mirror of
https://github.com/skyline-emu/skyline.git
synced 2024-10-16 21:56:50 +03:00
Serialize Maxwell3D Draw State for Subpass
All Maxwell3D state was passed by reference to the draw command lambda, this would break if there was more than one pass or the state was changed in any way before execution. All state has now been serialized by value into the draw command lambda capture, retaining state regardless of mutations of the class state.
This commit is contained in:
PixelyIon
parent
934130b3e6
commit
defbfe8f78
@ -48,10 +48,6 @@ namespace skyline::gpu::interconnect {
|
||||
public:
|
||||
GraphicsContext(GPU &gpu, soc::gm20b::ChannelContext &channelCtx, gpu::interconnect::CommandExecutor &executor) : gpu(gpu), channelCtx(channelCtx), executor(executor), pipelineCache(gpu.vkDevice, vk::PipelineCacheCreateInfo{}) {
|
||||
scissors.fill(DefaultScissor);
|
||||
if (!gpu.quirks.supportsMultipleViewports) {
|
||||
viewportState.viewportCount = 1;
|
||||
viewportState.scissorCount = 1;
|
||||
}
|
||||
|
||||
u32 bindingIndex{};
|
||||
for (auto &vertexBuffer : vertexBuffers) {
|
||||
@ -59,8 +55,6 @@ namespace skyline::gpu::interconnect {
|
||||
vertexBuffer.bindingDivisorDescription.binding = bindingIndex;
|
||||
bindingIndex++;
|
||||
}
|
||||
if (!gpu.quirks.supportsVertexAttributeDivisor)
|
||||
vertexState.unlink<vk::PipelineVertexInputDivisorStateCreateInfoEXT>();
|
||||
|
||||
u32 attributeIndex{};
|
||||
for (auto &vertexAttribute : vertexAttributes)
|
||||
@ -371,12 +365,6 @@ namespace skyline::gpu::interconnect {
|
||||
.extent.height = std::numeric_limits<i32>::max(),
|
||||
.extent.width = std::numeric_limits<i32>::max(),
|
||||
}; //!< A scissor which displays the entire viewport, utilized when the viewport scissor is disabled
|
||||
vk::PipelineViewportStateCreateInfo viewportState{
|
||||
.pViewports = viewports.data(),
|
||||
.viewportCount = maxwell3d::ViewportCount,
|
||||
.pScissors = scissors.data(),
|
||||
.scissorCount = maxwell3d::ViewportCount,
|
||||
};
|
||||
|
||||
public:
|
||||
/**
|
||||
@ -696,7 +684,7 @@ namespace skyline::gpu::interconnect {
|
||||
bool needsRecompile{}; //!< If the shader needs to be recompiled as runtime information has changed
|
||||
ShaderCompiler::VaryingState previousStageStores{};
|
||||
u32 bindingBase{}, bindingLast{}; //!< The base and last binding for descriptors bound to this stage
|
||||
std::optional<vk::raii::ShaderModule> vkModule;
|
||||
std::shared_ptr<vk::raii::ShaderModule> vkModule;
|
||||
|
||||
std::array<ConstantBuffer, maxwell3d::PipelineStageConstantBufferCount> constantBuffers{};
|
||||
|
||||
@ -725,9 +713,6 @@ namespace skyline::gpu::interconnect {
|
||||
ShaderSet shaders;
|
||||
PipelineStages pipelineStages;
|
||||
|
||||
std::array<vk::PipelineShaderStageCreateInfo, maxwell3d::PipelineStageCount> shaderStagesInfo{}; //!< Storage backing for the pipeline shader stage information for all shaders aside from 'VertexA' which uses the same stage as 'VertexB'
|
||||
std::optional<vk::raii::DescriptorSetLayout> descriptorSetLayout{}; //!< The descriptor set layout for the pipeline (Only valid when `activeShaderStagesInfoCount` is non-zero)
|
||||
|
||||
ShaderCompiler::RuntimeInfo runtimeInfo{};
|
||||
|
||||
constexpr static size_t MaxShaderBytecodeSize{1 * 1024 * 1024}; //!< The largest shader binary that we support (1 MiB)
|
||||
@ -746,8 +731,9 @@ namespace skyline::gpu::interconnect {
|
||||
* @note The `descriptorSetWrite` will have a null `dstSet` which needs to be assigned prior to usage
|
||||
*/
|
||||
struct ShaderProgramState {
|
||||
span<vk::PipelineShaderStageCreateInfo> shaders;
|
||||
vk::DescriptorSetLayout descriptorSetLayout;
|
||||
boost::container::static_vector<std::shared_ptr<vk::raii::ShaderModule>, maxwell3d::PipelineStageCount> shaderModules; //!< Shader modules for every pipeline stage
|
||||
boost::container::static_vector<vk::PipelineShaderStageCreateInfo, maxwell3d::PipelineStageCount> shaderStages; //!< Shader modules for every pipeline stage
|
||||
vk::raii::DescriptorSetLayout descriptorSetLayout; //!< The descriptor set layout for the pipeline (Only valid when `activeShaderStagesInfoCount` is non-zero)
|
||||
span<vk::WriteDescriptorSet> descriptorSetWrites; //!< The writes to the descriptor set that need to be done prior to executing a pipeline
|
||||
};
|
||||
|
||||
@ -861,7 +847,8 @@ namespace skyline::gpu::interconnect {
|
||||
runtimeInfo.previous_stage_stores.mask.set(); // First stage should always have all bits set
|
||||
ShaderCompiler::Backend::Bindings bindings{};
|
||||
|
||||
size_t count{};
|
||||
boost::container::static_vector<std::shared_ptr<vk::raii::ShaderModule>, maxwell3d::PipelineStageCount> shaderModules;
|
||||
boost::container::static_vector<vk::PipelineShaderStageCreateInfo, maxwell3d::PipelineStageCount> shaderStages;
|
||||
for (auto &pipelineStage : pipelineStages) {
|
||||
if (!pipelineStage.enabled)
|
||||
continue;
|
||||
@ -874,7 +861,7 @@ namespace skyline::gpu::interconnect {
|
||||
if (pipelineStage.needsRecompile || bindings.unified != pipelineStage.bindingBase || pipelineStage.previousStageStores.mask != runtimeInfo.previous_stage_stores.mask) {
|
||||
pipelineStage.previousStageStores = runtimeInfo.previous_stage_stores;
|
||||
pipelineStage.bindingBase = bindings.unified;
|
||||
pipelineStage.vkModule = gpu.shader.CompileShader(runtimeInfo, program, bindings);
|
||||
pipelineStage.vkModule = std::make_shared<vk::raii::ShaderModule>(gpu.shader.CompileShader(runtimeInfo, program, bindings));
|
||||
pipelineStage.bindingLast = bindings.unified;
|
||||
}
|
||||
|
||||
@ -951,21 +938,21 @@ namespace skyline::gpu::interconnect {
|
||||
}
|
||||
}
|
||||
|
||||
shaderStagesInfo[count++] = vk::PipelineShaderStageCreateInfo{
|
||||
shaderModules.emplace_back(pipelineStage.vkModule);
|
||||
shaderStages.emplace_back(vk::PipelineShaderStageCreateInfo{
|
||||
.stage = pipelineStage.vkStage,
|
||||
.module = **pipelineStage.vkModule,
|
||||
.pName = "main",
|
||||
};
|
||||
});
|
||||
}
|
||||
|
||||
descriptorSetLayout.emplace(gpu.vkDevice, vk::DescriptorSetLayoutCreateInfo{
|
||||
.pBindings = layoutBindings.data(),
|
||||
.bindingCount = static_cast<u32>(layoutBindings.size()),
|
||||
});
|
||||
|
||||
return {
|
||||
span(shaderStagesInfo.data(), count),
|
||||
**descriptorSetLayout,
|
||||
std::move(shaderModules),
|
||||
std::move(shaderStages),
|
||||
vk::raii::DescriptorSetLayout(gpu.vkDevice, vk::DescriptorSetLayoutCreateInfo{
|
||||
.pBindings = layoutBindings.data(),
|
||||
.bindingCount = static_cast<u32>(layoutBindings.size()),
|
||||
}),
|
||||
descriptorSetWrites,
|
||||
};
|
||||
}
|
||||
@ -1384,24 +1371,12 @@ namespace skyline::gpu::interconnect {
|
||||
std::shared_ptr<BufferView> view;
|
||||
};
|
||||
std::array<VertexBuffer, maxwell3d::VertexBufferCount> vertexBuffers{};
|
||||
boost::container::static_vector<vk::VertexInputBindingDescription, maxwell3d::VertexBufferCount> vertexBindingDescriptions{};
|
||||
boost::container::static_vector<vk::VertexInputBindingDivisorDescriptionEXT, maxwell3d::VertexBufferCount> vertexBindingDivisorsDescriptions{};
|
||||
|
||||
struct VertexAttribute {
|
||||
bool enabled{};
|
||||
vk::VertexInputAttributeDescription description;
|
||||
};
|
||||
std::array<VertexAttribute, maxwell3d::VertexAttributeCount> vertexAttributes{};
|
||||
boost::container::static_vector<vk::VertexInputAttributeDescription, maxwell3d::VertexAttributeCount> vertexAttributesDescriptions{};
|
||||
|
||||
vk::StructureChain<vk::PipelineVertexInputStateCreateInfo, vk::PipelineVertexInputDivisorStateCreateInfoEXT> vertexState{
|
||||
vk::PipelineVertexInputStateCreateInfo{
|
||||
.pVertexBindingDescriptions = vertexBindingDescriptions.data(),
|
||||
.pVertexAttributeDescriptions = vertexAttributesDescriptions.data(),
|
||||
}, vk::PipelineVertexInputDivisorStateCreateInfoEXT{
|
||||
.pVertexBindingDivisors = vertexBindingDivisorsDescriptions.data(),
|
||||
}
|
||||
};
|
||||
|
||||
public:
|
||||
void SetVertexBufferStride(u32 index, u32 stride) {
|
||||
@ -2078,86 +2053,21 @@ namespace skyline::gpu::interconnect {
|
||||
|
||||
/* Draws */
|
||||
private:
|
||||
vk::GraphicsPipelineCreateInfo pipelineState{
|
||||
.pVertexInputState = &vertexState.get<vk::PipelineVertexInputStateCreateInfo>(),
|
||||
.pInputAssemblyState = &inputAssemblyState,
|
||||
.pViewportState = &viewportState,
|
||||
.pRasterizationState = &rasterizerState.get<vk::PipelineRasterizationStateCreateInfo>(),
|
||||
.pMultisampleState = &multisampleState,
|
||||
.pDepthStencilState = &depthState,
|
||||
.pColorBlendState = &blendState,
|
||||
.pDynamicState = nullptr,
|
||||
};
|
||||
vk::raii::PipelineCache pipelineCache;
|
||||
|
||||
public:
|
||||
template<bool IsIndexed>
|
||||
void Draw(u32 count, u32 first, i32 vertexOffset = 0) {
|
||||
// Color Render Target Setup
|
||||
boost::container::static_vector<std::scoped_lock<TextureView>, maxwell3d::RenderTargetCount> colorRenderTargetLocks;
|
||||
boost::container::static_vector<TextureView *, maxwell3d::RenderTargetCount> activeColorRenderTargets;
|
||||
|
||||
for (u32 index{}; index < maxwell3d::RenderTargetCount; index++) {
|
||||
auto renderTarget{GetColorRenderTarget(index)};
|
||||
if (renderTarget) {
|
||||
colorRenderTargetLocks.emplace_back(*renderTarget);
|
||||
activeColorRenderTargets.push_back(renderTarget);
|
||||
}
|
||||
}
|
||||
|
||||
blendState.attachmentCount = static_cast<u32>(activeColorRenderTargets.size());
|
||||
|
||||
// Depth/Stencil Render Target Setup
|
||||
auto depthRenderTargetView{GetDepthRenderTarget()};
|
||||
std::optional<std::scoped_lock<TextureView>> depthTargetLock;
|
||||
if (depthRenderTargetView)
|
||||
depthTargetLock.emplace(*depthRenderTargetView);
|
||||
|
||||
// Vertex Buffer Setup
|
||||
std::array<vk::Buffer, maxwell3d::VertexBufferCount> vertexBufferHandles{};
|
||||
std::array<vk::DeviceSize, maxwell3d::VertexBufferCount> vertexBufferOffsets{};
|
||||
|
||||
vertexBindingDescriptions.clear();
|
||||
vertexBindingDivisorsDescriptions.clear();
|
||||
|
||||
for (u32 index{}; index < maxwell3d::VertexBufferCount; index++) {
|
||||
auto vertexBufferView{GetVertexBuffer(index)};
|
||||
if (vertexBufferView) {
|
||||
auto &vertexBuffer{vertexBuffers[index]};
|
||||
vertexBindingDescriptions.push_back(vertexBuffer.bindingDescription);
|
||||
vertexBindingDivisorsDescriptions.push_back(vertexBuffer.bindingDivisorDescription);
|
||||
|
||||
std::scoped_lock vertexBufferLock(*vertexBufferView);
|
||||
executor.AttachBuffer(vertexBufferView);
|
||||
vertexBufferHandles[index] = vertexBufferView->buffer->GetBacking();
|
||||
vertexBufferOffsets[index] = vertexBufferView->offset;
|
||||
}
|
||||
}
|
||||
|
||||
vertexState.get<vk::PipelineVertexInputStateCreateInfo>().vertexBindingDescriptionCount = static_cast<u32>(vertexBindingDescriptions.size());
|
||||
vertexState.get<vk::PipelineVertexInputDivisorStateCreateInfoEXT>().vertexBindingDivisorCount = static_cast<u32>(vertexBindingDivisorsDescriptions.size());
|
||||
|
||||
// Vertex Attribute Setup
|
||||
vertexAttributesDescriptions.clear();
|
||||
|
||||
for (auto &vertexAttribute : vertexAttributes)
|
||||
if (vertexAttribute.enabled)
|
||||
vertexAttributesDescriptions.push_back(vertexAttribute.description);
|
||||
|
||||
vertexState.get<vk::PipelineVertexInputStateCreateInfo>().vertexAttributeDescriptionCount = static_cast<u32>(vertexAttributesDescriptions.size());
|
||||
|
||||
// Shader + Binding Setup
|
||||
auto programState{CompileShaderProgramState()};
|
||||
pipelineState.pStages = programState.shaders.data();
|
||||
pipelineState.stageCount = static_cast<u32>(programState.shaders.size());
|
||||
|
||||
auto descriptorSet{gpu.descriptor.AllocateSet(programState.descriptorSetLayout)};
|
||||
auto descriptorSet{gpu.descriptor.AllocateSet(*programState.descriptorSetLayout)};
|
||||
for (auto &descriptorSetWrite : programState.descriptorSetWrites)
|
||||
descriptorSetWrite.dstSet = descriptorSet;
|
||||
gpu.vkDevice.updateDescriptorSets(programState.descriptorSetWrites, nullptr);
|
||||
|
||||
vk::raii::PipelineLayout pipelineLayout(gpu.vkDevice, vk::PipelineLayoutCreateInfo{
|
||||
.pSetLayouts = &programState.descriptorSetLayout,
|
||||
.pSetLayouts = &*programState.descriptorSetLayout,
|
||||
.setLayoutCount = 1,
|
||||
});
|
||||
|
||||
@ -2174,6 +2084,52 @@ namespace skyline::gpu::interconnect {
|
||||
indexBufferType = indexBuffer.type;
|
||||
}
|
||||
|
||||
// Vertex Buffer Setup
|
||||
std::array<vk::Buffer, maxwell3d::VertexBufferCount> vertexBufferHandles{};
|
||||
std::array<vk::DeviceSize, maxwell3d::VertexBufferCount> vertexBufferOffsets{};
|
||||
|
||||
boost::container::static_vector<vk::VertexInputBindingDescription, maxwell3d::VertexBufferCount> vertexBindingDescriptions{};
|
||||
boost::container::static_vector<vk::VertexInputBindingDivisorDescriptionEXT, maxwell3d::VertexBufferCount> vertexBindingDivisorsDescriptions{};
|
||||
|
||||
for (u32 index{}; index < maxwell3d::VertexBufferCount; index++) {
|
||||
auto vertexBufferView{GetVertexBuffer(index)};
|
||||
if (vertexBufferView) {
|
||||
auto &vertexBuffer{vertexBuffers[index]};
|
||||
vertexBindingDescriptions.push_back(vertexBuffer.bindingDescription);
|
||||
vertexBindingDivisorsDescriptions.push_back(vertexBuffer.bindingDivisorDescription);
|
||||
|
||||
std::scoped_lock vertexBufferLock(*vertexBufferView);
|
||||
vertexBufferHandles[index] = vertexBufferView->buffer->GetBacking();
|
||||
vertexBufferOffsets[index] = vertexBufferView->offset;
|
||||
executor.AttachBuffer(vertexBufferView);
|
||||
}
|
||||
}
|
||||
|
||||
// Vertex Attribute Setup
|
||||
boost::container::static_vector<vk::VertexInputAttributeDescription, maxwell3d::VertexAttributeCount> vertexAttributesDescriptions{};
|
||||
for (auto &vertexAttribute : vertexAttributes)
|
||||
if (vertexAttribute.enabled)
|
||||
vertexAttributesDescriptions.push_back(vertexAttribute.description);
|
||||
|
||||
// Color Render Target + Blending Setup
|
||||
boost::container::static_vector<TextureView *, maxwell3d::RenderTargetCount> activeColorRenderTargets;
|
||||
for (u32 index{}; index < maxwell3d::RenderTargetCount; index++) {
|
||||
auto renderTarget{GetColorRenderTarget(index)};
|
||||
if (renderTarget) {
|
||||
std::scoped_lock lock(*renderTarget);
|
||||
activeColorRenderTargets.push_back(renderTarget);
|
||||
executor.AttachTexture(renderTarget);
|
||||
}
|
||||
}
|
||||
|
||||
boost::container::static_vector<vk::PipelineColorBlendAttachmentState, maxwell3d::RenderTargetCount> blendAttachmentStates(blendState.pAttachments, blendState.pAttachments + activeColorRenderTargets.size());
|
||||
|
||||
// Depth/Stencil Render Target Setup
|
||||
auto depthRenderTargetView{GetDepthRenderTarget()};
|
||||
std::optional<std::scoped_lock<TextureView>> depthTargetLock;
|
||||
if (depthRenderTargetView)
|
||||
depthTargetLock.emplace(*depthRenderTargetView);
|
||||
|
||||
// Draw Persistent Storage
|
||||
struct Storage : FenceCycleDependency {
|
||||
vk::raii::PipelineLayout pipelineLayout;
|
||||
@ -2186,11 +2142,47 @@ namespace skyline::gpu::interconnect {
|
||||
auto storage{std::make_shared<Storage>(std::move(pipelineLayout), std::move(descriptorSet))};
|
||||
|
||||
// Submit Draw
|
||||
executor.AddSubpass([=, &vkDevice = gpu.vkDevice, pipelineCreateInfo = pipelineState, storage = std::move(storage), vertexBufferHandles = std::move(vertexBufferHandles), vertexBufferOffsets = std::move(vertexBufferOffsets), pipelineCache = *pipelineCache](vk::raii::CommandBuffer &commandBuffer, const std::shared_ptr<FenceCycle> &cycle, GPU &, vk::RenderPass renderPass, u32 subpassIndex) mutable {
|
||||
pipelineCreateInfo.layout = *storage->pipelineLayout;
|
||||
executor.AddSubpass([=, &vkDevice = gpu.vkDevice, shaderModules = programState.shaderModules, shaderStages = programState.shaderStages, inputAssemblyState = inputAssemblyState, multiViewport = gpu.quirks.supportsMultipleViewports, viewports = viewports, scissors = scissors, rasterizerState = rasterizerState, multisampleState = multisampleState, depthState = depthState, blendState = blendState, storage = std::move(storage), supportsVertexAttributeDivisor = gpu.quirks.supportsVertexAttributeDivisor, vertexBufferHandles = std::move(vertexBufferHandles), vertexBufferOffsets = std::move(vertexBufferOffsets), pipelineCache = *pipelineCache](vk::raii::CommandBuffer &commandBuffer, const std::shared_ptr<FenceCycle> &cycle, GPU &, vk::RenderPass renderPass, u32 subpassIndex) mutable {
|
||||
vk::StructureChain<vk::PipelineVertexInputStateCreateInfo, vk::PipelineVertexInputDivisorStateCreateInfoEXT> vertexState{
|
||||
vk::PipelineVertexInputStateCreateInfo{
|
||||
.pVertexBindingDescriptions = vertexBindingDescriptions.data(),
|
||||
.vertexBindingDescriptionCount = static_cast<u32>(vertexBindingDescriptions.size()),
|
||||
.pVertexAttributeDescriptions = vertexAttributesDescriptions.data(),
|
||||
.vertexAttributeDescriptionCount = static_cast<u32>(vertexAttributesDescriptions.size()),
|
||||
}, vk::PipelineVertexInputDivisorStateCreateInfoEXT{
|
||||
.pVertexBindingDivisors = vertexBindingDivisorsDescriptions.data(),
|
||||
.vertexBindingDivisorCount = static_cast<u32>(vertexBindingDivisorsDescriptions.size()),
|
||||
}
|
||||
};
|
||||
|
||||
pipelineCreateInfo.renderPass = renderPass;
|
||||
pipelineCreateInfo.subpass = subpassIndex;
|
||||
if (!supportsVertexAttributeDivisor)
|
||||
vertexState.unlink<vk::PipelineVertexInputDivisorStateCreateInfoEXT>();
|
||||
|
||||
vk::PipelineViewportStateCreateInfo viewportState{
|
||||
.pViewports = viewports.data(),
|
||||
.viewportCount = static_cast<u32>(multiViewport ? maxwell3d::ViewportCount : 1),
|
||||
.pScissors = scissors.data(),
|
||||
.scissorCount = static_cast<u32>(multiViewport ? maxwell3d::ViewportCount : 1),
|
||||
};
|
||||
|
||||
blendState.pAttachments = blendAttachmentStates.data();
|
||||
blendState.attachmentCount = static_cast<u32>(blendAttachmentStates.size());
|
||||
|
||||
vk::GraphicsPipelineCreateInfo pipelineCreateInfo{
|
||||
.pStages = shaderStages.data(),
|
||||
.stageCount = static_cast<u32>(shaderStages.size()),
|
||||
.pVertexInputState = &vertexState.get<vk::PipelineVertexInputStateCreateInfo>(),
|
||||
.pInputAssemblyState = &inputAssemblyState,
|
||||
.pViewportState = &viewportState,
|
||||
.pRasterizationState = &rasterizerState.get<vk::PipelineRasterizationStateCreateInfo>(),
|
||||
.pMultisampleState = &multisampleState,
|
||||
.pDepthStencilState = &depthState,
|
||||
.pColorBlendState = &blendState,
|
||||
.pDynamicState = nullptr,
|
||||
.layout = *storage->pipelineLayout,
|
||||
.renderPass = renderPass,
|
||||
.subpass = subpassIndex,
|
||||
};
|
||||
|
||||
auto pipeline{(*vkDevice).createGraphicsPipeline(pipelineCache, pipelineCreateInfo, nullptr, *vkDevice.getDispatcher())};
|
||||
if (pipeline.result != vk::Result::eSuccess)
|
||||
|
||||
Loading…
Reference in New Issue
Block a user