[SciPy-dev] newcore and gcc4/gfortran

Travis Oliphant oliphant at ee.byu.edu
Fri Nov 18 03:46:38 CST 2005


Arnd Baecker wrote:

>Hi,
>
>just for the fun of it  I tried to install newcore using gcc4.
>More precisely:
>64 Bit opteron, gcc version 4.0.2 20050901 (prerelease) (SUSE Linux).
>Newcore works fine!
>However for newscipy, built with
>  python setup.py config_fc --fcompiler=gnu95 build
>  python setup.py install --prefix=$DESTDIR
>, the following
>  import scipy
>  scipy.test(10,10)
>gives:
>
>check_bdtrik (scipy.special.basic.test_basic.test_cephes)
>Program received signal SIGSEGV, Segmentation fault.
>[Switching to Thread 46912509653888 (LWP 21533)]
>0x00007fffffbcb0e0 in ?? ()
>(gdb) bt
>#0  0x00007fffffbcb0e0 in ?? ()
>#1  0x00007fffffbcb0b0 in ?? ()
>#2  0x0000000000000000 in ?? ()
>#3  0x00007fffffbcb0b0 in ?? ()
>#4  0x00007fffffbcb0a8 in ?? ()
>#5  0x00007fffffbcb0d8 in ?? ()
>#6  0x00002aaab1240d41 in dinvr_ ()
>   from
>/home/abaecker/BUILDS2/Build_83_gcc4/inst_scipy_newcore/lib64/python2.4/site-packages/scipy/special/_cephes.so
>#7  0x00002aaab123b4cc in cdfbin_ ()
>   from
>/home/abaecker/BUILDS2/Build_83_gcc4/inst_scipy_newcore/lib64/python2.4/site-packages/scipy/special/_cephes.so
>#8  0x00002aaab1208600 in cdfbin2_wrap (p=1, xn=3, pr=0.5) at
>cdf_wrappers.c:98
>#9  0x00002aaab120ba99 in PyUFunc_ddd_d (args=<value optimized out>,
>dimensions=<value optimized out>,
>    steps=<value optimized out>, func=0x2aaab1208580) at
>ufunc_extras.c:303
>  
>
cdfbin_ is a Fortran-compiled subroutine so this looks like a compiler 
problem.  It doesn't look like the fortran compiler in gcc is really 
ready for primetime.  

At some point it may be possible to test the compiler on these fortran 
files in scipy and decide if it will work during configuration.  I'm not 
going to start debuggin the gcc fortran compiler at this point, though.

-Travis





More information about the Scipy-dev mailing list