battery::managed_allocator Class Reference

#include <allocator.hpp>

Public Member Functions
CUDA void *	allocate (size_t bytes)
CUDA void	deallocate (void *data)

Detailed Description

An allocator using the managed memory of CUDA when the memory is allocated on the host. We delegate the allocation to global_allocator when the allocation is done on the device (since managed memory cannot be allocated in device functions).

Member Function Documentation

allocate()

CUDA void* battery::managed_allocator::allocate ( size_t  bytes )

inline

deallocate()

CUDA void battery::managed_allocator::deallocate ( void *  data )

inline

---

The documentation for this class was generated from the following file:

• include/battery/allocator.hpp