lammps

History

Steve Plimpton 236ebf7fab Kokkos lib update		2016-09-08 13:56:18 -06:00
..
01_hello_world	Kokkos lib update	2016-09-08 13:56:18 -06:00
01_hello_world_lambda	Kokkos lib update	2016-09-08 13:56:18 -06:00
02_simple_reduce	Kokkos lib update	2016-09-08 13:56:18 -06:00
02_simple_reduce_lambda	Kokkos lib update	2016-09-08 13:56:18 -06:00
03_simple_view	Kokkos lib update	2016-09-08 13:56:18 -06:00
03_simple_view_lambda	Kokkos lib update	2016-09-08 13:56:18 -06:00
04_simple_memoryspaces	Kokkos lib update	2016-09-08 13:56:18 -06:00
05_simple_atomics	Kokkos lib update	2016-09-08 13:56:18 -06:00
Advanced_Views	Kokkos lib update	2016-09-08 13:56:18 -06:00
Algorithms	Kokkos lib update	2016-09-08 13:56:18 -06:00
Hierarchical_Parallelism	Kokkos lib update	2016-09-08 13:56:18 -06:00
CMakeLists.txt	Updating kokkos lib	2016-05-02 22:10:37 +00:00
Makefile	Updating kokkos lib	2016-05-02 22:10:37 +00:00
README	Updating kokkos lib	2016-05-02 22:10:37 +00:00

README

Build the examples by typing in each directory: 
make -j 16

To specify a target device:
make openmp -j 16
make pthreads -j 16
make serial -j 16
make cuda -j 16

The lambda variants can not be build with CUDA=yes at the moment, since
CUDA does not support lambdas from the host. 
Some of the advanced topics try to highlight performance impacts by timing 
different variants of doing the same thing.
Also some of the advanced topics (in particular hierarchical parallelism)
require C++11 even with out using host side lambdas. CUDA 6.5 can be used 
to compile those.