Many thanks for the quick response (as always!)
Removing $AMBERHOME/include/netcdf.mod (manually) indeed did the trick.
BTW, Jason's suggestion that would allow ./configure bypass the NetCDF
configure altogether if the existing NetCDF 'works', sounds great.
My incompatible netcdf.mod was indeed likely due to using different
compilers previously, but I think there may be good reasons to use
different compilers for different executables, e.g. sander.MPI and
pmemd.cuda.
Thanks,
Marc
On 12 July 2013 14:02, David A Case <case.biomaps.rutgers.edu> wrote:
> On Fri, Jul 12, 2013, Marc van der Kamp wrote:
>>
>> On a (linux) system where I previously successfully compiled
>> pmemd.cuda_SPFP (from amber12), I cannot re-compile it (after
>> updates).
>> I'm using cuda-4.2.9 (this is for Tesla M2090's).
>>
>> "configure -cuda gnu" is successful, but when I "make install", I'm
>> running into the following error:
>>
>> gfortran -DBINTRAJ -DDIRFRC_EFS -DDIRFRC_COMTRANS -DDIRFRC_NOVEC
>> -DFFTLOADBAL_2PROC -DPUBFFT -O3 -DCUDA -Duse_SPFP
>> -I/gpfs/cluster/chem/chmwvdk/amber12/include -c binrestart.F90
>> Fatal Error: Reading module netcdf at line 23 column 34: Expected integer
>
> It sounds like the netcdf module was earlier compiled with a different
> compiler (or different compiler version). In any event, remove
> $AMBERHOME/include/netcdf.mod, than try the compilation again.
>
> [Developers, esp. Jason: is there any reason that "make clean" (run
> automagically by configure) should not be doing this?]
>
> ...hope this helps...dac
>
>
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Fri Jul 12 2013 - 07:00:03 PDT
