[SciPy-user] g77 sse2 problem back(?) with fedora 5

Mark Heslep mark at mitre.org
Tue Mar 21 21:20:27 CST 2006


Mark Heslep wrote:
> Fix:  Removing the 'sse2' flag in ..numpy/distutils/fcompiler/gnu.py  
> clears the problem.  
>   
Well maybe not? scipy.test(9) on the Pentium M laptop gives
> Ran 1520 tests in 74.202s
> FAILED (failures=5, errors=145)
>
>
With errors like:
> ======================================================================
> ERROR: Does the matrix's sum() method work?
> ----------------------------------------------------------------------
> Traceback (most recent call last):
> File 
> "/usr/lib/python2.4/site-packages/scipy/sparse/tests/test_sparse.py", 
> line 42, in check_sum
> assert_array_equal(self.dat.sum(), self.datsp.sum())
> File "/usr/lib/python2.4/site-packages/scipy/sparse/sparse.py", line 
> 396, in sum
> return (o0 * (self * o1)).A.squeeze()
> File "/usr/lib/python2.4/site-packages/numpy/core/defmatrix.py", line 
> 138, in __mul__
> return N.dot(self, other)
> ValueError: objects are not aligned 
The Xeon box reports no errors, no failures.
> Ran 1081 tests in 84.241s
> FAILED (failures=2)
Checkdot caused the failure.



More information about the SciPy-user mailing list