alb::v_100::shared_heap< Allocator, NumberOfChunks, ChunkSize > Class Template Reference

#include <shared_heap.hpp>

Public Types
using	allocator = Allocator

Public Member Functions
	shared_heap (size_t numberOfChunks, size_t chunkSize) noexcept
	shared_heap (shared_heap &&x) noexcept
shared_heap &	operator= (shared_heap &&x) noexcept
size_t	number_of_chunk () const noexcept
size_t	chunk_size () const noexcept
bool	owns (const block &b) const noexcept
block	allocate (size_t n) noexcept
void	deallocate (block &b) noexcept
void	deallocate_all () noexcept
bool	reallocate (block &b, size_t n) noexcept
bool	expand (block &b, size_t delta) noexcept

Static Public Attributes
static constexpr bool	supports_truncated_deallocation = true
static constexpr unsigned	alignment = Allocator::alignment

Detailed Description

template<class Allocator, size_t NumberOfChunks, size_t ChunkSize>
class alb::v_100::shared_heap< Allocator, NumberOfChunks, ChunkSize >

The SharedHeap implements a classic heap with a pre-allocated size of numberOfChunks_.value() * chunk_size_.value() It has a overhead of one bit per block and linear complexity for allocation and deallocation operations. It is thread safe, except the moment of instantiation. As far as possible only a shared lock + an atomic operation is used during the memory operations

Definition at line 49 of file shared_heap.hpp.

---

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

• alb/shared_heap.hpp