Re: fntests fails in development sources (mingw gcc-4.3.2-dw2-TDM)

[Tatsuro MATSUOKA]

Hello

The previous post you can find in the following

http://www.nabble.com/fntests-fails-in-development-sources-(mingw-gcc-4.3.2-dw2-TDM)-to22368640.html#a22368640


At least A\b hangs :-(

octave.exe:1> A=rand(5)
A =

   0.406476   0.445073   0.362062   0.330964   0.872662
   0.509578   0.680379   0.189348   0.417418   0.268378
   0.602942   0.988650   0.169062   0.444177   0.361676
   0.455814   0.738912   0.297219   0.902869   0.982388
   0.312063   0.194874   0.977311   0.858924   0.055628

octave.exe:2> b=rand(5,1)
b =

   0.077008
   0.977694
   0.584754
   0.451279
   0.586048

octave.exe:3> A\b

Here octave hangs up. 

This may come from combitionaion ALTAS and Lapack.
I will try to build octave-3.1.53+ with refrenc

--- Tatsuro MATSUOKA  wrote:

> Hello
> 
> I have tried to build development sources on http://www.octave.org/hg/octave
> by mingw gcc-4.3.2-dw2-TDM.
> 
> Octave is now configured for i686-pc-mingw32
> 
>   Source directory:     ../octave-work
>   Installation prefix:  /usr/local/octave-3.1.x
>   C compiler:           gcc-dw2  -mieee-fp  -Wall -W -Wshadow -O3 
> -fomit-frame-pointer
>   C++ compiler:         g++-dw2  -mieee-fp 
> -I/WinDevTools/bin/../include/freetype2
> -I/WinDevTools/bin/../include  -Wall -W -Wshadow -Wold-style-cast -O3 
> -fomit-frame-pointer
>   Fortran compiler:     gfortran-dw2 -O3 -fomit-frame-pointer -mieee-fp
>   Fortran libraries:     -Lc:/Programs/mingw/lib/gcc/mingw32/4.3.2-dw2 
> -Lc:/Programs/mingw/lib
> -LC:/Programs/msys/1.0/local/octlib4.3.2/lib -Lc:/Programs/GnuWin32/lib
> -Lc:/programs/mingw/bin/../lib/gcc/mingw32/4.3.2-dw2 
> -Lc:/programs/mingw/bin/../lib/gcc
> -Lc:/programs/mingw/bin/../lib/gcc/mingw32/4.3.2-dw2/../../../../mingw32/lib
> -Lc:/programs/mingw/bin/../lib/gcc/mingw32/4.3.2-dw2/../../.. -lhdf5 -lz -lm 
> -lstdc++ -lg2c
> -luuid
> -lwsock32 -lws2_32 -lgmp -lregex -lwldap32 -lwininet -lgfortranbegin 
> -lgfortran -lmingw32
> -lmoldname
> -lmingwex -lmsvcrt -luser32 -lkernel32 -ladvapi32 -lshell32
>   BLAS libraries:       -llapack -lf77blas -lcblas -latlas 
>   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:      -Lc:/progra~1/Pcre/lib -lpcre -Wl,-s
>   QHULL libraries:    -lqhull
>   OPENGL libraries:     -lopengl32 -lglu32 -L/WinDevTools/bin/../lib 
> -lfreetype -lftgl
>   FLTK backend libs:    -L/usr/local/octlib4.3.2/lib -mwindows -L/mingw/lib 
> -L/GnuWin32/lib
> -L/usr/local/octave-3.1.51/lib -mno-cygwin -lfltk_gl -lglu32 -lopengl32 
> -lfltk -luuid -lwsock32
> -lws2_32 -lgmp -lole32 -luuid -lcomctl32 -lwsock32
>   X11 include flags:    
>   X11 libraries:        
>   CARBON libraries:     
>   LIBS:                 -lreadline  -ltermcap -liberty -lf77blas -lcblas 
> -latlas  -lhdf5 -lz -lm
> -lstdc++ -lg2c -luuid -lwsock32 -lws2_32 -lgmp -lregex -lwldap32 -lwininet 
> -lgdi32 -lws2_32
> -lkernel32
>   Default pager:        less
>   gnuplot:              pgnuplot
>   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
> 
>  
> 
> The latest ChangeLog is
> 2009-02-25  John W. Eaton  
>       * configure.in (doc_cache_file): New default variable
>       * Makeconf.in (@doc_cache_file, do-subst-default-vals): Substitute it.
> 
> The build itself seem to be successful, however many tests seem to hang up 
> (no response).
> 
> The lists of hang up tests are
> 
> ./scripts/geometry:
> delaunay.m
> delaunay3.m
> griddata.m
> griddata3.m
> griddatan.m
> 
> ./scripts/linear-algebra:
> cond.m
> condest.m
> 
> ./scripts/optimization:
> fsolve.m
> lsqnonneg.m
> sqp.m
> 
> ./scripts/polynomial:
> poly.m
> polyfit.m
> polyval.m
> residue.m
> roots.m
> 
> ./scripts/signal:
> detrend.m
> 
> ./scripts/sparse:
> pcg.m
> pcr.m
> svds.m
> 
> ./scripts/special-matrix:
> hadamard.m
> 
> ./scripts/statistics/base:
> cov.m
> 
> ./src:
> data.cc
> 
> ./src/DLD-FUNCTIONS:
> chol.cc
> dassl.cc
> det.cc
> eig.cc
> eigs.cc
> hess.cc
> inv.cc
> lsode.cc
> lu.cc
> qr.cc
> schur.cc
> svd.cc
> syl.cc
> 
> 
> ./test:
> test_io.m
> 
> The script test_io.m works after the modification at line 81
> from
> %!  opts = {"-z -text", "-z -binary", "-z -mat", "-v7"};
> to
> %!  opts = {"-text", "-binary", "-mat", "-v7"};
> 
> This is reported previously 
> http://www.nabble.com/Quesion-about-test_io.m-td22074114.html#a22074114
> 
> Perhaps reasons of most fails comes from faults in matrix calculation 
> routines.
> I will perphase reasons of the fails in the test indicate above.
> 
> Regards
> 
> Tatsuro
> 
> 
> 
> --------------------------------------
> Power up the Internet with Yahoo! Toolbar.
> http://pr.mail.yahoo.co.jp/toolbar/
> 


--------------------------------------
Power up the Internet with Yahoo! Toolbar.
http://pr.mail.yahoo.co.jp/toolbar/