cache_aligned_resource#
[memory_allocation.cache_aligned_resource]
A cache_aligned_resource is a general-purpose memory resource class, which acts as a wrapper to another memory resource
to ensure that all allocations are aligned on cache line boundaries to avoid false sharing.
See the cache_aligned_allocator template class section for more information about false sharing avoidance.
// Defined in header <oneapi/tbb/cache_aligned_allocator.h>
namespace oneapi {
namespace tbb {
    class cache_aligned_resource {
    public:
        cache_aligned_resource();
        explicit cache_aligned_resource( std::pmr::memory_resource* );
        std::pmr::memory_resource* upstream_resource() const;
    private:
        void* do_allocate(size_t n, size_t alignment) override;
        void do_deallocate(void* p, size_t n, size_t alignment) override;
        bool do_is_equal(const std::pmr::memory_resource& other) const noexcept override;
    };
} // namespace tbb
} // namespace oneapi
Member Functions#
- 
cache_aligned_resource()#
 Constructs a
cache_aligned_resourceoverstd::pmr::get_default_resource().
- 
explicit cache_aligned_resource(std::pmr::memory_resource *r)#
 Constructs a
cache_aligned_resourceover the memory resourcer.
- 
std::pmr::memory_resource *upstream_resource() const#
 Returns the pointer to the underlying memory resource.
- 
void *do_allocate(size_t n, size_t alignment) override#
 Allocates
nbytes of memory on a cache-line boundary, with alignment not less than requested. The allocation may include extra memory for padding. Returns pointer to the allocated memory.
- 
void do_deallocate(void *p, size_t n, size_t alignment) override#
 Deallocates memory pointed to by p and any extra padding. Pointer
pmust be obtained withdo_allocate(n, alignment). The memory must not be deallocated beforehand.
- 
bool do_is_equal(const std::pmr::memory_resource &other) const noexcept override#
 Compares upstream memory resources of
*thisandother. Ifotheris not acache_aligned_resource, returns false.