Re: [AMBER] Getting error messages during testing the installation.

From: David A Case <case.biomaps.rutgers.edu>
Date: Mon, 10 Dec 2012 08:39:55 -0500

On Mon, Dec 10, 2012, Sanjib Paul wrote:

> $CUDA_HOME is properly set. It is
> $CUDA_HOME=/usr/local/cuda. And libcurand.so.4 is in the directories I
> specify by the LD_LIBRARY_PATH. libcurand.so.4 is in the
> /usr/local/cuda/lib64 and in /usr/local/cuda/lib. And I set the
> LD_LIBRARY_PATH as export
> LD_LIBRARY_PATH=$LD_LIBRARY_PATH:$CUDA_HOME/lib:$CUDA_HOME/lib64. I did
> these things before my first mail and got those errors. Now what should I
> do?

Others on the list may know flags that give more detailed error messages.
I'd certainly recompile a fresh version, and make sure that the exectable
image you are running is the one you intend to run. You could try making a
static executable, but that is probably going in the wrong direction. I don't
know what would be causing the error below if the shared library is indeed in
the LD_LIBRARY_PATH.

> > > >> 1) "......: error while loading shared libraries: libcurand.so.4:
> > cannot
> > > >> open shared object file: No such file or directory".

...dac


_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Mon Dec 10 2012 - 06:00:04 PST
Custom Search