[SciPy-Dev] [SciPy-User] ANN: SciPy 0.11.0 release candidate 1

[Ralf Gommers]

On Tue, Aug 7, 2012 at 12:40 AM, Pauli Virtanen <pav at iki.fi> wrote:

> 06.08.2012 23:21, Ralf Gommers kirjoitti:
> > On Mon, Jul 30, 2012 at 10:36 PM, Ralf Gommers wrote:
> [clip]
> > I have been able to reproduce this now. Some tracebacks and details at
> > http://projects.scipy.org/scipy/ticket/1717. If anyone has some ideas,
> > shoot!
>
> If I understand correctly, this bug only occurs with the SSE3 binaries
> in our vendor repo. I should try to reproduce this also.
>
> Compiling with DEBUGFLAGS seems not to have worked out in your case. The
> f2py wrappers should spew out a lot of debug prints, if compiled with
> the correct debug #define enabled.
>

So where does f2py get its flags from? Since I can't set env variables in
wine the way you suggested, I added this flag to the following items in
"executables" in numpy/distutils/gnu.py (and checked that they were present
in the build log):
        'compiler_f77'
        'compiler_f90'
        'linker_so'
        'linker_exe'

Ralf


> The second thing is a longer shot, but wine does have a debugger, and
> obtaining a traceback from that could help. Or possibly not.
>
>         Pauli
>
>
> _______________________________________________
> SciPy-Dev mailing list
> SciPy-Dev at scipy.org
> http://mail.scipy.org/mailman/listinfo/scipy-dev
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scipy-dev/attachments/20120807/83c734aa/attachment.html>