Already tried the new numpy and scipy versions from #9808 instead?
I'm not sure if it helps, but give it a try. If the fortran compiler
is somehow involved the new versions could solve your problem.
On 24 Sep., 17:18, Niles wrote:
> Sorry for not seeing the other messages in this thread before m
Sorry for not seeing the other messages in this thread before my
previous post; SAGE_BINARY_BUILD is indeed the environment variable
that does the trick.
-Niles
On Sep 24, 11:11 am, Niles Johnson wrote:
> Oh, according to my install log I had the same problem; I was fooled
> by the fact that sa
On 24 Sep., 09:40, Minh Nguyen wrote:
> Hi Niles,
>
> On Fri, Sep 24, 2010 at 7:58 AM, Niles wrote:
> > Is there something easy I "forgot" to do?
>
> I couldn't even get Sage to compile successfully on that machine.
> Doing a serial compilation with
>
> $ make
>
> resulted in
>
> /usr/local/lib
it's probably some lapack/blas/atlas related. Hard to say without
seeing what
you replaced by ...
On Sep 24, 5:58 am, Niles wrote:
> I just compiled 4.6.alpha1 on the machine rosemary.math at UGA (24-
> core SUN machine running Redhat Linux Enterprise Edition) with (I
> thought) no problems. But
