On 04/14/2011 12:59 PM, Jason Swails wrote:
> On Thu, Apr 14, 2011 at 9:54 AM, Tyler Luchko<tluchko.rci.rutgers.edu>wrote:
>
>>
>> On 2011-04-14, at 12:22 PM, Jason Swails wrote:
>>
>>> Has anyone tested the new changes to RISM configure? sander.RISM wasn't
>>> compiling for me on my intel machine (not sure the exact cause). This
>> isn't
>>> generally an issue since sander.RISM isn't actually released.
>>>
>>> Just wondering if anyone else had problems.
>>>
>>
>> How are you compiling it? Are you using the master branch or is this using
>> AT 1.5 on top of Amber 11? The master branch works fine for me and Amber 11
>> shouldn't build sander.RISM.
>>
>
> It was the master branch. I may have been careless in building. It had to
> do with missing MPI symbols in fftw-2. Thinking back, I may have forgotten
> to do a clean in the AmberTools directory prior to trying a parallel build
> in the Amber11 src directory, which could explain those errors.
>
> It was late last night and I didn't look into it too much. The email was
> hasty and more just to see if I was the only one with those issues (which I
> only sent out since AT 1.5 is already released now ;) ). In any case, I'm
> confident that it's PEBKAC. If there's a problem when I try again (more
> coherently), I'll report back.
Sorry this took so long but it is now fixed. FFTW MPI libraries weren't
being built. I'm not sure why I couldn't reproduce the problem but Dan
helped me to find it.
Tyler
_______________________________________________
AMBER-Developers mailing list
AMBER-Developers.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber-developers
Received on Tue Apr 19 2011 - 13:00:02 PDT
