Kokkos Node API and Local Linear Algebra Kernels Version of the Day
Kokkos_ThrustGPUNode.cpp
00001 //@HEADER
00002 // ************************************************************************
00003 // 
00004 //          Kokkos: Node API and Parallel Node Kernels
00005 //              Copyright (2008) Sandia Corporation
00006 // 
00007 // Under the terms of Contract DE-AC04-94AL85000 with Sandia Corporation,
00008 // the U.S. Government retains certain rights in this software.
00009 // 
00010 // Redistribution and use in source and binary forms, with or without
00011 // modification, are permitted provided that the following conditions are
00012 // met:
00013 //
00014 // 1. Redistributions of source code must retain the above copyright
00015 // notice, this list of conditions and the following disclaimer.
00016 //
00017 // 2. Redistributions in binary form must reproduce the above copyright
00018 // notice, this list of conditions and the following disclaimer in the
00019 // documentation and/or other materials provided with the distribution.
00020 //
00021 // 3. Neither the name of the Corporation nor the names of the
00022 // contributors may be used to endorse or promote products derived from
00023 // this software without specific prior written permission.
00024 //
00025 // THIS SOFTWARE IS PROVIDED BY SANDIA CORPORATION "AS IS" AND ANY
00026 // EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
00027 // IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
00028 // PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL SANDIA CORPORATION OR THE
00029 // CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
00030 // EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
00031 // PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
00032 // PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF
00033 // LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING
00034 // NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
00035 // SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
00036 //
00037 // Questions? Contact Michael A. Heroux (maherou@sandia.gov) 
00038 // 
00039 // ************************************************************************
00040 //@HEADER
00041 
00042 #include "Kokkos_ThrustGPUNode.hpp"
00043 #include <Teuchos_Assert.hpp>
00044 #include <iostream>
00045 #include <cuda_runtime.h>
00046 
00047 namespace Kokkos {
00048 
00049   ThrustGPUNode::ThrustGPUNode(Teuchos::ParameterList &pl)
00050   {
00051     using std::cout;
00052     using std::cerr;
00053     using std::endl;
00054 
00055     // get node parameters
00056     int device = pl.get<int>("Device Number",0);
00057     int verbose = pl.get<int>("Verbose",0);
00058     // set device
00059     int deviceCount; cudaGetDeviceCount(&deviceCount); 
00060     TEUCHOS_TEST_FOR_EXCEPTION(deviceCount == 0, std::runtime_error,
00061         "ThrustGPUNode::ThrustGPUNode(): system has no CUDA devices.");
00062     if (device < 0 || device >= deviceCount) {
00063       cerr << "ThrustGPUNode::ThrustGPUNode(): specified device number not valid. Using device 0." << endl;
00064       device = 0;
00065     }
00066     cudaDeviceProp deviceProp; 
00067     cudaSetDevice(device);
00068     cudaGetDeviceProperties(&deviceProp, device); 
00069     // as of CUDA 2.1, device prop contains the following fields
00070     // char name[256]; 
00071     // size_t totalGlobalMem, sharedMemPerBlock; 
00072     // int regsPerBlock, warpSize; 
00073     // size_t memPitch; 
00074     // int maxThreadsPerBlock, maxThreadsDim[3], maxGridSize[3]; 
00075     // size_t totalConstMem; 
00076     // int major, minor;
00077     // int clockRate; 
00078     // size_t textureAlignment; 
00079     // int deviceOverlap; 
00080     // int multiProcessorCount; 
00081     // int kernelExecTimeoutEnabled; 
00082     if (verbose) {
00083       cout << "ThrustGPUNode attached to device #" << device << " \"" << deviceProp.name 
00084         << "\", of compute capability " << deviceProp.major << "." << deviceProp.minor
00085         << endl;
00086     }
00087     totalMem_ = deviceProp.totalGlobalMem;
00088   } 
00089 
00090   ThrustGPUNode::~ThrustGPUNode() {}
00091 
00092   void ThrustGPUNode::sync() const {
00093     cudaError err = cudaThreadSynchronize();
00094     TEUCHOS_TEST_FOR_EXCEPTION( cudaSuccess != err, std::runtime_error,
00095         "Kokkos::ThrustGPUNode::sync(): cudaThreadSynchronize() returned error " << err );
00096   }
00097 
00098 }
 All Classes Namespaces Files Functions Variables Typedefs Enumerations Enumerator Friends