problem: "octave-3.2.4" + "c++ (intel compiler)" + "x86-64 architecture" + "data.cc"

[Miguel Pan Fidalgo]

Hi, 

After compiling octave-3.2.4 in our itanium cluster (thank you for the patch 
Jaroslav Hajek)

I'm trying to install octave-3.2.4 in our x86-64 cluster 
[http://www.cesga.es/content/view/409/42/lang,en/ ]. 

The first problem was the compatibility between the icpc (intel c++ compiler) 
and the octave code. This tiny example is more expressive than me: 

-bash-3.00$ cat test.cc
class A {};

template <class X, class T> void method (double u, T v);

template <class X> void method (double u, long v);

int main (){
  double u(0); long v(0);
  method<A> (u, v);
}

-bash-3.00$ module load icc/11.0.083
-bash-3.00$ icpc -c test.cc
test.cc(9): error: more than one instance of overloaded function "method" 
matches the argument list:
            function template "void method<X,T>(double, T)"
            function template "void method<X>(double, long)"
            argument types are: (double, long)
    method<A> (u, v);
    ^

compilation aborted for test.cc (code 2)
-bash-3.00$ icpc -gcc-version=410 -c test.cc

With "-gcc-version=410" everything compiles properly, but I have precision 
problems in data.cc file.

src/data.cc ............................................ PASS  501/509  FAIL 8


***** assert(log2(complex(0,Inf)), Inf + log2(i));
!!!!! test failed
assert (log2 (complex (0, Inf)),Inf + log2 (i)) expected
Inf + 2.266i
but got
NaN + 2.266i
NaNs don't match  
***** test
 [f, e] = log2 (complex (zeros (3, 2), [0,-1; 2,-4; Inf,-Inf]));
 assert (f, complex (zeros (3, 2), [0,-0.5; 0.5,-0.5; Inf,-Inf]));
 assert (e(1:2,:), [0,1; 2,3]);
!!!!! test failed
assert (f,complex (zeros (3, 2), [0, -0.5; 0.5, -0.5; Inf, -Inf])) expected
   0.00000 + 0.00000i   0.00000 - 0.50000i
   0.00000 + 0.50000i   0.00000 - 0.50000i
   0.00000 +     Infi   0.00000 -     Infi
but got
   0.00000 + 0.00000i   0.00000 - 0.50000i
   0.00000 + 0.50000i   0.00000 - 0.50000i
       NaN -     NaNi       NaN -     NaNi
NaNs don't match 
****

I found the following thread very interesting, but I use intel compilers with 
no optimization: http://old.nabble.com/data.cc-failing-1-test-tc26822635.html 

Now I'm trying to compile with "-gcc-version=420" (with "-gcc-version=440" it 
fails). This is the configuration that I have previously used: 

 Installation prefix:  /opt/cesga/octave-3.2.4
  C compiler:           icc    -O0
  C++ compiler:         icpc   -I/usr/include/freetype2  -O0 -gcc-version=410
  Fortran compiler:     ifort  -O0
  Fortran 
libraries:     -lhdf5 -lz -L/opt/cesga/intel/Compiler/11.0/083/lib/intel64 
-limf -lm -L/opt/cesga/gnuplot-4.2.6/lib -L/opt/cesga/ftgl-2.1.3-rc5/lib -L/op
t/cesga/pcre-7.9/64/lib -L/opt/cesga/qhull-2003.1-patch/lib 
-L/opt/cesga/hdf5-1.8.1-api-1.6/lib -L/opt/cesga/fftw-3.2/64/lib 
-L/opt/cesga/gsl-1.12/lib -L/opt/cesga
/intel/Compiler/11.0/083/mkl/lib/em64t -L/opt/cesga/sge62/lib 
-L/opt/cesga/sge62/lib/lx24-x86 -L/usr/lib/gcc/x86_64-redhat-linux/3.4.6 
-L/usr/lib/gcc/x86_64-redhat
-linux/3.4.6/../../../../lib64 
-L/usr/lib/gcc/x86_64-redhat-linux/3.4.6/../../.. -L/lib64 -L/lib -L/usr/lib64 
-L/usr/lib -lifport -lifcore -lsvml -lipgo -lirc -lpt
hread -lirc_s -ldl
  BLAS 
libraries:       -L/opt/cesga/intel/Compiler/11.0/083/mkl/lib/em64t 
/opt/cesga/intel/Compiler/11.0/083/mkl/lib/em64t/libmkl_solver_lp64_sequential.a
 -Wl,--s
tart-group -lmkl_intel_lp64 -lmkl_sequential -lmkl_core -Wl,--end-group 
-lpthread  -L/opt/cesga/intel/Compiler/11.0/083/mkl/lib/em64t 
/opt/cesga/intel/Compiler/11.
0/083/mkl/lib/em64t/libmkl_solver_lp64_sequential.a -Wl,--start-group 
-lmkl_intel_lp64 -lmkl_sequential -lmkl_core -Wl,--end-group -lpthread
  FFTW libraries:       -lfftw3 -lfftw3f
  GLPK libraries:
  UMFPACK libraries:
  AMD libraries:
  CAMD libraries:
  COLAMD libraries:
  CCOLAMD libraries:
  CHOLMOD libraries:
  CXSPARSE libraries:
  ARPACK libraries:
  QRUPDATE libraries:
  HDF5 libraries:       -lhdf5
  CURL libraries:
  REGEX libraries:      -L/opt/cesga/pcre-7.9/64/lib -lpcre
  QHULL libraries:      -lqhull
  OPENGL 
libraries:     -lftgl -L/usr/lib64 -Wl,--rpath -Wl,/usr/lib64 -lfreetype -lz 
-L/usr/X11R6/lib -lGL -lGLU
  FLTK backend libs:
  X11 include flags:    /usr/X11R6/include
  X11 libraries:        -L/usr/X11R6/lib64 -lX11
  CARBON libraries:
  PTHREAD flags         -pthread
  PTHREAD libraries
  LIBS:                 -lreadline  -lncurses -ldl -lhdf5 -lz -lm
  Default pager:        less
  gnuplot:              gnuplot
  Magick config:

  Do internal array bounds checking:  true
  Build static libraries:             false
  Build shared libraries:             true
  Dynamic Linking:                    true (dlopen)
  Include support for GNU readline:   true
  64-bit array dims and indexing:     false

-- 
Atentamente, 

Miguel Pan Fidalgo (Applications Technician) 
mail: mpan at cesga.es 
web: http://www.cesga.es
Avda. de Vigo s/n 15705, Santiago de Compostela
Telf.: +34 981 569810 - Fax: 981 594616
-------------------------------------------------------------------------