src/gromacs/mdtypes/state_propagator_data_gpu_impl_gpu.cpp

   1 /*
   2  * This file is part of the GROMACS molecular simulation package.
   3  *
   4  * Copyright (c) 2019,2020,2021, 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.
   8  *
   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.
  13  *
  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.
  18  *
  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.
  23  *
  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.
  31  *
  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.
  34  */
  35 /*! \internal \file
  36  *
  37  * \brief Definitions of interfaces for GPU state data propagator object.
  38  *
  39  * \author Artem Zhmurov <zhmurov@gmail.com>
  40  *
  41  * \ingroup module_mdtypes
  42  */
  43 #include "gmxpre.h"
  44
  45 #include "config.h"
  46
  47 #if GMX_GPU
  48
  49 #    include "gromacs/gpu_utils/device_stream_manager.h"
  50 #    include "gromacs/gpu_utils/devicebuffer.h"
  51 #    include "gromacs/gpu_utils/gpueventsynchronizer.h"
  52 #    include "gromacs/math/vectypes.h"
  53 #    include "gromacs/mdtypes/state_propagator_data_gpu.h"
  54 #    include "gromacs/timing/wallcycle.h"
  55 #    include "gromacs/utility/classhelpers.h"
  56
  57 #    include "state_propagator_data_gpu_impl.h"
  58
  59
  60 namespace gmx
  61 {
  62
  63 StatePropagatorDataGpu::Impl::Impl(const DeviceStreamManager& deviceStreamManager,
  64                                    GpuApiCallBehavior         transferKind,
  65                                    int                        allocationBlockSizeDivisor,
  66                                    gmx_wallcycle*             wcycle) :
  67     deviceContext_(deviceStreamManager.context()),
  68     transferKind_(transferKind),
  69     allocationBlockSizeDivisor_(allocationBlockSizeDivisor),
  70     wcycle_(wcycle)
  71 {
  72     static_assert(
  73             GMX_GPU,
  74             "GPU state propagator data object should only be constructed on the GPU code-paths.");
  75
  76     // We need to keep local copies for re-initialization.
  77     pmeStream_      = &deviceStreamManager.stream(DeviceStreamType::Pme);
  78     localStream_    = &deviceStreamManager.stream(DeviceStreamType::NonBondedLocal);
  79     nonLocalStream_ = &deviceStreamManager.stream(DeviceStreamType::NonBondedNonLocal);
  80     // PME stream is used in OpenCL for H2D coordinate transfer
  81     updateStream_ = &deviceStreamManager.stream(
  82             GMX_GPU_OPENCL ? DeviceStreamType::Pme : DeviceStreamType::UpdateAndConstraints);
  83
  84     // Map the atom locality to the stream that will be used for coordinates,
  85     // velocities and forces transfers. Same streams are used for H2D and D2H copies.
  86     // Note, that nullptr stream is used here to indicate that the copy is not supported.
  87     xCopyStreams_[AtomLocality::Local]    = updateStream_;
  88     xCopyStreams_[AtomLocality::NonLocal] = nonLocalStream_;
  89     xCopyStreams_[AtomLocality::All]      = nullptr;
  90
  91     vCopyStreams_[AtomLocality::Local]    = updateStream_;
  92     vCopyStreams_[AtomLocality::NonLocal] = nullptr;
  93     vCopyStreams_[AtomLocality::All]      = nullptr;
  94
  95     fCopyStreams_[AtomLocality::Local]    = localStream_;
  96     fCopyStreams_[AtomLocality::NonLocal] = nonLocalStream_;
  97     fCopyStreams_[AtomLocality::All]      = updateStream_;
  98
  99     copyInStream_ = std::make_unique<DeviceStream>(deviceContext_, DeviceStreamPriority::Normal, false);
 100     memsetStream_ = std::make_unique<DeviceStream>(deviceContext_, DeviceStreamPriority::Normal, false);
 101 }
 102
 103 StatePropagatorDataGpu::Impl::Impl(const DeviceStream*  pmeStream,
 104                                    const DeviceContext& deviceContext,
 105                                    GpuApiCallBehavior   transferKind,
 106                                    int                  allocationBlockSizeDivisor,
 107                                    gmx_wallcycle*       wcycle) :
 108     deviceContext_(deviceContext),
 109     transferKind_(transferKind),
 110     allocationBlockSizeDivisor_(allocationBlockSizeDivisor),
 111     wcycle_(wcycle)
 112 {
 113     static_assert(
 114             GMX_GPU,
 115             "GPU state propagator data object should only be constructed on the GPU code-paths.");
 116
 117     GMX_ASSERT(pmeStream->isValid(), "GPU PME stream should be valid.");
 118     pmeStream_      = pmeStream;
 119     localStream_    = pmeStream; // For clearing the force buffer
 120     nonLocalStream_ = nullptr;
 121     updateStream_   = nullptr;
 122
 123
 124     // Only local/all coordinates are allowed to be copied in PME-only rank/ PME tests.
 125     // This it temporary measure to make it safe to use this class in those cases.
 126     xCopyStreams_[AtomLocality::Local]    = pmeStream_;
 127     xCopyStreams_[AtomLocality::NonLocal] = nullptr;
 128     xCopyStreams_[AtomLocality::All]      = nullptr;
 129
 130     vCopyStreams_[AtomLocality::Local]    = nullptr;
 131     vCopyStreams_[AtomLocality::NonLocal] = nullptr;
 132     vCopyStreams_[AtomLocality::All]      = nullptr;
 133
 134     fCopyStreams_[AtomLocality::Local]    = nullptr;
 135     fCopyStreams_[AtomLocality::NonLocal] = nullptr;
 136     fCopyStreams_[AtomLocality::All]      = nullptr;
 137 }
 138
 139 StatePropagatorDataGpu::Impl::~Impl() {}
 140
 141 void StatePropagatorDataGpu::Impl::reinit(int numAtomsLocal, int numAtomsAll)
 142 {
 143     wallcycle_start_nocount(wcycle_, WallCycleCounter::LaunchGpu);
 144     wallcycle_sub_start_nocount(wcycle_, WallCycleSubCounter::LaunchStatePropagatorData);
 145
 146     numAtomsLocal_ = numAtomsLocal;
 147     numAtomsAll_   = numAtomsAll;
 148
 149     int numAtomsPadded;
 150     if (allocationBlockSizeDivisor_ > 0)
 151     {
 152         numAtomsPadded = ((numAtomsAll_ + allocationBlockSizeDivisor_ - 1) / allocationBlockSizeDivisor_)
 153                          * allocationBlockSizeDivisor_;
 154     }
 155     else
 156     {
 157         numAtomsPadded = numAtomsAll_;
 158     }
 159
 160     reallocateDeviceBuffer(&d_x_, numAtomsPadded, &d_xSize_, &d_xCapacity_, deviceContext_);
 161
 162     const size_t paddingAllocationSize = numAtomsPadded - numAtomsAll_;
 163     if (paddingAllocationSize > 0)
 164     {
 165         // The PME stream is used here because the padding region of d_x_ is only in the PME task.
 166         clearDeviceBufferAsync(&d_x_, numAtomsAll_, paddingAllocationSize, *pmeStream_);
 167     }
 168
 169     reallocateDeviceBuffer(&d_v_, numAtomsAll_, &d_vSize_, &d_vCapacity_, deviceContext_);
 170     const int d_fOldCapacity = d_fCapacity_;
 171     reallocateDeviceBuffer(&d_f_, numAtomsAll_, &d_fSize_, &d_fCapacity_, deviceContext_);
 172
 173     // Clearing of the forces can be done in local stream since the nonlocal stream cannot reach
 174     // the force accumulation stage before syncing with the local stream. Only done in CUDA,
 175     // since the force buffer ops are not implemented in OpenCL.
 176     if (GMX_GPU_CUDA && d_fCapacity_ != d_fOldCapacity)
 177     {
 178         clearDeviceBufferAsync(&d_f_, 0, d_fCapacity_, *localStream_);
 179     }
 180
 181     wallcycle_sub_stop(wcycle_, WallCycleSubCounter::LaunchStatePropagatorData);
 182     wallcycle_stop(wcycle_, WallCycleCounter::LaunchGpu);
 183 }
 184
 185 std::tuple<int, int> StatePropagatorDataGpu::Impl::getAtomRangesFromAtomLocality(AtomLocality atomLocality) const
 186 {
 187     int atomsStartAt   = 0;
 188     int numAtomsToCopy = 0;
 189     switch (atomLocality)
 190     {
 191         case AtomLocality::All:
 192             atomsStartAt   = 0;
 193             numAtomsToCopy = numAtomsAll_;
 194             break;
 195         case AtomLocality::Local:
 196             atomsStartAt   = 0;
 197             numAtomsToCopy = numAtomsLocal_;
 198             break;
 199         case AtomLocality::NonLocal:
 200             atomsStartAt   = numAtomsLocal_;
 201             numAtomsToCopy = numAtomsAll_ - numAtomsLocal_;
 202             break;
 203         default:
 204             GMX_RELEASE_ASSERT(false,
 205                                "Wrong range of atoms requested in GPU state data manager. Should "
 206                                "be All, Local or NonLocal.");
 207     }
 208     GMX_ASSERT(atomsStartAt >= 0,
 209                "The first elemtnt to copy has negative index. Probably, the GPU propagator state "
 210                "was not initialized.");
 211     GMX_ASSERT(numAtomsToCopy >= 0,
 212                "Number of atoms to copy is negative. Probably, the GPU propagator state was not "
 213                "initialized.");
 214     return std::make_tuple(atomsStartAt, numAtomsToCopy);
 215 }
 216
 217 void StatePropagatorDataGpu::Impl::copyToDevice(DeviceBuffer<RVec>                   d_data,
 218                                                 const gmx::ArrayRef<const gmx::RVec> h_data,
 219                                                 int                                  dataSize,
 220                                                 AtomLocality                         atomLocality,
 221                                                 const DeviceStream&                  deviceStream)
 222 {
 223     GMX_UNUSED_VALUE(dataSize);
 224
 225     GMX_ASSERT(atomLocality < AtomLocality::Count, "Wrong atom locality.");
 226
 227     GMX_ASSERT(dataSize >= 0, "Trying to copy to device buffer before it was allocated.");
 228
 229     GMX_ASSERT(deviceStream.isValid(), "No stream is valid for copying with given atom locality.");
 230
 231     int atomsStartAt, numAtomsToCopy;
 232     std::tie(atomsStartAt, numAtomsToCopy) = getAtomRangesFromAtomLocality(atomLocality);
 233
 234     if (numAtomsToCopy != 0)
 235     {
 236         GMX_ASSERT(atomsStartAt + numAtomsToCopy <= dataSize,
 237                    "The device allocation is smaller than requested copy range.");
 238         GMX_ASSERT(atomsStartAt + numAtomsToCopy <= h_data.ssize(),
 239                    "The host buffer is smaller than the requested copy range.");
 240
 241         copyToDeviceBuffer(&d_data,
 242                            reinterpret_cast<const RVec*>(&h_data.data()[atomsStartAt]),
 243                            atomsStartAt,
 244                            numAtomsToCopy,
 245                            deviceStream,
 246                            transferKind_,
 247                            nullptr);
 248     }
 249 }
 250
 251 void StatePropagatorDataGpu::Impl::copyFromDevice(gmx::ArrayRef<gmx::RVec> h_data,
 252                                                   DeviceBuffer<RVec>       d_data,
 253                                                   int                      dataSize,
 254                                                   AtomLocality             atomLocality,
 255                                                   const DeviceStream&      deviceStream)
 256 {
 257     GMX_UNUSED_VALUE(dataSize);
 258
 259     GMX_ASSERT(atomLocality < AtomLocality::Count, "Wrong atom locality.");
 260
 261     GMX_ASSERT(dataSize >= 0, "Trying to copy from device buffer before it was allocated.");
 262
 263     GMX_ASSERT(deviceStream.isValid(), "No stream is valid for copying with given atom locality.");
 264
 265     int atomsStartAt, numAtomsToCopy;
 266     std::tie(atomsStartAt, numAtomsToCopy) = getAtomRangesFromAtomLocality(atomLocality);
 267
 268     if (numAtomsToCopy != 0)
 269     {
 270         GMX_ASSERT(atomsStartAt + numAtomsToCopy <= dataSize,
 271                    "The device allocation is smaller than requested copy range.");
 272         GMX_ASSERT(atomsStartAt + numAtomsToCopy <= h_data.ssize(),
 273                    "The host buffer is smaller than the requested copy range.");
 274
 275         copyFromDeviceBuffer(reinterpret_cast<RVec*>(&h_data.data()[atomsStartAt]),
 276                              &d_data,
 277                              atomsStartAt,
 278                              numAtomsToCopy,
 279                              deviceStream,
 280                              transferKind_,
 281                              nullptr);
 282     }
 283 }
 284
 285 void StatePropagatorDataGpu::Impl::clearOnDevice(DeviceBuffer<RVec>  d_data,
 286                                                  int                 dataSize,
 287                                                  AtomLocality        atomLocality,
 288                                                  const DeviceStream& deviceStream) const
 289 {
 290     GMX_UNUSED_VALUE(dataSize);
 291
 292     GMX_ASSERT(atomLocality < AtomLocality::Count, "Wrong atom locality.");
 293
 294     GMX_ASSERT(dataSize >= 0, "Trying to clear to device buffer before it was allocated.");
 295
 296     GMX_ASSERT(deviceStream.isValid(), "No stream is valid for clearing with given atom locality.");
 297
 298     int atomsStartAt, numAtomsToClear;
 299     std::tie(atomsStartAt, numAtomsToClear) = getAtomRangesFromAtomLocality(atomLocality);
 300
 301     if (numAtomsToClear != 0)
 302     {
 303         GMX_ASSERT(atomsStartAt + numAtomsToClear <= dataSize,
 304                    "The device allocation is smaller than requested clear range.");
 305
 306         clearDeviceBufferAsync(&d_data, atomsStartAt, numAtomsToClear, deviceStream);
 307     }
 308 }
 309
 310 DeviceBuffer<RVec> StatePropagatorDataGpu::Impl::getCoordinates()
 311 {
 312     return d_x_;
 313 }
 314
 315 void StatePropagatorDataGpu::Impl::copyCoordinatesToGpu(const gmx::ArrayRef<const gmx::RVec> h_x,
 316                                                         AtomLocality atomLocality)
 317 {
 318     GMX_ASSERT(atomLocality < AtomLocality::All,
 319                formatString("Wrong atom locality. Only Local and NonLocal are allowed for "
 320                             "coordinate transfers, passed value is \"%s\"",
 321                             enumValueToString(atomLocality))
 322                        .c_str());
 323
 324     const DeviceStream* deviceStream = xCopyStreams_[atomLocality];
 325     GMX_ASSERT(deviceStream != nullptr,
 326                "No stream is valid for copying positions with given atom locality.");
 327
 328     wallcycle_start_nocount(wcycle_, WallCycleCounter::LaunchGpu);
 329     wallcycle_sub_start(wcycle_, WallCycleSubCounter::LaunchStatePropagatorData);
 330
 331     copyToDevice(d_x_, h_x, d_xSize_, atomLocality, *deviceStream);
 332
 333     // markEvent is skipped in OpenCL as:
 334     //   - it's not needed, copy is done in the same stream as the only consumer task (PME)
 335     //   - we don't consume the events in OpenCL which is not allowed by GpuEventSynchronizer (would leak memory).
 336     // TODO: remove this by adding an event-mark free flavor of this function
 337     if (GMX_GPU_CUDA)
 338     {
 339         xReadyOnDevice_[atomLocality].markEvent(*deviceStream);
 340     }
 341
 342     wallcycle_sub_stop(wcycle_, WallCycleSubCounter::LaunchStatePropagatorData);
 343     wallcycle_stop(wcycle_, WallCycleCounter::LaunchGpu);
 344 }
 345
 346 GpuEventSynchronizer* StatePropagatorDataGpu::Impl::getCoordinatesReadyOnDeviceEvent(
 347         AtomLocality              atomLocality,
 348         const SimulationWorkload& simulationWork,
 349         const StepWorkload&       stepWork,
 350         GpuEventSynchronizer*     gpuCoordinateHaloLaunched)
 351 {
 352     // The provider of the coordinates may be different for local atoms. If the update is offloaded
 353     // and this is not a neighbor search step, then the consumer needs to wait for the update
 354     // to complete. Otherwise, the coordinates are copied from the host and we need to wait for
 355     // the copy event. Non-local coordinates are provided by the GPU halo exchange (if active), otherwise by H2D copy.
 356     //
 357     // In OpenCL no events are used as coordinate sync is not necessary
 358     if (GMX_GPU_OPENCL)
 359     {
 360         return nullptr;
 361     }
 362     if (atomLocality == AtomLocality::NonLocal && stepWork.useGpuXHalo)
 363     {
 364         GMX_ASSERT(gpuCoordinateHaloLaunched != nullptr,
 365                    "GPU halo exchange is active but its completion event is null.");
 366         return gpuCoordinateHaloLaunched;
 367     }
 368     if (atomLocality == AtomLocality::Local && simulationWork.useGpuUpdate && !stepWork.doNeighborSearch)
 369     {
 370         GMX_ASSERT(xUpdatedOnDeviceEvent_ != nullptr, "The event synchronizer can not be nullptr.");
 371         return xUpdatedOnDeviceEvent_;
 372     }
 373     else
 374     {
 375         if (stepWork.doNeighborSearch && xUpdatedOnDeviceEvent_)
 376         {
 377             /* On search steps, we do not consume the result of the GPU update
 378              * but rather that of a H2D transfer. So, we reset the event triggered after
 379              * update to avoid leaving it unconsumed.
 380              * Unfortunately, we don't always have the event marked either (e.g., on the
 381              * first step) so we just reset it here.
 382              * See Issue #3988. */
 383             xUpdatedOnDeviceEvent_->reset();
 384         }
 385         return &xReadyOnDevice_[atomLocality];
 386     }
 387 }
 388
 389 void StatePropagatorDataGpu::Impl::waitCoordinatesCopiedToDevice(AtomLocality atomLocality)
 390 {
 391     wallcycle_start(wcycle_, WallCycleCounter::WaitGpuStatePropagatorData);
 392     GMX_ASSERT(atomLocality < AtomLocality::Count, "Wrong atom locality.");
 393     xReadyOnDevice_[atomLocality].waitForEvent();
 394     wallcycle_stop(wcycle_, WallCycleCounter::WaitGpuStatePropagatorData);
 395 }
 396
 397 void StatePropagatorDataGpu::Impl::consumeCoordinatesCopiedToDeviceEvent(AtomLocality atomLocality)
 398 {
 399     GMX_ASSERT(atomLocality < AtomLocality::Count, "Wrong atom locality.");
 400     xReadyOnDevice_[atomLocality].consume();
 401 }
 402
 403 void StatePropagatorDataGpu::Impl::resetCoordinatesCopiedToDeviceEvent(AtomLocality atomLocality)
 404 {
 405     GMX_ASSERT(atomLocality < AtomLocality::Count, "Wrong atom locality.");
 406     xReadyOnDevice_[atomLocality].reset();
 407 }
 408
 409 void StatePropagatorDataGpu::Impl::setXUpdatedOnDeviceEvent(GpuEventSynchronizer* xUpdatedOnDeviceEvent)
 410 {
 411     GMX_ASSERT(xUpdatedOnDeviceEvent != nullptr, "The event synchronizer can not be nullptr.");
 412     xUpdatedOnDeviceEvent_ = xUpdatedOnDeviceEvent;
 413 }
 414
 415 void StatePropagatorDataGpu::Impl::copyCoordinatesFromGpu(gmx::ArrayRef<gmx::RVec> h_x,
 416                                                           AtomLocality             atomLocality,
 417                                                           GpuEventSynchronizer*    dependency)
 418 {
 419     GMX_ASSERT(atomLocality < AtomLocality::All,
 420                formatString("Wrong atom locality. Only Local and NonLocal are allowed for "
 421                             "coordinate transfers, passed value is \"%s\"",
 422                             enumValueToString(atomLocality))
 423                        .c_str());
 424     const DeviceStream* deviceStream = xCopyStreams_[atomLocality];
 425     GMX_ASSERT(deviceStream != nullptr,
 426                "No stream is valid for copying positions with given atom locality.");
 427
 428     if (dependency != nullptr)
 429     {
 430         dependency->enqueueWaitEvent(*deviceStream);
 431     }
 432
 433     wallcycle_start_nocount(wcycle_, WallCycleCounter::LaunchGpu);
 434     wallcycle_sub_start(wcycle_, WallCycleSubCounter::LaunchStatePropagatorData);
 435
 436     copyFromDevice(h_x, d_x_, d_xSize_, atomLocality, *deviceStream);
 437     // Note: unlike copyCoordinatesToGpu this is not used in OpenCL, and the conditional is not needed.
 438     xReadyOnHost_[atomLocality].markEvent(*deviceStream);
 439
 440     wallcycle_sub_stop(wcycle_, WallCycleSubCounter::LaunchStatePropagatorData);
 441     wallcycle_stop(wcycle_, WallCycleCounter::LaunchGpu);
 442 }
 443
 444 void StatePropagatorDataGpu::Impl::waitCoordinatesReadyOnHost(AtomLocality atomLocality)
 445 {
 446     wallcycle_start(wcycle_, WallCycleCounter::WaitGpuStatePropagatorData);
 447     xReadyOnHost_[atomLocality].waitForEvent();
 448     wallcycle_stop(wcycle_, WallCycleCounter::WaitGpuStatePropagatorData);
 449 }
 450
 451
 452 DeviceBuffer<RVec> StatePropagatorDataGpu::Impl::getVelocities()
 453 {
 454     return d_v_;
 455 }
 456
 457 void StatePropagatorDataGpu::Impl::copyVelocitiesToGpu(const gmx::ArrayRef<const gmx::RVec> h_v,
 458                                                        AtomLocality atomLocality)
 459 {
 460     GMX_ASSERT(atomLocality == AtomLocality::Local,
 461                formatString("Wrong atom locality. Only Local is allowed for "
 462                             "velocity transfers, passed value is \"%s\"",
 463                             enumValueToString(atomLocality))
 464                        .c_str());
 465     const DeviceStream* deviceStream = vCopyStreams_[atomLocality];
 466     GMX_ASSERT(deviceStream != nullptr,
 467                "No stream is valid for copying velocities with given atom locality.");
 468
 469     wallcycle_start_nocount(wcycle_, WallCycleCounter::LaunchGpu);
 470     wallcycle_sub_start(wcycle_, WallCycleSubCounter::LaunchStatePropagatorData);
 471
 472     copyToDevice(d_v_, h_v, d_vSize_, atomLocality, *deviceStream);
 473     /* Not marking the event, because it is not used anywhere.
 474      * Since we only use velocities on the device for update, and we launch the copy in
 475      * the "update" stream, that should be safe.
 476      */
 477
 478     wallcycle_sub_stop(wcycle_, WallCycleSubCounter::LaunchStatePropagatorData);
 479     wallcycle_stop(wcycle_, WallCycleCounter::LaunchGpu);
 480 }
 481
 482 void StatePropagatorDataGpu::Impl::copyVelocitiesFromGpu(gmx::ArrayRef<gmx::RVec> h_v, AtomLocality atomLocality)
 483 {
 484     GMX_ASSERT(atomLocality == AtomLocality::Local,
 485                formatString("Wrong atom locality. Only Local is allowed for "
 486                             "velocity transfers, passed value is \"%s\"",
 487                             enumValueToString(atomLocality))
 488                        .c_str());
 489     const DeviceStream* deviceStream = vCopyStreams_[atomLocality];
 490     GMX_ASSERT(deviceStream != nullptr,
 491                "No stream is valid for copying velocities with given atom locality.");
 492
 493     wallcycle_start_nocount(wcycle_, WallCycleCounter::LaunchGpu);
 494     wallcycle_sub_start(wcycle_, WallCycleSubCounter::LaunchStatePropagatorData);
 495
 496     copyFromDevice(h_v, d_v_, d_vSize_, atomLocality, *deviceStream);
 497     vReadyOnHost_[atomLocality].markEvent(*deviceStream);
 498
 499     wallcycle_sub_stop(wcycle_, WallCycleSubCounter::LaunchStatePropagatorData);
 500     wallcycle_stop(wcycle_, WallCycleCounter::LaunchGpu);
 501 }
 502
 503 void StatePropagatorDataGpu::Impl::waitVelocitiesReadyOnHost(AtomLocality atomLocality)
 504 {
 505     wallcycle_start(wcycle_, WallCycleCounter::WaitGpuStatePropagatorData);
 506     vReadyOnHost_[atomLocality].waitForEvent();
 507     wallcycle_stop(wcycle_, WallCycleCounter::WaitGpuStatePropagatorData);
 508 }
 509
 510
 511 DeviceBuffer<RVec> StatePropagatorDataGpu::Impl::getForces()
 512 {
 513     return d_f_;
 514 }
 515
 516 // Copy CPU forces to GPU using stream internal to this module to allow overlap
 517 // with GPU force calculations.
 518 void StatePropagatorDataGpu::Impl::copyForcesToGpu(const gmx::ArrayRef<const gmx::RVec> h_f,
 519                                                    AtomLocality atomLocality)
 520 {
 521     GMX_ASSERT(atomLocality < AtomLocality::Count, "Wrong atom locality.");
 522     DeviceStream* deviceStream = copyInStream_.get();
 523     GMX_ASSERT(deviceStream != nullptr,
 524                "No stream is valid for copying forces with given atom locality.");
 525
 526     wallcycle_start_nocount(wcycle_, WallCycleCounter::LaunchGpu);
 527     wallcycle_sub_start(wcycle_, WallCycleSubCounter::LaunchStatePropagatorData);
 528
 529     copyToDevice(d_f_, h_f, d_fSize_, atomLocality, *deviceStream);
 530     fReadyOnDevice_[atomLocality].markEvent(*deviceStream);
 531
 532     wallcycle_sub_stop(wcycle_, WallCycleSubCounter::LaunchStatePropagatorData);
 533     wallcycle_stop(wcycle_, WallCycleCounter::LaunchGpu);
 534 }
 535
 536 void StatePropagatorDataGpu::Impl::clearForcesOnGpu(AtomLocality atomLocality, GpuEventSynchronizer* dependency)
 537 {
 538     GMX_ASSERT(atomLocality < AtomLocality::Count, "Wrong atom locality.");
 539     DeviceStream* deviceStream = memsetStream_.get();
 540
 541     GMX_ASSERT(dependency != nullptr, "Dependency is not valid for clearing forces.");
 542     dependency->enqueueWaitEvent(*deviceStream);
 543
 544     GMX_ASSERT(deviceStream != nullptr,
 545                "No stream is valid for clearing forces with given atom locality.");
 546
 547     wallcycle_start_nocount(wcycle_, WallCycleCounter::LaunchGpu);
 548     wallcycle_sub_start(wcycle_, WallCycleSubCounter::LaunchStatePropagatorData);
 549
 550     clearOnDevice(d_f_, d_fSize_, atomLocality, *deviceStream);
 551
 552     fReadyOnDevice_[atomLocality].markEvent(*deviceStream);
 553
 554     wallcycle_sub_stop(wcycle_, WallCycleSubCounter::LaunchStatePropagatorData);
 555     wallcycle_stop(wcycle_, WallCycleCounter::LaunchGpu);
 556 }
 557
 558 GpuEventSynchronizer* StatePropagatorDataGpu::Impl::getLocalForcesReadyOnDeviceEvent(StepWorkload stepWork,
 559                                                                                      SimulationWorkload simulationWork)
 560 {
 561     if (stepWork.useGpuFBufferOps && !simulationWork.useCpuPmePpCommunication)
 562     {
 563         return &fReducedOnDevice_[AtomLocality::Local];
 564     }
 565     else
 566     {
 567         return &fReadyOnDevice_[AtomLocality::Local];
 568     }
 569 }
 570
 571 GpuEventSynchronizer* StatePropagatorDataGpu::Impl::fReducedOnDevice(AtomLocality atomLocality)
 572 {
 573     return &fReducedOnDevice_[atomLocality];
 574 }
 575
 576 void StatePropagatorDataGpu::Impl::consumeForcesReducedOnDeviceEvent(AtomLocality atomLocality)
 577 {
 578     fReducedOnDevice_[atomLocality].consume();
 579 }
 580
 581 GpuEventSynchronizer* StatePropagatorDataGpu::Impl::fReadyOnDevice(AtomLocality atomLocality)
 582 {
 583     return &fReadyOnDevice_[atomLocality];
 584 }
 585
 586 void StatePropagatorDataGpu::Impl::copyForcesFromGpu(gmx::ArrayRef<gmx::RVec> h_f, AtomLocality atomLocality)
 587 {
 588     GMX_ASSERT(atomLocality < AtomLocality::Count, "Wrong atom locality.");
 589     const DeviceStream* deviceStream = fCopyStreams_[atomLocality];
 590     GMX_ASSERT(deviceStream != nullptr,
 591                "No stream is valid for copying forces with given atom locality.");
 592
 593     wallcycle_start_nocount(wcycle_, WallCycleCounter::LaunchGpu);
 594     wallcycle_sub_start(wcycle_, WallCycleSubCounter::LaunchStatePropagatorData);
 595
 596     copyFromDevice(h_f, d_f_, d_fSize_, atomLocality, *deviceStream);
 597     fReadyOnHost_[atomLocality].markEvent(*deviceStream);
 598
 599     wallcycle_sub_stop(wcycle_, WallCycleSubCounter::LaunchStatePropagatorData);
 600     wallcycle_stop(wcycle_, WallCycleCounter::LaunchGpu);
 601 }
 602
 603 void StatePropagatorDataGpu::Impl::waitForcesReadyOnHost(AtomLocality atomLocality)
 604 {
 605     wallcycle_start(wcycle_, WallCycleCounter::WaitGpuStatePropagatorData);
 606     fReadyOnHost_[atomLocality].waitForEvent();
 607     wallcycle_stop(wcycle_, WallCycleCounter::WaitGpuStatePropagatorData);
 608 }
 609
 610 const DeviceStream* StatePropagatorDataGpu::Impl::getUpdateStream()
 611 {
 612     return updateStream_;
 613 }
 614
 615 int StatePropagatorDataGpu::Impl::numAtomsLocal() const
 616 {
 617     return numAtomsLocal_;
 618 }
 619
 620 int StatePropagatorDataGpu::Impl::numAtomsAll() const
 621 {
 622     return numAtomsAll_;
 623 }
 624
 625
 626 StatePropagatorDataGpu::StatePropagatorDataGpu(const DeviceStreamManager& deviceStreamManager,
 627                                                GpuApiCallBehavior         transferKind,
 628                                                int            allocationBlockSizeDivisor,
 629                                                gmx_wallcycle* wcycle) :
 630     impl_(new Impl(deviceStreamManager, transferKind, allocationBlockSizeDivisor, wcycle))
 631 {
 632 }
 633
 634 StatePropagatorDataGpu::StatePropagatorDataGpu(const DeviceStream*  pmeStream,
 635                                                const DeviceContext& deviceContext,
 636                                                GpuApiCallBehavior   transferKind,
 637                                                int                  allocationBlockSizeDivisor,
 638                                                gmx_wallcycle*       wcycle) :
 639     impl_(new Impl(pmeStream, deviceContext, transferKind, allocationBlockSizeDivisor, wcycle))
 640 {
 641 }
 642
 643 StatePropagatorDataGpu::StatePropagatorDataGpu(StatePropagatorDataGpu&& /* other */) noexcept = default;
 644
 645 StatePropagatorDataGpu& StatePropagatorDataGpu::operator=(StatePropagatorDataGpu&& /* other */) noexcept = default;
 646
 647 StatePropagatorDataGpu::~StatePropagatorDataGpu() = default;
 648
 649
 650 void StatePropagatorDataGpu::reinit(int numAtomsLocal, int numAtomsAll)
 651 {
 652     return impl_->reinit(numAtomsLocal, numAtomsAll);
 653 }
 654
 655 std::tuple<int, int> StatePropagatorDataGpu::getAtomRangesFromAtomLocality(AtomLocality atomLocality) const
 656 {
 657     return impl_->getAtomRangesFromAtomLocality(atomLocality);
 658 }
 659
 660
 661 DeviceBuffer<RVec> StatePropagatorDataGpu::getCoordinates()
 662 {
 663     return impl_->getCoordinates();
 664 }
 665
 666 void StatePropagatorDataGpu::copyCoordinatesToGpu(const gmx::ArrayRef<const gmx::RVec> h_x,
 667                                                   AtomLocality                         atomLocality)
 668 {
 669     return impl_->copyCoordinatesToGpu(h_x, atomLocality);
 670 }
 671
 672 GpuEventSynchronizer*
 673 StatePropagatorDataGpu::getCoordinatesReadyOnDeviceEvent(AtomLocality              atomLocality,
 674                                                          const SimulationWorkload& simulationWork,
 675                                                          const StepWorkload&       stepWork,
 676                                                          GpuEventSynchronizer* gpuCoordinateHaloLaunched)
 677 {
 678     return impl_->getCoordinatesReadyOnDeviceEvent(
 679             atomLocality, simulationWork, stepWork, gpuCoordinateHaloLaunched);
 680 }
 681
 682 void StatePropagatorDataGpu::waitCoordinatesCopiedToDevice(AtomLocality atomLocality)
 683 {
 684     return impl_->waitCoordinatesCopiedToDevice(atomLocality);
 685 }
 686
 687 void StatePropagatorDataGpu::consumeCoordinatesCopiedToDeviceEvent(AtomLocality atomLocality)
 688 {
 689     return impl_->consumeCoordinatesCopiedToDeviceEvent(atomLocality);
 690 }
 691
 692 void StatePropagatorDataGpu::resetCoordinatesCopiedToDeviceEvent(AtomLocality atomLocality)
 693 {
 694     return impl_->resetCoordinatesCopiedToDeviceEvent(atomLocality);
 695 }
 696
 697 void StatePropagatorDataGpu::setXUpdatedOnDeviceEvent(GpuEventSynchronizer* xUpdatedOnDeviceEvent)
 698 {
 699     impl_->setXUpdatedOnDeviceEvent(xUpdatedOnDeviceEvent);
 700 }
 701
 702 void StatePropagatorDataGpu::copyCoordinatesFromGpu(gmx::ArrayRef<RVec>   h_x,
 703                                                     AtomLocality          atomLocality,
 704                                                     GpuEventSynchronizer* dependency)
 705 {
 706     return impl_->copyCoordinatesFromGpu(h_x, atomLocality, dependency);
 707 }
 708
 709 void StatePropagatorDataGpu::waitCoordinatesReadyOnHost(AtomLocality atomLocality)
 710 {
 711     return impl_->waitCoordinatesReadyOnHost(atomLocality);
 712 }
 713
 714
 715 DeviceBuffer<RVec> StatePropagatorDataGpu::getVelocities()
 716 {
 717     return impl_->getVelocities();
 718 }
 719
 720 void StatePropagatorDataGpu::copyVelocitiesToGpu(const gmx::ArrayRef<const gmx::RVec> h_v,
 721                                                  AtomLocality                         atomLocality)
 722 {
 723     return impl_->copyVelocitiesToGpu(h_v, atomLocality);
 724 }
 725
 726 void StatePropagatorDataGpu::copyVelocitiesFromGpu(gmx::ArrayRef<RVec> h_v, AtomLocality atomLocality)
 727 {
 728     return impl_->copyVelocitiesFromGpu(h_v, atomLocality);
 729 }
 730
 731 void StatePropagatorDataGpu::waitVelocitiesReadyOnHost(AtomLocality atomLocality)
 732 {
 733     return impl_->waitVelocitiesReadyOnHost(atomLocality);
 734 }
 735
 736
 737 DeviceBuffer<RVec> StatePropagatorDataGpu::getForces()
 738 {
 739     return impl_->getForces();
 740 }
 741
 742 void StatePropagatorDataGpu::copyForcesToGpu(const gmx::ArrayRef<const gmx::RVec> h_f, AtomLocality atomLocality)
 743 {
 744     return impl_->copyForcesToGpu(h_f, atomLocality);
 745 }
 746
 747 void StatePropagatorDataGpu::clearForcesOnGpu(AtomLocality atomLocality, GpuEventSynchronizer* dependency)
 748 {
 749     return impl_->clearForcesOnGpu(atomLocality, dependency);
 750 }
 751
 752 GpuEventSynchronizer* StatePropagatorDataGpu::getLocalForcesReadyOnDeviceEvent(StepWorkload stepWork,
 753                                                                                SimulationWorkload simulationWork)
 754 {
 755     return impl_->getLocalForcesReadyOnDeviceEvent(stepWork, simulationWork);
 756 }
 757
 758 GpuEventSynchronizer* StatePropagatorDataGpu::fReducedOnDevice(AtomLocality atomLocality)
 759 {
 760     return impl_->fReducedOnDevice(atomLocality);
 761 }
 762
 763 void StatePropagatorDataGpu::consumeForcesReducedOnDeviceEvent(AtomLocality atomLocality)
 764 {
 765     impl_->consumeForcesReducedOnDeviceEvent(atomLocality);
 766 }
 767
 768 GpuEventSynchronizer* StatePropagatorDataGpu::fReadyOnDevice(AtomLocality atomLocality)
 769 {
 770     return impl_->fReadyOnDevice(atomLocality);
 771 }
 772
 773 void StatePropagatorDataGpu::copyForcesFromGpu(gmx::ArrayRef<RVec> h_f, AtomLocality atomLocality)
 774 {
 775     return impl_->copyForcesFromGpu(h_f, atomLocality);
 776 }
 777
 778 void StatePropagatorDataGpu::waitForcesReadyOnHost(AtomLocality atomLocality)
 779 {
 780     return impl_->waitForcesReadyOnHost(atomLocality);
 781 }
 782
 783
 784 const DeviceStream* StatePropagatorDataGpu::getUpdateStream()
 785 {
 786     return impl_->getUpdateStream();
 787 }
 788
 789 int StatePropagatorDataGpu::numAtomsLocal() const
 790 {
 791     return impl_->numAtomsLocal();
 792 }
 793
 794 int StatePropagatorDataGpu::numAtomsAll() const
 795 {
 796     return impl_->numAtomsAll();
 797 }
 798
 799 } // namespace gmx
 800
 801 #endif // GMX_GPU