mirror of
https://gitlab.com/libeigen/eigen.git
synced 2024-12-21 07:19:46 +08:00
8fbd47052b
This commit enables the use of Eigen on HIP kernels / AMD GPUs. Support has been added along the same lines as what already exists for using Eigen in CUDA kernels / NVidia GPUs. Application code needs to explicitly define EIGEN_USE_HIP when using Eigen in HIP kernels. This is because some of the CUDA headers get picked up by default during Eigen compile (irrespective of whether or not the underlying compiler is CUDACC/NVCC, for e.g. Eigen/src/Core/arch/CUDA/Half.h). In order to maintain this behavior, the EIGEN_USE_HIP macro is used to switch to using the HIP version of those header files (see Eigen/Core and unsupported/Eigen/CXX11/Tensor) Use the "-DEIGEN_TEST_HIP" cmake option to enable the HIP specific unit tests.
181 lines
4.7 KiB
Plaintext
181 lines
4.7 KiB
Plaintext
// This file is part of Eigen, a lightweight C++ template library
|
|
// for linear algebra.
|
|
//
|
|
// Copyright (C) 2014 Benoit Steiner <benoit.steiner.goog@gmail.com>
|
|
// Copyright (C) 2013 Christian Seiler <christian@iwakd.de>
|
|
//
|
|
// This Source Code Form is subject to the terms of the Mozilla
|
|
// Public License v. 2.0. If a copy of the MPL was not distributed
|
|
// with this file, You can obtain one at http://mozilla.org/MPL/2.0/.
|
|
|
|
//#ifndef EIGEN_CXX11_TENSOR_MODULE
|
|
//#define EIGEN_CXX11_TENSOR_MODULE
|
|
|
|
#include "../../../Eigen/Core"
|
|
|
|
#if defined(EIGEN_USE_SYCL)
|
|
#undef min
|
|
#undef max
|
|
#undef isnan
|
|
#undef isinf
|
|
#undef isfinite
|
|
#include <CL/sycl.hpp>
|
|
#include <iostream>
|
|
#include <map>
|
|
#include <memory>
|
|
#include <utility>
|
|
#endif
|
|
|
|
#include <Eigen/src/Core/util/DisableStupidWarnings.h>
|
|
|
|
#include "../SpecialFunctions"
|
|
#include "src/util/CXX11Meta.h"
|
|
#include "src/util/MaxSizeVector.h"
|
|
|
|
/** \defgroup CXX11_Tensor_Module Tensor Module
|
|
*
|
|
* This module provides a Tensor class for storing arbitrarily indexed
|
|
* objects.
|
|
*
|
|
* \code
|
|
* #include <Eigen/CXX11/Tensor>
|
|
* \endcode
|
|
*/
|
|
|
|
#include <cmath>
|
|
#include <cstddef>
|
|
#include <cstring>
|
|
|
|
#ifdef _WIN32
|
|
typedef __int16 int16_t;
|
|
typedef unsigned __int16 uint16_t;
|
|
typedef __int32 int32_t;
|
|
typedef unsigned __int32 uint32_t;
|
|
typedef __int64 int64_t;
|
|
typedef unsigned __int64 uint64_t;
|
|
#include <windows.h>
|
|
#else
|
|
#include <stdint.h>
|
|
#include <unistd.h>
|
|
#endif
|
|
|
|
#if __cplusplus > 199711 || EIGEN_COMP_MSVC >= 1900
|
|
#include <random>
|
|
#endif
|
|
|
|
#ifdef _WIN32
|
|
#include <windows.h>
|
|
#elif defined(__APPLE__)
|
|
#include <mach/mach_time.h>
|
|
#else
|
|
#include <time.h>
|
|
#endif
|
|
|
|
#if defined(EIGEN_USE_LIBXSMM)
|
|
#include "libxsmm.h"
|
|
#endif
|
|
|
|
#ifdef EIGEN_USE_THREADS
|
|
#include "ThreadPool"
|
|
#endif
|
|
|
|
#ifdef EIGEN_USE_GPU
|
|
#include <iostream>
|
|
#if defined(EIGEN_USE_HIP)
|
|
#include <hip/hip_runtime.h>
|
|
#else
|
|
#include <cuda_runtime.h>
|
|
#endif
|
|
#if __cplusplus >= 201103L
|
|
#include <atomic>
|
|
#include <unistd.h>
|
|
#endif
|
|
#endif
|
|
|
|
#include "src/Tensor/TensorMacros.h"
|
|
#include "src/Tensor/TensorForwardDeclarations.h"
|
|
#include "src/Tensor/TensorMeta.h"
|
|
#include "src/Tensor/TensorFunctors.h"
|
|
#include "src/Tensor/TensorCostModel.h"
|
|
#include "src/Tensor/TensorDeviceDefault.h"
|
|
#include "src/Tensor/TensorDeviceThreadPool.h"
|
|
#if defined(EIGEN_USE_HIP)
|
|
#include "src/Tensor/TensorDeviceHip.h"
|
|
#else
|
|
#include "src/Tensor/TensorDeviceCuda.h"
|
|
#endif
|
|
#include "src/Tensor/TensorDeviceSycl.h"
|
|
#include "src/Tensor/TensorIndexList.h"
|
|
#include "src/Tensor/TensorDimensionList.h"
|
|
#include "src/Tensor/TensorDimensions.h"
|
|
#include "src/Tensor/TensorInitializer.h"
|
|
#include "src/Tensor/TensorTraits.h"
|
|
#include "src/Tensor/TensorRandom.h"
|
|
#include "src/Tensor/TensorUInt128.h"
|
|
#include "src/Tensor/TensorIntDiv.h"
|
|
#include "src/Tensor/TensorGlobalFunctions.h"
|
|
|
|
#include "src/Tensor/TensorBase.h"
|
|
|
|
#include "src/Tensor/TensorEvaluator.h"
|
|
#include "src/Tensor/TensorExpr.h"
|
|
#include "src/Tensor/TensorReduction.h"
|
|
#if defined(EIGEN_USE_HIP)
|
|
#include "src/Tensor/TensorReductionHip.h"
|
|
#else
|
|
#include "src/Tensor/TensorReductionCuda.h"
|
|
#endif
|
|
#include "src/Tensor/TensorArgMax.h"
|
|
#include "src/Tensor/TensorConcatenation.h"
|
|
#include "src/Tensor/TensorContractionMapper.h"
|
|
#include "src/Tensor/TensorContractionBlocking.h"
|
|
#include "src/Tensor/TensorContraction.h"
|
|
#include "src/Tensor/TensorContractionThreadPool.h"
|
|
#if defined(EIGEN_USE_HIP)
|
|
#include "src/Tensor/TensorContractionHip.h"
|
|
#else
|
|
#include "src/Tensor/TensorContractionCuda.h"
|
|
#endif
|
|
#include "src/Tensor/TensorConversion.h"
|
|
#if defined(EIGEN_USE_HIP)
|
|
#include "src/Tensor/TensorConvolutionHip.h"
|
|
#else
|
|
#include "src/Tensor/TensorConvolution.h"
|
|
#endif
|
|
#include "src/Tensor/TensorFFT.h"
|
|
#include "src/Tensor/TensorPatch.h"
|
|
#include "src/Tensor/TensorImagePatch.h"
|
|
#include "src/Tensor/TensorVolumePatch.h"
|
|
#include "src/Tensor/TensorBroadcasting.h"
|
|
#include "src/Tensor/TensorChipping.h"
|
|
#include "src/Tensor/TensorInflation.h"
|
|
#include "src/Tensor/TensorLayoutSwap.h"
|
|
#include "src/Tensor/TensorMorphing.h"
|
|
#include "src/Tensor/TensorPadding.h"
|
|
#include "src/Tensor/TensorReverse.h"
|
|
#include "src/Tensor/TensorShuffling.h"
|
|
#include "src/Tensor/TensorStriding.h"
|
|
#include "src/Tensor/TensorCustomOp.h"
|
|
#include "src/Tensor/TensorEvalTo.h"
|
|
#include "src/Tensor/TensorForcedEval.h"
|
|
#include "src/Tensor/TensorGenerator.h"
|
|
#include "src/Tensor/TensorAssign.h"
|
|
#include "src/Tensor/TensorScan.h"
|
|
#include "src/Tensor/TensorTrace.h"
|
|
|
|
#include "src/Tensor/TensorSycl.h"
|
|
#include "src/Tensor/TensorExecutor.h"
|
|
#include "src/Tensor/TensorDevice.h"
|
|
|
|
#include "src/Tensor/TensorStorage.h"
|
|
#include "src/Tensor/Tensor.h"
|
|
#include "src/Tensor/TensorFixedSize.h"
|
|
#include "src/Tensor/TensorMap.h"
|
|
#include "src/Tensor/TensorRef.h"
|
|
|
|
#include "src/Tensor/TensorIO.h"
|
|
|
|
#include <Eigen/src/Core/util/ReenableStupidWarnings.h>
|
|
|
|
//#endif // EIGEN_CXX11_TENSOR_MODULE
|