VkRenderPipeline.hpp

#pragma once
#include <RGL/Types.hpp>
#include <RGL/Pipeline.hpp>
#include "VkDevice.hpp"
#include "VkRenderPass.hpp"
#include <unordered_map>
#include "VkPipelineLibraryShared.hpp"
 
namespace RGL {
 
    struct PipelineLayoutVk : public IPipelineLayout {
        const std::shared_ptr<DeviceVk> owningDevice;
        VkPipelineLayout layout = VK_NULL_HANDLE;
        VkDescriptorSetLayout descriptorSetLayout = VK_NULL_HANDLE;
 
        std::unordered_map<uint32_t, VkShaderStageFlags> pushConstantBindingStageFlags;
        std::unordered_map<uint32_t, VkDescriptorType> bindingDescriptorTypes;
 
        PipelineLayoutVk(decltype(owningDevice), const PipelineLayoutDescriptor&);
        virtual ~PipelineLayoutVk();
    };
 
    struct RenderPipelineVk : public IRenderPipeline {
        const std::shared_ptr<DeviceVk> owningDevice;
        const std::shared_ptr<PipelineLayoutVk> pipelineLayout;
        VkPipeline graphicsPipeline = VK_NULL_HANDLE;
 
        BufferBindingStore vsBufferBindings, fsBufferBindings;
 
        RenderPipelineVk(decltype(owningDevice), const RenderPipelineDescriptor&);
        virtual ~RenderPipelineVk();
    };
 
    
}