%bcond_without mpich

%ifarch s390 s390x

%bcond_with openmpi

%else

%bcond_without openmpi

%endif

%if %{with mpich}

%global mpi_list %{?mpi_list} mpich

%endif

%if %{with openmpi}

%global mpi_list %{?mpi_list} openmpi

%endif

Name:           fftw

Version:        3.3.8

Release:        12%{?dist}

Summary:        A Fast Fourier Transform library

License:        GPLv2+

URL:            http://www.fftw.org

Source0:        http://www.fftw.org/fftw-%{version}.tar.gz

BuildRequires:  gcc-gfortran

BuildRequires:  autoconf

BuildRequires:  automake

BuildRequires:  libtool

%global quad 0

# Quad precision support only available with gcc >= 4.6 (Fedora >= 15)

# and only on these arches

%ifarch %{ix86} x86_64 ia64

%global quad 1

%endif

# For check phase

BuildRequires:  time

BuildRequires:  perl-interpreter

%if %{with mpich}

BuildRequires:  mpich-devel

BuildRequires:  nss-myhostname

%endif

%if %{with openmpi}

BuildRequires:  openmpi-devel

%endif

%if %{with mpich} || %{with openmpi}

BuildRequires:  environment-modules

%endif

BuildRequires: make

%description

FFTW is a C subroutine library for computing the Discrete Fourier

Transform (DFT) in one or more dimensions, of both real and complex

data, and of arbitrary input size.

%package libs

Summary:        FFTW run-time library

Provides:       fftw3 = %{version}-%{release}

# Libs rearranged in 3.3.1-2

Obsoletes:      fftw-libs-threads < %{version}-%{release}

Obsoletes:      fftw-libs-openmp < %{version}-%{release}

# Pull in the actual libraries

Requires:       %{name}-libs-single%{?_isa} = %{version}-%{release}

Requires:       %{name}-libs-double%{?_isa} = %{version}-%{release}

Requires:       %{name}-libs-long%{?_isa} = %{version}-%{release}

%if %{quad}

Requires:       %{name}-libs-quad%{?_isa} = %{version}-%{release}

%endif

%description libs

This is a dummy package package, pulling in the individual FFTW

run-time libraries.

%package devel

Summary:        Headers, libraries and docs for the FFTW library

Requires:       pkgconfig

Requires:       %{name}%{?_isa} = %{version}-%{release}

Requires:       %{name}-libs%{?_isa} = %{version}-%{release}

Provides:       fftw3-devel%{?_isa} = %{version}-%{release}

Provides:       fftw3-devel = %{version}-%{release}

%description devel

FFTW is a C subroutine library for computing the Discrete Fourier

Transform (DFT) in one or more dimensions, of both real and complex

data, and of arbitrary input size.

This package contains header files and development libraries needed to

develop programs using the FFTW fast Fourier transform library.

%package libs-double

Summary:        FFTW library, double precision

%description libs-double

This package contains the FFTW library compiled in double precision.

%package libs-single

Summary:        FFTW library, single precision

%description libs-single

This package contains the FFTW library compiled in single precision.

%package libs-long

Summary:        FFTW library, long double precision 

%description libs-long

This package contains the FFTW library compiled in long double

precision.

%if %{quad}

%package libs-quad

Summary:        FFTW library, quadruple

%description libs-quad

This package contains the FFTW library compiled in quadruple

precision.

%endif

%package        static

Summary:        Static versions of the FFTW libraries

Requires:       %{name}-devel%{?_isa} = %{version}-%{release}

Provides:       fftw3-static%{?_isa} = %{version}-%{release}

Provides:       fftw3-static = %{version}-%{release}

%description static

The fftw-static package contains the statically linkable version of

the FFTW fast Fourier transform library.

%if %{with mpich}

%package mpich-libs

Summary:        FFTW MPICH run-time library

Provides:       fftw3-mpich = %{version}-%{release}

# Pull in the actual libraries

Requires:       %{name}-mpich-libs-single%{?_isa} = %{version}-%{release}

Requires:       %{name}-mpich-libs-double%{?_isa} = %{version}-%{release}

Requires:       %{name}-mpich-libs-long%{?_isa} = %{version}-%{release}

%description mpich-libs

This is a dummy package package, pulling in the individual FFTW

MPICH run-time libraries.

%package mpich-devel

Summary:        Headers, libraries and docs for the FFTW MPICH library

Requires:       mpich-devel

Requires:       pkgconfig

Requires:       %{name}-devel%{?_isa} = %{version}-%{release}

Requires:       %{name}-mpich-libs%{?_isa} = %{version}-%{release}

Provides:       fftw3-mpich-devel%{?_isa} = %{version}-%{release}

Provides:       fftw3-mpich-devel = %{version}-%{release}

%description mpich-devel

FFTW is a C subroutine library for computing the Discrete Fourier

Transform (DFT) in one or more dimensions, of both real and complex

data, and of arbitrary input size.

This package contains header files and development libraries needed to

develop programs using the FFTW fast Fourier transform library for MPICH.

%package mpich-libs-double

Summary:        FFTW MPICH library, double precision

Requires:       %{name}-libs-double%{?_isa} = %{version}-%{release}

%description mpich-libs-double

This package contains the FFTW MPICH library compiled in double precision.

%package mpich-libs-single

Summary:        FFTW MPICH library, single precision

Requires:       %{name}-libs-single%{?_isa} = %{version}-%{release}

%description mpich-libs-single

This package contains the FFTW MPICH library compiled in single precision.

%package mpich-libs-long

Summary:        FFTW MPICH library, long double precision 

Requires:       %{name}-libs-long%{?_isa} = %{version}-%{release}

%description mpich-libs-long

This package contains the FFTW MPICH library compiled in long double

precision.

%package        mpich-static

Summary:        Static versions of the FFTW MPICH libraries

Requires:       %{name}-mpich-devel%{?_isa} = %{version}-%{release}

Requires:       %{name}-static%{?_isa} = %{version}-%{release}

Provides:       fftw3-mpich-static%{?_isa} = %{version}-%{release}

Provides:       fftw3-mpich-static = %{version}-%{release}

%description mpich-static

The fftw-mpich-static package contains the statically linkable version of

the FFTW fast Fourier transform library for MPICh.

%endif

%if %{with openmpi}

%package openmpi-libs

Summary:        FFTW OpenMPI run-time library

Provides:       fftw3-openmpi = %{version}-%{release}

# Pull in the actual libraries

Requires:       %{name}-openmpi-libs-single%{?_isa} = %{version}-%{release}

Requires:       %{name}-openmpi-libs-double%{?_isa} = %{version}-%{release}

Requires:       %{name}-openmpi-libs-long%{?_isa} = %{version}-%{release}

%description openmpi-libs

This is a dummy package package, pulling in the individual FFTW

OpenMPI run-time libraries.

%package openmpi-devel

Summary:        Headers, libraries and docs for the FFTW OpenMPI library

Requires:       openmpi-devel

Requires:       pkgconfig

Requires:       %{name}-devel%{?_isa} = %{version}-%{release}

Requires:       %{name}-openmpi-libs%{?_isa} = %{version}-%{release}

Provides:       fftw3-openmpi-devel%{?_isa} = %{version}-%{release}

Provides:       fftw3-openmpi-devel = %{version}-%{release}

%description openmpi-devel

FFTW is a C subroutine library for computing the Discrete Fourier

Transform (DFT) in one or more dimensions, of both real and complex

data, and of arbitrary input size.

This package contains header files and development libraries needed to

develop programs using the FFTW fast Fourier transform library for OpenMPI.

%package openmpi-libs-double

Summary:        FFTW OpenMPI library, double precision

Requires:       %{name}-libs-double%{?_isa} = %{version}-%{release}

%description openmpi-libs-double

This package contains the FFTW OpenMPI library compiled in double precision.

%package openmpi-libs-single

Summary:        FFTW OpenMPI library, single precision

Requires:       %{name}-libs-single%{?_isa} = %{version}-%{release}

%description openmpi-libs-single

This package contains the FFTW OpenMPI library compiled in single precision.

%package openmpi-libs-long

Summary:        FFTW OpenMPI library, long double precision 

Requires:       %{name}-libs-long%{?_isa} = %{version}-%{release}

%description openmpi-libs-long

This package contains the FFTW OpenMPI library compiled in long double

precision.

%package        openmpi-static

Summary:        Static versions of the FFTW OpenMPI libraries

Requires:       %{name}-openmpi-devel%{?_isa} = %{version}-%{release}

Requires:       %{name}-static%{?_isa} = %{version}-%{release}

Provides:       fftw3-openmpi-static%{?_isa} = %{version}-%{release}

Provides:       fftw3-openmpi-static = %{version}-%{release}

%description openmpi-static

The fftw-openmpi-static package contains the statically linkable version of

the FFTW fast Fourier transform library for MPICh.

%endif

%package doc

Summary:        FFTW library manual

BuildArch:      noarch

%description doc

This package contains the manual for the FFTW fast Fourier transform

library.

%prep

%setup -q

%build

# Explicitly load shell support for the environment-modules package, used

# below via 'module' pseudo-command.

source /etc/profile.d/modules.sh

# Regenerate autoconf files using current tools so proper build flags

# from redhat-rpm-config are used

autoreconf -vfi

# Configure uses g77 by default, if present on system

export F77=gfortran

BASEFLAGS="--enable-shared --disable-dependency-tracking --enable-threads"

BASEFLAGS+=" --enable-openmp"

# Precisions to build

prec_name[0]=single

prec_name[1]=double

prec_name[2]=long

prec_name[3]=quad

# Corresponding flags

prec_flags[0]=--enable-single

prec_flags[1]=--enable-double

prec_flags[2]=--enable-long-double

prec_flags[3]=--enable-quad-precision

%ifarch x86_64

# Enable SSE2 and AVX support for x86_64

for((i=0;i<2;i++)); do

 prec_flags[i]+=" --enable-sse2 --enable-avx"

done

%endif

# No NEON run time detection, not all ARM SoCs have NEON

#%ifarch %{arm}

## Compile support for NEON instructions

#for((i=0;i<2;i++)); do

# prec_flags[i]+=" --enable-neon"

#done

#%endif

#%ifarch ppc ppc64

## Compile support for Altivec instructions

#for((i=0;i<2;i++)); do

 #prec_flags[i]+=" --enable-altivec"

#done

#%endif

# Loop over precisions

%if %{quad}

for((iprec=0;iprec<4;iprec++))

%else

for((iprec=0;iprec<3;iprec++))

%endif

do

 mkdir ${prec_name[iprec]}${ver_name[iver]}

 cd ${prec_name[iprec]}${ver_name[iver]}

 ln -s ../configure .

 %{configure} ${BASEFLAGS} ${prec_flags[iprec]}

 sed -i 's|^hardcode_libdir_flag_spec=.*|hardcode_libdir_flag_spec=""|g' libtool

 sed -i 's|^runpath_var=LD_RUN_PATH|runpath_var=DIE_RPATH_DIE|g' libtool

 make %{?_smp_mflags}

 cd ..

done

# MPI Builds - this duplicates the non-mpi builds, but oh well

for mpi in %{mpi_list}

do

 module load mpi/${mpi}-%{_arch}

 # Loop over precisions - no quad precision support with MPI

 for((iprec=0;iprec<3;iprec++))

 do

  mkdir ${mpi}-${prec_name[iprec]}${ver_name[iver]}

  cd ${mpi}-${prec_name[iprec]}${ver_name[iver]}

  ln -s ../configure .

  # Force linking the _mpi.so libraries with the mpi libs.  This works because

  # we get rid of all of the non-mpi components of these builds

  export CC=mpicc

  %{configure} ${BASEFLAGS} ${prec_flags[iprec]} --enable-mpi \

   --libdir=%{_libdir}/$mpi/lib \

   --bindir=%{_libdir}/$mpi/bin \

   --sbindir=%{_libdir}/$mpi/sbin \

   --includedir=%{_includedir}/$mpi-%{_arch} \

   --mandir=%{_libdir}/$mpi/share/man

  make %{?_smp_mflags}

  cd ..

 done

 module unload mpi/${mpi}-%{_arch}

done

%install

# Explicitly load shell support for the environment-modules package, used

# below via 'module' pseudo-command.

source /etc/profile.d/modules.sh

%if %{quad}

for ver in single double long quad

%else

for ver in single double long

%endif

do

 make -C $ver install DESTDIR=%{buildroot}

done

# MPI

for mpi in %{mpi_list}

do

 module load mpi/${mpi}-%{_arch}

 for ver in single double long

 do

  make -C ${mpi}-${ver} install DESTDIR=%{buildroot}

  # Remove duplicated non-mpi libraries, binaries, and data

  find %{buildroot}%{_libdir}/${mpi}/lib -name libfftw\* -a \! -name \*_mpi.\* -delete

  rm -r %{buildroot}%{_libdir}/${mpi}/{bin,share}

 done

 module unload mpi/${mpi}-%{_arch}

done

rm -f %{buildroot}%{_infodir}/dir

find %{buildroot} -name \*.la -delete

%check

# Explicitly load shell support for the environment-modules package, used

# below via 'module' pseudo-command.

source /etc/profile.d/modules.sh

bdir=`pwd`

%if %{quad}

for ver in single double long quad

%else

for ver in single double long

%endif

do 

 export LD_LIBRARY_PATH=$bdir/$ver/.libs:$bdir/$ver/threads/.libs

 make %{?_smp_mflags} -C $ver check

done

# MPI

%if %{with openmpi}

%ifarch %{ix86}

# disable Open MPI's vader byte transfer layer while running tests on 32-bit x86 platforms

# as it is known to be troublesome <https://github.com/open-mpi/ompi/issues/4260>

export OMPI_MCA_btl="^vader"

%endif

%endif

for mpi in %{mpi_list}

do

 module load mpi/${mpi}-%{_arch}

 for ver in single double long

 do 

  export LD_LIBRARY_PATH=$bdir/$ver/.libs:$bdir/$ver/threads/.libs

  make %{?_smp_mflags} -C ${mpi}-${ver}/mpi check

 done

 module unload mpi/${mpi}-%{_arch}

done

%ldconfig_scriptlets libs-single

%ldconfig_scriptlets libs-double

%ldconfig_scriptlets libs-long

%if %{quad}

%ldconfig_scriptlets libs-quad

%endif

%files

%{_mandir}/man1/fftw*.1*

%{_bindir}/fftw*-wisdom*

%files libs

%files libs-single

%license COPYING COPYRIGHT

%doc AUTHORS ChangeLog NEWS README* TODO

%{_libdir}/libfftw3f.so.*

%{_libdir}/libfftw3f_threads.so.*

%{_libdir}/libfftw3f_omp.so.*

%files libs-double

%license COPYING COPYRIGHT

%doc AUTHORS ChangeLog NEWS README* TODO

%{_libdir}/libfftw3.so.*

%{_libdir}/libfftw3_threads.so.*

%{_libdir}/libfftw3_omp.so.*

%files libs-long

%license COPYING COPYRIGHT

%doc AUTHORS ChangeLog NEWS README* TODO

%{_libdir}/libfftw3l.so.*

%{_libdir}/libfftw3l_threads.so.*

%{_libdir}/libfftw3l_omp.so.*

%if %{quad}

%files libs-quad

%license COPYING COPYRIGHT

%doc AUTHORS ChangeLog NEWS README* TODO

%{_libdir}/libfftw3q.so.*

%{_libdir}/libfftw3q_threads.so.*

%{_libdir}/libfftw3q_omp.so.*

%endif

%files devel

%doc doc/FAQ/fftw-faq.html/

%doc %{_infodir}/fftw3.info*

%{_includedir}/fftw3*

%dir %{_libdir}/cmake/fftw3/

%{_libdir}/cmake/fftw3/*.cmake

%{_libdir}/pkgconfig/fftw3*.pc

%{_libdir}/libfftw3*.so

%files static

%{_libdir}/libfftw3*.a

%files doc

%doc doc/*.pdf doc/html/

%if %{with mpich}

%files mpich-libs

%files mpich-libs-single

%license COPYING COPYRIGHT

%doc AUTHORS ChangeLog NEWS README* TODO

%{_libdir}/mpich/lib/libfftw3f_mpi.so.*

%files mpich-libs-double

%license COPYING COPYRIGHT

%doc AUTHORS ChangeLog NEWS README* TODO

%{_libdir}/mpich/lib/libfftw3_mpi.so.*

%files mpich-libs-long

%license COPYING COPYRIGHT

%doc AUTHORS ChangeLog NEWS README* TODO

%{_libdir}/mpich/lib/libfftw3l_mpi.so.*

%files mpich-devel

%doc doc/FAQ/fftw-faq.html/

%{_includedir}/mpich-%{_arch}

%dir %{_libdir}/mpich/lib/cmake/fftw3/

%{_libdir}/mpich/lib/cmake/fftw3/*.cmake

%{_libdir}/mpich/lib/pkgconfig/fftw3*.pc

%{_libdir}/mpich/lib/libfftw3*.so

%files mpich-static

%{_libdir}/mpich/lib/libfftw3*.a

%endif

%if %{with openmpi}

%files openmpi-libs

%files openmpi-libs-single

%license COPYING COPYRIGHT

%doc AUTHORS ChangeLog NEWS README* TODO

%{_libdir}/openmpi/lib/libfftw3f_mpi.so.*

%files openmpi-libs-double

%license COPYING COPYRIGHT

%doc AUTHORS ChangeLog NEWS README* TODO

%{_libdir}/openmpi/lib/libfftw3_mpi.so.*

%files openmpi-libs-long

%license COPYING COPYRIGHT

%doc AUTHORS ChangeLog NEWS README* TODO

%{_libdir}/openmpi/lib/libfftw3l_mpi.so.*

%files openmpi-devel

%doc doc/FAQ/fftw-faq.html/

%{_includedir}/openmpi-%{_arch}

%dir %{_libdir}/openmpi/lib/cmake/fftw3/

%{_libdir}/openmpi/lib/cmake/fftw3/*.cmake

%{_libdir}/openmpi/lib/pkgconfig/fftw3*.pc

%{_libdir}/openmpi/lib/libfftw3*.so

%files openmpi-static

%{_libdir}/openmpi/lib/libfftw3*.a

%endif

%changelog

* Mon Aug 09 2021 Mohan Boddu <mboddu@redhat.com> - 3.3.8-12

- Rebuilt for IMA sigs, glibc 2.34, aarch64 flags

  Related: rhbz#1991688

* Thu Apr 15 2021 Mohan Boddu <mboddu@redhat.com> - 3.3.8-11

- Rebuilt for RHEL 9 BETA on Apr 15th 2021. Related: rhbz#1947937

* Tue Jan 26 2021 Fedora Release Engineering <releng@fedoraproject.org> - 3.3.8-10

- Rebuilt for https://fedoraproject.org/wiki/Fedora_34_Mass_Rebuild

* Sat Aug 01 2020 Fedora Release Engineering <releng@fedoraproject.org> - 3.3.8-9

- Second attempt - Rebuilt for

  https://fedoraproject.org/wiki/Fedora_33_Mass_Rebuild

* Mon Jul 27 2020 Fedora Release Engineering <releng@fedoraproject.org> - 3.3.8-8

- Rebuilt for https://fedoraproject.org/wiki/Fedora_33_Mass_Rebuild

* Tue Jan 28 2020 Fedora Release Engineering <releng@fedoraproject.org> - 3.3.8-7

- Rebuilt for https://fedoraproject.org/wiki/Fedora_32_Mass_Rebuild

* Thu Jul 25 2019 Fedora Release Engineering <releng@fedoraproject.org> - 3.3.8-6

- Rebuilt for https://fedoraproject.org/wiki/Fedora_31_Mass_Rebuild

* Thu Mar  7 2019 Tim Landscheidt <tim@tim-landscheidt.de> - 3.3.8-5

- Remove obsolete requirements for %%post/%%preun scriptlets

* Thu Feb 14 2019 Orion Poplawski <orion@nwra.com> - 3.3.8-4

- Rebuild for openmpi 3.1.3

* Thu Jan 31 2019 Fedora Release Engineering <releng@fedoraproject.org> - 3.3.8-3

- Rebuilt for https://fedoraproject.org/wiki/Fedora_30_Mass_Rebuild

* Fri Jul 13 2018 Fedora Release Engineering <releng@fedoraproject.org> - 3.3.8-2

- Rebuilt for https://fedoraproject.org/wiki/Fedora_29_Mass_Rebuild

* Mon Jun 11 2018 Conrad Meyer <cemeyer@uw.edu> - 3.3.8-1

- Update to latest upstream, 3.3.8 (rhbz# 1413425)

- Add missing BuildRequires on environment-modules, used in 43b91c1c ("Build

  mpich and openmpi versions") without explicit BR dependency

- Add new cmake files to %%files

* Wed Apr 18 2018 Merlin Mathesius <mmathesi@redhat.com> - 3.3.5-11

- Regenerate autoconf files using current tools so proper build flags

  from redhat-rpm-config are used. This resolves BZ#1548473.

* Wed Feb 07 2018 Fedora Release Engineering <releng@fedoraproject.org> - 3.3.5-10

- Rebuilt for https://fedoraproject.org/wiki/Fedora_28_Mass_Rebuild

* Sat Feb 03 2018 Igor Gnatenko <ignatenkobrain@fedoraproject.org> - 3.3.5-9

- Switch to %%ldconfig_scriptlets

* Tue Nov 28 2017 Merlin Mathesius <mmathesi@redhat.com> - 3.3.5-8

- Disable Open MPI's vader byte transfer layer while running tests on 32-bit x86 platforms

  as it is known to be troublesome <https://github.com/open-mpi/ompi/issues/4260>.

  This resolves FTBFS issue (BZ#1518038).

* Wed Aug 02 2017 Fedora Release Engineering <releng@fedoraproject.org> - 3.3.5-7

- Rebuilt for https://fedoraproject.org/wiki/Fedora_27_Binutils_Mass_Rebuild

* Wed Jul 26 2017 Fedora Release Engineering <releng@fedoraproject.org> - 3.3.5-6

- Rebuilt for https://fedoraproject.org/wiki/Fedora_27_Mass_Rebuild

* Fri Feb 10 2017 Fedora Release Engineering <releng@fedoraproject.org> - 3.3.5-5

- Rebuilt for https://fedoraproject.org/wiki/Fedora_26_Mass_Rebuild

* Fri Oct 21 2016 Orion Poplawski <orion@cora.nwra.com> - 3.3.5-4

- Rebuild for openmpi 2.0

* Mon Oct 10 2016 Dan Horák <dan[at]danny.cz> - 3.3.5-3

- Update BRs so nss-myhostname (provided by systemd) is included also

  in mpich-only builds (Related: #1383271)

* Fri Oct 7 2016 Orion Poplawski <orion@cora.nwra.com> - 3.3.5-2

- Build mpich and openmpi versions

* Wed Oct 5 2016 Orion Poplawski <orion@cora.nwra.com> - 3.3.5-1

- Update to 3.3.5

- Cleanup spec

* Wed Feb 03 2016 Fedora Release Engineering <releng@fedoraproject.org> - 3.3.4-7

- Rebuilt for https://fedoraproject.org/wiki/Fedora_24_Mass_Rebuild

* Wed Jun 17 2015 Fedora Release Engineering <rel-eng@lists.fedoraproject.org> - 3.3.4-6

- Rebuilt for https://fedoraproject.org/wiki/Fedora_23_Mass_Rebuild

* Sat Aug 16 2014 Fedora Release Engineering <rel-eng@lists.fedoraproject.org> - 3.3.4-5

- Rebuilt for https://fedoraproject.org/wiki/Fedora_21_22_Mass_Rebuild

* Fri Jul  4 2014 Peter Robinson <pbrobinson@fedoraproject.org> 3.3.4-4

- Disable SSE2/AVX on x86(32) as we support Pentium Pro as base

- SSE2 only arrived with P-4

- https://fedoraproject.org/wiki/Features/F12X86Support

- Fix ARM macro

- Disable NEON on ARM (we don't enable by default, needs runtime detection)

* Thu Jul 03 2014 Conrad Meyer <cemeyer@uw.edu> - 3.3.4-3

- Build with --enable-avx (rhbz# 1114964)

* Sat Jun 07 2014 Fedora Release Engineering <rel-eng@lists.fedoraproject.org> - 3.3.4-2

- Rebuilt for https://fedoraproject.org/wiki/Fedora_21_Mass_Rebuild

* Tue Mar 18 2014 Susi Lehtola <jussilehtola@fedoraproject.org> - 3.3.4-1

- Update to 3.3.4.

* Sat Dec 14 2013 Conrad Meyer <cemeyer@uw.edu> - 3.3.3-8

- Remove non-OpenMP / g77 build for obsolete RHEL (< 5)

- Remove ancient Obsoletes (fftw < 3.3, Fedora < 16)

- Remove ancient non-Quad build (Fedora < 15) on supported arch(s)

- Remove obsolete conditional on BuildArch: noarch (RHEL < 6, Fedora < 13)

* Sat Aug 03 2013 Fedora Release Engineering <rel-eng@lists.fedoraproject.org> - 3.3.3-7

- Rebuilt for https://fedoraproject.org/wiki/Fedora_20_Mass_Rebuild

* Wed Feb 13 2013 Fedora Release Engineering <rel-eng@lists.fedoraproject.org> - 3.3.3-6

- Rebuilt for https://fedoraproject.org/wiki/Fedora_19_Mass_Rebuild

* Sat Jan 19 2013 PPC Secondary Arch Admin <karsten@redhat.com> 3.3.3-5

- disable altivec flag again, that works with single precision only

* Mon Dec 24 2012 Susi Lehtola <jussilehtola@fedoraproject.org> - 3.3.3-4

- Fix typo: should be %%ix86 instead of %%x86. Now should have SSE2 support

  on x86 as well.

* Tue Dec 18 2012 Susi Lehtola <jussilehtola@fedoraproject.org> - 3.3.3-3

- After consultation with upstream, enable SSE2 also on x86, altivec on ppc

  and ppc64 and NEON on arm.

* Tue Dec 18 2012 Susi Lehtola <jussilehtola@fedoraproject.org> - 3.3.3-2

- Enable SSE2 on x86_64.

* Mon Nov 26 2012 Jussi Lehtola <jussilehtola@fedoraproject.org> - 3.3.3-1

- Update to 3.3.3.

* Thu Jul 19 2012 Fedora Release Engineering <rel-eng@lists.fedoraproject.org> - 3.3.2-2

- Rebuilt for https://fedoraproject.org/wiki/Fedora_18_Mass_Rebuild

* Fri Jun 15 2012 Orion Poplawski <orion@cora.nwra.com> - 3.3.2-1

- Update to 3.3.2

- Drop alignment patch

* Fri Apr 27 2012 Jussi Lehtola <jussilehtola@fedoraproject.org> - 3.3.1-3

- Fix FTBFS with gcc 4.7.

* Thu Apr 26 2012 Jussi Lehtola <jussilehtola@fedoraproject.org> - 3.3.1-2

- Reorganized libraries (BZ #812981).

* Mon Feb 27 2012 Jussi Lehtola <jussilehtola@fedoraproject.org> - 3.3.1-1

- Update to 3.3.1.

* Fri Jan 13 2012 Fedora Release Engineering <rel-eng@lists.fedoraproject.org> - 3.3-5

- Rebuilt for https://fedoraproject.org/wiki/Fedora_17_Mass_Rebuild

* Tue Oct 11 2011 Dan Horák <dan[at]danny.cz> - 3.3-4

- libquadmath exists only on x86/x86_64 and ia64

* Mon Oct 10 2011 Rex Dieter <rdieter@fedoraproject.org> 3.3-3

- -devel: Provides: fftw3-devel (#744758)

- -static: Provides: fftw3-static

- drop %%_isa from Obsoletes

* Sat Jul 30 2011 Jussi Lehtola <jussilehtola@fedoraproject.org> - 3.3-2

- Conditionalize OpenMP and quadruple precision support based on capabilities

  of system compiler.

* Thu Jul 28 2011 Jussi Lehtola <jussilehtola@fedoraproject.org> - 3.3-1

- Update to 3.3.

* Tue Feb 08 2011 Fedora Release Engineering <rel-eng@lists.fedoraproject.org> - 3.2.2-5

- Rebuilt for https://fedoraproject.org/wiki/Fedora_15_Mass_Rebuild

* Sat Jan 9 2010 Jussi Lehtola <jussilehtola@fedoraproject.org> - 3.2.2-4

- Get rid of rpath.

* Sat Jan 9 2010 Jussi Lehtola <jussilehtola@fedoraproject.org> - 3.2.2-3