return *this;
}
+
+namespace gmx::internal
+{
+//! Shorthand alias to create a placeholder SYCL accessor with chosen data type and access mode.
+template<class T, enum cl::sycl::access::mode mode>
+using PlaceholderAccessor =
+ cl::sycl::accessor<T, 1, mode, cl::sycl::access::target::global_buffer, cl::sycl::access::placeholder::true_t>;
+} // namespace gmx::internal
+
+/** \brief
+ * Thin wrapper around placeholder accessor that allows implicit construction from \c DeviceBuffer.
+ *
+ * "Placeholder accessor" is an indicator of the intent to create an accessor for certain buffer
+ * of a certain type, that is not yet bound to a specific command group handler (device). Such
+ * accessors can be created outside SYCL kernels, which is helpful if we want to pass them as
+ * function arguments.
+ *
+ * \tparam T Type of buffer content.
+ * \tparam mode Access mode.
+ */
+template<class T, enum cl::sycl::access::mode mode>
+class DeviceAccessor : public gmx::internal::PlaceholderAccessor<T, mode>
+{
+public:
+ // Inherit all the constructors
+ using gmx::internal::PlaceholderAccessor<T, mode>::PlaceholderAccessor;
+ //! Construct Accessor from DeviceBuffer (must be initialized)
+ DeviceAccessor(DeviceBuffer<T>& buffer) :
+ gmx::internal::PlaceholderAccessor<T, mode>(getSyclBuffer(buffer))
+ {
+ }
+
+private:
+ //! Helper function to get sycl:buffer object from DeviceBuffer wrapper, with a sanity check.
+ static inline cl::sycl::buffer<T, 1>& getSyclBuffer(DeviceBuffer<T>& buffer)
+ {
+ GMX_ASSERT(bool(buffer), "Trying to construct accessor from an uninitialized buffer");
+ return *buffer.buffer_;
+ }
+};
+
+namespace gmx::internal
+{
+//! A "blackhole" class to be used when we want to ignore an argument to a function.
+struct EmptyClassThatIgnoresConstructorArguments
+{
+ template<class... Args>
+ [[maybe_unused]] EmptyClassThatIgnoresConstructorArguments(Args&&... /*args*/)
+ {
+ }
+};
+} // namespace gmx::internal
+
+/** \brief
+ * Helper class to be used as function argument. Will either correspond to a device accessor, or an empty class.
+ *
+ * Example usage:
+ * \code
+ template <bool doFoo>
+ void getBarKernel(handler& cgh, OptionalAccessor<float, mode::read, doFoo> a_fooPrms)
+ {
+ if constexpr (doFoo)
+ cgh.require(a_fooPrms);
+ // Can only use a_fooPrms if doFoo == true
+ }
+
+ template <bool doFoo>
+ void callBar(DeviceBuffer<float> b_fooPrms)
+ {
+ // If doFoo is false, b_fooPrms will be ignored (can be not initialized).
+ // Otherwise, an accessor will be built (b_fooPrms must be a valid buffer).
+ auto kernel = getBarKernel<doFoo>(b_fooPrms);
+ // If the accessor in not enabled, anything can be passed as its ctor argument.
+ auto kernel2 = getBarKernel<false>(nullptr_t);
+ }
+ * \endcode
+ *
+ * \tparam T Data type of the underlying buffer
+ * \tparam mode Access mode of the accessor
+ * \tparam enabled Compile-time flag indicating whether we want to actually create an accessor.
+ */
+template<class T, enum cl::sycl::access::mode mode, bool enabled>
+using OptionalAccessor =
+ std::conditional_t<enabled, DeviceAccessor<T, mode>, gmx::internal::EmptyClassThatIgnoresConstructorArguments>;
+
#endif // #ifndef DOXYGEN
+/*! \brief Check the validity of the device buffer.
+ *
+ * Checks if the buffer is valid and if its allocation is big enough.
+ *
+ * \param[in] buffer Device buffer to be checked.
+ * \param[in] requiredSize Number of elements that the buffer will have to accommodate.
+ *
+ * \returns Whether the device buffer exists and has enough capacity.
+ */
+template<typename T>
+static gmx_unused bool checkDeviceBuffer(const DeviceBuffer<T>& buffer, int requiredSize)
+{
+ return buffer.buffer_ && (static_cast<int>(buffer.buffer_->get_count()) >= requiredSize);
+}
+
/*! \libinternal \brief
* Allocates a device-side buffer.
* It is currently a caller's responsibility to call it only on not-yet allocated buffers.
return; // such calls are actually made with empty domains
}
GMX_ASSERT(buffer, "needs a buffer pointer");
- GMX_ASSERT(buffer->buffer_, "needs an initialized buffer pointer");
GMX_ASSERT(hostBuffer, "needs a host buffer pointer");
+ GMX_ASSERT(checkDeviceBuffer(*buffer, startingOffset + numValues),
+ "buffer too small or not initialized");
+
cl::sycl::buffer<ValueType>& syclBuffer = *buffer->buffer_;
cl::sycl::event ev = deviceStream.stream().submit([&](cl::sycl::handler& cgh) {
GMX_ASSERT(buffer, "needs a buffer pointer");
GMX_ASSERT(hostBuffer, "needs a host buffer pointer");
+ GMX_ASSERT(checkDeviceBuffer(*buffer, startingOffset + numValues),
+ "buffer too small or not initialized");
+
cl::sycl::buffer<ValueType>& syclBuffer = *buffer->buffer_;
cl::sycl::event ev = deviceStream.stream().submit([&](cl::sycl::handler& cgh) {
}
GMX_ASSERT(buffer, "needs a buffer pointer");
+ GMX_ASSERT(checkDeviceBuffer(*buffer, startingOffset + numValues),
+ "buffer too small or not initialized");
+
const ValueType pattern{};
cl::sycl::buffer<ValueType>& syclBuffer = *(buffer->buffer_);
});
}
-/*! \brief Check the validity of the device buffer.
- *
- * Checks if the buffer is valid and if its allocation is big enough.
- *
- * \param[in] buffer Device buffer to be checked.
- * \param[in] requiredSize Number of elements that the buffer will have to accommodate.
- *
- * \returns Whether the device buffer exists and has enough capacity.
- */
-template<typename T>
-static gmx_unused bool checkDeviceBuffer(DeviceBuffer<T> buffer, int requiredSize)
-{
- return buffer.buffer_ && (static_cast<int>(buffer.buffer_->get_count()) >= requiredSize);
-}
-
/*! \brief Create a texture object for an array of type ValueType.
*
* Creates the device buffer and copies read-only data for an array of type ValueType.
--- /dev/null
+/*
+ * This file is part of the GROMACS molecular simulation package.
+ *
+ * Copyright (c) 2020, by the GROMACS development team, led by
+ * Mark Abraham, David van der Spoel, Berk Hess, and Erik Lindahl,
+ * and including many others, as listed in the AUTHORS file in the
+ * top-level source directory and at http://www.gromacs.org.
+ *
+ * GROMACS is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public License
+ * as published by the Free Software Foundation; either version 2.1
+ * of the License, or (at your option) any later version.
+ *
+ * GROMACS is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with GROMACS; if not, see
+ * http://www.gnu.org/licenses, or write to the Free Software Foundation,
+ * Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
+ *
+ * If you want to redistribute modifications to GROMACS, please
+ * consider that scientific software is very special. Version
+ * control is crucial - bugs must be traceable. We will be happy to
+ * consider code for inclusion in the official distribution, but
+ * derived work must not be called official GROMACS. Details are found
+ * in the README & COPYING files - if they are missing, get the
+ * official version at http://www.gromacs.org.
+ *
+ * To help us fund GROMACS development, we humbly ask that you cite
+ * the research papers on the package. Check out http://www.gromacs.org.
+ */
+/*! \internal \file
+ *
+ * \brief Implements Leap-Frog using SYCL
+ *
+ * This file contains implementation of basic Leap-Frog integrator
+ * using SYCL, including class initialization, data-structures management
+ * and GPU kernel.
+ *
+ * \author Artem Zhmurov <zhmurov@gmail.com>
+ * \author Andrey Alekseenko <al42and@gmail.com>
+ *
+ * \ingroup module_mdlib
+ */
+#include "gmxpre.h"
+
+#include "gromacs/gpu_utils/devicebuffer.h"
+#include "gromacs/gpu_utils/gmxsycl.h"
+#include "gromacs/math/vec.h"
+#include "gromacs/mdlib/leapfrog_gpu.h"
+#include "gromacs/mdtypes/group.h"
+#include "gromacs/utility/arrayref.h"
+#include "gromacs/utility/fatalerror.h"
+#include "gromacs/utility/template_mp.h"
+
+namespace gmx
+{
+
+using cl::sycl::access::mode;
+
+/*! \brief Main kernel for the Leap-Frog integrator.
+ *
+ * The coordinates and velocities are updated on the GPU. Also saves the intermediate values of the coordinates for
+ * further use in constraints.
+ *
+ * Each GPU thread works with a single particle.
+ *
+ * \tparam numTempScaleValues The number of different T-couple values.
+ * \tparam velocityScaling Type of the Parrinello-Rahman velocity rescaling.
+ * \param cgh SYCL's command group handler.
+ * \param[in,out] a_x Coordinates to update upon integration.
+ * \param[out] a_xp A copy of the coordinates before the integration (for constraints).
+ * \param[in,out] a_v Velocities to update.
+ * \param[in] a_f Atomic forces.
+ * \param[in] a_inverseMasses Reciprocal masses.
+ * \param[in] dt Timestep.
+ * \param[in] a_lambdas Temperature scaling factors (one per group).
+ * \param[in] a_tempScaleGroups Mapping of atoms into groups.
+ * \param[in] prVelocityScalingMatrixDiagonal Diagonal elements of Parrinello-Rahman velocity scaling matrix
+ */
+template<NumTempScaleValues numTempScaleValues, VelocityScalingType velocityScaling>
+auto leapFrogKernel(
+ cl::sycl::handler& cgh,
+ DeviceAccessor<float3, mode::read_write> a_x,
+ DeviceAccessor<float3, mode::discard_write> a_xp,
+ DeviceAccessor<float3, mode::read_write> a_v,
+ DeviceAccessor<float3, mode::read> a_f,
+ DeviceAccessor<float, mode::read> a_inverseMasses,
+ float dt,
+ OptionalAccessor<float, mode::read, numTempScaleValues != NumTempScaleValues::None> a_lambdas,
+ OptionalAccessor<unsigned short, mode::read, numTempScaleValues == NumTempScaleValues::Multiple> a_tempScaleGroups,
+ float3 prVelocityScalingMatrixDiagonal)
+{
+ cgh.require(a_x);
+ cgh.require(a_xp);
+ cgh.require(a_v);
+ cgh.require(a_f);
+ cgh.require(a_inverseMasses);
+ if constexpr (numTempScaleValues != NumTempScaleValues::None)
+ {
+ cgh.require(a_lambdas);
+ }
+ if constexpr (numTempScaleValues == NumTempScaleValues::Multiple)
+ {
+ cgh.require(a_tempScaleGroups);
+ }
+
+ return [=](cl::sycl::id<1> itemIdx) {
+ const float3 x = a_x[itemIdx];
+ const float3 v = a_v[itemIdx];
+ const float3 f = a_f[itemIdx];
+ const float im = a_inverseMasses[itemIdx];
+ const float imdt = im * dt;
+
+ // Swapping places for xp and x so that the x will contain the updated coordinates and xp -
+ // the coordinates before update. This should be taken into account when (if) constraints
+ // are applied after the update: x and xp have to be passed to constraints in the 'wrong'
+ // order. See Issue #3727
+ a_xp[itemIdx] = x;
+
+ const float lambda = [=]() {
+ if constexpr (numTempScaleValues == NumTempScaleValues::None)
+ {
+ return 1.0F;
+ }
+ else if constexpr (numTempScaleValues == NumTempScaleValues::Single)
+ {
+ return a_lambdas[0];
+ }
+ else if constexpr (numTempScaleValues == NumTempScaleValues::Multiple)
+ {
+ const int tempScaleGroup = a_tempScaleGroups[itemIdx];
+ return a_lambdas[tempScaleGroup];
+ }
+ }();
+
+ const float3 prVelocityDelta = [=]() {
+ if constexpr (velocityScaling == VelocityScalingType::Diagonal)
+ {
+ return float3{ prVelocityScalingMatrixDiagonal[0] * v[0],
+ prVelocityScalingMatrixDiagonal[1] * v[1],
+ prVelocityScalingMatrixDiagonal[2] * v[2] };
+ }
+ else if constexpr (velocityScaling == VelocityScalingType::None)
+ {
+ return float3{ 0, 0, 0 };
+ }
+ }();
+
+ const float3 v_new = v * lambda - prVelocityDelta + f * imdt;
+ a_v[itemIdx] = v_new;
+ a_x[itemIdx] = x + v_new * dt;
+ };
+}
+
+// SYCL 1.2.1 requires providing a unique type for a kernel. Should not be needed for SYCL2020.
+template<NumTempScaleValues numTempScaleValues, VelocityScalingType velocityScaling>
+class LeapFrogKernelName;
+
+template<NumTempScaleValues numTempScaleValues, VelocityScalingType velocityScaling, class... Args>
+static cl::sycl::event launchLeapFrogKernel(const DeviceStream& deviceStream, int numAtoms, Args&&... args)
+{
+ // Should not be needed for SYCL2020.
+ using kernelNameType = LeapFrogKernelName<numTempScaleValues, velocityScaling>;
+
+ const cl::sycl::range<1> rangeAllAtoms(numAtoms);
+ cl::sycl::queue q = deviceStream.stream();
+
+ cl::sycl::event e = q.submit([&](cl::sycl::handler& cgh) {
+ auto kernel =
+ leapFrogKernel<numTempScaleValues, velocityScaling>(cgh, std::forward<Args>(args)...);
+ cgh.parallel_for<kernelNameType>(rangeAllAtoms, kernel);
+ });
+
+ return e;
+}
+
+static NumTempScaleValues getTempScalingType(bool doTemperatureScaling, int numTempScaleValues)
+{
+ if (!doTemperatureScaling)
+ {
+ return NumTempScaleValues::None;
+ }
+ else if (numTempScaleValues == 1)
+ {
+ return NumTempScaleValues::Single;
+ }
+ else if (numTempScaleValues > 1)
+ {
+ return NumTempScaleValues::Multiple;
+ }
+ else
+ {
+ gmx_incons("Temperature coupling was requested with no temperature coupling groups.");
+ }
+}
+
+/*! \brief Select templated kernel and launch it. */
+template<class... Args>
+static inline cl::sycl::event launchLeapFrogKernel(NumTempScaleValues tempScalingType,
+ VelocityScalingType prVelocityScalingType,
+ Args&&... args)
+{
+ GMX_ASSERT(prVelocityScalingType == VelocityScalingType::None
+ || prVelocityScalingType == VelocityScalingType::Diagonal,
+ "Only isotropic Parrinello-Rahman pressure coupling is supported.");
+
+ return dispatchTemplatedFunction(
+ [&](auto tempScalingType_, auto prScalingType_) {
+ return launchLeapFrogKernel<tempScalingType_, prScalingType_>(std::forward<Args>(args)...);
+ },
+ tempScalingType, prVelocityScalingType);
+}
+
+void LeapFrogGpu::integrate(DeviceBuffer<float3> d_x,
+ DeviceBuffer<float3> d_xp,
+ DeviceBuffer<float3> d_v,
+ DeviceBuffer<float3> d_f,
+ const real dt,
+ const bool doTemperatureScaling,
+ gmx::ArrayRef<const t_grp_tcstat> tcstat,
+ const bool doParrinelloRahman,
+ const float dtPressureCouple,
+ const matrix prVelocityScalingMatrix)
+{
+ if (doTemperatureScaling)
+ {
+ GMX_ASSERT(checkDeviceBuffer(d_lambdas_, numTempScaleValues_),
+ "Number of temperature scaling factors changed since it was set for the "
+ "last time.");
+ { // Explicitly limiting the scope of host accessor. Not strictly necessary here.
+ auto ha_lambdas_ = d_lambdas_.buffer_->get_access<mode::discard_write>();
+ for (int i = 0; i < numTempScaleValues_; i++)
+ {
+ ha_lambdas_[i] = tcstat[i].lambda;
+ }
+ }
+ }
+ NumTempScaleValues tempVelocityScalingType =
+ getTempScalingType(doTemperatureScaling, numTempScaleValues_);
+
+ VelocityScalingType prVelocityScalingType = VelocityScalingType::None;
+ if (doParrinelloRahman)
+ {
+ prVelocityScalingType = VelocityScalingType::Diagonal;
+ GMX_ASSERT(prVelocityScalingMatrix[YY][XX] == 0 && prVelocityScalingMatrix[ZZ][XX] == 0
+ && prVelocityScalingMatrix[ZZ][YY] == 0 && prVelocityScalingMatrix[XX][YY] == 0
+ && prVelocityScalingMatrix[XX][ZZ] == 0 && prVelocityScalingMatrix[YY][ZZ] == 0,
+ "Fully anisotropic Parrinello-Rahman pressure coupling is not yet supported "
+ "in GPU version of Leap-Frog integrator.");
+ prVelocityScalingMatrixDiagonal_ =
+ dtPressureCouple
+ * float3{ prVelocityScalingMatrix[XX][XX], prVelocityScalingMatrix[YY][YY],
+ prVelocityScalingMatrix[ZZ][ZZ] };
+ }
+
+ launchLeapFrogKernel(tempVelocityScalingType, prVelocityScalingType, deviceStream_, numAtoms_,
+ d_x, d_xp, d_v, d_f, d_inverseMasses_, dt, d_lambdas_, d_tempScaleGroups_,
+ prVelocityScalingMatrixDiagonal_);
+}
+
+LeapFrogGpu::LeapFrogGpu(const DeviceContext& deviceContext, const DeviceStream& deviceStream) :
+ deviceContext_(deviceContext),
+ deviceStream_(deviceStream),
+ numAtoms_(0)
+{
+}
+
+LeapFrogGpu::~LeapFrogGpu()
+{
+ freeDeviceBuffer(&d_inverseMasses_);
+}
+
+void LeapFrogGpu::set(const int numAtoms,
+ const real* inverseMasses,
+ const int numTempScaleValues,
+ const unsigned short* tempScaleGroups)
+{
+ numAtoms_ = numAtoms;
+ numTempScaleValues_ = numTempScaleValues;
+
+ reallocateDeviceBuffer(&d_inverseMasses_, numAtoms_, &numInverseMasses_,
+ &numInverseMassesAlloc_, deviceContext_);
+ copyToDeviceBuffer(&d_inverseMasses_, inverseMasses, 0, numAtoms_, deviceStream_,
+ GpuApiCallBehavior::Sync, nullptr);
+
+ // Temperature scale group map only used if there are more then one group
+ if (numTempScaleValues_ > 1)
+ {
+ reallocateDeviceBuffer(&d_tempScaleGroups_, numAtoms_, &numTempScaleGroups_,
+ &numTempScaleGroupsAlloc_, deviceContext_);
+ copyToDeviceBuffer(&d_tempScaleGroups_, tempScaleGroups, 0, numAtoms_, deviceStream_,
+ GpuApiCallBehavior::Sync, nullptr);
+ }
+
+ // If the temperature coupling is enabled, we need to make space for scaling factors
+ if (numTempScaleValues_ > 0)
+ {
+ reallocateDeviceBuffer(&d_lambdas_, numTempScaleValues_, &numLambdas_, &numLambdasAlloc_,
+ deviceContext_);
+ }
+}
+
+} // namespace gmx
biod.pnpi.spb.ru / alexxy / gromacs.git / commitdiff