[SciPy-dev] numpy 1.3.0 import error

[Forrest Sheng Bao]

Hi David, do you know how to check my CFLAGS or LDFLAGS before compiling? I
used the set.py to do everything.

Cheers,
Forrest

On Thu, Dec 3, 2009 at 10:56 PM, David Cournapeau <
david at ar.media.kyoto-u.ac.jp> wrote:

> Forrest Sheng Bao wrote:
> > Hi there,
> >
> > I just compiled and installed numpy 1.3.0 on my system. But when I try
> > to import numpy, I got this error:
> >
> > from numpy import *
> > Traceback (most recent call last):
> >   File "<stdin>", line 1, in ?
> >   File
> > "/home/sbao/apps/lib64/python2.4/site-packages/numpy/__init__.py",
> > line 130, in ?
> >     import add_newdocs
> >   File
> > "/home/sbao/apps/lib64/python2.4/site-packages/numpy/add_newdocs.py",
> > line 9, in ?
> >     from lib import add_newdoc
> >   File
> > "/home/sbao/apps/lib64/python2.4/site-packages/numpy/lib/__init__.py",
> > line 4, in ?
> >     from type_check import *
> >   File
> > "/home/sbao/apps/lib64/python2.4/site-packages/numpy/lib/type_check.py",
> > line 8, in ?
> >     import numpy.core.numeric as _nx
> >   File
> > "/home/sbao/apps/lib64/python2.4/site-packages/numpy/core/__init__.py",
> > line 5, in ?
> >     import multiarray
> > ImportError:
> > /home/sbao/apps/lib64/python2.4/site-packages/numpy/core/multiarray.so:
> > undefined symbol: __intel_security_cookie
> >
> > The CPU is an Intel Xeon. Any developer know what's the reason?
>
> Most likely a problem with building with icc. __intel_security_cookie is
> a canary against buffer overflow, but I cannot find much reference about
> it. Make sure neither CFLAGS nor LDFLAGS is set when you build numpy - I
> would not be surprised if this is specific to some security-related
> flags in icc. You could look which Intel library  define this symbol, by
> running nm + grep on the Intel compiler library directory.
>
> As I mentioned in your other thread, I could finally reproduce your
> error, fixed numpy sources, and 1.4.x should now be buildable. I can
> myself import numpy built with icc 11.0 on ia32.
>
> David
> _______________________________________________
> 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/20091204/3d41f5db/attachment.html>