2 * This file is part of the GROMACS molecular simulation package.
4 * Copyright (c) 2019, 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 Definitions of interfaces for GPU state data propagator object.
39 * \author Artem Zhmurov <zhmurov@gmail.com>
41 * \ingroup module_mdtypes
47 #if GMX_GPU != GMX_GPU_NONE
49 # if GMX_GPU == GMX_GPU_CUDA
50 # include "gromacs/gpu_utils/cudautils.cuh"
52 # include "gromacs/gpu_utils/devicebuffer.h"
53 # if GMX_GPU == GMX_GPU_OPENCL
54 # include "gromacs/gpu_utils/oclutils.h"
56 # include "gromacs/math/vectypes.h"
57 # include "gromacs/mdtypes/state_propagator_data_gpu.h"
58 # include "gromacs/utility/classhelpers.h"
60 # include "state_propagator_data_gpu_impl.h"
65 StatePropagatorDataGpu::Impl::Impl(const void* pmeStream,
66 const void* localStream,
67 const void* nonLocalStream,
68 const void* deviceContext,
69 GpuApiCallBehavior transferKind,
71 transferKind_(transferKind),
72 paddingSize_(paddingSize)
74 static_assert(GMX_GPU != GMX_GPU_NONE,
75 "This object should only be constructed on the GPU code-paths.");
77 // TODO: Refactor when the StreamManager is introduced.
78 if (GMX_GPU == GMX_GPU_OPENCL)
80 GMX_ASSERT(deviceContext != nullptr, "GPU context should be set in OpenCL builds.");
81 GMX_ASSERT(pmeStream != nullptr, "GPU PME stream should be set in OpenCL builds.");
83 // The update stream is set to the PME stream in OpenCL, since PME stream is the only stream created in the PME context.
84 pmeStream_ = *static_cast<const CommandStream*>(pmeStream);
85 updateStream_ = *static_cast<const CommandStream*>(pmeStream);
86 deviceContext_ = *static_cast<const DeviceContext*>(deviceContext);
87 GMX_UNUSED_VALUE(localStream);
88 GMX_UNUSED_VALUE(nonLocalStream);
91 if (GMX_GPU == GMX_GPU_CUDA)
93 if (pmeStream != nullptr)
95 pmeStream_ = *static_cast<const CommandStream*>(pmeStream);
97 if (localStream != nullptr)
99 localStream_ = *static_cast<const CommandStream*>(localStream);
101 if (nonLocalStream != nullptr)
103 nonLocalStream_ = *static_cast<const CommandStream*>(nonLocalStream);
106 // TODO: The update stream should be created only when it is needed.
107 # if (GMX_GPU == GMX_GPU_CUDA)
109 stat = cudaStreamCreate(&updateStream_);
110 CU_RET_ERR(stat, "CUDA stream creation failed in StatePropagatorDataGpu");
112 GMX_UNUSED_VALUE(deviceContext);
115 // Map the atom locality to the stream that will be used for coordinates,
116 // velocities and forces transfers. Same streams are used for H2D and D2H copies.
117 // Note, that nullptr stream is used here to indicate that the copy is not supported.
118 xCopyStreams_[AtomLocality::Local] = updateStream_;
119 xCopyStreams_[AtomLocality::NonLocal] = nonLocalStream_;
120 xCopyStreams_[AtomLocality::All] = updateStream_;
122 vCopyStreams_[AtomLocality::Local] = updateStream_;
123 vCopyStreams_[AtomLocality::NonLocal] = nullptr;
124 vCopyStreams_[AtomLocality::All] = updateStream_;
126 fCopyStreams_[AtomLocality::Local] = localStream_;
127 fCopyStreams_[AtomLocality::NonLocal] = nonLocalStream_;
128 fCopyStreams_[AtomLocality::All] = updateStream_;
131 StatePropagatorDataGpu::Impl::Impl(const void* pmeStream,
132 const void* deviceContext,
133 GpuApiCallBehavior transferKind,
135 transferKind_(transferKind),
136 paddingSize_(paddingSize)
138 static_assert(GMX_GPU != GMX_GPU_NONE,
139 "This object should only be constructed on the GPU code-paths.");
141 if (GMX_GPU == GMX_GPU_OPENCL)
143 GMX_ASSERT(deviceContext != nullptr, "GPU context should be set in OpenCL builds.");
144 deviceContext_ = *static_cast<const DeviceContext*>(deviceContext);
147 GMX_ASSERT(pmeStream != nullptr, "GPU PME stream should be set.");
148 pmeStream_ = *static_cast<const CommandStream*>(pmeStream);
150 localStream_ = nullptr;
151 nonLocalStream_ = nullptr;
152 updateStream_ = nullptr;
155 // Only local/all coordinates are allowed to be copied in PME-only rank/ PME tests.
156 // This it temporary measure to make it safe to use this class in those cases.
157 xCopyStreams_[AtomLocality::Local] = pmeStream_;
158 xCopyStreams_[AtomLocality::NonLocal] = nullptr;
159 xCopyStreams_[AtomLocality::All] = pmeStream_;
161 vCopyStreams_[AtomLocality::Local] = nullptr;
162 vCopyStreams_[AtomLocality::NonLocal] = nullptr;
163 vCopyStreams_[AtomLocality::All] = nullptr;
165 fCopyStreams_[AtomLocality::Local] = nullptr;
166 fCopyStreams_[AtomLocality::NonLocal] = nullptr;
167 fCopyStreams_[AtomLocality::All] = nullptr;
170 StatePropagatorDataGpu::Impl::~Impl() {}
172 void StatePropagatorDataGpu::Impl::reinit(int numAtomsLocal, int numAtomsAll)
174 numAtomsLocal_ = numAtomsLocal;
175 numAtomsAll_ = numAtomsAll;
178 if (paddingSize_ > 0)
180 numAtomsPadded = ((numAtomsAll_ + paddingSize_ - 1) / paddingSize_) * paddingSize_;
184 numAtomsPadded = numAtomsAll_;
187 reallocateDeviceBuffer(&d_x_, DIM * numAtomsPadded, &d_xSize_, &d_xCapacity_, deviceContext_);
189 const size_t paddingAllocationSize = numAtomsPadded - numAtomsAll_;
190 if (paddingAllocationSize > 0)
192 // The PME stream is used here because the padding region of d_x_ is only in the PME task.
193 clearDeviceBufferAsync(&d_x_, DIM * numAtomsAll_, DIM * paddingAllocationSize, pmeStream_);
196 reallocateDeviceBuffer(&d_v_, DIM * numAtomsAll_, &d_vSize_, &d_vCapacity_, deviceContext_);
197 reallocateDeviceBuffer(&d_f_, DIM * numAtomsAll_, &d_fSize_, &d_fCapacity_, deviceContext_);
200 std::tuple<int, int> StatePropagatorDataGpu::Impl::getAtomRangesFromAtomLocality(AtomLocality atomLocality)
202 int atomsStartAt = 0;
203 int numAtomsToCopy = 0;
204 switch (atomLocality)
206 case AtomLocality::All:
208 numAtomsToCopy = numAtomsAll_;
210 case AtomLocality::Local:
212 numAtomsToCopy = numAtomsLocal_;
214 case AtomLocality::NonLocal:
215 atomsStartAt = numAtomsLocal_;
216 numAtomsToCopy = numAtomsAll_ - numAtomsLocal_;
219 GMX_RELEASE_ASSERT(false,
220 "Wrong range of atoms requested in GPU state data manager. Should "
221 "be All, Local or NonLocal.");
223 GMX_ASSERT(atomsStartAt >= 0,
224 "The first elemtnt to copy has negative index. Probably, the GPU propagator state "
225 "was not initialized.");
226 GMX_ASSERT(numAtomsToCopy >= 0,
227 "Number of atoms to copy is negative. Probably, the GPU propagator state was not "
229 return std::make_tuple(atomsStartAt, numAtomsToCopy);
232 void StatePropagatorDataGpu::Impl::copyToDevice(DeviceBuffer<float> d_data,
233 const gmx::ArrayRef<const gmx::RVec> h_data,
235 AtomLocality atomLocality,
236 CommandStream commandStream)
239 GMX_UNUSED_VALUE(dataSize);
241 GMX_ASSERT(dataSize >= 0, "Trying to copy to device buffer before it was allocated.");
243 int atomsStartAt, numAtomsToCopy;
244 std::tie(atomsStartAt, numAtomsToCopy) = getAtomRangesFromAtomLocality(atomLocality);
246 int elementsStartAt = atomsStartAt * DIM;
247 int numElementsToCopy = numAtomsToCopy * DIM;
249 if (numAtomsToCopy != 0)
251 GMX_ASSERT(elementsStartAt + numElementsToCopy <= dataSize,
252 "The device allocation is smaller than requested copy range.");
253 GMX_ASSERT(atomsStartAt + numAtomsToCopy <= h_data.ssize(),
254 "The host buffer is smaller than the requested copy range.");
256 copyToDeviceBuffer(&d_data, reinterpret_cast<const float*>(&h_data.data()[atomsStartAt]),
257 elementsStartAt, numElementsToCopy, commandStream, transferKind_, nullptr);
261 void StatePropagatorDataGpu::Impl::copyFromDevice(gmx::ArrayRef<gmx::RVec> h_data,
262 DeviceBuffer<float> d_data,
264 AtomLocality atomLocality,
265 CommandStream commandStream)
268 GMX_UNUSED_VALUE(dataSize);
270 GMX_ASSERT(dataSize >= 0, "Trying to copy from device buffer before it was allocated.");
272 int atomsStartAt, numAtomsToCopy;
273 std::tie(atomsStartAt, numAtomsToCopy) = getAtomRangesFromAtomLocality(atomLocality);
275 int elementsStartAt = atomsStartAt * DIM;
276 int numElementsToCopy = numAtomsToCopy * DIM;
278 if (numAtomsToCopy != 0)
280 GMX_ASSERT(elementsStartAt + numElementsToCopy <= dataSize,
281 "The device allocation is smaller than requested copy range.");
282 GMX_ASSERT(atomsStartAt + numAtomsToCopy <= h_data.ssize(),
283 "The host buffer is smaller than the requested copy range.");
285 copyFromDeviceBuffer(reinterpret_cast<float*>(&h_data.data()[atomsStartAt]), &d_data,
286 elementsStartAt, numElementsToCopy, commandStream, transferKind_, nullptr);
290 DeviceBuffer<float> StatePropagatorDataGpu::Impl::getCoordinates()
295 void StatePropagatorDataGpu::Impl::copyCoordinatesToGpu(const gmx::ArrayRef<const gmx::RVec> h_x,
296 AtomLocality atomLocality)
298 GMX_ASSERT(atomLocality < AtomLocality::Count, "Wrong atom locality.");
299 CommandStream commandStream = xCopyStreams_[atomLocality];
300 GMX_ASSERT(commandStream != nullptr,
301 "No stream is valid for copying positions with given atom locality.");
303 copyToDevice(d_x_, h_x, d_xSize_, atomLocality, commandStream);
305 // markEvent is skipped in OpenCL as:
306 // - it's not needed, copy is done in the same stream as the only consumer task (PME)
307 // - we don't consume the events in OpenCL which is not allowed by GpuEventSynchronizer (would leak memory).
308 // TODO: remove this by adding an event-mark free flavor of this function
309 if (GMX_GPU == GMX_GPU_CUDA)
311 xReadyOnDevice_[atomLocality].markEvent(commandStream);
315 GpuEventSynchronizer*
316 StatePropagatorDataGpu::Impl::getCoordinatesReadyOnDeviceEvent(AtomLocality atomLocality,
317 const SimulationWorkload& simulationWork,
318 const StepWorkload& stepWork)
320 // The provider of the coordinates may be different for local atoms. If the update is offloaded
321 // and this is not a neighbor search step, then the consumer needs to wait for the update
322 // to complete. Otherwise, the coordinates are copied from the host and we need to wait for
323 // the copy event. Non-local coordinates are always provided by the H2D copy.
325 // TODO: This should be reconsidered to support the halo exchange.
327 // In OpenCL no events are used as coordinate sync is not necessary
328 if (GMX_GPU == GMX_GPU_OPENCL)
332 if (atomLocality == AtomLocality::Local && simulationWork.useGpuUpdate && !stepWork.doNeighborSearch)
334 return &xUpdatedOnDevice_;
338 return &xReadyOnDevice_[atomLocality];
342 void StatePropagatorDataGpu::Impl::waitCoordinatesCopiedToDevice(AtomLocality atomLocality)
344 GMX_ASSERT(atomLocality < AtomLocality::Count, "Wrong atom locality.");
345 xReadyOnDevice_[atomLocality].waitForEvent();
348 GpuEventSynchronizer* StatePropagatorDataGpu::Impl::xUpdatedOnDevice()
350 return &xUpdatedOnDevice_;
353 void StatePropagatorDataGpu::Impl::copyCoordinatesFromGpu(gmx::ArrayRef<gmx::RVec> h_x, AtomLocality atomLocality)
355 GMX_ASSERT(atomLocality < AtomLocality::Count, "Wrong atom locality.");
356 CommandStream commandStream = xCopyStreams_[atomLocality];
357 GMX_ASSERT(commandStream != nullptr,
358 "No stream is valid for copying positions with given atom locality.");
360 copyFromDevice(h_x, d_x_, d_xSize_, atomLocality, commandStream);
361 // Note: unlike copyCoordinatesToGpu this is not used in OpenCL, and the conditional is not needed.
362 xReadyOnHost_[atomLocality].markEvent(commandStream);
365 void StatePropagatorDataGpu::Impl::waitCoordinatesReadyOnHost(AtomLocality atomLocality)
367 xReadyOnHost_[atomLocality].waitForEvent();
371 DeviceBuffer<float> StatePropagatorDataGpu::Impl::getVelocities()
376 void StatePropagatorDataGpu::Impl::copyVelocitiesToGpu(const gmx::ArrayRef<const gmx::RVec> h_v,
377 AtomLocality atomLocality)
379 GMX_ASSERT(atomLocality < AtomLocality::Count, "Wrong atom locality.");
380 CommandStream commandStream = vCopyStreams_[atomLocality];
381 GMX_ASSERT(commandStream != nullptr,
382 "No stream is valid for copying velocities with given atom locality.");
384 copyToDevice(d_v_, h_v, d_vSize_, atomLocality, commandStream);
385 vReadyOnDevice_[atomLocality].markEvent(commandStream);
388 GpuEventSynchronizer* StatePropagatorDataGpu::Impl::getVelocitiesReadyOnDeviceEvent(AtomLocality atomLocality)
390 return &vReadyOnDevice_[atomLocality];
394 void StatePropagatorDataGpu::Impl::copyVelocitiesFromGpu(gmx::ArrayRef<gmx::RVec> h_v, AtomLocality atomLocality)
396 GMX_ASSERT(atomLocality < AtomLocality::Count, "Wrong atom locality.");
397 CommandStream commandStream = vCopyStreams_[atomLocality];
398 GMX_ASSERT(commandStream != nullptr,
399 "No stream is valid for copying velocities with given atom locality.");
401 copyFromDevice(h_v, d_v_, d_vSize_, atomLocality, commandStream);
402 vReadyOnHost_[atomLocality].markEvent(commandStream);
405 void StatePropagatorDataGpu::Impl::waitVelocitiesReadyOnHost(AtomLocality atomLocality)
407 vReadyOnHost_[atomLocality].waitForEvent();
411 DeviceBuffer<float> StatePropagatorDataGpu::Impl::getForces()
416 void StatePropagatorDataGpu::Impl::copyForcesToGpu(const gmx::ArrayRef<const gmx::RVec> h_f,
417 AtomLocality atomLocality)
419 GMX_ASSERT(atomLocality < AtomLocality::Count, "Wrong atom locality.");
420 CommandStream commandStream = fCopyStreams_[atomLocality];
421 GMX_ASSERT(commandStream != nullptr,
422 "No stream is valid for copying forces with given atom locality.");
424 copyToDevice(d_f_, h_f, d_fSize_, atomLocality, commandStream);
425 fReadyOnDevice_[atomLocality].markEvent(commandStream);
428 GpuEventSynchronizer* StatePropagatorDataGpu::Impl::getForcesReadyOnDeviceEvent(AtomLocality atomLocality,
429 bool useGpuFBufferOps)
431 if ((atomLocality == AtomLocality::Local || atomLocality == AtomLocality::NonLocal) && useGpuFBufferOps)
433 return &fReducedOnDevice_;
437 return &fReadyOnDevice_[atomLocality];
441 GpuEventSynchronizer* StatePropagatorDataGpu::Impl::fReducedOnDevice()
443 return &fReducedOnDevice_;
446 void StatePropagatorDataGpu::Impl::copyForcesFromGpu(gmx::ArrayRef<gmx::RVec> h_f, AtomLocality atomLocality)
448 GMX_ASSERT(atomLocality < AtomLocality::Count, "Wrong atom locality.");
449 CommandStream commandStream = fCopyStreams_[atomLocality];
450 GMX_ASSERT(commandStream != nullptr,
451 "No stream is valid for copying forces with given atom locality.");
453 copyFromDevice(h_f, d_f_, d_fSize_, atomLocality, commandStream);
454 fReadyOnHost_[atomLocality].markEvent(commandStream);
457 void StatePropagatorDataGpu::Impl::waitForcesReadyOnHost(AtomLocality atomLocality)
459 fReadyOnHost_[atomLocality].waitForEvent();
462 void* StatePropagatorDataGpu::Impl::getUpdateStream()
464 return &updateStream_;
467 int StatePropagatorDataGpu::Impl::numAtomsLocal()
469 return numAtomsLocal_;
472 int StatePropagatorDataGpu::Impl::numAtomsAll()
478 StatePropagatorDataGpu::StatePropagatorDataGpu(const void* pmeStream,
479 const void* localStream,
480 const void* nonLocalStream,
481 const void* deviceContext,
482 GpuApiCallBehavior transferKind,
484 impl_(new Impl(pmeStream, localStream, nonLocalStream, deviceContext, transferKind, paddingSize))
488 StatePropagatorDataGpu::StatePropagatorDataGpu(const void* pmeStream,
489 const void* deviceContext,
490 GpuApiCallBehavior transferKind,
492 impl_(new Impl(pmeStream, deviceContext, transferKind, paddingSize))
496 StatePropagatorDataGpu::StatePropagatorDataGpu(StatePropagatorDataGpu&& /* other */) noexcept = default;
498 StatePropagatorDataGpu& StatePropagatorDataGpu::operator=(StatePropagatorDataGpu&& /* other */) noexcept = default;
500 StatePropagatorDataGpu::~StatePropagatorDataGpu() = default;
503 void StatePropagatorDataGpu::reinit(int numAtomsLocal, int numAtomsAll)
505 return impl_->reinit(numAtomsLocal, numAtomsAll);
508 std::tuple<int, int> StatePropagatorDataGpu::getAtomRangesFromAtomLocality(AtomLocality atomLocality)
510 return impl_->getAtomRangesFromAtomLocality(atomLocality);
514 DeviceBuffer<float> StatePropagatorDataGpu::getCoordinates()
516 return impl_->getCoordinates();
519 void StatePropagatorDataGpu::copyCoordinatesToGpu(const gmx::ArrayRef<const gmx::RVec> h_x,
520 AtomLocality atomLocality)
522 return impl_->copyCoordinatesToGpu(h_x, atomLocality);
525 GpuEventSynchronizer*
526 StatePropagatorDataGpu::getCoordinatesReadyOnDeviceEvent(AtomLocality atomLocality,
527 const SimulationWorkload& simulationWork,
528 const StepWorkload& stepWork)
530 return impl_->getCoordinatesReadyOnDeviceEvent(atomLocality, simulationWork, stepWork);
533 void StatePropagatorDataGpu::waitCoordinatesCopiedToDevice(AtomLocality atomLocality)
535 return impl_->waitCoordinatesCopiedToDevice(atomLocality);
538 GpuEventSynchronizer* StatePropagatorDataGpu::xUpdatedOnDevice()
540 return impl_->xUpdatedOnDevice();
543 void StatePropagatorDataGpu::copyCoordinatesFromGpu(gmx::ArrayRef<RVec> h_x, AtomLocality atomLocality)
545 return impl_->copyCoordinatesFromGpu(h_x, atomLocality);
548 void StatePropagatorDataGpu::waitCoordinatesReadyOnHost(AtomLocality atomLocality)
550 return impl_->waitCoordinatesReadyOnHost(atomLocality);
554 DeviceBuffer<float> StatePropagatorDataGpu::getVelocities()
556 return impl_->getVelocities();
559 void StatePropagatorDataGpu::copyVelocitiesToGpu(const gmx::ArrayRef<const gmx::RVec> h_v,
560 AtomLocality atomLocality)
562 return impl_->copyVelocitiesToGpu(h_v, atomLocality);
565 GpuEventSynchronizer* StatePropagatorDataGpu::getVelocitiesReadyOnDeviceEvent(AtomLocality atomLocality)
567 return impl_->getVelocitiesReadyOnDeviceEvent(atomLocality);
570 void StatePropagatorDataGpu::copyVelocitiesFromGpu(gmx::ArrayRef<RVec> h_v, AtomLocality atomLocality)
572 return impl_->copyVelocitiesFromGpu(h_v, atomLocality);
575 void StatePropagatorDataGpu::waitVelocitiesReadyOnHost(AtomLocality atomLocality)
577 return impl_->waitVelocitiesReadyOnHost(atomLocality);
581 DeviceBuffer<float> StatePropagatorDataGpu::getForces()
583 return impl_->getForces();
586 void StatePropagatorDataGpu::copyForcesToGpu(const gmx::ArrayRef<const gmx::RVec> h_f, AtomLocality atomLocality)
588 return impl_->copyForcesToGpu(h_f, atomLocality);
591 GpuEventSynchronizer* StatePropagatorDataGpu::getForcesReadyOnDeviceEvent(AtomLocality atomLocality,
592 bool useGpuFBufferOps)
594 return impl_->getForcesReadyOnDeviceEvent(atomLocality, useGpuFBufferOps);
597 GpuEventSynchronizer* StatePropagatorDataGpu::fReducedOnDevice()
599 return impl_->fReducedOnDevice();
602 void StatePropagatorDataGpu::copyForcesFromGpu(gmx::ArrayRef<RVec> h_f, AtomLocality atomLocality)
604 return impl_->copyForcesFromGpu(h_f, atomLocality);
607 void StatePropagatorDataGpu::waitForcesReadyOnHost(AtomLocality atomLocality)
609 return impl_->waitForcesReadyOnHost(atomLocality);
613 void* StatePropagatorDataGpu::getUpdateStream()
615 return impl_->getUpdateStream();
618 int StatePropagatorDataGpu::numAtomsLocal()
620 return impl_->numAtomsLocal();
623 int StatePropagatorDataGpu::numAtomsAll()
625 return impl_->numAtomsAll();
630 #endif // GMX_GPU == GMX_GPU_NONE