34_android.cpp

#include <algorithm>
#include <array>
#include <assert.h>
#include <chrono>
#include <cstdlib>
#include <cstring>
#include <fstream>
#include <iostream>
#include <limits>
#include <memory>
#include <optional>
#include <stdexcept>
#include <vector>

#if defined(__INTELLISENSE__) || !defined(USE_CPP20_MODULES)
#	include <vulkan/vulkan_raii.hpp>
#else
import vulkan_hpp;
#endif
//clang-format off
#if defined(__ANDROID__)
#	include <vulkan/vulkan_core.h>
#	include <vulkan/vulkan_android.h>
#endif
//clang-format on
#include <vulkan/vulkan_profiles.hpp>

// Platform detection
#if defined(__ANDROID__)
#	define PLATFORM_ANDROID 1
#else
#	define PLATFORM_DESKTOP 1
#endif

#define STB_IMAGE_IMPLEMENTATION
#include <stb_image.h>

#define TINYOBJLOADER_IMPLEMENTATION
#include <tiny_obj_loader.h>

// Platform-specific includes
#if PLATFORM_ANDROID
// Android-specific includes
#	include <android/log.h>
#	include <game-activity/native_app_glue/android_native_app_glue.h>
#	include <android/asset_manager.h>
#	include <android/asset_manager_jni.h>

// Declare and implement app_dummy function from native_app_glue
extern "C" void app_dummy()
{
	// This is a dummy function that does nothing
	// It's used to prevent the linker from stripping out the native_app_glue code
}

// Define AAssetManager type for Android
typedef AAssetManager AssetManagerType;

// Define logging macros for Android
#	define LOGI(...) ((void) __android_log_print(ANDROID_LOG_INFO, "VulkanTutorial", __VA_ARGS__))
#	define LOGW(...) ((void) __android_log_print(ANDROID_LOG_WARN, "VulkanTutorial", __VA_ARGS__))
#	define LOGE(...) ((void) __android_log_print(ANDROID_LOG_ERROR, "VulkanTutorial", __VA_ARGS__))
#	define LOG_INFO(msg) LOGI("%s", msg)
#	define LOG_ERROR(msg) LOGE("%s", msg)
#else
// Define AAssetManager type for non-Android platforms
typedef void AssetManagerType;
// Desktop-specific includes
#	define GLFW_INCLUDE_VULKAN
#	include <GLFW/glfw3.h>

// Define logging macros for Desktop
#	define LOGI(...)        \
		printf(__VA_ARGS__); \
		printf("\n")
#	define LOGW(...)        \
		printf(__VA_ARGS__); \
		printf("\n")
#	define LOGE(...)                 \
		fprintf(stderr, __VA_ARGS__); \
		fprintf(stderr, "\n")
#	define LOG_INFO(msg) std::cout << msg << std::endl
#	define LOG_ERROR(msg) std::cerr << msg << std::endl
#endif

#define GLM_FORCE_RADIANS
#define GLM_FORCE_DEPTH_ZERO_TO_ONE
#define GLM_ENABLE_EXPERIMENTAL
#define GLM_FORCE_CXX11
#include <glm/glm.hpp>
#include <glm/gtc/matrix_transform.hpp>
#include <glm/gtx/hash.hpp>

constexpr uint32_t WIDTH                = 800;
constexpr uint32_t HEIGHT               = 600;
const std::string  MODEL_PATH           = "models/viking_room.obj";
const std::string  TEXTURE_PATH         = "textures/viking_room.png";
constexpr int      MAX_FRAMES_IN_FLIGHT = 2;

#if PLATFORM_ANDROID
// Define VpProfileProperties structure if not already defined
#	ifndef VP_PROFILE_PROPERTIES_DEFINED
#		define VP_PROFILE_PROPERTIES_DEFINED
struct VpProfileProperties
{
	char     name[256];
	uint32_t specVersion;
};
#	endif

// Define Vulkan Profile constants
#	ifndef VP_KHR_ROADMAP_2022_NAME
#		define VP_KHR_ROADMAP_2022_NAME "VP_KHR_roadmap_2022"
#	endif

#	ifndef VP_KHR_ROADMAP_2022_SPEC_VERSION
#		define VP_KHR_ROADMAP_2022_SPEC_VERSION 1
#	endif
#endif

// Application info structure to store profile support flags
struct AppInfo
{
	bool                profileSupported = false;
	VpProfileProperties profile;
};

struct Vertex
{
	glm::vec3 pos;
	glm::vec3 color;
	glm::vec2 texCoord;

	static vk::VertexInputBindingDescription getBindingDescription()
	{
		return {0, sizeof(Vertex), vk::VertexInputRate::eVertex};
	}

	static std::array<vk::VertexInputAttributeDescription, 3> getAttributeDescriptions()
	{
		return {
		    vk::VertexInputAttributeDescription(0, 0, vk::Format::eR32G32B32Sfloat, offsetof(Vertex, pos)),
		    vk::VertexInputAttributeDescription(1, 0, vk::Format::eR32G32B32Sfloat, offsetof(Vertex, color)),
		    vk::VertexInputAttributeDescription(2, 0, vk::Format::eR32G32Sfloat, offsetof(Vertex, texCoord))};
	}

	bool operator==(const Vertex &other) const
	{
		return pos == other.pos && color == other.color && texCoord == other.texCoord;
	}
};

template <>
struct std::hash<Vertex>
{
	size_t operator()(Vertex const &vertex) const noexcept
	{
		return ((hash<glm::vec3>()(vertex.pos) ^ (hash<glm::vec3>()(vertex.color) << 1)) >> 1) ^ (hash<glm::vec2>()(vertex.texCoord) << 1);
	}
};

struct UniformBufferObject
{
	alignas(16) glm::mat4 model;
	alignas(16) glm::mat4 view;
	alignas(16) glm::mat4 proj;
};

// Cross-platform file reading function
std::vector<char> readFile(const std::string &filename, std::optional<AssetManagerType *> assetManager = std::nullopt)
{
#if PLATFORM_ANDROID
	// On Android, use asset manager if provided
	if (assetManager.has_value() && *assetManager != nullptr)
	{
		// Open the asset
		AAsset *asset = AAssetManager_open(*assetManager, filename.c_str(), AASSET_MODE_BUFFER);
		if (!asset)
		{
			LOGE("Failed to open asset: %s", filename.c_str());
			throw std::runtime_error("Failed to open file: " + filename);
		}

		// Get the file size
		off_t             fileSize = AAsset_getLength(asset);
		std::vector<char> buffer(fileSize);

		// Read the file data
		AAsset_read(asset, buffer.data(), fileSize);

		// Close the asset
		AAsset_close(asset);

		return buffer;
	}
#endif

	// Desktop version or Android fallback to filesystem
	std::ifstream file(filename, std::ios::ate | std::ios::binary);

	if (!file.is_open())
	{
		throw std::runtime_error("Failed to open file: " + filename);
	}

	size_t            fileSize = static_cast<size_t>(file.tellg());
	std::vector<char> buffer(fileSize);

	file.seekg(0);
	file.read(buffer.data(), fileSize);
	file.close();

	return buffer;
}

// Cross-platform application class
class HelloTriangleApplication
{
  public:
#if PLATFORM_DESKTOP
	// Desktop constructor
	HelloTriangleApplication()
	{
		// No Android-specific initialization needed
	}
#else
	// Android constructor
	HelloTriangleApplication(android_app *app) :
	    androidApp(app)
	{
		androidApp->userData = this;
		androidApp->onAppCmd = handleAppCommand;
		// Note: onInputEvent is no longer a member of android_app in the current NDK version
		// Input events are now handled differently

		// Get the asset manager
		assetManager = androidApp->activity->assetManager;
	}
#endif

	void run()
	{
#if PLATFORM_DESKTOP
		// Desktop main loop
		initWindow();
		initVulkan();
		mainLoop();
		cleanup();
#else
		// Android main loop is handled by Android
		while (!initialized)
		{
			// Wait for app to initialize
			int                  events;
			android_poll_source *source;
			if (ALooper_pollOnce(0, nullptr, &events, (void **) &source) >= 0)
			{
				if (source != nullptr)
				{
					source->process(androidApp, source);
				}
			}
		}
#endif
	}

#if PLATFORM_DESKTOP
	// Initialize window (Desktop only)
	void initWindow()
	{
		glfwInit();
		glfwWindowHint(GLFW_CLIENT_API, GLFW_NO_API);
		glfwWindowHint(GLFW_RESIZABLE, GLFW_TRUE);

		window = glfwCreateWindow(WIDTH, HEIGHT, "Vulkan Cross-Platform", nullptr, nullptr);
		glfwSetWindowUserPointer(window, this);
		glfwSetFramebufferSizeCallback(window, framebufferResizeCallback);

		LOG_INFO("Desktop window created");
	}

	// Desktop main loop
	void mainLoop()
	{
		while (!glfwWindowShouldClose(window))
		{
			glfwPollEvents();
			drawFrame();
		}

		device.waitIdle();
	}

	// Desktop framebuffer resize callback
	static void framebufferResizeCallback(GLFWwindow *window, int, int)
	{
		auto app                = reinterpret_cast<HelloTriangleApplication *>(glfwGetWindowUserPointer(window));
		app->framebufferResized = true;
	}
#endif

	void cleanup()
	{
		if (initialized)
		{
			// Wait for device to finish operations
			if (*device)
			{
				device.waitIdle();
			}

			// Cleanup resources
			cleanupSwapChain();

			initialized = false;
		}
	}

  private:
#if PLATFORM_ANDROID
	// Android-specific members
	android_app      *androidApp   = nullptr;
	AssetManagerType *assetManager = nullptr;
#else
	// Desktop-specific members
	GLFWwindow *window = nullptr;
#endif
	bool initialized        = false;
	bool framebufferResized = false;

	// Vulkan objects
	vk::raii::Context                    context;
	vk::raii::Instance                   instance       = nullptr;
	vk::raii::DebugUtilsMessengerEXT     debugMessenger = nullptr;
	vk::raii::SurfaceKHR                 surface        = nullptr;
	vk::raii::PhysicalDevice             physicalDevice = nullptr;
	vk::raii::Device                     device         = nullptr;
	uint32_t                             queueIndex     = ~0;
	vk::raii::Queue                      queue          = nullptr;
	vk::raii::SwapchainKHR               swapChain      = nullptr;
	std::vector<vk::Image>               swapChainImages;
	vk::SurfaceFormatKHR                 swapChainSurfaceFormat;
	vk::Extent2D                         swapChainExtent;
	std::vector<vk::raii::ImageView>     swapChainImageViews;
	vk::raii::RenderPass                 renderPass          = nullptr;
	vk::raii::DescriptorSetLayout        descriptorSetLayout = nullptr;
	vk::raii::PipelineLayout             pipelineLayout      = nullptr;
	vk::raii::Pipeline                   graphicsPipeline    = nullptr;
	vk::Format                           depthFormat;
	vk::raii::Image                      depthImage       = nullptr;
	vk::raii::DeviceMemory               depthImageMemory = nullptr;
	vk::raii::ImageView                  depthImageView   = nullptr;
	std::vector<vk::raii::Framebuffer>   swapChainFramebuffers;
	vk::raii::CommandPool                commandPool = nullptr;
	std::vector<vk::raii::CommandBuffer> commandBuffers;
	vk::raii::Buffer                     vertexBuffer       = nullptr;
	vk::raii::DeviceMemory               vertexBufferMemory = nullptr;
	vk::raii::Buffer                     indexBuffer        = nullptr;
	vk::raii::DeviceMemory               indexBufferMemory  = nullptr;
	vk::raii::Image                      textureImage       = nullptr;
	vk::raii::DeviceMemory               textureImageMemory = nullptr;
	vk::raii::ImageView                  textureImageView   = nullptr;
	vk::raii::Sampler                    textureSampler     = nullptr;
	std::vector<vk::raii::Buffer>        uniformBuffers;
	std::vector<vk::raii::DeviceMemory>  uniformBuffersMemory;
	vk::raii::DescriptorPool             descriptorPool = nullptr;
	std::vector<vk::raii::DescriptorSet> descriptorSets;
	std::vector<vk::raii::Semaphore>     imageAvailableSemaphores;
	std::vector<vk::raii::Semaphore>     renderFinishedSemaphores;
	std::vector<vk::raii::Fence>         inFlightFences;
	uint32_t                             frameIndex = 0;

	// Application info
	AppInfo appInfo;

	// Model data
	std::vector<Vertex>   vertices;
	std::vector<uint32_t> indices;

	// Swap chain support details
	struct SwapChainSupportDetails
	{
		vk::SurfaceCapabilitiesKHR        capabilities;
		std::vector<vk::SurfaceFormatKHR> formats;
		std::vector<vk::PresentModeKHR>   presentModes;
	};

	// Required device extensions
	const std::vector<const char *> deviceExtensions = {
	    VK_KHR_SWAPCHAIN_EXTENSION_NAME};

	// Initialize Vulkan
	void initVulkan()
	{
		createInstance();
		createSurface();
		pickPhysicalDevice();
		checkFeatureSupport();
		createLogicalDevice();
		createSwapChain();
		createImageViews();
		createDepthResources();
		createRenderPass();
		createDescriptorSetLayout();
		createGraphicsPipeline();
		createFramebuffers();
		createCommandPool();
		createTextureImage();
		createTextureImageView();
		createTextureSampler();
		loadModel();
		createVertexBuffer();
		createIndexBuffer();
		createUniformBuffers();
		createDescriptorPool();
		createDescriptorSets();
		createCommandBuffers();
		createSyncObjects();

		initialized = true;
	}

	// Create Vulkan instance
	void createInstance()
	{
		// Application info
		vk::ApplicationInfo appInfo{
		    .pApplicationName   = "Vulkan Android",
		    .applicationVersion = VK_MAKE_VERSION(1, 0, 0),
		    .pEngineName        = "No Engine",
		    .engineVersion      = VK_MAKE_VERSION(1, 0, 0),
		    .apiVersion         = VK_API_VERSION_1_3};

		// Get required extensions
		std::vector<const char *> extensions = getRequiredExtensions();

		// Create instance
		vk::InstanceCreateInfo createInfo{
		    .pApplicationInfo        = &appInfo,
		    .enabledExtensionCount   = static_cast<uint32_t>(extensions.size()),
		    .ppEnabledExtensionNames = extensions.data()};

		instance = vk::raii::Instance(context, createInfo);
		LOGI("Vulkan instance created");
	}

	// Create platform-specific surface
	void createSurface()
	{
		VkSurfaceKHR _surface;

#if PLATFORM_ANDROID
		// Create Android surface
		VkAndroidSurfaceCreateInfoKHR createInfo = {
		    .sType  = VK_STRUCTURE_TYPE_ANDROID_SURFACE_CREATE_INFO_KHR,
		    .pNext  = nullptr,
		    .flags  = 0,
		    .window = androidApp->window};

		VkResult result = vkCreateAndroidSurfaceKHR(
		    *instance,
		    &createInfo,
		    nullptr,
		    &_surface);

		if (result != VK_SUCCESS)
		{
			throw std::runtime_error("Failed to create Android surface");
		}

		LOG_INFO("Android surface created");
#else
		// Create desktop surface using GLFW
		if (glfwCreateWindowSurface(*instance, window, nullptr, &_surface) != 0)
		{
			throw std::runtime_error("Failed to create window surface");
		}

		LOG_INFO("Desktop surface created");
#endif

		surface = vk::raii::SurfaceKHR(instance, _surface);
	}

	// Pick physical device
	void pickPhysicalDevice()
	{
		std::vector<vk::raii::PhysicalDevice> devices = instance.enumeratePhysicalDevices();
		const auto                            devIter = std::ranges::find_if(
            devices,
            [&](auto const &device) {
                // Check if any of the queue families support graphics operations
                auto queueFamilies = device.getQueueFamilyProperties();
                bool supportsGraphics =
                    std::ranges::any_of(queueFamilies, [](auto const &qfp) { return !!(qfp.queueFlags & vk::QueueFlagBits::eGraphics); });

                // Check if all required device extensions are available
                auto availableDeviceExtensions = device.enumerateDeviceExtensionProperties();
                bool supportsAllRequiredExtensions =
                    std::ranges::all_of(deviceExtensions,
			                                                       [&availableDeviceExtensions](auto const &requiredDeviceExtension) {
                                            return std::ranges::any_of(availableDeviceExtensions,
				                                                                                  [requiredDeviceExtension](auto const &availableDeviceExtension) {
                                                                           return strcmp(availableDeviceExtension.extensionName, requiredDeviceExtension) == 0;
                                                                       });
                                        });

                return supportsGraphics && supportsAllRequiredExtensions;
            });

		if (devIter != devices.end())
		{
			physicalDevice = *devIter;

			// Print device information
			vk::PhysicalDeviceProperties deviceProperties = physicalDevice.getProperties();
			LOGI("Selected GPU: %s", deviceProperties.deviceName.data());
		}
		else
		{
			throw std::runtime_error("Failed to find a suitable GPU");
		}
	}

	// Check feature support
	void checkFeatureSupport()
	{
		// Define the KHR roadmap 2022 profile
		appInfo.profile = {
		    VP_KHR_ROADMAP_2022_NAME,
		    VP_KHR_ROADMAP_2022_SPEC_VERSION};

		// Check if the profile is supported
		VkBool32 supported = VK_FALSE;

#ifdef PLATFORM_ANDROID
		// Create a vp::ProfileDesc from our VpProfileProperties
		vp::ProfileDesc profileDesc = {
		    appInfo.profile.name,
		    appInfo.profile.specVersion};

		// Use vp::GetProfileSupport instead of vpGetPhysicalDeviceProfileSupport
		bool result = vp::GetProfileSupport(
		    *physicalDevice,        // Pass the physical device directly
		    &profileDesc,           // Pass the profile description
		    &supported              // Output parameter for support status
		);
#else
		VkResult vk_result = vpGetPhysicalDeviceProfileSupport(
		    *instance,
		    *physicalDevice,
		    &appInfo.profile,
		    &supported);
		bool result = vk_result == VK_SUCCESS;
#endif

		if (result && supported == VK_TRUE)
		{
			appInfo.profileSupported = true;
			LOGI("Using KHR roadmap 2022 profile");
		}
		else
		{
			appInfo.profileSupported = false;
			LOGI("Falling back to traditional rendering (profile not supported)");
		}
	}

	// Create logical device
	void createLogicalDevice()
	{
		std::vector<vk::QueueFamilyProperties> queueFamilyProperties = physicalDevice.getQueueFamilyProperties();

		// get the first index into queueFamilyProperties which supports both graphics and present
		for (uint32_t qfpIndex = 0; qfpIndex < queueFamilyProperties.size(); qfpIndex++)
		{
			if ((queueFamilyProperties[qfpIndex].queueFlags & vk::QueueFlagBits::eGraphics) &&
			    physicalDevice.getSurfaceSupportKHR(qfpIndex, *surface))
			{
				// found a queue family that supports both graphics and present
				queueIndex = qfpIndex;
				break;
			}
		}
		if (queueIndex == ~0)
		{
			throw std::runtime_error("Could not find a queue for graphics and present -> terminating");
		}

		float                     queuePriority = 0.5f;
		vk::DeviceQueueCreateInfo deviceQueueCreateInfo{.queueFamilyIndex = queueIndex, .queueCount = 1, .pQueuePriorities = &queuePriority};

		if (appInfo.profileSupported)
		{
			// Enable required features
			vk::PhysicalDeviceFeatures2 features2;
			vk::PhysicalDeviceFeatures  deviceFeatures{};
			deviceFeatures.samplerAnisotropy = VK_TRUE;
			deviceFeatures.sampleRateShading = VK_TRUE;
			features2.features               = deviceFeatures;

			// Enable dynamic rendering
			vk::PhysicalDeviceDynamicRenderingFeatures dynamicRenderingFeatures;
			dynamicRenderingFeatures.dynamicRendering = VK_TRUE;
			features2.pNext                           = &dynamicRenderingFeatures;

			// Create a vk::DeviceCreateInfo with the required features
			vk::DeviceCreateInfo vkDeviceCreateInfo{
			    .pNext                   = &features2,
			    .queueCreateInfoCount    = 1,
			    .pQueueCreateInfos       = &deviceQueueCreateInfo,
			    .enabledExtensionCount   = static_cast<uint32_t>(deviceExtensions.size()),
			    .ppEnabledExtensionNames = deviceExtensions.data()};

			// Create the device with the vk::DeviceCreateInfo
			device = vk::raii::Device(physicalDevice, vkDeviceCreateInfo);
		}
		else
		{
			// Fallback to manual device creation
			vk::PhysicalDeviceFeatures deviceFeatures{};
			deviceFeatures.samplerAnisotropy = VK_TRUE;
			deviceFeatures.sampleRateShading = VK_TRUE;

			vk::DeviceCreateInfo createInfo{
			    .queueCreateInfoCount    = 1,
			    .pQueueCreateInfos       = &deviceQueueCreateInfo,
			    .enabledExtensionCount   = static_cast<uint32_t>(deviceExtensions.size()),
			    .ppEnabledExtensionNames = deviceExtensions.data(),
			    .pEnabledFeatures        = &deviceFeatures};

			device = vk::raii::Device(physicalDevice, createInfo);
		}

		queue = device.getQueue(queueIndex, 0);
	}

	// Create swap chain
	void createSwapChain()
	{
		SwapChainSupportDetails swapChainSupport = querySwapChainSupport(physicalDevice);
		swapChainExtent                          = chooseSwapExtent(swapChainSupport.capabilities);
		swapChainSurfaceFormat                   = chooseSwapSurfaceFormat(swapChainSupport.formats);
		vk::SwapchainCreateInfoKHR swapChainCreateInfo{.surface          = *surface,
		                                               .minImageCount    = chooseSwapMinImageCount(swapChainSupport.capabilities),
		                                               .imageFormat      = swapChainSurfaceFormat.format,
		                                               .imageColorSpace  = swapChainSurfaceFormat.colorSpace,
		                                               .imageExtent      = swapChainExtent,
		                                               .imageArrayLayers = 1,
		                                               .imageUsage       = vk::ImageUsageFlagBits::eColorAttachment,
		                                               .imageSharingMode = vk::SharingMode::eExclusive,
		                                               .preTransform     = swapChainSupport.capabilities.currentTransform,
		                                               .compositeAlpha   = vk::CompositeAlphaFlagBitsKHR::eOpaque,
		                                               .presentMode      = chooseSwapPresentMode(swapChainSupport.presentModes),
		                                               .clipped          = true};

		swapChain       = device.createSwapchainKHR(swapChainCreateInfo);
		swapChainImages = swapChain.getImages();
	}

	// Create image views
	void createImageViews()
	{
		assert(swapChainImageViews.empty());
		swapChainImageViews.reserve(swapChainImages.size());

		for (const auto &image : swapChainImages)
		{
			vk::ImageViewCreateInfo createInfo{
			    .image      = image,
			    .viewType   = vk::ImageViewType::e2D,
			    .format     = swapChainSurfaceFormat.format,
			    .components = {
			        .r = vk::ComponentSwizzle::eIdentity,
			        .g = vk::ComponentSwizzle::eIdentity,
			        .b = vk::ComponentSwizzle::eIdentity,
			        .a = vk::ComponentSwizzle::eIdentity},
			    .subresourceRange = {.aspectMask = vk::ImageAspectFlagBits::eColor, .baseMipLevel = 0, .levelCount = 1, .baseArrayLayer = 0, .layerCount = 1}};

			swapChainImageViews.push_back(device.createImageView(createInfo));
		}
	}

	// Create render pass
	void createRenderPass()
	{
		vk::AttachmentDescription colorAttachment{
		    .format         = swapChainSurfaceFormat.format,
		    .samples        = vk::SampleCountFlagBits::e1,
		    .loadOp         = vk::AttachmentLoadOp::eClear,
		    .storeOp        = vk::AttachmentStoreOp::eStore,
		    .stencilLoadOp  = vk::AttachmentLoadOp::eDontCare,
		    .stencilStoreOp = vk::AttachmentStoreOp::eDontCare,
		    .initialLayout  = vk::ImageLayout::eUndefined,
		    .finalLayout    = vk::ImageLayout::ePresentSrcKHR};

		vk::AttachmentReference colorAttachmentRef{
		    .attachment = 0,
		    .layout     = vk::ImageLayout::eColorAttachmentOptimal};

		vk::AttachmentDescription depthAttachment{
		    .format         = depthFormat,
		    .samples        = vk::SampleCountFlagBits::e1,
		    .loadOp         = vk::AttachmentLoadOp::eClear,
		    .storeOp        = vk::AttachmentStoreOp::eStore,
		    .stencilLoadOp  = vk::AttachmentLoadOp::eDontCare,
		    .stencilStoreOp = vk::AttachmentStoreOp::eDontCare,
		    .initialLayout  = vk::ImageLayout::eUndefined,
		    .finalLayout    = vk::ImageLayout::eDepthStencilAttachmentOptimal};

		vk::AttachmentReference depthAttachmentRef{
		    .attachment = 1,
		    .layout     = vk::ImageLayout::eDepthStencilAttachmentOptimal};

		vk::SubpassDescription subpass{
		    .pipelineBindPoint       = vk::PipelineBindPoint::eGraphics,
		    .colorAttachmentCount    = 1,
		    .pColorAttachments       = &colorAttachmentRef,
		    .pDepthStencilAttachment = &depthAttachmentRef};

		// @todo: barrier for deoth
		vk::SubpassDependency dependency{
		    .srcSubpass    = VK_SUBPASS_EXTERNAL,
		    .dstSubpass    = 0,
		    .srcStageMask  = vk::PipelineStageFlagBits::eColorAttachmentOutput | vk::PipelineStageFlagBits::eEarlyFragmentTests | vk::PipelineStageFlagBits::eLateFragmentTests,
		    .dstStageMask  = vk::PipelineStageFlagBits::eColorAttachmentOutput | vk::PipelineStageFlagBits::eEarlyFragmentTests | vk::PipelineStageFlagBits::eLateFragmentTests,
		    .srcAccessMask = vk::AccessFlagBits::eDepthStencilAttachmentWrite,
		    .dstAccessMask = vk::AccessFlagBits::eColorAttachmentWrite | vk::AccessFlagBits::eDepthStencilAttachmentWrite};

		vk::AttachmentDescription attachments[] = {
		    colorAttachment,
		    depthAttachment};

		vk::RenderPassCreateInfo renderPassInfo{
		    .attachmentCount = 2,
		    .pAttachments    = attachments,
		    .subpassCount    = 1,
		    .pSubpasses      = &subpass,
		    .dependencyCount = 1,
		    .pDependencies   = &dependency};

		renderPass = device.createRenderPass(renderPassInfo);
	}

	// Create descriptor set layout
	void createDescriptorSetLayout()
	{
		vk::DescriptorSetLayoutBinding uboLayoutBinding{
		    .binding         = 0,
		    .descriptorType  = vk::DescriptorType::eUniformBuffer,
		    .descriptorCount = 1,
		    .stageFlags      = vk::ShaderStageFlagBits::eVertex};

		vk::DescriptorSetLayoutBinding samplerLayoutBinding{
		    .binding         = 1,
		    .descriptorType  = vk::DescriptorType::eCombinedImageSampler,
		    .descriptorCount = 1,
		    .stageFlags      = vk::ShaderStageFlagBits::eFragment};

		std::array<vk::DescriptorSetLayoutBinding, 2> bindings = {uboLayoutBinding, samplerLayoutBinding};

		vk::DescriptorSetLayoutCreateInfo layoutInfo{
		    .bindingCount = static_cast<uint32_t>(bindings.size()),
		    .pBindings    = bindings.data()};

		descriptorSetLayout = device.createDescriptorSetLayout(layoutInfo);
	}

	// Create graphics pipeline
	void createGraphicsPipeline()
	{
		// Load shader code from asset files
		LOGI("Loading shaders from assets");

		// Load shader files using cross-platform function
#if PLATFORM_ANDROID
		std::optional<AssetManagerType *> optionalAssetManager = assetManager;
#else
		std::optional<void *> optionalAssetManager = std::nullopt;
#endif
		std::vector<char> vertShaderCode = readFile("shaders/vert.spv", optionalAssetManager);
		std::vector<char> fragShaderCode = readFile("shaders/frag.spv", optionalAssetManager);

		LOGI("Shaders loaded successfully");

		// Create shader modules
		vk::ShaderModuleCreateInfo vertShaderModuleInfo{
		    .codeSize = vertShaderCode.size(),
		    .pCode    = reinterpret_cast<const uint32_t *>(vertShaderCode.data())};
		vk::raii::ShaderModule vertShaderModule = device.createShaderModule(vertShaderModuleInfo);

		vk::ShaderModuleCreateInfo fragShaderModuleInfo{
		    .codeSize = fragShaderCode.size(),
		    .pCode    = reinterpret_cast<const uint32_t *>(fragShaderCode.data())};
		vk::raii::ShaderModule fragShaderModule = device.createShaderModule(fragShaderModuleInfo);

		// Create shader stages
		vk::PipelineShaderStageCreateInfo shaderStages[] = {
		    {.stage  = vk::ShaderStageFlagBits::eVertex,
		     .module = *vertShaderModule,
		     .pName  = "main"},
		    {.stage  = vk::ShaderStageFlagBits::eFragment,
		     .module = *fragShaderModule,
		     .pName  = "main"}};

		// Vertex input
		auto bindingDescription    = Vertex::getBindingDescription();
		auto attributeDescriptions = Vertex::getAttributeDescriptions();

		vk::PipelineVertexInputStateCreateInfo vertexInputInfo{
		    .vertexBindingDescriptionCount   = 1,
		    .pVertexBindingDescriptions      = &bindingDescription,
		    .vertexAttributeDescriptionCount = static_cast<uint32_t>(attributeDescriptions.size()),
		    .pVertexAttributeDescriptions    = attributeDescriptions.data()};

		// Input assembly
		vk::PipelineInputAssemblyStateCreateInfo inputAssembly{
		    .topology               = vk::PrimitiveTopology::eTriangleList,
		    .primitiveRestartEnable = VK_FALSE};

		// Viewport and scissor
		vk::PipelineViewportStateCreateInfo viewportState{
		    .viewportCount = 1,
		    .scissorCount  = 1};

		// Rasterization
		vk::PipelineRasterizationStateCreateInfo rasterizer{
		    .depthClampEnable        = VK_FALSE,
		    .rasterizerDiscardEnable = VK_FALSE,
		    .polygonMode             = vk::PolygonMode::eFill,
		    .cullMode                = vk::CullModeFlagBits::eBack,
		    .frontFace               = vk::FrontFace::eCounterClockwise,
		    .depthBiasEnable         = VK_FALSE,
		    .lineWidth               = 1.0f};

		// Depth/Stencil
		vk::PipelineDepthStencilStateCreateInfo depthStencil{
		    .depthTestEnable  = vk::True,
		    .depthWriteEnable = vk::True,
		    .depthCompareOp   = vk::CompareOp::eLessOrEqual};

		// Multisampling
		vk::PipelineMultisampleStateCreateInfo multisampling{
		    .rasterizationSamples = vk::SampleCountFlagBits::e1,
		    .sampleShadingEnable  = VK_FALSE};

		// Color blending
		vk::PipelineColorBlendAttachmentState colorBlendAttachment{
		    .blendEnable    = VK_FALSE,
		    .colorWriteMask = vk::ColorComponentFlagBits::eR | vk::ColorComponentFlagBits::eG | vk::ColorComponentFlagBits::eB | vk::ColorComponentFlagBits::eA};

		vk::PipelineColorBlendStateCreateInfo colorBlending{
		    .logicOpEnable   = VK_FALSE,
		    .logicOp         = vk::LogicOp::eCopy,
		    .attachmentCount = 1,
		    .pAttachments    = &colorBlendAttachment};

		// Dynamic states
		std::vector<vk::DynamicState> dynamicStates = {
		    vk::DynamicState::eViewport,
		    vk::DynamicState::eScissor};

		vk::PipelineDynamicStateCreateInfo dynamicState{
		    .dynamicStateCount = static_cast<uint32_t>(dynamicStates.size()),
		    .pDynamicStates    = dynamicStates.data()};

		// Pipeline layout
		vk::PipelineLayoutCreateInfo pipelineLayoutInfo{
		    .setLayoutCount = 1,
		    .pSetLayouts    = &*descriptorSetLayout};

		pipelineLayout = device.createPipelineLayout(pipelineLayoutInfo);

		// Create the graphics pipeline
		vk::GraphicsPipelineCreateInfo pipelineInfo{
		    .stageCount          = 2,
		    .pStages             = shaderStages,
		    .pVertexInputState   = &vertexInputInfo,
		    .pInputAssemblyState = &inputAssembly,
		    .pViewportState      = &viewportState,
		    .pRasterizationState = &rasterizer,
		    .pMultisampleState   = &multisampling,
		    .pDepthStencilState  = &depthStencil,
		    .pColorBlendState    = &colorBlending,
		    .pDynamicState       = &dynamicState,
		    .layout              = *pipelineLayout,
		    .renderPass          = *renderPass,
		    .subpass             = 0};

		// Create the pipeline
		graphicsPipeline = device.createGraphicsPipeline(nullptr, pipelineInfo);
	}

	// Create framebuffers
	void createFramebuffers()
	{
		swapChainFramebuffers.reserve(swapChainImageViews.size());

		for (size_t i = 0; i < swapChainImageViews.size(); i++)
		{
			vk::ImageView attachments[] = {
			    *swapChainImageViews[i],
			    *depthImageView};

			vk::FramebufferCreateInfo framebufferInfo{
			    .renderPass      = *renderPass,
			    .attachmentCount = 2,
			    .pAttachments    = attachments,
			    .width           = swapChainExtent.width,
			    .height          = swapChainExtent.height,
			    .layers          = 1};

			swapChainFramebuffers.push_back(device.createFramebuffer(framebufferInfo));
		}
	}

	// Create command pool
	void createCommandPool()
	{
		vk::CommandPoolCreateInfo poolInfo{
		    .flags            = vk::CommandPoolCreateFlagBits::eResetCommandBuffer,
		    .queueFamilyIndex = queueIndex};

		commandPool = device.createCommandPool(poolInfo);
	}

	vk::Format findSupportedFormat(const std::vector<vk::Format> &candidates, vk::ImageTiling tiling, vk::FormatFeatureFlags features) const
	{
		for (const auto format : candidates)
		{
			vk::FormatProperties props = physicalDevice.getFormatProperties(format);

			if (tiling == vk::ImageTiling::eLinear && (props.linearTilingFeatures & features) == features)
			{
				return format;
			}
			if (tiling == vk::ImageTiling::eOptimal && (props.optimalTilingFeatures & features) == features)
			{
				return format;
			}
		}

		throw std::runtime_error("failed to find supported format!");
	}

	void createDepthResources()
	{
		depthFormat = findSupportedFormat({vk::Format::eD32Sfloat, vk::Format::eD32SfloatS8Uint, vk::Format::eD24UnormS8Uint},
		                                  vk::ImageTiling::eOptimal,
		                                  vk::FormatFeatureFlagBits::eDepthStencilAttachment);

		createImage(swapChainExtent.width, swapChainExtent.height, 1, depthFormat, vk::ImageTiling::eOptimal, vk::ImageUsageFlagBits::eDepthStencilAttachment, vk::MemoryPropertyFlagBits::eDeviceLocal, depthImage, depthImageMemory);
		depthImageView = createImageView(depthImage, depthFormat, vk::ImageAspectFlagBits::eDepth, 1);
	}

	// Create texture image
	void createTextureImage()
	{
		// Load texture image
		int      texWidth, texHeight, texChannels;
		stbi_uc *pixels = nullptr;

#if PLATFORM_ANDROID
		// Load image from Android assets
		std::optional<AssetManagerType *> optionalAssetManager = assetManager;
		std::vector<char>                 imageData            = readFile(TEXTURE_PATH, optionalAssetManager);
		pixels                                                 = stbi_load_from_memory(
            reinterpret_cast<const stbi_uc *>(imageData.data()),
            static_cast<int>(imageData.size()),
            &texWidth, &texHeight, &texChannels, STBI_rgb_alpha);
#else
		// Load image from filesystem
		pixels = stbi_load(TEXTURE_PATH.c_str(), &texWidth, &texHeight, &texChannels, STBI_rgb_alpha);
#endif

		if (!pixels)
		{
			throw std::runtime_error("Failed to load texture image: " + TEXTURE_PATH);
		}

		LOG_INFO("Texture loaded successfully");

		vk::DeviceSize imageSize = texWidth * texHeight * 4;

		// Create staging buffer
		vk::raii::Buffer       stagingBuffer       = nullptr;
		vk::raii::DeviceMemory stagingBufferMemory = nullptr;

		createBuffer(imageSize, vk::BufferUsageFlagBits::eTransferSrc, vk::MemoryPropertyFlagBits::eHostVisible | vk::MemoryPropertyFlagBits::eHostCoherent, stagingBuffer, stagingBufferMemory);

		// Copy pixel data to staging buffer
		void *data;
		data = stagingBufferMemory.mapMemory(0, imageSize);
		memcpy(data, pixels, static_cast<size_t>(imageSize));
		stagingBufferMemory.unmapMemory();

		// Free the pixel data
		if (pixels != nullptr)
		{
			stbi_image_free(pixels);
		}

		// Create image
		createImage(texWidth, texHeight, 1, vk::Format::eR8G8B8A8Srgb, vk::ImageTiling::eOptimal, vk::ImageUsageFlagBits::eTransferSrc | vk::ImageUsageFlagBits::eTransferDst | vk::ImageUsageFlagBits::eSampled, vk::MemoryPropertyFlagBits::eDeviceLocal, textureImage, textureImageMemory);

		// Transition image layout and copy buffer to image
		transitionImageLayout(textureImage, vk::Format::eR8G8B8A8Srgb, vk::ImageLayout::eUndefined, vk::ImageLayout::eTransferDstOptimal);
		copyBufferToImage(stagingBuffer, textureImage, static_cast<uint32_t>(texWidth), static_cast<uint32_t>(texHeight));
		transitionImageLayout(textureImage, vk::Format::eR8G8B8A8Srgb, vk::ImageLayout::eTransferDstOptimal, vk::ImageLayout::eShaderReadOnlyOptimal);
	}

	// Create texture image view
	void createTextureImageView()
	{
		textureImageView = createImageView(textureImage, vk::Format::eR8G8B8A8Srgb, vk::ImageAspectFlagBits::eColor, 1);
	}

	// Create texture sampler
	void createTextureSampler()
	{
		vk::SamplerCreateInfo samplerInfo{
		    .magFilter               = vk::Filter::eLinear,
		    .minFilter               = vk::Filter::eLinear,
		    .mipmapMode              = vk::SamplerMipmapMode::eLinear,
		    .addressModeU            = vk::SamplerAddressMode::eRepeat,
		    .addressModeV            = vk::SamplerAddressMode::eRepeat,
		    .addressModeW            = vk::SamplerAddressMode::eRepeat,
		    .anisotropyEnable        = VK_TRUE,
		    .maxAnisotropy           = 16.0f,
		    .compareEnable           = VK_FALSE,
		    .compareOp               = vk::CompareOp::eAlways,
		    .borderColor             = vk::BorderColor::eIntOpaqueBlack,
		    .unnormalizedCoordinates = VK_FALSE};

		textureSampler = device.createSampler(samplerInfo);
	}

	// Load model
	void loadModel()
	{
		tinyobj::attrib_t                attrib;
		std::vector<tinyobj::shape_t>    shapes;
		std::vector<tinyobj::material_t> materials;
		std::string                      warn, err;

#if PLATFORM_ANDROID
		// Load OBJ file from Android assets
		std::optional<AssetManagerType *> optionalAssetManager = assetManager;
		std::vector<char>                 objData              = readFile(MODEL_PATH, optionalAssetManager);
		std::string                       objString(objData.begin(), objData.end());
		std::istringstream                objStream(objString);

		if (!tinyobj::LoadObj(&attrib, &shapes, &materials, &warn, &err, &objStream))
		{
			throw std::runtime_error("Failed to load model: " + MODEL_PATH + " - " + warn + err);
		}
#else
		// Load OBJ file from filesystem
		if (!tinyobj::LoadObj(&attrib, &shapes, &materials, &warn, &err, MODEL_PATH.c_str()))
		{
			throw std::runtime_error("Failed to load model: " + MODEL_PATH + " - " + warn + err);
		}
#endif

		std::unordered_map<Vertex, uint32_t> uniqueVertices{};

		for (const auto &shape : shapes)
		{
			for (const auto &index : shape.mesh.indices)
			{
				Vertex vertex{};

				vertex.pos = {
				    attrib.vertices[3 * index.vertex_index + 0],
				    attrib.vertices[3 * index.vertex_index + 1],
				    attrib.vertices[3 * index.vertex_index + 2]};

				vertex.texCoord = {
				    attrib.texcoords[2 * index.texcoord_index + 0],
				    1.0f - attrib.texcoords[2 * index.texcoord_index + 1]};

				vertex.color = {1.0f, 1.0f, 1.0f};

				if (uniqueVertices.count(vertex) == 0)
				{
					uniqueVertices[vertex] = static_cast<uint32_t>(vertices.size());
					vertices.push_back(vertex);
				}

				indices.push_back(uniqueVertices[vertex]);
			}
		}

		LOG_INFO("Model loaded successfully");
	}

	// Create vertex buffer
	void createVertexBuffer()
	{
		vk::DeviceSize bufferSize = sizeof(vertices[0]) * vertices.size();

		vk::raii::Buffer       stagingBuffer       = nullptr;
		vk::raii::DeviceMemory stagingBufferMemory = nullptr;
		createBuffer(bufferSize, vk::BufferUsageFlagBits::eTransferSrc, vk::MemoryPropertyFlagBits::eHostVisible | vk::MemoryPropertyFlagBits::eHostCoherent, stagingBuffer, stagingBufferMemory);

		void *data;
		data = stagingBufferMemory.mapMemory(0, bufferSize);
		memcpy(data, vertices.data(), (size_t) bufferSize);
		stagingBufferMemory.unmapMemory();

		createBuffer(bufferSize, vk::BufferUsageFlagBits::eTransferDst | vk::BufferUsageFlagBits::eVertexBuffer, vk::MemoryPropertyFlagBits::eDeviceLocal, vertexBuffer, vertexBufferMemory);

		copyBuffer(stagingBuffer, vertexBuffer, bufferSize);
	}

	// Create index buffer
	void createIndexBuffer()
	{
		vk::DeviceSize bufferSize = sizeof(indices[0]) * indices.size();

		vk::raii::Buffer       stagingBuffer       = nullptr;
		vk::raii::DeviceMemory stagingBufferMemory = nullptr;
		createBuffer(bufferSize, vk::BufferUsageFlagBits::eTransferSrc, vk::MemoryPropertyFlagBits::eHostVisible | vk::MemoryPropertyFlagBits::eHostCoherent, stagingBuffer, stagingBufferMemory);

		void *data;
		data = stagingBufferMemory.mapMemory(0, bufferSize);
		memcpy(data, indices.data(), (size_t) bufferSize);
		stagingBufferMemory.unmapMemory();

		createBuffer(bufferSize, vk::BufferUsageFlagBits::eTransferDst | vk::BufferUsageFlagBits::eIndexBuffer, vk::MemoryPropertyFlagBits::eDeviceLocal, indexBuffer, indexBufferMemory);

		copyBuffer(stagingBuffer, indexBuffer, bufferSize);
	}

	// Create uniform buffers
	void createUniformBuffers()
	{
		vk::DeviceSize bufferSize = sizeof(UniformBufferObject);

		uniformBuffers.clear();
		uniformBuffersMemory.clear();

		for (size_t i = 0; i < MAX_FRAMES_IN_FLIGHT; i++)
		{
			uniformBuffers.push_back(nullptr);
			uniformBuffersMemory.push_back(nullptr);
		}

		for (size_t i = 0; i < MAX_FRAMES_IN_FLIGHT; i++)
		{
			createBuffer(bufferSize, vk::BufferUsageFlagBits::eUniformBuffer, vk::MemoryPropertyFlagBits::eHostVisible | vk::MemoryPropertyFlagBits::eHostCoherent, uniformBuffers[i], uniformBuffersMemory[i]);
		}
	}

	// Create descriptor pool
	void createDescriptorPool()
	{
		std::array<vk::DescriptorPoolSize, 2> poolSizes = {
		    vk::DescriptorPoolSize{
		        .type            = vk::DescriptorType::eUniformBuffer,
		        .descriptorCount = static_cast<uint32_t>(MAX_FRAMES_IN_FLIGHT)},
		    vk::DescriptorPoolSize{
		        .type            = vk::DescriptorType::eCombinedImageSampler,
		        .descriptorCount = static_cast<uint32_t>(MAX_FRAMES_IN_FLIGHT)}};

		vk::DescriptorPoolCreateInfo poolInfo{
		    .maxSets       = static_cast<uint32_t>(MAX_FRAMES_IN_FLIGHT),
		    .poolSizeCount = static_cast<uint32_t>(poolSizes.size()),
		    .pPoolSizes    = poolSizes.data()};

		descriptorPool = device.createDescriptorPool(poolInfo);
	}

	// Create descriptor sets
	void createDescriptorSets()
	{
		std::vector<vk::DescriptorSetLayout> layouts(MAX_FRAMES_IN_FLIGHT, *descriptorSetLayout);
		vk::DescriptorSetAllocateInfo        allocInfo{
		           .descriptorPool     = *descriptorPool,
		           .descriptorSetCount = static_cast<uint32_t>(MAX_FRAMES_IN_FLIGHT),
		           .pSetLayouts        = layouts.data()};

		descriptorSets = device.allocateDescriptorSets(allocInfo);

		for (size_t i = 0; i < MAX_FRAMES_IN_FLIGHT; i++)
		{
			vk::DescriptorBufferInfo bufferInfo{
			    .buffer = *uniformBuffers[i],
			    .offset = 0,
			    .range  = sizeof(UniformBufferObject)};

			vk::DescriptorImageInfo imageInfo{
			    .sampler     = *textureSampler,
			    .imageView   = *textureImageView,
			    .imageLayout = vk::ImageLayout::eShaderReadOnlyOptimal};

			std::array<vk::WriteDescriptorSet, 2> descriptorWrites = {
			    vk::WriteDescriptorSet{
			        .dstSet          = *descriptorSets[i],
			        .dstBinding      = 0,
			        .dstArrayElement = 0,
			        .descriptorCount = 1,
			        .descriptorType  = vk::DescriptorType::eUniformBuffer,
			        .pBufferInfo     = &bufferInfo},
			    vk::WriteDescriptorSet{
			        .dstSet          = *descriptorSets[i],
			        .dstBinding      = 1,
			        .dstArrayElement = 0,
			        .descriptorCount = 1,
			        .descriptorType  = vk::DescriptorType::eCombinedImageSampler,
			        .pImageInfo      = &imageInfo}};

			device.updateDescriptorSets(descriptorWrites, nullptr);
		}
	}

	// Create command buffers
	void createCommandBuffers()
	{
		commandBuffers.reserve(MAX_FRAMES_IN_FLIGHT);

		vk::CommandBufferAllocateInfo allocInfo{
		    .commandPool        = *commandPool,
		    .level              = vk::CommandBufferLevel::ePrimary,
		    .commandBufferCount = static_cast<uint32_t>(MAX_FRAMES_IN_FLIGHT)};

		commandBuffers = device.allocateCommandBuffers(allocInfo);
	}

	// Create synchronization objects
	void createSyncObjects()
	{
		imageAvailableSemaphores.reserve(MAX_FRAMES_IN_FLIGHT);
		renderFinishedSemaphores.reserve(swapChainImages.size());
		inFlightFences.reserve(MAX_FRAMES_IN_FLIGHT);

		vk::SemaphoreCreateInfo semaphoreInfo{};
		vk::FenceCreateInfo     fenceInfo{
		        .flags = vk::FenceCreateFlagBits::eSignaled};

		for (size_t i = 0; i < MAX_FRAMES_IN_FLIGHT; i++)
		{
			imageAvailableSemaphores.push_back(device.createSemaphore(semaphoreInfo));
			inFlightFences.push_back(device.createFence(fenceInfo));
		}

		for (size_t i = 0; i < swapChainImages.size(); i++)
		{
			renderFinishedSemaphores.push_back(device.createSemaphore(semaphoreInfo));
		}
	}

	// Clean up swap chain
	void cleanupSwapChain()
	{
		swapChainFramebuffers.clear();
		swapChainImageViews.clear();

		// Semaphores tied to swapchain image indices need to be rebuilt on resize
		renderFinishedSemaphores.clear();
		for (auto &imageView : swapChainImageViews)
		{
			imageView = nullptr;
		}

		swapChainImageViews.clear();
		swapChain = nullptr;
	}

	// Record command buffer
	void recordCommandBuffer(vk::raii::CommandBuffer &commandBuffer, uint32_t imageIndex)
	{
		vk::CommandBufferBeginInfo beginInfo{};
		commandBuffer.begin(beginInfo);

		vk::ClearValue clearValues[]{
		    vk::ClearValue{vk::ClearColorValue(0.0f, 0.0f, 0.0f, 1.0f)},
		    vk::ClearValue{vk::ClearDepthStencilValue(1.0f, 0)}};

		vk::RenderPassBeginInfo renderPassInfo{
		    .renderPass  = *renderPass,
		    .framebuffer = *swapChainFramebuffers[imageIndex],
		    .renderArea  = {
		         .offset = {0, 0},
		         .extent = swapChainExtent},
		    .clearValueCount = 2,
		    .pClearValues    = clearValues};

		commandBuffer.beginRenderPass(renderPassInfo, vk::SubpassContents::eInline);
		commandBuffer.bindPipeline(vk::PipelineBindPoint::eGraphics, *graphicsPipeline);

		vk::Viewport viewport{
		    .x        = 0.0f,
		    .y        = 0.0f,
		    .width    = static_cast<float>(swapChainExtent.width),
		    .height   = static_cast<float>(swapChainExtent.height),
		    .minDepth = 0.0f,
		    .maxDepth = 1.0f};
		commandBuffer.setViewport(0, viewport);

		vk::Rect2D scissor{
		    .offset = {0, 0},
		    .extent = swapChainExtent};
		commandBuffer.setScissor(0, scissor);

		commandBuffer.bindVertexBuffers(0, {*vertexBuffer}, {0});
		commandBuffer.bindIndexBuffer(*indexBuffer, 0, vk::IndexType::eUint32);
		commandBuffer.bindDescriptorSets(vk::PipelineBindPoint::eGraphics, *pipelineLayout, 0, {*descriptorSets[frameIndex]}, nullptr);
		commandBuffer.drawIndexed(static_cast<uint32_t>(indices.size()), 1, 0, 0, 0);

		commandBuffer.endRenderPass();
		commandBuffer.end();
	}

	// Draw frame
	void drawFrame()
	{
		static_cast<void>(device.waitForFences({*inFlightFences[frameIndex]}, VK_TRUE, UINT64_MAX));

		auto [result, imageIndex] = swapChain.acquireNextImage(UINT64_MAX, *imageAvailableSemaphores[frameIndex], nullptr);

		// Due to VULKAN_HPP_HANDLE_ERROR_OUT_OF_DATE_AS_SUCCESS being defined, eErrorOutOfDateKHR can be checked as a result
		// here and does not need to be caught by an exception.
		if (result == vk::Result::eErrorOutOfDateKHR)
		{
			recreateSwapChain();
			return;
		}
		// On other success codes than eSuccess and eSuboptimalKHR we just throw an exception.
		// On any error code, aquireNextImage already threw an exception.
		if (result != vk::Result::eSuccess && result != vk::Result::eSuboptimalKHR)
		{
			assert(result == vk::Result::eTimeout || result == vk::Result::eNotReady);
			throw std::runtime_error("failed to acquire swap chain image!");
		}

		// Update uniform buffer with current transformation
		updateUniformBuffer(frameIndex);

		// Only reset the fence if we are submitting work
		device.resetFences(*inFlightFences[frameIndex]);

		commandBuffers[frameIndex].reset();
		recordCommandBuffer(commandBuffers[frameIndex], imageIndex);

		vk::PipelineStageFlags waitDestinationStageMask(vk::PipelineStageFlagBits::eColorAttachmentOutput);
		const vk::SubmitInfo   submitInfo{
		      .waitSemaphoreCount   = 1,
		      .pWaitSemaphores      = &*imageAvailableSemaphores[frameIndex],
		      .pWaitDstStageMask    = &waitDestinationStageMask,
		      .commandBufferCount   = 1,
		      .pCommandBuffers      = &*commandBuffers[frameIndex],
		      .signalSemaphoreCount = 1,
		      .pSignalSemaphores    = &*renderFinishedSemaphores[imageIndex]};
		queue.submit(submitInfo, *inFlightFences[frameIndex]);

		const vk::PresentInfoKHR presentInfoKHR{.waitSemaphoreCount = 1,
		                                        .pWaitSemaphores    = &*renderFinishedSemaphores[imageIndex],
		                                        .swapchainCount     = 1,
		                                        .pSwapchains        = &*swapChain,
		                                        .pImageIndices      = &imageIndex};
		result = queue.presentKHR(presentInfoKHR);
		// Due to VULKAN_HPP_HANDLE_ERROR_OUT_OF_DATE_AS_SUCCESS being defined, eErrorOutOfDateKHR can be checked as a result
		// here and does not need to be caught by an exception.
		if ((result == vk::Result::eSuboptimalKHR) || (result == vk::Result::eErrorOutOfDateKHR) || framebufferResized)
		{
			framebufferResized = false;
			recreateSwapChain();
		}
		else
		{
			// There are no other success codes than eSuccess; on any error code, presentKHR already threw an exception.
			assert(result == vk::Result::eSuccess);
		}

		frameIndex = (frameIndex + 1) % MAX_FRAMES_IN_FLIGHT;
	}

	// Recreate swap chain
	void recreateSwapChain()
	{
#if !PLATFORM_ANDROID
		// On desktop, wait until the framebuffer has a non-zero size (e.g., when window is minimized)
		int width = 0, height = 0;
		if (window)
		{
			glfwGetFramebufferSize(window, &width, &height);
			while (width == 0 || height == 0)
			{
				glfwGetFramebufferSize(window, &width, &height);
				glfwWaitEvents();
			}
		}
#endif
		// Wait for device to finish operations
		device.waitIdle();

		// Clean up old swap chain
		cleanupSwapChain();

		// Create new swap chain and dependent resources
		createSwapChain();
		createImageViews();
		createDepthResources();
		createFramebuffers();

		// Recreate per-swapchain-image present semaphores for presenting
		renderFinishedSemaphores.reserve(swapChainImages.size());
		vk::SemaphoreCreateInfo semaphoreInfo{};
		for (size_t i = 0; i < swapChainImages.size(); ++i)
		{
			renderFinishedSemaphores.push_back(device.createSemaphore(semaphoreInfo));
		}
	}

	// Get required extensions
	std::vector<const char *> getRequiredExtensions()
	{
#if PLATFORM_ANDROID
		// Android requires these extensions
		std::vector<const char *> extensions = {
		    VK_KHR_SURFACE_EXTENSION_NAME,
		    VK_KHR_ANDROID_SURFACE_EXTENSION_NAME};
#else
		// Get the required extensions from GLFW
		uint32_t                  glfwExtensionCount = 0;
		auto                      glfwExtensions     = glfwGetRequiredInstanceExtensions(&glfwExtensionCount);
		std::vector<const char *> extensions(glfwExtensions, glfwExtensions + glfwExtensionCount);
#endif

		// Check if the debug utils extension is available
		std::vector<vk::ExtensionProperties> props               = context.enumerateInstanceExtensionProperties();
		bool                                 debugUtilsAvailable = std::ranges::any_of(props,
		                                                                               [](vk::ExtensionProperties const &ep) {
                                                           return strcmp(ep.extensionName, vk::EXTDebugUtilsExtensionName) == 0;
                                                       });

		// Always include the debug utils extension if available
		if (debugUtilsAvailable)
		{
			extensions.push_back(vk::EXTDebugUtilsExtensionName);
#if PLATFORM_DESKTOP
		}
		else
		{
			LOG_INFO("VK_EXT_debug_utils extension not available. Validation layers may not work.");
#endif
		}

		return extensions;
	}

	static uint32_t chooseSwapMinImageCount(vk::SurfaceCapabilitiesKHR const &surfaceCapabilities)
	{
		auto minImageCount = std::max(3u, surfaceCapabilities.minImageCount);
		if ((0 < surfaceCapabilities.maxImageCount) && (surfaceCapabilities.maxImageCount < minImageCount))
		{
			minImageCount = surfaceCapabilities.maxImageCount;
		}
		return minImageCount;
	}

	// Choose swap surface format
	vk::SurfaceFormatKHR chooseSwapSurfaceFormat(const std::vector<vk::SurfaceFormatKHR> &availableFormats)
	{
		assert(!availableFormats.empty());
		const auto formatIt = std::ranges::find_if(
		    availableFormats,
		    [](const auto &format) { return format.format == vk::Format::eB8G8R8A8Srgb && format.colorSpace == vk::ColorSpaceKHR::eSrgbNonlinear; });
		return formatIt != availableFormats.end() ? *formatIt : availableFormats[0];
	}

	// Choose swap present mode
	vk::PresentModeKHR chooseSwapPresentMode(const std::vector<vk::PresentModeKHR> &availablePresentModes)
	{
		assert(std::ranges::any_of(availablePresentModes, [](auto presentMode) { return presentMode == vk::PresentModeKHR::eFifo; }));
		return std::ranges::any_of(availablePresentModes,
		                           [](const vk::PresentModeKHR value) { return vk::PresentModeKHR::eMailbox == value; }) ?
		           vk::PresentModeKHR::eMailbox :
		           vk::PresentModeKHR::eFifo;
	}

	// Choose swap extent
	vk::Extent2D chooseSwapExtent(const vk::SurfaceCapabilitiesKHR &capabilities)
	{
		if (capabilities.currentExtent.width != 0xFFFFFFFF)
		{
			return capabilities.currentExtent;
		}
		else
		{
#if PLATFORM_ANDROID
			// Get the window size from Android
			int32_t width  = ANativeWindow_getWidth(androidApp->window);
			int32_t height = ANativeWindow_getHeight(androidApp->window);
#else
			// Get the window size from GLFW
			int width, height;
			glfwGetFramebufferSize(window, &width, &height);
#endif

			vk::Extent2D actualExtent = {
			    static_cast<uint32_t>(width),
			    static_cast<uint32_t>(height)};

			actualExtent.width  = std::clamp(actualExtent.width,
			                                 capabilities.minImageExtent.width,
			                                 capabilities.maxImageExtent.width);
			actualExtent.height = std::clamp(actualExtent.height,
			                                 capabilities.minImageExtent.height,
			                                 capabilities.maxImageExtent.height);

			return actualExtent;
		}
	}

	// Query swap chain support
	SwapChainSupportDetails querySwapChainSupport(vk::raii::PhysicalDevice device)
	{
		SwapChainSupportDetails details;
		details.capabilities = device.getSurfaceCapabilitiesKHR(*surface);
		details.formats      = device.getSurfaceFormatsKHR(*surface);
		details.presentModes = device.getSurfacePresentModesKHR(*surface);
		return details;
	}

	// Create buffer
	void createBuffer(vk::DeviceSize size, vk::BufferUsageFlags usage, vk::MemoryPropertyFlags properties, vk::raii::Buffer &buffer, vk::raii::DeviceMemory &bufferMemory)
	{
		vk::BufferCreateInfo bufferInfo{
		    .size        = size,
		    .usage       = usage,
		    .sharingMode = vk::SharingMode::eExclusive};

		buffer = device.createBuffer(bufferInfo);

		vk::MemoryRequirements memRequirements = buffer.getMemoryRequirements();

		vk::MemoryAllocateInfo allocInfo{
		    .allocationSize  = memRequirements.size,
		    .memoryTypeIndex = findMemoryType(memRequirements.memoryTypeBits, properties)};

		bufferMemory = device.allocateMemory(allocInfo);
		buffer.bindMemory(*bufferMemory, 0);
	}

	// Copy buffer
	void copyBuffer(vk::raii::Buffer &srcBuffer, vk::raii::Buffer &dstBuffer, vk::DeviceSize size)
	{
		vk::CommandBufferAllocateInfo allocInfo{
		    .commandPool        = *commandPool,
		    .level              = vk::CommandBufferLevel::ePrimary,
		    .commandBufferCount = 1};

		vk::raii::CommandBuffer commandBuffer = std::move(device.allocateCommandBuffers(allocInfo)[0]);

		vk::CommandBufferBeginInfo beginInfo{
		    .flags = vk::CommandBufferUsageFlagBits::eOneTimeSubmit};
		commandBuffer.begin(beginInfo);

		vk::BufferCopy copyRegion{
		    .srcOffset = 0,
		    .dstOffset = 0,
		    .size      = size};
		commandBuffer.copyBuffer(*srcBuffer, *dstBuffer, copyRegion);

		commandBuffer.end();

		vk::SubmitInfo submitInfo{
		    .commandBufferCount = 1,
		    .pCommandBuffers    = &*commandBuffer};

		queue.submit(submitInfo, nullptr);
		queue.waitIdle();
	}

	// Find memory type
	uint32_t findMemoryType(uint32_t typeFilter, vk::MemoryPropertyFlags properties)
	{
		vk::PhysicalDeviceMemoryProperties memProperties = physicalDevice.getMemoryProperties();

		for (uint32_t i = 0; i < memProperties.memoryTypeCount; i++)
		{
			if ((typeFilter & (1 << i)) && (memProperties.memoryTypes[i].propertyFlags & properties) == properties)
			{
				return i;
			}
		}

		throw std::runtime_error("Failed to find suitable memory type");
	}

	// Create image view
	[[nodiscard]] vk::raii::ImageView createImageView(const vk::raii::Image &image, vk::Format format, vk::ImageAspectFlags aspectFlags, uint32_t mipLevels) const
	{
		vk::ImageViewCreateInfo viewInfo{
		    .image            = *image,
		    .viewType         = vk::ImageViewType::e2D,
		    .format           = format,
		    .subresourceRange = {aspectFlags, 0, mipLevels, 0, 1}};
		return vk::raii::ImageView(device, viewInfo);
	}

	void createImage(uint32_t width, uint32_t height, uint32_t mipLevels, vk::Format format, vk::ImageTiling tiling, vk::ImageUsageFlags usage, vk::MemoryPropertyFlags properties, vk::raii::Image &image, vk::raii::DeviceMemory &imageMemory)
	{
		vk::ImageCreateInfo imageInfo{
		    .imageType     = vk::ImageType::e2D,
		    .format        = format,
		    .extent        = {width, height, 1},
		    .mipLevels     = mipLevels,
		    .arrayLayers   = 1,
		    .samples       = vk::SampleCountFlagBits::e1,
		    .tiling        = tiling,
		    .usage         = usage,
		    .sharingMode   = vk::SharingMode::eExclusive,
		    .initialLayout = vk::ImageLayout::eUndefined};
		image = vk::raii::Image(device, imageInfo);

		vk::MemoryRequirements memRequirements = image.getMemoryRequirements();
		vk::MemoryAllocateInfo allocInfo{
		    .allocationSize  = memRequirements.size,
		    .memoryTypeIndex = findMemoryType(memRequirements.memoryTypeBits, properties)};
		imageMemory = vk::raii::DeviceMemory(device, allocInfo);
		image.bindMemory(*imageMemory, 0);
	}

	// Transition image layout
	void transitionImageLayout(vk::raii::Image &image, vk::Format format, vk::ImageLayout oldLayout, vk::ImageLayout newLayout)
	{
		vk::CommandBufferAllocateInfo allocInfo{
		    .commandPool        = *commandPool,
		    .level              = vk::CommandBufferLevel::ePrimary,
		    .commandBufferCount = 1};

		vk::raii::CommandBuffer commandBuffer = std::move(device.allocateCommandBuffers(allocInfo)[0]);

		vk::CommandBufferBeginInfo beginInfo{
		    .flags = vk::CommandBufferUsageFlagBits::eOneTimeSubmit};
		commandBuffer.begin(beginInfo);

		vk::ImageMemoryBarrier barrier{
		    .oldLayout           = oldLayout,
		    .newLayout           = newLayout,
		    .srcQueueFamilyIndex = VK_QUEUE_FAMILY_IGNORED,
		    .dstQueueFamilyIndex = VK_QUEUE_FAMILY_IGNORED,
		    .image               = *image,
		    .subresourceRange    = {
		           .aspectMask     = vk::ImageAspectFlagBits::eColor,
		           .baseMipLevel   = 0,
		           .levelCount     = 1,
		           .baseArrayLayer = 0,
		           .layerCount     = 1}};

		vk::PipelineStageFlags sourceStage;
		vk::PipelineStageFlags destinationStage;

		if (oldLayout == vk::ImageLayout::eUndefined && newLayout == vk::ImageLayout::eTransferDstOptimal)
		{
			barrier.srcAccessMask = vk::AccessFlagBits::eNone;
			barrier.dstAccessMask = vk::AccessFlagBits::eTransferWrite;

			sourceStage      = vk::PipelineStageFlagBits::eTopOfPipe;
			destinationStage = vk::PipelineStageFlagBits::eTransfer;
		}
		else if (oldLayout == vk::ImageLayout::eTransferDstOptimal && newLayout == vk::ImageLayout::eShaderReadOnlyOptimal)
		{
			barrier.srcAccessMask = vk::AccessFlagBits::eTransferWrite;
			barrier.dstAccessMask = vk::AccessFlagBits::eShaderRead;

			sourceStage      = vk::PipelineStageFlagBits::eTransfer;
			destinationStage = vk::PipelineStageFlagBits::eFragmentShader;
		}
		else
		{
			throw std::invalid_argument("Unsupported layout transition");
		}

		commandBuffer.pipelineBarrier(
		    sourceStage, destinationStage,
		    vk::DependencyFlagBits::eByRegion,
		    nullptr,
		    nullptr,
		    barrier);

		commandBuffer.end();

		vk::SubmitInfo submitInfo{
		    .commandBufferCount = 1,
		    .pCommandBuffers    = &*commandBuffer};

		queue.submit(submitInfo, nullptr);
		queue.waitIdle();
	}

	// Copy buffer to image
	void copyBufferToImage(vk::raii::Buffer &buffer, vk::raii::Image &image, uint32_t width, uint32_t height)
	{
		vk::CommandBufferAllocateInfo allocInfo{
		    .commandPool        = *commandPool,
		    .level              = vk::CommandBufferLevel::ePrimary,
		    .commandBufferCount = 1};

		vk::raii::CommandBuffer commandBuffer = std::move(device.allocateCommandBuffers(allocInfo)[0]);

		vk::CommandBufferBeginInfo beginInfo{
		    .flags = vk::CommandBufferUsageFlagBits::eOneTimeSubmit};
		commandBuffer.begin(beginInfo);

		vk::BufferImageCopy region{
		    .bufferOffset      = 0,
		    .bufferRowLength   = 0,
		    .bufferImageHeight = 0,
		    .imageSubresource  = {
		         .aspectMask     = vk::ImageAspectFlagBits::eColor,
		         .mipLevel       = 0,
		         .baseArrayLayer = 0,
		         .layerCount     = 1},
		    .imageOffset = {0, 0, 0},
		    .imageExtent = {width, height, 1}};

		commandBuffer.copyBufferToImage(
		    *buffer,
		    *image,
		    vk::ImageLayout::eTransferDstOptimal,
		    region);

		commandBuffer.end();

		vk::SubmitInfo submitInfo{
		    .commandBufferCount = 1,
		    .pCommandBuffers    = &*commandBuffer};

		queue.submit(submitInfo, nullptr);
		queue.waitIdle();
	}

	// Update uniform buffer
	void updateUniformBuffer(uint32_t currentImage)
	{
		static auto startTime = std::chrono::high_resolution_clock::now();

		auto  currentTime = std::chrono::high_resolution_clock::now();
		float time        = std::chrono::duration<float, std::chrono::seconds::period>(currentTime - startTime).count();

		UniformBufferObject ubo{};
		ubo.model = glm::rotate(glm::mat4(1.0f), time * glm::radians(90.0f), glm::vec3(0.0f, 0.0f, 1.0f));
		ubo.view  = glm::lookAt(glm::vec3(2.0f, 2.0f, 2.0f), glm::vec3(0.0f, 0.0f, 0.0f), glm::vec3(0.0f, 0.0f, 1.0f));
		ubo.proj  = glm::perspective(glm::radians(45.0f), swapChainExtent.width / (float) swapChainExtent.height, 0.1f, 10.0f);
		ubo.proj[1][1] *= -1;

		void *data;
		data = uniformBuffersMemory[currentImage].mapMemory(0, sizeof(ubo));
		memcpy(data, &ubo, sizeof(ubo));
		uniformBuffersMemory[currentImage].unmapMemory();
	}

#if PLATFORM_ANDROID
	// Handle app commands
	static void handleAppCommand(android_app *app, int32_t cmd)
	{
		auto *vulkanApp = static_cast<HelloTriangleApplication *>(app->userData);
		switch (cmd)
		{
			case APP_CMD_INIT_WINDOW:
				// Window created, initialize Vulkan
				if (app->window != nullptr)
				{
					vulkanApp->initVulkan();
				}
				break;
			case APP_CMD_TERM_WINDOW:
				// Window destroyed, clean up Vulkan
				vulkanApp->cleanup();
				break;
			default:
				break;
		}
	}

	// Handle input events
	static int32_t handleInputEvent(android_app *app, AInputEvent *event)
	{
		auto *vulkanApp = static_cast<HelloTriangleApplication *>(app->userData);
		if (AInputEvent_getType(event) == AINPUT_EVENT_TYPE_MOTION)
		{
			// Handle touch events
			float x = AMotionEvent_getX(event, 0);
			float y = AMotionEvent_getY(event, 0);

			// Process touch coordinates
			LOGI("Touch at: %f, %f", x, y);

			return 1;
		}
		return 0;
	}
#endif
};

// Platform-specific entry point
#if PLATFORM_ANDROID
// Android main entry point
void android_main(android_app *app)
{
	// Make sure glue isn't stripped
	app_dummy();

	try
	{
		// Create and run the Vulkan application
		HelloTriangleApplication vulkanApp(app);
		vulkanApp.run();
	}
	catch (const std::exception &e)
	{
		LOGE("Exception caught: %s", e.what());
	}
}
#else
// Desktop main entry point
int main()
{
	try
	{
		HelloTriangleApplication app;
		app.run();
	}
	catch (const std::exception &e)
	{
		std::cerr << e.what() << std::endl;
		return EXIT_FAILURE;
	}

	return EXIT_SUCCESS;
}
#endif