On 3/10/2012 9:31 PM, Sameer Grover wrote: > On 10 March 2012 02:23, Christoph Gohlke<cgoh...@uci.edu> wrote: >> >> >> On 3/9/2012 11:50 AM, Sameer Grover wrote: >>>>>> import gtk >>>>>> import foo # where foo is any f2py-wrapped program >>> >>> Subsequently, on exiting python interpreter, the interpreter crashes >>> with this error message - "This application has requested the Runtime >>> to terminate it in an unusual way. Please contact the application's >>> support team for more information." >>> >>> Strangely enough, interchanging the order of the import statements, >>> i.e. importing the f2py wrapped program before gtk works fine. >>> Furthermore, each module works fine individually. >>> >>> This is a windows-only problem. I'm using Windows 7, Python 2.7, >>> latest numpy, mingw32 compiler and the "pygtk all-in-one installer" >>> (mentioned on the pygtk download page). >>> >>> This happens even for very simpl fortran programs such as this one - >>> subroutine hello () >>> write(*,*)'Hello from Fortran90!!!' >>> end subroutine hello >>> >>> I don't know whether the problem is with f2py or with gtk or with >>> python but maybe somebody can shed some light on this. >>> >>> Regards, >>> Sameer Grover >> >> The error can be due to memory corruption. It works for me with >> msvc9/ifort builds of the pygtk and f2py extensions. >> >> Which DLLs does foo.pyd depend on (use DependencyWalker)? >> >> Christoph >> _______________________________________________ >> NumPy-Discussion mailing list >> NumPy-Discussion@scipy.org >> http://mail.scipy.org/mailman/listinfo/numpy-discussion > > The .pyd file generated by f2py depends on the following > > KERNEL32.DLL > MSVCRT.DLL > LIBGCC_S_DW2-1.DLL > LIBGFORTRAN-3.DLL > PYTHON27.DLL > > It is possible that this is a mingw32 problem. I haven't been able to > try using ifort/msvc.(mscv is somehow not getting installed on my > system, but that's a separate issue). > > Sameer
Can you try linking to the msvc9 runtime DLL? E.g. f2py.py -c --fcompiler=gnu95 --compiler=mingw32 -lmsvcr90 -m foo foo.f Christoph _______________________________________________ NumPy-Discussion mailing list NumPy-Discussion@scipy.org http://mail.scipy.org/mailman/listinfo/numpy-discussion
