#ifndef _VKRAYTRACINGUTIL_HPP
#define _VKRAYTRACINGUTIL_HPP
/*-------------------------------------------------------------------------
 * Vulkan CTS Framework
 * --------------------
 *
 * Copyright (c) 2020 The Khronos Group Inc.
 *
 * Licensed under the Apache License, Version 2.0 (the "License");
 * you may not use this file except in compliance with the License.
 * You may obtain a copy of the License at
 *
 *      http://www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 *
 *//*!
 * \file
 * \brief Vulkan ray tracing utility.
 *//*--------------------------------------------------------------------*/

#include "vkDefs.hpp"
#include "vkRef.hpp"
#include "vkMemUtil.hpp"
#include "vkBufferWithMemory.hpp"

#include "deFloat16.h"

#include "tcuVector.hpp"
#include "tcuVectorType.hpp"

#include <vector>

namespace vk
{
const VkTransformMatrixKHR identityMatrix3x4 = { { { 1.0f, 0.0f, 0.0f, 0.0f }, { 0.0f, 1.0f, 0.0f, 0.0f }, { 0.0f, 0.0f, 1.0f, 0.0f } } };

template<typename T>
inline de::SharedPtr<Move<T>> makeVkSharedPtr(Move<T> move)
{
	return de::SharedPtr<Move<T>>(new Move<T>(move));
}

template<typename T>
inline de::SharedPtr<de::MovePtr<T> > makeVkSharedPtr(de::MovePtr<T> movePtr)
{
	return de::SharedPtr<de::MovePtr<T> >(new de::MovePtr<T>(movePtr));
}

template<typename T>
inline const T* dataOrNullPtr(const std::vector<T>& v)
{
	return (v.empty() ? DE_NULL : v.data());
}

template<typename T>
inline T* dataOrNullPtr(std::vector<T>& v)
{
	return (v.empty() ? DE_NULL : v.data());
}

inline std::string updateRayTracingGLSL (const std::string& str)
{
	return str;
}

std::string getCommonRayGenerationShader (void);

// Get lowercase version of the format name with no VK_FORMAT_ prefix.
std::string getFormatSimpleName (vk::VkFormat format);

// Checks the given vertex buffer format is valid for acceleration structures.
// Note: VK_KHR_get_physical_device_properties2 and VK_KHR_acceleration_structure are supposed to be supported.
void checkAccelerationStructureVertexBufferFormat (const vk::InstanceInterface &vki, vk::VkPhysicalDevice physicalDevice, vk::VkFormat format);

class RaytracedGeometryBase
{
public:
								RaytracedGeometryBase			()										= delete;
								RaytracedGeometryBase			(const RaytracedGeometryBase& geometry)	= delete;
								RaytracedGeometryBase			(VkGeometryTypeKHR geometryType, VkFormat vertexFormat, VkIndexType indexType);
								virtual ~RaytracedGeometryBase	();

	inline VkGeometryTypeKHR	getGeometryType					(void) const								{ return m_geometryType; }
	inline bool					isTrianglesType					(void) const								{ return m_geometryType == VK_GEOMETRY_TYPE_TRIANGLES_KHR; }
	inline VkFormat				getVertexFormat					(void) const								{ return m_vertexFormat; }
	inline VkIndexType			getIndexType					(void) const								{ return m_indexType; }
	inline bool					usesIndices						(void) const								{ return m_indexType != VK_INDEX_TYPE_NONE_KHR; }
	inline VkGeometryFlagsKHR	getGeometryFlags				(void) const								{ return m_geometryFlags; }
	inline void					setGeometryFlags				(const VkGeometryFlagsKHR geometryFlags)	{ m_geometryFlags = geometryFlags; }
	virtual deUint32			getVertexCount					(void) const								= 0;
	virtual const deUint8*		getVertexPointer				(void) const								= 0;
	virtual VkDeviceSize		getVertexStride					(void) const								= 0;
	virtual VkDeviceSize		getAABBStride					(void) const								= 0;
	virtual size_t				getVertexByteSize				(void) const								= 0;
	virtual deUint32			getIndexCount					(void) const								= 0;
	virtual const deUint8*		getIndexPointer					(void) const								= 0;
	virtual VkDeviceSize		getIndexStride					(void) const								= 0;
	virtual size_t				getIndexByteSize				(void) const								= 0;
	virtual deUint32			getPrimitiveCount				(void) const								= 0;
	virtual void				addVertex						(const tcu::Vec3& vertex)					= 0;
	virtual void				addIndex						(const deUint32& index)						= 0;
private:
	VkGeometryTypeKHR			m_geometryType;
	VkFormat					m_vertexFormat;
	VkIndexType					m_indexType;
	VkGeometryFlagsKHR			m_geometryFlags;
};

template <typename T>
inline T convertSatRte (float f)
{
	// \note Doesn't work for 64-bit types
	DE_STATIC_ASSERT(sizeof(T) < sizeof(deUint64));
	DE_STATIC_ASSERT((-3 % 2 != 0) && (-4 % 2 == 0));

	deInt64	minVal	= std::numeric_limits<T>::min();
	deInt64 maxVal	= std::numeric_limits<T>::max();
	float	q		= deFloatFrac(f);
	deInt64 intVal	= (deInt64)(f-q);

	// Rounding.
	if (q == 0.5f)
	{
		if (intVal % 2 != 0)
			intVal++;
	}
	else if (q > 0.5f)
		intVal++;
	// else Don't add anything

	// Saturate.
	intVal = de::max(minVal, de::min(maxVal, intVal));

	return (T)intVal;
}

// Converts float to signed integer with variable width.
// Source float is assumed to be in the [-1, 1] range.
template <typename T>
inline T deFloat32ToSNorm (float src)
{
	DE_STATIC_ASSERT(std::numeric_limits<T>::is_integer && std::numeric_limits<T>::is_signed);
	const T range	= std::numeric_limits<T>::max();
	const T intVal	= convertSatRte<T>(src * static_cast<float>(range));
	return de::clamp<T>(intVal, -range, range);
}

typedef tcu::Vector<deFloat16, 2>			Vec2_16;
typedef tcu::Vector<deFloat16, 3>			Vec3_16;
typedef tcu::Vector<deFloat16, 4>			Vec4_16;
typedef tcu::Vector<deInt16, 2>				Vec2_16SNorm;
typedef tcu::Vector<deInt16, 3>				Vec3_16SNorm;
typedef tcu::Vector<deInt16, 4>				Vec4_16SNorm;
typedef tcu::Vector<deInt8, 2>				Vec2_8SNorm;
typedef tcu::Vector<deInt8, 3>				Vec3_8SNorm;
typedef tcu::Vector<deInt8, 4>				Vec4_8SNorm;

template<typename V>	VkFormat			vertexFormatFromType				();
template<>				inline VkFormat		vertexFormatFromType<tcu::Vec2>		()							{ return VK_FORMAT_R32G32_SFLOAT; }
template<>				inline VkFormat		vertexFormatFromType<tcu::Vec3>		()							{ return VK_FORMAT_R32G32B32_SFLOAT; }
template<>				inline VkFormat		vertexFormatFromType<tcu::Vec4>		()							{ return VK_FORMAT_R32G32B32A32_SFLOAT; }
template<>				inline VkFormat		vertexFormatFromType<Vec2_16>		()							{ return VK_FORMAT_R16G16_SFLOAT; }
template<>				inline VkFormat		vertexFormatFromType<Vec3_16>		()							{ return VK_FORMAT_R16G16B16_SFLOAT; }
template<>				inline VkFormat		vertexFormatFromType<Vec4_16>		()							{ return VK_FORMAT_R16G16B16A16_SFLOAT; }
template<>				inline VkFormat		vertexFormatFromType<Vec2_16SNorm>	()							{ return VK_FORMAT_R16G16_SNORM; }
template<>				inline VkFormat		vertexFormatFromType<Vec3_16SNorm>	()							{ return VK_FORMAT_R16G16B16_SNORM; }
template<>				inline VkFormat		vertexFormatFromType<Vec4_16SNorm>	()							{ return VK_FORMAT_R16G16B16A16_SNORM; }
template<>				inline VkFormat		vertexFormatFromType<tcu::DVec2>	()							{ return VK_FORMAT_R64G64_SFLOAT; }
template<>				inline VkFormat		vertexFormatFromType<tcu::DVec3>	()							{ return VK_FORMAT_R64G64B64_SFLOAT; }
template<>				inline VkFormat		vertexFormatFromType<tcu::DVec4>	()							{ return VK_FORMAT_R64G64B64A64_SFLOAT; }
template<>				inline VkFormat		vertexFormatFromType<Vec2_8SNorm>	()							{ return VK_FORMAT_R8G8_SNORM; }
template<>				inline VkFormat		vertexFormatFromType<Vec3_8SNorm>	()							{ return VK_FORMAT_R8G8B8_SNORM; }
template<>				inline VkFormat		vertexFormatFromType<Vec4_8SNorm>	()							{ return VK_FORMAT_R8G8B8A8_SNORM; }

struct EmptyIndex {};
template<typename I>	VkIndexType			indexTypeFromType					();
template<>				inline VkIndexType	indexTypeFromType<deUint16>			()							{ return VK_INDEX_TYPE_UINT16; }
template<>				inline VkIndexType	indexTypeFromType<deUint32>			()							{ return VK_INDEX_TYPE_UINT32; }
template<>				inline VkIndexType	indexTypeFromType<EmptyIndex>		()							{ return VK_INDEX_TYPE_NONE_KHR; }

template<typename V>	V					convertFloatTo						(const tcu::Vec3& vertex);
template<>				inline tcu::Vec2	convertFloatTo<tcu::Vec2>			(const tcu::Vec3& vertex)	{ return tcu::Vec2(vertex.x(), vertex.y()); }
template<>				inline tcu::Vec3	convertFloatTo<tcu::Vec3>			(const tcu::Vec3& vertex)	{ return vertex; }
template<>				inline tcu::Vec4	convertFloatTo<tcu::Vec4>			(const tcu::Vec3& vertex)	{ return tcu::Vec4(vertex.x(), vertex.y(), vertex.z(), 0.0f); }
template<>				inline Vec2_16		convertFloatTo<Vec2_16>				(const tcu::Vec3& vertex)	{ return Vec2_16(deFloat32To16(vertex.x()), deFloat32To16(vertex.y())); }
template<>				inline Vec3_16		convertFloatTo<Vec3_16>				(const tcu::Vec3& vertex)	{ return Vec3_16(deFloat32To16(vertex.x()), deFloat32To16(vertex.y()), deFloat32To16(vertex.z())); }
template<>				inline Vec4_16		convertFloatTo<Vec4_16>				(const tcu::Vec3& vertex)	{ return Vec4_16(deFloat32To16(vertex.x()), deFloat32To16(vertex.y()), deFloat32To16(vertex.z()), deFloat32To16(0.0f)); }
template<>				inline Vec2_16SNorm	convertFloatTo<Vec2_16SNorm>		(const tcu::Vec3& vertex)	{ return Vec2_16SNorm(deFloat32ToSNorm<deInt16>(vertex.x()), deFloat32ToSNorm<deInt16>(vertex.y())); }
template<>				inline Vec3_16SNorm	convertFloatTo<Vec3_16SNorm>		(const tcu::Vec3& vertex)	{ return Vec3_16SNorm(deFloat32ToSNorm<deInt16>(vertex.x()), deFloat32ToSNorm<deInt16>(vertex.y()), deFloat32ToSNorm<deInt16>(vertex.z())); }
template<>				inline Vec4_16SNorm	convertFloatTo<Vec4_16SNorm>		(const tcu::Vec3& vertex)	{ return Vec4_16SNorm(deFloat32ToSNorm<deInt16>(vertex.x()), deFloat32ToSNorm<deInt16>(vertex.y()), deFloat32ToSNorm<deInt16>(vertex.z()), deFloat32ToSNorm<deInt16>(0.0f)); }
template<>				inline tcu::DVec2	convertFloatTo<tcu::DVec2>			(const tcu::Vec3& vertex)	{ return tcu::DVec2(static_cast<double>(vertex.x()), static_cast<double>(vertex.y())); }
template<>				inline tcu::DVec3	convertFloatTo<tcu::DVec3>			(const tcu::Vec3& vertex)	{ return tcu::DVec3(static_cast<double>(vertex.x()), static_cast<double>(vertex.y()), static_cast<double>(vertex.z())); }
template<>				inline tcu::DVec4	convertFloatTo<tcu::DVec4>			(const tcu::Vec3& vertex)	{ return tcu::DVec4(static_cast<double>(vertex.x()), static_cast<double>(vertex.y()), static_cast<double>(vertex.z()), 0.0); }
template<>				inline Vec2_8SNorm	convertFloatTo<Vec2_8SNorm>			(const tcu::Vec3& vertex)	{ return Vec2_8SNorm(deFloat32ToSNorm<deInt8>(vertex.x()), deFloat32ToSNorm<deInt8>(vertex.y())); }
template<>				inline Vec3_8SNorm	convertFloatTo<Vec3_8SNorm>			(const tcu::Vec3& vertex)	{ return Vec3_8SNorm(deFloat32ToSNorm<deInt8>(vertex.x()), deFloat32ToSNorm<deInt8>(vertex.y()), deFloat32ToSNorm<deInt8>(vertex.z())); }
template<>				inline Vec4_8SNorm	convertFloatTo<Vec4_8SNorm>			(const tcu::Vec3& vertex)	{ return Vec4_8SNorm(deFloat32ToSNorm<deInt8>(vertex.x()), deFloat32ToSNorm<deInt8>(vertex.y()), deFloat32ToSNorm<deInt8>(vertex.z()), deFloat32ToSNorm<deInt8>(0.0f)); }

template<typename V>	V					convertIndexTo						(deUint32 index);
template<>				inline EmptyIndex	convertIndexTo<EmptyIndex>			(deUint32 index)			{ DE_UNREF(index); TCU_THROW(TestError, "Cannot add empty index"); }
template<>				inline deUint16		convertIndexTo<deUint16>			(deUint32 index)			{ return static_cast<deUint16>(index); }
template<>				inline deUint32		convertIndexTo<deUint32>			(deUint32 index)			{ return index; }

template<typename V, typename I>
class RaytracedGeometry : public RaytracedGeometryBase
{
public:
						RaytracedGeometry			()									= delete;
						RaytracedGeometry			(const RaytracedGeometry& geometry)	= delete;
						RaytracedGeometry			(VkGeometryTypeKHR geometryType, deUint32 paddingBlocks = 0u);
						RaytracedGeometry			(VkGeometryTypeKHR geometryType, const std::vector<V>& vertices, const std::vector<I>& indices = std::vector<I>(), deUint32 paddingBlocks = 0u);

	deUint32			getVertexCount				(void) const override;
	const deUint8*		getVertexPointer			(void) const override;
	VkDeviceSize		getVertexStride				(void) const override;
	VkDeviceSize		getAABBStride				(void) const override;
	size_t				getVertexByteSize			(void) const override;
	deUint32			getIndexCount				(void) const override;
	const deUint8*		getIndexPointer				(void) const override;
	VkDeviceSize		getIndexStride				(void) const override;
	size_t				getIndexByteSize			(void) const override;
	deUint32			getPrimitiveCount			(void) const override;

	void				addVertex					(const tcu::Vec3& vertex) override;
	void				addIndex					(const deUint32& index) override;

private:
	void				init						();					// To be run in constructors.
	void				checkGeometryType			() const;			// Checks geometry type is valid.
	void				calcBlockSize				();					// Calculates and saves vertex buffer block size.
	size_t				getBlockSize				() const;			// Return stored vertex buffer block size.
	void				addNativeVertex				(const V& vertex);	// Adds new vertex in native format.

	// The implementation below stores vertices as byte blocks to take the requested padding into account. m_vertices is the array
	// of bytes containing vertex data.
	//
	// For triangles, the padding block has a size that is a multiple of the vertex size and each vertex is stored in a byte block
	// equivalent to:
	//
	//	struct Vertex
	//	{
	//		V		vertex;
	//		deUint8	padding[m_paddingBlocks * sizeof(V)];
	//	};
	//
	// For AABBs, the padding block has a size that is a multiple of kAABBPadBaseSize (see below) and vertices are stored in pairs
	// before the padding block. This is equivalent to:
	//
	//		struct VertexPair
	//		{
	//			V		vertices[2];
	//			deUint8	padding[m_paddingBlocks * kAABBPadBaseSize];
	//		};
	//
	// The size of each pseudo-structure above is saved to one of the correspoding union members below.
	union BlockSize
	{
		size_t trianglesBlockSize;
		size_t aabbsBlockSize;
	};

	const deUint32			m_paddingBlocks;
	size_t					m_vertexCount;
	std::vector<deUint8>	m_vertices;			// Vertices are stored as byte blocks.
	std::vector<I>			m_indices;			// Indices are stored natively.
	BlockSize				m_blockSize;		// For m_vertices.

	// Data sizes.
	static constexpr size_t	kVertexSize			= sizeof(V);
	static constexpr size_t	kIndexSize			= sizeof(I);
	static constexpr size_t	kAABBPadBaseSize	= 8; // As required by the spec.
};

template<typename V, typename I>
RaytracedGeometry<V, I>::RaytracedGeometry (VkGeometryTypeKHR geometryType, deUint32 paddingBlocks)
	: RaytracedGeometryBase(geometryType, vertexFormatFromType<V>(), indexTypeFromType<I>())
	, m_paddingBlocks(paddingBlocks)
	, m_vertexCount(0)
{
	init();
}

template<typename V, typename I>
RaytracedGeometry<V,I>::RaytracedGeometry (VkGeometryTypeKHR geometryType, const std::vector<V>& vertices, const std::vector<I>& indices, deUint32 paddingBlocks)
	: RaytracedGeometryBase(geometryType, vertexFormatFromType<V>(), indexTypeFromType<I>())
	, m_paddingBlocks(paddingBlocks)
	, m_vertexCount(0)
	, m_vertices()
	, m_indices(indices)
{
	init();
	for (const auto& vertex : vertices)
		addNativeVertex(vertex);
}

template<typename V, typename I>
deUint32 RaytracedGeometry<V,I>::getVertexCount (void) const
{
	return (isTrianglesType() ? static_cast<deUint32>(m_vertexCount) : 0u);
}

template<typename V, typename I>
const deUint8* RaytracedGeometry<V, I>::getVertexPointer (void) const
{
	DE_ASSERT(!m_vertices.empty());
	return reinterpret_cast<const deUint8*>(m_vertices.data());
}

template<typename V, typename I>
VkDeviceSize RaytracedGeometry<V,I>::getVertexStride (void) const
{
	return ((!isTrianglesType()) ? 0ull : static_cast<VkDeviceSize>(getBlockSize()));
}

template<typename V, typename I>
VkDeviceSize RaytracedGeometry<V, I>::getAABBStride (void) const
{
	return (isTrianglesType() ? 0ull : static_cast<VkDeviceSize>(getBlockSize()));
}

template<typename V, typename I>
size_t RaytracedGeometry<V, I>::getVertexByteSize (void) const
{
	return m_vertices.size();
}

template<typename V, typename I>
deUint32 RaytracedGeometry<V, I>::getIndexCount (void) const
{
	return static_cast<deUint32>(isTrianglesType() ? m_indices.size() : 0);
}

template<typename V, typename I>
const deUint8* RaytracedGeometry<V, I>::getIndexPointer (void) const
{
	const auto indexCount = getIndexCount();
	DE_UNREF(indexCount); // For release builds.
	DE_ASSERT(indexCount > 0u);

	return reinterpret_cast<const deUint8*>(m_indices.data());
}

template<typename V, typename I>
VkDeviceSize RaytracedGeometry<V, I>::getIndexStride (void) const
{
	return static_cast<VkDeviceSize>(kIndexSize);
}

template<typename V, typename I>
size_t RaytracedGeometry<V, I>::getIndexByteSize (void) const
{
	const auto indexCount = getIndexCount();
	DE_ASSERT(indexCount > 0u);

	return (indexCount * kIndexSize);
}

template<typename V, typename I>
deUint32 RaytracedGeometry<V,I>::getPrimitiveCount (void) const
{
	return static_cast<deUint32>(isTrianglesType() ? (usesIndices() ? m_indices.size() / 3 : m_vertexCount / 3) : (m_vertexCount / 2));
}

template<typename V, typename I>
void RaytracedGeometry<V, I>::addVertex (const tcu::Vec3& vertex)
{
	addNativeVertex(convertFloatTo<V>(vertex));
}

template<typename V, typename I>
void RaytracedGeometry<V, I>::addNativeVertex (const V& vertex)
{
	const auto oldSize			= m_vertices.size();
	const auto blockSize		= getBlockSize();

	if (isTrianglesType())
	{
		// Reserve new block, copy vertex at the beginning of the new block.
		m_vertices.resize(oldSize + blockSize, deUint8{0});
		deMemcpy(&m_vertices[oldSize], &vertex, kVertexSize);
	}
	else // AABB
	{
		if (m_vertexCount % 2 == 0)
		{
			// New block needed.
			m_vertices.resize(oldSize + blockSize, deUint8{0});
			deMemcpy(&m_vertices[oldSize], &vertex, kVertexSize);
		}
		else
		{
			// Insert in the second position of last existing block.
			//
			//												Vertex Size
			//												+-------+
			//	+-------------+------------+----------------------------------------+
			//	|             |            |      ...       | vertex vertex padding |
			//	+-------------+------------+----------------+-----------------------+
			//												+-----------------------+
			//														Block Size
			//	+-------------------------------------------------------------------+
			//							Old Size
			//
			deMemcpy(&m_vertices[oldSize - blockSize + kVertexSize], &vertex, kVertexSize);
		}
	}

	++m_vertexCount;
}

template<typename V, typename I>
void RaytracedGeometry<V, I>::addIndex (const deUint32& index)
{
	m_indices.push_back(convertIndexTo<I>(index));
}

template<typename V, typename I>
void RaytracedGeometry<V, I>::init ()
{
	checkGeometryType();
	calcBlockSize();
}

template<typename V, typename I>
void RaytracedGeometry<V, I>::checkGeometryType () const
{
	const auto geometryType = getGeometryType();
	DE_UNREF(geometryType); // For release builds.
	DE_ASSERT(geometryType == VK_GEOMETRY_TYPE_TRIANGLES_KHR || geometryType == VK_GEOMETRY_TYPE_AABBS_KHR);
}

template<typename V, typename I>
void RaytracedGeometry<V, I>::calcBlockSize ()
{
	if (isTrianglesType())
		m_blockSize.trianglesBlockSize = kVertexSize * static_cast<size_t>(1u + m_paddingBlocks);
	else
		m_blockSize.aabbsBlockSize = 2 * kVertexSize + m_paddingBlocks * kAABBPadBaseSize;
}

template<typename V, typename I>
size_t RaytracedGeometry<V, I>::getBlockSize () const
{
	return (isTrianglesType() ? m_blockSize.trianglesBlockSize : m_blockSize.aabbsBlockSize);
}

de::SharedPtr<RaytracedGeometryBase> makeRaytracedGeometry (VkGeometryTypeKHR geometryType, VkFormat vertexFormat, VkIndexType indexType, bool padVertices = false);

VkDeviceAddress getBufferDeviceAddress ( const DeviceInterface&	vkd,
										 const VkDevice			device,
										 const VkBuffer			buffer,
										 VkDeviceSize			offset );

class SerialStorage
{
public:
	enum
	{
		DE_SERIALIZED_FIELD(DRIVER_UUID,		VK_UUID_SIZE),		// VK_UUID_SIZE bytes of data matching VkPhysicalDeviceIDProperties::driverUUID
		DE_SERIALIZED_FIELD(COMPAT_UUID,		VK_UUID_SIZE),		// VK_UUID_SIZE bytes of data identifying the compatibility for comparison using vkGetDeviceAccelerationStructureCompatibilityKHR
		DE_SERIALIZED_FIELD(SERIALIZED_SIZE,	sizeof(deUint64)),	// A 64-bit integer of the total size matching the value queried using VK_QUERY_TYPE_ACCELERATION_STRUCTURE_SERIALIZATION_SIZE_KHR
		DE_SERIALIZED_FIELD(DESERIALIZED_SIZE,	sizeof(deUint64)),	// A 64-bit integer of the deserialized size to be passed in to VkAccelerationStructureCreateInfoKHR::size
		DE_SERIALIZED_FIELD(HANDLES_COUNT,		sizeof(deUint64)),	// A 64-bit integer of the count of the number of acceleration structure handles following. This will be zero for a bottom-level acceleration structure.
		SERIAL_STORAGE_SIZE_MIN
	};

											SerialStorage		() = delete;
											SerialStorage		(const DeviceInterface&						vk,
																 const VkDevice								device,
																 Allocator&									allocator,
																 const VkAccelerationStructureBuildTypeKHR	buildType,
																 const VkDeviceSize							storageSize);

	VkDeviceOrHostAddressKHR				getAddress			(const DeviceInterface&	vk,
																 const VkDevice			device);
	VkDeviceOrHostAddressConstKHR			getAddressConst		(const DeviceInterface&	vk,
																 const VkDevice			device);
	VkDeviceSize							getStorageSize		();
	deUint64								getDeserializedSize	();

protected:
	VkAccelerationStructureBuildTypeKHR		m_buildType;
	de::MovePtr<BufferWithMemory>			m_buffer;
	VkDeviceSize							m_storageSize;

};

class BottomLevelAccelerationStructure
{
public:
	static deUint32										getRequiredAllocationCount				(void);

														BottomLevelAccelerationStructure		();
														BottomLevelAccelerationStructure		(const BottomLevelAccelerationStructure&		other) = delete;
	virtual												~BottomLevelAccelerationStructure		();

	virtual void										setGeometryData							(const std::vector<tcu::Vec3>&					geometryData,
																								 const bool										triangles,
																								 const VkGeometryFlagsKHR						geometryFlags			= 0u );
	virtual void										setDefaultGeometryData					(const VkShaderStageFlagBits					testStage,
																								 const VkGeometryFlagsKHR						geometryFlags			= 0u );
	virtual void										setGeometryCount						(const size_t									geometryCount);
	virtual void										addGeometry								(de::SharedPtr<RaytracedGeometryBase>&			raytracedGeometry);
	virtual void										addGeometry								(const std::vector<tcu::Vec3>&					geometryData,
																								 const bool										triangles,
																								 const VkGeometryFlagsKHR						geometryFlags			= 0u );

	virtual void										setBuildType							(const VkAccelerationStructureBuildTypeKHR		buildType) = DE_NULL;
	virtual void										setCreateFlags							(const VkAccelerationStructureCreateFlagsKHR	createFlags) = DE_NULL;
	virtual void										setCreateGeneric						(bool											createGeneric) = 0;
	virtual void										setBuildFlags							(const VkBuildAccelerationStructureFlagsKHR		buildFlags) = DE_NULL;
	virtual void										setBuildWithoutGeometries				(bool											buildWithoutGeometries) = 0;
	virtual void										setBuildWithoutPrimitives				(bool											buildWithoutPrimitives) = 0;
	virtual void										setDeferredOperation					(const bool										deferredOperation,
																								 const deUint32									workerThreadCount		= 0u ) = DE_NULL;
	virtual void										setUseArrayOfPointers					(const bool										useArrayOfPointers) = DE_NULL;
	virtual void										setIndirectBuildParameters				(const VkBuffer									indirectBuffer,
																								 const VkDeviceSize								indirectBufferOffset,
																								 const deUint32									indirectBufferStride) = DE_NULL;
	virtual VkBuildAccelerationStructureFlagsKHR		getBuildFlags							() const = DE_NULL;
	VkDeviceSize										getStructureSize						() const;

	// methods specific for each acceleration structure
	virtual void										create									(const DeviceInterface&							vk,
																								 const VkDevice									device,
																								 Allocator&										allocator,
																								 VkDeviceSize									structureSize,
																								 VkDeviceAddress								deviceAddress			= 0u) = DE_NULL;
	virtual void										build									(const DeviceInterface&							vk,
																								 const VkDevice									device,
																								 const VkCommandBuffer							cmdBuffer) = DE_NULL;
	virtual void										copyFrom								(const DeviceInterface&							vk,
																								 const VkDevice									device,
																								 const VkCommandBuffer							cmdBuffer,
																								 BottomLevelAccelerationStructure*				accelerationStructure,
																								 bool											compactCopy) = DE_NULL;

	virtual void										serialize								(const DeviceInterface&							vk,
																								 const VkDevice									device,
																								 const VkCommandBuffer							cmdBuffer,
																								 SerialStorage*									storage) = DE_NULL;
	virtual void										deserialize								(const DeviceInterface&							vk,
																								 const VkDevice									device,
																								 const VkCommandBuffer							cmdBuffer,
																								 SerialStorage*									storage) = DE_NULL;

	// helper methods for typical acceleration structure creation tasks
	void												createAndBuild							(const DeviceInterface&							vk,
																								 const VkDevice									device,
																								 const VkCommandBuffer							cmdBuffer,
																								 Allocator&										allocator,
																								 VkDeviceAddress								deviceAddress			= 0u );
	void												createAndCopyFrom						(const DeviceInterface&							vk,
																								 const VkDevice									device,
																								 const VkCommandBuffer							cmdBuffer,
																								 Allocator&										allocator,
																								 BottomLevelAccelerationStructure*				accelerationStructure,
																								 VkDeviceSize									compactCopySize			= 0u,
																								 VkDeviceAddress								deviceAddress			= 0u);
	void												createAndDeserializeFrom				(const DeviceInterface&							vk,
																								 const VkDevice									device,
																								 const VkCommandBuffer							cmdBuffer,
																								 Allocator&										allocator,
																								 SerialStorage*									storage,
																								 VkDeviceAddress								deviceAddress			= 0u);

	virtual const VkAccelerationStructureKHR*			getPtr									(void) const = DE_NULL;
protected:
	std::vector<de::SharedPtr<RaytracedGeometryBase>>	m_geometriesData;
	VkDeviceSize										m_structureSize;
	VkDeviceSize										m_updateScratchSize;
	VkDeviceSize										m_buildScratchSize;
};

de::MovePtr<BottomLevelAccelerationStructure> makeBottomLevelAccelerationStructure ();

struct InstanceData
{
								InstanceData (VkTransformMatrixKHR							matrix_,
											  deUint32										instanceCustomIndex_,
											  deUint32										mask_,
											  deUint32										instanceShaderBindingTableRecordOffset_,
											  VkGeometryInstanceFlagsKHR					flags_)
									: matrix(matrix_), instanceCustomIndex(instanceCustomIndex_), mask(mask_), instanceShaderBindingTableRecordOffset(instanceShaderBindingTableRecordOffset_), flags(flags_)
								{
								}
	VkTransformMatrixKHR		matrix;
	deUint32					instanceCustomIndex;
	deUint32					mask;
	deUint32					instanceShaderBindingTableRecordOffset;
	VkGeometryInstanceFlagsKHR	flags;
};

class TopLevelAccelerationStructure
{
public:
	static deUint32													getRequiredAllocationCount			(void);

																	TopLevelAccelerationStructure		();
																	TopLevelAccelerationStructure		(const TopLevelAccelerationStructure&				other) = delete;
	virtual															~TopLevelAccelerationStructure		();

	virtual void													setInstanceCount					(const size_t										instanceCount);
	virtual void													addInstance							(de::SharedPtr<BottomLevelAccelerationStructure>	bottomLevelStructure,
																										 const VkTransformMatrixKHR&						matrix									= identityMatrix3x4,
																										 deUint32											instanceCustomIndex						= 0,
																										 deUint32											mask									= 0xFF,
																										 deUint32											instanceShaderBindingTableRecordOffset	= 0,
																										 VkGeometryInstanceFlagsKHR							flags									= VkGeometryInstanceFlagBitsKHR(0u)	);

	virtual void													setBuildType						(const VkAccelerationStructureBuildTypeKHR			buildType) = DE_NULL;
	virtual void													setCreateFlags						(const VkAccelerationStructureCreateFlagsKHR		createFlags) = DE_NULL;
	virtual void													setCreateGeneric					(bool												createGeneric) = 0;
	virtual void													setBuildFlags						(const VkBuildAccelerationStructureFlagsKHR			buildFlags) = DE_NULL;
	virtual void													setBuildWithoutPrimitives			(bool												buildWithoutPrimitives) = 0;
	virtual void													setInactiveInstances				(bool												inactiveInstances) = 0;
	virtual void													setDeferredOperation				(const bool											deferredOperation,
																										 const deUint32										workerThreadCount = 0u) = DE_NULL;
	virtual void													setUseArrayOfPointers				(const bool											useArrayOfPointers) = DE_NULL;
	virtual void													setIndirectBuildParameters			(const VkBuffer										indirectBuffer,
																										 const VkDeviceSize									indirectBufferOffset,
																										 const deUint32										indirectBufferStride) = DE_NULL;
	virtual VkBuildAccelerationStructureFlagsKHR					getBuildFlags						() const = DE_NULL;
	VkDeviceSize													getStructureSize					() const;

	// methods specific for each acceleration structure
	virtual void													create								(const DeviceInterface&						vk,
																										 const VkDevice								device,
																										 Allocator&									allocator,
																										 VkDeviceSize								structureSize			= 0u,
																										 VkDeviceAddress							deviceAddress			= 0u ) = DE_NULL;
	virtual void													build								(const DeviceInterface&						vk,
																										 const VkDevice								device,
																										 const VkCommandBuffer						cmdBuffer) = DE_NULL;
	virtual void													copyFrom							(const DeviceInterface&						vk,
																										 const VkDevice								device,
																										 const VkCommandBuffer						cmdBuffer,
																										 TopLevelAccelerationStructure*				accelerationStructure,
																										 bool										compactCopy) = DE_NULL;

	virtual void													serialize							(const DeviceInterface&						vk,
																										 const VkDevice								device,
																										 const VkCommandBuffer						cmdBuffer,
																										 SerialStorage*								storage) = DE_NULL;
	virtual void													deserialize							(const DeviceInterface&						vk,
																										 const VkDevice								device,
																										 const VkCommandBuffer						cmdBuffer,
																										 SerialStorage*								storage) = DE_NULL;

	// helper methods for typical acceleration structure creation tasks
	void															createAndBuild						(const DeviceInterface&						vk,
																										 const VkDevice								device,
																										 const VkCommandBuffer						cmdBuffer,
																										 Allocator&									allocator,
																										 VkDeviceAddress							deviceAddress			= 0u );
	void															createAndCopyFrom					(const DeviceInterface&						vk,
																										 const VkDevice								device,
																										 const VkCommandBuffer						cmdBuffer,
																										 Allocator&									allocator,
																										 TopLevelAccelerationStructure*				accelerationStructure,
																										 VkDeviceSize								compactCopySize			= 0u,
																										 VkDeviceAddress							deviceAddress			= 0u);
	void															createAndDeserializeFrom			(const DeviceInterface&						vk,
																										 const VkDevice								device,
																										 const VkCommandBuffer						cmdBuffer,
																										 Allocator&									allocator,
																										 SerialStorage*								storage,
																										 VkDeviceAddress							deviceAddress			= 0u);

	virtual const VkAccelerationStructureKHR*						getPtr								(void) const = DE_NULL;

protected:
	std::vector<de::SharedPtr<BottomLevelAccelerationStructure> >	m_bottomLevelInstances;
	std::vector<InstanceData>										m_instanceData;
	VkDeviceSize													m_structureSize;
	VkDeviceSize													m_updateScratchSize;
	VkDeviceSize													m_buildScratchSize;
};

de::MovePtr<TopLevelAccelerationStructure> makeTopLevelAccelerationStructure ();

bool queryAccelerationStructureSize (const DeviceInterface&							vk,
									 const VkDevice									device,
									 const VkCommandBuffer							cmdBuffer,
									 const std::vector<VkAccelerationStructureKHR>&	accelerationStructureHandles,
									 VkAccelerationStructureBuildTypeKHR			buildType,
									 const VkQueryPool								queryPool,
									 VkQueryType									queryType,
									 deUint32										firstQuery,
									 std::vector<VkDeviceSize>&						results);

class RayTracingPipeline
{
public:
																RayTracingPipeline			();
																~RayTracingPipeline			();

	void														addShader					(VkShaderStageFlagBits									shaderStage,
																							 Move<VkShaderModule>									shaderModule,
																							 deUint32												group,
																							 const VkSpecializationInfo*							specializationInfo = nullptr);
	void														addShader					(VkShaderStageFlagBits									shaderStage,
																							 de::SharedPtr<Move<VkShaderModule>>					shaderModule,
																							 deUint32												group,
																							 const VkSpecializationInfo*							specializationInfoPtr = nullptr);
	void														addLibrary					(de::SharedPtr<de::MovePtr<RayTracingPipeline>>			pipelineLibrary);
	Move<VkPipeline>											createPipeline				(const DeviceInterface&									vk,
																							 const VkDevice											device,
																							 const VkPipelineLayout									pipelineLayout,
																							 const std::vector<de::SharedPtr<Move<VkPipeline>>>&	pipelineLibraries			= std::vector<de::SharedPtr<Move<VkPipeline>>>());
	std::vector<de::SharedPtr<Move<VkPipeline>>>				createPipelineWithLibraries	(const DeviceInterface&									vk,
																							 const VkDevice											device,
																							 const VkPipelineLayout									pipelineLayout);
	de::MovePtr<BufferWithMemory>								createShaderBindingTable	(const DeviceInterface&									vk,
																							 const VkDevice											device,
																							 const VkPipeline										pipeline,
																							 Allocator&												allocator,
																							 const deUint32&										shaderGroupHandleSize,
																							 const deUint32											shaderGroupBaseAlignment,
																							 const deUint32&										firstGroup,
																							 const deUint32&										groupCount,
																							 const VkBufferCreateFlags&								additionalBufferCreateFlags	= VkBufferCreateFlags(0u),
																							 const VkBufferUsageFlags&								additionalBufferUsageFlags	= VkBufferUsageFlags(0u),
																							 const MemoryRequirement&								additionalMemoryRequirement	= MemoryRequirement::Any,
																							 const VkDeviceAddress&									opaqueCaptureAddress		= 0u,
																							 const deUint32											shaderBindingTableOffset	= 0u,
																							 const deUint32											shaderRecordSize			= 0u,
																							 const void**											shaderGroupDataPtrPerGroup	= nullptr);
	void														setCreateFlags				(const VkPipelineCreateFlags&							pipelineCreateFlags);
	void														setMaxRecursionDepth		(const deUint32&										maxRecursionDepth);
	void														setMaxPayloadSize			(const deUint32&										maxPayloadSize);
	void														setMaxAttributeSize			(const deUint32&										maxAttributeSize);
	void														setDeferredOperation		(const bool												deferredOperation,
																							 const deUint32											workerThreadCount = 0);
	void														addDynamicState				(const VkDynamicState&									dynamicState);


protected:
	Move<VkPipeline>											createPipelineKHR			(const DeviceInterface&									vk,
																							 const VkDevice											device,
																							 const VkPipelineLayout									pipelineLayout,
																							 const std::vector<de::SharedPtr<Move<VkPipeline>>>&	pipelineLibraries);

	std::vector<de::SharedPtr<Move<VkShaderModule> > >			m_shadersModules;
	std::vector<de::SharedPtr<de::MovePtr<RayTracingPipeline>>>	m_pipelineLibraries;
	std::vector<VkPipelineShaderStageCreateInfo>				m_shaderCreateInfos;
	std::vector<VkRayTracingShaderGroupCreateInfoKHR>			m_shadersGroupCreateInfos;
	VkPipelineCreateFlags										m_pipelineCreateFlags;
	deUint32													m_maxRecursionDepth;
	deUint32													m_maxPayloadSize;
	deUint32													m_maxAttributeSize;
	bool														m_deferredOperation;
	deUint32													m_workerThreadCount;
	std::vector<VkDynamicState>									m_dynamicStates;
};

class RayTracingProperties
{
protected:
									RayTracingProperties						() {};

public:
									RayTracingProperties						(const InstanceInterface&	vki,
																				 const VkPhysicalDevice		physicalDevice) { DE_UNREF(vki); DE_UNREF(physicalDevice); };
	virtual							~RayTracingProperties						() {};

	virtual deUint32				getShaderGroupHandleSize					(void)	= DE_NULL;
	virtual deUint32				getMaxRecursionDepth						(void)	= DE_NULL;
	virtual deUint32				getMaxShaderGroupStride						(void)	= DE_NULL;
	virtual deUint32				getShaderGroupBaseAlignment					(void)	= DE_NULL;
	virtual deUint64				getMaxGeometryCount							(void)	= DE_NULL;
	virtual deUint64				getMaxInstanceCount							(void)	= DE_NULL;
	virtual deUint64				getMaxPrimitiveCount						(void)	= DE_NULL;
	virtual deUint32				getMaxDescriptorSetAccelerationStructures	(void)	= DE_NULL;
	virtual deUint32				getMaxRayDispatchInvocationCount			(void)	= DE_NULL;
	virtual deUint32				getMaxRayHitAttributeSize					(void)	= DE_NULL;
};

de::MovePtr<RayTracingProperties> makeRayTracingProperties (const InstanceInterface&	vki,
															const VkPhysicalDevice		physicalDevice);

void cmdTraceRays	(const DeviceInterface&					vk,
					 VkCommandBuffer						commandBuffer,
					 const VkStridedDeviceAddressRegionKHR*	raygenShaderBindingTableRegion,
					 const VkStridedDeviceAddressRegionKHR*	missShaderBindingTableRegion,
					 const VkStridedDeviceAddressRegionKHR*	hitShaderBindingTableRegion,
					 const VkStridedDeviceAddressRegionKHR*	callableShaderBindingTableRegion,
					 deUint32								width,
					 deUint32								height,
					 deUint32								depth);

void cmdTraceRaysIndirect	(const DeviceInterface&					vk,
							 VkCommandBuffer						commandBuffer,
							 const VkStridedDeviceAddressRegionKHR*	raygenShaderBindingTableRegion,
							 const VkStridedDeviceAddressRegionKHR*	missShaderBindingTableRegion,
							 const VkStridedDeviceAddressRegionKHR*	hitShaderBindingTableRegion,
							 const VkStridedDeviceAddressRegionKHR*	callableShaderBindingTableRegion,
							 VkDeviceAddress						indirectDeviceAddress);
} // vk

#endif // _VKRAYTRACINGUTIL_HPP