[Numpy-discussion] import scipy.linalg is hanging on Marvell armada 370

[Charles R Harris]

On Tue, May 5, 2015 at 1:52 PM, Alexander Brezinov <
alexander.brezinov at mmbresearch.com> wrote:

> Hello
>
> The import of scipy.linalg is hanging in DOUBLE_mutiply function
> (BINARY_LOOP) in umath.so. After attaching the gdb and dumping the local
> varibles the args are empty strings. Could you please advise if this is
> known issue? I just search the mailing list and could not find any solution
> for the problem.
>
> I am running:
>
> kernel 3.2.36  + Debian wheezy on ARMv71 armhf
> CPU Armada 370 Marvell
> python 2.7.3
> scipy 0.15.1
> numpy 1.9.2
>
> The problem could be reproduced by launching python and importing
> scipy.linalg(import linalg)
>
> I also run the same OS on qemu and was not able to reproduce the issue.
> Similar architecture such as rasbery pi (ARMv7 armhf) is fine. Also if
> using software floating point intead of hardware floating point on the same
> Armada 370 (ARMv7) working just fine.
>
>
> Thank you for any comments or suggestions in advance,
> Alex
>
>
Almost sounds like a compiler problem, are you using the correctly compiled
version of umath.so? Not that there couldn't be other sources of the
problem...

Chuck
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/numpy-discussion/attachments/20150505/a28bf9d8/attachment.html>