Memory resource that wraps page-locked CUDA host allocation. More...
#include <vecmem/memory/cuda/host_memory_resource.hpp>
Inheritance diagram for vecmem::cuda::host_memory_resource:
Public Member Functions | |
| VECMEM_CUDA_EXPORT | host_memory_resource () |
| Default constructor. | |
| VECMEM_CUDA_EXPORT | ~host_memory_resource () |
| Destructor. | |
Private Member Functions | |
Function(s) implementing @c vecmem::memory_resource | |
| virtual VECMEM_CUDA_EXPORT void * | do_allocate (std::size_t, std::size_t) override final |
| Allocate page-locked host memory. | |
| virtual VECMEM_CUDA_EXPORT void | do_deallocate (void *p, std::size_t, std::size_t) override final |
| De-allocate a previously allocated page-locked memory block. | |
| virtual VECMEM_CUDA_EXPORT bool | do_is_equal (const memory_resource &other) const noexcept override final |
Compares *this for equality with other. | |
Detailed Description
Memory resource that wraps page-locked CUDA host allocation.
This is an allocator-type memory resource that allocates CUDA host memory, which is page-locked by default to allow faster transfer to the CUDA devices.
The documentation for this class was generated from the following files:
- cuda/include/vecmem/memory/cuda/host_memory_resource.hpp
- cuda/src/memory/host_memory_resource.cpp
