fortran (FFTW 3.3.10)">
<meta name="keywords" content="Accessing the wisdom API from Fortran (FFTW 3.3.10)">
<meta name="resource-type" content="document">
<meta name="distribution" content="global">
<meta name="Generator" content="makeinfo">
<link href="index.html" rel="start" title="Top">
<link href="Concept-Index.html" rel="index" title="Concept Index">
<link href="index.html#SEC_Contents" rel="contents" title="Table of Contents">
<link href="Calling-FFTW-from-Modern-Fortran.html" rel="up" title="Calling FFTW from Modern Fortran">
<link href="Wisdom-File-Export_002fImport-from-Fortran.html" rel="next" title="Wisdom File Export/Import from Fortran">
<link href="Allocating-aligned-memory-in-Fortran.html" rel="prev" title="Allocating aligned memory in Fortran">
<style type="text/css">
<!--
a.summary-letter {text-decoration: none}
blockquote.indentedblock {margin-right: 0em}
div.display {margin-left: 3.2em}
div.example {margin-left: 3.2em}
div.lisp {margin-left: 3.2em}
kbd {font-style: oblique}
pre.display {font-family: inherit}
pre.format {font-family: inherit}
pre.menu-comment {font-family: serif}
pre.menu-preformatted {font-family: serif}
span.nolinebreak {white-space: nowrap}
span.roman {font-family: initial; font-weight: normal}
span.sansserif {font-family: sans-serif; font-weight: normal}
ul.no-bullet {list-style: none}
-->
</style>
</head>
<body lang="en">
<span id="Accessing-the-wisdom-API-from-Fortran"></span><div class="header">
<p>
Next: <a href="Defining-an-FFTW-module.html" accesskey="n" rel="next">Defining an FFTW module</a>, Previous: <a href="Allocating-aligned-memory-in-Fortran.html" accesskey="p" rel="prev">Allocating aligned memory in Fortran</a>, Up: <a href="Calling-FFTW-from-Modern-Fortran.html" accesskey="u" rel="up">Calling FFTW from Modern Fortran</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
<hr>
<span id="Accessing-the-wisdom-API-from-Fortran-1"></span><h3 class="section">7.6 Accessing the wisdom API from Fortran</h3>
<span id="index-wisdom-3"></span>
<span id="index-saving-plans-to-disk-3"></span>
<p>As explained in <a href="Words-of-Wisdom_002dSaving-Plans.html">Words of Wisdom-Saving Plans</a>, FFTW provides a
&ldquo;wisdom&rdquo; API for saving plans to disk so that they can be recreated
quickly. The C API for exporting (see <a href="Wisdom-Export.html">Wisdom Export</a>) and
importing (see <a href="Wisdom-Import.html">Wisdom Import</a>) wisdom is somewhat tricky to use
from Fortran, however, because of differences in file I/O and string
types between C and Fortran.
</p>
<table class="menu" border="0" cellspacing="0">
<tr><td align="left" valign="top">&bull; <a href="Wisdom-File-Export_002fImport-from-Fortran.html" accesskey="1">Wisdom File Export/Import from Fortran</a></td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="Wisdom-String-Export_002fImport-from-Fortran.html" accesskey="2">Wisdom String Export/Import from Fortran</a></td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="Wisdom-Generic-Export_002fImport-from-Fortran.html" accesskey="3">Wisdom Generic Export/Import from Fortran</a></td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
</table>
</body>
</html>

View File

@@ -0,0 +1,164 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html>
<!-- This manual is for FFTW
(version 3.3.10, 10 December 2020).
Copyright (C) 2003 Matteo Frigo.
Copyright (C) 2003 Massachusetts Institute of Technology.
Permission is granted to make and distribute verbatim copies of this
manual provided the copyright notice and this permission notice are
preserved on all copies.
Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided that the
entire resulting derived work is distributed under the terms of a
permission notice identical to this one.
Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions,
except that this permission notice may be stated in a translation
approved by the Free Software Foundation. -->
<!-- Created by GNU Texinfo 6.7, http://www.gnu.org/software/texinfo/ -->
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<title>Acknowledgments (FFTW 3.3.10)</title>
<meta name="description" content="Acknowledgments (FFTW 3.3.10)">
<meta name="keywords" content="Acknowledgments (FFTW 3.3.10)">
<meta name="resource-type" content="document">
<meta name="distribution" content="global">
<meta name="Generator" content="makeinfo">
<link href="index.html" rel="start" title="Top">
<link href="Concept-Index.html" rel="index" title="Concept Index">
<link href="index.html#SEC_Contents" rel="contents" title="Table of Contents">
<link href="index.html" rel="up" title="Top">
<link href="License-and-Copyright.html" rel="next" title="License and Copyright">
<link href="Generating-your-own-code.html" rel="prev" title="Generating your own code">
<style type="text/css">
<!--
a.summary-letter {text-decoration: none}
blockquote.indentedblock {margin-right: 0em}
div.display {margin-left: 3.2em}
div.example {margin-left: 3.2em}
div.lisp {margin-left: 3.2em}
kbd {font-style: oblique}
pre.display {font-family: inherit}
pre.format {font-family: inherit}
pre.menu-comment {font-family: serif}
pre.menu-preformatted {font-family: serif}
span.nolinebreak {white-space: nowrap}
span.roman {font-family: initial; font-weight: normal}
span.sansserif {font-family: sans-serif; font-weight: normal}
ul.no-bullet {list-style: none}
-->
</style>
</head>
<body lang="en">
<span id="Acknowledgments"></span><div class="header">
<p>
Next: <a href="License-and-Copyright.html" accesskey="n" rel="next">License and Copyright</a>, Previous: <a href="Installation-and-Customization.html" accesskey="p" rel="prev">Installation and Customization</a>, Up: <a href="index.html" accesskey="u" rel="up">Top</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
<hr>
<span id="Acknowledgments-1"></span><h2 class="chapter">11 Acknowledgments</h2>
<p>Matteo Frigo was supported in part by the Special Research Program SFB
F011 &ldquo;AURORA&rdquo; of the Austrian Science Fund FWF and by MIT Lincoln
Laboratory. For previous versions of FFTW, he was supported in part by the
Defense Advanced Research Projects Agency (DARPA), under Grants
N00014-94-1-0985 and F30602-97-1-0270, and by a Digital Equipment
Corporation Fellowship.
</p>
<p>Steven G. Johnson was supported in part by a Dept.&nbsp;of Defense NDSEG
Fellowship, an MIT Karl Taylor Compton Fellowship, and by the Materials
Research Science and Engineering Center program of the National Science
Foundation under award DMR-9400334.
</p>
<p>Code for the Cell Broadband Engine was graciously donated to the FFTW
project by the IBM Austin Research Lab and included in fftw-3.2. (This
code was removed in fftw-3.3.)
</p>
<p>Code for the MIPS paired-single SIMD support was graciously donated to
the FFTW project by CodeSourcery, Inc.
</p>
<p>We are grateful to Sun Microsystems Inc.&nbsp;for its donation of a
cluster of 9 8-processor Ultra HPC 5000 SMPs (24 Gflops peak). These
machines served as the primary platform for the development of early
versions of FFTW.
</p>
<p>We thank Intel Corporation for donating a four-processor Pentium Pro
machine. We thank the GNU/Linux community for giving us a decent OS to
run on that machine.
</p>
<p>We are thankful to the AMD corporation for donating an AMD Athlon XP 1700+
computer to the FFTW project.
</p>
<p>We thank the Compaq/HP testdrive program and VA Software Corporation
(SourceForge.net) for providing remote access to machines that were used
to test FFTW.
</p>
<p>The <code>genfft</code> suite of code generators was written using Objective
Caml, a dialect of ML. Objective Caml is a small and elegant language
developed by Xavier Leroy. The implementation is available from
<a href="http://caml.inria.fr/"><code>http://caml.inria.fr/</code></a>. In previous
releases of FFTW, <code>genfft</code> was written in Caml Light, by the same
authors. An even earlier implementation of <code>genfft</code> was written in
Scheme, but Caml is definitely better for this kind of application.
<span id="index-Caml-1"></span>
<span id="index-LISP"></span>
</p>
<p>FFTW uses many tools from the GNU project, including <code>automake</code>,
<code>texinfo</code>, and <code>libtool</code>.
</p>
<p>Prof.&nbsp;Charles E.&nbsp;Leiserson of MIT provided continuous support and
encouragement. This program would not exist without him. Charles also
proposed the name &ldquo;codelets&rdquo; for the basic FFT blocks.
<span id="index-codelet-3"></span>
</p>
<p>Prof.&nbsp;John D.&nbsp;Joannopoulos of MIT demonstrated continuing tolerance of
Steven&rsquo;s &ldquo;extra-curricular&rdquo; computer-science activities, as well as
remarkable creativity in working them into his grant proposals.
Steven&rsquo;s physics degree would not exist without him.
</p>
<p>Franz Franchetti wrote SIMD extensions to FFTW 2, which eventually
led to the SIMD support in FFTW 3.
</p>
<p>Stefan Kral wrote most of the K7 code generator distributed with FFTW
3.0.x and 3.1.x.
</p>
<p>Andrew Sterian contributed the Windows timing code in FFTW 2.
</p>
<p>Didier Miras reported a bug in the test procedure used in FFTW 1.2. We
now use a completely different test algorithm by Funda Ergun that does
not require a separate FFT program to compare against.
</p>
<p>Wolfgang Reimer contributed the Pentium cycle counter and a few fixes
that help portability.
</p>
<p>Ming-Chang Liu uncovered a well-hidden bug in the complex transforms of
FFTW 2.0 and supplied a patch to correct it.
</p>
<p>The FFTW FAQ was written in <code>bfnn</code> (Bizarre Format With No Name)
and formatted using the tools developed by Ian Jackson for the Linux
FAQ.
</p>
<p><em>We are especially thankful to all of our users for their
continuing support, feedback, and interest during our development of
FFTW.</em>
</p>
<hr>
<div class="header">
<p>
Next: <a href="License-and-Copyright.html" accesskey="n" rel="next">License and Copyright</a>, Previous: <a href="Installation-and-Customization.html" accesskey="p" rel="prev">Installation and Customization</a>, Up: <a href="index.html" accesskey="u" rel="up">Top</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
</body>
</html>

View File

@@ -0,0 +1,177 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html>
<!-- This manual is for FFTW
(version 3.3.10, 10 December 2020).
Copyright (C) 2003 Matteo Frigo.
Copyright (C) 2003 Massachusetts Institute of Technology.
Permission is granted to make and distribute verbatim copies of this
manual provided the copyright notice and this permission notice are
preserved on all copies.
Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided that the
entire resulting derived work is distributed under the terms of a
permission notice identical to this one.
Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions,
except that this permission notice may be stated in a translation
approved by the Free Software Foundation. -->
<!-- Created by GNU Texinfo 6.7, http://www.gnu.org/software/texinfo/ -->
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<title>Advanced Complex DFTs (FFTW 3.3.10)</title>
<meta name="description" content="Advanced Complex DFTs (FFTW 3.3.10)">
<meta name="keywords" content="Advanced Complex DFTs (FFTW 3.3.10)">
<meta name="resource-type" content="document">
<meta name="distribution" content="global">
<meta name="Generator" content="makeinfo">
<link href="index.html" rel="start" title="Top">
<link href="Concept-Index.html" rel="index" title="Concept Index">
<link href="index.html#SEC_Contents" rel="contents" title="Table of Contents">
<link href="Advanced-Interface.html" rel="up" title="Advanced Interface">
<link href="Advanced-Real_002ddata-DFTs.html" rel="next" title="Advanced Real-data DFTs">
<link href="Advanced-Interface.html" rel="prev" title="Advanced Interface">
<style type="text/css">
<!--
a.summary-letter {text-decoration: none}
blockquote.indentedblock {margin-right: 0em}
div.display {margin-left: 3.2em}
div.example {margin-left: 3.2em}
div.lisp {margin-left: 3.2em}
kbd {font-style: oblique}
pre.display {font-family: inherit}
pre.format {font-family: inherit}
pre.menu-comment {font-family: serif}
pre.menu-preformatted {font-family: serif}
span.nolinebreak {white-space: nowrap}
span.roman {font-family: initial; font-weight: normal}
span.sansserif {font-family: sans-serif; font-weight: normal}
ul.no-bullet {list-style: none}
-->
</style>
</head>
<body lang="en">
<span id="Advanced-Complex-DFTs"></span><div class="header">
<p>
Next: <a href="Advanced-Real_002ddata-DFTs.html" accesskey="n" rel="next">Advanced Real-data DFTs</a>, Previous: <a href="Advanced-Interface.html" accesskey="p" rel="prev">Advanced Interface</a>, Up: <a href="Advanced-Interface.html" accesskey="u" rel="up">Advanced Interface</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
<hr>
<span id="Advanced-Complex-DFTs-1"></span><h4 class="subsection">4.4.1 Advanced Complex DFTs</h4>
<div class="example">
<pre class="example">fftw_plan fftw_plan_many_dft(int rank, const int *n, int howmany,
fftw_complex *in, const int *inembed,
int istride, int idist,
fftw_complex *out, const int *onembed,
int ostride, int odist,
int sign, unsigned flags);
</pre></div>
<span id="index-fftw_005fplan_005fmany_005fdft"></span>
<p>This routine plans multiple multidimensional complex DFTs, and it
extends the <code>fftw_plan_dft</code> routine (see <a href="Complex-DFTs.html">Complex DFTs</a>) to
compute <code>howmany</code> transforms, each having rank <code>rank</code> and size
<code>n</code>. In addition, the transform data need not be contiguous, but
it may be laid out in memory with an arbitrary stride. To account for
these possibilities, <code>fftw_plan_many_dft</code> adds the new parameters
<code>howmany</code>, {<code>i</code>,<code>o</code>}<code>nembed</code>,
{<code>i</code>,<code>o</code>}<code>stride</code>, and
{<code>i</code>,<code>o</code>}<code>dist</code>. The FFTW basic interface
(see <a href="Complex-DFTs.html">Complex DFTs</a>) provides routines specialized for ranks 1, 2,
and&nbsp;3, but the advanced interface handles only the general-rank
case.
</p>
<p><code>howmany</code> is the (nonnegative) number of transforms to compute. The resulting
plan computes <code>howmany</code> transforms, where the input of the
<code>k</code>-th transform is at location <code>in+k*idist</code> (in C pointer
arithmetic), and its output is at location <code>out+k*odist</code>. Plans
obtained in this way can often be faster than calling FFTW multiple
times for the individual transforms. The basic <code>fftw_plan_dft</code>
interface corresponds to <code>howmany=1</code> (in which case the <code>dist</code>
parameters are ignored).
<span id="index-howmany-parameter"></span>
<span id="index-dist"></span>
</p>
<p>Each of the <code>howmany</code> transforms has rank <code>rank</code> and size
<code>n</code>, as in the basic interface. In addition, the advanced
interface allows the input and output arrays of each transform to be
row-major subarrays of larger rank-<code>rank</code> arrays, described by
<code>inembed</code> and <code>onembed</code> parameters, respectively.
{<code>i</code>,<code>o</code>}<code>nembed</code> must be arrays of length <code>rank</code>,
and <code>n</code> should be elementwise less than or equal to
{<code>i</code>,<code>o</code>}<code>nembed</code>. Passing <code>NULL</code> for an
<code>nembed</code> parameter is equivalent to passing <code>n</code> (i.e. same
physical and logical dimensions, as in the basic interface.)
</p>
<p>The <code>stride</code> parameters indicate that the <code>j</code>-th element of
the input or output arrays is located at <code>j*istride</code> or
<code>j*ostride</code>, respectively. (For a multi-dimensional array,
<code>j</code> is the ordinary row-major index.) When combined with the
<code>k</code>-th transform in a <code>howmany</code> loop, from above, this means
that the (<code>j</code>,<code>k</code>)-th element is at <code>j*stride+k*dist</code>.
(The basic <code>fftw_plan_dft</code> interface corresponds to a stride of 1.)
<span id="index-stride-1"></span>
</p>
<p>For in-place transforms, the input and output <code>stride</code> and
<code>dist</code> parameters should be the same; otherwise, the planner may
return <code>NULL</code>.
</p>
<p>Arrays <code>n</code>, <code>inembed</code>, and <code>onembed</code> are not used after
this function returns. You can safely free or reuse them.
</p>
<p><strong>Examples</strong>:
One transform of one 5 by 6 array contiguous in memory:
</p><div class="example">
<pre class="example"> int rank = 2;
int n[] = {5, 6};
int howmany = 1;
int idist = odist = 0; /* unused because howmany = 1 */
int istride = ostride = 1; /* array is contiguous in memory */
int *inembed = n, *onembed = n;
</pre></div>
<p>Transform of three 5 by 6 arrays, each contiguous in memory,
stored in memory one after another:
</p><div class="example">
<pre class="example"> int rank = 2;
int n[] = {5, 6};
int howmany = 3;
int idist = odist = n[0]*n[1]; /* = 30, the distance in memory
between the first element
of the first array and the
first element of the second array */
int istride = ostride = 1; /* array is contiguous in memory */
int *inembed = n, *onembed = n;
</pre></div>
<p>Transform each column of a 2d array with 10 rows and 3 columns:
</p><div class="example">
<pre class="example"> int rank = 1; /* not 2: we are computing 1d transforms */
int n[] = {10}; /* 1d transforms of length 10 */
int howmany = 3;
int idist = odist = 1;
int istride = ostride = 3; /* distance between two elements in
the same column */
int *inembed = n, *onembed = n;
</pre></div>
<hr>
<div class="header">
<p>
Next: <a href="Advanced-Real_002ddata-DFTs.html" accesskey="n" rel="next">Advanced Real-data DFTs</a>, Previous: <a href="Advanced-Interface.html" accesskey="p" rel="prev">Advanced Interface</a>, Up: <a href="Advanced-Interface.html" accesskey="u" rel="up">Advanced Interface</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
</body>
</html>

View File

@@ -0,0 +1,91 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html>
<!-- This manual is for FFTW
(version 3.3.10, 10 December 2020).
Copyright (C) 2003 Matteo Frigo.
Copyright (C) 2003 Massachusetts Institute of Technology.
Permission is granted to make and distribute verbatim copies of this
manual provided the copyright notice and this permission notice are
preserved on all copies.
Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided that the
entire resulting derived work is distributed under the terms of a
permission notice identical to this one.
Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions,
except that this permission notice may be stated in a translation
approved by the Free Software Foundation. -->
<!-- Created by GNU Texinfo 6.7, http://www.gnu.org/software/texinfo/ -->
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<title>Advanced Interface (FFTW 3.3.10)</title>
<meta name="description" content="Advanced Interface (FFTW 3.3.10)">
<meta name="keywords" content="Advanced Interface (FFTW 3.3.10)">
<meta name="resource-type" content="document">
<meta name="distribution" content="global">
<meta name="Generator" content="makeinfo">
<link href="index.html" rel="start" title="Top">
<link href="Concept-Index.html" rel="index" title="Concept Index">
<link href="index.html#SEC_Contents" rel="contents" title="Table of Contents">
<link href="FFTW-Reference.html" rel="up" title="FFTW Reference">
<link href="Advanced-Complex-DFTs.html" rel="next" title="Advanced Complex DFTs">
<link href="Real_002dto_002dReal-Transform-Kinds.html" rel="prev" title="Real-to-Real Transform Kinds">
<style type="text/css">
<!--
a.summary-letter {text-decoration: none}
blockquote.indentedblock {margin-right: 0em}
div.display {margin-left: 3.2em}
div.example {margin-left: 3.2em}
div.lisp {margin-left: 3.2em}
kbd {font-style: oblique}
pre.display {font-family: inherit}
pre.format {font-family: inherit}
pre.menu-comment {font-family: serif}
pre.menu-preformatted {font-family: serif}
span.nolinebreak {white-space: nowrap}
span.roman {font-family: initial; font-weight: normal}
span.sansserif {font-family: sans-serif; font-weight: normal}
ul.no-bullet {list-style: none}
-->
</style>
</head>
<body lang="en">
<span id="Advanced-Interface"></span><div class="header">
<p>
Next: <a href="Guru-Interface.html" accesskey="n" rel="next">Guru Interface</a>, Previous: <a href="Basic-Interface.html" accesskey="p" rel="prev">Basic Interface</a>, Up: <a href="FFTW-Reference.html" accesskey="u" rel="up">FFTW Reference</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
<hr>
<span id="Advanced-Interface-1"></span><h3 class="section">4.4 Advanced Interface</h3>
<span id="index-advanced-interface-3"></span>
<p>FFTW&rsquo;s &ldquo;advanced&rdquo; interface supplements the basic interface with four
new planner routines, providing a new level of flexibility: you can plan
a transform of multiple arrays simultaneously, operate on non-contiguous
(strided) data, and transform a subset of a larger multi-dimensional
array. Other than these additional features, the planner operates in
the same fashion as in the basic interface, and the resulting
<code>fftw_plan</code> is used in the same way (see <a href="Using-Plans.html">Using Plans</a>).
</p>
<table class="menu" border="0" cellspacing="0">
<tr><td align="left" valign="top">&bull; <a href="Advanced-Complex-DFTs.html" accesskey="1">Advanced Complex DFTs</a></td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="Advanced-Real_002ddata-DFTs.html" accesskey="2">Advanced Real-data DFTs</a></td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="Advanced-Real_002dto_002dreal-Transforms.html" accesskey="3">Advanced Real-to-real Transforms</a></td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
</table>
</body>
</html>

View File

@@ -0,0 +1,122 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html>
<!-- This manual is for FFTW
(version 3.3.10, 10 December 2020).
Copyright (C) 2003 Matteo Frigo.
Copyright (C) 2003 Massachusetts Institute of Technology.
Permission is granted to make and distribute verbatim copies of this
manual provided the copyright notice and this permission notice are
preserved on all copies.
Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided that the
entire resulting derived work is distributed under the terms of a
permission notice identical to this one.
Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions,
except that this permission notice may be stated in a translation
approved by the Free Software Foundation. -->
<!-- Created by GNU Texinfo 6.7, http://www.gnu.org/software/texinfo/ -->
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<title>Advanced Real-data DFTs (FFTW 3.3.10)</title>
<meta name="description" content="Advanced Real-data DFTs (FFTW 3.3.10)">
<meta name="keywords" content="Advanced Real-data DFTs (FFTW 3.3.10)">
<meta name="resource-type" content="document">
<meta name="distribution" content="global">
<meta name="Generator" content="makeinfo">
<link href="index.html" rel="start" title="Top">
<link href="Concept-Index.html" rel="index" title="Concept Index">
<link href="index.html#SEC_Contents" rel="contents" title="Table of Contents">
<link href="Advanced-Interface.html" rel="up" title="Advanced Interface">
<link href="Advanced-Real_002dto_002dreal-Transforms.html" rel="next" title="Advanced Real-to-real Transforms">
<link href="Advanced-Complex-DFTs.html" rel="prev" title="Advanced Complex DFTs">
<style type="text/css">
<!--
a.summary-letter {text-decoration: none}
blockquote.indentedblock {margin-right: 0em}
div.display {margin-left: 3.2em}
div.example {margin-left: 3.2em}
div.lisp {margin-left: 3.2em}
kbd {font-style: oblique}
pre.display {font-family: inherit}
pre.format {font-family: inherit}
pre.menu-comment {font-family: serif}
pre.menu-preformatted {font-family: serif}
span.nolinebreak {white-space: nowrap}
span.roman {font-family: initial; font-weight: normal}
span.sansserif {font-family: sans-serif; font-weight: normal}
ul.no-bullet {list-style: none}
-->
</style>
</head>
<body lang="en">
<span id="Advanced-Real_002ddata-DFTs"></span><div class="header">
<p>
Next: <a href="Advanced-Real_002dto_002dreal-Transforms.html" accesskey="n" rel="next">Advanced Real-to-real Transforms</a>, Previous: <a href="Advanced-Complex-DFTs.html" accesskey="p" rel="prev">Advanced Complex DFTs</a>, Up: <a href="Advanced-Interface.html" accesskey="u" rel="up">Advanced Interface</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
<hr>
<span id="Advanced-Real_002ddata-DFTs-1"></span><h4 class="subsection">4.4.2 Advanced Real-data DFTs</h4>
<div class="example">
<pre class="example">fftw_plan fftw_plan_many_dft_r2c(int rank, const int *n, int howmany,
double *in, const int *inembed,
int istride, int idist,
fftw_complex *out, const int *onembed,
int ostride, int odist,
unsigned flags);
fftw_plan fftw_plan_many_dft_c2r(int rank, const int *n, int howmany,
fftw_complex *in, const int *inembed,
int istride, int idist,
double *out, const int *onembed,
int ostride, int odist,
unsigned flags);
</pre></div>
<span id="index-fftw_005fplan_005fmany_005fdft_005fr2c"></span>
<span id="index-fftw_005fplan_005fmany_005fdft_005fc2r"></span>
<p>Like <code>fftw_plan_many_dft</code>, these two functions add <code>howmany</code>,
<code>nembed</code>, <code>stride</code>, and <code>dist</code> parameters to the
<code>fftw_plan_dft_r2c</code> and <code>fftw_plan_dft_c2r</code> functions, but
otherwise behave the same as the basic interface.
</p>
<p>The interpretation of <code>howmany</code>, <code>stride</code>, and <code>dist</code> are
the same as for <code>fftw_plan_many_dft</code>, above. Note that the
<code>stride</code> and <code>dist</code> for the real array are in units of
<code>double</code>, and for the complex array are in units of
<code>fftw_complex</code>.
</p>
<p>If an <code>nembed</code> parameter is <code>NULL</code>, it is interpreted as what
it would be in the basic interface, as described in <a href="Real_002ddata-DFT-Array-Format.html">Real-data DFT Array Format</a>. That is, for the complex array the size is assumed to be
the same as <code>n</code>, but with the last dimension cut roughly in half.
For the real array, the size is assumed to be <code>n</code> if the transform
is out-of-place, or <code>n</code> with the last dimension &ldquo;padded&rdquo; if the
transform is in-place.
</p>
<p>If an <code>nembed</code> parameter is non-<code>NULL</code>, it is interpreted as
the physical size of the corresponding array, in row-major order, just
as for <code>fftw_plan_many_dft</code>. In this case, each dimension of
<code>nembed</code> should be <code>&gt;=</code> what it would be in the basic
interface (e.g. the halved or padded <code>n</code>).
</p>
<p>Arrays <code>n</code>, <code>inembed</code>, and <code>onembed</code> are not used after
this function returns. You can safely free or reuse them.
</p>
<hr>
<div class="header">
<p>
Next: <a href="Advanced-Real_002dto_002dreal-Transforms.html" accesskey="n" rel="next">Advanced Real-to-real Transforms</a>, Previous: <a href="Advanced-Complex-DFTs.html" accesskey="p" rel="prev">Advanced Complex DFTs</a>, Up: <a href="Advanced-Interface.html" accesskey="u" rel="up">Advanced Interface</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
</body>
</html>

View File

@@ -0,0 +1,94 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html>
<!-- This manual is for FFTW
(version 3.3.10, 10 December 2020).
Copyright (C) 2003 Matteo Frigo.
Copyright (C) 2003 Massachusetts Institute of Technology.
Permission is granted to make and distribute verbatim copies of this
manual provided the copyright notice and this permission notice are
preserved on all copies.
Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided that the
entire resulting derived work is distributed under the terms of a
permission notice identical to this one.
Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions,
except that this permission notice may be stated in a translation
approved by the Free Software Foundation. -->
<!-- Created by GNU Texinfo 6.7, http://www.gnu.org/software/texinfo/ -->
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<title>Advanced Real-to-real Transforms (FFTW 3.3.10)</title>
<meta name="description" content="Advanced Real-to-real Transforms (FFTW 3.3.10)">
<meta name="keywords" content="Advanced Real-to-real Transforms (FFTW 3.3.10)">
<meta name="resource-type" content="document">
<meta name="distribution" content="global">
<meta name="Generator" content="makeinfo">
<link href="index.html" rel="start" title="Top">
<link href="Concept-Index.html" rel="index" title="Concept Index">
<link href="index.html#SEC_Contents" rel="contents" title="Table of Contents">
<link href="Advanced-Interface.html" rel="up" title="Advanced Interface">
<link href="Guru-Interface.html" rel="next" title="Guru Interface">
<link href="Advanced-Real_002ddata-DFTs.html" rel="prev" title="Advanced Real-data DFTs">
<style type="text/css">
<!--
a.summary-letter {text-decoration: none}
blockquote.indentedblock {margin-right: 0em}
div.display {margin-left: 3.2em}
div.example {margin-left: 3.2em}
div.lisp {margin-left: 3.2em}
kbd {font-style: oblique}
pre.display {font-family: inherit}
pre.format {font-family: inherit}
pre.menu-comment {font-family: serif}
pre.menu-preformatted {font-family: serif}
span.nolinebreak {white-space: nowrap}
span.roman {font-family: initial; font-weight: normal}
span.sansserif {font-family: sans-serif; font-weight: normal}
ul.no-bullet {list-style: none}
-->
</style>
</head>
<body lang="en">
<span id="Advanced-Real_002dto_002dreal-Transforms"></span><div class="header">
<p>
Previous: <a href="Advanced-Real_002ddata-DFTs.html" accesskey="p" rel="prev">Advanced Real-data DFTs</a>, Up: <a href="Advanced-Interface.html" accesskey="u" rel="up">Advanced Interface</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
<hr>
<span id="Advanced-Real_002dto_002dreal-Transforms-1"></span><h4 class="subsection">4.4.3 Advanced Real-to-real Transforms</h4>
<div class="example">
<pre class="example">fftw_plan fftw_plan_many_r2r(int rank, const int *n, int howmany,
double *in, const int *inembed,
int istride, int idist,
double *out, const int *onembed,
int ostride, int odist,
const fftw_r2r_kind *kind, unsigned flags);
</pre></div>
<span id="index-fftw_005fplan_005fmany_005fr2r"></span>
<p>Like <code>fftw_plan_many_dft</code>, this functions adds <code>howmany</code>,
<code>nembed</code>, <code>stride</code>, and <code>dist</code> parameters to the
<code>fftw_plan_r2r</code> function, but otherwise behave the same as the
basic interface. The interpretation of those additional parameters are
the same as for <code>fftw_plan_many_dft</code>. (Of course, the
<code>stride</code> and <code>dist</code> parameters are now in units of
<code>double</code>, not <code>fftw_complex</code>.)
</p>
<p>Arrays <code>n</code>, <code>inembed</code>, <code>onembed</code>, and <code>kind</code> are not
used after this function returns. You can safely free or reuse them.
</p>
</body>
</html>

View File

@@ -0,0 +1,97 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html>
<!-- This manual is for FFTW
(version 3.3.10, 10 December 2020).
Copyright (C) 2003 Matteo Frigo.
Copyright (C) 2003 Massachusetts Institute of Technology.
Permission is granted to make and distribute verbatim copies of this
manual provided the copyright notice and this permission notice are
preserved on all copies.
Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided that the
entire resulting derived work is distributed under the terms of a
permission notice identical to this one.
Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions,
except that this permission notice may be stated in a translation
approved by the Free Software Foundation. -->
<!-- Created by GNU Texinfo 6.7, http://www.gnu.org/software/texinfo/ -->
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<title>Advanced distributed-transpose interface (FFTW 3.3.10)</title>
<meta name="description" content="Advanced distributed-transpose interface (FFTW 3.3.10)">
<meta name="keywords" content="Advanced distributed-transpose interface (FFTW 3.3.10)">
<meta name="resource-type" content="document">
<meta name="distribution" content="global">
<meta name="Generator" content="makeinfo">
<link href="index.html" rel="start" title="Top">
<link href="Concept-Index.html" rel="index" title="Concept Index">
<link href="index.html#SEC_Contents" rel="contents" title="Table of Contents">
<link href="FFTW-MPI-Transposes.html" rel="up" title="FFTW MPI Transposes">
<link href="An-improved-replacement-for-MPI_005fAlltoall.html" rel="next" title="An improved replacement for MPI_Alltoall">
<link href="Basic-distributed_002dtranspose-interface.html" rel="prev" title="Basic distributed-transpose interface">
<style type="text/css">
<!--
a.summary-letter {text-decoration: none}
blockquote.indentedblock {margin-right: 0em}
div.display {margin-left: 3.2em}
div.example {margin-left: 3.2em}
div.lisp {margin-left: 3.2em}
kbd {font-style: oblique}
pre.display {font-family: inherit}
pre.format {font-family: inherit}
pre.menu-comment {font-family: serif}
pre.menu-preformatted {font-family: serif}
span.nolinebreak {white-space: nowrap}
span.roman {font-family: initial; font-weight: normal}
span.sansserif {font-family: sans-serif; font-weight: normal}
ul.no-bullet {list-style: none}
-->
</style>
</head>
<body lang="en">
<span id="Advanced-distributed_002dtranspose-interface"></span><div class="header">
<p>
Next: <a href="An-improved-replacement-for-MPI_005fAlltoall.html" accesskey="n" rel="next">An improved replacement for MPI_Alltoall</a>, Previous: <a href="Basic-distributed_002dtranspose-interface.html" accesskey="p" rel="prev">Basic distributed-transpose interface</a>, Up: <a href="FFTW-MPI-Transposes.html" accesskey="u" rel="up">FFTW MPI Transposes</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
<hr>
<span id="Advanced-distributed_002dtranspose-interface-1"></span><h4 class="subsection">6.7.2 Advanced distributed-transpose interface</h4>
<p>The above routines are for a transpose of a matrix of numbers (of type
<code>double</code>), using FFTW&rsquo;s default block sizes. More generally, one
can perform transposes of <em>tuples</em> of numbers, with
user-specified block sizes for the input and output:
</p>
<div class="example">
<pre class="example">fftw_plan fftw_mpi_plan_many_transpose
(ptrdiff_t n0, ptrdiff_t n1, ptrdiff_t howmany,
ptrdiff_t block0, ptrdiff_t block1,
double *in, double *out, MPI_Comm comm, unsigned flags);
</pre></div>
<span id="index-fftw_005fmpi_005fplan_005fmany_005ftranspose"></span>
<p>In this case, one is transposing an <code>n0</code> by <code>n1</code> matrix of
<code>howmany</code>-tuples (e.g. <code>howmany = 2</code> for complex numbers).
The input is distributed along the <code>n0</code> dimension with block size
<code>block0</code>, and the <code>n1</code> by <code>n0</code> output is distributed
along the <code>n1</code> dimension with block size <code>block1</code>. If
<code>FFTW_MPI_DEFAULT_BLOCK</code> (0) is passed for a block size then FFTW
uses its default block size. To get the local size of the data on
each process, you should then call <code>fftw_mpi_local_size_many_transposed</code>.
<span id="index-FFTW_005fMPI_005fDEFAULT_005fBLOCK-1"></span>
<span id="index-fftw_005fmpi_005flocal_005fsize_005fmany_005ftransposed"></span>
</p>
</body>
</html>

View File

@@ -0,0 +1,157 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html>
<!-- This manual is for FFTW
(version 3.3.10, 10 December 2020).
Copyright (C) 2003 Matteo Frigo.
Copyright (C) 2003 Massachusetts Institute of Technology.
Permission is granted to make and distribute verbatim copies of this
manual provided the copyright notice and this permission notice are
preserved on all copies.
Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided that the
entire resulting derived work is distributed under the terms of a
permission notice identical to this one.
Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions,
except that this permission notice may be stated in a translation
approved by the Free Software Foundation. -->
<!-- Created by GNU Texinfo 6.7, http://www.gnu.org/software/texinfo/ -->
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<title>Allocating aligned memory in Fortran (FFTW 3.3.10)</title>
<meta name="description" content="Allocating aligned memory in Fortran (FFTW 3.3.10)">
<meta name="keywords" content="Allocating aligned memory in Fortran (FFTW 3.3.10)">
<meta name="resource-type" content="document">
<meta name="distribution" content="global">
<meta name="Generator" content="makeinfo">
<link href="index.html" rel="start" title="Top">
<link href="Concept-Index.html" rel="index" title="Concept Index">
<link href="index.html#SEC_Contents" rel="contents" title="Table of Contents">
<link href="Calling-FFTW-from-Modern-Fortran.html" rel="up" title="Calling FFTW from Modern Fortran">
<link href="Accessing-the-wisdom-API-from-Fortran.html" rel="next" title="Accessing the wisdom API from Fortran">
<link href="Plan-execution-in-Fortran.html" rel="prev" title="Plan execution in Fortran">
<style type="text/css">
<!--
a.summary-letter {text-decoration: none}
blockquote.indentedblock {margin-right: 0em}
div.display {margin-left: 3.2em}
div.example {margin-left: 3.2em}
div.lisp {margin-left: 3.2em}
kbd {font-style: oblique}
pre.display {font-family: inherit}
pre.format {font-family: inherit}
pre.menu-comment {font-family: serif}
pre.menu-preformatted {font-family: serif}
span.nolinebreak {white-space: nowrap}
span.roman {font-family: initial; font-weight: normal}
span.sansserif {font-family: sans-serif; font-weight: normal}
ul.no-bullet {list-style: none}
-->
</style>
</head>
<body lang="en">
<span id="Allocating-aligned-memory-in-Fortran"></span><div class="header">
<p>
Next: <a href="Accessing-the-wisdom-API-from-Fortran.html" accesskey="n" rel="next">Accessing the wisdom API from Fortran</a>, Previous: <a href="Plan-execution-in-Fortran.html" accesskey="p" rel="prev">Plan execution in Fortran</a>, Up: <a href="Calling-FFTW-from-Modern-Fortran.html" accesskey="u" rel="up">Calling FFTW from Modern Fortran</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
<hr>
<span id="Allocating-aligned-memory-in-Fortran-1"></span><h3 class="section">7.5 Allocating aligned memory in Fortran</h3>
<span id="index-alignment-5"></span>
<span id="index-fftw_005falloc_005freal-5"></span>
<span id="index-fftw_005falloc_005fcomplex-5"></span>
<p>In order to obtain maximum performance in FFTW, you should store your
data in arrays that have been specially aligned in memory (see <a href="SIMD-alignment-and-fftw_005fmalloc.html">SIMD alignment and fftw_malloc</a>). Enforcing alignment also permits you to
safely use the new-array execute functions (see <a href="New_002darray-Execute-Functions.html">New-array Execute Functions</a>) to apply a given plan to more than one pair of in/out
arrays. Unfortunately, standard Fortran arrays do <em>not</em> provide
any alignment guarantees. The <em>only</em> way to allocate aligned
memory in standard Fortran is to allocate it with an external C
function, like the <code>fftw_alloc_real</code> and
<code>fftw_alloc_complex</code> functions. Fortunately, Fortran 2003 provides
a simple way to associate such allocated memory with a standard Fortran
array pointer that you can then use normally.
</p>
<p>We therefore recommend allocating all your input/output arrays using
the following technique:
</p>
<ol>
<li> Declare a <code>pointer</code>, <code>arr</code>, to your array of the desired type
and dimensions. For example, <code>real(C_DOUBLE), pointer :: a(:,:)</code>
for a 2d real array, or <code>complex(C_DOUBLE_COMPLEX), pointer ::
a(:,:,:)</code> for a 3d complex array.
</li><li> The number of elements to allocate must be an
<code>integer(C_SIZE_T)</code>. You can either declare a variable of this
type, e.g. <code>integer(C_SIZE_T) :: sz</code>, to store the number of
elements to allocate, or you can use the <code>int(..., C_SIZE_T)</code>
intrinsic function. e.g. set <code>sz = L * M * N</code> or use
<code>int(L * M * N, C_SIZE_T)</code> for an L&nbsp;&times;&nbsp;M&nbsp;&times;&nbsp;N
array.
</li><li> Declare a <code>type(C_PTR) :: p</code> to hold the return value from
FFTW&rsquo;s allocation routine. Set <code>p = fftw_alloc_real(sz)</code> for a real array, or <code>p = fftw_alloc_complex(sz)</code> for a complex array.
</li><li> <span id="index-c_005ff_005fpointer-2"></span>
Associate your pointer <code>arr</code> with the allocated memory <code>p</code>
using the standard <code>c_f_pointer</code> subroutine: <code>call
c_f_pointer(p, arr, [...dimensions...])</code>, where
<code>[...dimensions...])</code> are an array of the dimensions of the array
(in the usual Fortran order). e.g. <code>call c_f_pointer(p, arr,
[L,M,N])</code> for an L&nbsp;&times;&nbsp;M&nbsp;&times;&nbsp;N
array. (Alternatively, you can
omit the dimensions argument if you specified the shape explicitly
when declaring <code>arr</code>.) You can now use <code>arr</code> as a usual
multidimensional array.
</li><li> When you are done using the array, deallocate the memory by <code>call
fftw_free(p)</code> on <code>p</code>.
</li></ol>
<p>For example, here is how we would allocate an L&nbsp;&times;&nbsp;M
2d real array:
</p>
<div class="example">
<pre class="example"> real(C_DOUBLE), pointer :: arr(:,:)
type(C_PTR) :: p
p = fftw_alloc_real(int(L * M, C_SIZE_T))
call c_f_pointer(p, arr, [L,M])
<em>...use arr and arr(i,j) as usual...</em>
call fftw_free(p)
</pre></div>
<p>and here is an L&nbsp;&times;&nbsp;M&nbsp;&times;&nbsp;N
3d complex array:
</p>
<div class="example">
<pre class="example"> complex(C_DOUBLE_COMPLEX), pointer :: arr(:,:,:)
type(C_PTR) :: p
p = fftw_alloc_complex(int(L * M * N, C_SIZE_T))
call c_f_pointer(p, arr, [L,M,N])
<em>...use arr and arr(i,j,k) as usual...</em>
call fftw_free(p)
</pre></div>
<p>See <a href="Reversing-array-dimensions.html">Reversing array dimensions</a> for an example allocating a
single array and associating both real and complex array pointers with
it, for in-place real-to-complex transforms.
</p>
<hr>
<div class="header">
<p>
Next: <a href="Accessing-the-wisdom-API-from-Fortran.html" accesskey="n" rel="next">Accessing the wisdom API from Fortran</a>, Previous: <a href="Plan-execution-in-Fortran.html" accesskey="p" rel="prev">Plan execution in Fortran</a>, Up: <a href="Calling-FFTW-from-Modern-Fortran.html" accesskey="u" rel="up">Calling FFTW from Modern Fortran</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
</body>
</html>

View File

@@ -0,0 +1,127 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html>
<!-- This manual is for FFTW
(version 3.3.10, 10 December 2020).
Copyright (C) 2003 Matteo Frigo.
Copyright (C) 2003 Massachusetts Institute of Technology.
Permission is granted to make and distribute verbatim copies of this
manual provided the copyright notice and this permission notice are
preserved on all copies.
Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided that the
entire resulting derived work is distributed under the terms of a
permission notice identical to this one.
Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions,
except that this permission notice may be stated in a translation
approved by the Free Software Foundation. -->
<!-- Created by GNU Texinfo 6.7, http://www.gnu.org/software/texinfo/ -->
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<title>An improved replacement for MPI_Alltoall (FFTW 3.3.10)</title>
<meta name="description" content="An improved replacement for MPI_Alltoall (FFTW 3.3.10)">
<meta name="keywords" content="An improved replacement for MPI_Alltoall (FFTW 3.3.10)">
<meta name="resource-type" content="document">
<meta name="distribution" content="global">
<meta name="Generator" content="makeinfo">
<link href="index.html" rel="start" title="Top">
<link href="Concept-Index.html" rel="index" title="Concept Index">
<link href="index.html#SEC_Contents" rel="contents" title="Table of Contents">
<link href="FFTW-MPI-Transposes.html" rel="up" title="FFTW MPI Transposes">
<link href="FFTW-MPI-Wisdom.html" rel="next" title="FFTW MPI Wisdom">
<link href="Advanced-distributed_002dtranspose-interface.html" rel="prev" title="Advanced distributed-transpose interface">
<style type="text/css">
<!--
a.summary-letter {text-decoration: none}
blockquote.indentedblock {margin-right: 0em}
div.display {margin-left: 3.2em}
div.example {margin-left: 3.2em}
div.lisp {margin-left: 3.2em}
kbd {font-style: oblique}
pre.display {font-family: inherit}
pre.format {font-family: inherit}
pre.menu-comment {font-family: serif}
pre.menu-preformatted {font-family: serif}
span.nolinebreak {white-space: nowrap}
span.roman {font-family: initial; font-weight: normal}
span.sansserif {font-family: sans-serif; font-weight: normal}
ul.no-bullet {list-style: none}
-->
</style>
</head>
<body lang="en">
<span id="An-improved-replacement-for-MPI_005fAlltoall"></span><div class="header">
<p>
Previous: <a href="Advanced-distributed_002dtranspose-interface.html" accesskey="p" rel="prev">Advanced distributed-transpose interface</a>, Up: <a href="FFTW-MPI-Transposes.html" accesskey="u" rel="up">FFTW MPI Transposes</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
<hr>
<span id="An-improved-replacement-for-MPI_005fAlltoall-1"></span><h4 class="subsection">6.7.3 An improved replacement for MPI_Alltoall</h4>
<p>We close this section by noting that FFTW&rsquo;s MPI transpose routines can
be thought of as a generalization for the <code>MPI_Alltoall</code> function
(albeit only for floating-point types), and in some circumstances can
function as an improved replacement.
<span id="index-MPI_005fAlltoall"></span>
</p>
<p><code>MPI_Alltoall</code> is defined by the MPI standard as:
</p>
<div class="example">
<pre class="example">int MPI_Alltoall(void *sendbuf, int sendcount, MPI_Datatype sendtype,
void *recvbuf, int recvcnt, MPI_Datatype recvtype,
MPI_Comm comm);
</pre></div>
<p>In particular, for <code>double*</code> arrays <code>in</code> and <code>out</code>,
consider the call:
</p>
<div class="example">
<pre class="example">MPI_Alltoall(in, howmany, MPI_DOUBLE, out, howmany MPI_DOUBLE, comm);
</pre></div>
<p>This is completely equivalent to:
</p>
<div class="example">
<pre class="example">MPI_Comm_size(comm, &amp;P);
plan = fftw_mpi_plan_many_transpose(P, P, howmany, 1, 1, in, out, comm, FFTW_ESTIMATE);
fftw_execute(plan);
fftw_destroy_plan(plan);
</pre></div>
<p>That is, computing a P&nbsp;&times;&nbsp;P
transpose on <code>P</code> processes,
with a block size of 1, is just a standard all-to-all communication.
</p>
<p>However, using the FFTW routine instead of <code>MPI_Alltoall</code> may
have certain advantages. First of all, FFTW&rsquo;s routine can operate
in-place (<code>in == out</code>) whereas <code>MPI_Alltoall</code> can only
operate out-of-place.
<span id="index-in_002dplace-8"></span>
</p>
<p>Second, even for out-of-place plans, FFTW&rsquo;s routine may be faster,
especially if you need to perform the all-to-all communication many
times and can afford to use <code>FFTW_MEASURE</code> or
<code>FFTW_PATIENT</code>. It should certainly be no slower, not including
the time to create the plan, since one of the possible algorithms that
FFTW uses for an out-of-place transpose <em>is</em> simply to call
<code>MPI_Alltoall</code>. However, FFTW also considers several other
possible algorithms that, depending on your MPI implementation and
your hardware, may be faster.
<span id="index-FFTW_005fMEASURE-3"></span>
<span id="index-FFTW_005fPATIENT-4"></span>
</p>
</body>
</html>

View File

@@ -0,0 +1,96 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html>
<!-- This manual is for FFTW
(version 3.3.10, 10 December 2020).
Copyright (C) 2003 Matteo Frigo.
Copyright (C) 2003 Massachusetts Institute of Technology.
Permission is granted to make and distribute verbatim copies of this
manual provided the copyright notice and this permission notice are
preserved on all copies.
Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided that the
entire resulting derived work is distributed under the terms of a
permission notice identical to this one.
Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions,
except that this permission notice may be stated in a translation
approved by the Free Software Foundation. -->
<!-- Created by GNU Texinfo 6.7, http://www.gnu.org/software/texinfo/ -->
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<title>Avoiding MPI Deadlocks (FFTW 3.3.10)</title>
<meta name="description" content="Avoiding MPI Deadlocks (FFTW 3.3.10)">
<meta name="keywords" content="Avoiding MPI Deadlocks (FFTW 3.3.10)">
<meta name="resource-type" content="document">
<meta name="distribution" content="global">
<meta name="Generator" content="makeinfo">
<link href="index.html" rel="start" title="Top">
<link href="Concept-Index.html" rel="index" title="Concept Index">
<link href="index.html#SEC_Contents" rel="contents" title="Table of Contents">
<link href="Distributed_002dmemory-FFTW-with-MPI.html" rel="up" title="Distributed-memory FFTW with MPI">
<link href="FFTW-MPI-Performance-Tips.html" rel="next" title="FFTW MPI Performance Tips">
<link href="FFTW-MPI-Wisdom.html" rel="prev" title="FFTW MPI Wisdom">
<style type="text/css">
<!--
a.summary-letter {text-decoration: none}
blockquote.indentedblock {margin-right: 0em}
div.display {margin-left: 3.2em}
div.example {margin-left: 3.2em}
div.lisp {margin-left: 3.2em}
kbd {font-style: oblique}
pre.display {font-family: inherit}
pre.format {font-family: inherit}
pre.menu-comment {font-family: serif}
pre.menu-preformatted {font-family: serif}
span.nolinebreak {white-space: nowrap}
span.roman {font-family: initial; font-weight: normal}
span.sansserif {font-family: sans-serif; font-weight: normal}
ul.no-bullet {list-style: none}
-->
</style>
</head>
<body lang="en">
<span id="Avoiding-MPI-Deadlocks"></span><div class="header">
<p>
Next: <a href="FFTW-MPI-Performance-Tips.html" accesskey="n" rel="next">FFTW MPI Performance Tips</a>, Previous: <a href="FFTW-MPI-Wisdom.html" accesskey="p" rel="prev">FFTW MPI Wisdom</a>, Up: <a href="Distributed_002dmemory-FFTW-with-MPI.html" accesskey="u" rel="up">Distributed-memory FFTW with MPI</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
<hr>
<span id="Avoiding-MPI-Deadlocks-1"></span><h3 class="section">6.9 Avoiding MPI Deadlocks</h3>
<span id="index-deadlock"></span>
<p>An MPI program can <em>deadlock</em> if one process is waiting for a
message from another process that never gets sent. To avoid deadlocks
when using FFTW&rsquo;s MPI routines, it is important to know which
functions are <em>collective</em>: that is, which functions must
<em>always</em> be called in the <em>same order</em> from <em>every</em>
process in a given communicator. (For example, <code>MPI_Barrier</code> is
the canonical example of a collective function in the MPI standard.)
<span id="index-collective-function-2"></span>
<span id="index-MPI_005fBarrier"></span>
</p>
<p>The functions in FFTW that are <em>always</em> collective are: every
function beginning with &lsquo;<samp>fftw_mpi_plan</samp>&rsquo;, as well as
<code>fftw_mpi_broadcast_wisdom</code> and <code>fftw_mpi_gather_wisdom</code>.
Also, the following functions from the ordinary FFTW interface are
collective when they are applied to a plan created by an
&lsquo;<samp>fftw_mpi_plan</samp>&rsquo; function: <code>fftw_execute</code>,
<code>fftw_destroy_plan</code>, and <code>fftw_flops</code>.
<span id="index-fftw_005fexecute-4"></span>
<span id="index-fftw_005fdestroy_005fplan-2"></span>
<span id="index-fftw_005fflops-1"></span>
</p>
</body>
</html>

View File

@@ -0,0 +1,104 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html>
<!-- This manual is for FFTW
(version 3.3.10, 10 December 2020).
Copyright (C) 2003 Matteo Frigo.
Copyright (C) 2003 Massachusetts Institute of Technology.
Permission is granted to make and distribute verbatim copies of this
manual provided the copyright notice and this permission notice are
preserved on all copies.
Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided that the
entire resulting derived work is distributed under the terms of a
permission notice identical to this one.
Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions,
except that this permission notice may be stated in a translation
approved by the Free Software Foundation. -->
<!-- Created by GNU Texinfo 6.7, http://www.gnu.org/software/texinfo/ -->
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<title>Basic Interface (FFTW 3.3.10)</title>
<meta name="description" content="Basic Interface (FFTW 3.3.10)">
<meta name="keywords" content="Basic Interface (FFTW 3.3.10)">
<meta name="resource-type" content="document">
<meta name="distribution" content="global">
<meta name="Generator" content="makeinfo">
<link href="index.html" rel="start" title="Top">
<link href="Concept-Index.html" rel="index" title="Concept Index">
<link href="index.html#SEC_Contents" rel="contents" title="Table of Contents">
<link href="FFTW-Reference.html" rel="up" title="FFTW Reference">
<link href="Complex-DFTs.html" rel="next" title="Complex DFTs">
<link href="Using-Plans.html" rel="prev" title="Using Plans">
<style type="text/css">
<!--
a.summary-letter {text-decoration: none}
blockquote.indentedblock {margin-right: 0em}
div.display {margin-left: 3.2em}
div.example {margin-left: 3.2em}
div.lisp {margin-left: 3.2em}
kbd {font-style: oblique}
pre.display {font-family: inherit}
pre.format {font-family: inherit}
pre.menu-comment {font-family: serif}
pre.menu-preformatted {font-family: serif}
span.nolinebreak {white-space: nowrap}
span.roman {font-family: initial; font-weight: normal}
span.sansserif {font-family: sans-serif; font-weight: normal}
ul.no-bullet {list-style: none}
-->
</style>
</head>
<body lang="en">
<span id="Basic-Interface"></span><div class="header">
<p>
Next: <a href="Advanced-Interface.html" accesskey="n" rel="next">Advanced Interface</a>, Previous: <a href="Using-Plans.html" accesskey="p" rel="prev">Using Plans</a>, Up: <a href="FFTW-Reference.html" accesskey="u" rel="up">FFTW Reference</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
<hr>
<span id="Basic-Interface-1"></span><h3 class="section">4.3 Basic Interface</h3>
<span id="index-basic-interface-2"></span>
<p>Recall that the FFTW API is divided into three parts<a id="DOCF6" href="#FOOT6"><sup>6</sup></a>: the <em>basic interface</em>
computes a single transform of contiguous data, the <em>advanced
interface</em> computes transforms of multiple or strided arrays, and the
<em>guru interface</em> supports the most general data layouts,
multiplicities, and strides. This section describes the basic
interface, which we expect to satisfy the needs of most users.
</p>
<table class="menu" border="0" cellspacing="0">
<tr><td align="left" valign="top">&bull; <a href="Complex-DFTs.html" accesskey="1">Complex DFTs</a></td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="Planner-Flags.html" accesskey="2">Planner Flags</a></td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="Real_002ddata-DFTs.html" accesskey="3">Real-data DFTs</a></td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="Real_002ddata-DFT-Array-Format.html" accesskey="4">Real-data DFT Array Format</a></td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="Real_002dto_002dReal-Transforms.html" accesskey="5">Real-to-Real Transforms</a></td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="Real_002dto_002dReal-Transform-Kinds.html" accesskey="6">Real-to-Real Transform Kinds</a></td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
</table>
<div class="footnote">
<hr>
<h4 class="footnotes-heading">Footnotes</h4>
<h5><a id="FOOT6" href="#DOCF6">(6)</a></h3>
<p><i>Gallia est
omnis divisa in partes tres</i> (Julius Caesar).</p>
</div>
</body>
</html>

View File

@@ -0,0 +1,177 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html>
<!-- This manual is for FFTW
(version 3.3.10, 10 December 2020).
Copyright (C) 2003 Matteo Frigo.
Copyright (C) 2003 Massachusetts Institute of Technology.
Permission is granted to make and distribute verbatim copies of this
manual provided the copyright notice and this permission notice are
preserved on all copies.
Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided that the
entire resulting derived work is distributed under the terms of a
permission notice identical to this one.
Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions,
except that this permission notice may be stated in a translation
approved by the Free Software Foundation. -->
<!-- Created by GNU Texinfo 6.7, http://www.gnu.org/software/texinfo/ -->
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<title>Basic and advanced distribution interfaces (FFTW 3.3.10)</title>
<meta name="description" content="Basic and advanced distribution interfaces (FFTW 3.3.10)">
<meta name="keywords" content="Basic and advanced distribution interfaces (FFTW 3.3.10)">
<meta name="resource-type" content="document">
<meta name="distribution" content="global">
<meta name="Generator" content="makeinfo">
<link href="index.html" rel="start" title="Top">
<link href="Concept-Index.html" rel="index" title="Concept Index">
<link href="index.html#SEC_Contents" rel="contents" title="Table of Contents">
<link href="MPI-Data-Distribution.html" rel="up" title="MPI Data Distribution">
<link href="Load-balancing.html" rel="next" title="Load balancing">
<link href="MPI-Data-Distribution.html" rel="prev" title="MPI Data Distribution">
<style type="text/css">
<!--
a.summary-letter {text-decoration: none}
blockquote.indentedblock {margin-right: 0em}
div.display {margin-left: 3.2em}
div.example {margin-left: 3.2em}
div.lisp {margin-left: 3.2em}
kbd {font-style: oblique}
pre.display {font-family: inherit}
pre.format {font-family: inherit}
pre.menu-comment {font-family: serif}
pre.menu-preformatted {font-family: serif}
span.nolinebreak {white-space: nowrap}
span.roman {font-family: initial; font-weight: normal}
span.sansserif {font-family: sans-serif; font-weight: normal}
ul.no-bullet {list-style: none}
-->
</style>
</head>
<body lang="en">
<span id="Basic-and-advanced-distribution-interfaces"></span><div class="header">
<p>
Next: <a href="Load-balancing.html" accesskey="n" rel="next">Load balancing</a>, Previous: <a href="MPI-Data-Distribution.html" accesskey="p" rel="prev">MPI Data Distribution</a>, Up: <a href="MPI-Data-Distribution.html" accesskey="u" rel="up">MPI Data Distribution</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
<hr>
<span id="Basic-and-advanced-distribution-interfaces-1"></span><h4 class="subsection">6.4.1 Basic and advanced distribution interfaces</h4>
<p>As with the planner interface, the &lsquo;<samp>fftw_mpi_local_size</samp>&rsquo;
distribution interface is broken into basic and advanced
(&lsquo;<samp>_many</samp>&rsquo;) interfaces, where the latter allows you to specify the
block size manually and also to request block sizes when computing
multiple transforms simultaneously. These functions are documented
more exhaustively by the FFTW MPI Reference, but we summarize the
basic ideas here using a couple of two-dimensional examples.
</p>
<p>For the 100&nbsp;&times;&nbsp;200
complex-DFT example, above, we would find
the distribution by calling the following function in the basic
interface:
</p>
<div class="example">
<pre class="example">ptrdiff_t fftw_mpi_local_size_2d(ptrdiff_t n0, ptrdiff_t n1, MPI_Comm comm,
ptrdiff_t *local_n0, ptrdiff_t *local_0_start);
</pre></div>
<span id="index-fftw_005fmpi_005flocal_005fsize_005f2d-1"></span>
<p>Given the total size of the data to be transformed (here, <code>n0 =
100</code> and <code>n1 = 200</code>) and an MPI communicator (<code>comm</code>), this
function provides three numbers.
</p>
<p>First, it describes the shape of the local data: the current process
should store a <code>local_n0</code> by <code>n1</code> slice of the overall
dataset, in row-major order (<code>n1</code> dimension contiguous), starting
at index <code>local_0_start</code>. That is, if the total dataset is
viewed as a <code>n0</code> by <code>n1</code> matrix, the current process should
store the rows <code>local_0_start</code> to
<code>local_0_start+local_n0-1</code>. Obviously, if you are running with
only a single MPI process, that process will store the entire array:
<code>local_0_start</code> will be zero and <code>local_n0</code> will be
<code>n0</code>. See <a href="Row_002dmajor-Format.html">Row-major Format</a>.
<span id="index-row_002dmajor-4"></span>
</p>
<p>Second, the return value is the total number of data elements (e.g.,
complex numbers for a complex DFT) that should be allocated for the
input and output arrays on the current process (ideally with
<code>fftw_malloc</code> or an &lsquo;<samp>fftw_alloc</samp>&rsquo; function, to ensure optimal
alignment). It might seem that this should always be equal to
<code>local_n0 * n1</code>, but this is <em>not</em> the case. FFTW&rsquo;s
distributed FFT algorithms require data redistributions at
intermediate stages of the transform, and in some circumstances this
may require slightly larger local storage. This is discussed in more
detail below, under <a href="Load-balancing.html">Load balancing</a>.
<span id="index-fftw_005fmalloc-5"></span>
<span id="index-fftw_005falloc_005fcomplex-3"></span>
</p>
<span id="index-advanced-interface-4"></span>
<p>The advanced-interface &lsquo;<samp>local_size</samp>&rsquo; function for multidimensional
transforms returns the same three things (<code>local_n0</code>,
<code>local_0_start</code>, and the total number of elements to allocate),
but takes more inputs:
</p>
<div class="example">
<pre class="example">ptrdiff_t fftw_mpi_local_size_many(int rnk, const ptrdiff_t *n,
ptrdiff_t howmany,
ptrdiff_t block0,
MPI_Comm comm,
ptrdiff_t *local_n0,
ptrdiff_t *local_0_start);
</pre></div>
<span id="index-fftw_005fmpi_005flocal_005fsize_005fmany"></span>
<p>The two-dimensional case above corresponds to <code>rnk = 2</code> and an
array <code>n</code> of length 2 with <code>n[0] = n0</code> and <code>n[1] = n1</code>.
This routine is for any <code>rnk &gt; 1</code>; one-dimensional transforms
have their own interface because they work slightly differently, as
discussed below.
</p>
<p>First, the advanced interface allows you to perform multiple
transforms at once, of interleaved data, as specified by the
<code>howmany</code> parameter. (<code>hoamany</code> is 1 for a single
transform.)
</p>
<p>Second, here you can specify your desired block size in the <code>n0</code>
dimension, <code>block0</code>. To use FFTW&rsquo;s default block size, pass
<code>FFTW_MPI_DEFAULT_BLOCK</code> (0) for <code>block0</code>. Otherwise, on
<code>P</code> processes, FFTW will return <code>local_n0</code> equal to
<code>block0</code> on the first <code>P / block0</code> processes (rounded down),
return <code>local_n0</code> equal to <code>n0 - block0 * (P / block0)</code> on
the next process, and <code>local_n0</code> equal to zero on any remaining
processes. In general, we recommend using the default block size
(which corresponds to <code>n0 / P</code>, rounded up).
<span id="index-FFTW_005fMPI_005fDEFAULT_005fBLOCK"></span>
<span id="index-block-distribution-1"></span>
</p>
<p>For example, suppose you have <code>P = 4</code> processes and <code>n0 =
21</code>. The default will be a block size of <code>6</code>, which will give
<code>local_n0 = 6</code> on the first three processes and <code>local_n0 =
3</code> on the last process. Instead, however, you could specify
<code>block0 = 5</code> if you wanted, which would give <code>local_n0 = 5</code>
on processes 0 to 2, <code>local_n0 = 6</code> on process 3. (This choice,
while it may look superficially more &ldquo;balanced,&rdquo; has the same
critical path as FFTW&rsquo;s default but requires more communications.)
</p>
<hr>
<div class="header">
<p>
Next: <a href="Load-balancing.html" accesskey="n" rel="next">Load balancing</a>, Previous: <a href="MPI-Data-Distribution.html" accesskey="p" rel="prev">MPI Data Distribution</a>, Up: <a href="MPI-Data-Distribution.html" accesskey="u" rel="up">MPI Data Distribution</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
</body>
</html>

View File

@@ -0,0 +1,129 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html>
<!-- This manual is for FFTW
(version 3.3.10, 10 December 2020).
Copyright (C) 2003 Matteo Frigo.
Copyright (C) 2003 Massachusetts Institute of Technology.
Permission is granted to make and distribute verbatim copies of this
manual provided the copyright notice and this permission notice are
preserved on all copies.
Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided that the
entire resulting derived work is distributed under the terms of a
permission notice identical to this one.
Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions,
except that this permission notice may be stated in a translation
approved by the Free Software Foundation. -->
<!-- Created by GNU Texinfo 6.7, http://www.gnu.org/software/texinfo/ -->
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<title>Basic distributed-transpose interface (FFTW 3.3.10)</title>
<meta name="description" content="Basic distributed-transpose interface (FFTW 3.3.10)">
<meta name="keywords" content="Basic distributed-transpose interface (FFTW 3.3.10)">
<meta name="resource-type" content="document">
<meta name="distribution" content="global">
<meta name="Generator" content="makeinfo">
<link href="index.html" rel="start" title="Top">
<link href="Concept-Index.html" rel="index" title="Concept Index">
<link href="index.html#SEC_Contents" rel="contents" title="Table of Contents">
<link href="FFTW-MPI-Transposes.html" rel="up" title="FFTW MPI Transposes">
<link href="Advanced-distributed_002dtranspose-interface.html" rel="next" title="Advanced distributed-transpose interface">
<link href="FFTW-MPI-Transposes.html" rel="prev" title="FFTW MPI Transposes">
<style type="text/css">
<!--
a.summary-letter {text-decoration: none}
blockquote.indentedblock {margin-right: 0em}
div.display {margin-left: 3.2em}
div.example {margin-left: 3.2em}
div.lisp {margin-left: 3.2em}
kbd {font-style: oblique}
pre.display {font-family: inherit}
pre.format {font-family: inherit}
pre.menu-comment {font-family: serif}
pre.menu-preformatted {font-family: serif}
span.nolinebreak {white-space: nowrap}
span.roman {font-family: initial; font-weight: normal}
span.sansserif {font-family: sans-serif; font-weight: normal}
ul.no-bullet {list-style: none}
-->
</style>
</head>
<body lang="en">
<span id="Basic-distributed_002dtranspose-interface"></span><div class="header">
<p>
Next: <a href="Advanced-distributed_002dtranspose-interface.html" accesskey="n" rel="next">Advanced distributed-transpose interface</a>, Previous: <a href="FFTW-MPI-Transposes.html" accesskey="p" rel="prev">FFTW MPI Transposes</a>, Up: <a href="FFTW-MPI-Transposes.html" accesskey="u" rel="up">FFTW MPI Transposes</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
<hr>
<span id="Basic-distributed_002dtranspose-interface-1"></span><h4 class="subsection">6.7.1 Basic distributed-transpose interface</h4>
<p>In particular, suppose that we have an <code>n0</code> by <code>n1</code> array in
row-major order, block-distributed across the <code>n0</code> dimension. To
transpose this into an <code>n1</code> by <code>n0</code> array block-distributed
across the <code>n1</code> dimension, we would create a plan by calling the
following function:
</p>
<div class="example">
<pre class="example">fftw_plan fftw_mpi_plan_transpose(ptrdiff_t n0, ptrdiff_t n1,
double *in, double *out,
MPI_Comm comm, unsigned flags);
</pre></div>
<span id="index-fftw_005fmpi_005fplan_005ftranspose"></span>
<p>The input and output arrays (<code>in</code> and <code>out</code>) can be the
same. The transpose is actually executed by calling
<code>fftw_execute</code> on the plan, as usual.
<span id="index-fftw_005fexecute-3"></span>
</p>
<p>The <code>flags</code> are the usual FFTW planner flags, but support
two additional flags: <code>FFTW_MPI_TRANSPOSED_OUT</code> and/or
<code>FFTW_MPI_TRANSPOSED_IN</code>. What these flags indicate, for
transpose plans, is that the output and/or input, respectively, are
<em>locally</em> transposed. That is, on each process input data is
normally stored as a <code>local_n0</code> by <code>n1</code> array in row-major
order, but for an <code>FFTW_MPI_TRANSPOSED_IN</code> plan the input data is
stored as <code>n1</code> by <code>local_n0</code> in row-major order. Similarly,
<code>FFTW_MPI_TRANSPOSED_OUT</code> means that the output is <code>n0</code> by
<code>local_n1</code> instead of <code>local_n1</code> by <code>n0</code>.
<span id="index-FFTW_005fMPI_005fTRANSPOSED_005fOUT-1"></span>
<span id="index-FFTW_005fMPI_005fTRANSPOSED_005fIN-1"></span>
</p>
<p>To determine the local size of the array on each process before and
after the transpose, as well as the amount of storage that must be
allocated, one should call <code>fftw_mpi_local_size_2d_transposed</code>,
just as for a 2d DFT as described in the previous section:
<span id="index-data-distribution-4"></span>
</p>
<div class="example">
<pre class="example">ptrdiff_t fftw_mpi_local_size_2d_transposed
(ptrdiff_t n0, ptrdiff_t n1, MPI_Comm comm,
ptrdiff_t *local_n0, ptrdiff_t *local_0_start,
ptrdiff_t *local_n1, ptrdiff_t *local_1_start);
</pre></div>
<span id="index-fftw_005fmpi_005flocal_005fsize_005f2d_005ftransposed"></span>
<p>Again, the return value is the local storage to allocate, which in
this case is the number of <em>real</em> (<code>double</code>) values rather
than complex numbers as in the previous examples.
</p>
<hr>
<div class="header">
<p>
Next: <a href="Advanced-distributed_002dtranspose-interface.html" accesskey="n" rel="next">Advanced distributed-transpose interface</a>, Previous: <a href="FFTW-MPI-Transposes.html" accesskey="p" rel="prev">FFTW MPI Transposes</a>, Up: <a href="FFTW-MPI-Transposes.html" accesskey="u" rel="up">FFTW MPI Transposes</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
</body>
</html>

View File

@@ -0,0 +1,115 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html>
<!-- This manual is for FFTW
(version 3.3.10, 10 December 2020).
Copyright (C) 2003 Matteo Frigo.
Copyright (C) 2003 Massachusetts Institute of Technology.
Permission is granted to make and distribute verbatim copies of this
manual provided the copyright notice and this permission notice are
preserved on all copies.
Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided that the
entire resulting derived work is distributed under the terms of a
permission notice identical to this one.
Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions,
except that this permission notice may be stated in a translation
approved by the Free Software Foundation. -->
<!-- Created by GNU Texinfo 6.7, http://www.gnu.org/software/texinfo/ -->
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<title>Calling FFTW from Legacy Fortran (FFTW 3.3.10)</title>
<meta name="description" content="Calling FFTW from Legacy Fortran (FFTW 3.3.10)">
<meta name="keywords" content="Calling FFTW from Legacy Fortran (FFTW 3.3.10)">
<meta name="resource-type" content="document">
<meta name="distribution" content="global">
<meta name="Generator" content="makeinfo">
<link href="index.html" rel="start" title="Top">
<link href="Concept-Index.html" rel="index" title="Concept Index">
<link href="index.html#SEC_Contents" rel="contents" title="Table of Contents">
<link href="index.html" rel="up" title="Top">
<link href="Fortran_002dinterface-routines.html" rel="next" title="Fortran-interface routines">
<link href="Defining-an-FFTW-module.html" rel="prev" title="Defining an FFTW module">
<style type="text/css">
<!--
a.summary-letter {text-decoration: none}
blockquote.indentedblock {margin-right: 0em}
div.display {margin-left: 3.2em}
div.example {margin-left: 3.2em}
div.lisp {margin-left: 3.2em}
kbd {font-style: oblique}
pre.display {font-family: inherit}
pre.format {font-family: inherit}
pre.menu-comment {font-family: serif}
pre.menu-preformatted {font-family: serif}
span.nolinebreak {white-space: nowrap}
span.roman {font-family: initial; font-weight: normal}
span.sansserif {font-family: sans-serif; font-weight: normal}
ul.no-bullet {list-style: none}
-->
</style>
</head>
<body lang="en">
<span id="Calling-FFTW-from-Legacy-Fortran"></span><div class="header">
<p>
Next: <a href="Upgrading-from-FFTW-version-2.html" accesskey="n" rel="next">Upgrading from FFTW version 2</a>, Previous: <a href="Calling-FFTW-from-Modern-Fortran.html" accesskey="p" rel="prev">Calling FFTW from Modern Fortran</a>, Up: <a href="index.html" accesskey="u" rel="up">Top</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
<hr>
<span id="Calling-FFTW-from-Legacy-Fortran-1"></span><h2 class="chapter">8 Calling FFTW from Legacy Fortran</h2>
<span id="index-Fortran-interface-3"></span>
<p>This chapter describes the interface to FFTW callable by Fortran code
in older compilers not supporting the Fortran 2003 C interoperability
features (see <a href="Calling-FFTW-from-Modern-Fortran.html">Calling FFTW from Modern Fortran</a>). This interface
has the major disadvantage that it is not type-checked, so if you
mistake the argument types or ordering then your program will not have
any compiler errors, and will likely crash at runtime. So, greater
care is needed. Also, technically interfacing older Fortran versions
to C is nonstandard, but in practice we have found that the techniques
used in this chapter have worked with all known Fortran compilers for
many years.
</p>
<p>The legacy Fortran interface differs from the C interface only in the
prefix (&lsquo;<samp>dfftw_</samp>&rsquo; instead of &lsquo;<samp>fftw_</samp>&rsquo; in double precision) and
a few other minor details. This Fortran interface is included in the
FFTW libraries by default, unless a Fortran compiler isn&rsquo;t found on
your system or <code>--disable-fortran</code> is included in the
<code>configure</code> flags. We assume here that the reader is already
familiar with the usage of FFTW in C, as described elsewhere in this
manual.
</p>
<p>The MPI parallel interface to FFTW is <em>not</em> currently available
to legacy Fortran.
</p>
<table class="menu" border="0" cellspacing="0">
<tr><td align="left" valign="top">&bull; <a href="Fortran_002dinterface-routines.html" accesskey="1">Fortran-interface routines</a></td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="FFTW-Constants-in-Fortran.html" accesskey="2">FFTW Constants in Fortran</a></td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="FFTW-Execution-in-Fortran.html" accesskey="3">FFTW Execution in Fortran</a></td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="Fortran-Examples.html" accesskey="4">Fortran Examples</a></td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="Wisdom-of-Fortran_003f.html" accesskey="5">Wisdom of Fortran?</a></td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
</table>
<hr>
<div class="header">
<p>
Next: <a href="Upgrading-from-FFTW-version-2.html" accesskey="n" rel="next">Upgrading from FFTW version 2</a>, Previous: <a href="Calling-FFTW-from-Modern-Fortran.html" accesskey="p" rel="prev">Calling FFTW from Modern Fortran</a>, Up: <a href="index.html" accesskey="u" rel="up">Top</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
</body>
</html>

View File

@@ -0,0 +1,108 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html>
<!-- This manual is for FFTW
(version 3.3.10, 10 December 2020).
Copyright (C) 2003 Matteo Frigo.
Copyright (C) 2003 Massachusetts Institute of Technology.
Permission is granted to make and distribute verbatim copies of this
manual provided the copyright notice and this permission notice are
preserved on all copies.
Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided that the
entire resulting derived work is distributed under the terms of a
permission notice identical to this one.
Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions,
except that this permission notice may be stated in a translation
approved by the Free Software Foundation. -->
<!-- Created by GNU Texinfo 6.7, http://www.gnu.org/software/texinfo/ -->
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<title>Calling FFTW from Modern Fortran (FFTW 3.3.10)</title>
<meta name="description" content="Calling FFTW from Modern Fortran (FFTW 3.3.10)">
<meta name="keywords" content="Calling FFTW from Modern Fortran (FFTW 3.3.10)">
<meta name="resource-type" content="document">
<meta name="distribution" content="global">
<meta name="Generator" content="makeinfo">
<link href="index.html" rel="start" title="Top">
<link href="Concept-Index.html" rel="index" title="Concept Index">
<link href="index.html#SEC_Contents" rel="contents" title="Table of Contents">
<link href="index.html" rel="up" title="Top">
<link href="Overview-of-Fortran-interface.html" rel="next" title="Overview of Fortran interface">
<link href="FFTW-MPI-Fortran-Interface.html" rel="prev" title="FFTW MPI Fortran Interface">
<style type="text/css">
<!--
a.summary-letter {text-decoration: none}
blockquote.indentedblock {margin-right: 0em}
div.display {margin-left: 3.2em}
div.example {margin-left: 3.2em}
div.lisp {margin-left: 3.2em}
kbd {font-style: oblique}
pre.display {font-family: inherit}
pre.format {font-family: inherit}
pre.menu-comment {font-family: serif}
pre.menu-preformatted {font-family: serif}
span.nolinebreak {white-space: nowrap}
span.roman {font-family: initial; font-weight: normal}
span.sansserif {font-family: sans-serif; font-weight: normal}
ul.no-bullet {list-style: none}
-->
</style>
</head>
<body lang="en">
<span id="Calling-FFTW-from-Modern-Fortran"></span><div class="header">
<p>
Next: <a href="Calling-FFTW-from-Legacy-Fortran.html" accesskey="n" rel="next">Calling FFTW from Legacy Fortran</a>, Previous: <a href="Distributed_002dmemory-FFTW-with-MPI.html" accesskey="p" rel="prev">Distributed-memory FFTW with MPI</a>, Up: <a href="index.html" accesskey="u" rel="up">Top</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
<hr>
<span id="Calling-FFTW-from-Modern-Fortran-1"></span><h2 class="chapter">7 Calling FFTW from Modern Fortran</h2>
<span id="index-Fortran-interface-2"></span>
<p>Fortran 2003 standardized ways for Fortran code to call C libraries,
and this allows us to support a direct translation of the FFTW C API
into Fortran. Compared to the legacy Fortran 77 interface
(see <a href="Calling-FFTW-from-Legacy-Fortran.html">Calling FFTW from Legacy Fortran</a>), this direct interface
offers many advantages, especially compile-time type-checking and
aligned memory allocation. As of this writing, support for these C
interoperability features seems widespread, having been implemented in
nearly all major Fortran compilers (e.g. GNU, Intel, IBM,
Oracle/Solaris, Portland Group, NAG).
<span id="index-portability-4"></span>
</p>
<p>This chapter documents that interface. For the most part, since this
interface allows Fortran to call the C interface directly, the usage
is identical to C translated to Fortran syntax. However, there are a
few subtle points such as memory allocation, wisdom, and data types
that deserve closer attention.
</p>
<table class="menu" border="0" cellspacing="0">
<tr><td align="left" valign="top">&bull; <a href="Overview-of-Fortran-interface.html" accesskey="1">Overview of Fortran interface</a></td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="Reversing-array-dimensions.html" accesskey="2">Reversing array dimensions</a></td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="FFTW-Fortran-type-reference.html" accesskey="3">FFTW Fortran type reference</a></td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="Plan-execution-in-Fortran.html" accesskey="4">Plan execution in Fortran</a></td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="Allocating-aligned-memory-in-Fortran.html" accesskey="5">Allocating aligned memory in Fortran</a></td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="Accessing-the-wisdom-API-from-Fortran.html" accesskey="6">Accessing the wisdom API from Fortran</a></td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="Defining-an-FFTW-module.html" accesskey="7">Defining an FFTW module</a></td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
</table>
</body>
</html>

View File

@@ -0,0 +1,128 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html>
<!-- This manual is for FFTW
(version 3.3.10, 10 December 2020).
Copyright (C) 2003 Matteo Frigo.
Copyright (C) 2003 Massachusetts Institute of Technology.
Permission is granted to make and distribute verbatim copies of this
manual provided the copyright notice and this permission notice are
preserved on all copies.
Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided that the
entire resulting derived work is distributed under the terms of a
permission notice identical to this one.
Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions,
except that this permission notice may be stated in a translation
approved by the Free Software Foundation. -->
<!-- Created by GNU Texinfo 6.7, http://www.gnu.org/software/texinfo/ -->
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<title>Caveats in Using Wisdom (FFTW 3.3.10)</title>
<meta name="description" content="Caveats in Using Wisdom (FFTW 3.3.10)">
<meta name="keywords" content="Caveats in Using Wisdom (FFTW 3.3.10)">
<meta name="resource-type" content="document">
<meta name="distribution" content="global">
<meta name="Generator" content="makeinfo">
<link href="index.html" rel="start" title="Top">
<link href="Concept-Index.html" rel="index" title="Concept Index">
<link href="index.html#SEC_Contents" rel="contents" title="Table of Contents">
<link href="Other-Important-Topics.html" rel="up" title="Other Important Topics">
<link href="FFTW-Reference.html" rel="next" title="FFTW Reference">
<link href="Words-of-Wisdom_002dSaving-Plans.html" rel="prev" title="Words of Wisdom-Saving Plans">
<style type="text/css">
<!--
a.summary-letter {text-decoration: none}
blockquote.indentedblock {margin-right: 0em}
div.display {margin-left: 3.2em}
div.example {margin-left: 3.2em}
div.lisp {margin-left: 3.2em}
kbd {font-style: oblique}
pre.display {font-family: inherit}
pre.format {font-family: inherit}
pre.menu-comment {font-family: serif}
pre.menu-preformatted {font-family: serif}
span.nolinebreak {white-space: nowrap}
span.roman {font-family: initial; font-weight: normal}
span.sansserif {font-family: sans-serif; font-weight: normal}
ul.no-bullet {list-style: none}
-->
</style>
</head>
<body lang="en">
<span id="Caveats-in-Using-Wisdom"></span><div class="header">
<p>
Previous: <a href="Words-of-Wisdom_002dSaving-Plans.html" accesskey="p" rel="prev">Words of Wisdom-Saving Plans</a>, Up: <a href="Other-Important-Topics.html" accesskey="u" rel="up">Other Important Topics</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
<hr>
<span id="Caveats-in-Using-Wisdom-1"></span><h3 class="section">3.4 Caveats in Using Wisdom</h3>
<span id="index-wisdom_002c-problems-with"></span>
<blockquote>
<i>
<p>For in much wisdom is much grief, and he that increaseth knowledge
increaseth sorrow.
</i>
[Ecclesiastes 1:18]
<span id="index-Ecclesiastes"></span>
</p></blockquote>
<span id="index-portability-1"></span>
<p>There are pitfalls to using wisdom, in that it can negate FFTW&rsquo;s
ability to adapt to changing hardware and other conditions. For
example, it would be perfectly possible to export wisdom from a
program running on one processor and import it into a program running
on another processor. Doing so, however, would mean that the second
program would use plans optimized for the first processor, instead of
the one it is running on.
</p>
<p>It should be safe to reuse wisdom as long as the hardware and program
binaries remain unchanged. (Actually, the optimal plan may change even
between runs of the same binary on identical hardware, due to
differences in the virtual memory environment, etcetera. Users
seriously interested in performance should worry about this problem,
too.) It is likely that, if the same wisdom is used for two
different program binaries, even running on the same machine, the
plans may be sub-optimal because of differing code alignments. It is
therefore wise to recreate wisdom every time an application is
recompiled. The more the underlying hardware and software changes
between the creation of wisdom and its use, the greater grows
the risk of sub-optimal plans.
</p>
<p>Nevertheless, if the choice is between using <code>FFTW_ESTIMATE</code> or
using possibly-suboptimal wisdom (created on the same machine, but for a
different binary), the wisdom is likely to be better. For this reason,
we provide a function to import wisdom from a standard system-wide
location (<code>/etc/fftw/wisdom</code> on Unix):
<span id="index-wisdom_002c-system_002dwide"></span>
</p>
<div class="example">
<pre class="example">int fftw_import_system_wisdom(void);
</pre></div>
<span id="index-fftw_005fimport_005fsystem_005fwisdom"></span>
<p>FFTW also provides a standalone program, <code>fftw-wisdom</code> (described
by its own <code>man</code> page on Unix) with which users can create wisdom,
e.g. for a canonical set of sizes to store in the system wisdom file.
See <a href="Wisdom-Utilities.html">Wisdom Utilities</a>.
<span id="index-fftw_002dwisdom-utility"></span>
</p>
<hr>
<div class="header">
<p>
Previous: <a href="Words-of-Wisdom_002dSaving-Plans.html" accesskey="p" rel="prev">Words of Wisdom-Saving Plans</a>, Up: <a href="Other-Important-Topics.html" accesskey="u" rel="up">Other Important Topics</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
</body>
</html>

View File

@@ -0,0 +1,91 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html>
<!-- This manual is for FFTW
(version 3.3.10, 10 December 2020).
Copyright (C) 2003 Matteo Frigo.
Copyright (C) 2003 Massachusetts Institute of Technology.
Permission is granted to make and distribute verbatim copies of this
manual provided the copyright notice and this permission notice are
preserved on all copies.
Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided that the
entire resulting derived work is distributed under the terms of a
permission notice identical to this one.
Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions,
except that this permission notice may be stated in a translation
approved by the Free Software Foundation. -->
<!-- Created by GNU Texinfo 6.7, http://www.gnu.org/software/texinfo/ -->
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<title>Column-major Format (FFTW 3.3.10)</title>
<meta name="description" content="Column-major Format (FFTW 3.3.10)">
<meta name="keywords" content="Column-major Format (FFTW 3.3.10)">
<meta name="resource-type" content="document">
<meta name="distribution" content="global">
<meta name="Generator" content="makeinfo">
<link href="index.html" rel="start" title="Top">
<link href="Concept-Index.html" rel="index" title="Concept Index">
<link href="index.html#SEC_Contents" rel="contents" title="Table of Contents">
<link href="Multi_002ddimensional-Array-Format.html" rel="up" title="Multi-dimensional Array Format">
<link href="Fixed_002dsize-Arrays-in-C.html" rel="next" title="Fixed-size Arrays in C">
<link href="Row_002dmajor-Format.html" rel="prev" title="Row-major Format">
<style type="text/css">
<!--
a.summary-letter {text-decoration: none}
blockquote.indentedblock {margin-right: 0em}
div.display {margin-left: 3.2em}
div.example {margin-left: 3.2em}
div.lisp {margin-left: 3.2em}
kbd {font-style: oblique}
pre.display {font-family: inherit}
pre.format {font-family: inherit}
pre.menu-comment {font-family: serif}
pre.menu-preformatted {font-family: serif}
span.nolinebreak {white-space: nowrap}
span.roman {font-family: initial; font-weight: normal}
span.sansserif {font-family: sans-serif; font-weight: normal}
ul.no-bullet {list-style: none}
-->
</style>
</head>
<body lang="en">
<span id="Column_002dmajor-Format"></span><div class="header">
<p>
Next: <a href="Fixed_002dsize-Arrays-in-C.html" accesskey="n" rel="next">Fixed-size Arrays in C</a>, Previous: <a href="Row_002dmajor-Format.html" accesskey="p" rel="prev">Row-major Format</a>, Up: <a href="Multi_002ddimensional-Array-Format.html" accesskey="u" rel="up">Multi-dimensional Array Format</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
<hr>
<span id="Column_002dmajor-Format-1"></span><h4 class="subsection">3.2.2 Column-major Format</h4>
<span id="index-column_002dmajor"></span>
<p>Readers from the Fortran world are used to arrays stored in
<em>column-major</em> order (sometimes called &ldquo;Fortran order&rdquo;). This is
essentially the exact opposite of row-major order in that, here, the
<em>first</em> dimension&rsquo;s index varies most quickly.
</p>
<p>If you have an array stored in column-major order and wish to
transform it using FFTW, it is quite easy to do. When creating the
plan, simply pass the dimensions of the array to the planner in
<em>reverse order</em>. For example, if your array is a rank three
<code>N x M x L</code> matrix in column-major order, you should pass the
dimensions of the array as if it were an <code>L x M x N</code> matrix
(which it is, from the perspective of FFTW). This is done for you
<em>automatically</em> by the FFTW legacy-Fortran interface
(see <a href="Calling-FFTW-from-Legacy-Fortran.html">Calling FFTW from Legacy Fortran</a>), but you must do it
manually with the modern Fortran interface (see <a href="Reversing-array-dimensions.html">Reversing array dimensions</a>).
<span id="index-Fortran-interface"></span>
</p>
</body>
</html>

View File

@@ -0,0 +1,158 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html>
<!-- This manual is for FFTW
(version 3.3.10, 10 December 2020).
Copyright (C) 2003 Matteo Frigo.
Copyright (C) 2003 Massachusetts Institute of Technology.
Permission is granted to make and distribute verbatim copies of this
manual provided the copyright notice and this permission notice are
preserved on all copies.
Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided that the
entire resulting derived work is distributed under the terms of a
permission notice identical to this one.
Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions,
except that this permission notice may be stated in a translation
approved by the Free Software Foundation. -->
<!-- Created by GNU Texinfo 6.7, http://www.gnu.org/software/texinfo/ -->
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<title>Combining MPI and Threads (FFTW 3.3.10)</title>
<meta name="description" content="Combining MPI and Threads (FFTW 3.3.10)">
<meta name="keywords" content="Combining MPI and Threads (FFTW 3.3.10)">
<meta name="resource-type" content="document">
<meta name="distribution" content="global">
<meta name="Generator" content="makeinfo">
<link href="index.html" rel="start" title="Top">
<link href="Concept-Index.html" rel="index" title="Concept Index">
<link href="index.html#SEC_Contents" rel="contents" title="Table of Contents">
<link href="Distributed_002dmemory-FFTW-with-MPI.html" rel="up" title="Distributed-memory FFTW with MPI">
<link href="FFTW-MPI-Reference.html" rel="next" title="FFTW MPI Reference">
<link href="FFTW-MPI-Performance-Tips.html" rel="prev" title="FFTW MPI Performance Tips">
<style type="text/css">
<!--
a.summary-letter {text-decoration: none}
blockquote.indentedblock {margin-right: 0em}
div.display {margin-left: 3.2em}
div.example {margin-left: 3.2em}
div.lisp {margin-left: 3.2em}
kbd {font-style: oblique}
pre.display {font-family: inherit}
pre.format {font-family: inherit}
pre.menu-comment {font-family: serif}
pre.menu-preformatted {font-family: serif}
span.nolinebreak {white-space: nowrap}
span.roman {font-family: initial; font-weight: normal}
span.sansserif {font-family: sans-serif; font-weight: normal}
ul.no-bullet {list-style: none}
-->
</style>
</head>
<body lang="en">
<span id="Combining-MPI-and-Threads"></span><div class="header">
<p>
Next: <a href="FFTW-MPI-Reference.html" accesskey="n" rel="next">FFTW MPI Reference</a>, Previous: <a href="FFTW-MPI-Performance-Tips.html" accesskey="p" rel="prev">FFTW MPI Performance Tips</a>, Up: <a href="Distributed_002dmemory-FFTW-with-MPI.html" accesskey="u" rel="up">Distributed-memory FFTW with MPI</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
<hr>
<span id="Combining-MPI-and-Threads-1"></span><h3 class="section">6.11 Combining MPI and Threads</h3>
<span id="index-threads-2"></span>
<p>In certain cases, it may be advantageous to combine MPI
(distributed-memory) and threads (shared-memory) parallelization.
FFTW supports this, with certain caveats. For example, if you have a
cluster of 4-processor shared-memory nodes, you may want to use
threads within the nodes and MPI between the nodes, instead of MPI for
all parallelization.
</p>
<p>In particular, it is possible to seamlessly combine the MPI FFTW
routines with the multi-threaded FFTW routines (see <a href="Multi_002dthreaded-FFTW.html">Multi-threaded FFTW</a>). However, some care must be taken in the initialization code,
which should look something like this:
</p>
<div class="example">
<pre class="example">int threads_ok;
int main(int argc, char **argv)
{
int provided;
MPI_Init_thread(&amp;argc, &amp;argv, MPI_THREAD_FUNNELED, &amp;provided);
threads_ok = provided &gt;= MPI_THREAD_FUNNELED;
if (threads_ok) threads_ok = fftw_init_threads();
fftw_mpi_init();
...
if (threads_ok) fftw_plan_with_nthreads(...);
...
MPI_Finalize();
}
</pre></div>
<span id="index-fftw_005fmpi_005finit-3"></span>
<span id="index-fftw_005finit_005fthreads-2"></span>
<span id="index-fftw_005fplan_005fwith_005fnthreads-1"></span>
<p>First, note that instead of calling <code>MPI_Init</code>, you should call
<code>MPI_Init_threads</code>, which is the initialization routine defined
by the MPI-2 standard to indicate to MPI that your program will be
multithreaded. We pass <code>MPI_THREAD_FUNNELED</code>, which indicates
that we will only call MPI routines from the main thread. (FFTW will
launch additional threads internally, but the extra threads will not
call MPI code.) (You may also pass <code>MPI_THREAD_SERIALIZED</code> or
<code>MPI_THREAD_MULTIPLE</code>, which requests additional multithreading
support from the MPI implementation, but this is not required by
FFTW.) The <code>provided</code> parameter returns what level of threads
support is actually supported by your MPI implementation; this
<em>must</em> be at least <code>MPI_THREAD_FUNNELED</code> if you want to call
the FFTW threads routines, so we define a global variable
<code>threads_ok</code> to record this. You should only call
<code>fftw_init_threads</code> or <code>fftw_plan_with_nthreads</code> if
<code>threads_ok</code> is true. For more information on thread safety in
MPI, see the
<a href="http://www.mpi-forum.org/docs/mpi-20-html/node162.htm">MPI and
Threads</a> section of the MPI-2 standard.
<span id="index-thread-safety-2"></span>
</p>
<p>Second, we must call <code>fftw_init_threads</code> <em>before</em>
<code>fftw_mpi_init</code>. This is critical for technical reasons having
to do with how FFTW initializes its list of algorithms.
</p>
<p>Then, if you call <code>fftw_plan_with_nthreads(N)</code>, <em>every</em> MPI
process will launch (up to) <code>N</code> threads to parallelize its transforms.
</p>
<p>For example, in the hypothetical cluster of 4-processor nodes, you
might wish to launch only a single MPI process per node, and then call
<code>fftw_plan_with_nthreads(4)</code> on each process to use all
processors in the nodes.
</p>
<p>This may or may not be faster than simply using as many MPI processes
as you have processors, however. On the one hand, using threads
within a node eliminates the need for explicit message passing within
the node. On the other hand, FFTW&rsquo;s transpose routines are not
multi-threaded, and this means that the communications that do take
place will not benefit from parallelization within the node.
Moreover, many MPI implementations already have optimizations to
exploit shared memory when it is available, so adding the
multithreaded FFTW on top of this may be superfluous.
<span id="index-transpose-4"></span>
</p>
<hr>
<div class="header">
<p>
Next: <a href="FFTW-MPI-Reference.html" accesskey="n" rel="next">FFTW MPI Reference</a>, Previous: <a href="FFTW-MPI-Performance-Tips.html" accesskey="p" rel="prev">FFTW MPI Performance Tips</a>, Up: <a href="Distributed_002dmemory-FFTW-with-MPI.html" accesskey="u" rel="up">Distributed-memory FFTW with MPI</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
</body>
</html>

View File

@@ -0,0 +1,170 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html>
<!-- This manual is for FFTW
(version 3.3.10, 10 December 2020).
Copyright (C) 2003 Matteo Frigo.
Copyright (C) 2003 Massachusetts Institute of Technology.
Permission is granted to make and distribute verbatim copies of this
manual provided the copyright notice and this permission notice are
preserved on all copies.
Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided that the
entire resulting derived work is distributed under the terms of a
permission notice identical to this one.
Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions,
except that this permission notice may be stated in a translation
approved by the Free Software Foundation. -->
<!-- Created by GNU Texinfo 6.7, http://www.gnu.org/software/texinfo/ -->
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<title>Complex DFTs (FFTW 3.3.10)</title>
<meta name="description" content="Complex DFTs (FFTW 3.3.10)">
<meta name="keywords" content="Complex DFTs (FFTW 3.3.10)">
<meta name="resource-type" content="document">
<meta name="distribution" content="global">
<meta name="Generator" content="makeinfo">
<link href="index.html" rel="start" title="Top">
<link href="Concept-Index.html" rel="index" title="Concept Index">
<link href="index.html#SEC_Contents" rel="contents" title="Table of Contents">
<link href="Basic-Interface.html" rel="up" title="Basic Interface">
<link href="Planner-Flags.html" rel="next" title="Planner Flags">
<link href="Basic-Interface.html" rel="prev" title="Basic Interface">
<style type="text/css">
<!--
a.summary-letter {text-decoration: none}
blockquote.indentedblock {margin-right: 0em}
div.display {margin-left: 3.2em}
div.example {margin-left: 3.2em}
div.lisp {margin-left: 3.2em}
kbd {font-style: oblique}
pre.display {font-family: inherit}
pre.format {font-family: inherit}
pre.menu-comment {font-family: serif}
pre.menu-preformatted {font-family: serif}
span.nolinebreak {white-space: nowrap}
span.roman {font-family: initial; font-weight: normal}
span.sansserif {font-family: sans-serif; font-weight: normal}
ul.no-bullet {list-style: none}
-->
</style>
</head>
<body lang="en">
<span id="Complex-DFTs"></span><div class="header">
<p>
Next: <a href="Planner-Flags.html" accesskey="n" rel="next">Planner Flags</a>, Previous: <a href="Basic-Interface.html" accesskey="p" rel="prev">Basic Interface</a>, Up: <a href="Basic-Interface.html" accesskey="u" rel="up">Basic Interface</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
<hr>
<span id="Complex-DFTs-1"></span><h4 class="subsection">4.3.1 Complex DFTs</h4>
<div class="example">
<pre class="example">fftw_plan fftw_plan_dft_1d(int n0,
fftw_complex *in, fftw_complex *out,
int sign, unsigned flags);
fftw_plan fftw_plan_dft_2d(int n0, int n1,
fftw_complex *in, fftw_complex *out,
int sign, unsigned flags);
fftw_plan fftw_plan_dft_3d(int n0, int n1, int n2,
fftw_complex *in, fftw_complex *out,
int sign, unsigned flags);
fftw_plan fftw_plan_dft(int rank, const int *n,
fftw_complex *in, fftw_complex *out,
int sign, unsigned flags);
</pre></div>
<span id="index-fftw_005fplan_005fdft_005f1d-1"></span>
<span id="index-fftw_005fplan_005fdft_005f2d-1"></span>
<span id="index-fftw_005fplan_005fdft_005f3d-1"></span>
<span id="index-fftw_005fplan_005fdft-1"></span>
<p>Plan a complex input/output discrete Fourier transform (DFT) in zero or
more dimensions, returning an <code>fftw_plan</code> (see <a href="Using-Plans.html">Using Plans</a>).
</p>
<p>Once you have created a plan for a certain transform type and
parameters, then creating another plan of the same type and parameters,
but for different arrays, is fast and shares constant data with the
first plan (if it still exists).
</p>
<p>The planner returns <code>NULL</code> if the plan cannot be created. In the
standard FFTW distribution, the basic interface is guaranteed to return
a non-<code>NULL</code> plan. A plan may be <code>NULL</code>, however, if you are
using a customized FFTW configuration supporting a restricted set of
transforms.
</p>
<span id="Arguments"></span><h4 class="subsubheading">Arguments</h4>
<ul>
<li> <code>rank</code> is the rank of the transform (it should be the size of the
array <code>*n</code>), and can be any non-negative integer. (See <a href="Complex-Multi_002dDimensional-DFTs.html">Complex Multi-Dimensional DFTs</a>, for the definition of &ldquo;rank&rdquo;.) The
&lsquo;<samp>_1d</samp>&rsquo;, &lsquo;<samp>_2d</samp>&rsquo;, and &lsquo;<samp>_3d</samp>&rsquo; planners correspond to a
<code>rank</code> of <code>1</code>, <code>2</code>, and <code>3</code>, respectively. The rank
may be zero, which is equivalent to a rank-1 transform of size 1, i.e. a
copy of one number from input to output.
</li><li> <code>n0</code>, <code>n1</code>, <code>n2</code>, or <code>n[0..rank-1]</code> (as appropriate
for each routine) specify the size of the transform dimensions. They
can be any positive integer.
<ul class="no-bullet">
<li>- <span id="index-row_002dmajor-1"></span>
Multi-dimensional arrays are stored in row-major order with dimensions:
<code>n0</code> x <code>n1</code>; or <code>n0</code> x <code>n1</code> x <code>n2</code>; or
<code>n[0]</code> x <code>n[1]</code> x ... x <code>n[rank-1]</code>.
See <a href="Multi_002ddimensional-Array-Format.html">Multi-dimensional Array Format</a>.
</li><li>- FFTW is best at handling sizes of the form
2<sup>a</sup> 3<sup>b</sup> 5<sup>c</sup> 7<sup>d</sup>
11<sup>e</sup> 13<sup>f</sup>,
where <em>e+f</em> is either <em>0</em> or <em>1</em>, and the other exponents
are arbitrary. Other sizes are computed by means of a slow,
general-purpose algorithm (which nevertheless retains <i>O</i>(<i>n</i>&nbsp;log&nbsp;<i>n</i>)
performance even for prime sizes). It is possible to customize FFTW
for different array sizes; see <a href="Installation-and-Customization.html">Installation and Customization</a>.
Transforms whose sizes are powers of <em>2</em> are especially fast.
</li></ul>
</li><li> <code>in</code> and <code>out</code> point to the input and output arrays of the
transform, which may be the same (yielding an in-place transform).
<span id="index-in_002dplace-2"></span>
These arrays are overwritten during planning, unless
<code>FFTW_ESTIMATE</code> is used in the flags. (The arrays need not be
initialized, but they must be allocated.)
<p>If <code>in == out</code>, the transform is <em>in-place</em> and the input
array is overwritten. If <code>in != out</code>, the two arrays must
not overlap (but FFTW does not check for this condition).
</p>
</li><li> <span id="index-FFTW_005fFORWARD-2"></span>
<span id="index-FFTW_005fBACKWARD-2"></span>
<code>sign</code> is the sign of the exponent in the formula that defines the
Fourier transform. It can be <em>-1</em> (= <code>FFTW_FORWARD</code>) or
<em>+1</em> (= <code>FFTW_BACKWARD</code>).
</li><li> <span id="index-flags-2"></span>
<code>flags</code> is a bitwise OR (&lsquo;<samp>|</samp>&rsquo;) of zero or more planner flags,
as defined in <a href="Planner-Flags.html">Planner Flags</a>.
</li></ul>
<p>FFTW computes an unnormalized transform: computing a forward followed by
a backward transform (or vice versa) will result in the original data
multiplied by the size of the transform (the product of the dimensions).
<span id="index-normalization-5"></span>
For more information, see <a href="What-FFTW-Really-Computes.html">What FFTW Really Computes</a>.
</p>
<hr>
<div class="header">
<p>
Next: <a href="Planner-Flags.html" accesskey="n" rel="next">Planner Flags</a>, Previous: <a href="Basic-Interface.html" accesskey="p" rel="prev">Basic Interface</a>, Up: <a href="Basic-Interface.html" accesskey="u" rel="up">Basic Interface</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
</body>
</html>

View File

@@ -0,0 +1,166 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html>
<!-- This manual is for FFTW
(version 3.3.10, 10 December 2020).
Copyright (C) 2003 Matteo Frigo.
Copyright (C) 2003 Massachusetts Institute of Technology.
Permission is granted to make and distribute verbatim copies of this
manual provided the copyright notice and this permission notice are
preserved on all copies.
Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided that the
entire resulting derived work is distributed under the terms of a
permission notice identical to this one.
Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions,
except that this permission notice may be stated in a translation
approved by the Free Software Foundation. -->
<!-- Created by GNU Texinfo 6.7, http://www.gnu.org/software/texinfo/ -->
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<title>Complex Multi-Dimensional DFTs (FFTW 3.3.10)</title>
<meta name="description" content="Complex Multi-Dimensional DFTs (FFTW 3.3.10)">
<meta name="keywords" content="Complex Multi-Dimensional DFTs (FFTW 3.3.10)">
<meta name="resource-type" content="document">
<meta name="distribution" content="global">
<meta name="Generator" content="makeinfo">
<link href="index.html" rel="start" title="Top">
<link href="Concept-Index.html" rel="index" title="Concept Index">
<link href="index.html#SEC_Contents" rel="contents" title="Table of Contents">
<link href="Tutorial.html" rel="up" title="Tutorial">
<link href="One_002dDimensional-DFTs-of-Real-Data.html" rel="next" title="One-Dimensional DFTs of Real Data">
<link href="Complex-One_002dDimensional-DFTs.html" rel="prev" title="Complex One-Dimensional DFTs">
<style type="text/css">
<!--
a.summary-letter {text-decoration: none}
blockquote.indentedblock {margin-right: 0em}
div.display {margin-left: 3.2em}
div.example {margin-left: 3.2em}
div.lisp {margin-left: 3.2em}
kbd {font-style: oblique}
pre.display {font-family: inherit}
pre.format {font-family: inherit}
pre.menu-comment {font-family: serif}
pre.menu-preformatted {font-family: serif}
span.nolinebreak {white-space: nowrap}
span.roman {font-family: initial; font-weight: normal}
span.sansserif {font-family: sans-serif; font-weight: normal}
ul.no-bullet {list-style: none}
-->
</style>
</head>
<body lang="en">
<span id="Complex-Multi_002dDimensional-DFTs"></span><div class="header">
<p>
Next: <a href="One_002dDimensional-DFTs-of-Real-Data.html" accesskey="n" rel="next">One-Dimensional DFTs of Real Data</a>, Previous: <a href="Complex-One_002dDimensional-DFTs.html" accesskey="p" rel="prev">Complex One-Dimensional DFTs</a>, Up: <a href="Tutorial.html" accesskey="u" rel="up">Tutorial</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
<hr>
<span id="Complex-Multi_002dDimensional-DFTs-1"></span><h3 class="section">2.2 Complex Multi-Dimensional DFTs</h3>
<p>Multi-dimensional transforms work much the same way as one-dimensional
transforms: you allocate arrays of <code>fftw_complex</code> (preferably
using <code>fftw_malloc</code>), create an <code>fftw_plan</code>, execute it as
many times as you want with <code>fftw_execute(plan)</code>, and clean up
with <code>fftw_destroy_plan(plan)</code> (and <code>fftw_free</code>).
</p>
<p>FFTW provides two routines for creating plans for 2d and 3d transforms,
and one routine for creating plans of arbitrary dimensionality.
The 2d and 3d routines have the following signature:
</p><div class="example">
<pre class="example">fftw_plan fftw_plan_dft_2d(int n0, int n1,
fftw_complex *in, fftw_complex *out,
int sign, unsigned flags);
fftw_plan fftw_plan_dft_3d(int n0, int n1, int n2,
fftw_complex *in, fftw_complex *out,
int sign, unsigned flags);
</pre></div>
<span id="index-fftw_005fplan_005fdft_005f2d"></span>
<span id="index-fftw_005fplan_005fdft_005f3d"></span>
<p>These routines create plans for <code>n0</code> by <code>n1</code> two-dimensional
(2d) transforms and <code>n0</code> by <code>n1</code> by <code>n2</code> 3d transforms,
respectively. All of these transforms operate on contiguous arrays in
the C-standard <em>row-major</em> order, so that the last dimension has the
fastest-varying index in the array. This layout is described further in
<a href="Multi_002ddimensional-Array-Format.html">Multi-dimensional Array Format</a>.
</p>
<p>FFTW can also compute transforms of higher dimensionality. In order to
avoid confusion between the various meanings of the the word
&ldquo;dimension&rdquo;, we use the term <em>rank</em>
<span id="index-rank"></span>
to denote the number of independent indices in an array.<a id="DOCF2" href="#FOOT2"><sup>2</sup></a> For
example, we say that a 2d transform has rank&nbsp;2, a 3d transform has
rank&nbsp;3, and so on. You can plan transforms of arbitrary rank by
means of the following function:
</p>
<div class="example">
<pre class="example">fftw_plan fftw_plan_dft(int rank, const int *n,
fftw_complex *in, fftw_complex *out,
int sign, unsigned flags);
</pre></div>
<span id="index-fftw_005fplan_005fdft"></span>
<p>Here, <code>n</code> is a pointer to an array <code>n[rank]</code> denoting an
<code>n[0]</code> by <code>n[1]</code> by &hellip; by <code>n[rank-1]</code> transform.
Thus, for example, the call
</p><div class="example">
<pre class="example">fftw_plan_dft_2d(n0, n1, in, out, sign, flags);
</pre></div>
<p>is equivalent to the following code fragment:
</p><div class="example">
<pre class="example">int n[2];
n[0] = n0;
n[1] = n1;
fftw_plan_dft(2, n, in, out, sign, flags);
</pre></div>
<p><code>fftw_plan_dft</code> is not restricted to 2d and 3d transforms,
however, but it can plan transforms of arbitrary rank.
</p>
<p>You may have noticed that all the planner routines described so far
have overlapping functionality. For example, you can plan a 1d or 2d
transform by using <code>fftw_plan_dft</code> with a <code>rank</code> of <code>1</code>
or <code>2</code>, or even by calling <code>fftw_plan_dft_3d</code> with <code>n0</code>
and/or <code>n1</code> equal to <code>1</code> (with no loss in efficiency). This
pattern continues, and FFTW&rsquo;s planning routines in general form a
&ldquo;partial order,&rdquo; sequences of
<span id="index-partial-order"></span>
interfaces with strictly increasing generality but correspondingly
greater complexity.
</p>
<p><code>fftw_plan_dft</code> is the most general complex-DFT routine that we
describe in this tutorial, but there are also the advanced and guru interfaces,
<span id="index-advanced-interface-1"></span>
<span id="index-guru-interface-1"></span>
which allow one to efficiently combine multiple/strided transforms
into a single FFTW plan, transform a subset of a larger
multi-dimensional array, and/or to handle more general complex-number
formats. For more information, see <a href="FFTW-Reference.html">FFTW Reference</a>.
</p>
<div class="footnote">
<hr>
<h4 class="footnotes-heading">Footnotes</h4>
<h5><a id="FOOT2" href="#DOCF2">(2)</a></h3>
<p>The
term &ldquo;rank&rdquo; is commonly used in the APL, FORTRAN, and Common Lisp
traditions, although it is not so common in the C&nbsp;world.</p>
</div>
<hr>
<div class="header">
<p>
Next: <a href="One_002dDimensional-DFTs-of-Real-Data.html" accesskey="n" rel="next">One-Dimensional DFTs of Real Data</a>, Previous: <a href="Complex-One_002dDimensional-DFTs.html" accesskey="p" rel="prev">Complex One-Dimensional DFTs</a>, Up: <a href="Tutorial.html" accesskey="u" rel="up">Tutorial</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
</body>
</html>

View File

@@ -0,0 +1,239 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html>
<!-- This manual is for FFTW
(version 3.3.10, 10 December 2020).
Copyright (C) 2003 Matteo Frigo.
Copyright (C) 2003 Massachusetts Institute of Technology.
Permission is granted to make and distribute verbatim copies of this
manual provided the copyright notice and this permission notice are
preserved on all copies.
Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided that the
entire resulting derived work is distributed under the terms of a
permission notice identical to this one.
Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions,
except that this permission notice may be stated in a translation
approved by the Free Software Foundation. -->
<!-- Created by GNU Texinfo 6.7, http://www.gnu.org/software/texinfo/ -->
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<title>Complex One-Dimensional DFTs (FFTW 3.3.10)</title>
<meta name="description" content="Complex One-Dimensional DFTs (FFTW 3.3.10)">
<meta name="keywords" content="Complex One-Dimensional DFTs (FFTW 3.3.10)">
<meta name="resource-type" content="document">
<meta name="distribution" content="global">
<meta name="Generator" content="makeinfo">
<link href="index.html" rel="start" title="Top">
<link href="Concept-Index.html" rel="index" title="Concept Index">
<link href="index.html#SEC_Contents" rel="contents" title="Table of Contents">
<link href="Tutorial.html" rel="up" title="Tutorial">
<link href="Complex-Multi_002dDimensional-DFTs.html" rel="next" title="Complex Multi-Dimensional DFTs">
<link href="Tutorial.html" rel="prev" title="Tutorial">
<style type="text/css">
<!--
a.summary-letter {text-decoration: none}
blockquote.indentedblock {margin-right: 0em}
div.display {margin-left: 3.2em}
div.example {margin-left: 3.2em}
div.lisp {margin-left: 3.2em}
kbd {font-style: oblique}
pre.display {font-family: inherit}
pre.format {font-family: inherit}
pre.menu-comment {font-family: serif}
pre.menu-preformatted {font-family: serif}
span.nolinebreak {white-space: nowrap}
span.roman {font-family: initial; font-weight: normal}
span.sansserif {font-family: sans-serif; font-weight: normal}
ul.no-bullet {list-style: none}
-->
</style>
</head>
<body lang="en">
<span id="Complex-One_002dDimensional-DFTs"></span><div class="header">
<p>
Next: <a href="Complex-Multi_002dDimensional-DFTs.html" accesskey="n" rel="next">Complex Multi-Dimensional DFTs</a>, Previous: <a href="Tutorial.html" accesskey="p" rel="prev">Tutorial</a>, Up: <a href="Tutorial.html" accesskey="u" rel="up">Tutorial</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
<hr>
<span id="Complex-One_002dDimensional-DFTs-1"></span><h3 class="section">2.1 Complex One-Dimensional DFTs</h3>
<blockquote>
<p>Plan: To bother about the best method of accomplishing an accidental result.
[Ambrose Bierce, <cite>The Enlarged Devil&rsquo;s Dictionary</cite>.]
<span id="index-Devil"></span>
</p></blockquote>
<p>The basic usage of FFTW to compute a one-dimensional DFT of size
<code>N</code> is simple, and it typically looks something like this code:
</p>
<div class="example">
<pre class="example">#include &lt;fftw3.h&gt;
...
{
fftw_complex *in, *out;
fftw_plan p;
...
in = (fftw_complex*) fftw_malloc(sizeof(fftw_complex) * N);
out = (fftw_complex*) fftw_malloc(sizeof(fftw_complex) * N);
p = fftw_plan_dft_1d(N, in, out, FFTW_FORWARD, FFTW_ESTIMATE);
...
fftw_execute(p); /* <span class="roman">repeat as needed</span> */
...
fftw_destroy_plan(p);
fftw_free(in); fftw_free(out);
}
</pre></div>
<p>You must link this code with the <code>fftw3</code> library. On Unix systems,
link with <code>-lfftw3 -lm</code>.
</p>
<p>The example code first allocates the input and output arrays. You can
allocate them in any way that you like, but we recommend using
<code>fftw_malloc</code>, which behaves like
<span id="index-fftw_005fmalloc"></span>
<code>malloc</code> except that it properly aligns the array when SIMD
instructions (such as SSE and Altivec) are available (see <a href="SIMD-alignment-and-fftw_005fmalloc.html">SIMD alignment and fftw_malloc</a>). [Alternatively, we provide a convenient wrapper function <code>fftw_alloc_complex(N)</code> which has the same effect.]
<span id="index-fftw_005falloc_005fcomplex"></span>
<span id="index-SIMD"></span>
</p>
<p>The data is an array of type <code>fftw_complex</code>, which is by default a
<code>double[2]</code> composed of the real (<code>in[i][0]</code>) and imaginary
(<code>in[i][1]</code>) parts of a complex number.
<span id="index-fftw_005fcomplex"></span>
</p>
<p>The next step is to create a <em>plan</em>, which is an object
<span id="index-plan-1"></span>
that contains all the data that FFTW needs to compute the FFT.
This function creates the plan:
</p>
<div class="example">
<pre class="example">fftw_plan fftw_plan_dft_1d(int n, fftw_complex *in, fftw_complex *out,
int sign, unsigned flags);
</pre></div>
<span id="index-fftw_005fplan_005fdft_005f1d"></span>
<span id="index-fftw_005fplan"></span>
<p>The first argument, <code>n</code>, is the size of the transform you are
trying to compute. The size <code>n</code> can be any positive integer, but
sizes that are products of small factors are transformed most
efficiently (although prime sizes still use an <i>O</i>(<i>n</i>&nbsp;log&nbsp;<i>n</i>)
algorithm).
</p>
<p>The next two arguments are pointers to the input and output arrays of
the transform. These pointers can be equal, indicating an
<em>in-place</em> transform.
<span id="index-in_002dplace"></span>
</p>
<p>The fourth argument, <code>sign</code>, can be either <code>FFTW_FORWARD</code>
(<code>-1</code>) or <code>FFTW_BACKWARD</code> (<code>+1</code>),
<span id="index-FFTW_005fFORWARD"></span>
<span id="index-FFTW_005fBACKWARD"></span>
and indicates the direction of the transform you are interested in;
technically, it is the sign of the exponent in the transform.
</p>
<p>The <code>flags</code> argument is usually either <code>FFTW_MEASURE</code> or
<span id="index-flags"></span>
<code>FFTW_ESTIMATE</code>. <code>FFTW_MEASURE</code> instructs FFTW to run
<span id="index-FFTW_005fMEASURE"></span>
and measure the execution time of several FFTs in order to find the
best way to compute the transform of size <code>n</code>. This process takes
some time (usually a few seconds), depending on your machine and on
the size of the transform. <code>FFTW_ESTIMATE</code>, on the contrary,
does not run any computation and just builds a
<span id="index-FFTW_005fESTIMATE"></span>
reasonable plan that is probably sub-optimal. In short, if your
program performs many transforms of the same size and initialization
time is not important, use <code>FFTW_MEASURE</code>; otherwise use the
estimate.
</p>
<p><em>You must create the plan before initializing the input</em>, because
<code>FFTW_MEASURE</code> overwrites the <code>in</code>/<code>out</code> arrays.
(Technically, <code>FFTW_ESTIMATE</code> does not touch your arrays, but you
should always create plans first just to be sure.)
</p>
<p>Once the plan has been created, you can use it as many times as you
like for transforms on the specified <code>in</code>/<code>out</code> arrays,
computing the actual transforms via <code>fftw_execute(plan)</code>:
</p><div class="example">
<pre class="example">void fftw_execute(const fftw_plan plan);
</pre></div>
<span id="index-fftw_005fexecute"></span>
<p>The DFT results are stored in-order in the array <code>out</code>, with the
zero-frequency (DC) component in <code>out[0]</code>.
<span id="index-frequency"></span>
If <code>in != out</code>, the transform is <em>out-of-place</em> and the input
array <code>in</code> is not modified. Otherwise, the input array is
overwritten with the transform.
</p>
<span id="index-execute-1"></span>
<p>If you want to transform a <em>different</em> array of the same size, you
can create a new plan with <code>fftw_plan_dft_1d</code> and FFTW
automatically reuses the information from the previous plan, if
possible. Alternatively, with the &ldquo;guru&rdquo; interface you can apply a
given plan to a different array, if you are careful.
See <a href="FFTW-Reference.html">FFTW Reference</a>.
</p>
<p>When you are done with the plan, you deallocate it by calling
<code>fftw_destroy_plan(plan)</code>:
</p><div class="example">
<pre class="example">void fftw_destroy_plan(fftw_plan plan);
</pre></div>
<span id="index-fftw_005fdestroy_005fplan"></span>
<p>If you allocate an array with <code>fftw_malloc()</code> you must deallocate
it with <code>fftw_free()</code>. Do not use <code>free()</code> or, heaven
forbid, <code>delete</code>.
<span id="index-fftw_005ffree"></span>
</p>
<p>FFTW computes an <em>unnormalized</em> DFT. Thus, computing a forward
followed by a backward transform (or vice versa) results in the original
array scaled by <code>n</code>. For the definition of the DFT, see <a href="What-FFTW-Really-Computes.html">What FFTW Really Computes</a>.
<span id="index-DFT-1"></span>
<span id="index-normalization"></span>
</p>
<p>If you have a C compiler, such as <code>gcc</code>, that supports the
C99 standard, and you <code>#include &lt;complex.h&gt;</code> <em>before</em>
<code>&lt;fftw3.h&gt;</code>, then <code>fftw_complex</code> is the native
double-precision complex type and you can manipulate it with ordinary
arithmetic. Otherwise, FFTW defines its own complex type, which is
bit-compatible with the C99 complex type. See <a href="Complex-numbers.html">Complex numbers</a>.
(The C++ <code>&lt;complex&gt;</code> template class may also be usable via a
typecast.)
<span id="index-C_002b_002b"></span>
</p>
<p>To use single or long-double precision versions of FFTW, replace the
<code>fftw_</code> prefix by <code>fftwf_</code> or <code>fftwl_</code> and link with
<code>-lfftw3f</code> or <code>-lfftw3l</code>, but use the <em>same</em>
<code>&lt;fftw3.h&gt;</code> header file.
<span id="index-precision"></span>
</p>
<p>Many more flags exist besides <code>FFTW_MEASURE</code> and
<code>FFTW_ESTIMATE</code>. For example, use <code>FFTW_PATIENT</code> if you&rsquo;re
willing to wait even longer for a possibly even faster plan (see <a href="FFTW-Reference.html">FFTW Reference</a>).
<span id="index-FFTW_005fPATIENT"></span>
You can also save plans for future use, as described by <a href="Words-of-Wisdom_002dSaving-Plans.html">Words of Wisdom-Saving Plans</a>.
</p>
<hr>
<div class="header">
<p>
Next: <a href="Complex-Multi_002dDimensional-DFTs.html" accesskey="n" rel="next">Complex Multi-Dimensional DFTs</a>, Previous: <a href="Tutorial.html" accesskey="p" rel="prev">Tutorial</a>, Up: <a href="Tutorial.html" accesskey="u" rel="up">Tutorial</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
</body>
</html>

View File

@@ -0,0 +1,114 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html>
<!-- This manual is for FFTW
(version 3.3.10, 10 December 2020).
Copyright (C) 2003 Matteo Frigo.
Copyright (C) 2003 Massachusetts Institute of Technology.
Permission is granted to make and distribute verbatim copies of this
manual provided the copyright notice and this permission notice are
preserved on all copies.
Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided that the
entire resulting derived work is distributed under the terms of a
permission notice identical to this one.
Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions,
except that this permission notice may be stated in a translation
approved by the Free Software Foundation. -->
<!-- Created by GNU Texinfo 6.7, http://www.gnu.org/software/texinfo/ -->
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<title>Complex numbers (FFTW 3.3.10)</title>
<meta name="description" content="Complex numbers (FFTW 3.3.10)">
<meta name="keywords" content="Complex numbers (FFTW 3.3.10)">
<meta name="resource-type" content="document">
<meta name="distribution" content="global">
<meta name="Generator" content="makeinfo">
<link href="index.html" rel="start" title="Top">
<link href="Concept-Index.html" rel="index" title="Concept Index">
<link href="index.html#SEC_Contents" rel="contents" title="Table of Contents">
<link href="Data-Types-and-Files.html" rel="up" title="Data Types and Files">
<link href="Precision.html" rel="next" title="Precision">
<link href="Data-Types-and-Files.html" rel="prev" title="Data Types and Files">
<style type="text/css">
<!--
a.summary-letter {text-decoration: none}
blockquote.indentedblock {margin-right: 0em}
div.display {margin-left: 3.2em}
div.example {margin-left: 3.2em}
div.lisp {margin-left: 3.2em}
kbd {font-style: oblique}
pre.display {font-family: inherit}
pre.format {font-family: inherit}
pre.menu-comment {font-family: serif}
pre.menu-preformatted {font-family: serif}
span.nolinebreak {white-space: nowrap}
span.roman {font-family: initial; font-weight: normal}
span.sansserif {font-family: sans-serif; font-weight: normal}
ul.no-bullet {list-style: none}
-->
</style>
</head>
<body lang="en">
<span id="Complex-numbers"></span><div class="header">
<p>
Next: <a href="Precision.html" accesskey="n" rel="next">Precision</a>, Previous: <a href="Data-Types-and-Files.html" accesskey="p" rel="prev">Data Types and Files</a>, Up: <a href="Data-Types-and-Files.html" accesskey="u" rel="up">Data Types and Files</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
<hr>
<span id="Complex-numbers-1"></span><h4 class="subsection">4.1.1 Complex numbers</h4>
<p>The default FFTW interface uses <code>double</code> precision for all
floating-point numbers, and defines a <code>fftw_complex</code> type to hold
complex numbers as:
</p>
<div class="example">
<pre class="example">typedef double fftw_complex[2];
</pre></div>
<span id="index-fftw_005fcomplex-1"></span>
<p>Here, the <code>[0]</code> element holds the real part and the <code>[1]</code>
element holds the imaginary part.
</p>
<p>Alternatively, if you have a C compiler (such as <code>gcc</code>) that
supports the C99 revision of the ANSI C standard, you can use C&rsquo;s new
native complex type (which is binary-compatible with the typedef above).
In particular, if you <code>#include &lt;complex.h&gt;</code> <em>before</em>
<code>&lt;fftw3.h&gt;</code>, then <code>fftw_complex</code> is defined to be the native
complex type and you can manipulate it with ordinary arithmetic
(e.g. <code>x = y * (3+4*I)</code>, where <code>x</code> and <code>y</code> are
<code>fftw_complex</code> and <code>I</code> is the standard symbol for the
imaginary unit);
<span id="index-C99-1"></span>
</p>
<p>C++ has its own <code>complex&lt;T&gt;</code> template class, defined in the
standard <code>&lt;complex&gt;</code> header file. Reportedly, the C++ standards
committee has recently agreed to mandate that the storage format used
for this type be binary-compatible with the C99 type, i.e. an array
<code>T[2]</code> with consecutive real <code>[0]</code> and imaginary <code>[1]</code>
parts. (See report
<a href="http://www.open-std.org/jtc1/sc22/WG21/docs/papers/2002/n1388.pdf
WG21/N1388">http://www.open-std.org/jtc1/sc22/WG21/docs/papers/2002/n1388.pdf
WG21/N1388</a>.) Although not part of the official standard as of this
writing, the proposal stated that: &ldquo;This solution has been tested with
all current major implementations of the standard library and shown to
be working.&rdquo; To the extent that this is true, if you have a variable
<code>complex&lt;double&gt; *x</code>, you can pass it directly to FFTW via
<code>reinterpret_cast&lt;fftw_complex*&gt;(x)</code>.
<span id="index-C_002b_002b-3"></span>
<span id="index-portability-2"></span>
</p>
</body>
</html>

View File

@@ -0,0 +1,517 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html>
<!-- This manual is for FFTW
(version 3.3.10, 10 December 2020).
Copyright (C) 2003 Matteo Frigo.
Copyright (C) 2003 Massachusetts Institute of Technology.
Permission is granted to make and distribute verbatim copies of this
manual provided the copyright notice and this permission notice are
preserved on all copies.
Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided that the
entire resulting derived work is distributed under the terms of a
permission notice identical to this one.
Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions,
except that this permission notice may be stated in a translation
approved by the Free Software Foundation. -->
<!-- Created by GNU Texinfo 6.7, http://www.gnu.org/software/texinfo/ -->
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<title>Concept Index (FFTW 3.3.10)</title>
<meta name="description" content="Concept Index (FFTW 3.3.10)">
<meta name="keywords" content="Concept Index (FFTW 3.3.10)">
<meta name="resource-type" content="document">
<meta name="distribution" content="global">
<meta name="Generator" content="makeinfo">
<link href="index.html" rel="start" title="Top">
<link href="#Concept-Index" rel="index" title="Concept Index">
<link href="index.html#SEC_Contents" rel="contents" title="Table of Contents">
<link href="index.html" rel="up" title="Top">
<link href="Library-Index.html" rel="next" title="Library Index">
<link href="License-and-Copyright.html" rel="prev" title="License and Copyright">
<style type="text/css">
<!--
a.summary-letter {text-decoration: none}
blockquote.indentedblock {margin-right: 0em}
div.display {margin-left: 3.2em}
div.example {margin-left: 3.2em}
div.lisp {margin-left: 3.2em}
kbd {font-style: oblique}
pre.display {font-family: inherit}
pre.format {font-family: inherit}
pre.menu-comment {font-family: serif}
pre.menu-preformatted {font-family: serif}
span.nolinebreak {white-space: nowrap}
span.roman {font-family: initial; font-weight: normal}
span.sansserif {font-family: sans-serif; font-weight: normal}
ul.no-bullet {list-style: none}
-->
</style>
</head>
<body lang="en">
<span id="Concept-Index"></span><div class="header">
<p>
Next: <a href="Library-Index.html" accesskey="n" rel="next">Library Index</a>, Previous: <a href="License-and-Copyright.html" accesskey="p" rel="prev">License and Copyright</a>, Up: <a href="index.html" accesskey="u" rel="up">Top</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Concept-Index" title="Index" rel="index">Index</a>]</p>
</div>
<hr>
<span id="Concept-Index-1"></span><h2 class="chapter">13 Concept Index</h2>
<table><tr><th valign="top">Jump to: &nbsp; </th><td><a class="summary-letter" href="#Concept-Index_cp_symbol-1"><b>6</b></a>
&nbsp;
<br>
<a class="summary-letter" href="#Concept-Index_cp_letter-A"><b>A</b></a>
&nbsp;
<a class="summary-letter" href="#Concept-Index_cp_letter-B"><b>B</b></a>
&nbsp;
<a class="summary-letter" href="#Concept-Index_cp_letter-C"><b>C</b></a>
&nbsp;
<a class="summary-letter" href="#Concept-Index_cp_letter-D"><b>D</b></a>
&nbsp;
<a class="summary-letter" href="#Concept-Index_cp_letter-E"><b>E</b></a>
&nbsp;
<a class="summary-letter" href="#Concept-Index_cp_letter-F"><b>F</b></a>
&nbsp;
<a class="summary-letter" href="#Concept-Index_cp_letter-G"><b>G</b></a>
&nbsp;
<a class="summary-letter" href="#Concept-Index_cp_letter-H"><b>H</b></a>
&nbsp;
<a class="summary-letter" href="#Concept-Index_cp_letter-I"><b>I</b></a>
&nbsp;
<a class="summary-letter" href="#Concept-Index_cp_letter-K"><b>K</b></a>
&nbsp;
<a class="summary-letter" href="#Concept-Index_cp_letter-L"><b>L</b></a>
&nbsp;
<a class="summary-letter" href="#Concept-Index_cp_letter-M"><b>M</b></a>
&nbsp;
<a class="summary-letter" href="#Concept-Index_cp_letter-N"><b>N</b></a>
&nbsp;
<a class="summary-letter" href="#Concept-Index_cp_letter-O"><b>O</b></a>
&nbsp;
<a class="summary-letter" href="#Concept-Index_cp_letter-P"><b>P</b></a>
&nbsp;
<a class="summary-letter" href="#Concept-Index_cp_letter-R"><b>R</b></a>
&nbsp;
<a class="summary-letter" href="#Concept-Index_cp_letter-S"><b>S</b></a>
&nbsp;
<a class="summary-letter" href="#Concept-Index_cp_letter-T"><b>T</b></a>
&nbsp;
<a class="summary-letter" href="#Concept-Index_cp_letter-V"><b>V</b></a>
&nbsp;
<a class="summary-letter" href="#Concept-Index_cp_letter-W"><b>W</b></a>
&nbsp;
</td></tr></table>
<table class="index-cp" border="0">
<tr><td></td><th align="left">Index Entry</th><td>&nbsp;</td><th align="left"> Section</th></tr>
<tr><td colspan="4"> <hr></td></tr>
<tr><th id="Concept-Index_cp_symbol-1">6</th><td></td><td></td></tr>
<tr><td></td><td valign="top"><a href="64_002dbit-Guru-Interface.html#index-64_002dbit-architecture">64-bit architecture</a>:</td><td>&nbsp;</td><td valign="top"><a href="64_002dbit-Guru-Interface.html">64-bit Guru Interface</a></td></tr>
<tr><td></td><td valign="top"><a href="2d-MPI-example.html#index-64_002dbit-architecture-1">64-bit architecture</a>:</td><td>&nbsp;</td><td valign="top"><a href="2d-MPI-example.html">2d MPI example</a></td></tr>
<tr><td></td><td valign="top"><a href="FFTW-Fortran-type-reference.html#index-64_002dbit-architecture-2">64-bit architecture</a>:</td><td>&nbsp;</td><td valign="top"><a href="FFTW-Fortran-type-reference.html">FFTW Fortran type reference</a></td></tr>
<tr><td colspan="4"> <hr></td></tr>
<tr><th id="Concept-Index_cp_letter-A">A</th><td></td><td></td></tr>
<tr><td></td><td valign="top"><a href="Introduction.html#index-advanced-interface">advanced interface</a>:</td><td>&nbsp;</td><td valign="top"><a href="Introduction.html">Introduction</a></td></tr>
<tr><td></td><td valign="top"><a href="Complex-Multi_002dDimensional-DFTs.html#index-advanced-interface-1">advanced interface</a>:</td><td>&nbsp;</td><td valign="top"><a href="Complex-Multi_002dDimensional-DFTs.html">Complex Multi-Dimensional DFTs</a></td></tr>
<tr><td></td><td valign="top"><a href="Row_002dmajor-Format.html#index-advanced-interface-2">advanced interface</a>:</td><td>&nbsp;</td><td valign="top"><a href="Row_002dmajor-Format.html">Row-major Format</a></td></tr>
<tr><td></td><td valign="top"><a href="Advanced-Interface.html#index-advanced-interface-3">advanced interface</a>:</td><td>&nbsp;</td><td valign="top"><a href="Advanced-Interface.html">Advanced Interface</a></td></tr>
<tr><td></td><td valign="top"><a href="Basic-and-advanced-distribution-interfaces.html#index-advanced-interface-4">advanced interface</a>:</td><td>&nbsp;</td><td valign="top"><a href="Basic-and-advanced-distribution-interfaces.html">Basic and advanced distribution interfaces</a></td></tr>
<tr><td></td><td valign="top"><a href="MPI-Data-Distribution-Functions.html#index-advanced-interface-5">advanced interface</a>:</td><td>&nbsp;</td><td valign="top"><a href="MPI-Data-Distribution-Functions.html">MPI Data Distribution Functions</a></td></tr>
<tr><td></td><td valign="top"><a href="MPI-Plan-Creation.html#index-advanced-interface-6">advanced interface</a>:</td><td>&nbsp;</td><td valign="top"><a href="MPI-Plan-Creation.html">MPI Plan Creation</a></td></tr>
<tr><td></td><td valign="top"><a href="Introduction.html#index-algorithm">algorithm</a>:</td><td>&nbsp;</td><td valign="top"><a href="Introduction.html">Introduction</a></td></tr>
<tr><td></td><td valign="top"><a href="Memory-Allocation.html#index-alignment">alignment</a>:</td><td>&nbsp;</td><td valign="top"><a href="Memory-Allocation.html">Memory Allocation</a></td></tr>
<tr><td></td><td valign="top"><a href="Planner-Flags.html#index-alignment-1">alignment</a>:</td><td>&nbsp;</td><td valign="top"><a href="Planner-Flags.html">Planner Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="New_002darray-Execute-Functions.html#index-alignment-2">alignment</a>:</td><td>&nbsp;</td><td valign="top"><a href="New_002darray-Execute-Functions.html">New-array Execute Functions</a></td></tr>
<tr><td></td><td valign="top"><a href="Using-MPI-Plans.html#index-alignment-3">alignment</a>:</td><td>&nbsp;</td><td valign="top"><a href="Using-MPI-Plans.html">Using MPI Plans</a></td></tr>
<tr><td></td><td valign="top"><a href="Overview-of-Fortran-interface.html#index-alignment-4">alignment</a>:</td><td>&nbsp;</td><td valign="top"><a href="Overview-of-Fortran-interface.html">Overview of Fortran interface</a></td></tr>
<tr><td></td><td valign="top"><a href="Allocating-aligned-memory-in-Fortran.html#index-alignment-5">alignment</a>:</td><td>&nbsp;</td><td valign="top"><a href="Allocating-aligned-memory-in-Fortran.html">Allocating aligned memory in Fortran</a></td></tr>
<tr><td></td><td valign="top"><a href="SIMD-alignment-and-fftw_005fmalloc.html#index-AltiVec">AltiVec</a>:</td><td>&nbsp;</td><td valign="top"><a href="SIMD-alignment-and-fftw_005fmalloc.html">SIMD alignment and fftw_malloc</a></td></tr>
<tr><td></td><td valign="top"><a href="SIMD-alignment-and-fftw_005fmalloc.html#index-AVX">AVX</a>:</td><td>&nbsp;</td><td valign="top"><a href="SIMD-alignment-and-fftw_005fmalloc.html">SIMD alignment and fftw_malloc</a></td></tr>
<tr><td></td><td valign="top"><a href="SIMD-alignment-and-fftw_005fmalloc.html#index-AVX2">AVX2</a>:</td><td>&nbsp;</td><td valign="top"><a href="SIMD-alignment-and-fftw_005fmalloc.html">SIMD alignment and fftw_malloc</a></td></tr>
<tr><td></td><td valign="top"><a href="SIMD-alignment-and-fftw_005fmalloc.html#index-AVX512">AVX512</a>:</td><td>&nbsp;</td><td valign="top"><a href="SIMD-alignment-and-fftw_005fmalloc.html">SIMD alignment and fftw_malloc</a></td></tr>
<tr><td colspan="4"> <hr></td></tr>
<tr><th id="Concept-Index_cp_letter-B">B</th><td></td><td></td></tr>
<tr><td></td><td valign="top"><a href="Introduction.html#index-basic-interface">basic interface</a>:</td><td>&nbsp;</td><td valign="top"><a href="Introduction.html">Introduction</a></td></tr>
<tr><td></td><td valign="top"><a href="Tutorial.html#index-basic-interface-1">basic interface</a>:</td><td>&nbsp;</td><td valign="top"><a href="Tutorial.html">Tutorial</a></td></tr>
<tr><td></td><td valign="top"><a href="Basic-Interface.html#index-basic-interface-2">basic interface</a>:</td><td>&nbsp;</td><td valign="top"><a href="Basic-Interface.html">Basic Interface</a></td></tr>
<tr><td></td><td valign="top"><a href="MPI-Data-Distribution.html#index-block-distribution">block distribution</a>:</td><td>&nbsp;</td><td valign="top"><a href="MPI-Data-Distribution.html">MPI Data Distribution</a></td></tr>
<tr><td></td><td valign="top"><a href="Basic-and-advanced-distribution-interfaces.html#index-block-distribution-1">block distribution</a>:</td><td>&nbsp;</td><td valign="top"><a href="Basic-and-advanced-distribution-interfaces.html">Basic and advanced distribution interfaces</a></td></tr>
<tr><td></td><td valign="top"><a href="FFTW-MPI-Performance-Tips.html#index-block-distribution-2">block distribution</a>:</td><td>&nbsp;</td><td valign="top"><a href="FFTW-MPI-Performance-Tips.html">FFTW MPI Performance Tips</a></td></tr>
<tr><td colspan="4"> <hr></td></tr>
<tr><th id="Concept-Index_cp_letter-C">C</th><td></td><td></td></tr>
<tr><td></td><td valign="top"><a href="Fixed_002dsize-Arrays-in-C.html#index-C-multi_002ddimensional-arrays">C multi-dimensional arrays</a>:</td><td>&nbsp;</td><td valign="top"><a href="Fixed_002dsize-Arrays-in-C.html">Fixed-size Arrays in C</a></td></tr>
<tr><td></td><td valign="top"><a href="Complex-One_002dDimensional-DFTs.html#index-C_002b_002b">C++</a>:</td><td>&nbsp;</td><td valign="top"><a href="Complex-One_002dDimensional-DFTs.html">Complex One-Dimensional DFTs</a></td></tr>
<tr><td></td><td valign="top"><a href="SIMD-alignment-and-fftw_005fmalloc.html#index-C_002b_002b-1">C++</a>:</td><td>&nbsp;</td><td valign="top"><a href="SIMD-alignment-and-fftw_005fmalloc.html">SIMD alignment and fftw_malloc</a></td></tr>
<tr><td></td><td valign="top"><a href="Dynamic-Arrays-in-C.html#index-C_002b_002b-2">C++</a>:</td><td>&nbsp;</td><td valign="top"><a href="Dynamic-Arrays-in-C.html">Dynamic Arrays in C</a></td></tr>
<tr><td></td><td valign="top"><a href="Complex-numbers.html#index-C_002b_002b-3">C++</a>:</td><td>&nbsp;</td><td valign="top"><a href="Complex-numbers.html">Complex numbers</a></td></tr>
<tr><td></td><td valign="top"><a href="Memory-Allocation.html#index-C_002b_002b-4">C++</a>:</td><td>&nbsp;</td><td valign="top"><a href="Memory-Allocation.html">Memory Allocation</a></td></tr>
<tr><td></td><td valign="top"><a href="One_002dDimensional-DFTs-of-Real-Data.html#index-c2r">c2r</a>:</td><td>&nbsp;</td><td valign="top"><a href="One_002dDimensional-DFTs-of-Real-Data.html">One-Dimensional DFTs of Real Data</a></td></tr>
<tr><td></td><td valign="top"><a href="Planner-Flags.html#index-c2r-1">c2r</a>:</td><td>&nbsp;</td><td valign="top"><a href="Planner-Flags.html">Planner Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="Real_002ddata-DFTs.html#index-c2r-2">c2r</a>:</td><td>&nbsp;</td><td valign="top"><a href="Real_002ddata-DFTs.html">Real-data DFTs</a></td></tr>
<tr><td></td><td valign="top"><a href="Dynamic-Arrays-in-C.html#index-C99">C99</a>:</td><td>&nbsp;</td><td valign="top"><a href="Dynamic-Arrays-in-C.html">Dynamic Arrays in C</a></td></tr>
<tr><td></td><td valign="top"><a href="Complex-numbers.html#index-C99-1">C99</a>:</td><td>&nbsp;</td><td valign="top"><a href="Complex-numbers.html">Complex numbers</a></td></tr>
<tr><td></td><td valign="top"><a href="Precision.html#index-C99-2">C99</a>:</td><td>&nbsp;</td><td valign="top"><a href="Precision.html">Precision</a></td></tr>
<tr><td></td><td valign="top"><a href="Generating-your-own-code.html#index-Caml">Caml</a>:</td><td>&nbsp;</td><td valign="top"><a href="Generating-your-own-code.html">Generating your own code</a></td></tr>
<tr><td></td><td valign="top"><a href="Acknowledgments.html#index-Caml-1">Caml</a>:</td><td>&nbsp;</td><td valign="top"><a href="Acknowledgments.html">Acknowledgments</a></td></tr>
<tr><td></td><td valign="top"><a href="Introduction.html#index-code-generator">code generator</a>:</td><td>&nbsp;</td><td valign="top"><a href="Introduction.html">Introduction</a></td></tr>
<tr><td></td><td valign="top"><a href="Generating-your-own-code.html#index-code-generator-1">code generator</a>:</td><td>&nbsp;</td><td valign="top"><a href="Generating-your-own-code.html">Generating your own code</a></td></tr>
<tr><td></td><td valign="top"><a href="Introduction.html#index-codelet">codelet</a>:</td><td>&nbsp;</td><td valign="top"><a href="Introduction.html">Introduction</a></td></tr>
<tr><td></td><td valign="top"><a href="Installation-and-Customization.html#index-codelet-1">codelet</a>:</td><td>&nbsp;</td><td valign="top"><a href="Installation-and-Customization.html">Installation and Customization</a></td></tr>
<tr><td></td><td valign="top"><a href="Generating-your-own-code.html#index-codelet-2">codelet</a>:</td><td>&nbsp;</td><td valign="top"><a href="Generating-your-own-code.html">Generating your own code</a></td></tr>
<tr><td></td><td valign="top"><a href="Acknowledgments.html#index-codelet-3">codelet</a>:</td><td>&nbsp;</td><td valign="top"><a href="Acknowledgments.html">Acknowledgments</a></td></tr>
<tr><td></td><td valign="top"><a href="2d-MPI-example.html#index-collective-function">collective function</a>:</td><td>&nbsp;</td><td valign="top"><a href="2d-MPI-example.html">2d MPI example</a></td></tr>
<tr><td></td><td valign="top"><a href="FFTW-MPI-Wisdom.html#index-collective-function-1">collective function</a>:</td><td>&nbsp;</td><td valign="top"><a href="FFTW-MPI-Wisdom.html">FFTW MPI Wisdom</a></td></tr>
<tr><td></td><td valign="top"><a href="Avoiding-MPI-Deadlocks.html#index-collective-function-2">collective function</a>:</td><td>&nbsp;</td><td valign="top"><a href="Avoiding-MPI-Deadlocks.html">Avoiding MPI Deadlocks</a></td></tr>
<tr><td></td><td valign="top"><a href="Using-MPI-Plans.html#index-collective-function-3">collective function</a>:</td><td>&nbsp;</td><td valign="top"><a href="Using-MPI-Plans.html">Using MPI Plans</a></td></tr>
<tr><td></td><td valign="top"><a href="MPI-Plan-Creation.html#index-collective-function-4">collective function</a>:</td><td>&nbsp;</td><td valign="top"><a href="MPI-Plan-Creation.html">MPI Plan Creation</a></td></tr>
<tr><td></td><td valign="top"><a href="Column_002dmajor-Format.html#index-column_002dmajor">column-major</a>:</td><td>&nbsp;</td><td valign="top"><a href="Column_002dmajor-Format.html">Column-major Format</a></td></tr>
<tr><td></td><td valign="top"><a href="Reversing-array-dimensions.html#index-column_002dmajor-1">column-major</a>:</td><td>&nbsp;</td><td valign="top"><a href="Reversing-array-dimensions.html">Reversing array dimensions</a></td></tr>
<tr><td></td><td valign="top"><a href="Fortran_002dinterface-routines.html#index-column_002dmajor-2">column-major</a>:</td><td>&nbsp;</td><td valign="top"><a href="Fortran_002dinterface-routines.html">Fortran-interface routines</a></td></tr>
<tr><td></td><td valign="top"><a href="Fortran-Examples.html#index-column_002dmajor-3">column-major</a>:</td><td>&nbsp;</td><td valign="top"><a href="Fortran-Examples.html">Fortran Examples</a></td></tr>
<tr><td></td><td valign="top"><a href="Introduction.html#index-compiler">compiler</a>:</td><td>&nbsp;</td><td valign="top"><a href="Introduction.html">Introduction</a></td></tr>
<tr><td></td><td valign="top"><a href="Installation-and-Customization.html#index-compiler-1">compiler</a>:</td><td>&nbsp;</td><td valign="top"><a href="Installation-and-Customization.html">Installation and Customization</a></td></tr>
<tr><td></td><td valign="top"><a href="Installation-on-Unix.html#index-compiler-2">compiler</a>:</td><td>&nbsp;</td><td valign="top"><a href="Installation-on-Unix.html">Installation on Unix</a></td></tr>
<tr><td></td><td valign="top"><a href="Cycle-Counters.html#index-compiler-3">compiler</a>:</td><td>&nbsp;</td><td valign="top"><a href="Cycle-Counters.html">Cycle Counters</a></td></tr>
<tr><td></td><td valign="top"><a href="Installation-on-Unix.html#index-compiler-flags">compiler flags</a>:</td><td>&nbsp;</td><td valign="top"><a href="Installation-on-Unix.html">Installation on Unix</a></td></tr>
<tr><td></td><td valign="top"><a href="Installation-on-Unix.html#index-compiler-flags-1">compiler flags</a>:</td><td>&nbsp;</td><td valign="top"><a href="Installation-on-Unix.html">Installation on Unix</a></td></tr>
<tr><td></td><td valign="top"><a href="Wisdom-Utilities.html#index-configuration-routines">configuration routines</a>:</td><td>&nbsp;</td><td valign="top"><a href="Wisdom-Utilities.html">Wisdom Utilities</a></td></tr>
<tr><td></td><td valign="top"><a href="Installation-and-Supported-Hardware_002fSoftware.html#index-configure"><code>configure</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="Installation-and-Supported-Hardware_002fSoftware.html">Installation and Supported Hardware/Software</a></td></tr>
<tr><td></td><td valign="top"><a href="FFTW-MPI-Installation.html#index-configure-1"><code>configure</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="FFTW-MPI-Installation.html">FFTW MPI Installation</a></td></tr>
<tr><td></td><td valign="top"><a href="Installation-on-Unix.html#index-configure-2"><code>configure</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="Installation-on-Unix.html">Installation on Unix</a></td></tr>
<tr><td></td><td valign="top"><a href="Installation-and-Customization.html#index-cycle-counter">cycle counter</a>:</td><td>&nbsp;</td><td valign="top"><a href="Installation-and-Customization.html">Installation and Customization</a></td></tr>
<tr><td></td><td valign="top"><a href="Cycle-Counters.html#index-cycle-counter-1">cycle counter</a>:</td><td>&nbsp;</td><td valign="top"><a href="Cycle-Counters.html">Cycle Counters</a></td></tr>
<tr><td colspan="4"> <hr></td></tr>
<tr><th id="Concept-Index_cp_letter-D">D</th><td></td><td></td></tr>
<tr><td></td><td valign="top"><a href="Distributed_002dmemory-FFTW-with-MPI.html#index-data-distribution">data distribution</a>:</td><td>&nbsp;</td><td valign="top"><a href="Distributed_002dmemory-FFTW-with-MPI.html">Distributed-memory FFTW with MPI</a></td></tr>
<tr><td></td><td valign="top"><a href="2d-MPI-example.html#index-data-distribution-1">data distribution</a>:</td><td>&nbsp;</td><td valign="top"><a href="2d-MPI-example.html">2d MPI example</a></td></tr>
<tr><td></td><td valign="top"><a href="MPI-Data-Distribution.html#index-data-distribution-2">data distribution</a>:</td><td>&nbsp;</td><td valign="top"><a href="MPI-Data-Distribution.html">MPI Data Distribution</a></td></tr>
<tr><td></td><td valign="top"><a href="Multi_002ddimensional-MPI-DFTs-of-Real-Data.html#index-data-distribution-3">data distribution</a>:</td><td>&nbsp;</td><td valign="top"><a href="Multi_002ddimensional-MPI-DFTs-of-Real-Data.html">Multi-dimensional MPI DFTs of Real Data</a></td></tr>
<tr><td></td><td valign="top"><a href="Basic-distributed_002dtranspose-interface.html#index-data-distribution-4">data distribution</a>:</td><td>&nbsp;</td><td valign="top"><a href="Basic-distributed_002dtranspose-interface.html">Basic distributed-transpose interface</a></td></tr>
<tr><td></td><td valign="top"><a href="MPI-Data-Distribution-Functions.html#index-data-distribution-5">data distribution</a>:</td><td>&nbsp;</td><td valign="top"><a href="MPI-Data-Distribution-Functions.html">MPI Data Distribution Functions</a></td></tr>
<tr><td></td><td valign="top"><a href="Real-even_002fodd-DFTs-_0028cosine_002fsine-transforms_0029.html#index-DCT">DCT</a>:</td><td>&nbsp;</td><td valign="top"><a href="Real-even_002fodd-DFTs-_0028cosine_002fsine-transforms_0029.html">Real even/odd DFTs (cosine/sine transforms)</a></td></tr>
<tr><td></td><td valign="top"><a href="Real_002dto_002dReal-Transform-Kinds.html#index-DCT-1">DCT</a>:</td><td>&nbsp;</td><td valign="top"><a href="Real_002dto_002dReal-Transform-Kinds.html">Real-to-Real Transform Kinds</a></td></tr>
<tr><td></td><td valign="top"><a href="1d-Real_002deven-DFTs-_0028DCTs_0029.html#index-DCT-2">DCT</a>:</td><td>&nbsp;</td><td valign="top"><a href="1d-Real_002deven-DFTs-_0028DCTs_0029.html">1d Real-even DFTs (DCTs)</a></td></tr>
<tr><td></td><td valign="top"><a href="Avoiding-MPI-Deadlocks.html#index-deadlock">deadlock</a>:</td><td>&nbsp;</td><td valign="top"><a href="Avoiding-MPI-Deadlocks.html">Avoiding MPI Deadlocks</a></td></tr>
<tr><td></td><td valign="top"><a href="Complex-One_002dDimensional-DFTs.html#index-Devil">Devil</a>:</td><td>&nbsp;</td><td valign="top"><a href="Complex-One_002dDimensional-DFTs.html">Complex One-Dimensional DFTs</a></td></tr>
<tr><td></td><td valign="top"><a href="Introduction.html#index-DFT">DFT</a>:</td><td>&nbsp;</td><td valign="top"><a href="Introduction.html">Introduction</a></td></tr>
<tr><td></td><td valign="top"><a href="Complex-One_002dDimensional-DFTs.html#index-DFT-1">DFT</a>:</td><td>&nbsp;</td><td valign="top"><a href="Complex-One_002dDimensional-DFTs.html">Complex One-Dimensional DFTs</a></td></tr>
<tr><td></td><td valign="top"><a href="The-1d-Discrete-Fourier-Transform-_0028DFT_0029.html#index-DFT-2">DFT</a>:</td><td>&nbsp;</td><td valign="top"><a href="The-1d-Discrete-Fourier-Transform-_0028DFT_0029.html">The 1d Discrete Fourier Transform (DFT)</a></td></tr>
<tr><td></td><td valign="top"><a href="The-Discrete-Hartley-Transform.html#index-DHT">DHT</a>:</td><td>&nbsp;</td><td valign="top"><a href="The-Discrete-Hartley-Transform.html">The Discrete Hartley Transform</a></td></tr>
<tr><td></td><td valign="top"><a href="1d-Discrete-Hartley-Transforms-_0028DHTs_0029.html#index-DHT-1">DHT</a>:</td><td>&nbsp;</td><td valign="top"><a href="1d-Discrete-Hartley-Transforms-_0028DHTs_0029.html">1d Discrete Hartley Transforms (DHTs)</a></td></tr>
<tr><td></td><td valign="top"><a href="Real-even_002fodd-DFTs-_0028cosine_002fsine-transforms_0029.html#index-discrete-cosine-transform">discrete cosine transform</a>:</td><td>&nbsp;</td><td valign="top"><a href="Real-even_002fodd-DFTs-_0028cosine_002fsine-transforms_0029.html">Real even/odd DFTs (cosine/sine transforms)</a></td></tr>
<tr><td></td><td valign="top"><a href="Real_002dto_002dReal-Transform-Kinds.html#index-discrete-cosine-transform-1">discrete cosine transform</a>:</td><td>&nbsp;</td><td valign="top"><a href="Real_002dto_002dReal-Transform-Kinds.html">Real-to-Real Transform Kinds</a></td></tr>
<tr><td></td><td valign="top"><a href="1d-Real_002deven-DFTs-_0028DCTs_0029.html#index-discrete-cosine-transform-2">discrete cosine transform</a>:</td><td>&nbsp;</td><td valign="top"><a href="1d-Real_002deven-DFTs-_0028DCTs_0029.html">1d Real-even DFTs (DCTs)</a></td></tr>
<tr><td></td><td valign="top"><a href="Introduction.html#index-discrete-Fourier-transform">discrete Fourier transform</a>:</td><td>&nbsp;</td><td valign="top"><a href="Introduction.html">Introduction</a></td></tr>
<tr><td></td><td valign="top"><a href="The-1d-Discrete-Fourier-Transform-_0028DFT_0029.html#index-discrete-Fourier-transform-1">discrete Fourier transform</a>:</td><td>&nbsp;</td><td valign="top"><a href="The-1d-Discrete-Fourier-Transform-_0028DFT_0029.html">The 1d Discrete Fourier Transform (DFT)</a></td></tr>
<tr><td></td><td valign="top"><a href="The-Discrete-Hartley-Transform.html#index-discrete-Hartley-transform">discrete Hartley transform</a>:</td><td>&nbsp;</td><td valign="top"><a href="The-Discrete-Hartley-Transform.html">The Discrete Hartley Transform</a></td></tr>
<tr><td></td><td valign="top"><a href="Real_002dto_002dReal-Transform-Kinds.html#index-discrete-Hartley-transform-1">discrete Hartley transform</a>:</td><td>&nbsp;</td><td valign="top"><a href="Real_002dto_002dReal-Transform-Kinds.html">Real-to-Real Transform Kinds</a></td></tr>
<tr><td></td><td valign="top"><a href="1d-Discrete-Hartley-Transforms-_0028DHTs_0029.html#index-discrete-Hartley-transform-2">discrete Hartley transform</a>:</td><td>&nbsp;</td><td valign="top"><a href="1d-Discrete-Hartley-Transforms-_0028DHTs_0029.html">1d Discrete Hartley Transforms (DHTs)</a></td></tr>
<tr><td></td><td valign="top"><a href="Real-even_002fodd-DFTs-_0028cosine_002fsine-transforms_0029.html#index-discrete-sine-transform">discrete sine transform</a>:</td><td>&nbsp;</td><td valign="top"><a href="Real-even_002fodd-DFTs-_0028cosine_002fsine-transforms_0029.html">Real even/odd DFTs (cosine/sine transforms)</a></td></tr>
<tr><td></td><td valign="top"><a href="Real_002dto_002dReal-Transform-Kinds.html#index-discrete-sine-transform-1">discrete sine transform</a>:</td><td>&nbsp;</td><td valign="top"><a href="Real_002dto_002dReal-Transform-Kinds.html">Real-to-Real Transform Kinds</a></td></tr>
<tr><td></td><td valign="top"><a href="1d-Real_002dodd-DFTs-_0028DSTs_0029.html#index-discrete-sine-transform-2">discrete sine transform</a>:</td><td>&nbsp;</td><td valign="top"><a href="1d-Real_002dodd-DFTs-_0028DSTs_0029.html">1d Real-odd DFTs (DSTs)</a></td></tr>
<tr><td></td><td valign="top"><a href="Advanced-Complex-DFTs.html#index-dist">dist</a>:</td><td>&nbsp;</td><td valign="top"><a href="Advanced-Complex-DFTs.html">Advanced Complex DFTs</a></td></tr>
<tr><td></td><td valign="top"><a href="Guru-vector-and-transform-sizes.html#index-dist-1">dist</a>:</td><td>&nbsp;</td><td valign="top"><a href="Guru-vector-and-transform-sizes.html">Guru vector and transform sizes</a></td></tr>
<tr><td></td><td valign="top"><a href="Real-even_002fodd-DFTs-_0028cosine_002fsine-transforms_0029.html#index-DST">DST</a>:</td><td>&nbsp;</td><td valign="top"><a href="Real-even_002fodd-DFTs-_0028cosine_002fsine-transforms_0029.html">Real even/odd DFTs (cosine/sine transforms)</a></td></tr>
<tr><td></td><td valign="top"><a href="Real_002dto_002dReal-Transform-Kinds.html#index-DST-1">DST</a>:</td><td>&nbsp;</td><td valign="top"><a href="Real_002dto_002dReal-Transform-Kinds.html">Real-to-Real Transform Kinds</a></td></tr>
<tr><td></td><td valign="top"><a href="1d-Real_002dodd-DFTs-_0028DSTs_0029.html#index-DST-2">DST</a>:</td><td>&nbsp;</td><td valign="top"><a href="1d-Real_002dodd-DFTs-_0028DSTs_0029.html">1d Real-odd DFTs (DSTs)</a></td></tr>
<tr><td colspan="4"> <hr></td></tr>
<tr><th id="Concept-Index_cp_letter-E">E</th><td></td><td></td></tr>
<tr><td></td><td valign="top"><a href="Caveats-in-Using-Wisdom.html#index-Ecclesiastes">Ecclesiastes</a>:</td><td>&nbsp;</td><td valign="top"><a href="Caveats-in-Using-Wisdom.html">Caveats in Using Wisdom</a></td></tr>
<tr><td></td><td valign="top"><a href="Introduction.html#index-execute">execute</a>:</td><td>&nbsp;</td><td valign="top"><a href="Introduction.html">Introduction</a></td></tr>
<tr><td></td><td valign="top"><a href="Complex-One_002dDimensional-DFTs.html#index-execute-1">execute</a>:</td><td>&nbsp;</td><td valign="top"><a href="Complex-One_002dDimensional-DFTs.html">Complex One-Dimensional DFTs</a></td></tr>
<tr><td></td><td valign="top"><a href="New_002darray-Execute-Functions.html#index-execute-2">execute</a>:</td><td>&nbsp;</td><td valign="top"><a href="New_002darray-Execute-Functions.html">New-array Execute Functions</a></td></tr>
<tr><td colspan="4"> <hr></td></tr>
<tr><th id="Concept-Index_cp_letter-F">F</th><td></td><td></td></tr>
<tr><td></td><td valign="top"><a href="Introduction.html#index-FFTW">FFTW</a>:</td><td>&nbsp;</td><td valign="top"><a href="Introduction.html">Introduction</a></td></tr>
<tr><td></td><td valign="top"><a href="Caveats-in-Using-Wisdom.html#index-fftw_002dwisdom-utility">fftw-wisdom utility</a>:</td><td>&nbsp;</td><td valign="top"><a href="Caveats-in-Using-Wisdom.html">Caveats in Using Wisdom</a></td></tr>
<tr><td></td><td valign="top"><a href="Wisdom-Utilities.html#index-fftw_002dwisdom-utility-1">fftw-wisdom utility</a>:</td><td>&nbsp;</td><td valign="top"><a href="Wisdom-Utilities.html">Wisdom Utilities</a></td></tr>
<tr><td></td><td valign="top"><a href="Wisdom-Utilities.html#index-fftw_002dwisdom_002dto_002dconf-utility">fftw-wisdom-to-conf utility</a>:</td><td>&nbsp;</td><td valign="top"><a href="Wisdom-Utilities.html">Wisdom Utilities</a></td></tr>
<tr><td></td><td valign="top"><a href="Complex-One_002dDimensional-DFTs.html#index-flags">flags</a>:</td><td>&nbsp;</td><td valign="top"><a href="Complex-One_002dDimensional-DFTs.html">Complex One-Dimensional DFTs</a></td></tr>
<tr><td></td><td valign="top"><a href="One_002dDimensional-DFTs-of-Real-Data.html#index-flags-1">flags</a>:</td><td>&nbsp;</td><td valign="top"><a href="One_002dDimensional-DFTs-of-Real-Data.html">One-Dimensional DFTs of Real Data</a></td></tr>
<tr><td></td><td valign="top"><a href="Complex-DFTs.html#index-flags-2">flags</a>:</td><td>&nbsp;</td><td valign="top"><a href="Complex-DFTs.html">Complex DFTs</a></td></tr>
<tr><td></td><td valign="top"><a href="Real_002ddata-DFTs.html#index-flags-3">flags</a>:</td><td>&nbsp;</td><td valign="top"><a href="Real_002ddata-DFTs.html">Real-data DFTs</a></td></tr>
<tr><td></td><td valign="top"><a href="Real_002dto_002dReal-Transforms.html#index-flags-4">flags</a>:</td><td>&nbsp;</td><td valign="top"><a href="Real_002dto_002dReal-Transforms.html">Real-to-Real Transforms</a></td></tr>
<tr><td></td><td valign="top"><a href="Guru-Complex-DFTs.html#index-flags-5">flags</a>:</td><td>&nbsp;</td><td valign="top"><a href="Guru-Complex-DFTs.html">Guru Complex DFTs</a></td></tr>
<tr><td></td><td valign="top"><a href="Guru-Real_002ddata-DFTs.html#index-flags-6">flags</a>:</td><td>&nbsp;</td><td valign="top"><a href="Guru-Real_002ddata-DFTs.html">Guru Real-data DFTs</a></td></tr>
<tr><td></td><td valign="top"><a href="Guru-Real_002dto_002dreal-Transforms.html#index-flags-7">flags</a>:</td><td>&nbsp;</td><td valign="top"><a href="Guru-Real_002dto_002dreal-Transforms.html">Guru Real-to-real Transforms</a></td></tr>
<tr><td></td><td valign="top"><a href="Overview-of-Fortran-interface.html#index-flags-8">flags</a>:</td><td>&nbsp;</td><td valign="top"><a href="Overview-of-Fortran-interface.html">Overview of Fortran interface</a></td></tr>
<tr><td></td><td valign="top"><a href="FFTW-Constants-in-Fortran.html#index-flags-9">flags</a>:</td><td>&nbsp;</td><td valign="top"><a href="FFTW-Constants-in-Fortran.html">FFTW Constants in Fortran</a></td></tr>
<tr><td></td><td valign="top"><a href="Column_002dmajor-Format.html#index-Fortran-interface">Fortran interface</a>:</td><td>&nbsp;</td><td valign="top"><a href="Column_002dmajor-Format.html">Column-major Format</a></td></tr>
<tr><td></td><td valign="top"><a href="FFTW-MPI-Fortran-Interface.html#index-Fortran-interface-1">Fortran interface</a>:</td><td>&nbsp;</td><td valign="top"><a href="FFTW-MPI-Fortran-Interface.html">FFTW MPI Fortran Interface</a></td></tr>
<tr><td></td><td valign="top"><a href="Calling-FFTW-from-Modern-Fortran.html#index-Fortran-interface-2">Fortran interface</a>:</td><td>&nbsp;</td><td valign="top"><a href="Calling-FFTW-from-Modern-Fortran.html">Calling FFTW from Modern Fortran</a></td></tr>
<tr><td></td><td valign="top"><a href="Calling-FFTW-from-Legacy-Fortran.html#index-Fortran-interface-3">Fortran interface</a>:</td><td>&nbsp;</td><td valign="top"><a href="Calling-FFTW-from-Legacy-Fortran.html">Calling FFTW from Legacy Fortran</a></td></tr>
<tr><td></td><td valign="top"><a href="Installation-on-Unix.html#index-Fortran_002dcallable-wrappers">Fortran-callable wrappers</a>:</td><td>&nbsp;</td><td valign="top"><a href="Installation-on-Unix.html">Installation on Unix</a></td></tr>
<tr><td></td><td valign="top"><a href="Complex-One_002dDimensional-DFTs.html#index-frequency">frequency</a>:</td><td>&nbsp;</td><td valign="top"><a href="Complex-One_002dDimensional-DFTs.html">Complex One-Dimensional DFTs</a></td></tr>
<tr><td></td><td valign="top"><a href="The-1d-Discrete-Fourier-Transform-_0028DFT_0029.html#index-frequency-1">frequency</a>:</td><td>&nbsp;</td><td valign="top"><a href="The-1d-Discrete-Fourier-Transform-_0028DFT_0029.html">The 1d Discrete Fourier Transform (DFT)</a></td></tr>
<tr><td colspan="4"> <hr></td></tr>
<tr><th id="Concept-Index_cp_letter-G">G</th><td></td><td></td></tr>
<tr><td></td><td valign="top"><a href="Installation-on-Unix.html#index-g77"><code>g77</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="Installation-on-Unix.html">Installation on Unix</a></td></tr>
<tr><td></td><td valign="top"><a href="Introduction.html#index-guru-interface">guru interface</a>:</td><td>&nbsp;</td><td valign="top"><a href="Introduction.html">Introduction</a></td></tr>
<tr><td></td><td valign="top"><a href="Complex-Multi_002dDimensional-DFTs.html#index-guru-interface-1">guru interface</a>:</td><td>&nbsp;</td><td valign="top"><a href="Complex-Multi_002dDimensional-DFTs.html">Complex Multi-Dimensional DFTs</a></td></tr>
<tr><td></td><td valign="top"><a href="Guru-Interface.html#index-guru-interface-2">guru interface</a>:</td><td>&nbsp;</td><td valign="top"><a href="Guru-Interface.html">Guru Interface</a></td></tr>
<tr><td></td><td valign="top"><a href="FFTW-Fortran-type-reference.html#index-guru-interface-3">guru interface</a>:</td><td>&nbsp;</td><td valign="top"><a href="FFTW-Fortran-type-reference.html">FFTW Fortran type reference</a></td></tr>
<tr><td></td><td valign="top"><a href="Fortran_002dinterface-routines.html#index-guru-interface-4">guru interface</a>:</td><td>&nbsp;</td><td valign="top"><a href="Fortran_002dinterface-routines.html">Fortran-interface routines</a></td></tr>
<tr><td colspan="4"> <hr></td></tr>
<tr><th id="Concept-Index_cp_letter-H">H</th><td></td><td></td></tr>
<tr><td></td><td valign="top"><a href="One_002dDimensional-DFTs-of-Real-Data.html#index-halfcomplex-format">halfcomplex format</a>:</td><td>&nbsp;</td><td valign="top"><a href="One_002dDimensional-DFTs-of-Real-Data.html">One-Dimensional DFTs of Real Data</a></td></tr>
<tr><td></td><td valign="top"><a href="The-Halfcomplex_002dformat-DFT.html#index-halfcomplex-format-1">halfcomplex format</a>:</td><td>&nbsp;</td><td valign="top"><a href="The-Halfcomplex_002dformat-DFT.html">The Halfcomplex-format DFT</a></td></tr>
<tr><td></td><td valign="top"><a href="The-1d-Real_002ddata-DFT.html#index-halfcomplex-format-2">halfcomplex format</a>:</td><td>&nbsp;</td><td valign="top"><a href="The-1d-Real_002ddata-DFT.html">The 1d Real-data DFT</a></td></tr>
<tr><td></td><td valign="top"><a href="The-Halfcomplex_002dformat-DFT.html#index-hc2r">hc2r</a>:</td><td>&nbsp;</td><td valign="top"><a href="The-Halfcomplex_002dformat-DFT.html">The Halfcomplex-format DFT</a></td></tr>
<tr><td></td><td valign="top"><a href="Planner-Flags.html#index-hc2r-1">hc2r</a>:</td><td>&nbsp;</td><td valign="top"><a href="Planner-Flags.html">Planner Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="2d-MPI-example.html#index-HDF5">HDF5</a>:</td><td>&nbsp;</td><td valign="top"><a href="2d-MPI-example.html">2d MPI example</a></td></tr>
<tr><td></td><td valign="top"><a href="One_002dDimensional-DFTs-of-Real-Data.html#index-Hermitian">Hermitian</a>:</td><td>&nbsp;</td><td valign="top"><a href="One_002dDimensional-DFTs-of-Real-Data.html">One-Dimensional DFTs of Real Data</a></td></tr>
<tr><td></td><td valign="top"><a href="The-1d-Real_002ddata-DFT.html#index-Hermitian-1">Hermitian</a>:</td><td>&nbsp;</td><td valign="top"><a href="The-1d-Real_002ddata-DFT.html">The 1d Real-data DFT</a></td></tr>
<tr><td></td><td valign="top"><a href="Guru-vector-and-transform-sizes.html#index-howmany-loop">howmany loop</a>:</td><td>&nbsp;</td><td valign="top"><a href="Guru-vector-and-transform-sizes.html">Guru vector and transform sizes</a></td></tr>
<tr><td></td><td valign="top"><a href="Advanced-Complex-DFTs.html#index-howmany-parameter">howmany parameter</a>:</td><td>&nbsp;</td><td valign="top"><a href="Advanced-Complex-DFTs.html">Advanced Complex DFTs</a></td></tr>
<tr><td colspan="4"> <hr></td></tr>
<tr><th id="Concept-Index_cp_letter-I">I</th><td></td><td></td></tr>
<tr><td></td><td valign="top"><a href="Real-even_002fodd-DFTs-_0028cosine_002fsine-transforms_0029.html#index-IDCT">IDCT</a>:</td><td>&nbsp;</td><td valign="top"><a href="Real-even_002fodd-DFTs-_0028cosine_002fsine-transforms_0029.html">Real even/odd DFTs (cosine/sine transforms)</a></td></tr>
<tr><td></td><td valign="top"><a href="Real-even_002fodd-DFTs-_0028cosine_002fsine-transforms_0029.html#index-IDCT-1">IDCT</a>:</td><td>&nbsp;</td><td valign="top"><a href="Real-even_002fodd-DFTs-_0028cosine_002fsine-transforms_0029.html">Real even/odd DFTs (cosine/sine transforms)</a></td></tr>
<tr><td></td><td valign="top"><a href="Real_002dto_002dReal-Transform-Kinds.html#index-IDCT-2">IDCT</a>:</td><td>&nbsp;</td><td valign="top"><a href="Real_002dto_002dReal-Transform-Kinds.html">Real-to-Real Transform Kinds</a></td></tr>
<tr><td></td><td valign="top"><a href="1d-Real_002deven-DFTs-_0028DCTs_0029.html#index-IDCT-3">IDCT</a>:</td><td>&nbsp;</td><td valign="top"><a href="1d-Real_002deven-DFTs-_0028DCTs_0029.html">1d Real-even DFTs (DCTs)</a></td></tr>
<tr><td></td><td valign="top"><a href="Complex-One_002dDimensional-DFTs.html#index-in_002dplace">in-place</a>:</td><td>&nbsp;</td><td valign="top"><a href="Complex-One_002dDimensional-DFTs.html">Complex One-Dimensional DFTs</a></td></tr>
<tr><td></td><td valign="top"><a href="One_002dDimensional-DFTs-of-Real-Data.html#index-in_002dplace-1">in-place</a>:</td><td>&nbsp;</td><td valign="top"><a href="One_002dDimensional-DFTs-of-Real-Data.html">One-Dimensional DFTs of Real Data</a></td></tr>
<tr><td></td><td valign="top"><a href="Complex-DFTs.html#index-in_002dplace-2">in-place</a>:</td><td>&nbsp;</td><td valign="top"><a href="Complex-DFTs.html">Complex DFTs</a></td></tr>
<tr><td></td><td valign="top"><a href="Real_002ddata-DFTs.html#index-in_002dplace-3">in-place</a>:</td><td>&nbsp;</td><td valign="top"><a href="Real_002ddata-DFTs.html">Real-data DFTs</a></td></tr>
<tr><td></td><td valign="top"><a href="Real_002ddata-DFT-Array-Format.html#index-in_002dplace-4">in-place</a>:</td><td>&nbsp;</td><td valign="top"><a href="Real_002ddata-DFT-Array-Format.html">Real-data DFT Array Format</a></td></tr>
<tr><td></td><td valign="top"><a href="Real_002dto_002dReal-Transforms.html#index-in_002dplace-5">in-place</a>:</td><td>&nbsp;</td><td valign="top"><a href="Real_002dto_002dReal-Transforms.html">Real-to-Real Transforms</a></td></tr>
<tr><td></td><td valign="top"><a href="Guru-Real_002ddata-DFTs.html#index-in_002dplace-6">in-place</a>:</td><td>&nbsp;</td><td valign="top"><a href="Guru-Real_002ddata-DFTs.html">Guru Real-data DFTs</a></td></tr>
<tr><td></td><td valign="top"><a href="Guru-Real_002ddata-DFTs.html#index-in_002dplace-7">in-place</a>:</td><td>&nbsp;</td><td valign="top"><a href="Guru-Real_002ddata-DFTs.html">Guru Real-data DFTs</a></td></tr>
<tr><td></td><td valign="top"><a href="An-improved-replacement-for-MPI_005fAlltoall.html#index-in_002dplace-8">in-place</a>:</td><td>&nbsp;</td><td valign="top"><a href="An-improved-replacement-for-MPI_005fAlltoall.html">An improved replacement for MPI_Alltoall</a></td></tr>
<tr><td></td><td valign="top"><a href="Reversing-array-dimensions.html#index-in_002dplace-9">in-place</a>:</td><td>&nbsp;</td><td valign="top"><a href="Reversing-array-dimensions.html">Reversing array dimensions</a></td></tr>
<tr><td></td><td valign="top"><a href="FFTW-Fortran-type-reference.html#index-in_002dplace-10">in-place</a>:</td><td>&nbsp;</td><td valign="top"><a href="FFTW-Fortran-type-reference.html">FFTW Fortran type reference</a></td></tr>
<tr><td></td><td valign="top"><a href="Installation-and-Customization.html#index-installation">installation</a>:</td><td>&nbsp;</td><td valign="top"><a href="Installation-and-Customization.html">Installation and Customization</a></td></tr>
<tr><td></td><td valign="top"><a href="Interleaved-and-split-arrays.html#index-interleaved-format">interleaved format</a>:</td><td>&nbsp;</td><td valign="top"><a href="Interleaved-and-split-arrays.html">Interleaved and split arrays</a></td></tr>
<tr><td></td><td valign="top"><a href="FFTW-MPI-Fortran-Interface.html#index-iso_005fc_005fbinding">iso_c_binding</a>:</td><td>&nbsp;</td><td valign="top"><a href="FFTW-MPI-Fortran-Interface.html">FFTW MPI Fortran Interface</a></td></tr>
<tr><td></td><td valign="top"><a href="Overview-of-Fortran-interface.html#index-iso_005fc_005fbinding-1">iso_c_binding</a>:</td><td>&nbsp;</td><td valign="top"><a href="Overview-of-Fortran-interface.html">Overview of Fortran interface</a></td></tr>
<tr><td></td><td valign="top"><a href="Extended-and-quadruple-precision-in-Fortran.html#index-iso_005fc_005fbinding-2">iso_c_binding</a>:</td><td>&nbsp;</td><td valign="top"><a href="Extended-and-quadruple-precision-in-Fortran.html">Extended and quadruple precision in Fortran</a></td></tr>
<tr><td colspan="4"> <hr></td></tr>
<tr><th id="Concept-Index_cp_letter-K">K</th><td></td><td></td></tr>
<tr><td></td><td valign="top"><a href="More-DFTs-of-Real-Data.html#index-kind-_0028r2r_0029">kind (r2r)</a>:</td><td>&nbsp;</td><td valign="top"><a href="More-DFTs-of-Real-Data.html">More DFTs of Real Data</a></td></tr>
<tr><td></td><td valign="top"><a href="Real_002dto_002dReal-Transform-Kinds.html#index-kind-_0028r2r_0029-1">kind (r2r)</a>:</td><td>&nbsp;</td><td valign="top"><a href="Real_002dto_002dReal-Transform-Kinds.html">Real-to-Real Transform Kinds</a></td></tr>
<tr><td colspan="4"> <hr></td></tr>
<tr><th id="Concept-Index_cp_letter-L">L</th><td></td><td></td></tr>
<tr><td></td><td valign="top"><a href="Usage-of-Multi_002dthreaded-FFTW.html#index-linking-on-Unix">linking on Unix</a>:</td><td>&nbsp;</td><td valign="top"><a href="Usage-of-Multi_002dthreaded-FFTW.html">Usage of Multi-threaded FFTW</a></td></tr>
<tr><td></td><td valign="top"><a href="Linking-and-Initializing-MPI-FFTW.html#index-linking-on-Unix-1">linking on Unix</a>:</td><td>&nbsp;</td><td valign="top"><a href="Linking-and-Initializing-MPI-FFTW.html">Linking and Initializing MPI FFTW</a></td></tr>
<tr><td></td><td valign="top"><a href="Acknowledgments.html#index-LISP">LISP</a>:</td><td>&nbsp;</td><td valign="top"><a href="Acknowledgments.html">Acknowledgments</a></td></tr>
<tr><td></td><td valign="top"><a href="Load-balancing.html#index-load-balancing">load balancing</a>:</td><td>&nbsp;</td><td valign="top"><a href="Load-balancing.html">Load balancing</a></td></tr>
<tr><td></td><td valign="top"><a href="FFTW-MPI-Performance-Tips.html#index-load-balancing-1">load balancing</a>:</td><td>&nbsp;</td><td valign="top"><a href="FFTW-MPI-Performance-Tips.html">FFTW MPI Performance Tips</a></td></tr>
<tr><td colspan="4"> <hr></td></tr>
<tr><th id="Concept-Index_cp_letter-M">M</th><td></td><td></td></tr>
<tr><td></td><td valign="top"><a href="Generating-your-own-code.html#index-monadic-programming">monadic programming</a>:</td><td>&nbsp;</td><td valign="top"><a href="Generating-your-own-code.html">Generating your own code</a></td></tr>
<tr><td></td><td valign="top"><a href="Distributed_002dmemory-FFTW-with-MPI.html#index-MPI">MPI</a>:</td><td>&nbsp;</td><td valign="top"><a href="Distributed_002dmemory-FFTW-with-MPI.html">Distributed-memory FFTW with MPI</a></td></tr>
<tr><td></td><td valign="top"><a href="Installation-on-Unix.html#index-MPI-1">MPI</a>:</td><td>&nbsp;</td><td valign="top"><a href="Installation-on-Unix.html">Installation on Unix</a></td></tr>
<tr><td></td><td valign="top"><a href="Distributed_002dmemory-FFTW-with-MPI.html#index-MPI-communicator">MPI communicator</a>:</td><td>&nbsp;</td><td valign="top"><a href="Distributed_002dmemory-FFTW-with-MPI.html">Distributed-memory FFTW with MPI</a></td></tr>
<tr><td></td><td valign="top"><a href="Using-MPI-Plans.html#index-MPI-communicator-1">MPI communicator</a>:</td><td>&nbsp;</td><td valign="top"><a href="Using-MPI-Plans.html">Using MPI Plans</a></td></tr>
<tr><td></td><td valign="top"><a href="MPI-Plan-Creation.html#index-MPI-communicator-2">MPI communicator</a>:</td><td>&nbsp;</td><td valign="top"><a href="MPI-Plan-Creation.html">MPI Plan Creation</a></td></tr>
<tr><td></td><td valign="top"><a href="FFTW-MPI-Fortran-Interface.html#index-MPI-communicator-3">MPI communicator</a>:</td><td>&nbsp;</td><td valign="top"><a href="FFTW-MPI-Fortran-Interface.html">FFTW MPI Fortran Interface</a></td></tr>
<tr><td></td><td valign="top"><a href="2d-MPI-example.html#index-MPI-I_002fO">MPI I/O</a>:</td><td>&nbsp;</td><td valign="top"><a href="2d-MPI-example.html">2d MPI example</a></td></tr>
<tr><td></td><td valign="top"><a href="FFTW-MPI-Wisdom.html#index-MPI-I_002fO-1">MPI I/O</a>:</td><td>&nbsp;</td><td valign="top"><a href="FFTW-MPI-Wisdom.html">FFTW MPI Wisdom</a></td></tr>
<tr><td></td><td valign="top"><a href="FFTW-MPI-Installation.html#index-mpicc"><code>mpicc</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="FFTW-MPI-Installation.html">FFTW MPI Installation</a></td></tr>
<tr><td></td><td valign="top"><a href="Linking-and-Initializing-MPI-FFTW.html#index-mpicc-1"><code>mpicc</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="Linking-and-Initializing-MPI-FFTW.html">Linking and Initializing MPI FFTW</a></td></tr>
<tr><td colspan="4"> <hr></td></tr>
<tr><th id="Concept-Index_cp_letter-N">N</th><td></td><td></td></tr>
<tr><td></td><td valign="top"><a href="New_002darray-Execute-Functions.html#index-new_002darray-execution">new-array execution</a>:</td><td>&nbsp;</td><td valign="top"><a href="New_002darray-Execute-Functions.html">New-array Execute Functions</a></td></tr>
<tr><td></td><td valign="top"><a href="Using-MPI-Plans.html#index-new_002darray-execution-1">new-array execution</a>:</td><td>&nbsp;</td><td valign="top"><a href="Using-MPI-Plans.html">Using MPI Plans</a></td></tr>
<tr><td></td><td valign="top"><a href="MPI-Plan-Creation.html#index-new_002darray-execution-2">new-array execution</a>:</td><td>&nbsp;</td><td valign="top"><a href="MPI-Plan-Creation.html">MPI Plan Creation</a></td></tr>
<tr><td></td><td valign="top"><a href="FFTW-MPI-Fortran-Interface.html#index-new_002darray-execution-3">new-array execution</a>:</td><td>&nbsp;</td><td valign="top"><a href="FFTW-MPI-Fortran-Interface.html">FFTW MPI Fortran Interface</a></td></tr>
<tr><td></td><td valign="top"><a href="Complex-One_002dDimensional-DFTs.html#index-normalization">normalization</a>:</td><td>&nbsp;</td><td valign="top"><a href="Complex-One_002dDimensional-DFTs.html">Complex One-Dimensional DFTs</a></td></tr>
<tr><td></td><td valign="top"><a href="Multi_002dDimensional-DFTs-of-Real-Data.html#index-normalization-1">normalization</a>:</td><td>&nbsp;</td><td valign="top"><a href="Multi_002dDimensional-DFTs-of-Real-Data.html">Multi-Dimensional DFTs of Real Data</a></td></tr>
<tr><td></td><td valign="top"><a href="The-Halfcomplex_002dformat-DFT.html#index-normalization-2">normalization</a>:</td><td>&nbsp;</td><td valign="top"><a href="The-Halfcomplex_002dformat-DFT.html">The Halfcomplex-format DFT</a></td></tr>
<tr><td></td><td valign="top"><a href="Real-even_002fodd-DFTs-_0028cosine_002fsine-transforms_0029.html#index-normalization-3">normalization</a>:</td><td>&nbsp;</td><td valign="top"><a href="Real-even_002fodd-DFTs-_0028cosine_002fsine-transforms_0029.html">Real even/odd DFTs (cosine/sine transforms)</a></td></tr>
<tr><td></td><td valign="top"><a href="The-Discrete-Hartley-Transform.html#index-normalization-4">normalization</a>:</td><td>&nbsp;</td><td valign="top"><a href="The-Discrete-Hartley-Transform.html">The Discrete Hartley Transform</a></td></tr>
<tr><td></td><td valign="top"><a href="Complex-DFTs.html#index-normalization-5">normalization</a>:</td><td>&nbsp;</td><td valign="top"><a href="Complex-DFTs.html">Complex DFTs</a></td></tr>
<tr><td></td><td valign="top"><a href="Real_002ddata-DFTs.html#index-normalization-6">normalization</a>:</td><td>&nbsp;</td><td valign="top"><a href="Real_002ddata-DFTs.html">Real-data DFTs</a></td></tr>
<tr><td></td><td valign="top"><a href="Real_002dto_002dReal-Transform-Kinds.html#index-normalization-7">normalization</a>:</td><td>&nbsp;</td><td valign="top"><a href="Real_002dto_002dReal-Transform-Kinds.html">Real-to-Real Transform Kinds</a></td></tr>
<tr><td></td><td valign="top"><a href="The-1d-Discrete-Fourier-Transform-_0028DFT_0029.html#index-normalization-8">normalization</a>:</td><td>&nbsp;</td><td valign="top"><a href="The-1d-Discrete-Fourier-Transform-_0028DFT_0029.html">The 1d Discrete Fourier Transform (DFT)</a></td></tr>
<tr><td></td><td valign="top"><a href="The-1d-Real_002ddata-DFT.html#index-normalization-9">normalization</a>:</td><td>&nbsp;</td><td valign="top"><a href="The-1d-Real_002ddata-DFT.html">The 1d Real-data DFT</a></td></tr>
<tr><td></td><td valign="top"><a href="1d-Real_002deven-DFTs-_0028DCTs_0029.html#index-normalization-10">normalization</a>:</td><td>&nbsp;</td><td valign="top"><a href="1d-Real_002deven-DFTs-_0028DCTs_0029.html">1d Real-even DFTs (DCTs)</a></td></tr>
<tr><td></td><td valign="top"><a href="1d-Real_002dodd-DFTs-_0028DSTs_0029.html#index-normalization-11">normalization</a>:</td><td>&nbsp;</td><td valign="top"><a href="1d-Real_002dodd-DFTs-_0028DSTs_0029.html">1d Real-odd DFTs (DSTs)</a></td></tr>
<tr><td></td><td valign="top"><a href="1d-Discrete-Hartley-Transforms-_0028DHTs_0029.html#index-normalization-12">normalization</a>:</td><td>&nbsp;</td><td valign="top"><a href="1d-Discrete-Hartley-Transforms-_0028DHTs_0029.html">1d Discrete Hartley Transforms (DHTs)</a></td></tr>
<tr><td></td><td valign="top"><a href="How-Many-Threads-to-Use_003f.html#index-number-of-threads">number of threads</a>:</td><td>&nbsp;</td><td valign="top"><a href="How-Many-Threads-to-Use_003f.html">How Many Threads to Use?</a></td></tr>
<tr><td colspan="4"> <hr></td></tr>
<tr><th id="Concept-Index_cp_letter-O">O</th><td></td><td></td></tr>
<tr><td></td><td valign="top"><a href="Installation-and-Supported-Hardware_002fSoftware.html#index-OpenMP">OpenMP</a>:</td><td>&nbsp;</td><td valign="top"><a href="Installation-and-Supported-Hardware_002fSoftware.html">Installation and Supported Hardware/Software</a></td></tr>
<tr><td></td><td valign="top"><a href="Usage-of-Multi_002dthreaded-FFTW.html#index-OpenMP-1">OpenMP</a>:</td><td>&nbsp;</td><td valign="top"><a href="Usage-of-Multi_002dthreaded-FFTW.html">Usage of Multi-threaded FFTW</a></td></tr>
<tr><td></td><td valign="top"><a href="Usage-of-Multi_002dthreaded-FFTW.html#index-OpenMP-2">OpenMP</a>:</td><td>&nbsp;</td><td valign="top"><a href="Usage-of-Multi_002dthreaded-FFTW.html">Usage of Multi-threaded FFTW</a></td></tr>
<tr><td></td><td valign="top"><a href="Thread-safety.html#index-OpenMP-3">OpenMP</a>:</td><td>&nbsp;</td><td valign="top"><a href="Thread-safety.html">Thread safety</a></td></tr>
<tr><td></td><td valign="top"><a href="Planner-Flags.html#index-out_002dof_002dplace">out-of-place</a>:</td><td>&nbsp;</td><td valign="top"><a href="Planner-Flags.html">Planner Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="Real_002ddata-DFT-Array-Format.html#index-out_002dof_002dplace-1">out-of-place</a>:</td><td>&nbsp;</td><td valign="top"><a href="Real_002ddata-DFT-Array-Format.html">Real-data DFT Array Format</a></td></tr>
<tr><td colspan="4"> <hr></td></tr>
<tr><th id="Concept-Index_cp_letter-P">P</th><td></td><td></td></tr>
<tr><td></td><td valign="top"><a href="One_002dDimensional-DFTs-of-Real-Data.html#index-padding">padding</a>:</td><td>&nbsp;</td><td valign="top"><a href="One_002dDimensional-DFTs-of-Real-Data.html">One-Dimensional DFTs of Real Data</a></td></tr>
<tr><td></td><td valign="top"><a href="Multi_002dDimensional-DFTs-of-Real-Data.html#index-padding-1">padding</a>:</td><td>&nbsp;</td><td valign="top"><a href="Multi_002dDimensional-DFTs-of-Real-Data.html">Multi-Dimensional DFTs of Real Data</a></td></tr>
<tr><td></td><td valign="top"><a href="Real_002ddata-DFTs.html#index-padding-2">padding</a>:</td><td>&nbsp;</td><td valign="top"><a href="Real_002ddata-DFTs.html">Real-data DFTs</a></td></tr>
<tr><td></td><td valign="top"><a href="Real_002ddata-DFT-Array-Format.html#index-padding-3">padding</a>:</td><td>&nbsp;</td><td valign="top"><a href="Real_002ddata-DFT-Array-Format.html">Real-data DFT Array Format</a></td></tr>
<tr><td></td><td valign="top"><a href="Multi_002ddimensional-MPI-DFTs-of-Real-Data.html#index-padding-4">padding</a>:</td><td>&nbsp;</td><td valign="top"><a href="Multi_002ddimensional-MPI-DFTs-of-Real-Data.html">Multi-dimensional MPI DFTs of Real Data</a></td></tr>
<tr><td></td><td valign="top"><a href="Reversing-array-dimensions.html#index-padding-5">padding</a>:</td><td>&nbsp;</td><td valign="top"><a href="Reversing-array-dimensions.html">Reversing array dimensions</a></td></tr>
<tr><td></td><td valign="top"><a href="Multi_002dthreaded-FFTW.html#index-parallel-transform">parallel transform</a>:</td><td>&nbsp;</td><td valign="top"><a href="Multi_002dthreaded-FFTW.html">Multi-threaded FFTW</a></td></tr>
<tr><td></td><td valign="top"><a href="Distributed_002dmemory-FFTW-with-MPI.html#index-parallel-transform-1">parallel transform</a>:</td><td>&nbsp;</td><td valign="top"><a href="Distributed_002dmemory-FFTW-with-MPI.html">Distributed-memory FFTW with MPI</a></td></tr>
<tr><td></td><td valign="top"><a href="Complex-Multi_002dDimensional-DFTs.html#index-partial-order">partial order</a>:</td><td>&nbsp;</td><td valign="top"><a href="Complex-Multi_002dDimensional-DFTs.html">Complex Multi-Dimensional DFTs</a></td></tr>
<tr><td></td><td valign="top"><a href="Introduction.html#index-plan">plan</a>:</td><td>&nbsp;</td><td valign="top"><a href="Introduction.html">Introduction</a></td></tr>
<tr><td></td><td valign="top"><a href="Complex-One_002dDimensional-DFTs.html#index-plan-1">plan</a>:</td><td>&nbsp;</td><td valign="top"><a href="Complex-One_002dDimensional-DFTs.html">Complex One-Dimensional DFTs</a></td></tr>
<tr><td></td><td valign="top"><a href="Introduction.html#index-planner">planner</a>:</td><td>&nbsp;</td><td valign="top"><a href="Introduction.html">Introduction</a></td></tr>
<tr><td></td><td valign="top"><a href="SIMD-alignment-and-fftw_005fmalloc.html#index-portability">portability</a>:</td><td>&nbsp;</td><td valign="top"><a href="SIMD-alignment-and-fftw_005fmalloc.html">SIMD alignment and fftw_malloc</a></td></tr>
<tr><td></td><td valign="top"><a href="Caveats-in-Using-Wisdom.html#index-portability-1">portability</a>:</td><td>&nbsp;</td><td valign="top"><a href="Caveats-in-Using-Wisdom.html">Caveats in Using Wisdom</a></td></tr>
<tr><td></td><td valign="top"><a href="Complex-numbers.html#index-portability-2">portability</a>:</td><td>&nbsp;</td><td valign="top"><a href="Complex-numbers.html">Complex numbers</a></td></tr>
<tr><td></td><td valign="top"><a href="Installation-and-Supported-Hardware_002fSoftware.html#index-portability-3">portability</a>:</td><td>&nbsp;</td><td valign="top"><a href="Installation-and-Supported-Hardware_002fSoftware.html">Installation and Supported Hardware/Software</a></td></tr>
<tr><td></td><td valign="top"><a href="Calling-FFTW-from-Modern-Fortran.html#index-portability-4">portability</a>:</td><td>&nbsp;</td><td valign="top"><a href="Calling-FFTW-from-Modern-Fortran.html">Calling FFTW from Modern Fortran</a></td></tr>
<tr><td></td><td valign="top"><a href="FFTW-Fortran-type-reference.html#index-portability-5">portability</a>:</td><td>&nbsp;</td><td valign="top"><a href="FFTW-Fortran-type-reference.html">FFTW Fortran type reference</a></td></tr>
<tr><td></td><td valign="top"><a href="Fortran_002dinterface-routines.html#index-portability-6">portability</a>:</td><td>&nbsp;</td><td valign="top"><a href="Fortran_002dinterface-routines.html">Fortran-interface routines</a></td></tr>
<tr><td></td><td valign="top"><a href="Fortran_002dinterface-routines.html#index-portability-7">portability</a>:</td><td>&nbsp;</td><td valign="top"><a href="Fortran_002dinterface-routines.html">Fortran-interface routines</a></td></tr>
<tr><td></td><td valign="top"><a href="Wisdom-of-Fortran_003f.html#index-portability-8">portability</a>:</td><td>&nbsp;</td><td valign="top"><a href="Wisdom-of-Fortran_003f.html">Wisdom of Fortran?</a></td></tr>
<tr><td></td><td valign="top"><a href="Installation-and-Customization.html#index-portability-9">portability</a>:</td><td>&nbsp;</td><td valign="top"><a href="Installation-and-Customization.html">Installation and Customization</a></td></tr>
<tr><td></td><td valign="top"><a href="Complex-One_002dDimensional-DFTs.html#index-precision">precision</a>:</td><td>&nbsp;</td><td valign="top"><a href="Complex-One_002dDimensional-DFTs.html">Complex One-Dimensional DFTs</a></td></tr>
<tr><td></td><td valign="top"><a href="One_002dDimensional-DFTs-of-Real-Data.html#index-precision-1">precision</a>:</td><td>&nbsp;</td><td valign="top"><a href="One_002dDimensional-DFTs-of-Real-Data.html">One-Dimensional DFTs of Real Data</a></td></tr>
<tr><td></td><td valign="top"><a href="SIMD-alignment-and-fftw_005fmalloc.html#index-precision-2">precision</a>:</td><td>&nbsp;</td><td valign="top"><a href="SIMD-alignment-and-fftw_005fmalloc.html">SIMD alignment and fftw_malloc</a></td></tr>
<tr><td></td><td valign="top"><a href="Precision.html#index-precision-3">precision</a>:</td><td>&nbsp;</td><td valign="top"><a href="Precision.html">Precision</a></td></tr>
<tr><td></td><td valign="top"><a href="Memory-Allocation.html#index-precision-4">precision</a>:</td><td>&nbsp;</td><td valign="top"><a href="Memory-Allocation.html">Memory Allocation</a></td></tr>
<tr><td></td><td valign="top"><a href="Linking-and-Initializing-MPI-FFTW.html#index-precision-5">precision</a>:</td><td>&nbsp;</td><td valign="top"><a href="Linking-and-Initializing-MPI-FFTW.html">Linking and Initializing MPI FFTW</a></td></tr>
<tr><td></td><td valign="top"><a href="MPI-Files-and-Data-Types.html#index-precision-6">precision</a>:</td><td>&nbsp;</td><td valign="top"><a href="MPI-Files-and-Data-Types.html">MPI Files and Data Types</a></td></tr>
<tr><td></td><td valign="top"><a href="Extended-and-quadruple-precision-in-Fortran.html#index-precision-7">precision</a>:</td><td>&nbsp;</td><td valign="top"><a href="Extended-and-quadruple-precision-in-Fortran.html">Extended and quadruple precision in Fortran</a></td></tr>
<tr><td></td><td valign="top"><a href="FFTW-Fortran-type-reference.html#index-precision-8">precision</a>:</td><td>&nbsp;</td><td valign="top"><a href="FFTW-Fortran-type-reference.html">FFTW Fortran type reference</a></td></tr>
<tr><td></td><td valign="top"><a href="Installation-on-Unix.html#index-precision-9">precision</a>:</td><td>&nbsp;</td><td valign="top"><a href="Installation-on-Unix.html">Installation on Unix</a></td></tr>
<tr><td></td><td valign="top"><a href="Installation-on-Unix.html#index-precision-10">precision</a>:</td><td>&nbsp;</td><td valign="top"><a href="Installation-on-Unix.html">Installation on Unix</a></td></tr>
<tr><td></td><td valign="top"><a href="Installation-on-Unix.html#index-precision-11">precision</a>:</td><td>&nbsp;</td><td valign="top"><a href="Installation-on-Unix.html">Installation on Unix</a></td></tr>
<tr><td colspan="4"> <hr></td></tr>
<tr><th id="Concept-Index_cp_letter-R">R</th><td></td><td></td></tr>
<tr><td></td><td valign="top"><a href="One_002dDimensional-DFTs-of-Real-Data.html#index-r2c">r2c</a>:</td><td>&nbsp;</td><td valign="top"><a href="One_002dDimensional-DFTs-of-Real-Data.html">One-Dimensional DFTs of Real Data</a></td></tr>
<tr><td></td><td valign="top"><a href="The-Halfcomplex_002dformat-DFT.html#index-r2c-1">r2c</a>:</td><td>&nbsp;</td><td valign="top"><a href="The-Halfcomplex_002dformat-DFT.html">The Halfcomplex-format DFT</a></td></tr>
<tr><td></td><td valign="top"><a href="Real_002ddata-DFTs.html#index-r2c-2">r2c</a>:</td><td>&nbsp;</td><td valign="top"><a href="Real_002ddata-DFTs.html">Real-data DFTs</a></td></tr>
<tr><td></td><td valign="top"><a href="Multi_002ddimensional-Transforms.html#index-r2c-3">r2c</a>:</td><td>&nbsp;</td><td valign="top"><a href="Multi_002ddimensional-Transforms.html">Multi-dimensional Transforms</a></td></tr>
<tr><td></td><td valign="top"><a href="MPI-Plan-Creation.html#index-r2c-4">r2c</a>:</td><td>&nbsp;</td><td valign="top"><a href="MPI-Plan-Creation.html">MPI Plan Creation</a></td></tr>
<tr><td></td><td valign="top"><a href="Multi_002dDimensional-DFTs-of-Real-Data.html#index-r2c_002fc2r-multi_002ddimensional-array-format">r2c/c2r multi-dimensional array format</a>:</td><td>&nbsp;</td><td valign="top"><a href="Multi_002dDimensional-DFTs-of-Real-Data.html">Multi-Dimensional DFTs of Real Data</a></td></tr>
<tr><td></td><td valign="top"><a href="Real_002ddata-DFT-Array-Format.html#index-r2c_002fc2r-multi_002ddimensional-array-format-1">r2c/c2r multi-dimensional array format</a>:</td><td>&nbsp;</td><td valign="top"><a href="Real_002ddata-DFT-Array-Format.html">Real-data DFT Array Format</a></td></tr>
<tr><td></td><td valign="top"><a href="Reversing-array-dimensions.html#index-r2c_002fc2r-multi_002ddimensional-array-format-2">r2c/c2r multi-dimensional array format</a>:</td><td>&nbsp;</td><td valign="top"><a href="Reversing-array-dimensions.html">Reversing array dimensions</a></td></tr>
<tr><td></td><td valign="top"><a href="Fortran-Examples.html#index-r2c_002fc2r-multi_002ddimensional-array-format-3">r2c/c2r multi-dimensional array format</a>:</td><td>&nbsp;</td><td valign="top"><a href="Fortran-Examples.html">Fortran Examples</a></td></tr>
<tr><td></td><td valign="top"><a href="The-Halfcomplex_002dformat-DFT.html#index-r2hc">r2hc</a>:</td><td>&nbsp;</td><td valign="top"><a href="The-Halfcomplex_002dformat-DFT.html">The Halfcomplex-format DFT</a></td></tr>
<tr><td></td><td valign="top"><a href="More-DFTs-of-Real-Data.html#index-r2r">r2r</a>:</td><td>&nbsp;</td><td valign="top"><a href="More-DFTs-of-Real-Data.html">More DFTs of Real Data</a></td></tr>
<tr><td></td><td valign="top"><a href="Real_002dto_002dReal-Transforms.html#index-r2r-1">r2r</a>:</td><td>&nbsp;</td><td valign="top"><a href="Real_002dto_002dReal-Transforms.html">Real-to-Real Transforms</a></td></tr>
<tr><td></td><td valign="top"><a href="The-1d-Real_002ddata-DFT.html#index-r2r-2">r2r</a>:</td><td>&nbsp;</td><td valign="top"><a href="The-1d-Real_002ddata-DFT.html">The 1d Real-data DFT</a></td></tr>
<tr><td></td><td valign="top"><a href="Other-Multi_002ddimensional-Real_002ddata-MPI-Transforms.html#index-r2r-3">r2r</a>:</td><td>&nbsp;</td><td valign="top"><a href="Other-Multi_002ddimensional-Real_002ddata-MPI-Transforms.html">Other Multi-dimensional Real-data MPI Transforms</a></td></tr>
<tr><td></td><td valign="top"><a href="MPI-Plan-Creation.html#index-r2r-4">r2r</a>:</td><td>&nbsp;</td><td valign="top"><a href="MPI-Plan-Creation.html">MPI Plan Creation</a></td></tr>
<tr><td></td><td valign="top"><a href="Complex-Multi_002dDimensional-DFTs.html#index-rank">rank</a>:</td><td>&nbsp;</td><td valign="top"><a href="Complex-Multi_002dDimensional-DFTs.html">Complex Multi-Dimensional DFTs</a></td></tr>
<tr><td></td><td valign="top"><a href="Real-even_002fodd-DFTs-_0028cosine_002fsine-transforms_0029.html#index-real_002deven-DFT">real-even DFT</a>:</td><td>&nbsp;</td><td valign="top"><a href="Real-even_002fodd-DFTs-_0028cosine_002fsine-transforms_0029.html">Real even/odd DFTs (cosine/sine transforms)</a></td></tr>
<tr><td></td><td valign="top"><a href="1d-Real_002deven-DFTs-_0028DCTs_0029.html#index-real_002deven-DFT-1">real-even DFT</a>:</td><td>&nbsp;</td><td valign="top"><a href="1d-Real_002deven-DFTs-_0028DCTs_0029.html">1d Real-even DFTs (DCTs)</a></td></tr>
<tr><td></td><td valign="top"><a href="Real-even_002fodd-DFTs-_0028cosine_002fsine-transforms_0029.html#index-real_002dodd-DFT">real-odd DFT</a>:</td><td>&nbsp;</td><td valign="top"><a href="Real-even_002fodd-DFTs-_0028cosine_002fsine-transforms_0029.html">Real even/odd DFTs (cosine/sine transforms)</a></td></tr>
<tr><td></td><td valign="top"><a href="1d-Real_002dodd-DFTs-_0028DSTs_0029.html#index-real_002dodd-DFT-1">real-odd DFT</a>:</td><td>&nbsp;</td><td valign="top"><a href="1d-Real_002dodd-DFTs-_0028DSTs_0029.html">1d Real-odd DFTs (DSTs)</a></td></tr>
<tr><td></td><td valign="top"><a href="Real-even_002fodd-DFTs-_0028cosine_002fsine-transforms_0029.html#index-REDFT">REDFT</a>:</td><td>&nbsp;</td><td valign="top"><a href="Real-even_002fodd-DFTs-_0028cosine_002fsine-transforms_0029.html">Real even/odd DFTs (cosine/sine transforms)</a></td></tr>
<tr><td></td><td valign="top"><a href="1d-Real_002deven-DFTs-_0028DCTs_0029.html#index-REDFT-1">REDFT</a>:</td><td>&nbsp;</td><td valign="top"><a href="1d-Real_002deven-DFTs-_0028DCTs_0029.html">1d Real-even DFTs (DCTs)</a></td></tr>
<tr><td></td><td valign="top"><a href="Generating-your-own-code.html#index-REDFT-2">REDFT</a>:</td><td>&nbsp;</td><td valign="top"><a href="Generating-your-own-code.html">Generating your own code</a></td></tr>
<tr><td></td><td valign="top"><a href="Real-even_002fodd-DFTs-_0028cosine_002fsine-transforms_0029.html#index-RODFT">RODFT</a>:</td><td>&nbsp;</td><td valign="top"><a href="Real-even_002fodd-DFTs-_0028cosine_002fsine-transforms_0029.html">Real even/odd DFTs (cosine/sine transforms)</a></td></tr>
<tr><td></td><td valign="top"><a href="1d-Real_002dodd-DFTs-_0028DSTs_0029.html#index-RODFT-1">RODFT</a>:</td><td>&nbsp;</td><td valign="top"><a href="1d-Real_002dodd-DFTs-_0028DSTs_0029.html">1d Real-odd DFTs (DSTs)</a></td></tr>
<tr><td></td><td valign="top"><a href="Row_002dmajor-Format.html#index-row_002dmajor">row-major</a>:</td><td>&nbsp;</td><td valign="top"><a href="Row_002dmajor-Format.html">Row-major Format</a></td></tr>
<tr><td></td><td valign="top"><a href="Complex-DFTs.html#index-row_002dmajor-1">row-major</a>:</td><td>&nbsp;</td><td valign="top"><a href="Complex-DFTs.html">Complex DFTs</a></td></tr>
<tr><td></td><td valign="top"><a href="Real_002dto_002dReal-Transforms.html#index-row_002dmajor-2">row-major</a>:</td><td>&nbsp;</td><td valign="top"><a href="Real_002dto_002dReal-Transforms.html">Real-to-Real Transforms</a></td></tr>
<tr><td></td><td valign="top"><a href="Guru-vector-and-transform-sizes.html#index-row_002dmajor-3">row-major</a>:</td><td>&nbsp;</td><td valign="top"><a href="Guru-vector-and-transform-sizes.html">Guru vector and transform sizes</a></td></tr>
<tr><td></td><td valign="top"><a href="Basic-and-advanced-distribution-interfaces.html#index-row_002dmajor-4">row-major</a>:</td><td>&nbsp;</td><td valign="top"><a href="Basic-and-advanced-distribution-interfaces.html">Basic and advanced distribution interfaces</a></td></tr>
<tr><td></td><td valign="top"><a href="Multi_002ddimensional-MPI-DFTs-of-Real-Data.html#index-row_002dmajor-5">row-major</a>:</td><td>&nbsp;</td><td valign="top"><a href="Multi_002ddimensional-MPI-DFTs-of-Real-Data.html">Multi-dimensional MPI DFTs of Real Data</a></td></tr>
<tr><td></td><td valign="top"><a href="Reversing-array-dimensions.html#index-row_002dmajor-6">row-major</a>:</td><td>&nbsp;</td><td valign="top"><a href="Reversing-array-dimensions.html">Reversing array dimensions</a></td></tr>
<tr><td colspan="4"> <hr></td></tr>
<tr><th id="Concept-Index_cp_letter-S">S</th><td></td><td></td></tr>
<tr><td></td><td valign="top"><a href="Words-of-Wisdom_002dSaving-Plans.html#index-saving-plans-to-disk">saving plans to disk</a>:</td><td>&nbsp;</td><td valign="top"><a href="Words-of-Wisdom_002dSaving-Plans.html">Words of Wisdom-Saving Plans</a></td></tr>
<tr><td></td><td valign="top"><a href="Wisdom.html#index-saving-plans-to-disk-1">saving plans to disk</a>:</td><td>&nbsp;</td><td valign="top"><a href="Wisdom.html">Wisdom</a></td></tr>
<tr><td></td><td valign="top"><a href="FFTW-MPI-Wisdom.html#index-saving-plans-to-disk-2">saving plans to disk</a>:</td><td>&nbsp;</td><td valign="top"><a href="FFTW-MPI-Wisdom.html">FFTW MPI Wisdom</a></td></tr>
<tr><td></td><td valign="top"><a href="Accessing-the-wisdom-API-from-Fortran.html#index-saving-plans-to-disk-3">saving plans to disk</a>:</td><td>&nbsp;</td><td valign="top"><a href="Accessing-the-wisdom-API-from-Fortran.html">Accessing the wisdom API from Fortran</a></td></tr>
<tr><td></td><td valign="top"><a href="Multi_002dthreaded-FFTW.html#index-shared_002dmemory">shared-memory</a>:</td><td>&nbsp;</td><td valign="top"><a href="Multi_002dthreaded-FFTW.html">Multi-threaded FFTW</a></td></tr>
<tr><td></td><td valign="top"><a href="Complex-One_002dDimensional-DFTs.html#index-SIMD">SIMD</a>:</td><td>&nbsp;</td><td valign="top"><a href="Complex-One_002dDimensional-DFTs.html">Complex One-Dimensional DFTs</a></td></tr>
<tr><td></td><td valign="top"><a href="SIMD-alignment-and-fftw_005fmalloc.html#index-SIMD-1">SIMD</a>:</td><td>&nbsp;</td><td valign="top"><a href="SIMD-alignment-and-fftw_005fmalloc.html">SIMD alignment and fftw_malloc</a></td></tr>
<tr><td></td><td valign="top"><a href="Overview-of-Fortran-interface.html#index-SIMD-2">SIMD</a>:</td><td>&nbsp;</td><td valign="top"><a href="Overview-of-Fortran-interface.html">Overview of Fortran interface</a></td></tr>
<tr><td></td><td valign="top"><a href="Interleaved-and-split-arrays.html#index-split-format">split format</a>:</td><td>&nbsp;</td><td valign="top"><a href="Interleaved-and-split-arrays.html">Interleaved and split arrays</a></td></tr>
<tr><td></td><td valign="top"><a href="SIMD-alignment-and-fftw_005fmalloc.html#index-SSE">SSE</a>:</td><td>&nbsp;</td><td valign="top"><a href="SIMD-alignment-and-fftw_005fmalloc.html">SIMD alignment and fftw_malloc</a></td></tr>
<tr><td></td><td valign="top"><a href="SIMD-alignment-and-fftw_005fmalloc.html#index-SSE2">SSE2</a>:</td><td>&nbsp;</td><td valign="top"><a href="SIMD-alignment-and-fftw_005fmalloc.html">SIMD alignment and fftw_malloc</a></td></tr>
<tr><td></td><td valign="top"><a href="Row_002dmajor-Format.html#index-stride">stride</a>:</td><td>&nbsp;</td><td valign="top"><a href="Row_002dmajor-Format.html">Row-major Format</a></td></tr>
<tr><td></td><td valign="top"><a href="Advanced-Complex-DFTs.html#index-stride-1">stride</a>:</td><td>&nbsp;</td><td valign="top"><a href="Advanced-Complex-DFTs.html">Advanced Complex DFTs</a></td></tr>
<tr><td></td><td valign="top"><a href="Guru-vector-and-transform-sizes.html#index-stride-2">stride</a>:</td><td>&nbsp;</td><td valign="top"><a href="Guru-vector-and-transform-sizes.html">Guru vector and transform sizes</a></td></tr>
<tr><td></td><td valign="top"><a href="MPI-Plan-Creation.html#index-stride-3">stride</a>:</td><td>&nbsp;</td><td valign="top"><a href="MPI-Plan-Creation.html">MPI Plan Creation</a></td></tr>
<tr><td colspan="4"> <hr></td></tr>
<tr><th id="Concept-Index_cp_letter-T">T</th><td></td><td></td></tr>
<tr><td></td><td valign="top"><a href="Usage-of-Multi_002dthreaded-FFTW.html#index-thread-safety">thread safety</a>:</td><td>&nbsp;</td><td valign="top"><a href="Usage-of-Multi_002dthreaded-FFTW.html">Usage of Multi-threaded FFTW</a></td></tr>
<tr><td></td><td valign="top"><a href="Thread-safety.html#index-thread-safety-1">thread safety</a>:</td><td>&nbsp;</td><td valign="top"><a href="Thread-safety.html">Thread safety</a></td></tr>
<tr><td></td><td valign="top"><a href="Combining-MPI-and-Threads.html#index-thread-safety-2">thread safety</a>:</td><td>&nbsp;</td><td valign="top"><a href="Combining-MPI-and-Threads.html">Combining MPI and Threads</a></td></tr>
<tr><td></td><td valign="top"><a href="Multi_002dthreaded-FFTW.html#index-threads">threads</a>:</td><td>&nbsp;</td><td valign="top"><a href="Multi_002dthreaded-FFTW.html">Multi-threaded FFTW</a></td></tr>
<tr><td></td><td valign="top"><a href="Thread-safety.html#index-threads-1">threads</a>:</td><td>&nbsp;</td><td valign="top"><a href="Thread-safety.html">Thread safety</a></td></tr>
<tr><td></td><td valign="top"><a href="Combining-MPI-and-Threads.html#index-threads-2">threads</a>:</td><td>&nbsp;</td><td valign="top"><a href="Combining-MPI-and-Threads.html">Combining MPI and Threads</a></td></tr>
<tr><td></td><td valign="top"><a href="Installation-on-Unix.html#index-threads-3">threads</a>:</td><td>&nbsp;</td><td valign="top"><a href="Installation-on-Unix.html">Installation on Unix</a></td></tr>
<tr><td></td><td valign="top"><a href="Transposed-distributions.html#index-transpose">transpose</a>:</td><td>&nbsp;</td><td valign="top"><a href="Transposed-distributions.html">Transposed distributions</a></td></tr>
<tr><td></td><td valign="top"><a href="Multi_002ddimensional-MPI-DFTs-of-Real-Data.html#index-transpose-1">transpose</a>:</td><td>&nbsp;</td><td valign="top"><a href="Multi_002ddimensional-MPI-DFTs-of-Real-Data.html">Multi-dimensional MPI DFTs of Real Data</a></td></tr>
<tr><td></td><td valign="top"><a href="FFTW-MPI-Transposes.html#index-transpose-2">transpose</a>:</td><td>&nbsp;</td><td valign="top"><a href="FFTW-MPI-Transposes.html">FFTW MPI Transposes</a></td></tr>
<tr><td></td><td valign="top"><a href="FFTW-MPI-Performance-Tips.html#index-transpose-3">transpose</a>:</td><td>&nbsp;</td><td valign="top"><a href="FFTW-MPI-Performance-Tips.html">FFTW MPI Performance Tips</a></td></tr>
<tr><td></td><td valign="top"><a href="Combining-MPI-and-Threads.html#index-transpose-4">transpose</a>:</td><td>&nbsp;</td><td valign="top"><a href="Combining-MPI-and-Threads.html">Combining MPI and Threads</a></td></tr>
<tr><td></td><td valign="top"><a href="MPI-Plan-Creation.html#index-transpose-5">transpose</a>:</td><td>&nbsp;</td><td valign="top"><a href="MPI-Plan-Creation.html">MPI Plan Creation</a></td></tr>
<tr><td colspan="4"> <hr></td></tr>
<tr><th id="Concept-Index_cp_letter-V">V</th><td></td><td></td></tr>
<tr><td></td><td valign="top"><a href="Guru-Interface.html#index-vector">vector</a>:</td><td>&nbsp;</td><td valign="top"><a href="Guru-Interface.html">Guru Interface</a></td></tr>
<tr><td></td><td valign="top"><a href="SIMD-alignment-and-fftw_005fmalloc.html#index-VSX">VSX</a>:</td><td>&nbsp;</td><td valign="top"><a href="SIMD-alignment-and-fftw_005fmalloc.html">SIMD alignment and fftw_malloc</a></td></tr>
<tr><td colspan="4"> <hr></td></tr>
<tr><th id="Concept-Index_cp_letter-W">W</th><td></td><td></td></tr>
<tr><td></td><td valign="top"><a href="Words-of-Wisdom_002dSaving-Plans.html#index-wisdom">wisdom</a>:</td><td>&nbsp;</td><td valign="top"><a href="Words-of-Wisdom_002dSaving-Plans.html">Words of Wisdom-Saving Plans</a></td></tr>
<tr><td></td><td valign="top"><a href="Wisdom.html#index-wisdom-1">wisdom</a>:</td><td>&nbsp;</td><td valign="top"><a href="Wisdom.html">Wisdom</a></td></tr>
<tr><td></td><td valign="top"><a href="FFTW-MPI-Wisdom.html#index-wisdom-2">wisdom</a>:</td><td>&nbsp;</td><td valign="top"><a href="FFTW-MPI-Wisdom.html">FFTW MPI Wisdom</a></td></tr>
<tr><td></td><td valign="top"><a href="Accessing-the-wisdom-API-from-Fortran.html#index-wisdom-3">wisdom</a>:</td><td>&nbsp;</td><td valign="top"><a href="Accessing-the-wisdom-API-from-Fortran.html">Accessing the wisdom API from Fortran</a></td></tr>
<tr><td></td><td valign="top"><a href="Caveats-in-Using-Wisdom.html#index-wisdom_002c-problems-with">wisdom, problems with</a>:</td><td>&nbsp;</td><td valign="top"><a href="Caveats-in-Using-Wisdom.html">Caveats in Using Wisdom</a></td></tr>
<tr><td></td><td valign="top"><a href="Caveats-in-Using-Wisdom.html#index-wisdom_002c-system_002dwide">wisdom, system-wide</a>:</td><td>&nbsp;</td><td valign="top"><a href="Caveats-in-Using-Wisdom.html">Caveats in Using Wisdom</a></td></tr>
<tr><td></td><td valign="top"><a href="Wisdom-Import.html#index-wisdom_002c-system_002dwide-1">wisdom, system-wide</a>:</td><td>&nbsp;</td><td valign="top"><a href="Wisdom-Import.html">Wisdom Import</a></td></tr>
<tr><td colspan="4"> <hr></td></tr>
</table>
<table><tr><th valign="top">Jump to: &nbsp; </th><td><a class="summary-letter" href="#Concept-Index_cp_symbol-1"><b>6</b></a>
&nbsp;
<br>
<a class="summary-letter" href="#Concept-Index_cp_letter-A"><b>A</b></a>
&nbsp;
<a class="summary-letter" href="#Concept-Index_cp_letter-B"><b>B</b></a>
&nbsp;
<a class="summary-letter" href="#Concept-Index_cp_letter-C"><b>C</b></a>
&nbsp;
<a class="summary-letter" href="#Concept-Index_cp_letter-D"><b>D</b></a>
&nbsp;
<a class="summary-letter" href="#Concept-Index_cp_letter-E"><b>E</b></a>
&nbsp;
<a class="summary-letter" href="#Concept-Index_cp_letter-F"><b>F</b></a>
&nbsp;
<a class="summary-letter" href="#Concept-Index_cp_letter-G"><b>G</b></a>
&nbsp;
<a class="summary-letter" href="#Concept-Index_cp_letter-H"><b>H</b></a>
&nbsp;
<a class="summary-letter" href="#Concept-Index_cp_letter-I"><b>I</b></a>
&nbsp;
<a class="summary-letter" href="#Concept-Index_cp_letter-K"><b>K</b></a>
&nbsp;
<a class="summary-letter" href="#Concept-Index_cp_letter-L"><b>L</b></a>
&nbsp;
<a class="summary-letter" href="#Concept-Index_cp_letter-M"><b>M</b></a>
&nbsp;
<a class="summary-letter" href="#Concept-Index_cp_letter-N"><b>N</b></a>
&nbsp;
<a class="summary-letter" href="#Concept-Index_cp_letter-O"><b>O</b></a>
&nbsp;
<a class="summary-letter" href="#Concept-Index_cp_letter-P"><b>P</b></a>
&nbsp;
<a class="summary-letter" href="#Concept-Index_cp_letter-R"><b>R</b></a>
&nbsp;
<a class="summary-letter" href="#Concept-Index_cp_letter-S"><b>S</b></a>
&nbsp;
<a class="summary-letter" href="#Concept-Index_cp_letter-T"><b>T</b></a>
&nbsp;
<a class="summary-letter" href="#Concept-Index_cp_letter-V"><b>V</b></a>
&nbsp;
<a class="summary-letter" href="#Concept-Index_cp_letter-W"><b>W</b></a>
&nbsp;
</td></tr></table>
<hr>
<div class="header">
<p>
Next: <a href="Library-Index.html" accesskey="n" rel="next">Library Index</a>, Previous: <a href="License-and-Copyright.html" accesskey="p" rel="prev">License and Copyright</a>, Up: <a href="index.html" accesskey="u" rel="up">Top</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Concept-Index" title="Index" rel="index">Index</a>]</p>
</div>
</body>
</html>

View File

@@ -0,0 +1,109 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html>
<!-- This manual is for FFTW
(version 3.3.10, 10 December 2020).
Copyright (C) 2003 Matteo Frigo.
Copyright (C) 2003 Massachusetts Institute of Technology.
Permission is granted to make and distribute verbatim copies of this
manual provided the copyright notice and this permission notice are
preserved on all copies.
Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided that the
entire resulting derived work is distributed under the terms of a
permission notice identical to this one.
Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions,
except that this permission notice may be stated in a translation
approved by the Free Software Foundation. -->
<!-- Created by GNU Texinfo 6.7, http://www.gnu.org/software/texinfo/ -->
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<title>Cycle Counters (FFTW 3.3.10)</title>
<meta name="description" content="Cycle Counters (FFTW 3.3.10)">
<meta name="keywords" content="Cycle Counters (FFTW 3.3.10)">
<meta name="resource-type" content="document">
<meta name="distribution" content="global">
<meta name="Generator" content="makeinfo">
<link href="index.html" rel="start" title="Top">
<link href="Concept-Index.html" rel="index" title="Concept Index">
<link href="index.html#SEC_Contents" rel="contents" title="Table of Contents">
<link href="Installation-and-Customization.html" rel="up" title="Installation and Customization">
<link href="Generating-your-own-code.html" rel="next" title="Generating your own code">
<link href="Installation-on-non_002dUnix-systems.html" rel="prev" title="Installation on non-Unix systems">
<style type="text/css">
<!--
a.summary-letter {text-decoration: none}
blockquote.indentedblock {margin-right: 0em}
div.display {margin-left: 3.2em}
div.example {margin-left: 3.2em}
div.lisp {margin-left: 3.2em}
kbd {font-style: oblique}
pre.display {font-family: inherit}
pre.format {font-family: inherit}
pre.menu-comment {font-family: serif}
pre.menu-preformatted {font-family: serif}
span.nolinebreak {white-space: nowrap}
span.roman {font-family: initial; font-weight: normal}
span.sansserif {font-family: sans-serif; font-weight: normal}
ul.no-bullet {list-style: none}
-->
</style>
</head>
<body lang="en">
<span id="Cycle-Counters"></span><div class="header">
<p>
Next: <a href="Generating-your-own-code.html" accesskey="n" rel="next">Generating your own code</a>, Previous: <a href="Installation-on-non_002dUnix-systems.html" accesskey="p" rel="prev">Installation on non-Unix systems</a>, Up: <a href="Installation-and-Customization.html" accesskey="u" rel="up">Installation and Customization</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
<hr>
<span id="Cycle-Counters-1"></span><h3 class="section">10.3 Cycle Counters</h3>
<span id="index-cycle-counter-1"></span>
<p>FFTW&rsquo;s planner actually executes and times different possible FFT
algorithms in order to pick the fastest plan for a given <em>n</em>. In
order to do this in as short a time as possible, however, the timer must
have a very high resolution, and to accomplish this we employ the
hardware <em>cycle counters</em> that are available on most CPUs.
Currently, FFTW supports the cycle counters on x86, PowerPC/POWER, Alpha,
UltraSPARC (SPARC v9), IA64, PA-RISC, and MIPS processors.
</p>
<span id="index-compiler-3"></span>
<p>Access to the cycle counters, unfortunately, is a compiler and/or
operating-system dependent task, often requiring inline assembly
language, and it may be that your compiler is not supported. If you are
<em>not</em> supported, FFTW will by default fall back on its estimator
(effectively using <code>FFTW_ESTIMATE</code> for all plans).
<span id="index-FFTW_005fESTIMATE-3"></span>
</p>
<p>You can add support by editing the file <code>kernel/cycle.h</code>; normally,
this will involve adapting one of the examples already present in order
to use the inline-assembler syntax for your C compiler, and will only
require a couple of lines of code. Anyone adding support for a new
system to <code>cycle.h</code> is encouraged to email us at <a href="mailto:fftw@fftw.org">fftw@fftw.org</a>.
</p>
<p>If a cycle counter is not available on your system (e.g. some embedded
processor), and you don&rsquo;t want to use estimated plans, as a last resort
you can use the <code>--with-slow-timer</code> option to <code>configure</code> (on
Unix) or <code>#define WITH_SLOW_TIMER</code> in <code>config.h</code> (elsewhere).
This will use the much lower-resolution <code>gettimeofday</code> function, or even
<code>clock</code> if the former is unavailable, and planning will be
extremely slow.
</p>
<hr>
<div class="header">
<p>
Next: <a href="Generating-your-own-code.html" accesskey="n" rel="next">Generating your own code</a>, Previous: <a href="Installation-on-non_002dUnix-systems.html" accesskey="p" rel="prev">Installation on non-Unix systems</a>, Up: <a href="Installation-and-Customization.html" accesskey="u" rel="up">Installation and Customization</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
</body>
</html>

View File

@@ -0,0 +1,91 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html>
<!-- This manual is for FFTW
(version 3.3.10, 10 December 2020).
Copyright (C) 2003 Matteo Frigo.
Copyright (C) 2003 Massachusetts Institute of Technology.
Permission is granted to make and distribute verbatim copies of this
manual provided the copyright notice and this permission notice are
preserved on all copies.
Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided that the
entire resulting derived work is distributed under the terms of a
permission notice identical to this one.
Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions,
except that this permission notice may be stated in a translation
approved by the Free Software Foundation. -->
<!-- Created by GNU Texinfo 6.7, http://www.gnu.org/software/texinfo/ -->
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<title>Data Types and Files (FFTW 3.3.10)</title>
<meta name="description" content="Data Types and Files (FFTW 3.3.10)">
<meta name="keywords" content="Data Types and Files (FFTW 3.3.10)">
<meta name="resource-type" content="document">
<meta name="distribution" content="global">
<meta name="Generator" content="makeinfo">
<link href="index.html" rel="start" title="Top">
<link href="Concept-Index.html" rel="index" title="Concept Index">
<link href="index.html#SEC_Contents" rel="contents" title="Table of Contents">
<link href="FFTW-Reference.html" rel="up" title="FFTW Reference">
<link href="Complex-numbers.html" rel="next" title="Complex numbers">
<link href="FFTW-Reference.html" rel="prev" title="FFTW Reference">
<style type="text/css">
<!--
a.summary-letter {text-decoration: none}
blockquote.indentedblock {margin-right: 0em}
div.display {margin-left: 3.2em}
div.example {margin-left: 3.2em}
div.lisp {margin-left: 3.2em}
kbd {font-style: oblique}
pre.display {font-family: inherit}
pre.format {font-family: inherit}
pre.menu-comment {font-family: serif}
pre.menu-preformatted {font-family: serif}
span.nolinebreak {white-space: nowrap}
span.roman {font-family: initial; font-weight: normal}
span.sansserif {font-family: sans-serif; font-weight: normal}
ul.no-bullet {list-style: none}
-->
</style>
</head>
<body lang="en">
<span id="Data-Types-and-Files"></span><div class="header">
<p>
Next: <a href="Using-Plans.html" accesskey="n" rel="next">Using Plans</a>, Previous: <a href="FFTW-Reference.html" accesskey="p" rel="prev">FFTW Reference</a>, Up: <a href="FFTW-Reference.html" accesskey="u" rel="up">FFTW Reference</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
<hr>
<span id="Data-Types-and-Files-1"></span><h3 class="section">4.1 Data Types and Files</h3>
<p>All programs using FFTW should include its header file:
</p>
<div class="example">
<pre class="example">#include &lt;fftw3.h&gt;
</pre></div>
<p>You must also link to the FFTW library. On Unix, this
means adding <code>-lfftw3 -lm</code> at the <em>end</em> of the link command.
</p>
<table class="menu" border="0" cellspacing="0">
<tr><td align="left" valign="top">&bull; <a href="Complex-numbers.html" accesskey="1">Complex numbers</a></td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="Precision.html" accesskey="2">Precision</a></td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="Memory-Allocation.html" accesskey="3">Memory Allocation</a></td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
</table>
</body>
</html>

View File

@@ -0,0 +1,99 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html>
<!-- This manual is for FFTW
(version 3.3.10, 10 December 2020).
Copyright (C) 2003 Matteo Frigo.
Copyright (C) 2003 Massachusetts Institute of Technology.
Permission is granted to make and distribute verbatim copies of this
manual provided the copyright notice and this permission notice are
preserved on all copies.
Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided that the
entire resulting derived work is distributed under the terms of a
permission notice identical to this one.
Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions,
except that this permission notice may be stated in a translation
approved by the Free Software Foundation. -->
<!-- Created by GNU Texinfo 6.7, http://www.gnu.org/software/texinfo/ -->
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<title>Defining an FFTW module (FFTW 3.3.10)</title>
<meta name="description" content="Defining an FFTW module (FFTW 3.3.10)">
<meta name="keywords" content="Defining an FFTW module (FFTW 3.3.10)">
<meta name="resource-type" content="document">
<meta name="distribution" content="global">
<meta name="Generator" content="makeinfo">
<link href="index.html" rel="start" title="Top">
<link href="Concept-Index.html" rel="index" title="Concept Index">
<link href="index.html#SEC_Contents" rel="contents" title="Table of Contents">
<link href="Calling-FFTW-from-Modern-Fortran.html" rel="up" title="Calling FFTW from Modern Fortran">
<link href="Calling-FFTW-from-Legacy-Fortran.html" rel="next" title="Calling FFTW from Legacy Fortran">
<link href="Wisdom-Generic-Export_002fImport-from-Fortran.html" rel="prev" title="Wisdom Generic Export/Import from Fortran">
<style type="text/css">
<!--
a.summary-letter {text-decoration: none}
blockquote.indentedblock {margin-right: 0em}
div.display {margin-left: 3.2em}
div.example {margin-left: 3.2em}
div.lisp {margin-left: 3.2em}
kbd {font-style: oblique}
pre.display {font-family: inherit}
pre.format {font-family: inherit}
pre.menu-comment {font-family: serif}
pre.menu-preformatted {font-family: serif}
span.nolinebreak {white-space: nowrap}
span.roman {font-family: initial; font-weight: normal}
span.sansserif {font-family: sans-serif; font-weight: normal}
ul.no-bullet {list-style: none}
-->
</style>
</head>
<body lang="en">
<span id="Defining-an-FFTW-module"></span><div class="header">
<p>
Previous: <a href="Accessing-the-wisdom-API-from-Fortran.html" accesskey="p" rel="prev">Accessing the wisdom API from Fortran</a>, Up: <a href="Calling-FFTW-from-Modern-Fortran.html" accesskey="u" rel="up">Calling FFTW from Modern Fortran</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
<hr>
<span id="Defining-an-FFTW-module-1"></span><h3 class="section">7.7 Defining an FFTW module</h3>
<p>Rather than using the <code>include</code> statement to include the
<code>fftw3.f03</code> interface file in any subroutine where you want to
use FFTW, you might prefer to define an FFTW Fortran module. FFTW
does not install itself as a module, primarily because
<code>fftw3.f03</code> can be shared between different Fortran compilers while
modules (in general) cannot. However, it is trivial to define your
own FFTW module if you want. Just create a file containing:
</p>
<div class="example">
<pre class="example"> module FFTW3
use, intrinsic :: iso_c_binding
include 'fftw3.f03'
end module
</pre></div>
<p>Compile this file into a module as usual for your compiler (e.g. with
<code>gfortran -c</code> you will get a file <code>fftw3.mod</code>). Now,
instead of <code>include 'fftw3.f03'</code>, whenever you want to use FFTW
routines you can just do:
</p>
<div class="example">
<pre class="example"> use FFTW3
</pre></div>
<p>as usual for Fortran modules. (You still need to link to the FFTW
library, of course.)
</p>
</body>
</html>

View File

@@ -0,0 +1,150 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html>
<!-- This manual is for FFTW
(version 3.3.10, 10 December 2020).
Copyright (C) 2003 Matteo Frigo.
Copyright (C) 2003 Massachusetts Institute of Technology.
Permission is granted to make and distribute verbatim copies of this
manual provided the copyright notice and this permission notice are
preserved on all copies.
Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided that the
entire resulting derived work is distributed under the terms of a
permission notice identical to this one.
Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions,
except that this permission notice may be stated in a translation
approved by the Free Software Foundation. -->
<!-- Created by GNU Texinfo 6.7, http://www.gnu.org/software/texinfo/ -->
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<title>Distributed-memory FFTW with MPI (FFTW 3.3.10)</title>
<meta name="description" content="Distributed-memory FFTW with MPI (FFTW 3.3.10)">
<meta name="keywords" content="Distributed-memory FFTW with MPI (FFTW 3.3.10)">
<meta name="resource-type" content="document">
<meta name="distribution" content="global">
<meta name="Generator" content="makeinfo">
<link href="index.html" rel="start" title="Top">
<link href="Concept-Index.html" rel="index" title="Concept Index">
<link href="index.html#SEC_Contents" rel="contents" title="Table of Contents">
<link href="index.html" rel="up" title="Top">
<link href="FFTW-MPI-Installation.html" rel="next" title="FFTW MPI Installation">
<link href="Thread-safety.html" rel="prev" title="Thread safety">
<style type="text/css">
<!--
a.summary-letter {text-decoration: none}
blockquote.indentedblock {margin-right: 0em}
div.display {margin-left: 3.2em}
div.example {margin-left: 3.2em}
div.lisp {margin-left: 3.2em}
kbd {font-style: oblique}
pre.display {font-family: inherit}
pre.format {font-family: inherit}
pre.menu-comment {font-family: serif}
pre.menu-preformatted {font-family: serif}
span.nolinebreak {white-space: nowrap}
span.roman {font-family: initial; font-weight: normal}
span.sansserif {font-family: sans-serif; font-weight: normal}
ul.no-bullet {list-style: none}
-->
</style>
</head>
<body lang="en">
<span id="Distributed_002dmemory-FFTW-with-MPI"></span><div class="header">
<p>
Next: <a href="Calling-FFTW-from-Modern-Fortran.html" accesskey="n" rel="next">Calling FFTW from Modern Fortran</a>, Previous: <a href="Multi_002dthreaded-FFTW.html" accesskey="p" rel="prev">Multi-threaded FFTW</a>, Up: <a href="index.html" accesskey="u" rel="up">Top</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
<hr>
<span id="Distributed_002dmemory-FFTW-with-MPI-1"></span><h2 class="chapter">6 Distributed-memory FFTW with MPI</h2>
<span id="index-MPI"></span>
<span id="index-parallel-transform-1"></span>
<p>In this chapter we document the parallel FFTW routines for parallel
systems supporting the MPI message-passing interface. Unlike the
shared-memory threads described in the previous chapter, MPI allows
you to use <em>distributed-memory</em> parallelism, where each CPU has
its own separate memory, and which can scale up to clusters of many
thousands of processors. This capability comes at a price, however:
each process only stores a <em>portion</em> of the data to be
transformed, which means that the data structures and
programming-interface are quite different from the serial or threads
versions of FFTW.
<span id="index-data-distribution"></span>
</p>
<p>Distributed-memory parallelism is especially useful when you are
transforming arrays so large that they do not fit into the memory of a
single processor. The storage per-process required by FFTW&rsquo;s MPI
routines is proportional to the total array size divided by the number
of processes. Conversely, distributed-memory parallelism can easily
pose an unacceptably high communications overhead for small problems;
the threshold problem size for which parallelism becomes advantageous
will depend on the precise problem you are interested in, your
hardware, and your MPI implementation.
</p>
<p>A note on terminology: in MPI, you divide the data among a set of
&ldquo;processes&rdquo; which each run in their own memory address space.
Generally, each process runs on a different physical processor, but
this is not required. A set of processes in MPI is described by an
opaque data structure called a &ldquo;communicator,&rdquo; the most common of
which is the predefined communicator <code>MPI_COMM_WORLD</code> which
refers to <em>all</em> processes. For more information on these and
other concepts common to all MPI programs, we refer the reader to the
documentation at <a href="http://www.mcs.anl.gov/research/projects/mpi/">the MPI home
page</a>.
<span id="index-MPI-communicator"></span>
<span id="index-MPI_005fCOMM_005fWORLD"></span>
</p>
<p>We assume in this chapter that the reader is familiar with the usage
of the serial (uniprocessor) FFTW, and focus only on the concepts new
to the MPI interface.
</p>
<table class="menu" border="0" cellspacing="0">
<tr><td align="left" valign="top">&bull; <a href="FFTW-MPI-Installation.html" accesskey="1">FFTW MPI Installation</a></td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="Linking-and-Initializing-MPI-FFTW.html" accesskey="2">Linking and Initializing MPI FFTW</a></td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="2d-MPI-example.html" accesskey="3">2d MPI example</a></td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="MPI-Data-Distribution.html" accesskey="4">MPI Data Distribution</a></td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="Multi_002ddimensional-MPI-DFTs-of-Real-Data.html" accesskey="5">Multi-dimensional MPI DFTs of Real Data</a></td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="Other-Multi_002ddimensional-Real_002ddata-MPI-Transforms.html" accesskey="6">Other Multi-dimensional Real-data MPI Transforms</a></td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="FFTW-MPI-Transposes.html" accesskey="7">FFTW MPI Transposes</a></td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="FFTW-MPI-Wisdom.html" accesskey="8">FFTW MPI Wisdom</a></td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="Avoiding-MPI-Deadlocks.html" accesskey="9">Avoiding MPI Deadlocks</a></td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="FFTW-MPI-Performance-Tips.html">FFTW MPI Performance Tips</a></td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="Combining-MPI-and-Threads.html">Combining MPI and Threads</a></td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="FFTW-MPI-Reference.html">FFTW MPI Reference</a></td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="FFTW-MPI-Fortran-Interface.html">FFTW MPI Fortran Interface</a></td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
</table>
<hr>
<div class="header">
<p>
Next: <a href="Calling-FFTW-from-Modern-Fortran.html" accesskey="n" rel="next">Calling FFTW from Modern Fortran</a>, Previous: <a href="Multi_002dthreaded-FFTW.html" accesskey="p" rel="prev">Multi-threaded FFTW</a>, Up: <a href="index.html" accesskey="u" rel="up">Top</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
</body>
</html>

View File

@@ -0,0 +1,108 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html>
<!-- This manual is for FFTW
(version 3.3.10, 10 December 2020).
Copyright (C) 2003 Matteo Frigo.
Copyright (C) 2003 Massachusetts Institute of Technology.
Permission is granted to make and distribute verbatim copies of this
manual provided the copyright notice and this permission notice are
preserved on all copies.
Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided that the
entire resulting derived work is distributed under the terms of a
permission notice identical to this one.
Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions,
except that this permission notice may be stated in a translation
approved by the Free Software Foundation. -->
<!-- Created by GNU Texinfo 6.7, http://www.gnu.org/software/texinfo/ -->
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<title>Dynamic Arrays in C (FFTW 3.3.10)</title>
<meta name="description" content="Dynamic Arrays in C (FFTW 3.3.10)">
<meta name="keywords" content="Dynamic Arrays in C (FFTW 3.3.10)">
<meta name="resource-type" content="document">
<meta name="distribution" content="global">
<meta name="Generator" content="makeinfo">
<link href="index.html" rel="start" title="Top">
<link href="Concept-Index.html" rel="index" title="Concept Index">
<link href="index.html#SEC_Contents" rel="contents" title="Table of Contents">
<link href="Multi_002ddimensional-Array-Format.html" rel="up" title="Multi-dimensional Array Format">
<link href="Dynamic-Arrays-in-C_002dThe-Wrong-Way.html" rel="next" title="Dynamic Arrays in C-The Wrong Way">
<link href="Fixed_002dsize-Arrays-in-C.html" rel="prev" title="Fixed-size Arrays in C">
<style type="text/css">
<!--
a.summary-letter {text-decoration: none}
blockquote.indentedblock {margin-right: 0em}
div.display {margin-left: 3.2em}
div.example {margin-left: 3.2em}
div.lisp {margin-left: 3.2em}
kbd {font-style: oblique}
pre.display {font-family: inherit}
pre.format {font-family: inherit}
pre.menu-comment {font-family: serif}
pre.menu-preformatted {font-family: serif}
span.nolinebreak {white-space: nowrap}
span.roman {font-family: initial; font-weight: normal}
span.sansserif {font-family: sans-serif; font-weight: normal}
ul.no-bullet {list-style: none}
-->
</style>
</head>
<body lang="en">
<span id="Dynamic-Arrays-in-C"></span><div class="header">
<p>
Next: <a href="Dynamic-Arrays-in-C_002dThe-Wrong-Way.html" accesskey="n" rel="next">Dynamic Arrays in C-The Wrong Way</a>, Previous: <a href="Fixed_002dsize-Arrays-in-C.html" accesskey="p" rel="prev">Fixed-size Arrays in C</a>, Up: <a href="Multi_002ddimensional-Array-Format.html" accesskey="u" rel="up">Multi-dimensional Array Format</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
<hr>
<span id="Dynamic-Arrays-in-C-1"></span><h4 class="subsection">3.2.4 Dynamic Arrays in C</h4>
<p>We recommend allocating most arrays dynamically, with
<code>fftw_malloc</code>. This isn&rsquo;t too hard to do, although it is not as
straightforward for multi-dimensional arrays as it is for
one-dimensional arrays.
</p>
<p>Creating the array is simple: using a dynamic-allocation routine like
<code>fftw_malloc</code>, allocate an array big enough to store N
<code>fftw_complex</code> values (for a complex DFT), where N is the product
of the sizes of the array dimensions (i.e. the total number of complex
values in the array). For example, here is code to allocate a
5&nbsp;&times;&nbsp;12&nbsp;&times;&nbsp;27
rank-3 array:
<span id="index-fftw_005fmalloc-2"></span>
</p>
<div class="example">
<pre class="example">fftw_complex *an_array;
an_array = (fftw_complex*) fftw_malloc(5*12*27 * sizeof(fftw_complex));
</pre></div>
<p>Accessing the array elements, however, is more tricky&mdash;you can&rsquo;t
simply use multiple applications of the &lsquo;<samp>[]</samp>&rsquo; operator like you
could for fixed-size arrays. Instead, you have to explicitly compute
the offset into the array using the formula given earlier for
row-major arrays. For example, to reference the <em>(i,j,k)</em>-th
element of the array allocated above, you would use the expression
<code>an_array[k + 27 * (j + 12 * i)]</code>.
</p>
<p>This pain can be alleviated somewhat by defining appropriate macros,
or, in C++, creating a class and overloading the &lsquo;<samp>()</samp>&rsquo; operator.
The recent C99 standard provides a way to reinterpret the dynamic
array as a &ldquo;variable-length&rdquo; multi-dimensional array amenable to
&lsquo;<samp>[]</samp>&rsquo;, but this feature is not yet widely supported by compilers.
<span id="index-C99"></span>
<span id="index-C_002b_002b-2"></span>
</p>
</body>
</html>

View File

@@ -0,0 +1,108 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html>
<!-- This manual is for FFTW
(version 3.3.10, 10 December 2020).
Copyright (C) 2003 Matteo Frigo.
Copyright (C) 2003 Massachusetts Institute of Technology.
Permission is granted to make and distribute verbatim copies of this
manual provided the copyright notice and this permission notice are
preserved on all copies.
Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided that the
entire resulting derived work is distributed under the terms of a
permission notice identical to this one.
Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions,
except that this permission notice may be stated in a translation
approved by the Free Software Foundation. -->
<!-- Created by GNU Texinfo 6.7, http://www.gnu.org/software/texinfo/ -->
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<title>Dynamic Arrays in C-The Wrong Way (FFTW 3.3.10)</title>
<meta name="description" content="Dynamic Arrays in C-The Wrong Way (FFTW 3.3.10)">
<meta name="keywords" content="Dynamic Arrays in C-The Wrong Way (FFTW 3.3.10)">
<meta name="resource-type" content="document">
<meta name="distribution" content="global">
<meta name="Generator" content="makeinfo">
<link href="index.html" rel="start" title="Top">
<link href="Concept-Index.html" rel="index" title="Concept Index">
<link href="index.html#SEC_Contents" rel="contents" title="Table of Contents">
<link href="Multi_002ddimensional-Array-Format.html" rel="up" title="Multi-dimensional Array Format">
<link href="Words-of-Wisdom_002dSaving-Plans.html" rel="next" title="Words of Wisdom-Saving Plans">
<link href="Dynamic-Arrays-in-C.html" rel="prev" title="Dynamic Arrays in C">
<style type="text/css">
<!--
a.summary-letter {text-decoration: none}
blockquote.indentedblock {margin-right: 0em}
div.display {margin-left: 3.2em}
div.example {margin-left: 3.2em}
div.lisp {margin-left: 3.2em}
kbd {font-style: oblique}
pre.display {font-family: inherit}
pre.format {font-family: inherit}
pre.menu-comment {font-family: serif}
pre.menu-preformatted {font-family: serif}
span.nolinebreak {white-space: nowrap}
span.roman {font-family: initial; font-weight: normal}
span.sansserif {font-family: sans-serif; font-weight: normal}
ul.no-bullet {list-style: none}
-->
</style>
</head>
<body lang="en">
<span id="Dynamic-Arrays-in-C_002dThe-Wrong-Way"></span><div class="header">
<p>
Previous: <a href="Dynamic-Arrays-in-C.html" accesskey="p" rel="prev">Dynamic Arrays in C</a>, Up: <a href="Multi_002ddimensional-Array-Format.html" accesskey="u" rel="up">Multi-dimensional Array Format</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
<hr>
<span id="Dynamic-Arrays-in-C_002d_002d_002dThe-Wrong-Way"></span><h4 class="subsection">3.2.5 Dynamic Arrays in C&mdash;The Wrong Way</h4>
<p>A different method for allocating multi-dimensional arrays in C is
often suggested that is incompatible with FFTW: <em>using it will
cause FFTW to die a painful death</em>. We discuss the technique here,
however, because it is so commonly known and used. This method is to
create arrays of pointers of arrays of pointers of &hellip;etcetera.
For example, the analogue in this method to the example above is:
</p>
<div class="example">
<pre class="example">int i,j;
fftw_complex ***a_bad_array; /* <span class="roman">another way to make a 5x12x27 array</span> */
a_bad_array = (fftw_complex ***) malloc(5 * sizeof(fftw_complex **));
for (i = 0; i &lt; 5; ++i) {
a_bad_array[i] =
(fftw_complex **) malloc(12 * sizeof(fftw_complex *));
for (j = 0; j &lt; 12; ++j)
a_bad_array[i][j] =
(fftw_complex *) malloc(27 * sizeof(fftw_complex));
}
</pre></div>
<p>As you can see, this sort of array is inconvenient to allocate (and
deallocate). On the other hand, it has the advantage that the
<em>(i,j,k)</em>-th element can be referenced simply by
<code>a_bad_array[i][j][k]</code>.
</p>
<p>If you like this technique and want to maximize convenience in accessing
the array, but still want to pass the array to FFTW, you can use a
hybrid method. Allocate the array as one contiguous block, but also
declare an array of arrays of pointers that point to appropriate places
in the block. That sort of trick is beyond the scope of this
documentation; for more information on multi-dimensional arrays in C,
see the <code>comp.lang.c</code>
<a href="http://c-faq.com/aryptr/dynmuldimary.html">FAQ</a>.
</p>
</body>
</html>

View File

@@ -0,0 +1,102 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html>
<!-- This manual is for FFTW
(version 3.3.10, 10 December 2020).
Copyright (C) 2003 Matteo Frigo.
Copyright (C) 2003 Massachusetts Institute of Technology.
Permission is granted to make and distribute verbatim copies of this
manual provided the copyright notice and this permission notice are
preserved on all copies.
Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided that the
entire resulting derived work is distributed under the terms of a
permission notice identical to this one.
Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions,
except that this permission notice may be stated in a translation
approved by the Free Software Foundation. -->
<!-- Created by GNU Texinfo 6.7, http://www.gnu.org/software/texinfo/ -->
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<title>Extended and quadruple precision in Fortran (FFTW 3.3.10)</title>
<meta name="description" content="Extended and quadruple precision in Fortran (FFTW 3.3.10)">
<meta name="keywords" content="Extended and quadruple precision in Fortran (FFTW 3.3.10)">
<meta name="resource-type" content="document">
<meta name="distribution" content="global">
<meta name="Generator" content="makeinfo">
<link href="index.html" rel="start" title="Top">
<link href="Concept-Index.html" rel="index" title="Concept Index">
<link href="index.html#SEC_Contents" rel="contents" title="Table of Contents">
<link href="Overview-of-Fortran-interface.html" rel="up" title="Overview of Fortran interface">
<link href="Reversing-array-dimensions.html" rel="next" title="Reversing array dimensions">
<link href="Overview-of-Fortran-interface.html" rel="prev" title="Overview of Fortran interface">
<style type="text/css">
<!--
a.summary-letter {text-decoration: none}
blockquote.indentedblock {margin-right: 0em}
div.display {margin-left: 3.2em}
div.example {margin-left: 3.2em}
div.lisp {margin-left: 3.2em}
kbd {font-style: oblique}
pre.display {font-family: inherit}
pre.format {font-family: inherit}
pre.menu-comment {font-family: serif}
pre.menu-preformatted {font-family: serif}
span.nolinebreak {white-space: nowrap}
span.roman {font-family: initial; font-weight: normal}
span.sansserif {font-family: sans-serif; font-weight: normal}
ul.no-bullet {list-style: none}
-->
</style>
</head>
<body lang="en">
<span id="Extended-and-quadruple-precision-in-Fortran"></span><div class="header">
<p>
Previous: <a href="Overview-of-Fortran-interface.html" accesskey="p" rel="prev">Overview of Fortran interface</a>, Up: <a href="Overview-of-Fortran-interface.html" accesskey="u" rel="up">Overview of Fortran interface</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
<hr>
<span id="Extended-and-quadruple-precision-in-Fortran-1"></span><h4 class="subsection">7.1.1 Extended and quadruple precision in Fortran</h4>
<span id="index-precision-7"></span>
<p>If FFTW is compiled in <code>long double</code> (extended) precision
(see <a href="Installation-and-Customization.html">Installation and Customization</a>), you may be able to call the
resulting <code>fftwl_</code> routines (see <a href="Precision.html">Precision</a>) from Fortran if
your compiler supports the <code>C_LONG_DOUBLE_COMPLEX</code> type code.
</p>
<p>Because some Fortran compilers do not support
<code>C_LONG_DOUBLE_COMPLEX</code>, the <code>fftwl_</code> declarations are
segregated into a separate interface file <code>fftw3l.f03</code>, which you
should include <em>in addition</em> to <code>fftw3.f03</code> (which declares
precision-independent &lsquo;<samp>FFTW_</samp>&rsquo; constants):
</p>
<span id="index-iso_005fc_005fbinding-2"></span>
<div class="example">
<pre class="example"> use, intrinsic :: iso_c_binding
include 'fftw3.f03'
include 'fftw3l.f03'
</pre></div>
<p>We also support using the nonstandard <code>__float128</code>
quadruple-precision type provided by recent versions of <code>gcc</code> on
32- and 64-bit x86 hardware (see <a href="Installation-and-Customization.html">Installation and Customization</a>),
using the corresponding <code>real(16)</code> and <code>complex(16)</code> types
supported by <code>gfortran</code>. The quadruple-precision &lsquo;<samp>fftwq_</samp>&rsquo;
functions (see <a href="Precision.html">Precision</a>) are declared in a <code>fftw3q.f03</code>
interface file, which should be included in addition to
<code>fftw3.f03</code>, as above. You should also link with
<code>-lfftw3q -lquadmath -lm</code> as in C.
</p>
</body>
</html>

View File

@@ -0,0 +1,93 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html>
<!-- This manual is for FFTW
(version 3.3.10, 10 December 2020).
Copyright (C) 2003 Matteo Frigo.
Copyright (C) 2003 Massachusetts Institute of Technology.
Permission is granted to make and distribute verbatim copies of this
manual provided the copyright notice and this permission notice are
preserved on all copies.
Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided that the
entire resulting derived work is distributed under the terms of a
permission notice identical to this one.
Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions,
except that this permission notice may be stated in a translation
approved by the Free Software Foundation. -->
<!-- Created by GNU Texinfo 6.7, http://www.gnu.org/software/texinfo/ -->
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<title>FFTW Constants in Fortran (FFTW 3.3.10)</title>
<meta name="description" content="FFTW Constants in Fortran (FFTW 3.3.10)">
<meta name="keywords" content="FFTW Constants in Fortran (FFTW 3.3.10)">
<meta name="resource-type" content="document">
<meta name="distribution" content="global">
<meta name="Generator" content="makeinfo">
<link href="index.html" rel="start" title="Top">
<link href="Concept-Index.html" rel="index" title="Concept Index">
<link href="index.html#SEC_Contents" rel="contents" title="Table of Contents">
<link href="Calling-FFTW-from-Legacy-Fortran.html" rel="up" title="Calling FFTW from Legacy Fortran">
<link href="FFTW-Execution-in-Fortran.html" rel="next" title="FFTW Execution in Fortran">
<link href="Fortran_002dinterface-routines.html" rel="prev" title="Fortran-interface routines">
<style type="text/css">
<!--
a.summary-letter {text-decoration: none}
blockquote.indentedblock {margin-right: 0em}
div.display {margin-left: 3.2em}
div.example {margin-left: 3.2em}
div.lisp {margin-left: 3.2em}
kbd {font-style: oblique}
pre.display {font-family: inherit}
pre.format {font-family: inherit}
pre.menu-comment {font-family: serif}
pre.menu-preformatted {font-family: serif}
span.nolinebreak {white-space: nowrap}
span.roman {font-family: initial; font-weight: normal}
span.sansserif {font-family: sans-serif; font-weight: normal}
ul.no-bullet {list-style: none}
-->
</style>
</head>
<body lang="en">
<span id="FFTW-Constants-in-Fortran"></span><div class="header">
<p>
Next: <a href="FFTW-Execution-in-Fortran.html" accesskey="n" rel="next">FFTW Execution in Fortran</a>, Previous: <a href="Fortran_002dinterface-routines.html" accesskey="p" rel="prev">Fortran-interface routines</a>, Up: <a href="Calling-FFTW-from-Legacy-Fortran.html" accesskey="u" rel="up">Calling FFTW from Legacy Fortran</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
<hr>
<span id="FFTW-Constants-in-Fortran-1"></span><h3 class="section">8.2 FFTW Constants in Fortran</h3>
<p>When creating plans in FFTW, a number of constants are used to specify
options, such as <code>FFTW_MEASURE</code> or <code>FFTW_ESTIMATE</code>. The
same constants must be used with the wrapper routines, but of course the
C header files where the constants are defined can&rsquo;t be incorporated
directly into Fortran code.
</p>
<p>Instead, we have placed Fortran equivalents of the FFTW constant
definitions in the file <code>fftw3.f</code>, which can be found in the same
directory as <code>fftw3.h</code>. If your Fortran compiler supports a
preprocessor of some sort, you should be able to <code>include</code> or
<code>#include</code> this file; otherwise, you can paste it directly into
your code.
</p>
<span id="index-flags-9"></span>
<p>In C, you combine different flags (like <code>FFTW_PRESERVE_INPUT</code> and
<code>FFTW_MEASURE</code>) using the &lsquo;<samp><code>|</code></samp>&rsquo; operator; in Fortran
you should just use &lsquo;<samp><code>+</code></samp>&rsquo;. (Take care not to add in the
same flag more than once, though. Alternatively, you can use the
<code>ior</code> intrinsic function standardized in Fortran 95.)
</p>
</body>
</html>

View File

@@ -0,0 +1,134 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html>
<!-- This manual is for FFTW
(version 3.3.10, 10 December 2020).
Copyright (C) 2003 Matteo Frigo.
Copyright (C) 2003 Massachusetts Institute of Technology.
Permission is granted to make and distribute verbatim copies of this
manual provided the copyright notice and this permission notice are
preserved on all copies.
Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided that the
entire resulting derived work is distributed under the terms of a
permission notice identical to this one.
Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions,
except that this permission notice may be stated in a translation
approved by the Free Software Foundation. -->
<!-- Created by GNU Texinfo 6.7, http://www.gnu.org/software/texinfo/ -->
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<title>FFTW Execution in Fortran (FFTW 3.3.10)</title>
<meta name="description" content="FFTW Execution in Fortran (FFTW 3.3.10)">
<meta name="keywords" content="FFTW Execution in Fortran (FFTW 3.3.10)">
<meta name="resource-type" content="document">
<meta name="distribution" content="global">
<meta name="Generator" content="makeinfo">
<link href="index.html" rel="start" title="Top">
<link href="Concept-Index.html" rel="index" title="Concept Index">
<link href="index.html#SEC_Contents" rel="contents" title="Table of Contents">
<link href="Calling-FFTW-from-Legacy-Fortran.html" rel="up" title="Calling FFTW from Legacy Fortran">
<link href="Fortran-Examples.html" rel="next" title="Fortran Examples">
<link href="FFTW-Constants-in-Fortran.html" rel="prev" title="FFTW Constants in Fortran">
<style type="text/css">
<!--
a.summary-letter {text-decoration: none}
blockquote.indentedblock {margin-right: 0em}
div.display {margin-left: 3.2em}
div.example {margin-left: 3.2em}
div.lisp {margin-left: 3.2em}
kbd {font-style: oblique}
pre.display {font-family: inherit}
pre.format {font-family: inherit}
pre.menu-comment {font-family: serif}
pre.menu-preformatted {font-family: serif}
span.nolinebreak {white-space: nowrap}
span.roman {font-family: initial; font-weight: normal}
span.sansserif {font-family: sans-serif; font-weight: normal}
ul.no-bullet {list-style: none}
-->
</style>
</head>
<body lang="en">
<span id="FFTW-Execution-in-Fortran"></span><div class="header">
<p>
Next: <a href="Fortran-Examples.html" accesskey="n" rel="next">Fortran Examples</a>, Previous: <a href="FFTW-Constants-in-Fortran.html" accesskey="p" rel="prev">FFTW Constants in Fortran</a>, Up: <a href="Calling-FFTW-from-Legacy-Fortran.html" accesskey="u" rel="up">Calling FFTW from Legacy Fortran</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
<hr>
<span id="FFTW-Execution-in-Fortran-1"></span><h3 class="section">8.3 FFTW Execution in Fortran</h3>
<p>In C, in order to use a plan, one normally calls <code>fftw_execute</code>,
which executes the plan to perform the transform on the input/output
arrays passed when the plan was created (see <a href="Using-Plans.html">Using Plans</a>). The
corresponding subroutine call in legacy Fortran is:
</p><div class="example">
<pre class="example"> call dfftw_execute(plan)
</pre></div>
<span id="index-dfftw_005fexecute"></span>
<p>However, we have had reports that this causes problems with some
recent optimizing Fortran compilers. The problem is, because the
input/output arrays are not passed as explicit arguments to
<code>dfftw_execute</code>, the semantics of Fortran (unlike C) allow the
compiler to assume that the input/output arrays are not changed by
<code>dfftw_execute</code>. As a consequence, certain compilers end up
optimizing out or repositioning the call to <code>dfftw_execute</code>,
assuming incorrectly that it does nothing.
</p>
<p>There are various workarounds to this, but the safest and simplest
thing is to not use <code>dfftw_execute</code> in Fortran. Instead, use the
functions described in <a href="New_002darray-Execute-Functions.html">New-array Execute Functions</a>, which take
the input/output arrays as explicit arguments. For example, if the
plan is for a complex-data DFT and was created for the arrays
<code>in</code> and <code>out</code>, you would do:
</p><div class="example">
<pre class="example"> call dfftw_execute_dft(plan, in, out)
</pre></div>
<span id="index-dfftw_005fexecute_005fdft"></span>
<p>There are a few things to be careful of, however:
</p>
<ul>
<li> You must use the correct type of execute function, matching the way
the plan was created. Complex DFT plans should use
<code>dfftw_execute_dft</code>, Real-input (r2c) DFT plans should use use
<code>dfftw_execute_dft_r2c</code>, and real-output (c2r) DFT plans should
use <code>dfftw_execute_dft_c2r</code>. The various r2r plans should use
<code>dfftw_execute_r2r</code>.
</li><li> You should normally pass the same input/output arrays that were used when
creating the plan. This is always safe.
</li><li> <em>If</em> you pass <em>different</em> input/output arrays compared to
those used when creating the plan, you must abide by all the
restrictions of the new-array execute functions (see <a href="New_002darray-Execute-Functions.html">New-array Execute Functions</a>). The most difficult of these, in Fortran, is the
requirement that the new arrays have the same alignment as the
original arrays, because there seems to be no way in legacy Fortran to obtain
guaranteed-aligned arrays (analogous to <code>fftw_malloc</code> in C). You
can, of course, use the <code>FFTW_UNALIGNED</code> flag when creating the
plan, in which case the plan does not depend on the alignment, but
this may sacrifice substantial performance on architectures (like x86)
with SIMD instructions (see <a href="SIMD-alignment-and-fftw_005fmalloc.html">SIMD alignment and fftw_malloc</a>).
<span id="index-FFTW_005fUNALIGNED-3"></span>
</li></ul>
<hr>
<div class="header">
<p>
Next: <a href="Fortran-Examples.html" accesskey="n" rel="next">Fortran Examples</a>, Previous: <a href="FFTW-Constants-in-Fortran.html" accesskey="p" rel="prev">FFTW Constants in Fortran</a>, Up: <a href="Calling-FFTW-from-Legacy-Fortran.html" accesskey="u" rel="up">Calling FFTW from Legacy Fortran</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
</body>
</html>

View File

@@ -0,0 +1,173 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html>
<!-- This manual is for FFTW
(version 3.3.10, 10 December 2020).
Copyright (C) 2003 Matteo Frigo.
Copyright (C) 2003 Massachusetts Institute of Technology.
Permission is granted to make and distribute verbatim copies of this
manual provided the copyright notice and this permission notice are
preserved on all copies.
Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided that the
entire resulting derived work is distributed under the terms of a
permission notice identical to this one.
Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions,
except that this permission notice may be stated in a translation
approved by the Free Software Foundation. -->
<!-- Created by GNU Texinfo 6.7, http://www.gnu.org/software/texinfo/ -->
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<title>FFTW Fortran type reference (FFTW 3.3.10)</title>
<meta name="description" content="FFTW Fortran type reference (FFTW 3.3.10)">
<meta name="keywords" content="FFTW Fortran type reference (FFTW 3.3.10)">
<meta name="resource-type" content="document">
<meta name="distribution" content="global">
<meta name="Generator" content="makeinfo">
<link href="index.html" rel="start" title="Top">
<link href="Concept-Index.html" rel="index" title="Concept Index">
<link href="index.html#SEC_Contents" rel="contents" title="Table of Contents">
<link href="Calling-FFTW-from-Modern-Fortran.html" rel="up" title="Calling FFTW from Modern Fortran">
<link href="Plan-execution-in-Fortran.html" rel="next" title="Plan execution in Fortran">
<link href="Reversing-array-dimensions.html" rel="prev" title="Reversing array dimensions">
<style type="text/css">
<!--
a.summary-letter {text-decoration: none}
blockquote.indentedblock {margin-right: 0em}
div.display {margin-left: 3.2em}
div.example {margin-left: 3.2em}
div.lisp {margin-left: 3.2em}
kbd {font-style: oblique}
pre.display {font-family: inherit}
pre.format {font-family: inherit}
pre.menu-comment {font-family: serif}
pre.menu-preformatted {font-family: serif}
span.nolinebreak {white-space: nowrap}
span.roman {font-family: initial; font-weight: normal}
span.sansserif {font-family: sans-serif; font-weight: normal}
ul.no-bullet {list-style: none}
-->
</style>
</head>
<body lang="en">
<span id="FFTW-Fortran-type-reference"></span><div class="header">
<p>
Next: <a href="Plan-execution-in-Fortran.html" accesskey="n" rel="next">Plan execution in Fortran</a>, Previous: <a href="Reversing-array-dimensions.html" accesskey="p" rel="prev">Reversing array dimensions</a>, Up: <a href="Calling-FFTW-from-Modern-Fortran.html" accesskey="u" rel="up">Calling FFTW from Modern Fortran</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
<hr>
<span id="FFTW-Fortran-type-reference-1"></span><h3 class="section">7.3 FFTW Fortran type reference</h3>
<p>The following are the most important type correspondences between the
C interface and Fortran:
</p>
<ul>
<li> <span id="index-fftw_005fplan-2"></span>
Plans (<code>fftw_plan</code> and variants) are <code>type(C_PTR)</code> (i.e. an
opaque pointer).
</li><li> <span id="index-fftw_005fcomplex-3"></span>
<span id="index-precision-8"></span>
<span id="index-C_005fDOUBLE-1"></span>
<span id="index-C_005fFLOAT"></span>
<span id="index-C_005fLONG_005fDOUBLE"></span>
<span id="index-C_005fDOUBLE_005fCOMPLEX-1"></span>
<span id="index-C_005fFLOAT_005fCOMPLEX"></span>
<span id="index-C_005fLONG_005fDOUBLE_005fCOMPLEX"></span>
The C floating-point types <code>double</code>, <code>float</code>, and <code>long
double</code> correspond to <code>real(C_DOUBLE)</code>, <code>real(C_FLOAT)</code>, and
<code>real(C_LONG_DOUBLE)</code>, respectively. The C complex types
<code>fftw_complex</code>, <code>fftwf_complex</code>, and <code>fftwl_complex</code>
correspond in Fortran to <code>complex(C_DOUBLE_COMPLEX)</code>,
<code>complex(C_FLOAT_COMPLEX)</code>, and
<code>complex(C_LONG_DOUBLE_COMPLEX)</code>, respectively.
Just as in C
(see <a href="Precision.html">Precision</a>), the FFTW subroutines and types are prefixed with
&lsquo;<samp>fftw_</samp>&rsquo;, <code>fftwf_</code>, and <code>fftwl_</code> for the different precisions, and link to different libraries (<code>-lfftw3</code>, <code>-lfftw3f</code>, and <code>-lfftw3l</code> on Unix), but use the <em>same</em> include file <code>fftw3.f03</code> and the <em>same</em> constants (all of which begin with &lsquo;<samp>FFTW_</samp>&rsquo;). The exception is <code>long double</code> precision, for which you should <em>also</em> include <code>fftw3l.f03</code> (see <a href="Extended-and-quadruple-precision-in-Fortran.html">Extended and quadruple precision in Fortran</a>).
</li><li> <span id="index-ptrdiff_005ft-2"></span>
<span id="index-C_005fINT-1"></span>
<span id="index-C_005fINTPTR_005fT"></span>
<span id="index-C_005fSIZE_005fT"></span>
<span id="index-fftw_005fmalloc-7"></span>
The C integer types <code>int</code> and <code>unsigned</code> (used for planner
flags) become <code>integer(C_INT)</code>. The C integer type <code>ptrdiff_t</code> (e.g. in the <a href="64_002dbit-Guru-Interface.html">64-bit Guru Interface</a>) becomes <code>integer(C_INTPTR_T)</code>, and <code>size_t</code> (in <code>fftw_malloc</code> etc.) becomes <code>integer(C_SIZE_T)</code>.
</li><li> <span id="index-fftw_005fr2r_005fkind-2"></span>
<span id="index-C_005fFFTW_005fR2R_005fKIND"></span>
The <code>fftw_r2r_kind</code> type (see <a href="Real_002dto_002dReal-Transform-Kinds.html">Real-to-Real Transform Kinds</a>)
becomes <code>integer(C_FFTW_R2R_KIND)</code>. The various constant values
of the C enumerated type (<code>FFTW_R2HC</code> etc.) become simply integer
constants of the same names in Fortran.
</li><li> <span id="index-FFTW_005fDESTROY_005fINPUT-2"></span>
<span id="index-in_002dplace-10"></span>
<span id="index-fftw_005fflops-2"></span>
Numeric array pointer arguments (e.g. <code>double *</code>)
become <code>dimension(*), intent(out)</code> arrays of the same type, or
<code>dimension(*), intent(in)</code> if they are pointers to constant data
(e.g. <code>const int *</code>). There are a few exceptions where numeric
pointers refer to scalar outputs (e.g. for <code>fftw_flops</code>), in which
case they are <code>intent(out)</code> scalar arguments in Fortran too.
For the new-array execute functions (see <a href="New_002darray-Execute-Functions.html">New-array Execute Functions</a>),
the input arrays are declared <code>dimension(*), intent(inout)</code>, since
they can be modified in the case of in-place or <code>FFTW_DESTROY_INPUT</code>
transforms.
</li><li> <span id="index-fftw_005falloc_005freal-4"></span>
<span id="index-c_005ff_005fpointer-1"></span>
Pointer <em>return</em> values (e.g <code>double *</code>) become
<code>type(C_PTR)</code>. (If they are pointers to arrays, as for
<code>fftw_alloc_real</code>, you can convert them back to Fortran array
pointers with the standard intrinsic function <code>c_f_pointer</code>.)
</li><li> <span id="index-guru-interface-3"></span>
<span id="index-fftw_005fiodim-1"></span>
<span id="index-fftw_005fiodim64-1"></span>
<span id="index-64_002dbit-architecture-2"></span>
The <code>fftw_iodim</code> type in the guru interface (see <a href="Guru-vector-and-transform-sizes.html">Guru vector and transform sizes</a>) becomes <code>type(fftw_iodim)</code> in Fortran, a
derived data type (the Fortran analogue of C&rsquo;s <code>struct</code>) with
three <code>integer(C_INT)</code> components: <code>n</code>, <code>is</code>, and
<code>os</code>, with the same meanings as in C. The <code>fftw_iodim64</code> type in the 64-bit guru interface (see <a href="64_002dbit-Guru-Interface.html">64-bit Guru Interface</a>) is the same, except that its components are of type <code>integer(C_INTPTR_T)</code>.
</li><li> <span id="index-C_005fFUNPTR"></span>
Using the wisdom import/export functions from Fortran is a bit tricky,
and is discussed in <a href="Accessing-the-wisdom-API-from-Fortran.html">Accessing the wisdom API from Fortran</a>. In
brief, the <code>FILE *</code> arguments map to <code>type(C_PTR)</code>, <code>const char *</code> to <code>character(C_CHAR), dimension(*), intent(in)</code> (null-terminated!), and the generic read-char/write-char functions map to <code>type(C_FUNPTR)</code>.
</li></ul>
<span id="index-portability-5"></span>
<p>You may be wondering if you need to search-and-replace
<code>real(kind(0.0d0))</code> (or whatever your favorite Fortran spelling
of &ldquo;double precision&rdquo; is) with <code>real(C_DOUBLE)</code> everywhere in
your program, and similarly for <code>complex</code> and <code>integer</code>
types. The answer is no; you can still use your existing types. As
long as these types match their C counterparts, things should work
without a hitch. The worst that can happen, e.g. in the (unlikely)
event of a system where <code>real(kind(0.0d0))</code> is different from
<code>real(C_DOUBLE)</code>, is that the compiler will give you a
type-mismatch error. That is, if you don&rsquo;t use the
<code>iso_c_binding</code> kinds you need to accept at least the theoretical
possibility of having to change your code in response to compiler
errors on some future machine, but you don&rsquo;t need to worry about
silently compiling incorrect code that yields runtime errors.
</p>
<hr>
<div class="header">
<p>
Next: <a href="Plan-execution-in-Fortran.html" accesskey="n" rel="next">Plan execution in Fortran</a>, Previous: <a href="Reversing-array-dimensions.html" accesskey="p" rel="prev">Reversing array dimensions</a>, Up: <a href="Calling-FFTW-from-Modern-Fortran.html" accesskey="u" rel="up">Calling FFTW from Modern Fortran</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
</body>
</html>

View File

@@ -0,0 +1,226 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html>
<!-- This manual is for FFTW
(version 3.3.10, 10 December 2020).
Copyright (C) 2003 Matteo Frigo.
Copyright (C) 2003 Massachusetts Institute of Technology.
Permission is granted to make and distribute verbatim copies of this
manual provided the copyright notice and this permission notice are
preserved on all copies.
Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided that the
entire resulting derived work is distributed under the terms of a
permission notice identical to this one.
Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions,
except that this permission notice may be stated in a translation
approved by the Free Software Foundation. -->
<!-- Created by GNU Texinfo 6.7, http://www.gnu.org/software/texinfo/ -->
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<title>FFTW MPI Fortran Interface (FFTW 3.3.10)</title>
<meta name="description" content="FFTW MPI Fortran Interface (FFTW 3.3.10)">
<meta name="keywords" content="FFTW MPI Fortran Interface (FFTW 3.3.10)">
<meta name="resource-type" content="document">
<meta name="distribution" content="global">
<meta name="Generator" content="makeinfo">
<link href="index.html" rel="start" title="Top">
<link href="Concept-Index.html" rel="index" title="Concept Index">
<link href="index.html#SEC_Contents" rel="contents" title="Table of Contents">
<link href="Distributed_002dmemory-FFTW-with-MPI.html" rel="up" title="Distributed-memory FFTW with MPI">
<link href="Calling-FFTW-from-Modern-Fortran.html" rel="next" title="Calling FFTW from Modern Fortran">
<link href="MPI-Wisdom-Communication.html" rel="prev" title="MPI Wisdom Communication">
<style type="text/css">
<!--
a.summary-letter {text-decoration: none}
blockquote.indentedblock {margin-right: 0em}
div.display {margin-left: 3.2em}
div.example {margin-left: 3.2em}
div.lisp {margin-left: 3.2em}
kbd {font-style: oblique}
pre.display {font-family: inherit}
pre.format {font-family: inherit}
pre.menu-comment {font-family: serif}
pre.menu-preformatted {font-family: serif}
span.nolinebreak {white-space: nowrap}
span.roman {font-family: initial; font-weight: normal}
span.sansserif {font-family: sans-serif; font-weight: normal}
ul.no-bullet {list-style: none}
-->
</style>
</head>
<body lang="en">
<span id="FFTW-MPI-Fortran-Interface"></span><div class="header">
<p>
Previous: <a href="FFTW-MPI-Reference.html" accesskey="p" rel="prev">FFTW MPI Reference</a>, Up: <a href="Distributed_002dmemory-FFTW-with-MPI.html" accesskey="u" rel="up">Distributed-memory FFTW with MPI</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
<hr>
<span id="FFTW-MPI-Fortran-Interface-1"></span><h3 class="section">6.13 FFTW MPI Fortran Interface</h3>
<span id="index-Fortran-interface-1"></span>
<span id="index-iso_005fc_005fbinding"></span>
<p>The FFTW MPI interface is callable from modern Fortran compilers
supporting the Fortran 2003 <code>iso_c_binding</code> standard for calling
C functions. As described in <a href="Calling-FFTW-from-Modern-Fortran.html">Calling FFTW from Modern Fortran</a>,
this means that you can directly call FFTW&rsquo;s C interface from Fortran
with only minor changes in syntax. There are, however, a few things
specific to the MPI interface to keep in mind:
</p>
<ul>
<li> Instead of including <code>fftw3.f03</code> as in <a href="Overview-of-Fortran-interface.html">Overview of Fortran interface</a>, you should <code>include 'fftw3-mpi.f03'</code> (after
<code>use, intrinsic :: iso_c_binding</code> as before). The
<code>fftw3-mpi.f03</code> file includes <code>fftw3.f03</code>, so you should
<em>not</em> <code>include</code> them both yourself. (You will also want to
include the MPI header file, usually via <code>include 'mpif.h'</code> or
similar, although though this is not needed by <code>fftw3-mpi.f03</code>
<i>per se</i>.) (To use the &lsquo;<samp>fftwl_</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</a>.)
</li><li> Because of the different storage conventions between C and Fortran,
you reverse the order of your array dimensions when passing them to
FFTW (see <a href="Reversing-array-dimensions.html">Reversing array dimensions</a>). This is merely a
difference in notation and incurs no performance overhead. However,
it means that, whereas in C the <em>first</em> dimension is distributed,
in Fortran the <em>last</em> dimension of your array is distributed.
</li><li> <span id="index-MPI-communicator-3"></span>
In Fortran, communicators are stored as <code>integer</code> types; there is
no <code>MPI_Comm</code> type, nor is there any way to access a C
<code>MPI_Comm</code>. Fortunately, this is taken care of for you by the
FFTW Fortran interface: whenever the C interface expects an
<code>MPI_Comm</code> type, you should pass the Fortran communicator as an
<code>integer</code>.<a id="DOCF8" href="#FOOT8"><sup>8</sup></a>
</li><li> Because you need to call the &lsquo;<samp>local_size</samp>&rsquo; function to find out
how much space to allocate, and this may be <em>larger</em> than the
local portion of the array (see <a href="MPI-Data-Distribution.html">MPI Data Distribution</a>), you should
<em>always</em> allocate your arrays dynamically using FFTW&rsquo;s allocation
routines as described in <a href="Allocating-aligned-memory-in-Fortran.html">Allocating aligned memory in Fortran</a>.
(Coincidentally, this also provides the best performance by
guaranteeding proper data alignment.)
</li><li> Because all sizes in the MPI FFTW interface are declared as
<code>ptrdiff_t</code> in C, you should use <code>integer(C_INTPTR_T)</code> in
Fortran (see <a href="FFTW-Fortran-type-reference.html">FFTW Fortran type reference</a>).
</li><li> <span id="index-fftw_005fexecute_005fdft-1"></span>
<span id="index-fftw_005fmpi_005fexecute_005fdft-1"></span>
<span id="index-new_002darray-execution-3"></span>
In Fortran, because of the language semantics, we generally recommend
using the new-array execute functions for all plans, even in the
common case where you are executing the plan on the same arrays for
which the plan was created (see <a href="Plan-execution-in-Fortran.html">Plan execution in Fortran</a>).
However, note that in the MPI interface these functions are changed:
<code>fftw_execute_dft</code> becomes <code>fftw_mpi_execute_dft</code>,
etcetera. See <a href="Using-MPI-Plans.html">Using MPI Plans</a>.
</li></ul>
<p>For example, here is a Fortran code snippet to perform a distributed
L&nbsp;&times;&nbsp;M
complex DFT in-place. (This assumes you have already
initialized MPI with <code>MPI_init</code> and have also performed
<code>call fftw_mpi_init</code>.)
</p>
<div class="example">
<pre class="example"> use, intrinsic :: iso_c_binding
include 'fftw3-mpi.f03'
integer(C_INTPTR_T), parameter :: L = ...
integer(C_INTPTR_T), parameter :: M = ...
type(C_PTR) :: plan, cdata
complex(C_DOUBLE_COMPLEX), pointer :: data(:,:)
integer(C_INTPTR_T) :: i, j, alloc_local, local_M, local_j_offset
! <span class="roman">get local data size and allocate (note dimension reversal)</span>
alloc_local = fftw_mpi_local_size_2d(M, L, MPI_COMM_WORLD, &amp;
local_M, local_j_offset)
cdata = fftw_alloc_complex(alloc_local)
call c_f_pointer(cdata, data, [L,local_M])
! <span class="roman">create MPI plan for in-place forward DFT (note dimension reversal)</span>
plan = fftw_mpi_plan_dft_2d(M, L, data, data, MPI_COMM_WORLD, &amp;
FFTW_FORWARD, FFTW_MEASURE)
! <span class="roman">initialize data to some function</span> my_function(i,j)
do j = 1, local_M
do i = 1, L
data(i, j) = my_function(i, j + local_j_offset)
end do
end do
! <span class="roman">compute transform (as many times as desired)</span>
call fftw_mpi_execute_dft(plan, data, data)
call fftw_destroy_plan(plan)
call fftw_free(cdata)
</pre></div>
<p>Note that when we called <code>fftw_mpi_local_size_2d</code> and
<code>fftw_mpi_plan_dft_2d</code> with the dimensions in reversed order,
since a L&nbsp;&times;&nbsp;M
Fortran array is viewed by FFTW in C as a
M&nbsp;&times;&nbsp;L
array. This means that the array was distributed over
the <code>M</code> dimension, the local portion of which is a
L&nbsp;&times;&nbsp;local_M
array in Fortran. (You must <em>not</em> use an
<code>allocate</code> statement to allocate an L&nbsp;&times;&nbsp;local_M
array,
however; you must allocate <code>alloc_local</code> complex numbers, which
may be greater than <code>L * local_M</code>, in order to reserve space for
intermediate steps of the transform.) Finally, we mention that
because C&rsquo;s array indices are zero-based, the <code>local_j_offset</code>
argument can conveniently be interpreted as an offset in the 1-based
<code>j</code> index (rather than as a starting index as in C).
</p>
<p>If instead you had used the <code>ior(FFTW_MEASURE,
FFTW_MPI_TRANSPOSED_OUT)</code> flag, the output of the transform would be a
transposed M&nbsp;&times;&nbsp;local_L
array, associated with the <em>same</em>
<code>cdata</code> allocation (since the transform is in-place), and which
you could declare with:
</p>
<div class="example">
<pre class="example"> complex(C_DOUBLE_COMPLEX), pointer :: tdata(:,:)
...
call c_f_pointer(cdata, tdata, [M,local_L])
</pre></div>
<p>where <code>local_L</code> would have been obtained by changing the
<code>fftw_mpi_local_size_2d</code> call to:
</p>
<div class="example">
<pre class="example"> alloc_local = fftw_mpi_local_size_2d_transposed(M, L, MPI_COMM_WORLD, &amp;
local_M, local_j_offset, local_L, local_i_offset)
</pre></div>
<div class="footnote">
<hr>
<h4 class="footnotes-heading">Footnotes</h4>
<h5><a id="FOOT8" href="#DOCF8">(8)</a></h3>
<p>Technically, this is because you aren&rsquo;t
actually calling the C functions directly. You are calling wrapper
functions that translate the communicator with <code>MPI_Comm_f2c</code>
before calling the ordinary C interface. This is all done
transparently, however, since the <code>fftw3-mpi.f03</code> interface file
renames the wrappers so that they are called in Fortran with the same
names as the C interface functions.</p>
</div>
<hr>
<div class="header">
<p>
Previous: <a href="FFTW-MPI-Reference.html" accesskey="p" rel="prev">FFTW MPI Reference</a>, Up: <a href="Distributed_002dmemory-FFTW-with-MPI.html" accesskey="u" rel="up">Distributed-memory FFTW with MPI</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
</body>
</html>

View File

@@ -0,0 +1,106 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html>
<!-- This manual is for FFTW
(version 3.3.10, 10 December 2020).
Copyright (C) 2003 Matteo Frigo.
Copyright (C) 2003 Massachusetts Institute of Technology.
Permission is granted to make and distribute verbatim copies of this
manual provided the copyright notice and this permission notice are
preserved on all copies.
Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided that the
entire resulting derived work is distributed under the terms of a
permission notice identical to this one.
Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions,
except that this permission notice may be stated in a translation
approved by the Free Software Foundation. -->
<!-- Created by GNU Texinfo 6.7, http://www.gnu.org/software/texinfo/ -->
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<title>FFTW MPI Installation (FFTW 3.3.10)</title>
<meta name="description" content="FFTW MPI Installation (FFTW 3.3.10)">
<meta name="keywords" content="FFTW MPI Installation (FFTW 3.3.10)">
<meta name="resource-type" content="document">
<meta name="distribution" content="global">
<meta name="Generator" content="makeinfo">
<link href="index.html" rel="start" title="Top">
<link href="Concept-Index.html" rel="index" title="Concept Index">
<link href="index.html#SEC_Contents" rel="contents" title="Table of Contents">
<link href="Distributed_002dmemory-FFTW-with-MPI.html" rel="up" title="Distributed-memory FFTW with MPI">
<link href="Linking-and-Initializing-MPI-FFTW.html" rel="next" title="Linking and Initializing MPI FFTW">
<link href="Distributed_002dmemory-FFTW-with-MPI.html" rel="prev" title="Distributed-memory FFTW with MPI">
<style type="text/css">
<!--
a.summary-letter {text-decoration: none}
blockquote.indentedblock {margin-right: 0em}
div.display {margin-left: 3.2em}
div.example {margin-left: 3.2em}
div.lisp {margin-left: 3.2em}
kbd {font-style: oblique}
pre.display {font-family: inherit}
pre.format {font-family: inherit}
pre.menu-comment {font-family: serif}
pre.menu-preformatted {font-family: serif}
span.nolinebreak {white-space: nowrap}
span.roman {font-family: initial; font-weight: normal}
span.sansserif {font-family: sans-serif; font-weight: normal}
ul.no-bullet {list-style: none}
-->
</style>
</head>
<body lang="en">
<span id="FFTW-MPI-Installation"></span><div class="header">
<p>
Next: <a href="Linking-and-Initializing-MPI-FFTW.html" accesskey="n" rel="next">Linking and Initializing MPI FFTW</a>, Previous: <a href="Distributed_002dmemory-FFTW-with-MPI.html" accesskey="p" rel="prev">Distributed-memory FFTW with MPI</a>, Up: <a href="Distributed_002dmemory-FFTW-with-MPI.html" accesskey="u" rel="up">Distributed-memory FFTW with MPI</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
<hr>
<span id="FFTW-MPI-Installation-1"></span><h3 class="section">6.1 FFTW MPI Installation</h3>
<p>All of the FFTW MPI code is located in the <code>mpi</code> subdirectory of
the FFTW package. On Unix systems, the FFTW MPI libraries and header
files are automatically configured, compiled, and installed along with
the uniprocessor FFTW libraries simply by including
<code>--enable-mpi</code> in the flags to the <code>configure</code> script
(see <a href="Installation-on-Unix.html">Installation on Unix</a>).
<span id="index-configure-1"></span>
</p>
<p>Any implementation of the MPI standard, version 1 or later, should
work with FFTW. The <code>configure</code> script will attempt to
automatically detect how to compile and link code using your MPI
implementation. In some cases, especially if you have multiple
different MPI implementations installed or have an unusual MPI
software package, you may need to provide this information explicitly.
</p>
<p>Most commonly, one compiles MPI code by invoking a special compiler
command, typically <code>mpicc</code> for C code. The <code>configure</code>
script knows the most common names for this command, but you can
specify the MPI compilation command explicitly by setting the
<code>MPICC</code> variable, as in &lsquo;<samp>./configure MPICC=mpicc ...</samp>&rsquo;.
<span id="index-mpicc"></span>
</p>
<p>If, instead of a special compiler command, you need to link a certain
library, you can specify the link command via the <code>MPILIBS</code>
variable, as in &lsquo;<samp>./configure MPILIBS=-lmpi ...</samp>&rsquo;. Note that if
your MPI library is installed in a non-standard location (one the
compiler does not know about by default), you may also have to specify
the location of the library and header files via <code>LDFLAGS</code> and
<code>CPPFLAGS</code> variables, respectively, as in &lsquo;<samp>./configure
LDFLAGS=-L/path/to/mpi/libs CPPFLAGS=-I/path/to/mpi/include ...</samp>&rsquo;.
</p>
</body>
</html>

View File

@@ -0,0 +1,104 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html>
<!-- This manual is for FFTW
(version 3.3.10, 10 December 2020).
Copyright (C) 2003 Matteo Frigo.
Copyright (C) 2003 Massachusetts Institute of Technology.
Permission is granted to make and distribute verbatim copies of this
manual provided the copyright notice and this permission notice are
preserved on all copies.
Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided that the
entire resulting derived work is distributed under the terms of a
permission notice identical to this one.
Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions,
except that this permission notice may be stated in a translation
approved by the Free Software Foundation. -->
<!-- Created by GNU Texinfo 6.7, http://www.gnu.org/software/texinfo/ -->
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<title>FFTW MPI Performance Tips (FFTW 3.3.10)</title>
<meta name="description" content="FFTW MPI Performance Tips (FFTW 3.3.10)">
<meta name="keywords" content="FFTW MPI Performance Tips (FFTW 3.3.10)">
<meta name="resource-type" content="document">
<meta name="distribution" content="global">
<meta name="Generator" content="makeinfo">
<link href="index.html" rel="start" title="Top">
<link href="Concept-Index.html" rel="index" title="Concept Index">
<link href="index.html#SEC_Contents" rel="contents" title="Table of Contents">
<link href="Distributed_002dmemory-FFTW-with-MPI.html" rel="up" title="Distributed-memory FFTW with MPI">
<link href="Combining-MPI-and-Threads.html" rel="next" title="Combining MPI and Threads">
<link href="Avoiding-MPI-Deadlocks.html" rel="prev" title="Avoiding MPI Deadlocks">
<style type="text/css">
<!--
a.summary-letter {text-decoration: none}
blockquote.indentedblock {margin-right: 0em}
div.display {margin-left: 3.2em}
div.example {margin-left: 3.2em}
div.lisp {margin-left: 3.2em}
kbd {font-style: oblique}
pre.display {font-family: inherit}
pre.format {font-family: inherit}
pre.menu-comment {font-family: serif}
pre.menu-preformatted {font-family: serif}
span.nolinebreak {white-space: nowrap}
span.roman {font-family: initial; font-weight: normal}
span.sansserif {font-family: sans-serif; font-weight: normal}
ul.no-bullet {list-style: none}
-->
</style>
</head>
<body lang="en">
<span id="FFTW-MPI-Performance-Tips"></span><div class="header">
<p>
Next: <a href="Combining-MPI-and-Threads.html" accesskey="n" rel="next">Combining MPI and Threads</a>, Previous: <a href="Avoiding-MPI-Deadlocks.html" accesskey="p" rel="prev">Avoiding MPI Deadlocks</a>, Up: <a href="Distributed_002dmemory-FFTW-with-MPI.html" accesskey="u" rel="up">Distributed-memory FFTW with MPI</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
<hr>
<span id="FFTW-MPI-Performance-Tips-1"></span><h3 class="section">6.10 FFTW MPI Performance Tips</h3>
<p>In this section, we collect a few tips on getting the best performance
out of FFTW&rsquo;s MPI transforms.
</p>
<p>First, because of the 1d block distribution, FFTW&rsquo;s parallelization is
currently limited by the size of the first dimension.
(Multidimensional block distributions may be supported by a future
version.) More generally, you should ideally arrange the dimensions so
that FFTW can divide them equally among the processes. See <a href="Load-balancing.html">Load balancing</a>.
<span id="index-block-distribution-2"></span>
<span id="index-load-balancing-1"></span>
</p>
<p>Second, if it is not too inconvenient, you should consider working
with transposed output for multidimensional plans, as this saves a
considerable amount of communications. See <a href="Transposed-distributions.html">Transposed distributions</a>.
<span id="index-transpose-3"></span>
</p>
<p>Third, the fastest choices are generally either an in-place transform
or an out-of-place transform with the <code>FFTW_DESTROY_INPUT</code> flag
(which allows the input array to be used as scratch space). In-place
is especially beneficial if the amount of data per process is large.
<span id="index-FFTW_005fDESTROY_005fINPUT-1"></span>
</p>
<p>Fourth, if you have multiple arrays to transform at once, rather than
calling FFTW&rsquo;s MPI transforms several times it usually seems to be
faster to interleave the data and use the advanced interface. (This
groups the communications together instead of requiring separate
messages for each transform.)
</p>
</body>
</html>

View File

@@ -0,0 +1,92 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html>
<!-- This manual is for FFTW
(version 3.3.10, 10 December 2020).
Copyright (C) 2003 Matteo Frigo.
Copyright (C) 2003 Massachusetts Institute of Technology.
Permission is granted to make and distribute verbatim copies of this
manual provided the copyright notice and this permission notice are
preserved on all copies.
Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided that the
entire resulting derived work is distributed under the terms of a
permission notice identical to this one.
Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions,
except that this permission notice may be stated in a translation
approved by the Free Software Foundation. -->
<!-- Created by GNU Texinfo 6.7, http://www.gnu.org/software/texinfo/ -->
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<title>FFTW MPI Reference (FFTW 3.3.10)</title>
<meta name="description" content="FFTW MPI Reference (FFTW 3.3.10)">
<meta name="keywords" content="FFTW MPI Reference (FFTW 3.3.10)">
<meta name="resource-type" content="document">
<meta name="distribution" content="global">
<meta name="Generator" content="makeinfo">
<link href="index.html" rel="start" title="Top">
<link href="Concept-Index.html" rel="index" title="Concept Index">
<link href="index.html#SEC_Contents" rel="contents" title="Table of Contents">
<link href="Distributed_002dmemory-FFTW-with-MPI.html" rel="up" title="Distributed-memory FFTW with MPI">
<link href="MPI-Files-and-Data-Types.html" rel="next" title="MPI Files and Data Types">
<link href="Combining-MPI-and-Threads.html" rel="prev" title="Combining MPI and Threads">
<style type="text/css">
<!--
a.summary-letter {text-decoration: none}
blockquote.indentedblock {margin-right: 0em}
div.display {margin-left: 3.2em}
div.example {margin-left: 3.2em}
div.lisp {margin-left: 3.2em}
kbd {font-style: oblique}
pre.display {font-family: inherit}
pre.format {font-family: inherit}
pre.menu-comment {font-family: serif}
pre.menu-preformatted {font-family: serif}
span.nolinebreak {white-space: nowrap}
span.roman {font-family: initial; font-weight: normal}
span.sansserif {font-family: sans-serif; font-weight: normal}
ul.no-bullet {list-style: none}
-->
</style>
</head>
<body lang="en">
<span id="FFTW-MPI-Reference"></span><div class="header">
<p>
Next: <a href="FFTW-MPI-Fortran-Interface.html" accesskey="n" rel="next">FFTW MPI Fortran Interface</a>, Previous: <a href="Combining-MPI-and-Threads.html" accesskey="p" rel="prev">Combining MPI and Threads</a>, Up: <a href="Distributed_002dmemory-FFTW-with-MPI.html" accesskey="u" rel="up">Distributed-memory FFTW with MPI</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
<hr>
<span id="FFTW-MPI-Reference-1"></span><h3 class="section">6.12 FFTW MPI Reference</h3>
<p>This chapter provides a complete reference to all FFTW MPI functions,
datatypes, and constants. See also <a href="FFTW-Reference.html">FFTW Reference</a> for information
on functions and types in common with the serial interface.
</p>
<table class="menu" border="0" cellspacing="0">
<tr><td align="left" valign="top">&bull; <a href="MPI-Files-and-Data-Types.html" accesskey="1">MPI Files and Data Types</a></td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="MPI-Initialization.html" accesskey="2">MPI Initialization</a></td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="Using-MPI-Plans.html" accesskey="3">Using MPI Plans</a></td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="MPI-Data-Distribution-Functions.html" accesskey="4">MPI Data Distribution Functions</a></td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="MPI-Plan-Creation.html" accesskey="5">MPI Plan Creation</a></td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="MPI-Wisdom-Communication.html" accesskey="6">MPI Wisdom Communication</a></td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
</table>
</body>
</html>

View File

@@ -0,0 +1,92 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html>
<!-- This manual is for FFTW
(version 3.3.10, 10 December 2020).
Copyright (C) 2003 Matteo Frigo.
Copyright (C) 2003 Massachusetts Institute of Technology.
Permission is granted to make and distribute verbatim copies of this
manual provided the copyright notice and this permission notice are
preserved on all copies.
Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided that the
entire resulting derived work is distributed under the terms of a
permission notice identical to this one.
Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions,
except that this permission notice may be stated in a translation
approved by the Free Software Foundation. -->
<!-- Created by GNU Texinfo 6.7, http://www.gnu.org/software/texinfo/ -->
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<title>FFTW MPI Transposes (FFTW 3.3.10)</title>
<meta name="description" content="FFTW MPI Transposes (FFTW 3.3.10)">
<meta name="keywords" content="FFTW MPI Transposes (FFTW 3.3.10)">
<meta name="resource-type" content="document">
<meta name="distribution" content="global">
<meta name="Generator" content="makeinfo">
<link href="index.html" rel="start" title="Top">
<link href="Concept-Index.html" rel="index" title="Concept Index">
<link href="index.html#SEC_Contents" rel="contents" title="Table of Contents">
<link href="Distributed_002dmemory-FFTW-with-MPI.html" rel="up" title="Distributed-memory FFTW with MPI">
<link href="Basic-distributed_002dtranspose-interface.html" rel="next" title="Basic distributed-transpose interface">
<link href="Other-Multi_002ddimensional-Real_002ddata-MPI-Transforms.html" rel="prev" title="Other Multi-dimensional Real-data MPI Transforms">
<style type="text/css">
<!--
a.summary-letter {text-decoration: none}
blockquote.indentedblock {margin-right: 0em}
div.display {margin-left: 3.2em}
div.example {margin-left: 3.2em}
div.lisp {margin-left: 3.2em}
kbd {font-style: oblique}
pre.display {font-family: inherit}
pre.format {font-family: inherit}
pre.menu-comment {font-family: serif}
pre.menu-preformatted {font-family: serif}
span.nolinebreak {white-space: nowrap}
span.roman {font-family: initial; font-weight: normal}
span.sansserif {font-family: sans-serif; font-weight: normal}
ul.no-bullet {list-style: none}
-->
</style>
</head>
<body lang="en">
<span id="FFTW-MPI-Transposes"></span><div class="header">
<p>
Next: <a href="FFTW-MPI-Wisdom.html" accesskey="n" rel="next">FFTW MPI Wisdom</a>, Previous: <a href="Other-Multi_002ddimensional-Real_002ddata-MPI-Transforms.html" accesskey="p" rel="prev">Other Multi-dimensional Real-data MPI Transforms</a>, Up: <a href="Distributed_002dmemory-FFTW-with-MPI.html" accesskey="u" rel="up">Distributed-memory FFTW with MPI</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
<hr>
<span id="FFTW-MPI-Transposes-1"></span><h3 class="section">6.7 FFTW MPI Transposes</h3>
<span id="index-transpose-2"></span>
<p>The FFTW&rsquo;s MPI Fourier transforms rely on one or more <em>global
transposition</em> step for their communications. For example, the
multidimensional transforms work by transforming along some
dimensions, then transposing to make the first dimension local and
transforming that, then transposing back. Because global
transposition of a block-distributed matrix has many other potential
uses besides FFTs, FFTW&rsquo;s transpose routines can be called directly,
as documented in this section.
</p>
<table class="menu" border="0" cellspacing="0">
<tr><td align="left" valign="top">&bull; <a href="Basic-distributed_002dtranspose-interface.html" accesskey="1">Basic distributed-transpose interface</a></td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="Advanced-distributed_002dtranspose-interface.html" accesskey="2">Advanced distributed-transpose interface</a></td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="An-improved-replacement-for-MPI_005fAlltoall.html" accesskey="3">An improved replacement for MPI_Alltoall</a></td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
</table>
</body>
</html>

View File

@@ -0,0 +1,171 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html>
<!-- This manual is for FFTW
(version 3.3.10, 10 December 2020).
Copyright (C) 2003 Matteo Frigo.
Copyright (C) 2003 Massachusetts Institute of Technology.
Permission is granted to make and distribute verbatim copies of this
manual provided the copyright notice and this permission notice are
preserved on all copies.
Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided that the
entire resulting derived work is distributed under the terms of a
permission notice identical to this one.
Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions,
except that this permission notice may be stated in a translation
approved by the Free Software Foundation. -->
<!-- Created by GNU Texinfo 6.7, http://www.gnu.org/software/texinfo/ -->
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<title>FFTW MPI Wisdom (FFTW 3.3.10)</title>
<meta name="description" content="FFTW MPI Wisdom (FFTW 3.3.10)">
<meta name="keywords" content="FFTW MPI Wisdom (FFTW 3.3.10)">
<meta name="resource-type" content="document">
<meta name="distribution" content="global">
<meta name="Generator" content="makeinfo">
<link href="index.html" rel="start" title="Top">
<link href="Concept-Index.html" rel="index" title="Concept Index">
<link href="index.html#SEC_Contents" rel="contents" title="Table of Contents">
<link href="Distributed_002dmemory-FFTW-with-MPI.html" rel="up" title="Distributed-memory FFTW with MPI">
<link href="Avoiding-MPI-Deadlocks.html" rel="next" title="Avoiding MPI Deadlocks">
<link href="An-improved-replacement-for-MPI_005fAlltoall.html" rel="prev" title="An improved replacement for MPI_Alltoall">
<style type="text/css">
<!--
a.summary-letter {text-decoration: none}
blockquote.indentedblock {margin-right: 0em}
div.display {margin-left: 3.2em}
div.example {margin-left: 3.2em}
div.lisp {margin-left: 3.2em}
kbd {font-style: oblique}
pre.display {font-family: inherit}
pre.format {font-family: inherit}
pre.menu-comment {font-family: serif}
pre.menu-preformatted {font-family: serif}
span.nolinebreak {white-space: nowrap}
span.roman {font-family: initial; font-weight: normal}
span.sansserif {font-family: sans-serif; font-weight: normal}
ul.no-bullet {list-style: none}
-->
</style>
</head>
<body lang="en">
<span id="FFTW-MPI-Wisdom"></span><div class="header">
<p>
Next: <a href="Avoiding-MPI-Deadlocks.html" accesskey="n" rel="next">Avoiding MPI Deadlocks</a>, Previous: <a href="FFTW-MPI-Transposes.html" accesskey="p" rel="prev">FFTW MPI Transposes</a>, Up: <a href="Distributed_002dmemory-FFTW-with-MPI.html" accesskey="u" rel="up">Distributed-memory FFTW with MPI</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
<hr>
<span id="FFTW-MPI-Wisdom-1"></span><h3 class="section">6.8 FFTW MPI Wisdom</h3>
<span id="index-wisdom-2"></span>
<span id="index-saving-plans-to-disk-2"></span>
<p>FFTW&rsquo;s &ldquo;wisdom&rdquo; facility (see <a href="Words-of-Wisdom_002dSaving-Plans.html">Words of Wisdom-Saving Plans</a>) can
be used to save MPI plans as well as to save uniprocessor plans.
However, for MPI there are several unavoidable complications.
</p>
<span id="index-MPI-I_002fO-1"></span>
<p>First, the MPI standard does not guarantee that every process can
perform file I/O (at least, not using C stdio routines)&mdash;in general,
we may only assume that process 0 is capable of I/O.<a id="DOCF7" href="#FOOT7"><sup>7</sup></a> So, if we
want to export the wisdom from a single process to a file, we must
first export the wisdom to a string, then send it to process 0, then
write it to a file.
</p>
<p>Second, in principle we may want to have separate wisdom for every
process, since in general the processes may run on different hardware
even for a single MPI program. However, in practice FFTW&rsquo;s MPI code
is designed for the case of homogeneous hardware (see <a href="Load-balancing.html">Load balancing</a>), and in this case it is convenient to use the same wisdom
for every process. Thus, we need a mechanism to synchronize the wisdom.
</p>
<p>To address both of these problems, FFTW provides the following two
functions:
</p>
<div class="example">
<pre class="example">void fftw_mpi_broadcast_wisdom(MPI_Comm comm);
void fftw_mpi_gather_wisdom(MPI_Comm comm);
</pre></div>
<span id="index-fftw_005fmpi_005fgather_005fwisdom"></span>
<span id="index-fftw_005fmpi_005fbroadcast_005fwisdom"></span>
<p>Given a communicator <code>comm</code>, <code>fftw_mpi_broadcast_wisdom</code>
will broadcast the wisdom from process 0 to all other processes.
Conversely, <code>fftw_mpi_gather_wisdom</code> will collect wisdom from all
processes onto process 0. (If the plans created for the same problem
by different processes are not the same, <code>fftw_mpi_gather_wisdom</code>
will arbitrarily choose one of the plans.) Both of these functions
may result in suboptimal plans for different processes if the
processes are running on non-identical hardware. Both of these
functions are <em>collective</em> calls, which means that they must be
executed by all processes in the communicator.
<span id="index-collective-function-1"></span>
</p>
<p>So, for example, a typical code snippet to import wisdom from a file
and use it on all processes would be:
</p>
<div class="example">
<pre class="example">{
int rank;
fftw_mpi_init();
MPI_Comm_rank(MPI_COMM_WORLD, &amp;rank);
if (rank == 0) fftw_import_wisdom_from_filename(&quot;mywisdom&quot;);
fftw_mpi_broadcast_wisdom(MPI_COMM_WORLD);
}
</pre></div>
<p>(Note that we must call <code>fftw_mpi_init</code> before importing any
wisdom that might contain MPI plans.) Similarly, a typical code
snippet to export wisdom from all processes to a file is:
<span id="index-fftw_005fmpi_005finit-2"></span>
</p>
<div class="example">
<pre class="example">{
int rank;
fftw_mpi_gather_wisdom(MPI_COMM_WORLD);
MPI_Comm_rank(MPI_COMM_WORLD, &amp;rank);
if (rank == 0) fftw_export_wisdom_to_filename(&quot;mywisdom&quot;);
}
</pre></div>
<div class="footnote">
<hr>
<h4 class="footnotes-heading">Footnotes</h4>
<h5><a id="FOOT7" href="#DOCF7">(7)</a></h3>
<p>In fact,
even this assumption is not technically guaranteed by the standard,
although it seems to be universal in actual MPI implementations and is
widely assumed by MPI-using software. Technically, you need to query
the <code>MPI_IO</code> attribute of <code>MPI_COMM_WORLD</code> with
<code>MPI_Attr_get</code>. If this attribute is <code>MPI_PROC_NULL</code>, no
I/O is possible. If it is <code>MPI_ANY_SOURCE</code>, any process can
perform I/O. Otherwise, it is the rank of a process that can perform
I/O ... but since it is not guaranteed to yield the <em>same</em> rank
on all processes, you have to do an <code>MPI_Allreduce</code> of some kind
if you want all processes to agree about which is going to do I/O.
And even then, the standard only guarantees that this process can
perform output, but not input. See e.g. <cite>Parallel Programming
with MPI</cite> by P. S. Pacheco, section 8.1.3. Needless to say, in our
experience virtually no MPI programmers worry about this.</p>
</div>
<hr>
<div class="header">
<p>
Next: <a href="Avoiding-MPI-Deadlocks.html" accesskey="n" rel="next">Avoiding MPI Deadlocks</a>, Previous: <a href="FFTW-MPI-Transposes.html" accesskey="p" rel="prev">FFTW MPI Transposes</a>, Up: <a href="Distributed_002dmemory-FFTW-with-MPI.html" accesskey="u" rel="up">Distributed-memory FFTW with MPI</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
</body>
</html>

View File

@@ -0,0 +1,96 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html>
<!-- This manual is for FFTW
(version 3.3.10, 10 December 2020).
Copyright (C) 2003 Matteo Frigo.
Copyright (C) 2003 Massachusetts Institute of Technology.
Permission is granted to make and distribute verbatim copies of this
manual provided the copyright notice and this permission notice are
preserved on all copies.
Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided that the
entire resulting derived work is distributed under the terms of a
permission notice identical to this one.
Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions,
except that this permission notice may be stated in a translation
approved by the Free Software Foundation. -->
<!-- Created by GNU Texinfo 6.7, http://www.gnu.org/software/texinfo/ -->
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<title>FFTW Reference (FFTW 3.3.10)</title>
<meta name="description" content="FFTW Reference (FFTW 3.3.10)">
<meta name="keywords" content="FFTW Reference (FFTW 3.3.10)">
<meta name="resource-type" content="document">
<meta name="distribution" content="global">
<meta name="Generator" content="makeinfo">
<link href="index.html" rel="start" title="Top">
<link href="Concept-Index.html" rel="index" title="Concept Index">
<link href="index.html#SEC_Contents" rel="contents" title="Table of Contents">
<link href="index.html" rel="up" title="Top">
<link href="Data-Types-and-Files.html" rel="next" title="Data Types and Files">
<link href="Caveats-in-Using-Wisdom.html" rel="prev" title="Caveats in Using Wisdom">
<style type="text/css">
<!--
a.summary-letter {text-decoration: none}
blockquote.indentedblock {margin-right: 0em}
div.display {margin-left: 3.2em}
div.example {margin-left: 3.2em}
div.lisp {margin-left: 3.2em}
kbd {font-style: oblique}
pre.display {font-family: inherit}
pre.format {font-family: inherit}
pre.menu-comment {font-family: serif}
pre.menu-preformatted {font-family: serif}
span.nolinebreak {white-space: nowrap}
span.roman {font-family: initial; font-weight: normal}
span.sansserif {font-family: sans-serif; font-weight: normal}
ul.no-bullet {list-style: none}
-->
</style>
</head>
<body lang="en">
<span id="FFTW-Reference"></span><div class="header">
<p>
Next: <a href="Multi_002dthreaded-FFTW.html" accesskey="n" rel="next">Multi-threaded FFTW</a>, Previous: <a href="Other-Important-Topics.html" accesskey="p" rel="prev">Other Important Topics</a>, Up: <a href="index.html" accesskey="u" rel="up">Top</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
<hr>
<span id="FFTW-Reference-1"></span><h2 class="chapter">4 FFTW Reference</h2>
<p>This chapter provides a complete reference for all sequential (i.e.,
one-processor) FFTW functions. Parallel transforms are described in
later chapters.
</p>
<table class="menu" border="0" cellspacing="0">
<tr><td align="left" valign="top">&bull; <a href="Data-Types-and-Files.html" accesskey="1">Data Types and Files</a></td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="Using-Plans.html" accesskey="2">Using Plans</a></td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="Basic-Interface.html" accesskey="3">Basic Interface</a></td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="Advanced-Interface.html" accesskey="4">Advanced Interface</a></td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="Guru-Interface.html" accesskey="5">Guru Interface</a></td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="New_002darray-Execute-Functions.html" accesskey="6">New-array Execute Functions</a></td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="Wisdom.html" accesskey="7">Wisdom</a></td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="What-FFTW-Really-Computes.html" accesskey="8">What FFTW Really Computes</a></td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
</table>
</body>
</html>

View File

@@ -0,0 +1,104 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html>
<!-- This manual is for FFTW
(version 3.3.10, 10 December 2020).
Copyright (C) 2003 Matteo Frigo.
Copyright (C) 2003 Massachusetts Institute of Technology.
Permission is granted to make and distribute verbatim copies of this
manual provided the copyright notice and this permission notice are
preserved on all copies.
Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided that the
entire resulting derived work is distributed under the terms of a
permission notice identical to this one.
Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions,
except that this permission notice may be stated in a translation
approved by the Free Software Foundation. -->
<!-- Created by GNU Texinfo 6.7, http://www.gnu.org/software/texinfo/ -->
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<title>Fixed-size Arrays in C (FFTW 3.3.10)</title>
<meta name="description" content="Fixed-size Arrays in C (FFTW 3.3.10)">
<meta name="keywords" content="Fixed-size Arrays in C (FFTW 3.3.10)">
<meta name="resource-type" content="document">
<meta name="distribution" content="global">
<meta name="Generator" content="makeinfo">
<link href="index.html" rel="start" title="Top">
<link href="Concept-Index.html" rel="index" title="Concept Index">
<link href="index.html#SEC_Contents" rel="contents" title="Table of Contents">
<link href="Multi_002ddimensional-Array-Format.html" rel="up" title="Multi-dimensional Array Format">
<link href="Dynamic-Arrays-in-C.html" rel="next" title="Dynamic Arrays in C">
<link href="Column_002dmajor-Format.html" rel="prev" title="Column-major Format">
<style type="text/css">
<!--
a.summary-letter {text-decoration: none}
blockquote.indentedblock {margin-right: 0em}
div.display {margin-left: 3.2em}
div.example {margin-left: 3.2em}
div.lisp {margin-left: 3.2em}
kbd {font-style: oblique}
pre.display {font-family: inherit}
pre.format {font-family: inherit}
pre.menu-comment {font-family: serif}
pre.menu-preformatted {font-family: serif}
span.nolinebreak {white-space: nowrap}
span.roman {font-family: initial; font-weight: normal}
span.sansserif {font-family: sans-serif; font-weight: normal}
ul.no-bullet {list-style: none}
-->
</style>
</head>
<body lang="en">
<span id="Fixed_002dsize-Arrays-in-C"></span><div class="header">
<p>
Next: <a href="Dynamic-Arrays-in-C.html" accesskey="n" rel="next">Dynamic Arrays in C</a>, Previous: <a href="Column_002dmajor-Format.html" accesskey="p" rel="prev">Column-major Format</a>, Up: <a href="Multi_002ddimensional-Array-Format.html" accesskey="u" rel="up">Multi-dimensional Array Format</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
<hr>
<span id="Fixed_002dsize-Arrays-in-C-1"></span><h4 class="subsection">3.2.3 Fixed-size Arrays in C</h4>
<span id="index-C-multi_002ddimensional-arrays"></span>
<p>A multi-dimensional array whose size is declared at compile time in C
is <em>already</em> in row-major order. You don&rsquo;t have to do anything
special to transform it. For example:
</p>
<div class="example">
<pre class="example">{
fftw_complex data[N0][N1][N2];
fftw_plan plan;
...
plan = fftw_plan_dft_3d(N0, N1, N2, &amp;data[0][0][0], &amp;data[0][0][0],
FFTW_FORWARD, FFTW_ESTIMATE);
...
}
</pre></div>
<p>This will plan a 3d in-place transform of size <code>N0 x N1 x N2</code>.
Notice how we took the address of the zero-th element to pass to the
planner (we could also have used a typecast).
</p>
<p>However, we tend to <em>discourage</em> users from declaring their
arrays in this way, for two reasons. First, this allocates the array
on the stack (&ldquo;automatic&rdquo; storage), which has a very limited size on
most operating systems (declaring an array with more than a few
thousand elements will often cause a crash). (You can get around this
limitation on many systems by declaring the array as
<code>static</code> and/or global, but that has its own drawbacks.)
Second, it may not optimally align the array for use with a SIMD
FFTW (see <a href="SIMD-alignment-and-fftw_005fmalloc.html">SIMD alignment and fftw_malloc</a>). Instead, we recommend
using <code>fftw_malloc</code>, as described below.
</p>
</body>
</html>

View File

@@ -0,0 +1,82 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html>
<!-- This manual is for FFTW
(version 3.3.10, 10 December 2020).
Copyright (C) 2003 Matteo Frigo.
Copyright (C) 2003 Massachusetts Institute of Technology.
Permission is granted to make and distribute verbatim copies of this
manual provided the copyright notice and this permission notice are
preserved on all copies.
Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided that the
entire resulting derived work is distributed under the terms of a
permission notice identical to this one.
Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions,
except that this permission notice may be stated in a translation
approved by the Free Software Foundation. -->
<!-- Created by GNU Texinfo 6.7, http://www.gnu.org/software/texinfo/ -->
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<title>Forgetting Wisdom (FFTW 3.3.10)</title>
<meta name="description" content="Forgetting Wisdom (FFTW 3.3.10)">
<meta name="keywords" content="Forgetting Wisdom (FFTW 3.3.10)">
<meta name="resource-type" content="document">
<meta name="distribution" content="global">
<meta name="Generator" content="makeinfo">
<link href="index.html" rel="start" title="Top">
<link href="Concept-Index.html" rel="index" title="Concept Index">
<link href="index.html#SEC_Contents" rel="contents" title="Table of Contents">
<link href="Wisdom.html" rel="up" title="Wisdom">
<link href="Wisdom-Utilities.html" rel="next" title="Wisdom Utilities">
<link href="Wisdom-Import.html" rel="prev" title="Wisdom Import">
<style type="text/css">
<!--
a.summary-letter {text-decoration: none}
blockquote.indentedblock {margin-right: 0em}
div.display {margin-left: 3.2em}
div.example {margin-left: 3.2em}
div.lisp {margin-left: 3.2em}
kbd {font-style: oblique}
pre.display {font-family: inherit}
pre.format {font-family: inherit}
pre.menu-comment {font-family: serif}
pre.menu-preformatted {font-family: serif}
span.nolinebreak {white-space: nowrap}
span.roman {font-family: initial; font-weight: normal}
span.sansserif {font-family: sans-serif; font-weight: normal}
ul.no-bullet {list-style: none}
-->
</style>
</head>
<body lang="en">
<span id="Forgetting-Wisdom"></span><div class="header">
<p>
Next: <a href="Wisdom-Utilities.html" accesskey="n" rel="next">Wisdom Utilities</a>, Previous: <a href="Wisdom-Import.html" accesskey="p" rel="prev">Wisdom Import</a>, Up: <a href="Wisdom.html" accesskey="u" rel="up">Wisdom</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
<hr>
<span id="Forgetting-Wisdom-1"></span><h4 class="subsection">4.7.3 Forgetting Wisdom</h4>
<div class="example">
<pre class="example">void fftw_forget_wisdom(void);
</pre></div>
<span id="index-fftw_005fforget_005fwisdom-1"></span>
<p>Calling <code>fftw_forget_wisdom</code> causes all accumulated <code>wisdom</code>
to be discarded and its associated memory to be freed. (New
<code>wisdom</code> can still be gathered subsequently, however.)
</p>
</body>
</html>

View File

@@ -0,0 +1,200 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html>
<!-- This manual is for FFTW
(version 3.3.10, 10 December 2020).
Copyright (C) 2003 Matteo Frigo.
Copyright (C) 2003 Massachusetts Institute of Technology.
Permission is granted to make and distribute verbatim copies of this
manual provided the copyright notice and this permission notice are
preserved on all copies.
Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided that the
entire resulting derived work is distributed under the terms of a
permission notice identical to this one.
Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions,
except that this permission notice may be stated in a translation
approved by the Free Software Foundation. -->
<!-- Created by GNU Texinfo 6.7, http://www.gnu.org/software/texinfo/ -->
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<title>Fortran Examples (FFTW 3.3.10)</title>
<meta name="description" content="Fortran Examples (FFTW 3.3.10)">
<meta name="keywords" content="Fortran Examples (FFTW 3.3.10)">
<meta name="resource-type" content="document">
<meta name="distribution" content="global">
<meta name="Generator" content="makeinfo">
<link href="index.html" rel="start" title="Top">
<link href="Concept-Index.html" rel="index" title="Concept Index">
<link href="index.html#SEC_Contents" rel="contents" title="Table of Contents">
<link href="Calling-FFTW-from-Legacy-Fortran.html" rel="up" title="Calling FFTW from Legacy Fortran">
<link href="Wisdom-of-Fortran_003f.html" rel="next" title="Wisdom of Fortran?">
<link href="FFTW-Execution-in-Fortran.html" rel="prev" title="FFTW Execution in Fortran">
<style type="text/css">
<!--
a.summary-letter {text-decoration: none}
blockquote.indentedblock {margin-right: 0em}
div.display {margin-left: 3.2em}
div.example {margin-left: 3.2em}
div.lisp {margin-left: 3.2em}
kbd {font-style: oblique}
pre.display {font-family: inherit}
pre.format {font-family: inherit}
pre.menu-comment {font-family: serif}
pre.menu-preformatted {font-family: serif}
span.nolinebreak {white-space: nowrap}
span.roman {font-family: initial; font-weight: normal}
span.sansserif {font-family: sans-serif; font-weight: normal}
ul.no-bullet {list-style: none}
-->
</style>
</head>
<body lang="en">
<span id="Fortran-Examples"></span><div class="header">
<p>
Next: <a href="Wisdom-of-Fortran_003f.html" accesskey="n" rel="next">Wisdom of Fortran?</a>, Previous: <a href="FFTW-Execution-in-Fortran.html" accesskey="p" rel="prev">FFTW Execution in Fortran</a>, Up: <a href="Calling-FFTW-from-Legacy-Fortran.html" accesskey="u" rel="up">Calling FFTW from Legacy Fortran</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
<hr>
<span id="Fortran-Examples-1"></span><h3 class="section">8.4 Fortran Examples</h3>
<p>In C, you might have something like the following to transform a
one-dimensional complex array:
</p>
<div class="example">
<pre class="example"> fftw_complex in[N], out[N];
fftw_plan plan;
plan = fftw_plan_dft_1d(N,in,out,FFTW_FORWARD,FFTW_ESTIMATE);
fftw_execute(plan);
fftw_destroy_plan(plan);
</pre></div>
<p>In Fortran, you would use the following to accomplish the same thing:
</p>
<div class="example">
<pre class="example"> double complex in, out
dimension in(N), out(N)
integer*8 plan
call dfftw_plan_dft_1d(plan,N,in,out,FFTW_FORWARD,FFTW_ESTIMATE)
call dfftw_execute_dft(plan, in, out)
call dfftw_destroy_plan(plan)
</pre></div>
<span id="index-dfftw_005fplan_005fdft_005f1d"></span>
<span id="index-dfftw_005fexecute_005fdft-1"></span>
<span id="index-dfftw_005fdestroy_005fplan"></span>
<p>Notice how all routines are called as Fortran subroutines, and the
plan is returned via the first argument to <code>dfftw_plan_dft_1d</code>.
Notice also that we changed <code>fftw_execute</code> to
<code>dfftw_execute_dft</code> (see <a href="FFTW-Execution-in-Fortran.html">FFTW Execution in Fortran</a>). To do
the same thing, but using 8 threads in parallel (see <a href="Multi_002dthreaded-FFTW.html">Multi-threaded FFTW</a>), you would simply prefix these calls with:
</p>
<div class="example">
<pre class="example"> integer iret
call dfftw_init_threads(iret)
call dfftw_plan_with_nthreads(8)
</pre></div>
<span id="index-dfftw_005finit_005fthreads"></span>
<span id="index-dfftw_005fplan_005fwith_005fnthreads"></span>
<p>(You might want to check the value of <code>iret</code>: if it is zero, it
indicates an unlikely error during thread initialization.)
</p>
<p>To check the number of threads currently being used by the planner, you
can do the following:
</p>
<div class="example">
<pre class="example"> integer iret
call dfftw_planner_nthreads(iret)
</pre></div>
<span id="index-dfftw_005fplanner_005fnthreads"></span>
<p>To transform a three-dimensional array in-place with C, you might do:
</p>
<div class="example">
<pre class="example"> fftw_complex arr[L][M][N];
fftw_plan plan;
plan = fftw_plan_dft_3d(L,M,N, arr,arr,
FFTW_FORWARD, FFTW_ESTIMATE);
fftw_execute(plan);
fftw_destroy_plan(plan);
</pre></div>
<p>In Fortran, you would use this instead:
</p>
<div class="example">
<pre class="example"> double complex arr
dimension arr(L,M,N)
integer*8 plan
call dfftw_plan_dft_3d(plan, L,M,N, arr,arr,
&amp; FFTW_FORWARD, FFTW_ESTIMATE)
call dfftw_execute_dft(plan, arr, arr)
call dfftw_destroy_plan(plan)
</pre></div>
<span id="index-dfftw_005fplan_005fdft_005f3d"></span>
<p>Note that we pass the array dimensions in the &ldquo;natural&rdquo; order in both C
and Fortran.
</p>
<p>To transform a one-dimensional real array in Fortran, you might do:
</p>
<div class="example">
<pre class="example"> double precision in
dimension in(N)
double complex out
dimension out(N/2 + 1)
integer*8 plan
call dfftw_plan_dft_r2c_1d(plan,N,in,out,FFTW_ESTIMATE)
call dfftw_execute_dft_r2c(plan, in, out)
call dfftw_destroy_plan(plan)
</pre></div>
<span id="index-dfftw_005fplan_005fdft_005fr2c_005f1d"></span>
<span id="index-dfftw_005fexecute_005fdft_005fr2c"></span>
<p>To transform a two-dimensional real array, out of place, you might use
the following:
</p>
<div class="example">
<pre class="example"> double precision in
dimension in(M,N)
double complex out
dimension out(M/2 + 1, N)
integer*8 plan
call dfftw_plan_dft_r2c_2d(plan,M,N,in,out,FFTW_ESTIMATE)
call dfftw_execute_dft_r2c(plan, in, out)
call dfftw_destroy_plan(plan)
</pre></div>
<span id="index-dfftw_005fplan_005fdft_005fr2c_005f2d"></span>
<p><strong>Important:</strong> Notice that it is the <em>first</em> dimension of the
complex output array that is cut in half in Fortran, rather than the
last dimension as in C. This is a consequence of the interface routines
reversing the order of the array dimensions passed to FFTW so that the
Fortran program can use its ordinary column-major order.
<span id="index-column_002dmajor-3"></span>
<span id="index-r2c_002fc2r-multi_002ddimensional-array-format-3"></span>
</p>
<hr>
<div class="header">
<p>
Next: <a href="Wisdom-of-Fortran_003f.html" accesskey="n" rel="next">Wisdom of Fortran?</a>, Previous: <a href="FFTW-Execution-in-Fortran.html" accesskey="p" rel="prev">FFTW Execution in Fortran</a>, Up: <a href="Calling-FFTW-from-Legacy-Fortran.html" accesskey="u" rel="up">Calling FFTW from Legacy Fortran</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
</body>
</html>

View File

@@ -0,0 +1,162 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html>
<!-- This manual is for FFTW
(version 3.3.10, 10 December 2020).
Copyright (C) 2003 Matteo Frigo.
Copyright (C) 2003 Massachusetts Institute of Technology.
Permission is granted to make and distribute verbatim copies of this
manual provided the copyright notice and this permission notice are
preserved on all copies.
Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided that the
entire resulting derived work is distributed under the terms of a
permission notice identical to this one.
Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions,
except that this permission notice may be stated in a translation
approved by the Free Software Foundation. -->
<!-- Created by GNU Texinfo 6.7, http://www.gnu.org/software/texinfo/ -->
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<title>Fortran-interface routines (FFTW 3.3.10)</title>
<meta name="description" content="Fortran-interface routines (FFTW 3.3.10)">
<meta name="keywords" content="Fortran-interface routines (FFTW 3.3.10)">
<meta name="resource-type" content="document">
<meta name="distribution" content="global">
<meta name="Generator" content="makeinfo">
<link href="index.html" rel="start" title="Top">
<link href="Concept-Index.html" rel="index" title="Concept Index">
<link href="index.html#SEC_Contents" rel="contents" title="Table of Contents">
<link href="Calling-FFTW-from-Legacy-Fortran.html" rel="up" title="Calling FFTW from Legacy Fortran">
<link href="FFTW-Constants-in-Fortran.html" rel="next" title="FFTW Constants in Fortran">
<link href="Calling-FFTW-from-Legacy-Fortran.html" rel="prev" title="Calling FFTW from Legacy Fortran">
<style type="text/css">
<!--
a.summary-letter {text-decoration: none}
blockquote.indentedblock {margin-right: 0em}
div.display {margin-left: 3.2em}
div.example {margin-left: 3.2em}
div.lisp {margin-left: 3.2em}
kbd {font-style: oblique}
pre.display {font-family: inherit}
pre.format {font-family: inherit}
pre.menu-comment {font-family: serif}
pre.menu-preformatted {font-family: serif}
span.nolinebreak {white-space: nowrap}
span.roman {font-family: initial; font-weight: normal}
span.sansserif {font-family: sans-serif; font-weight: normal}
ul.no-bullet {list-style: none}
-->
</style>
</head>
<body lang="en">
<span id="Fortran_002dinterface-routines"></span><div class="header">
<p>
Next: <a href="FFTW-Constants-in-Fortran.html" accesskey="n" rel="next">FFTW Constants in Fortran</a>, Previous: <a href="Calling-FFTW-from-Legacy-Fortran.html" accesskey="p" rel="prev">Calling FFTW from Legacy Fortran</a>, Up: <a href="Calling-FFTW-from-Legacy-Fortran.html" accesskey="u" rel="up">Calling FFTW from Legacy Fortran</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
<hr>
<span id="Fortran_002dinterface-routines-1"></span><h3 class="section">8.1 Fortran-interface routines</h3>
<p>Nearly all of the FFTW functions have Fortran-callable equivalents.
The name of the legacy Fortran routine is the same as that of the
corresponding C routine, but with the &lsquo;<samp>fftw_</samp>&rsquo; prefix replaced by
&lsquo;<samp>dfftw_</samp>&rsquo;.<a id="DOCF9" href="#FOOT9"><sup>9</sup></a> The single and long-double precision
versions use &lsquo;<samp>sfftw_</samp>&rsquo; and &lsquo;<samp>lfftw_</samp>&rsquo;, respectively, instead of
&lsquo;<samp>fftwf_</samp>&rsquo; and &lsquo;<samp>fftwl_</samp>&rsquo;; quadruple precision (<code>real*16</code>)
is available on some systems as &lsquo;<samp>fftwq_</samp>&rsquo; (see <a href="Precision.html">Precision</a>).
(Note that <code>long double</code> on x86 hardware is usually at most
80-bit extended precision, <em>not</em> quadruple precision.)
</p>
<p>For the most part, all of the arguments to the functions are the same,
with the following exceptions:
</p>
<ul>
<li> <code>plan</code> variables (what would be of type <code>fftw_plan</code> in C),
must be declared as a type that is at least as big as a pointer
(address) on your machine. We recommend using <code>integer*8</code> everywhere,
since this should always be big enough.
<span id="index-portability-6"></span>
</li><li> Any function that returns a value (e.g. <code>fftw_plan_dft</code>) is
converted into a <em>subroutine</em>. The return value is converted into
an additional <em>first</em> parameter of this subroutine.<a id="DOCF10" href="#FOOT10"><sup>10</sup></a>
</li><li> <span id="index-column_002dmajor-2"></span>
The Fortran routines expect multi-dimensional arrays to be in
<em>column-major</em> order, which is the ordinary format of Fortran
arrays (see <a href="Multi_002ddimensional-Array-Format.html">Multi-dimensional Array Format</a>). They do this
transparently and costlessly simply by reversing the order of the
dimensions passed to FFTW, but this has one important consequence for
multi-dimensional real-complex transforms, discussed below.
</li><li> Wisdom import and export is somewhat more tricky because one cannot
easily pass files or strings between C and Fortran; see <a href="Wisdom-of-Fortran_003f.html">Wisdom of Fortran?</a>.
</li><li> Legacy Fortran cannot use the <code>fftw_malloc</code> dynamic-allocation routine.
If you want to exploit the SIMD FFTW (see <a href="SIMD-alignment-and-fftw_005fmalloc.html">SIMD alignment and fftw_malloc</a>), you&rsquo;ll
need to figure out some other way to ensure that your arrays are at
least 16-byte aligned.
</li><li> <span id="index-fftw_005fiodim-2"></span>
<span id="index-guru-interface-4"></span>
Since Fortran 77 does not have data structures, the <code>fftw_iodim</code>
structure from the guru interface (see <a href="Guru-vector-and-transform-sizes.html">Guru vector and transform sizes</a>) must be split into separate arguments. In particular, any
<code>fftw_iodim</code> array arguments in the C guru interface become three
integer array arguments (<code>n</code>, <code>is</code>, and <code>os</code>) in the
Fortran guru interface, all of whose lengths should be equal to the
corresponding <code>rank</code> argument.
</li><li> The guru planner interface in Fortran does <em>not</em> do any automatic
translation between column-major and row-major; you are responsible
for setting the strides etcetera to correspond to your Fortran arrays.
However, as a slight bug that we are preserving for backwards
compatibility, the &lsquo;<samp>plan_guru_r2r</samp>&rsquo; in Fortran <em>does</em> reverse the
order of its <code>kind</code> array parameter, so the <code>kind</code> array
of that routine should be in the reverse of the order of the iodim
arrays (see above).
</li></ul>
<p>In general, you should take care to use Fortran data types that
correspond to (i.e. are the same size as) the C types used by FFTW.
In practice, this correspondence is usually straightforward
(i.e. <code>integer</code> corresponds to <code>int</code>, <code>real</code>
corresponds to <code>float</code>, etcetera). The native Fortran
double/single-precision complex type should be compatible with
<code>fftw_complex</code>/<code>fftwf_complex</code>. Such simple correspondences
are assumed in the examples below.
<span id="index-portability-7"></span>
</p>
<div class="footnote">
<hr>
<h4 class="footnotes-heading">Footnotes</h4>
<h5><a id="FOOT9" href="#DOCF9">(9)</a></h3>
<p>Technically, Fortran 77 identifiers are not
allowed to have more than 6 characters, nor may they contain
underscores. Any compiler that enforces this limitation doesn&rsquo;t
deserve to link to FFTW.</p>
<h5><a id="FOOT10" href="#DOCF10">(10)</a></h3>
<p>The
reason for this is that some Fortran implementations seem to have
trouble with C function return values, and vice versa.</p>
</div>
<hr>
<div class="header">
<p>
Next: <a href="FFTW-Constants-in-Fortran.html" accesskey="n" rel="next">FFTW Constants in Fortran</a>, Previous: <a href="Calling-FFTW-from-Legacy-Fortran.html" accesskey="p" rel="prev">Calling FFTW from Legacy Fortran</a>, Up: <a href="Calling-FFTW-from-Legacy-Fortran.html" accesskey="u" rel="up">Calling FFTW from Legacy Fortran</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
</body>
</html>

View File

@@ -0,0 +1,119 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html>
<!-- This manual is for FFTW
(version 3.3.10, 10 December 2020).
Copyright (C) 2003 Matteo Frigo.
Copyright (C) 2003 Massachusetts Institute of Technology.
Permission is granted to make and distribute verbatim copies of this
manual provided the copyright notice and this permission notice are
preserved on all copies.
Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided that the
entire resulting derived work is distributed under the terms of a
permission notice identical to this one.
Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions,
except that this permission notice may be stated in a translation
approved by the Free Software Foundation. -->
<!-- Created by GNU Texinfo 6.7, http://www.gnu.org/software/texinfo/ -->
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<title>Generating your own code (FFTW 3.3.10)</title>
<meta name="description" content="Generating your own code (FFTW 3.3.10)">
<meta name="keywords" content="Generating your own code (FFTW 3.3.10)">
<meta name="resource-type" content="document">
<meta name="distribution" content="global">
<meta name="Generator" content="makeinfo">
<link href="index.html" rel="start" title="Top">
<link href="Concept-Index.html" rel="index" title="Concept Index">
<link href="index.html#SEC_Contents" rel="contents" title="Table of Contents">
<link href="Installation-and-Customization.html" rel="up" title="Installation and Customization">
<link href="Acknowledgments.html" rel="next" title="Acknowledgments">
<link href="Cycle-Counters.html" rel="prev" title="Cycle Counters">
<style type="text/css">
<!--
a.summary-letter {text-decoration: none}
blockquote.indentedblock {margin-right: 0em}
div.display {margin-left: 3.2em}
div.example {margin-left: 3.2em}
div.lisp {margin-left: 3.2em}
kbd {font-style: oblique}
pre.display {font-family: inherit}
pre.format {font-family: inherit}
pre.menu-comment {font-family: serif}
pre.menu-preformatted {font-family: serif}
span.nolinebreak {white-space: nowrap}
span.roman {font-family: initial; font-weight: normal}
span.sansserif {font-family: sans-serif; font-weight: normal}
ul.no-bullet {list-style: none}
-->
</style>
</head>
<body lang="en">
<span id="Generating-your-own-code"></span><div class="header">
<p>
Previous: <a href="Cycle-Counters.html" accesskey="p" rel="prev">Cycle Counters</a>, Up: <a href="Installation-and-Customization.html" accesskey="u" rel="up">Installation and Customization</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
<hr>
<span id="Generating-your-own-code-1"></span><h3 class="section">10.4 Generating your own code</h3>
<span id="index-code-generator-1"></span>
<p>The directory <code>genfft</code> contains the programs that were used to
generate FFTW&rsquo;s &ldquo;codelets,&rdquo; which are hard-coded transforms of small
sizes.
<span id="index-codelet-2"></span>
We do not expect casual users to employ the generator, which is a rather
sophisticated program that generates directed acyclic graphs of FFT
algorithms and performs algebraic simplifications on them. It was
written in Objective Caml, a dialect of ML, which is available at
<a href="http://caml.inria.fr/ocaml/index.en.html">http://caml.inria.fr/ocaml/index.en.html</a>.
<span id="index-Caml"></span>
</p>
<p>If you have Objective Caml installed (along with recent versions of
GNU <code>autoconf</code>, <code>automake</code>, and <code>libtool</code>), then you
can change the set of codelets that are generated or play with the
generation options. The set of generated codelets is specified by the
<code>{dft,rdft}/{codelets,simd}/*/Makefile.am</code> files. For example, you can add
efficient REDFT codelets of small sizes by modifying
<code>rdft/codelets/r2r/Makefile.am</code>.
<span id="index-REDFT-2"></span>
After you modify any <code>Makefile.am</code> files, you can type <code>sh
bootstrap.sh</code> in the top-level directory followed by <code>make</code> to
re-generate the files.
</p>
<p>We do not provide more details about the code-generation process, since
we do not expect that most users will need to generate their own code.
However, feel free to contact us at <a href="mailto:fftw@fftw.org">fftw@fftw.org</a> if
you are interested in the subject.
</p>
<span id="index-monadic-programming"></span>
<p>You might find it interesting to learn Caml and/or some modern
programming techniques that we used in the generator (including monadic
programming), especially if you heard the rumor that Java and
object-oriented programming are the latest advancement in the field.
The internal operation of the codelet generator is described in the
paper, &ldquo;A Fast Fourier Transform Compiler,&rdquo; by M. Frigo, which is
available from the <a href="http://www.fftw.org">FFTW home page</a> and also
appeared in the <cite>Proceedings of the 1999 ACM SIGPLAN Conference on
Programming Language Design and Implementation (PLDI)</cite>.
</p>
<hr>
<div class="header">
<p>
Previous: <a href="Cycle-Counters.html" accesskey="p" rel="prev">Cycle Counters</a>, Up: <a href="Installation-and-Customization.html" accesskey="u" rel="up">Installation and Customization</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
</body>
</html>

View File

@@ -0,0 +1,149 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html>
<!-- This manual is for FFTW
(version 3.3.10, 10 December 2020).
Copyright (C) 2003 Matteo Frigo.
Copyright (C) 2003 Massachusetts Institute of Technology.
Permission is granted to make and distribute verbatim copies of this
manual provided the copyright notice and this permission notice are
preserved on all copies.
Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided that the
entire resulting derived work is distributed under the terms of a
permission notice identical to this one.
Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions,
except that this permission notice may be stated in a translation
approved by the Free Software Foundation. -->
<!-- Created by GNU Texinfo 6.7, http://www.gnu.org/software/texinfo/ -->
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<title>Guru Complex DFTs (FFTW 3.3.10)</title>
<meta name="description" content="Guru Complex DFTs (FFTW 3.3.10)">
<meta name="keywords" content="Guru Complex DFTs (FFTW 3.3.10)">
<meta name="resource-type" content="document">
<meta name="distribution" content="global">
<meta name="Generator" content="makeinfo">
<link href="index.html" rel="start" title="Top">
<link href="Concept-Index.html" rel="index" title="Concept Index">
<link href="index.html#SEC_Contents" rel="contents" title="Table of Contents">
<link href="Guru-Interface.html" rel="up" title="Guru Interface">
<link href="Guru-Real_002ddata-DFTs.html" rel="next" title="Guru Real-data DFTs">
<link href="Guru-vector-and-transform-sizes.html" rel="prev" title="Guru vector and transform sizes">
<style type="text/css">
<!--
a.summary-letter {text-decoration: none}
blockquote.indentedblock {margin-right: 0em}
div.display {margin-left: 3.2em}
div.example {margin-left: 3.2em}
div.lisp {margin-left: 3.2em}
kbd {font-style: oblique}
pre.display {font-family: inherit}
pre.format {font-family: inherit}
pre.menu-comment {font-family: serif}
pre.menu-preformatted {font-family: serif}
span.nolinebreak {white-space: nowrap}
span.roman {font-family: initial; font-weight: normal}
span.sansserif {font-family: sans-serif; font-weight: normal}
ul.no-bullet {list-style: none}
-->
</style>
</head>
<body lang="en">
<span id="Guru-Complex-DFTs"></span><div class="header">
<p>
Next: <a href="Guru-Real_002ddata-DFTs.html" accesskey="n" rel="next">Guru Real-data DFTs</a>, Previous: <a href="Guru-vector-and-transform-sizes.html" accesskey="p" rel="prev">Guru vector and transform sizes</a>, Up: <a href="Guru-Interface.html" accesskey="u" rel="up">Guru Interface</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
<hr>
<span id="Guru-Complex-DFTs-1"></span><h4 class="subsection">4.5.3 Guru Complex DFTs</h4>
<div class="example">
<pre class="example">fftw_plan fftw_plan_guru_dft(
int rank, const fftw_iodim *dims,
int howmany_rank, const fftw_iodim *howmany_dims,
fftw_complex *in, fftw_complex *out,
int sign, unsigned flags);
fftw_plan fftw_plan_guru_split_dft(
int rank, const fftw_iodim *dims,
int howmany_rank, const fftw_iodim *howmany_dims,
double *ri, double *ii, double *ro, double *io,
unsigned flags);
</pre></div>
<span id="index-fftw_005fplan_005fguru_005fdft"></span>
<span id="index-fftw_005fplan_005fguru_005fsplit_005fdft"></span>
<p>These two functions plan a complex-data, multi-dimensional DFT
for the interleaved and split format, respectively.
Transform dimensions are given by (<code>rank</code>, <code>dims</code>) over a
multi-dimensional vector (loop) of dimensions (<code>howmany_rank</code>,
<code>howmany_dims</code>). <code>dims</code> and <code>howmany_dims</code> should point
to <code>fftw_iodim</code> arrays of length <code>rank</code> and
<code>howmany_rank</code>, respectively.
</p>
<span id="index-flags-5"></span>
<p><code>flags</code> is a bitwise OR (&lsquo;<samp>|</samp>&rsquo;) of zero or more planner flags,
as defined in <a href="Planner-Flags.html">Planner Flags</a>.
</p>
<p>In the <code>fftw_plan_guru_dft</code> function, the pointers <code>in</code> and
<code>out</code> point to the interleaved input and output arrays,
respectively. The sign can be either <em>-1</em> (=
<code>FFTW_FORWARD</code>) or <em>+1</em> (= <code>FFTW_BACKWARD</code>). If the
pointers are equal, the transform is in-place.
</p>
<p>In the <code>fftw_plan_guru_split_dft</code> function,
<code>ri</code> and <code>ii</code> point to the real and imaginary input arrays,
and <code>ro</code> and <code>io</code> point to the real and imaginary output
arrays. The input and output pointers may be the same, indicating an
in-place transform. For example, for <code>fftw_complex</code> pointers
<code>in</code> and <code>out</code>, the corresponding parameters are:
</p>
<div class="example">
<pre class="example">ri = (double *) in;
ii = (double *) in + 1;
ro = (double *) out;
io = (double *) out + 1;
</pre></div>
<p>Because <code>fftw_plan_guru_split_dft</code> accepts split arrays, strides
are expressed in units of <code>double</code>. For a contiguous
<code>fftw_complex</code> array, the overall stride of the transform should
be 2, the distance between consecutive real parts or between
consecutive imaginary parts; see <a href="Guru-vector-and-transform-sizes.html">Guru vector and transform sizes</a>. Note that the dimension strides are applied equally to the
real and imaginary parts; real and imaginary arrays with different
strides are not supported.
</p>
<p>There is no <code>sign</code> parameter in <code>fftw_plan_guru_split_dft</code>.
This function always plans for an <code>FFTW_FORWARD</code> transform. To
plan for an <code>FFTW_BACKWARD</code> transform, you can exploit the
identity that the backwards DFT is equal to the forwards DFT with the
real and imaginary parts swapped. For example, in the case of the
<code>fftw_complex</code> arrays above, the <code>FFTW_BACKWARD</code> transform
is computed by the parameters:
</p>
<div class="example">
<pre class="example">ri = (double *) in + 1;
ii = (double *) in;
ro = (double *) out + 1;
io = (double *) out;
</pre></div>
<hr>
<div class="header">
<p>
Next: <a href="Guru-Real_002ddata-DFTs.html" accesskey="n" rel="next">Guru Real-data DFTs</a>, Previous: <a href="Guru-vector-and-transform-sizes.html" accesskey="p" rel="prev">Guru vector and transform sizes</a>, Up: <a href="Guru-Interface.html" accesskey="u" rel="up">Guru Interface</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
</body>
</html>

View File

@@ -0,0 +1,102 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html>
<!-- This manual is for FFTW
(version 3.3.10, 10 December 2020).
Copyright (C) 2003 Matteo Frigo.
Copyright (C) 2003 Massachusetts Institute of Technology.
Permission is granted to make and distribute verbatim copies of this
manual provided the copyright notice and this permission notice are
preserved on all copies.
Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided that the
entire resulting derived work is distributed under the terms of a
permission notice identical to this one.
Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions,
except that this permission notice may be stated in a translation
approved by the Free Software Foundation. -->
<!-- Created by GNU Texinfo 6.7, http://www.gnu.org/software/texinfo/ -->
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<title>Guru Interface (FFTW 3.3.10)</title>
<meta name="description" content="Guru Interface (FFTW 3.3.10)">
<meta name="keywords" content="Guru Interface (FFTW 3.3.10)">
<meta name="resource-type" content="document">
<meta name="distribution" content="global">
<meta name="Generator" content="makeinfo">
<link href="index.html" rel="start" title="Top">
<link href="Concept-Index.html" rel="index" title="Concept Index">
<link href="index.html#SEC_Contents" rel="contents" title="Table of Contents">
<link href="FFTW-Reference.html" rel="up" title="FFTW Reference">
<link href="Interleaved-and-split-arrays.html" rel="next" title="Interleaved and split arrays">
<link href="Advanced-Real_002dto_002dreal-Transforms.html" rel="prev" title="Advanced Real-to-real Transforms">
<style type="text/css">
<!--
a.summary-letter {text-decoration: none}
blockquote.indentedblock {margin-right: 0em}
div.display {margin-left: 3.2em}
div.example {margin-left: 3.2em}
div.lisp {margin-left: 3.2em}
kbd {font-style: oblique}
pre.display {font-family: inherit}
pre.format {font-family: inherit}
pre.menu-comment {font-family: serif}
pre.menu-preformatted {font-family: serif}
span.nolinebreak {white-space: nowrap}
span.roman {font-family: initial; font-weight: normal}
span.sansserif {font-family: sans-serif; font-weight: normal}
ul.no-bullet {list-style: none}
-->
</style>
</head>
<body lang="en">
<span id="Guru-Interface"></span><div class="header">
<p>
Next: <a href="New_002darray-Execute-Functions.html" accesskey="n" rel="next">New-array Execute Functions</a>, Previous: <a href="Advanced-Interface.html" accesskey="p" rel="prev">Advanced Interface</a>, Up: <a href="FFTW-Reference.html" accesskey="u" rel="up">FFTW Reference</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
<hr>
<span id="Guru-Interface-1"></span><h3 class="section">4.5 Guru Interface</h3>
<span id="index-guru-interface-2"></span>
<p>The &ldquo;guru&rdquo; interface to FFTW is intended to expose as much as possible
of the flexibility in the underlying FFTW architecture. It allows one
to compute multi-dimensional &ldquo;vectors&rdquo; (loops) of multi-dimensional
transforms, where each vector/transform dimension has an independent
size and stride.
<span id="index-vector"></span>
One can also use more general complex-number formats, e.g. separate real
and imaginary arrays.
</p>
<p>For those users who require the flexibility of the guru interface, it is
important that they pay special attention to the documentation lest they
shoot themselves in the foot.
</p>
<table class="menu" border="0" cellspacing="0">
<tr><td align="left" valign="top">&bull; <a href="Interleaved-and-split-arrays.html" accesskey="1">Interleaved and split arrays</a></td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="Guru-vector-and-transform-sizes.html" accesskey="2">Guru vector and transform sizes</a></td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="Guru-Complex-DFTs.html" accesskey="3">Guru Complex DFTs</a></td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="Guru-Real_002ddata-DFTs.html" accesskey="4">Guru Real-data DFTs</a></td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="Guru-Real_002dto_002dreal-Transforms.html" accesskey="5">Guru Real-to-real Transforms</a></td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="64_002dbit-Guru-Interface.html" accesskey="6">64-bit Guru Interface</a></td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
</table>
</body>
</html>

View File

@@ -0,0 +1,148 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html>
<!-- This manual is for FFTW
(version 3.3.10, 10 December 2020).
Copyright (C) 2003 Matteo Frigo.
Copyright (C) 2003 Massachusetts Institute of Technology.
Permission is granted to make and distribute verbatim copies of this
manual provided the copyright notice and this permission notice are
preserved on all copies.
Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided that the
entire resulting derived work is distributed under the terms of a
permission notice identical to this one.
Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions,
except that this permission notice may be stated in a translation
approved by the Free Software Foundation. -->
<!-- Created by GNU Texinfo 6.7, http://www.gnu.org/software/texinfo/ -->
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<title>Guru Real-data DFTs (FFTW 3.3.10)</title>
<meta name="description" content="Guru Real-data DFTs (FFTW 3.3.10)">
<meta name="keywords" content="Guru Real-data DFTs (FFTW 3.3.10)">
<meta name="resource-type" content="document">
<meta name="distribution" content="global">
<meta name="Generator" content="makeinfo">
<link href="index.html" rel="start" title="Top">
<link href="Concept-Index.html" rel="index" title="Concept Index">
<link href="index.html#SEC_Contents" rel="contents" title="Table of Contents">
<link href="Guru-Interface.html" rel="up" title="Guru Interface">
<link href="Guru-Real_002dto_002dreal-Transforms.html" rel="next" title="Guru Real-to-real Transforms">
<link href="Guru-Complex-DFTs.html" rel="prev" title="Guru Complex DFTs">
<style type="text/css">
<!--
a.summary-letter {text-decoration: none}
blockquote.indentedblock {margin-right: 0em}
div.display {margin-left: 3.2em}
div.example {margin-left: 3.2em}
div.lisp {margin-left: 3.2em}
kbd {font-style: oblique}
pre.display {font-family: inherit}
pre.format {font-family: inherit}
pre.menu-comment {font-family: serif}
pre.menu-preformatted {font-family: serif}
span.nolinebreak {white-space: nowrap}
span.roman {font-family: initial; font-weight: normal}
span.sansserif {font-family: sans-serif; font-weight: normal}
ul.no-bullet {list-style: none}
-->
</style>
</head>
<body lang="en">
<span id="Guru-Real_002ddata-DFTs"></span><div class="header">
<p>
Next: <a href="Guru-Real_002dto_002dreal-Transforms.html" accesskey="n" rel="next">Guru Real-to-real Transforms</a>, Previous: <a href="Guru-Complex-DFTs.html" accesskey="p" rel="prev">Guru Complex DFTs</a>, Up: <a href="Guru-Interface.html" accesskey="u" rel="up">Guru Interface</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
<hr>
<span id="Guru-Real_002ddata-DFTs-1"></span><h4 class="subsection">4.5.4 Guru Real-data DFTs</h4>
<div class="example">
<pre class="example">fftw_plan fftw_plan_guru_dft_r2c(
int rank, const fftw_iodim *dims,
int howmany_rank, const fftw_iodim *howmany_dims,
double *in, fftw_complex *out,
unsigned flags);
fftw_plan fftw_plan_guru_split_dft_r2c(
int rank, const fftw_iodim *dims,
int howmany_rank, const fftw_iodim *howmany_dims,
double *in, double *ro, double *io,
unsigned flags);
fftw_plan fftw_plan_guru_dft_c2r(
int rank, const fftw_iodim *dims,
int howmany_rank, const fftw_iodim *howmany_dims,
fftw_complex *in, double *out,
unsigned flags);
fftw_plan fftw_plan_guru_split_dft_c2r(
int rank, const fftw_iodim *dims,
int howmany_rank, const fftw_iodim *howmany_dims,
double *ri, double *ii, double *out,
unsigned flags);
</pre></div>
<span id="index-fftw_005fplan_005fguru_005fdft_005fr2c"></span>
<span id="index-fftw_005fplan_005fguru_005fsplit_005fdft_005fr2c"></span>
<span id="index-fftw_005fplan_005fguru_005fdft_005fc2r"></span>
<span id="index-fftw_005fplan_005fguru_005fsplit_005fdft_005fc2r"></span>
<p>Plan a real-input (r2c) or real-output (c2r), multi-dimensional DFT with
transform dimensions given by (<code>rank</code>, <code>dims</code>) over a
multi-dimensional vector (loop) of dimensions (<code>howmany_rank</code>,
<code>howmany_dims</code>). <code>dims</code> and <code>howmany_dims</code> should point
to <code>fftw_iodim</code> arrays of length <code>rank</code> and
<code>howmany_rank</code>, respectively. As for the basic and advanced
interfaces, an r2c transform is <code>FFTW_FORWARD</code> and a c2r transform
is <code>FFTW_BACKWARD</code>.
</p>
<p>The <em>last</em> dimension of <code>dims</code> is interpreted specially:
that dimension of the real array has size <code>dims[rank-1].n</code>, but
that dimension of the complex array has size <code>dims[rank-1].n/2+1</code>
(division rounded down). The strides, on the other hand, are taken to
be exactly as specified. It is up to the user to specify the strides
appropriately for the peculiar dimensions of the data, and we do not
guarantee that the planner will succeed (return non-<code>NULL</code>) for
any dimensions other than those described in <a href="Real_002ddata-DFT-Array-Format.html">Real-data DFT Array Format</a> and generalized in <a href="Advanced-Real_002ddata-DFTs.html">Advanced Real-data DFTs</a>. (That is,
for an in-place transform, each individual dimension should be able to
operate in place.)
<span id="index-in_002dplace-6"></span>
</p>
<p><code>in</code> and <code>out</code> point to the input and output arrays for r2c
and c2r transforms, respectively. For split arrays, <code>ri</code> and
<code>ii</code> point to the real and imaginary input arrays for a c2r
transform, and <code>ro</code> and <code>io</code> point to the real and imaginary
output arrays for an r2c transform. <code>in</code> and <code>ro</code> or
<code>ri</code> and <code>out</code> may be the same, indicating an in-place
transform. (In-place transforms where <code>in</code> and <code>io</code> or
<code>ii</code> and <code>out</code> are the same are not currently supported.)
</p>
<span id="index-flags-6"></span>
<p><code>flags</code> is a bitwise OR (&lsquo;<samp>|</samp>&rsquo;) of zero or more planner flags,
as defined in <a href="Planner-Flags.html">Planner Flags</a>.
</p>
<p>In-place transforms of rank greater than 1 are currently only
supported for interleaved arrays. For split arrays, the planner will
return <code>NULL</code>.
<span id="index-in_002dplace-7"></span>
</p>
<hr>
<div class="header">
<p>
Next: <a href="Guru-Real_002dto_002dreal-Transforms.html" accesskey="n" rel="next">Guru Real-to-real Transforms</a>, Previous: <a href="Guru-Complex-DFTs.html" accesskey="p" rel="prev">Guru Complex DFTs</a>, Up: <a href="Guru-Interface.html" accesskey="u" rel="up">Guru Interface</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
</body>
</html>

View File

@@ -0,0 +1,101 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html>
<!-- This manual is for FFTW
(version 3.3.10, 10 December 2020).
Copyright (C) 2003 Matteo Frigo.
Copyright (C) 2003 Massachusetts Institute of Technology.
Permission is granted to make and distribute verbatim copies of this
manual provided the copyright notice and this permission notice are
preserved on all copies.
Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided that the
entire resulting derived work is distributed under the terms of a
permission notice identical to this one.
Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions,
except that this permission notice may be stated in a translation
approved by the Free Software Foundation. -->
<!-- Created by GNU Texinfo 6.7, http://www.gnu.org/software/texinfo/ -->
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<title>Guru Real-to-real Transforms (FFTW 3.3.10)</title>
<meta name="description" content="Guru Real-to-real Transforms (FFTW 3.3.10)">
<meta name="keywords" content="Guru Real-to-real Transforms (FFTW 3.3.10)">
<meta name="resource-type" content="document">
<meta name="distribution" content="global">
<meta name="Generator" content="makeinfo">
<link href="index.html" rel="start" title="Top">
<link href="Concept-Index.html" rel="index" title="Concept Index">
<link href="index.html#SEC_Contents" rel="contents" title="Table of Contents">
<link href="Guru-Interface.html" rel="up" title="Guru Interface">
<link href="64_002dbit-Guru-Interface.html" rel="next" title="64-bit Guru Interface">
<link href="Guru-Real_002ddata-DFTs.html" rel="prev" title="Guru Real-data DFTs">
<style type="text/css">
<!--
a.summary-letter {text-decoration: none}
blockquote.indentedblock {margin-right: 0em}
div.display {margin-left: 3.2em}
div.example {margin-left: 3.2em}
div.lisp {margin-left: 3.2em}
kbd {font-style: oblique}
pre.display {font-family: inherit}
pre.format {font-family: inherit}
pre.menu-comment {font-family: serif}
pre.menu-preformatted {font-family: serif}
span.nolinebreak {white-space: nowrap}
span.roman {font-family: initial; font-weight: normal}
span.sansserif {font-family: sans-serif; font-weight: normal}
ul.no-bullet {list-style: none}
-->
</style>
</head>
<body lang="en">
<span id="Guru-Real_002dto_002dreal-Transforms"></span><div class="header">
<p>
Next: <a href="64_002dbit-Guru-Interface.html" accesskey="n" rel="next">64-bit Guru Interface</a>, Previous: <a href="Guru-Real_002ddata-DFTs.html" accesskey="p" rel="prev">Guru Real-data DFTs</a>, Up: <a href="Guru-Interface.html" accesskey="u" rel="up">Guru Interface</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
<hr>
<span id="Guru-Real_002dto_002dreal-Transforms-1"></span><h4 class="subsection">4.5.5 Guru Real-to-real Transforms</h4>
<div class="example">
<pre class="example">fftw_plan fftw_plan_guru_r2r(int rank, const fftw_iodim *dims,
int howmany_rank,
const fftw_iodim *howmany_dims,
double *in, double *out,
const fftw_r2r_kind *kind,
unsigned flags);
</pre></div>
<span id="index-fftw_005fplan_005fguru_005fr2r"></span>
<p>Plan a real-to-real (r2r) multi-dimensional <code>FFTW_FORWARD</code>
transform with transform dimensions given by (<code>rank</code>, <code>dims</code>)
over a multi-dimensional vector (loop) of dimensions
(<code>howmany_rank</code>, <code>howmany_dims</code>). <code>dims</code> and
<code>howmany_dims</code> should point to <code>fftw_iodim</code> arrays of length
<code>rank</code> and <code>howmany_rank</code>, respectively.
</p>
<p>The transform kind of each dimension is given by the <code>kind</code>
parameter, which should point to an array of length <code>rank</code>. Valid
<code>fftw_r2r_kind</code> constants are given in <a href="Real_002dto_002dReal-Transform-Kinds.html">Real-to-Real Transform Kinds</a>.
</p>
<p><code>in</code> and <code>out</code> point to the real input and output arrays; they
may be the same, indicating an in-place transform.
</p>
<span id="index-flags-7"></span>
<p><code>flags</code> is a bitwise OR (&lsquo;<samp>|</samp>&rsquo;) of zero or more planner flags,
as defined in <a href="Planner-Flags.html">Planner Flags</a>.
</p>
</body>
</html>

View File

@@ -0,0 +1,142 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html>
<!-- This manual is for FFTW
(version 3.3.10, 10 December 2020).
Copyright (C) 2003 Matteo Frigo.
Copyright (C) 2003 Massachusetts Institute of Technology.
Permission is granted to make and distribute verbatim copies of this
manual provided the copyright notice and this permission notice are
preserved on all copies.
Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided that the
entire resulting derived work is distributed under the terms of a
permission notice identical to this one.
Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions,
except that this permission notice may be stated in a translation
approved by the Free Software Foundation. -->
<!-- Created by GNU Texinfo 6.7, http://www.gnu.org/software/texinfo/ -->
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<title>Guru vector and transform sizes (FFTW 3.3.10)</title>
<meta name="description" content="Guru vector and transform sizes (FFTW 3.3.10)">
<meta name="keywords" content="Guru vector and transform sizes (FFTW 3.3.10)">
<meta name="resource-type" content="document">
<meta name="distribution" content="global">
<meta name="Generator" content="makeinfo">
<link href="index.html" rel="start" title="Top">
<link href="Concept-Index.html" rel="index" title="Concept Index">
<link href="index.html#SEC_Contents" rel="contents" title="Table of Contents">
<link href="Guru-Interface.html" rel="up" title="Guru Interface">
<link href="Guru-Complex-DFTs.html" rel="next" title="Guru Complex DFTs">
<link href="Interleaved-and-split-arrays.html" rel="prev" title="Interleaved and split arrays">
<style type="text/css">
<!--
a.summary-letter {text-decoration: none}
blockquote.indentedblock {margin-right: 0em}
div.display {margin-left: 3.2em}
div.example {margin-left: 3.2em}
div.lisp {margin-left: 3.2em}
kbd {font-style: oblique}
pre.display {font-family: inherit}
pre.format {font-family: inherit}
pre.menu-comment {font-family: serif}
pre.menu-preformatted {font-family: serif}
span.nolinebreak {white-space: nowrap}
span.roman {font-family: initial; font-weight: normal}
span.sansserif {font-family: sans-serif; font-weight: normal}
ul.no-bullet {list-style: none}
-->
</style>
</head>
<body lang="en">
<span id="Guru-vector-and-transform-sizes"></span><div class="header">
<p>
Next: <a href="Guru-Complex-DFTs.html" accesskey="n" rel="next">Guru Complex DFTs</a>, Previous: <a href="Interleaved-and-split-arrays.html" accesskey="p" rel="prev">Interleaved and split arrays</a>, Up: <a href="Guru-Interface.html" accesskey="u" rel="up">Guru Interface</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
<hr>
<span id="Guru-vector-and-transform-sizes-1"></span><h4 class="subsection">4.5.2 Guru vector and transform sizes</h4>
<p>The guru interface introduces one basic new data structure,
<code>fftw_iodim</code>, that is used to specify sizes and strides for
multi-dimensional transforms and vectors:
</p>
<div class="example">
<pre class="example">typedef struct {
int n;
int is;
int os;
} fftw_iodim;
</pre></div>
<span id="index-fftw_005fiodim"></span>
<p>Here, <code>n</code> is the size of the dimension, and <code>is</code> and <code>os</code>
are the strides of that dimension for the input and output arrays. (The
stride is the separation of consecutive elements along this dimension.)
</p>
<p>The meaning of the stride parameter depends on the type of the array
that the stride refers to. <em>If the array is interleaved complex,
strides are expressed in units of complex numbers
(<code>fftw_complex</code>). If the array is split complex or real, strides
are expressed in units of real numbers (<code>double</code>).</em> This
convention is consistent with the usual pointer arithmetic in the C
language. An interleaved array is denoted by a pointer <code>p</code> to
<code>fftw_complex</code>, so that <code>p+1</code> points to the next complex
number. Split arrays are denoted by pointers to <code>double</code>, in
which case pointer arithmetic operates in units of
<code>sizeof(double)</code>.
<span id="index-stride-2"></span>
</p>
<p>The guru planner interfaces all take a (<code>rank</code>, <code>dims[rank]</code>)
pair describing the transform size, and a (<code>howmany_rank</code>,
<code>howmany_dims[howmany_rank]</code>) pair describing the &ldquo;vector&rdquo; size (a
multi-dimensional loop of transforms to perform), where <code>dims</code> and
<code>howmany_dims</code> are arrays of <code>fftw_iodim</code>. Each <code>n</code> field must
be positive for <code>dims</code> and nonnegative for <code>howmany_dims</code>, while both
<code>rank</code> and <code>howmany_rank</code> must be nonnegative.
</p>
<p>For example, the <code>howmany</code> parameter in the advanced complex-DFT
interface corresponds to <code>howmany_rank</code> = 1,
<code>howmany_dims[0].n</code> = <code>howmany</code>, <code>howmany_dims[0].is</code> =
<code>idist</code>, and <code>howmany_dims[0].os</code> = <code>odist</code>.
<span id="index-howmany-loop"></span>
<span id="index-dist-1"></span>
(To compute a single transform, you can just use <code>howmany_rank</code> = 0.)
</p>
<p>A row-major multidimensional array with dimensions <code>n[rank]</code>
(see <a href="Row_002dmajor-Format.html">Row-major Format</a>) corresponds to <code>dims[i].n</code> =
<code>n[i]</code> and the recurrence <code>dims[i].is</code> = <code>n[i+1] *
dims[i+1].is</code> (similarly for <code>os</code>). The stride of the last
(<code>i=rank-1</code>) dimension is the overall stride of the array.
e.g. to be equivalent to the advanced complex-DFT interface, you would
have <code>dims[rank-1].is</code> = <code>istride</code> and
<code>dims[rank-1].os</code> = <code>ostride</code>.
<span id="index-row_002dmajor-3"></span>
</p>
<p>In general, we only guarantee FFTW to return a non-<code>NULL</code> plan if
the vector and transform dimensions correspond to a set of distinct
indices, and for in-place transforms the input/output strides should
be the same.
</p>
<hr>
<div class="header">
<p>
Next: <a href="Guru-Complex-DFTs.html" accesskey="n" rel="next">Guru Complex DFTs</a>, Previous: <a href="Interleaved-and-split-arrays.html" accesskey="p" rel="prev">Interleaved and split arrays</a>, Up: <a href="Guru-Interface.html" accesskey="u" rel="up">Guru Interface</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
</body>
</html>

View File

@@ -0,0 +1,91 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html>
<!-- This manual is for FFTW
(version 3.3.10, 10 December 2020).
Copyright (C) 2003 Matteo Frigo.
Copyright (C) 2003 Massachusetts Institute of Technology.
Permission is granted to make and distribute verbatim copies of this
manual provided the copyright notice and this permission notice are
preserved on all copies.
Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided that the
entire resulting derived work is distributed under the terms of a
permission notice identical to this one.
Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions,
except that this permission notice may be stated in a translation
approved by the Free Software Foundation. -->
<!-- Created by GNU Texinfo 6.7, http://www.gnu.org/software/texinfo/ -->
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<title>How Many Threads to Use? (FFTW 3.3.10)</title>
<meta name="description" content="How Many Threads to Use? (FFTW 3.3.10)">
<meta name="keywords" content="How Many Threads to Use? (FFTW 3.3.10)">
<meta name="resource-type" content="document">
<meta name="distribution" content="global">
<meta name="Generator" content="makeinfo">
<link href="index.html" rel="start" title="Top">
<link href="Concept-Index.html" rel="index" title="Concept Index">
<link href="index.html#SEC_Contents" rel="contents" title="Table of Contents">
<link href="Multi_002dthreaded-FFTW.html" rel="up" title="Multi-threaded FFTW">
<link href="Thread-safety.html" rel="next" title="Thread safety">
<link href="Usage-of-Multi_002dthreaded-FFTW.html" rel="prev" title="Usage of Multi-threaded FFTW">
<style type="text/css">
<!--
a.summary-letter {text-decoration: none}
blockquote.indentedblock {margin-right: 0em}
div.display {margin-left: 3.2em}
div.example {margin-left: 3.2em}
div.lisp {margin-left: 3.2em}
kbd {font-style: oblique}
pre.display {font-family: inherit}
pre.format {font-family: inherit}
pre.menu-comment {font-family: serif}
pre.menu-preformatted {font-family: serif}
span.nolinebreak {white-space: nowrap}
span.roman {font-family: initial; font-weight: normal}
span.sansserif {font-family: sans-serif; font-weight: normal}
ul.no-bullet {list-style: none}
-->
</style>
</head>
<body lang="en">
<span id="How-Many-Threads-to-Use_003f"></span><div class="header">
<p>
Next: <a href="Thread-safety.html" accesskey="n" rel="next">Thread safety</a>, Previous: <a href="Usage-of-Multi_002dthreaded-FFTW.html" accesskey="p" rel="prev">Usage of Multi-threaded FFTW</a>, Up: <a href="Multi_002dthreaded-FFTW.html" accesskey="u" rel="up">Multi-threaded FFTW</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
<hr>
<span id="How-Many-Threads-to-Use_003f-1"></span><h3 class="section">5.3 How Many Threads to Use?</h3>
<span id="index-number-of-threads"></span>
<p>There is a fair amount of overhead involved in synchronizing threads,
so the optimal number of threads to use depends upon the size of the
transform as well as on the number of processors you have.
</p>
<p>As a general rule, you don&rsquo;t want to use more threads than you have
processors. (Using more threads will work, but there will be extra
overhead with no benefit.) In fact, if the problem size is too small,
you may want to use fewer threads than you have processors.
</p>
<p>You will have to experiment with your system to see what level of
parallelization is best for your problem size. Typically, the problem
will have to involve at least a few thousand data points before threads
become beneficial. If you plan with <code>FFTW_PATIENT</code>, it will
automatically disable threads for sizes that don&rsquo;t benefit from
parallelization.
<span id="index-FFTW_005fPATIENT-3"></span>
</p>
</body>
</html>

View File

@@ -0,0 +1,121 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html>
<!-- This manual is for FFTW
(version 3.3.10, 10 December 2020).
Copyright (C) 2003 Matteo Frigo.
Copyright (C) 2003 Massachusetts Institute of Technology.
Permission is granted to make and distribute verbatim copies of this
manual provided the copyright notice and this permission notice are
preserved on all copies.
Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided that the
entire resulting derived work is distributed under the terms of a
permission notice identical to this one.
Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions,
except that this permission notice may be stated in a translation
approved by the Free Software Foundation. -->
<!-- Created by GNU Texinfo 6.7, http://www.gnu.org/software/texinfo/ -->
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<title>Installation and Customization (FFTW 3.3.10)</title>
<meta name="description" content="Installation and Customization (FFTW 3.3.10)">
<meta name="keywords" content="Installation and Customization (FFTW 3.3.10)">
<meta name="resource-type" content="document">
<meta name="distribution" content="global">
<meta name="Generator" content="makeinfo">
<link href="index.html" rel="start" title="Top">
<link href="Concept-Index.html" rel="index" title="Concept Index">
<link href="index.html#SEC_Contents" rel="contents" title="Table of Contents">
<link href="index.html" rel="up" title="Top">
<link href="Installation-on-Unix.html" rel="next" title="Installation on Unix">
<link href="Upgrading-from-FFTW-version-2.html" rel="prev" title="Upgrading from FFTW version 2">
<style type="text/css">
<!--
a.summary-letter {text-decoration: none}
blockquote.indentedblock {margin-right: 0em}
div.display {margin-left: 3.2em}
div.example {margin-left: 3.2em}
div.lisp {margin-left: 3.2em}
kbd {font-style: oblique}
pre.display {font-family: inherit}
pre.format {font-family: inherit}
pre.menu-comment {font-family: serif}
pre.menu-preformatted {font-family: serif}
span.nolinebreak {white-space: nowrap}
span.roman {font-family: initial; font-weight: normal}
span.sansserif {font-family: sans-serif; font-weight: normal}
ul.no-bullet {list-style: none}
-->
</style>
</head>
<body lang="en">
<span id="Installation-and-Customization"></span><div class="header">
<p>
Next: <a href="Acknowledgments.html" accesskey="n" rel="next">Acknowledgments</a>, Previous: <a href="Upgrading-from-FFTW-version-2.html" accesskey="p" rel="prev">Upgrading from FFTW version 2</a>, Up: <a href="index.html" accesskey="u" rel="up">Top</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
<hr>
<span id="Installation-and-Customization-1"></span><h2 class="chapter">10 Installation and Customization</h2>
<span id="index-installation"></span>
<p>This chapter describes the installation and customization of FFTW, the
latest version of which may be downloaded from
<a href="http://www.fftw.org">the FFTW home page</a>.
</p>
<p>In principle, FFTW should work on any system with an ANSI C compiler
(<code>gcc</code> is fine). However, planner time is drastically reduced if
FFTW can exploit a hardware cycle counter; FFTW comes with cycle-counter
support for all modern general-purpose CPUs, but you may need to add a
couple of lines of code if your compiler is not yet supported
(see <a href="Cycle-Counters.html">Cycle Counters</a>). (On Unix, there will be a warning at the end
of the <code>configure</code> output if no cycle counter is found.)
<span id="index-cycle-counter"></span>
<span id="index-compiler-1"></span>
<span id="index-portability-9"></span>
</p>
<p>Installation of FFTW is simplest if you have a Unix or a GNU system,
such as GNU/Linux, and we describe this case in the first section below,
including the use of special configuration options to e.g. install
different precisions or exploit optimizations for particular
architectures (e.g. SIMD). Compilation on non-Unix systems is a more
manual process, but we outline the procedure in the second section. It
is also likely that pre-compiled binaries will be available for popular
systems.
</p>
<p>Finally, we describe how you can customize FFTW for particular needs by
generating <em>codelets</em> for fast transforms of sizes not supported
efficiently by the standard FFTW distribution.
<span id="index-codelet-1"></span>
</p>
<table class="menu" border="0" cellspacing="0">
<tr><td align="left" valign="top">&bull; <a href="Installation-on-Unix.html" accesskey="1">Installation on Unix</a></td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="Installation-on-non_002dUnix-systems.html" accesskey="2">Installation on non-Unix systems</a></td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="Cycle-Counters.html" accesskey="3">Cycle Counters</a></td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="Generating-your-own-code.html" accesskey="4">Generating your own code</a></td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
</table>
<hr>
<div class="header">
<p>
Next: <a href="Acknowledgments.html" accesskey="n" rel="next">Acknowledgments</a>, Previous: <a href="Upgrading-from-FFTW-version-2.html" accesskey="p" rel="prev">Upgrading from FFTW version 2</a>, Up: <a href="index.html" accesskey="u" rel="up">Top</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
</body>
</html>

View File

@@ -0,0 +1,106 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html>
<!-- This manual is for FFTW
(version 3.3.10, 10 December 2020).
Copyright (C) 2003 Matteo Frigo.
Copyright (C) 2003 Massachusetts Institute of Technology.
Permission is granted to make and distribute verbatim copies of this
manual provided the copyright notice and this permission notice are
preserved on all copies.
Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided that the
entire resulting derived work is distributed under the terms of a
permission notice identical to this one.
Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions,
except that this permission notice may be stated in a translation
approved by the Free Software Foundation. -->
<!-- Created by GNU Texinfo 6.7, http://www.gnu.org/software/texinfo/ -->
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<title>Installation and Supported Hardware/Software (FFTW 3.3.10)</title>
<meta name="description" content="Installation and Supported Hardware/Software (FFTW 3.3.10)">
<meta name="keywords" content="Installation and Supported Hardware/Software (FFTW 3.3.10)">
<meta name="resource-type" content="document">
<meta name="distribution" content="global">
<meta name="Generator" content="makeinfo">
<link href="index.html" rel="start" title="Top">
<link href="Concept-Index.html" rel="index" title="Concept Index">
<link href="index.html#SEC_Contents" rel="contents" title="Table of Contents">
<link href="Multi_002dthreaded-FFTW.html" rel="up" title="Multi-threaded FFTW">
<link href="Usage-of-Multi_002dthreaded-FFTW.html" rel="next" title="Usage of Multi-threaded FFTW">
<link href="Multi_002dthreaded-FFTW.html" rel="prev" title="Multi-threaded FFTW">
<style type="text/css">
<!--
a.summary-letter {text-decoration: none}
blockquote.indentedblock {margin-right: 0em}
div.display {margin-left: 3.2em}
div.example {margin-left: 3.2em}
div.lisp {margin-left: 3.2em}
kbd {font-style: oblique}
pre.display {font-family: inherit}
pre.format {font-family: inherit}
pre.menu-comment {font-family: serif}
pre.menu-preformatted {font-family: serif}
span.nolinebreak {white-space: nowrap}
span.roman {font-family: initial; font-weight: normal}
span.sansserif {font-family: sans-serif; font-weight: normal}
ul.no-bullet {list-style: none}
-->
</style>
</head>
<body lang="en">
<span id="Installation-and-Supported-Hardware_002fSoftware"></span><div class="header">
<p>
Next: <a href="Usage-of-Multi_002dthreaded-FFTW.html" accesskey="n" rel="next">Usage of Multi-threaded FFTW</a>, Previous: <a href="Multi_002dthreaded-FFTW.html" accesskey="p" rel="prev">Multi-threaded FFTW</a>, Up: <a href="Multi_002dthreaded-FFTW.html" accesskey="u" rel="up">Multi-threaded FFTW</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
<hr>
<span id="Installation-and-Supported-Hardware_002fSoftware-1"></span><h3 class="section">5.1 Installation and Supported Hardware/Software</h3>
<p>All of the FFTW threads code is located in the <code>threads</code>
subdirectory of the FFTW package. On Unix systems, the FFTW threads
libraries and header files can be automatically configured, compiled,
and installed along with the uniprocessor FFTW libraries simply by
including <code>--enable-threads</code> in the flags to the <code>configure</code>
script (see <a href="Installation-on-Unix.html">Installation on Unix</a>), or <code>--enable-openmp</code> to use
<a href="http://www.openmp.org">OpenMP</a> threads.
<span id="index-configure"></span>
</p>
<span id="index-portability-3"></span>
<span id="index-OpenMP"></span>
<p>The threads routines require your operating system to have some sort
of shared-memory threads support. Specifically, the FFTW threads
package works with POSIX threads (available on most Unix variants,
from GNU/Linux to MacOS X) and Win32 threads. OpenMP threads, which
are supported in many common compilers (e.g. gcc) are also supported,
and may give better performance on some systems. (OpenMP threads are
also useful if you are employing OpenMP in your own code, in order to
minimize conflicts between threading models.) If you have a
shared-memory machine that uses a different threads API, it should be
a simple matter of programming to include support for it; see the file
<code>threads/threads.c</code> for more detail.
</p>
<p>You can compile FFTW with <em>both</em> <code>--enable-threads</code> and
<code>--enable-openmp</code> at the same time, since they install libraries
with different names (&lsquo;<samp>fftw3_threads</samp>&rsquo; and &lsquo;<samp>fftw3_omp</samp>&rsquo;, as
described below). However, your programs may only link to <em>one</em>
of these two libraries at a time.
</p>
<p>Ideally, of course, you should also have multiple processors in order to
get any benefit from the threaded transforms.
</p>
</body>
</html>

View File

@@ -0,0 +1,253 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html>
<!-- This manual is for FFTW
(version 3.3.10, 10 December 2020).
Copyright (C) 2003 Matteo Frigo.
Copyright (C) 2003 Massachusetts Institute of Technology.
Permission is granted to make and distribute verbatim copies of this
manual provided the copyright notice and this permission notice are
preserved on all copies.
Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided that the
entire resulting derived work is distributed under the terms of a
permission notice identical to this one.
Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions,
except that this permission notice may be stated in a translation
approved by the Free Software Foundation. -->
<!-- Created by GNU Texinfo 6.7, http://www.gnu.org/software/texinfo/ -->
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<title>Installation on Unix (FFTW 3.3.10)</title>
<meta name="description" content="Installation on Unix (FFTW 3.3.10)">
<meta name="keywords" content="Installation on Unix (FFTW 3.3.10)">
<meta name="resource-type" content="document">
<meta name="distribution" content="global">
<meta name="Generator" content="makeinfo">
<link href="index.html" rel="start" title="Top">
<link href="Concept-Index.html" rel="index" title="Concept Index">
<link href="index.html#SEC_Contents" rel="contents" title="Table of Contents">
<link href="Installation-and-Customization.html" rel="up" title="Installation and Customization">
<link href="Installation-on-non_002dUnix-systems.html" rel="next" title="Installation on non-Unix systems">
<link href="Installation-and-Customization.html" rel="prev" title="Installation and Customization">
<style type="text/css">
<!--
a.summary-letter {text-decoration: none}
blockquote.indentedblock {margin-right: 0em}
div.display {margin-left: 3.2em}
div.example {margin-left: 3.2em}
div.lisp {margin-left: 3.2em}
kbd {font-style: oblique}
pre.display {font-family: inherit}
pre.format {font-family: inherit}
pre.menu-comment {font-family: serif}
pre.menu-preformatted {font-family: serif}
span.nolinebreak {white-space: nowrap}
span.roman {font-family: initial; font-weight: normal}
span.sansserif {font-family: sans-serif; font-weight: normal}
ul.no-bullet {list-style: none}
-->
</style>
</head>
<body lang="en">
<span id="Installation-on-Unix"></span><div class="header">
<p>
Next: <a href="Installation-on-non_002dUnix-systems.html" accesskey="n" rel="next">Installation on non-Unix systems</a>, Previous: <a href="Installation-and-Customization.html" accesskey="p" rel="prev">Installation and Customization</a>, Up: <a href="Installation-and-Customization.html" accesskey="u" rel="up">Installation and Customization</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
<hr>
<span id="Installation-on-Unix-1"></span><h3 class="section">10.1 Installation on Unix</h3>
<p>FFTW comes with a <code>configure</code> program in the GNU style.
Installation can be as simple as:
<span id="index-configure-2"></span>
</p>
<div class="example">
<pre class="example">./configure
make
make install
</pre></div>
<p>This will build the uniprocessor complex and real transform libraries
along with the test programs. (We recommend that you use GNU
<code>make</code> if it is available; on some systems it is called
<code>gmake</code>.) The &ldquo;<code>make install</code>&rdquo; command installs the fftw
and rfftw libraries in standard places, and typically requires root
privileges (unless you specify a different install directory with the
<code>--prefix</code> flag to <code>configure</code>). You can also type
&ldquo;<code>make check</code>&rdquo; to put the FFTW test programs through their paces.
If you have problems during configuration or compilation, you may want
to run &ldquo;<code>make distclean</code>&rdquo; before trying again; this ensures that
you don&rsquo;t have any stale files left over from previous compilation
attempts.
</p>
<p>The <code>configure</code> script chooses the <code>gcc</code> compiler by default,
if it is available; you can select some other compiler with:
</p><div class="example">
<pre class="example">./configure CC=&quot;<span class="roman"><i>&lt;the name of your C compiler&gt;</i></span>&quot;
</pre></div>
<p>The <code>configure</code> script knows good <code>CFLAGS</code> (C compiler flags)
<span id="index-compiler-flags"></span>
for a few systems. If your system is not known, the <code>configure</code>
script will print out a warning. In this case, you should re-configure
FFTW with the command
</p><div class="example">
<pre class="example">./configure CFLAGS=&quot;<span class="roman"><i>&lt;write your CFLAGS here&gt;</i></span>&quot;
</pre></div>
<p>and then compile as usual. If you do find an optimal set of
<code>CFLAGS</code> for your system, please let us know what they are (along
with the output of <code>config.guess</code>) so that we can include them in
future releases.
</p>
<p><code>configure</code> supports all the standard flags defined by the GNU
Coding Standards; see the <code>INSTALL</code> file in FFTW or
<a href="http://www.gnu.org/prep/standards/html_node/index.html">the GNU web page</a>.
Note especially <code>--help</code> to list all flags and
<code>--enable-shared</code> to create shared, rather than static, libraries.
<code>configure</code> also accepts a few FFTW-specific flags, particularly:
</p>
<ul>
<li> <span id="index-precision-9"></span>
<code>--enable-float</code>: Produces a single-precision version of FFTW
(<code>float</code>) instead of the default double-precision (<code>double</code>).
See <a href="Precision.html">Precision</a>.
</li><li> <span id="index-precision-10"></span>
<code>--enable-long-double</code>: Produces a long-double precision version of
FFTW (<code>long double</code>) instead of the default double-precision
(<code>double</code>). The <code>configure</code> script will halt with an error
message if <code>long double</code> is the same size as <code>double</code> on your
machine/compiler. See <a href="Precision.html">Precision</a>.
</li><li> <span id="index-precision-11"></span>
<code>--enable-quad-precision</code>: Produces a quadruple-precision version
of FFTW using the nonstandard <code>__float128</code> type provided by
<code>gcc</code> 4.6 or later on x86, x86-64, and Itanium architectures,
instead of the default double-precision (<code>double</code>). The
<code>configure</code> script will halt with an error message if the
compiler is not <code>gcc</code> version 4.6 or later or if <code>gcc</code>&rsquo;s
<code>libquadmath</code> library is not installed. See <a href="Precision.html">Precision</a>.
</li><li> <span id="index-threads-3"></span>
<code>--enable-threads</code>: Enables compilation and installation of the
FFTW threads library (see <a href="Multi_002dthreaded-FFTW.html">Multi-threaded FFTW</a>), which provides a
simple interface to parallel transforms for SMP systems. By default,
the threads routines are not compiled.
</li><li> <code>--enable-openmp</code>: Like <code>--enable-threads</code>, but using OpenMP
compiler directives in order to induce parallelism rather than
spawning its own threads directly, and installing an &lsquo;<samp>fftw3_omp</samp>&rsquo; library
rather than an &lsquo;<samp>fftw3_threads</samp>&rsquo; library (see <a href="Multi_002dthreaded-FFTW.html">Multi-threaded FFTW</a>). You can use both <code>--enable-openmp</code> and <code>--enable-threads</code>
since they compile/install libraries with different names. By default,
the OpenMP routines are not compiled.
</li><li> <code>--with-combined-threads</code>: By default, if <code>--enable-threads</code>
is used, the threads support is compiled into a separate library that
must be linked in addition to the main FFTW library. This is so that
users of the serial library do not need to link the system threads
libraries. If <code>--with-combined-threads</code> is specified, however,
then no separate threads library is created, and threads are included
in the main FFTW library. This is mainly useful under Windows, where
no system threads library is required and inter-library dependencies
are problematic.
</li><li> <span id="index-MPI-1"></span>
<code>--enable-mpi</code>: Enables compilation and installation of the FFTW
MPI library (see <a href="Distributed_002dmemory-FFTW-with-MPI.html">Distributed-memory FFTW with MPI</a>), which provides
parallel transforms for distributed-memory systems with MPI. (By
default, the MPI routines are not compiled.) See <a href="FFTW-MPI-Installation.html">FFTW MPI Installation</a>.
</li><li> <span id="index-Fortran_002dcallable-wrappers"></span>
<code>--disable-fortran</code>: Disables inclusion of legacy-Fortran
wrapper routines (see <a href="Calling-FFTW-from-Legacy-Fortran.html">Calling FFTW from Legacy Fortran</a>) in the standard
FFTW libraries. These wrapper routines increase the library size by
only a negligible amount, so they are included by default as long as
the <code>configure</code> script finds a Fortran compiler on your system.
(To specify a particular Fortran compiler <i>foo</i>, pass
<code>F77=</code><i>foo</i> to <code>configure</code>.)
</li><li> <code>--with-g77-wrappers</code>: By default, when Fortran wrappers are
included, the wrappers employ the linking conventions of the Fortran
compiler detected by the <code>configure</code> script. If this compiler is
GNU <code>g77</code>, however, then <em>two</em> versions of the wrappers are
included: one with <code>g77</code>&rsquo;s idiosyncratic convention of appending
two underscores to identifiers, and one with the more common
convention of appending only a single underscore. This way, the same
FFTW library will work with both <code>g77</code> and other Fortran
compilers, such as GNU <code>gfortran</code>. However, the converse is not
true: if you configure with a different compiler, then the
<code>g77</code>-compatible wrappers are not included. By specifying
<code>--with-g77-wrappers</code>, the <code>g77</code>-compatible wrappers are
included in addition to wrappers for whatever Fortran compiler
<code>configure</code> finds.
<span id="index-g77"></span>
</li><li> <code>--with-slow-timer</code>: Disables the use of hardware cycle counters,
and falls back on <code>gettimeofday</code> or <code>clock</code>. This greatly
worsens performance, and should generally not be used (unless you don&rsquo;t
have a cycle counter but still really want an optimized plan regardless
of the time). See <a href="Cycle-Counters.html">Cycle Counters</a>.
</li><li> <code>--enable-sse</code> (single precision),
<code>--enable-sse2</code> (single, double),
<code>--enable-avx</code> (single, double),
<code>--enable-avx2</code> (single, double),
<code>--enable-avx512</code> (single, double),
<code>--enable-avx-128-fma</code>,
<code>--enable-kcvi</code> (single),
<code>--enable-altivec</code> (single),
<code>--enable-vsx</code> (single, double),
<code>--enable-neon</code> (single, double on aarch64),
<code>--enable-generic-simd128</code>,
and
<code>--enable-generic-simd256</code>:
<p>Enable various SIMD instruction sets. You need compiler that supports
the given SIMD extensions, but FFTW will try to detect at runtime
whether the CPU supports these extensions. That is, you can compile
with<code>--enable-avx</code> and the code will still run on a CPU without AVX
support.
</p>
<ul class="no-bullet">
<li>- These options require a compiler supporting SIMD extensions, and
compiler support is always a bit flaky: see the FFTW FAQ for a list of
compiler versions that have problems compiling FFTW.
</li><li>- Because of the large variety of ARM processors and ABIs, FFTW
does not attempt to guess the correct <code>gcc</code> flags for generating
NEON code. In general, you will have to provide them on the command line.
This command line is known to have worked at least once:
<div class="example">
<pre class="example">./configure --with-slow-timer --host=arm-linux-gnueabi \
--enable-single --enable-neon \
&quot;CC=arm-linux-gnueabi-gcc -march=armv7-a -mfloat-abi=softfp&quot;
</pre></div>
</li></ul>
</li></ul>
<span id="index-compiler-2"></span>
<p>To force <code>configure</code> to use a particular C compiler <i>foo</i>
(instead of the default, usually <code>gcc</code>), pass <code>CC=</code><i>foo</i> to the
<code>configure</code> script; you may also need to set the flags via the variable
<code>CFLAGS</code> as described above.
<span id="index-compiler-flags-1"></span>
</p>
<hr>
<div class="header">
<p>
Next: <a href="Installation-on-non_002dUnix-systems.html" accesskey="n" rel="next">Installation on non-Unix systems</a>, Previous: <a href="Installation-and-Customization.html" accesskey="p" rel="prev">Installation and Customization</a>, Up: <a href="Installation-and-Customization.html" accesskey="u" rel="up">Installation and Customization</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
</body>
</html>

View File

@@ -0,0 +1,118 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html>
<!-- This manual is for FFTW
(version 3.3.10, 10 December 2020).
Copyright (C) 2003 Matteo Frigo.
Copyright (C) 2003 Massachusetts Institute of Technology.
Permission is granted to make and distribute verbatim copies of this
manual provided the copyright notice and this permission notice are
preserved on all copies.
Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided that the
entire resulting derived work is distributed under the terms of a
permission notice identical to this one.
Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions,
except that this permission notice may be stated in a translation
approved by the Free Software Foundation. -->
<!-- Created by GNU Texinfo 6.7, http://www.gnu.org/software/texinfo/ -->
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<title>Installation on non-Unix systems (FFTW 3.3.10)</title>
<meta name="description" content="Installation on non-Unix systems (FFTW 3.3.10)">
<meta name="keywords" content="Installation on non-Unix systems (FFTW 3.3.10)">
<meta name="resource-type" content="document">
<meta name="distribution" content="global">
<meta name="Generator" content="makeinfo">
<link href="index.html" rel="start" title="Top">
<link href="Concept-Index.html" rel="index" title="Concept Index">
<link href="index.html#SEC_Contents" rel="contents" title="Table of Contents">
<link href="Installation-and-Customization.html" rel="up" title="Installation and Customization">
<link href="Cycle-Counters.html" rel="next" title="Cycle Counters">
<link href="Installation-on-Unix.html" rel="prev" title="Installation on Unix">
<style type="text/css">
<!--
a.summary-letter {text-decoration: none}
blockquote.indentedblock {margin-right: 0em}
div.display {margin-left: 3.2em}
div.example {margin-left: 3.2em}
div.lisp {margin-left: 3.2em}
kbd {font-style: oblique}
pre.display {font-family: inherit}
pre.format {font-family: inherit}
pre.menu-comment {font-family: serif}
pre.menu-preformatted {font-family: serif}
span.nolinebreak {white-space: nowrap}
span.roman {font-family: initial; font-weight: normal}
span.sansserif {font-family: sans-serif; font-weight: normal}
ul.no-bullet {list-style: none}
-->
</style>
</head>
<body lang="en">
<span id="Installation-on-non_002dUnix-systems"></span><div class="header">
<p>
Next: <a href="Cycle-Counters.html" accesskey="n" rel="next">Cycle Counters</a>, Previous: <a href="Installation-on-Unix.html" accesskey="p" rel="prev">Installation on Unix</a>, Up: <a href="Installation-and-Customization.html" accesskey="u" rel="up">Installation and Customization</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
<hr>
<span id="Installation-on-non_002dUnix-systems-1"></span><h3 class="section">10.2 Installation on non-Unix systems</h3>
<p>It should be relatively straightforward to compile FFTW even on non-Unix
systems lacking the niceties of a <code>configure</code> script. Basically,
you need to edit the <code>config.h</code> header (copy it from
<code>config.h.in</code>) to <code>#define</code> the various options and compiler
characteristics, and then compile all the &lsquo;<samp>.c</samp>&rsquo; files in the
relevant directories.
</p>
<p>The <code>config.h</code> header contains about 100 options to set, each one
initially an <code>#undef</code>, each documented with a comment, and most of
them fairly obvious. For most of the options, you should simply
<code>#define</code> them to <code>1</code> if they are applicable, although a few
options require a particular value (e.g. <code>SIZEOF_LONG_LONG</code> should
be defined to the size of the <code>long long</code> type, in bytes, or zero
if it is not supported). We will likely post some sample
<code>config.h</code> files for various operating systems and compilers for
you to use (at least as a starting point). Please let us know if you
have to hand-create a configuration file (and/or a pre-compiled binary)
that you want to share.
</p>
<p>To create the FFTW library, you will then need to compile all of the
&lsquo;<samp>.c</samp>&rsquo; files in the <code>kernel</code>, <code>dft</code>, <code>dft/scalar</code>,
<code>dft/scalar/codelets</code>, <code>rdft</code>, <code>rdft/scalar</code>,
<code>rdft/scalar/r2cf</code>, <code>rdft/scalar/r2cb</code>,
<code>rdft/scalar/r2r</code>, <code>reodft</code>, and <code>api</code> directories.
If you are compiling with SIMD support (e.g. you defined
<code>HAVE_SSE2</code> in <code>config.h</code>), then you also need to compile
the <code>.c</code> files in the <code>simd-support</code>,
<code>{dft,rdft}/simd</code>, <code>{dft,rdft}/simd/*</code> directories.
</p>
<p>Once these files are all compiled, link them into a library, or a shared
library, or directly into your program.
</p>
<p>To compile the FFTW test program, additionally compile the code in the
<code>libbench2/</code> directory, and link it into a library. Then compile
the code in the <code>tests/</code> directory and link it to the
<code>libbench2</code> and FFTW libraries. To compile the <code>fftw-wisdom</code>
(command-line) tool (see <a href="Wisdom-Utilities.html">Wisdom Utilities</a>), compile
<code>tools/fftw-wisdom.c</code> and link it to the <code>libbench2</code> and FFTW
libraries
</p>
<hr>
<div class="header">
<p>
Next: <a href="Cycle-Counters.html" accesskey="n" rel="next">Cycle Counters</a>, Previous: <a href="Installation-on-Unix.html" accesskey="p" rel="prev">Installation on Unix</a>, Up: <a href="Installation-and-Customization.html" accesskey="u" rel="up">Installation and Customization</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
</body>
</html>

View File

@@ -0,0 +1,95 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html>
<!-- This manual is for FFTW
(version 3.3.10, 10 December 2020).
Copyright (C) 2003 Matteo Frigo.
Copyright (C) 2003 Massachusetts Institute of Technology.
Permission is granted to make and distribute verbatim copies of this
manual provided the copyright notice and this permission notice are
preserved on all copies.
Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided that the
entire resulting derived work is distributed under the terms of a
permission notice identical to this one.
Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions,
except that this permission notice may be stated in a translation
approved by the Free Software Foundation. -->
<!-- Created by GNU Texinfo 6.7, http://www.gnu.org/software/texinfo/ -->
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<title>Interleaved and split arrays (FFTW 3.3.10)</title>
<meta name="description" content="Interleaved and split arrays (FFTW 3.3.10)">
<meta name="keywords" content="Interleaved and split arrays (FFTW 3.3.10)">
<meta name="resource-type" content="document">
<meta name="distribution" content="global">
<meta name="Generator" content="makeinfo">
<link href="index.html" rel="start" title="Top">
<link href="Concept-Index.html" rel="index" title="Concept Index">
<link href="index.html#SEC_Contents" rel="contents" title="Table of Contents">
<link href="Guru-Interface.html" rel="up" title="Guru Interface">
<link href="Guru-vector-and-transform-sizes.html" rel="next" title="Guru vector and transform sizes">
<link href="Guru-Interface.html" rel="prev" title="Guru Interface">
<style type="text/css">
<!--
a.summary-letter {text-decoration: none}
blockquote.indentedblock {margin-right: 0em}
div.display {margin-left: 3.2em}
div.example {margin-left: 3.2em}
div.lisp {margin-left: 3.2em}
kbd {font-style: oblique}
pre.display {font-family: inherit}
pre.format {font-family: inherit}
pre.menu-comment {font-family: serif}
pre.menu-preformatted {font-family: serif}
span.nolinebreak {white-space: nowrap}
span.roman {font-family: initial; font-weight: normal}
span.sansserif {font-family: sans-serif; font-weight: normal}
ul.no-bullet {list-style: none}
-->
</style>
</head>
<body lang="en">
<span id="Interleaved-and-split-arrays"></span><div class="header">
<p>
Next: <a href="Guru-vector-and-transform-sizes.html" accesskey="n" rel="next">Guru vector and transform sizes</a>, Previous: <a href="Guru-Interface.html" accesskey="p" rel="prev">Guru Interface</a>, Up: <a href="Guru-Interface.html" accesskey="u" rel="up">Guru Interface</a> &nbsp; [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html" title="Index" rel="index">Index</a>]</p>
</div>
<hr>
<span id="Interleaved-and-split-arrays-1"></span><h4 class="subsection">4.5.1 Interleaved and split arrays</h4>
<p>The guru interface supports two representations of complex numbers,
which we call the interleaved and the split format.
</p>
<p>The <em>interleaved</em> format is the same one used by the basic and
advanced interfaces, and it is documented in <a href="Complex-numbers.html">Complex numbers</a>.
In the interleaved format, you provide pointers to the real part of a
complex number, and the imaginary part understood to be stored in the
next memory location.
<span id="index-interleaved-format"></span>
</p>
<p>The <em>split</em> format allows separate pointers to the real and
imaginary parts of a complex array.
<span id="index-split-format"></span>
</p>
<p>Technically, the interleaved format is redundant, because you can
always express an interleaved array in terms of a split array with
appropriate pointers and strides. On the other hand, the interleaved
format is simpler to use, and it is common in practice. Hence, FFTW
supports it as a special case.
</p>
</body>
</html>

Some files were not shown because too many files have changed in this diff Show More