2 * This file is part of the GROMACS molecular simulation package.
4 * Copyright (c) 2016,2017,2018, 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.
37 * \brief Implements CUDA FFT routines for PME GPU.
39 * \author Aleksei Iupinov <a.yupinov@gmail.com>
40 * \ingroup module_ewald
45 #include "pme-gpu-3dfft.h"
47 #include "gromacs/utility/fatalerror.h"
48 #include "gromacs/utility/gmxassert.h"
51 #include "pme-gpu-types.h"
52 #include "pme-gpu-types-host.h"
53 #include "pme-gpu-types-host-impl.h"
55 static void handleCufftError(cufftResult_t status, const char *msg)
57 if (status != CUFFT_SUCCESS)
59 gmx_fatal(FARGS, "%s (error code %d)\n", msg, status);
63 GpuParallel3dFft::GpuParallel3dFft(const PmeGpu *pmeGpu)
65 const PmeGpuCudaKernelParams *kernelParamsPtr = pmeGpu->kernelParams.get();
66 ivec realGridSize, realGridSizePadded, complexGridSizePadded;
67 for (int i = 0; i < DIM; i++)
69 realGridSize[i] = kernelParamsPtr->grid.realGridSize[i];
70 realGridSizePadded[i] = kernelParamsPtr->grid.realGridSizePadded[i];
71 complexGridSizePadded[i] = kernelParamsPtr->grid.complexGridSizePadded[i];
74 GMX_RELEASE_ASSERT(!pme_gpu_uses_dd(pmeGpu), "FFT decomposition not implemented");
76 const int complexGridSizePaddedTotal = complexGridSizePadded[XX] * complexGridSizePadded[YY] * complexGridSizePadded[ZZ];
77 const int realGridSizePaddedTotal = realGridSizePadded[XX] * realGridSizePadded[YY] * realGridSizePadded[ZZ];
79 realGrid_ = (cufftReal *)kernelParamsPtr->grid.d_realGrid;
80 GMX_RELEASE_ASSERT(realGrid_, "Bad (null) input real-space grid");
81 complexGrid_ = (cufftComplex *)kernelParamsPtr->grid.d_fourierGrid;
82 GMX_RELEASE_ASSERT(complexGrid_, "Bad (null) input complex grid");
85 /* Commented code for a simple 3D grid with no padding */
87 result = cufftPlan3d(&planR2C_, realGridSize[XX], realGridSize[YY], realGridSize[ZZ], CUFFT_R2C);
88 handleCufftError(result, "cufftPlan3d R2C plan failure");
90 result = cufftPlan3d(&planC2R_, realGridSize[XX], realGridSize[YY], realGridSize[ZZ], CUFFT_C2R);
91 handleCufftError(result, "cufftPlan3d C2R plan failure");
94 const int rank = 3, batch = 1;
95 result = cufftPlanMany(&planR2C_, rank, realGridSize,
96 realGridSizePadded, 1, realGridSizePaddedTotal,
97 complexGridSizePadded, 1, complexGridSizePaddedTotal,
100 handleCufftError(result, "cufftPlanMany R2C plan failure");
102 result = cufftPlanMany(&planC2R_, rank, realGridSize,
103 complexGridSizePadded, 1, complexGridSizePaddedTotal,
104 realGridSizePadded, 1, realGridSizePaddedTotal,
107 handleCufftError(result, "cufftPlanMany C2R plan failure");
109 cudaStream_t stream = pmeGpu->archSpecific->pmeStream;
110 GMX_RELEASE_ASSERT(stream, "Using the default CUDA stream for PME cuFFT");
112 result = cufftSetStream(planR2C_, stream);
113 handleCufftError(result, "cufftSetStream R2C failure");
115 result = cufftSetStream(planC2R_, stream);
116 handleCufftError(result, "cufftSetStream C2R failure");
119 GpuParallel3dFft::~GpuParallel3dFft()
121 cufftResult_t result;
122 result = cufftDestroy(planR2C_);
123 handleCufftError(result, "cufftDestroy R2C failure");
124 result = cufftDestroy(planC2R_);
125 handleCufftError(result, "cufftDestroy C2R failure");
128 void GpuParallel3dFft::perform3dFft(gmx_fft_direction dir,
129 CommandEvent */*timingEvent*/)
131 cufftResult_t result;
132 if (dir == GMX_FFT_REAL_TO_COMPLEX)
134 result = cufftExecR2C(planR2C_, realGrid_, complexGrid_);
135 handleCufftError(result, "cuFFT R2C execution failure");
139 result = cufftExecC2R(planC2R_, complexGrid_, realGrid_);
140 handleCufftError(result, "cuFFT C2R execution failure");