[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Building for Microsoft Windows using MSYS
|
From: |
Tatsuro MATSUOKA |
|
Subject: |
Re: Building for Microsoft Windows using MSYS |
|
Date: |
Sun, 14 Feb 2010 11:09:05 +0900 (JST) |
Hello
I have tried compile with the msys bundled with octave-3.2.3.
>From octave prompt, I execute sh shell
octave:1>system('sh')
sh-3.1$
cd "C:\usr\tatsu\mingwhome\octaves\OctBuild\octave-3.2.4B"
#FLAG setting
export CC='mingw32-gcc-4.4.0-dw2 -shared-libgcc'
export CXX='mingw32-g++-4.4.0-dw2 -shared-libgcc'
export F77='mingw32-gfortran-4.4.0-dw2 -shared-libgcc'
#if you have installed gcc-4.4.0 you can use
# export CC='gcc -shared-libgcc'
# instead the above
#export CXXFLAGS='-D_GLIBCXX_DLL'
export LDFLAGS='-L/c/Programs/Octave/3.2.3_gcc4.4.0/lib
-Wl,--allow-multiple-definition'
export CPPFLAGS='-I/c/Programs/Octave/3.2.3_gcc4.4.0/include'
export CXXLIBS='-lstdc++_s'
../../OctaveSource/octave-3.2.4/configure
--prefix=/c/Programs/Octave/3.2.4_gcc4.4.0
Octave is now configured for i686-pc-mingw32
Source directory: ../../OctaveSource/octave-3.2.4
Installation prefix: /c/Programs/Octave/3.2.4_gcc4.4.0
C compiler: mingw32-gcc-4.4.0-dw2 -shared-libgcc -mieee-fp -Wall
-W -Wshadow -Wformat -g
-O2
C++ compiler: mingw32-g++-4.4.0-dw2 -shared-libgcc -mieee-fp -Wall
-W -Wshadow
-Wold-style-cast -Wformat -g -O2
Fortran compiler: mingw32-gfortran-4.4.0-dw2 -shared-libgcc -O -mieee-fp
Fortran libraries: -Lc:/Programs/Octave/3.2.3_gcc4.4.0/lib
-Lc:/programs/octave/3.2.3_gcc4.4.0/mingw32/bin/../lib/gcc/mingw32/4.4.0
-Lc:/programs/octave/3.2.3_gcc4.4.0/mingw32/bin/../lib/gcc
-Lc:/programs/octave/3.2.3_gcc4.4.0/mingw32/bin/../lib/gcc/mingw32/4.4.0/../../../../mingw32/lib
-Lc:/programs/octave/3.2.3_gcc4.4.0/mingw32/bin/../lib/gcc/mingw32/4.4.0/../../..
-lhdf5 -lz -lm
-lgfortranbegin -lgfortran -lmingw32 -lmoldname -lmingwex -lmsvcrt -luser32
-lkernel32 -ladvapi32
-lshell32
BLAS libraries: -llapack -lblas
FFTW libraries: -lfftw3 -lfftw3f
GLPK libraries: -lglpk
UMFPACK libraries: -lumfpack
AMD libraries: -lamd
CAMD libraries: -lcamd
COLAMD libraries: -lcolamd
CCOLAMD libraries: -lccolamd
CHOLMOD libraries: -lcholmod
CXSPARSE libraries: -lcxsparse
ARPACK libraries: -larpack
QRUPDATE libraries: -lqrupdate
HDF5 libraries: -lhdf5
CURL libraries: -lcurl
REGEX libraries: -lregex -lpcre
QHULL libraries: -lqhull
OPENGL libraries: -lftgl -lfreetype -lz -lopengl32 -lglu32
FLTK backend libs:
X11 include flags:
X11 libraries:
CARBON libraries:
PTHREAD flags
PTHREAD libraries -lpthread
LIBS: -lreadline -lncurses -liberty -lhdf5 -lz -lm -lgdi32
-lws2_32 -luser32
-lkernel32
Default pager: less
gnuplot: gnuplot
Magick config: GraphicsMagick++-config
Do internal array bounds checking: false
Build static libraries: false
Build shared libraries: true
Dynamic Linking: true (LoadLibrary)
Include support for GNU readline: true
64-bit array dims and indexing: false
configure:36157: WARNING: I didn't find bison, but it's only a problem if you
need to reconstruct
parse.cc
configure:36383: WARNING: "FLTK does not have OpenGL support. Native graphics
will be disabled."
configure:36392: WARNING:
configure:36394: WARNING: I didn't find the necessary libraries to compile
native
configure:36396: WARNING: graphics. It isn't necessary to have native graphics
configure:36398: WARNING: but you will have to use gnuplot or you won't be able
configure:36400: WARNING: to use any of Octave's plotting commands
configure:36402: WARNING:
configure:36412:
NOTE: libraries may be skipped if a library is not found OR
if the library on your system is missing required features.
(Note that ./autogen.sh was carried out with my msys environments because it
requires autotools.)
and make
Compile of oct-norm.cc was completed without error.
mingw32-g++-4.4.0-dw2 -shared-libgcc -c
-I/c/Programs/Octave/3.2.3_gcc-4.4.0/include -I. -I..
-I../liboctave -I../src -I../libcruft/misc -DHAVE_CONFIG_H -mieee-fp -Wall -W
-Wshadow
-Wold-style-cast -Wformat -lstdc++_s oct-norm.cc -o oct-norm.o
In your case complier obviously claims the oct-norm.cc has errrors. However, in
my case, even waring
appeared. Perhaps something is wrong with your development environments but I
do not figure out what
are wrong.
Regards
Tatsuro
--- address@hidden wrote:
> Hi Tatsuro,
>
> Thanks again for the help.
>
> I saw the same output when I applied the patch. I manually applied the
> rejected hunks that hadn't already been applied. Afterwards, I ran
> autogen.sh.
>
> I just recompiled without -D_GLIBCXX_DLL and it failed in a new way:
>
> mingw32-g++-4.4.0-dw2 -shared-libgcc -c
> -I/c/Octave/3.2.3_gcc-4.4.0/include -I. -I.. -I../liboctave -I../src
> -I../libcruft/misc -DHAVE_CONFIG_H -mieee-fp -Wall -W -Wshadow
> -Wold-style-cast -Wformat -lstdc++_s oct-norm.cc -o oct-norm.o
> oct-norm.cc: In function 'R higham(const MatrixT&, R, R, int, VectorT&)
> [with MatrixT = ComplexMatrix, VectorT = ComplexMatrix, R = double]':
> oct-norm.cc:480: instantiated from 'R matrix_norm(const MatrixT&, R,
> VectorT, SVDT) [with MatrixT = ComplexMatrix, VectorT = ComplexMatrix,
> SVDT = ComplexSVD, R = double]'
> oct-norm.cc:522: instantiated from here
> oct-norm.cc:427: error: ambiguous overload for 'operator/' in 'x /
> vector_norm [with T = std::complex<double>, R = double](((const
> MArray2<std::complex<double> >&)((const MArray2<std::complex<double>
> >*)(&((ComplexMatrix*)x)->ComplexMatrix::<anonymous>))), p)'
> CMatrix.h:420: note: candidates are: ComplexMatrix operator/(const
> ComplexMatrix&, const Complex&)
> dMatrix.h:369: note: Matrix operator/(const Matrix&, const
> double&)
> fCMatrix.h:420: note: FloatComplexMatrix operator/(const
> FloatComplexMatrix&, const FloatComplex&)
> fMatrix.h:370: note: FloatMatrix operator/(const
> FloatMatrix&, const float&)
> CNDArray.h:172: note: ComplexNDArray operator/(const
> ComplexNDArray&, const Complex&)
> oct-norm.cc: In function 'R higham(const MatrixT&, R, R, int, VectorT&)
> [with MatrixT = FloatComplexMatrix, VectorT = FloatComplexMatrix, R =
> float]':
> oct-norm.cc:480: instantiated from 'R matrix_norm(const MatrixT&, R,
> VectorT, SVDT) [with MatrixT = FloatComplexMatrix, VectorT =
> FloatComplexMatrix, SVDT = FloatComplexSVD, R = float]'
> oct-norm.cc:524: instantiated from here
> oct-norm.cc:427: error: ambiguous overload for 'operator/' in 'x /
> vector_norm [with T = std::complex<float>, R = float](((const
> MArray2<std::complex<float> >&)((const MArray2<std::complex<float>
> >*)(&((FloatComplexMatrix*)x)->FloatComplexMatrix::<anonymous>))), p)'
> CMatrix.h:420: note: candidates are: ComplexMatrix operator/(const
> ComplexMatrix&, const Complex&)
> dMatrix.h:369: note: Matrix operator/(const Matrix&, const
> double&)
> fCMatrix.h:420: note: FloatComplexMatrix operator/(const
> FloatComplexMatrix&, const FloatComplex&)
> fMatrix.h:370: note: FloatMatrix operator/(const
> FloatMatrix&, const float&)
> oct-norm.cc: In function 'R higham(const MatrixT&, R, R, int, VectorT&)
> [with MatrixT = SparseComplexMatrix, VectorT = ComplexMatrix, R =
> double]':
> oct-norm.cc:501: instantiated from 'R matrix_norm(const MatrixT&, R,
> VectorT) [with MatrixT = SparseComplexMatrix, VectorT = ComplexMatrix, R =
> double]'
> oct-norm.cc:548: instantiated from here
> oct-norm.cc:427: error: ambiguous overload for 'operator/' in 'x /
> vector_norm [with T = std::complex<double>, R = double](((const
> MArray2<std::complex<double> >&)((const MArray2<std::complex<double>
> >*)(&((ComplexMatrix*)x)->ComplexMatrix::<anonymous>))), p)'
> CMatrix.h:420: note: candidates are: ComplexMatrix operator/(const
> ComplexMatrix&, const Complex&)
> dMatrix.h:369: note: Matrix operator/(const Matrix&, const
> double&)
> fCMatrix.h:420: note: FloatComplexMatrix operator/(const
> FloatComplexMatrix&, const FloatComplex&)
> fMatrix.h:370: note: FloatMatrix operator/(const
> FloatMatrix&, const float&)
> CNDArray.h:172: note: ComplexNDArray operator/(const
> ComplexNDArray&, const Complex&)
> oct-norm.cc: In function 'void higham_subp(const ColVectorT&, const
> ColVectorT&, octave_idx_type, R, std::complex<_Up>&, std::complex<_Up>&)
> [with ColVectorT = ComplexMatrix, R = double]':
> oct-norm.cc:419: instantiated from 'R higham(const MatrixT&, R, R, int,
> VectorT&) [with MatrixT = ComplexMatrix, VectorT = ComplexMatrix, R =
> double]'
> oct-norm.cc:480: instantiated from 'R matrix_norm(const MatrixT&, R,
> VectorT, SVDT) [with MatrixT = ComplexMatrix, VectorT = ComplexMatrix,
> SVDT = ComplexSVD, R = double]'
> oct-norm.cc:522: instantiated from here
> oct-norm.cc:361: error: ambiguous overload for 'operator*' in 'mu1 * col'
> CNDArray.h:172: note: candidates are: ComplexNDArray operator*(const
> Complex&, const ComplexNDArray&)
> fMatrix.h:370: note: FloatMatrix operator*(const float&,
> const FloatMatrix&)
> fCMatrix.h:420: note: FloatComplexMatrix operator*(const
> FloatComplex&, const FloatComplexMatrix&)
> dMatrix.h:369: note: Matrix operator*(const double&, const
> Matrix&)
> CMatrix.h:420: note: ComplexMatrix operator*(const
> Complex&, const ComplexMatrix&)
> oct-norm.cc: In function 'VectorT dual_p(const VectorT&, R, R) [with
> VectorT = ComplexMatrix, R = double]':
> oct-norm.cc:438: instantiated from 'R higham(const MatrixT&, R, R, int,
> VectorT&) [with MatrixT = ComplexMatrix, VectorT = ComplexMatrix, R =
> double]'
> oct-norm.cc:480: instantiated from 'R matrix_norm(const MatrixT&, R,
> VectorT, SVDT) [with MatrixT = ComplexMatrix, VectorT = ComplexMatrix,
> SVDT = ComplexSVD, R = double]'
> oct-norm.cc:522: instantiated from here
> oct-norm.cc:401: error: ambiguous overload for 'operator/' in 'res /
> vector_norm [with T = std::complex<double>, R = double](((const
> MArray2<std::complex<double> >&)((const MArray2<std::complex<double> >*)(&
> res.ComplexMatrix::<anonymous>))), q)'
> CMatrix.h:420: note: candidates are: ComplexMatrix operator/(const
> ComplexMatrix&, const Complex&)
> dMatrix.h:369: note: Matrix operator/(const Matrix&, const
> double&)
> fCMatrix.h:420: note: FloatComplexMatrix operator/(const
> FloatComplexMatrix&, const FloatComplex&)
> fMatrix.h:370: note: FloatMatrix operator/(const
> FloatMatrix&, const float&)
> CNDArray.h:172: note: ComplexNDArray operator/(const
> ComplexNDArray&, const Complex&)
> oct-norm.cc: In function 'void higham_subp(const ColVectorT&, const
> ColVectorT&, octave_idx_type, R, std::complex<_Up>&, std::complex<_Up>&)
> [with ColVectorT = FloatComplexMatrix, R = float]':
> oct-norm.cc:419: instantiated from 'R higham(const MatrixT&, R, R, int,
> VectorT&) [with MatrixT = FloatComplexMatrix, VectorT =
> FloatComplexMatrix, R = float]'
> oct-norm.cc:480: instantiated from 'R matrix_norm(const MatrixT&, R,
> VectorT, SVDT) [with MatrixT = FloatComplexMatrix, VectorT =
> FloatComplexMatrix, SVDT = FloatComplexSVD, R = float]'
> oct-norm.cc:524: instantiated from here
> oct-norm.cc:361: error: ambiguous overload for 'operator*' in 'mu1 * col'
> fMatrix.h:370: note: candidates are: FloatMatrix operator*(const float&,
> const FloatMatrix&)
> fCMatrix.h:420: note: FloatComplexMatrix operator*(const
> FloatComplex&, const FloatComplexMatrix&)
> dMatrix.h:369: note: Matrix operator*(const double&, const
> Matrix&)
> CMatrix.h:420: note: ComplexMatrix operator*(const
> Complex&, const ComplexMatrix&)
> oct-norm.cc: In function 'VectorT dual_p(const VectorT&, R, R) [with
> VectorT = FloatComplexMatrix, R = float]':
> oct-norm.cc:438: instantiated from 'R higham(const MatrixT&, R, R, int,
> VectorT&) [with MatrixT = FloatComplexMatrix, VectorT =
> FloatComplexMatrix, R = float]'
> oct-norm.cc:480: instantiated from 'R matrix_norm(const MatrixT&, R,
> VectorT, SVDT) [with MatrixT = FloatComplexMatrix, VectorT =
> FloatComplexMatrix, SVDT = FloatComplexSVD, R = float]'
> oct-norm.cc:524: instantiated from here
> oct-norm.cc:401: error: ambiguous overload for 'operator/' in 'res /
> vector_norm [with T = std::complex<float>, R = float](((const
> MArray2<std::complex<float> >&)((const MArray2<std::complex<float> >*)(&
> res.FloatComplexMatrix::<anonymous>))), q)'
> CMatrix.h:420: note: candidates are: ComplexMatrix operator/(const
> ComplexMatrix&, const Complex&)
> dMatrix.h:369: note: Matrix operator/(const Matrix&, const
> double&)
> fCMatrix.h:420: note: FloatComplexMatrix operator/(const
> FloatComplexMatrix&, const FloatComplex&)
> fMatrix.h:370: note: FloatMatrix operator/(const
> FloatMatrix&, const float&)
> make[2]: *** [oct-norm.o] Error 1
> make[2]: Leaving directory `/c/Documents and
> Settings/mjkraai/Desktop/octave-3.2.4/liboctave'
> make[1]: *** [liboctave] Error 2
> make[1]: Leaving directory `/c/Documents and
> Settings/mjkraai/Desktop/octave-3.2.4'
> make: *** [all] Error 2
>
> Matt
>
>
>
>
> Tatsuro MATSUOKA <address@hidden>
> 02/11/2010 05:25 PM
> Please respond to
> address@hidden
>
>
> To
> address@hidden
> cc
> address@hidden
> Subject
> Re: Building for Microsoft Windows using MSYS
>
>
>
>
>
>
> Hello
>
> To be honest, I cannot find out that what was wrong with you.
> I am trying to build by myself.
>
> Perhaps the patch command seems to be in partly in failure.
> I have forced to patch with some rejections.
> How were the patch results?
>
> For me
> Apply anyway? [n] y
> Hunk #1 succeeded at 1079 with fuzz 2 (offset 22 lines).
> Hunk #2 FAILED at 1123.
> Hunk #3 FAILED at 1142.
> 2 out of 3 hunks FAILED -- saving rejects to file config.h.in.rej
> patching file configure.in
> Hunk #1 succeeded at 1272 (offset 3 lines).
> patching file doc/interpreter/Makefile.in
> Hunk #1 FAILED at 135.
> Hunk #2 succeeded at 182 (offset -1 lines).
> Hunk #3 FAILED at 205.
> 2 out of 3 hunks FAILED -- saving rejects to file
> doc/interpreter/Makefile.in.rej
> patching file doc/liboctave/Makefile.in
> patching file libcruft/Makefile.in
> patching file liboctave/Makefile.in
> Hunk #1 FAILED at 38.
> 1 out of 1 hunk FAILED -- saving rejects to file liboctave/Makefile.in.rej
> patching file liboctave/syswait.h
> patching file mkoctfile.cc.in
> patching file mkoctfile.in
> patching file octMakefile.in
> patching file scripts/Makefile.in
> patching file scripts/pkg/pkg.m
> patching file scripts/startup/main-rcfile
> patching file src/ChangeLog
> Hunk #1 succeeded at 415 (offset 42 lines).
> patching file src/DLD-FUNCTIONS/__magick_read__.cc
> patching file src/Makefile.in
> patching file src/pager.cc
>
>
> Did you carry out ./autogen.sh after applied patch before configure?
>
> I have met quit.cc problem from -D_GLIBCXX_DLL so I stopped to use
> -D_GLIBCXX_DLL
> For this problem, see
>
>
http://old.nabble.com/undefined-reference-to-%60_imp___ZTVSt9bad_alloc-and-_imp___ZNSt9bad_allocD1Ev-to27337166.html
>
>
>
> For configure
> cd "D:\usr\Tatsu\mingwhome\octaves\OctBuild\octave-3.2.4B"
>
> #FLAG setting
> export CC='gcc -shared-libgcc'
> export CXX='g++ -shared-libgcc'
> export F77='gfortran -shared-libgcc'
> #export CXXFLAGS='-D_GLIBCXX_DLL'
> export LDFLAGS='-L/c/Programs/Octave/3.2.3_gcc4.4.0/lib'
> export CPPFLAGS='-I/c/Programs/Octave/3.2.3_gcc4.4.0/include'
> export CXXLIBS='-lstdc++_s'
>
> ../../octavesrc/octave-3.2.4/configure
> --prefix=/c/Programs/OctaveBuild/3.2.4B
>
> In my case, from my preference, I made the working directory for build but
> it is not essential
> problem.
>
> To linking liboctave.dll was done without failure.
>
>
> g++ -shared-libgcc -shared -Wl,--export-all-symbols
> -Wl,--enable-auto-import
> -Wl,--enable-auto-image-base -L/c/Programs/
> Octave/3.2.3_gcc4.4.0/lib -Wl,--out-implib=liboctave.dll.a -o
> liboctave.dll \
> oct-locbuf.o CollocWt.o DASPK.o DASRT.o DASSL.o LSODE.o ODES.o
> Quad.o Range.o data-conv.o
> dir-ops.o file-ops.o file-stat.o glob-match.o idx-vector.o lo-ieee.o
> lo-mappers.o lo-specfun.o lo-
> :
> (skip)
> :
> rename.o rmdir.o strftime.o strptime.o strcasecmp.o strncase.o tempname.o
> tempnam.o cmd-edit.o
> cmd-hist.o oct-rl-edit.o oct-rl-hist.o pathsearch.o kpse-xfns.o
> -L../libcruft -L. -lcruft -lcholmod
> -lumfpack -lamd -lcamd -lcolamd -lccolamd -lcxsparse -larpack -lqrupdate
> -llapack -lblas -lfftw3
> -lfftw3f -lreadline -lglob -lregex -lpcre -lreadline -lncurses -liberty
> -lhdf5 -lz -lm -lgdi32
> -lws2_32 -luser32 -lkernel32 -Lc:/Programs/Octave/3.2.3_gcc4.4.0/lib
> -Lc:/programs/mingw/bin/../lib/gcc/mingw32/4.4.0
> -Lc:/programs/mingw/bin/../lib/gcc
> -Lc:/programs/mingw/bin/../lib/gcc/mingw32/4.4.0/../../../../mingw32/lib
> -Lc:/programs/mingw/bin/../lib/gcc/mingw32/4.4.0/../../.. -lhdf5 -lz -lm
> -lgfortranbegin -lgfortran
> -lmingw32 -lmoldname -lmingwex -lmsvcrt -luser32 -lkernel32 -ladvapi32
> -lshell32 -lpthread
> Creating library file: liboctave.dll.a
> rm -f liboctave.dll.3.2.4
> cp -p liboctave.dll liboctave.dll.3.2.4
>
> I hope the above helpful for you.
>
> Regards
>
> Tatsuro
>
> --- address@hidden wrote:
>
> > Hi Tatsuro,
> >
> > Thanks for the instructions. I tried following them, but the build
> > eventually failed with the following errors:
> >
> > mingw32-g++-4.4.0-dw2 -shared-libgcc -shared -Wl,--export-all-symbols
> > -Wl,--enable-auto-import -Wl,--enable-auto-image-base
> > -L/c/Octave/3.2.3_gcc-4.4.0/lib -Wl,--out-implib=liboctave.dll.a -o
> > liboctave.dll \
> > oct-locbuf.o CollocWt.o DASPK.o DASRT.o DASSL.o LSODE.o ODES.o
> > Quad.o Range.o data-conv.o dir-ops.o file-ops.o file-stat.o glob-match.o
>
> > idx-vector.o lo-ieee.o lo-mappers.o lo-specfun.o lo-sysdep.o lo-utils.o
> > mach-info.o oct-alloc.o oct-env.o oct-fftw.o oct-group.o oct-mutex.o
> > oct-md5.o oct-norm.o oct-passwd.o oct-rand.o oct-shlib.o oct-spparms.o
> > oct-syscalls.o oct-time.o oct-uname.o prog-args.o regex-match.o
> > sparse-sort.o sparse-util.o str-vec.o Array-C.o Array-b.o Array-ch.o
> > Array-i.o Array-d.o Array-f.o Array-fC.o Array-s.o Array-str.o
> > Array-voidp.o Array-idx-vec.o MArray-C.o MArray-ch.o MArray-i.o
> MArray-d.o
> > MArray-f.o MArray-fC.o MArray-s.o MSparse-C.o MSparse-d.o Sparse-C.o
> > Sparse-b.o Sparse-d.o oct-inttypes.o Array-util.o CColVector.o
> > CDiagMatrix.o CMatrix.o CNDArray.o CRowVector.o CmplxAEPBAL.o
> CmplxCHOL.o
> > CmplxGEPBAL.o CmplxHESS.o CmplxLU.o CmplxQR.o CmplxQRP.o CmplxSCHUR.o
> > CmplxSVD.o EIG.o fEIG.o boolMatrix.o boolNDArray.o chMatrix.o
> chNDArray.o
> > dColVector.o dDiagMatrix.o dMatrix.o dNDArray.o dRowVector.o
> dbleAEPBAL.o
> > dbleCHOL.o dbleGEPBAL.o dbleHESS.o dbleLU.o dbleQR.o dbleQRP.o
> dbleSCHUR.o
> > dbleSVD.o boolSparse.o CSparse.o dSparse.o MSparse.o Sparse.o
> > SparseCmplxLU.o SparsedbleLU.o SparseCmplxCHOL.o SparsedbleCHOL.o
> > SparseCmplxQR.o SparseQR.o MatrixType.o PermMatrix.o int8NDArray.o
> > uint8NDArray.o int16NDArray.o uint16NDArray.o int32NDArray.o
> > uint32NDArray.o int64NDArray.o uint64NDArray.o fCColVector.o
> fCRowVector.o
> > fCDiagMatrix.o fCMatrix.o fCNDArray.o fColVector.o fRowVector.o
> > fDiagMatrix.o fMatrix.o fNDArray.o fCmplxAEPBAL.o fCmplxCHOL.o
> > fCmplxGEPBAL.o fCmplxHESS.o fCmplxLU.o fCmplxSCHUR.o fCmplxSVD.o
> > fCmplxQR.o fCmplxQRP.o floatAEPBAL.o floatCHOL.o floatGEPBAL.o
> floatHESS.o
> > floatLU.o floatSCHUR.o floatSVD.o floatQR.o floatQRP.o f2c-main.o
> > filemode.o getopt.o getopt1.o lo-cieee.o lo-cutils.o md5.o mkdir.o
> > oct-getopt.o randgamma.o randmtzig.o randpoisson.o rename.o rmdir.o
> > strftime.o strptime.o strcasecmp.o strncase.o tempname.o tempnam.o
> > cmd-edit.o cmd-hist.o oct-rl-edit.o oct-rl-hist.o pathsearch.o
> kpse-xfns.o
> > -L../libcruft -L. -lcruft -lcholmod -lumfpack -lamd -lcamd -lcolamd
> > -lccolamd -lcxsparse -larpack -lqrupdate -llapack -lblas -lfftw3
> -lfftw3f
> > -lreadline -lglob -lregex -lpcre -lreadline -lncurses -liberty -lhdf5
> -lz
> > -lm -lgdi32 -lws2_32 -luser32 -lkernel32
> -Lc:/Octave/3.2.3_gcc-4.4.0/lib
> > -Lc:/octave/3.2.3_gcc-4.4.0/mingw32/bin/../lib/gcc/mingw32/4.4.0
> > -Lc:/octave/3.2.3_gcc-4.4.0/mingw32/bin/../lib/gcc
> >
> -Lc:/octave/3.2.3_gcc-4.4.0/mingw32/bin/../lib/gcc/mingw32/4.4.0/../../../../mingw32/lib
>
>
> >
> -Lc:/octave/3.2.3_gcc-4.4.0/mingw32/bin/../lib/gcc/mingw32/4.4.0/../../..
> > -L/mingw/lib -lhdf5 -lz -lm -lgfortranbegin -lgfortran -lmingw32
> > -lmoldname -lmingwex -lmsvcrt -luser32 -lkernel32 -ladvapi32 -lshell32
> > -lstdc++_s -lpthread
> > Creating library file: liboctave.dll.a
> > oct-norm.o:oct-norm.cc:(.text+0x103b): undefined reference to
> > `operator*(double const&, ComplexMatrix const&)'
> > oct-norm.o:oct-norm.cc:(.text+0x176a): undefined reference to
> > `operator*(float const&, FloatComplexMatrix const&)'
> >
> oct-norm.o:oct-norm.cc:(.text$_Z6highamI13ComplexMatrixS0_dET1_RKT_S1_S1_iRT0_[double
>
>
> > higham<ComplexMatrix, ComplexMatrix, double>(ComplexMatrix const&,
> double,
> > double, int, ComplexMatrix&)]+0x414): undefined reference to
> > `operator/(ComplexMatrix const&, double const&)'
> >
> oct-norm.o:oct-norm.cc:(.text$_Z6highamI18FloatComplexMatrixS0_fET1_RKT_S1_S1_iRT0_[float
>
>
> > higham<FloatComplexMatrix, FloatComplexMatrix, float>(FloatComplexMatrix
>
> > const&, float, float, int, FloatComplexMatrix&)]+0x38c): undefined
> > reference to `operator/(FloatComplexMatrix const&, float const&)'
> >
>
oct-norm.o:oct-norm.cc:(.text$_Z6highamI19SparseComplexMatrix13ComplexMatrixdET1_RKT_S2_S2_iRT0_[double
> >
> > higham<SparseComplexMatrix, ComplexMatrix, double>(SparseComplexMatrix
> > const&, double, double, int, ComplexMatrix&)]+0x414): undefined
> reference
> > to `operator/(ComplexMatrix const&, double const&)'
> >
> oct-norm.o:oct-norm.cc:(.text$_Z6dual_pI13ComplexMatrixdET_RKS1_T0_S4_[ComplexMatrix
>
>
> > dual_p<ComplexMatrix, double>(ComplexMatrix const&, double,
> > double)]+0x114): undefined reference to `operator/(ComplexMatrix const&,
>
> > double const&)'
> >
> oct-norm.o:oct-norm.cc:(.text$_Z6dual_pI18FloatComplexMatrixfET_RKS1_T0_S4_[FloatComplexMatrix
>
>
> > dual_p<FloatComplexMatrix, float>(FloatComplexMatrix const&, float,
> > float)]+0xbe): undefined reference to `operator/(FloatComplexMatrix
> > const&, float const&)'
> > CMatrix.o:CMatrix.cc:(.text+0x48b1): undefined reference to
> > `operator*(ComplexMatrix const&, DiagMatrix const&)'
> > dMatrix.o:dMatrix.cc:(.text+0x3223): undefined reference to
> > `operator*(Matrix const&, DiagMatrix const&)'
> > CSparse.o:CSparse.cc:(.text+0x2756b): undefined reference to
> > `operator*(ComplexMatrix const&, double const&)'
> > CSparse.o:CSparse.cc:(.text+0x27851): undefined reference to
> > `operator*(Matrix const&, std::complex<double> const&)'
> > CSparse.o:CSparse.cc:(.text+0x2843e): undefined reference to
> > `operator*(std::complex<double> const&, Matrix const&)'
> > CSparse.o:CSparse.cc:(.text+0x28717): undefined reference to
> > `operator*(double const&, ComplexMatrix const&)'
> > fCMatrix.o:fCMatrix.cc:(.text+0x4661): undefined reference to
> > `operator*(FloatComplexMatrix const&, FloatDiagMatrix const&)'
> > fMatrix.o:fMatrix.cc:(.text+0x3138): undefined reference to
> > `operator*(FloatMatrix const&, FloatDiagMatrix const&)'
> > collect2: ld returned 1 exit status
> > make[2]: *** [liboctave.dll] Error 1
> > make[2]: Leaving directory `/c/Documents and
> > Settings/mjkraai/Desktop/octave-3.2.4/liboctave'
> > make[1]: *** [liboctave] Error 2
> > make[1]: Leaving directory `/c/Documents and
> > Settings/mjkraai/Desktop/octave-3.2.4'
> > make: *** [all] Error 2
> >
> > Do you have any suggestions?
> >
> > Matt
> >
> >
> >
> >
> > Tatsuro MATSUOKA <address@hidden>
> > 02/10/2010 05:10 PM
> > Please respond to
> > address@hidden
> >
> >
> > To
> > address@hidden, address@hidden, Benjamin Lindner
> > <address@hidden>
> > cc
> > address@hidden
> > Subject
> > Re: Building for Microsoft Windows using MSYS
> >
> >
> >
> >
> >
> >
> > Please execute ./autogen,sh before configure. It is required to update
> > configure script and the
> > related files to reflect the modification by the patches.
> >
> > Good luck!
> >
> > Regards
> >
> > Tatsuro
> > --- Tatsuro MATSUOKA wrote:
> >
> > > BTW, in practical sense, it is better to use build libraries in
> > octave-3.2.3 binary package.
> > >
> > > 1. download and extract octave source
> > > 2. download patch
> > >
> > >
> >
>
http://octave.svn.sourceforge.net/viewvc/octave/trunk/octave-forge/admin/Windows/mingw32/octave/octave-3.2.3-4.patch?view=log
>
> >
> > > 3. modify patch. version number '3.2.3' should be replaced by '3.2.4'
> > > 4. apply patch with patch -p 1 < (patch file name) at the base
> directory
> > of source trees
> > > 5. correct related source files manually if the some part of the patch
>
> > fail
> >
> >
> >
> > > 6. configure
> > > #example
> > > cd (source directry)
> > > # add PATH to use dll libraries, gnuplot, and gcc
> > > export
> > >
> >
> PATH=$PATH:/c/Programs/Octave/3.2.3_gcc4.4.0/bin:/c/Programs/Octave/3.2.3_gcc4.4.0/mingw32/bin
> > >
> > > #FLAG setting
> > > export CC='mingw32-gcc-4.4.0-dw2 -shared-libgcc'
> > > export CXX='mingw32-g++-4.4.0-dw2 -shared-libgcc'
> > > export F77='mingw32-gfortran-4.4.0-dw2 gfortran -shared-libgcc'
> > > #if you have installed gcc-4.4.0 you can use
> > > # export CC='gcc -shared-libgcc'
> > > # instead the above
> > > export CXXFLAGS='-D_GLIBCXX_DLL'
> > > export LDFLAGS='-L/c/Programs/Octave/3.2.3_gcc4.4.0/lib'
> > > export CPPFLAGS='-I/c/Programs/Octave/3.2.3_gcc4.4.0/include'
> > > export CXXLIBS='-lstdc++_s'
> > >
> > > ./configure --prefix=(octave install directory path)
> > >
> > > 5. make
> > > 6. make check
> > > 7. make install-strip
> > >
> > > Regards
> > >
> > > Tatsuro
> > >
> > > --------------------------------------
> > > VANCOUVER 2010 Olympic News [Yahoo! Sports/sportsnavi]
> > > http://pr.mail.yahoo.co.jp/olympic/
> > >
> >
> >
> > --------------------------------------
> > VANCOUVER 2010 Olympic News [Yahoo! Sports/sportsnavi]
> > http://pr.mail.yahoo.co.jp/olympic/
> >
> >
> > The server made the following annotations
> >
> >
> ---------------------------------------------------------------------------------
> >
> > This message contains information that may be privileged or confidential
> and is the property of
> > Beckman Coulter, Inc. It is intended only for the person to whom it is
> addressed. If you are
> > not the intended recipient, you are not authorized to read, print,
> retain, copy, disseminate,
> > distribute or use this message or any part thereof. If you receive this
> message in error,
> > please notify the sender immediately and delete all copies of this
> message.
> >
> >
> ---------------------------------------------------------------------------------
> >
> >
>
>
> --------------------------------------
> VANCOUVER 2010 Olympic News [Yahoo! Sports/sportsnavi]
> http://pr.mail.yahoo.co.jp/olympic/
>
>
> The server made the following annotations
>
> ---------------------------------------------------------------------------------
>
> This message contains information that may be privileged or confidential and
> is the property of
> Beckman Coulter, Inc. It is intended only for the person to whom it is
> addressed. If you are
> not the intended recipient, you are not authorized to read, print, retain,
> copy, disseminate,
> distribute or use this message or any part thereof. If you receive this
> message in error,
> please notify the sender immediately and delete all copies of this message.
>
> ---------------------------------------------------------------------------------
>
>
--------------------------------------
VANCOUVER 2010 Olympic News [Yahoo! Sports/sportsnavi]
http://pr.mail.yahoo.co.jp/olympic/
Re: Building for Microsoft Windows using MSYS, Benjamin Lindner, 2010/02/15