Re: [AMBER] Only -mpi option creates the python modules. Is this right?

From: Jason Swails <jason.swails.gmail.com>
Date: Tue, 2 Jun 2015 10:58:42 -0400

On Tue, Jun 2, 2015 at 10:26 AM, Eduardo Almeida Costa <eacosta.uesc.br>
wrote:

> Hi, I'm Eduardo. Greetings.
>
> I can't test the CUDA, OpenMP ou MPI+CUDA installation. (make test.xxxx).
> I have the Amber14 and AmberTools15, and followed the Amber 2015 Reference
> Manual and http://jswails.wikidot.com/installing-amber14-and-ambertools14.
>
> When I use the options:
>
> 1. ./configure -mpi -cuda gnu
> 2. ./configure -mpi -openmp gnu
> 3. ./configure -openmp gnu
> 4. ./configure -cuda gnu
>
> In these cases, the make option doesn't create the folder "lib/python2.6/".
> The folder is created only using the option "./configure -mpi".
> In all cases I set the AMBERHOME correctly and I load the amber.sh script.
>
> Only -mpi option creates the python modules. Is this right?
>

C
​ompiling in serial creates the python modules as well, but you're right.
I hadn't thought about that.

I'll work on a fix.

Thanks,
Jason

-- 
Jason M. Swails
BioMaPS,
Rutgers University
Postdoctoral Researcher
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Tue Jun 02 2015 - 08:00:02 PDT
Custom Search