[SciPy-user] SciPy with gcc4 and gfortran on OS X

[Robert Kern]

Zachary Pincus wrote:
> Robert -
> 
> Thanks for the feedback about the errors, etc.
> Do you have any feel for the persistent errors on scipy.test() that  
> I've been seeing with gfortran? You commented on the sporadic errors  
> earlier, but the consistent ones worry me the most, given that it  
> appears to be a simple dot product (on complex numbers, reproduced  
> below for convenience) that is failing badly, and consistently. Is  
> this indicative of gfortran problems, do you think, or something else  
> that I can hunt down?

I can't think of anything offhand. You will want to check whether the function
that is failing is cdotu or zdotu. Also, check if {c,z}dotc is giving incorrect
results, too. {c,z}dotu seems to be used in _dotblas, so you should check
numpy.dot() for the same bug.

-- 
Robert Kern
robert.kern at gmail.com

"I have come to believe that the whole world is an enigma, a harmless enigma
 that is made terrible by our own mad attempt to interpret it as though it had
 an underlying truth."
  -- Umberto Eco