Dear Sir,
I have get the copy of LAM-MPS for GPU on , windows as well as Linux, from svn checkout http://gpulammps.googlecode.com/svn/trunk/gpulammps-read-only``
``
lammps-3Feb10]$ ls lib/gpu/
gb_gpu.cu gb_gpu_memory.h lj_gpu_memory.h nvc_get_devices.cu nvc_traits.h pair_gpu_nbor.h
gb_gpu_extra.h lj_gpu.cu Makefile.nvidia nvc_macros.h pair_gpu_atom.cu pair_gpu_texture.h
gb_gpu_kernel.h lj_gpu_kernel.h nvc_device.cu nvc_memory.h pair_gpu_atom.h pair_tex_tar.cu
gb_gpu_memory.cu lj_gpu_memory.cu nvc_device.h nvc_timer.h pair_gpu_nbor.cu README
But I am not able to get an overall idea to compile LAM-MPS with CUDA libraries.
Can you please write me the whole procedure to compile LAMMPS on GPU.
Thanks & Regards,
Shraddha Desai
Dear Sir,
I have get the copy of LAM\-MPS for GPU on , windows as well as
Linux, from svn checkout
http://gpulammps.googlecode.com/svn/trunk/gpulammps-read-only
lammps-3Feb10]$ ls lib/gpu/
gb_gpu.cu gb_gpu_memory.h lj_gpu_memory.h nvc_get_devices.cu
nvc_traits.h pair_gpu_nbor.h
gb_gpu_extra.h lj_gpu.cu Makefile.nvidia nvc_macros.h
pair_gpu_atom.cu pair_gpu_texture.h
gb_gpu_kernel.h lj_gpu_kernel.h nvc_device.cu nvc_memory.h
pair_gpu_atom.h pair_tex_tar.cu
gb_gpu_memory.cu lj_gpu_memory.cu nvc_device.h nvc_timer.h
pair_gpu_nbor.cu README
But I am not able to get an overall idea to compile LAM\-MPS with CUDA
libraries.
disregard this checkout if you want to use the GPU package.
this part of the gpulammps repository is outdated.
the gpulammps svn contains the up-to-date version of the USER-CUDA
package, an alternate approach to GPU acceleration in lammps.
you can find the up-to-date version of the GPU in the standard lammps
distribution.
Can you please write me the whole procedure to compile LAMMPS on GPU.
the instructions to compile lammps with the GPU package are contained
in the lammps online documentation.
the corresponding instructions for the USER-CUDA package are in the
gpulammps wiki.
cheers,
axel.