targetDP
========

Copyright 2015 The University of Edinburgh

Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with the License. You may obtain a copy of the License at

http://www.apache.org/licenses/LICENSE-2.0

Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an "AS IS" BASIS, WITHOUT  ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License. 

About
----------

targetDP (target Data Parallel) is a lightweight programming abstraction layer, designed to allow the same application source code to be able to target multiple architectures, e.g. NVIDIA GPUs and multicore/manycore CPUs, in a performance portable manner.

See:

 Alan Gray and Kevin Stratford. “targetDP: an Abstraction of Lattice
 Based Parallelism with Portable Performance”. In: Proceedings of 2014
 IEEE International Conference on High Performance Computing and
 Communications (HPCC) (2014), pp. 312–315. doi:
 10.1109/HPCC.2014.212, http://arxiv.org/abs/1405.6162


Obtaining
----------

To obtain a copy of targetDP:
svn checkout http://ccpforge.cse.rl.ac.uk/svn/ludwig/trunk/targetDP

Note that this is the latest development version, and is subject to change.


Compiling the targetDP libraries
---------- 

to create the library libtarget.a:

Edit the Makefile to set CC to the desired compiler (and CFLAGS to the desired compiler flags). Then type 

```
make
```

If CC is a standard C/C++ compiler, then the CPU version of targetDP.a
will be created. If CC=nvcc, then the GPU version will be created.


Documentation/Specification
----------

See targetDPspec.pdf in the doc directory, and the example (below).


Example
----------

See simpleExample.c (which has compilation instructions for CPU and GPU near the top of the file).