2 * This file is part of the GROMACS molecular simulation package.
4 * Copyright (c) 2018,2019,2020, by the GROMACS development team, led by
5 * Mark Abraham, David van der Spoel, Berk Hess, and Erik Lindahl,
6 * and including many others, as listed in the AUTHORS file in the
7 * top-level source directory and at http://www.gromacs.org.
9 * GROMACS is free software; you can redistribute it and/or
10 * modify it under the terms of the GNU Lesser General Public License
11 * as published by the Free Software Foundation; either version 2.1
12 * of the License, or (at your option) any later version.
14 * GROMACS is distributed in the hope that it will be useful,
15 * but WITHOUT ANY WARRANTY; without even the implied warranty of
16 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
17 * Lesser General Public License for more details.
19 * You should have received a copy of the GNU Lesser General Public
20 * License along with GROMACS; if not, see
21 * http://www.gnu.org/licenses, or write to the Free Software Foundation,
22 * Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
24 * If you want to redistribute modifications to GROMACS, please
25 * consider that scientific software is very special. Version
26 * control is crucial - bugs must be traceable. We will be happy to
27 * consider code for inclusion in the official distribution, but
28 * derived work must not be called official GROMACS. Details are found
29 * in the README & COPYING files - if they are missing, get the
30 * official version at http://www.gromacs.org.
32 * To help us fund GROMACS development, we humbly ask that you cite
33 * the research papers on the package. Check out http://www.gromacs.org.
36 * \brief Defines the runner for CUDA version of SETTLE.
38 * \author Artem Zhmurov <zhmurov@gmail.com>
39 * \ingroup module_mdlib
43 #include "settletestrunners.h"
54 #include "gromacs/gpu_utils/devicebuffer.cuh"
55 #include "gromacs/gpu_utils/gpu_utils.h"
56 #include "gromacs/mdlib/settle_gpu.cuh"
57 #include "gromacs/utility/unique_cptr.h"
64 /*! \brief Apply SETTLE using GPU version of the algorithm.
66 * Initializes SETTLE object, copied data to the GPU, applies algorithm, copies the data back,
67 * destroys the object. The coordinates, velocities and virial are updated in the testData object.
69 * \param[in,out] testData An object, containing all the data structures needed by SETTLE.
70 * \param[in] pbc Periodic boundary setup.
71 * \param[in] updateVelocities If the velocities should be updated.
72 * \param[in] calcVirial If the virial should be computed.
73 * \param[in] testDescription Brief description that will be printed in case of test failure.
75 void applySettleGpu(SettleTestData* testData,
77 const bool updateVelocities,
78 const bool calcVirial,
79 gmx_unused const std::string& testDescription)
81 // These should never fail since this function should only be called if CUDA is enabled and
82 // there is a CUDA-capable device available.
83 GMX_RELEASE_ASSERT(GMX_GPU == GMX_GPU_CUDA,
84 "CUDA version of SETTLE was called from non-CUDA build.");
86 // TODO: Here we should check that at least 1 suitable GPU is available
87 GMX_RELEASE_ASSERT(canPerformGpuDetection(), "Can't detect CUDA-capable GPUs.");
89 DeviceInformation deviceInfo;
90 const DeviceContext deviceContext(deviceInfo);
91 const DeviceStream deviceStream(deviceContext, DeviceStreamPriority::Normal, false);
93 auto settleGpu = std::make_unique<SettleGpu>(testData->mtop_, deviceContext, deviceStream);
95 settleGpu->set(*testData->idef_, testData->mdatoms_);
97 setPbcAiuc(pbc.ndim_ePBC, pbc.box, &pbcAiuc);
99 int numAtoms = testData->mdatoms_.homenr;
101 float3 *d_x, *d_xp, *d_v;
103 float3* h_x = (float3*)(as_rvec_array(testData->x_.data()));
104 float3* h_xp = (float3*)(as_rvec_array(testData->xPrime_.data()));
105 float3* h_v = (float3*)(as_rvec_array(testData->v_.data()));
107 allocateDeviceBuffer(&d_x, numAtoms, deviceContext);
108 allocateDeviceBuffer(&d_xp, numAtoms, deviceContext);
109 allocateDeviceBuffer(&d_v, numAtoms, deviceContext);
111 copyToDeviceBuffer(&d_x, (float3*)h_x, 0, numAtoms, deviceStream, GpuApiCallBehavior::Sync, nullptr);
112 copyToDeviceBuffer(&d_xp, (float3*)h_xp, 0, numAtoms, deviceStream, GpuApiCallBehavior::Sync, nullptr);
113 if (updateVelocities)
115 copyToDeviceBuffer(&d_v, (float3*)h_v, 0, numAtoms, deviceStream, GpuApiCallBehavior::Sync, nullptr);
117 settleGpu->apply(d_x, d_xp, updateVelocities, d_v, testData->reciprocalTimeStep_, calcVirial,
118 testData->virial_, pbcAiuc);
120 copyFromDeviceBuffer((float3*)h_xp, &d_xp, 0, numAtoms, deviceStream, GpuApiCallBehavior::Sync, nullptr);
121 if (updateVelocities)
123 copyFromDeviceBuffer((float3*)h_v, &d_v, 0, numAtoms, deviceStream,
124 GpuApiCallBehavior::Sync, nullptr);
127 freeDeviceBuffer(&d_x);
128 freeDeviceBuffer(&d_xp);
129 freeDeviceBuffer(&d_v);