Re: [AMBER] problem in running AMBER12 in GPU

From: Sanjib Paul <sanjib88paul.gmail.com>
Date: Wed, 11 Sep 2013 18:58:26 +0530

Hello,
         I have downloaded a file named 'mpich2-1.5-3.fc20.x86_64.rpm',
extracted it. And then kept the extracted file 'mpich2-1.5-3.fc20.x86_64'
in $AMBERHOME/AmberTools/src folder. Then run the script configure_mpich2.

./configure_mpich2 gnu

Setting AMBERHOME to /home/software/AMBER12/amber12

./configure_mpich2: line 141: ./configure: No such file or directory
    MPICH2 configure failed, returning 127

I am getting the above error. I have searched google, but did not get any
problem or answer regarding this. So, please help.

Sanjib




On Wed, Sep 4, 2013 at 8:26 PM, Jason Swails <jason.swails.gmail.com> wrote:

> On Wed, Sep 4, 2013 at 10:21 AM, Sanjib Paul <sanjib88paul.gmail.com>
> wrote:
>
> > Hello,
> > Thanks for your valuable suggestion. After updating we are able
> to
> > run AMBER12 (sander, pmemd, sander.MPI & pmemd.MPI, pmemd.cuda). But
> > unfortunately when we are not able to install pmemd.cuda.MPI
> successfully.
> > Those are few following errors which we are getting at the end of
> > installation process.
> >
> > ./cuda/cuda.a(gpu.o): In function `MPI::Intracomm::Clone() const':
> > gpu.cpp:(.text._ZNK3MPI9Intracomm5CloneEv[MPI::Intracomm::Clone()
> > const]+0x27): undefined reference to `MPI::Comm::Comm()'
> >
> >
> ./cuda/cuda.a(gpu.o):gpu.cpp:(.text._ZNK3MPI9Intracomm5SplitEii[MPI::Intracomm::Split(int,
> > int) const]+0x24): more undefined references to `MPI::Comm::Comm()'
> follow
> > ./cuda/cuda.a(gpu.o):(.rodata._ZTVN3MPI3WinE[vtable for MPI::Win]+0x48):
> > undefined reference to `MPI::Win::Free()'
> > ./cuda/cuda.a(gpu.o):(.rodata._ZTVN3MPI8DatatypeE[vtable for
> > MPI::Datatype]+0x78): undefined reference to `MPI::Datatype::Free()'
> > collect2: ld returned 1 exit status
> > make[3]: *** [pmemd.cuda.MPI] Error 1
> > make[3]: Leaving directory `/home/test/amber12/src/pmemd/src'
> > make[2]: *** [cuda_parallel] Error 2
> > make[2]: Leaving directory `/home/test/amber12/src/pmemd'
> > make[1]: *** [cuda_parallel] Error 2
> > make[1]: Leaving directory `/home/test/amber12/src'
> > make: *** [install] Error 2
> >
> > We are using cuda version 5.5 with openmpi 1.7.2 built with installed
> cuda.
> > We did not understand what the problem is. Please give some suggestion.
> >
>
> It's likely that your MPI was not built with the appropriate support that
> pmemd.cuda.MPI needs (make sure the MPI is built with C++ support, I
> think). I've been using mpich2 (now just 'mpich') for some time and I've
> never had a problem building pmemd.cuda.MPI before. Note, if you google
> the last little bit of your error message:
>
> ./cuda/cuda.a(gpu.o):(.rodata._ZTVN3MPI8DatatypeE[vtable for
> MPI::Datatype]+0x78): undefined reference to `MPI::Datatype::Free()'
>
> it brings you to this link: http://archive.ambermd.org/201106/0678.html
>
> which provides useful things to try.
>
> Try adding -lmpi_cxx to the PMEMD_CU_LIBS line in config.h. If that still
> doesn't work, rebuild your MPI with C++ support. If you are unsure what to
> do, use the configure_mpich2 script in $AMBERHOME/AmberTools/src to build
> your own MPI (see the manual for instructions -- you need to download
> mpich2 first to use it). You also need to set up PATH and LD_LIBRARY_PATH
> so that "which mpif90" returns the version in $AMBERHOME/bin.
>
> HTH,
> Jason
>
> --
> Jason M. Swails
> BioMaPS,
> Rutgers University
> Postdoctoral Researcher
> _______________________________________________
> 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 Wed Sep 11 2013 - 06:30:04 PDT
Custom Search