atdunbg
d9b1742327
vendored meshoptimizer 1.2（submodule 环境故障改直接 vendor）。Mesh 多 LOD：各级索引连续拼进同一 index buffer（共享顶点，LodRange 区间 选择），解析期 meshopt_simplify 生成链（LOD0=原始，每级目标减半， 相对误差 1e-2，小网格复用上级）；都在 worker 线程，加载零额外卡顿。 SceneRenderer 按 (mesh, LOD) 聚合 batch（lookup key = 指针<<8|lod）： 主视图按屏占比近似 radius/dist 选级（阈值 10/20 倍半径）；阴影级联 按级联序降级（越远越简）。Draw 按 batch.lod 选索引区间。 A/B 对比（100 背包 10×10，RTX 5060 Laptop，vsync off）： - 纯背包：LOD 开 93-95 fps / 关 90-107 fps —— 无显著差异 - 加 SunTemple：LOD 开 52-56 / 关 51-55 —— 无显著差异 结论：当前场景瓶颈不在顶点数（5060 顶点吞吐富余），LOD 基础设施 保留（几何复杂资产/低端 GPU/更大场景生效），帧率问题转向 Tracy GPU 时间线定位真实瓶颈。 坑：LodRange 记录"LOD 不足复用最后一级"时 first 不得继续累加 （buffer 无重复数据，累加必越界——VUID-08798 抓的）。
159 lines
7.4 KiB
C++
159 lines
7.4 KiB
C++
//
|
||
// Created by atdunbg on 2026/9/23.
|
||
//
|
||
|
||
#include <Engine/Renderer/Passes/ForwardPass.h>
|
||
#include <Engine/Renderer/Vertex.h>
|
||
#include <Engine/Renderer/Mesh.h>
|
||
#include <Engine/Renderer/Material.h>
|
||
|
||
#include <array>
|
||
#include <span>
|
||
|
||
namespace Engine::Passes
|
||
{
|
||
void ForwardPass::Register(RG::RenderGraph& rg)
|
||
{
|
||
rg.AddPass("Forward", [this](RG::RenderGraphContext& ctx) {
|
||
const vk::CommandBuffer cmd = ctx.cmd;
|
||
|
||
vk::ClearValue clearValue;
|
||
clearValue.color = vk::ClearColorValue(
|
||
std::array<float, 4>{0.1f, 0.1f, 0.1f, 0.0f});
|
||
|
||
vk::RenderingAttachmentInfo colorAttachments[2];
|
||
// HDR 附件由 Skybox pass 清屏并画背景，这里 Load 在其上合成几何
|
||
colorAttachments[0].setImageView(ctx.View(m_HDRColor))
|
||
.setImageLayout(vk::ImageLayout::eColorAttachmentOptimal)
|
||
.setLoadOp(vk::AttachmentLoadOp::eLoad)
|
||
.setStoreOp(vk::AttachmentStoreOp::eStore)
|
||
.setClearValue(clearValue);
|
||
|
||
// velocity 附件：清 0 = 静止（背景无几何 → 历史原位采样）
|
||
vk::ClearValue velocityClear;
|
||
velocityClear.color = vk::ClearColorValue(std::array<float, 4>{0.f, 0.f, 0.f, 0.f});
|
||
colorAttachments[1].setImageView(ctx.View(m_Velocity))
|
||
.setImageLayout(vk::ImageLayout::eColorAttachmentOptimal)
|
||
.setLoadOp(vk::AttachmentLoadOp::eClear)
|
||
.setStoreOp(vk::AttachmentStoreOp::eStore)
|
||
.setClearValue(velocityClear);
|
||
|
||
vk::RenderingAttachmentInfo depthAttachment;
|
||
depthAttachment.setImageView(ctx.View(m_Depth))
|
||
.setImageLayout(vk::ImageLayout::eDepthAttachmentOptimal)
|
||
.setLoadOp(vk::AttachmentLoadOp::eClear)
|
||
.setStoreOp(vk::AttachmentStoreOp::eDontCare)
|
||
.setClearValue(vk::ClearValue(
|
||
vk::ClearDepthStencilValue(1.0f, 0)));
|
||
|
||
std::array<vk::RenderingAttachmentInfo, 2> colorArray = {
|
||
colorAttachments[0], colorAttachments[1]};
|
||
|
||
vk::RenderingInfo renderingInfo;
|
||
renderingInfo.setRenderArea({{0, 0}, ctx.extent})
|
||
.setLayerCount(1)
|
||
.setColorAttachments(colorArray)
|
||
.setPDepthAttachment(&depthAttachment);
|
||
|
||
cmd.beginRendering(renderingInfo);
|
||
|
||
// 绑定 shader object
|
||
std::array<vk::ShaderStageFlagBits, 2> stages = {
|
||
vk::ShaderStageFlagBits::eVertex,
|
||
vk::ShaderStageFlagBits::eFragment
|
||
};
|
||
std::array<vk::ShaderEXT, 2> shaders = {m_Gpu.forwardVert, m_Gpu.forwardFrag};
|
||
cmd.bindShadersEXT(stages, shaders);
|
||
|
||
// Set 0：本帧槽位的 UBO；Set 2：本帧槽位的阴影图
|
||
cmd.bindDescriptorSets(vk::PipelineBindPoint::eGraphics,
|
||
m_Gpu.forwardLayout, 0,
|
||
m_Gpu.uboSets[ctx.frame.frameSlot], {});
|
||
cmd.bindDescriptorSets(vk::PipelineBindPoint::eGraphics,
|
||
m_Gpu.forwardLayout, 2,
|
||
m_Gpu.shadowSets[ctx.frame.frameSlot], {});
|
||
|
||
// ---- 动态状态 ----
|
||
cmd.setRasterizerDiscardEnable(VK_FALSE);
|
||
|
||
cmd.setDepthTestEnable(VK_TRUE);
|
||
cmd.setDepthWriteEnable(VK_TRUE);
|
||
cmd.setDepthCompareOp(vk::CompareOp::eLess);
|
||
cmd.setDepthBoundsTestEnable(VK_FALSE);
|
||
|
||
cmd.setStencilTestEnable(VK_FALSE);
|
||
cmd.setDepthBiasEnable(VK_FALSE);
|
||
cmd.setPolygonModeEXT(vk::PolygonMode::eFill);
|
||
cmd.setCullMode(vk::CullModeFlagBits::eNone);
|
||
cmd.setFrontFace(vk::FrontFace::eCounterClockwise);
|
||
cmd.setPrimitiveTopology(vk::PrimitiveTopology::eTriangleList);
|
||
cmd.setPrimitiveRestartEnable(VK_FALSE);
|
||
|
||
auto binding = Vertex::GetBindingDescription();
|
||
auto attributes = Vertex::GetAttributeDescriptions();
|
||
cmd.setVertexInputEXT(binding, attributes);
|
||
|
||
cmd.setRasterizationSamplesEXT(vk::SampleCountFlagBits::e1);
|
||
vk::SampleMask sampleMask = 0xFFFFFFFF;
|
||
cmd.setSampleMaskEXT(vk::SampleCountFlagBits::e1, sampleMask);
|
||
cmd.setAlphaToCoverageEnableEXT(VK_FALSE);
|
||
cmd.setAlphaToOneEnableEXT(VK_FALSE);
|
||
|
||
vk::Viewport viewport(0.0f, 0.0f,
|
||
static_cast<float>(ctx.extent.width),
|
||
static_cast<float>(ctx.extent.height),
|
||
0.0f, 1.0f);
|
||
cmd.setViewportWithCount(viewport);
|
||
|
||
vk::Rect2D scissor({0, 0}, ctx.extent);
|
||
cmd.setScissorWithCount(scissor);
|
||
|
||
vk::Bool32 blendEnable = VK_FALSE;
|
||
vk::ColorBlendEquationEXT blendEquation;
|
||
blendEquation.setSrcColorBlendFactor(vk::BlendFactor::eOne)
|
||
.setDstColorBlendFactor(vk::BlendFactor::eZero)
|
||
.setColorBlendOp(vk::BlendOp::eAdd)
|
||
.setSrcAlphaBlendFactor(vk::BlendFactor::eOne)
|
||
.setDstAlphaBlendFactor(vk::BlendFactor::eZero)
|
||
.setAlphaBlendOp(vk::BlendOp::eAdd);
|
||
|
||
vk::ColorComponentFlags colorWriteMask =
|
||
vk::ColorComponentFlagBits::eR |
|
||
vk::ColorComponentFlagBits::eG |
|
||
vk::ColorComponentFlagBits::eB |
|
||
vk::ColorComponentFlagBits::eA;
|
||
|
||
// 两个颜色附件（HDR + velocity）都是无混合直写
|
||
cmd.setColorBlendEnableEXT(0,
|
||
std::array<vk::Bool32, 2>{blendEnable, blendEnable});
|
||
cmd.setColorBlendEquationEXT(0,
|
||
std::array<vk::ColorBlendEquationEXT, 2>{blendEquation, blendEquation});
|
||
cmd.setColorWriteMaskEXT(0,
|
||
std::array<vk::ColorComponentFlags, 2>{colorWriteMask, colorWriteMask});
|
||
|
||
// ---- 绘制：GPUDriven 按 mesh 分 batch（相机视锥剔除后）----
|
||
// 实例矩阵已由 Renderer 写入 ObjectBuffer（per-slot SSBO）；
|
||
// 同 mesh 的 N 个可见实体合并为一次 drawIndexed(instanceCount=N)，
|
||
// draw 数 = 可见唯一 mesh 数，与实体数量解耦
|
||
for (const auto& batch : m_Frame.mainView.batches) {
|
||
cmd.bindDescriptorSets(vk::PipelineBindPoint::eGraphics,
|
||
m_Gpu.forwardLayout, 1,
|
||
batch.mesh->GetMaterial()->GetDescriptorSet(), {});
|
||
batch.mesh->Draw(cmd, batch.count, batch.first, batch.lod);
|
||
}
|
||
|
||
cmd.endRendering();
|
||
})
|
||
.Write(m_HDRColor, RG::ResourceUse::ColorAttachment)
|
||
.Write(m_Velocity, RG::ResourceUse::ColorAttachment)
|
||
.Write(m_Depth, RG::ResourceUse::DepthAttachment)
|
||
.Read(m_ShadowMaps[0], RG::ResourceUse::ShaderRead,
|
||
vk::ShaderStageFlagBits::eFragment)
|
||
.Read(m_ShadowMaps[1], RG::ResourceUse::ShaderRead,
|
||
vk::ShaderStageFlagBits::eFragment)
|
||
.Read(m_ShadowMaps[2], RG::ResourceUse::ShaderRead,
|
||
vk::ShaderStageFlagBits::eFragment);
|
||
}
|
||
|
||
} // namespace Engine::Passes
|