CeresEngine::TGPUBufferPool< T, B > Class Template Reference

A template class that automatically manages a uniform buffer pool. More...

#include <CeresEngine/RenderAPI/GPUBufferObject.hpp>

Public Types
using	BufferType = B< UInt8 >
	The buffer type.

Public Member Functions
	TGPUBufferPool (GPUDevice &device)
	Creates a new TBufferPool object.
void	reserve (const size_t length)
	Reserves space in the pool.
T &	operator[] (const size_t index)
	Gets a reference to the Uniform Buffer block stored at index.
void	flush (const GPUBufferSize offset=0, const GPUBufferSize size=GPUBuffer::whole)
	Flushes the cached memory data starting at offset with size bytes.
void	invalidate (const GPUBufferSize offset=0, const GPUBufferSize size=GPUBuffer::whole)
	Invalidates the cached memory data starting at offset with size bytes.

Private Member Functions
void	allocateBuffers (const size_t length)

Private Attributes
GPUDevicePtr	device
	The device to be used when allocating memory.
Vector< BufferType >	buffers
	A vector of allocated buffers.
const size_t	baseAllocationSize = 4 * 1024 * 1024
	The base allocation size used by the pool.
size_t	mLength = 0

Static Private Attributes
static constexpr size_t	alignment = ((sizeof(T) + 256 - 1) / 256) * 256
	The alignment to be used for each object in pool.

Detailed Description

template<typename T, template< typename > class B = GPUDynamicUniformBufferArray>
class CeresEngine::TGPUBufferPool< T, B >

A template class that automatically manages a uniform buffer pool.

This class will compute the required alignment (required by the GPU) and return references for objects of type T.

Template Parameters

T	the object type stored inside the uniform buffer

Member Typedef Documentation

BufferType

template<typename T , template< typename > class B = GPUDynamicUniformBufferArray>

using CeresEngine::TGPUBufferPool< T, B >::BufferType = B<UInt8>

The buffer type.

Constructor & Destructor Documentation

TGPUBufferPool()

template<typename T , template< typename > class B = GPUDynamicUniformBufferArray>

CeresEngine::TGPUBufferPool< T, B >::TGPUBufferPool ( GPUDevice &  device )

inlineexplicit

Creates a new TBufferPool object.

Parameters

device

The device to be used when allocating memory.

Member Function Documentation

allocateBuffers()

template<typename T , template< typename > class B = GPUDynamicUniformBufferArray>

void CeresEngine::TGPUBufferPool< T, B >::allocateBuffers ( const size_t  length )

inlineprivate

flush()

template<typename T , template< typename > class B = GPUDynamicUniformBufferArray>

void CeresEngine::TGPUBufferPool< T, B >::flush ( const GPUBufferSize  offset = 0, const GPUBufferSize  size = GPUBuffer::whole  )

inline

Flushes the cached memory data starting at offset with size bytes.

Parameters

offset	The starting offset to be flushed
size	The amount of bytes to be flushed

invalidate()

template<typename T , template< typename > class B = GPUDynamicUniformBufferArray>

void CeresEngine::TGPUBufferPool< T, B >::invalidate ( const GPUBufferSize  offset = 0, const GPUBufferSize  size = GPUBuffer::whole  )

inline

Invalidates the cached memory data starting at offset with size bytes.

Parameters

offset	The starting offset to be flushed
size	The amount of bytes to be flushed

operator[]()

template<typename T , template< typename > class B = GPUDynamicUniformBufferArray>

T & CeresEngine::TGPUBufferPool< T, B >::operator[] ( const size_t  index )

inline

Gets a reference to the Uniform Buffer block stored at index.

Parameters

index

The uniform buffer index to fetch

Returns

A reference to the uniform block

reserve()

template<typename T , template< typename > class B = GPUDynamicUniformBufferArray>

void CeresEngine::TGPUBufferPool< T, B >::reserve ( const size_t  length )

inline

Reserves space in the pool.

Parameters

length

The number of uniform objects to fit in the buffer

Member Data Documentation

alignment

template<typename T , template< typename > class B = GPUDynamicUniformBufferArray>

constexpr size_t CeresEngine::TGPUBufferPool< T, B >::alignment = ((sizeof(T) + 256 - 1) / 256) * 256

staticconstexprprivate

The alignment to be used for each object in pool.

baseAllocationSize

template<typename T , template< typename > class B = GPUDynamicUniformBufferArray>

const size_t CeresEngine::TGPUBufferPool< T, B >::baseAllocationSize = 4 * 1024 * 1024

private

The base allocation size used by the pool.

buffers

template<typename T , template< typename > class B = GPUDynamicUniformBufferArray>

Vector<BufferType> CeresEngine::TGPUBufferPool< T, B >::buffers

private

A vector of allocated buffers.

Each buffer has mBaseAllocationSize bytes.

device

template<typename T , template< typename > class B = GPUDynamicUniformBufferArray>

GPUDevicePtr CeresEngine::TGPUBufferPool< T, B >::device

private

The device to be used when allocating memory.

mLength

template<typename T , template< typename > class B = GPUDynamicUniformBufferArray>

size_t CeresEngine::TGPUBufferPool< T, B >::mLength = 0

private

---

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

• /Users/Rogiel/Developer/CeresEngine/Engine/Sources/CeresEngine/RenderAPI/Forward.hpp
• /Users/Rogiel/Developer/CeresEngine/Engine/Sources/CeresEngine/RenderAPI/GPUBufferObject.hpp