mirror of
https://gitlab.kitware.com/vtk/vtk-m
synced 2024-09-08 21:33:55 +00:00
89 lines
3.1 KiB
C++
89 lines
3.1 KiB
C++
//============================================================================
|
|
// Copyright (c) Kitware, Inc.
|
|
// All rights reserved.
|
|
// See LICENSE.txt for details.
|
|
//
|
|
// This software is distributed WITHOUT ANY WARRANTY; without even
|
|
// the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR
|
|
// PURPOSE. See the above copyright notice for more information.
|
|
//============================================================================
|
|
|
|
#ifndef vtk_m_cont_cuda_internal_CudaAllocator_h
|
|
#define vtk_m_cont_cuda_internal_CudaAllocator_h
|
|
|
|
#include <vtkm/cont/vtkm_cont_export.h>
|
|
#include <vtkm/internal/ExportMacros.h>
|
|
|
|
#include <cstddef>
|
|
|
|
namespace vtkm
|
|
{
|
|
namespace cont
|
|
{
|
|
namespace cuda
|
|
{
|
|
namespace internal
|
|
{
|
|
|
|
/// Collection of cuda-specific memory management operations.
|
|
struct VTKM_CONT_EXPORT CudaAllocator
|
|
{
|
|
/// Returns true if all detected CUDA devices support pageable managed memory
|
|
/// that can be accessed concurrently by the CPU and GPUs.
|
|
static VTKM_CONT bool UsingManagedMemory();
|
|
|
|
/// Force CUDA allocations to occur with unmanaged memory (aka cudaMalloc).
|
|
static VTKM_CONT void ForceManagedMemoryOff();
|
|
|
|
/// Force CUDA allocations to occur with pageable managed memory.
|
|
/// If the current hardware doesn't support pageable managed memory
|
|
/// VTK-m will ignore the request and continue to use unmanaged memory (aka cudaMalloc).
|
|
static VTKM_CONT void ForceManagedMemoryOn();
|
|
|
|
static VTKM_CONT void ForceSyncMemoryAllocator();
|
|
static VTKM_CONT void ForceAsyncMemoryAllocator();
|
|
|
|
/// Returns true if the pointer is accessible from a CUDA device.
|
|
static VTKM_CONT bool IsDevicePointer(const void* ptr);
|
|
|
|
/// Returns true if the pointer is a CUDA pointer allocated with
|
|
/// cudaMallocManaged.
|
|
static VTKM_CONT bool IsManagedPointer(const void* ptr);
|
|
|
|
/// Will allocate memory that could be managed or unmanaged
|
|
static VTKM_CONT void* Allocate(std::size_t numBytes);
|
|
|
|
/// Explicitly allocate unmanaged memory even when the device supports
|
|
/// managed memory
|
|
static VTKM_CONT void* AllocateUnManaged(std::size_t numBytes);
|
|
|
|
/// Explicitly deallocate memory immediately.
|
|
static VTKM_CONT void Free(void* ptr);
|
|
|
|
/// \brief Defer deallocation of some memory
|
|
///
|
|
/// Keeps a pool of pointers to free until such a time of as we have
|
|
/// meet a threshold in total memory or number of pointers.
|
|
/// Currently the threshold to free all the pointers is 16MB
|
|
///
|
|
/// The reason for using this is that cudaFree causes a cudaSync call
|
|
/// to occur across all cuda devices and streams. This causes lots of stalls
|
|
/// when we are constructing small objects like virtuals and function pointers.
|
|
static VTKM_CONT void FreeDeferred(void* ptr, std::size_t numBytes);
|
|
|
|
static VTKM_CONT void PrepareForControl(const void* ptr, std::size_t numBytes);
|
|
|
|
static VTKM_CONT void PrepareForInput(const void* ptr, std::size_t numBytes);
|
|
static VTKM_CONT void PrepareForOutput(const void* ptr, std::size_t numBytes);
|
|
static VTKM_CONT void PrepareForInPlace(const void* ptr, std::size_t numBytes);
|
|
|
|
private:
|
|
static VTKM_CONT void Initialize();
|
|
};
|
|
}
|
|
}
|
|
}
|
|
} // end namespace vtkm::cont::cuda::internal
|
|
|
|
#endif // vtk_m_cont_cuda_internal_CudaAllocator_h
|