raphiaura/KytyPS5
1
0
Fork 0
mirror of https://github.com/KytyPS5/KytyPS5.git synced 2026-10-20 10:56:58 +02:00
Code Issues Packages Projects Releases Wiki Activity

render: remove redundant cache setup

Browse Source
This commit is contained in:
nmzik
2026-09-05 14:37:04 +04:00
parent c66060be10
commit 1c68c94c05
5 changed files with 23 additions and 47 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/graphics/host_gpu/renderer/cache/bufferCache.cpp
Vendored
+20 -23
View File
@@ -408,7 +408,6 @@ bool BufferCache::SynchronizeBuffer(Buffer& buffer, uint64_t vaddr, uint64_t siz
command.EndRendering();
const auto native = command.Handle();
vk::BufferMemoryBarrier before {};
before.sType = vk::StructureType::eBufferMemoryBarrier;
before.srcAccessMask = vk::AccessFlagBits::eMemoryRead | vk::AccessFlagBits::eMemoryWrite |
vk::AccessFlagBits::eTransferRead |
vk::AccessFlagBits::eTransferWrite;
@@ -643,29 +642,27 @@ void BufferCache::CopyBuffer(uint64_t dst_vaddr, uint64_t src_vaddr, uint64_t si
" size=0x%016" PRIx64 " src_gds=%d dst_gds=%d\n",
src_vaddr, dst_vaddr, size, static_cast<int>(src_gds), static_cast<int>(dst_gds));
}
if (src_memory || dst_memory) {
const auto src_region =
src_memory ? m_texture_cache.QueryRegion(src_vaddr, size) : TextureCache::RegionInfo {};
const auto dst_region =
dst_memory ? m_texture_cache.QueryRegion(dst_vaddr, size) : TextureCache::RegionInfo {};
if (src_memory && dst_memory && !HasGpuDirtyBytes(src_vaddr, size) &&
!HasGpuDirtyBytes(dst_vaddr, size) && !src_region.gpu_image_bytes &&
!dst_region.gpu_image_bytes) {
if (dst_region.image_bytes) {
m_texture_cache.InvalidateMemory(dst_vaddr, size);
}
std::array<uint8_t, 64 * 1024> bytes;
for (uint64_t offset = 0; offset < size;) {
const auto chunk = std::min<uint64_t>(size - offset, bytes.size());
if (!Libs::LibKernel::Memory::TryReadBacking(src_vaddr + offset, bytes.data(),
chunk)) {
EXIT("BufferCache: host DMA source has no direct backing\n");
}
WriteHostMemory(dst_vaddr + offset, std::span {bytes}.first(chunk));
offset += chunk;
}
return;
const auto src_region =
src_memory ? m_texture_cache.QueryRegion(src_vaddr, size) : TextureCache::RegionInfo {};
const auto dst_region =
dst_memory ? m_texture_cache.QueryRegion(dst_vaddr, size) : TextureCache::RegionInfo {};
if (src_memory && dst_memory && !HasGpuDirtyBytes(src_vaddr, size) &&
!HasGpuDirtyBytes(dst_vaddr, size) && !src_region.gpu_image_bytes &&
!dst_region.gpu_image_bytes) {
if (dst_region.image_bytes) {
m_texture_cache.InvalidateMemory(dst_vaddr, size);
}
std::array<uint8_t, 64 * 1024> bytes;
for (uint64_t offset = 0; offset < size;) {
const auto chunk = std::min<uint64_t>(size - offset, bytes.size());
if (!Libs::LibKernel::Memory::TryReadBacking(src_vaddr + offset, bytes.data(),
chunk)) {
EXIT("BufferCache: host DMA source has no direct backing\n");
}
WriteHostMemory(dst_vaddr + offset, std::span {bytes}.first(chunk));
offset += chunk;
}
return;
}
auto& command = m_scheduler.Current();
src/graphics/host_gpu/renderer/cache/faultManager.cpp
Vendored
-8
View File
@@ -39,7 +39,6 @@ FaultManager::FaultManager(GraphicContext& graphics, CommandScheduler& scheduler
{1, vk::DescriptorType::eStorageBuffer, 1, vk::ShaderStageFlagBits::eCompute, nullptr},
};
vk::DescriptorSetLayoutCreateInfo layout_info {};
layout_info.sType = vk::StructureType::eDescriptorSetLayoutCreateInfo;
layout_info.flags = vk::DescriptorSetLayoutCreateFlagBits::ePushDescriptorKHR;
layout_info.bindingCount = std::size(bindings);
layout_info.pBindings = bindings;
@@ -49,7 +48,6 @@ FaultManager::FaultManager(GraphicContext& graphics, CommandScheduler& scheduler
"create fault-buffer descriptor layout");
vk::ShaderModuleCreateInfo module_info {};
module_info.sType = vk::StructureType::eShaderModuleCreateInfo;
module_info.codeSize = std::size(FAULT_BUFFER_PROCESS_SPV) * sizeof(uint32_t);
module_info.pCode = FAULT_BUFFER_PROCESS_SPV;
vk::ShaderModule module = nullptr;
@@ -57,7 +55,6 @@ FaultManager::FaultManager(GraphicContext& graphics, CommandScheduler& scheduler
"create fault-buffer shader module");
vk::PipelineLayoutCreateInfo pipeline_layout_info {};
pipeline_layout_info.sType = vk::StructureType::ePipelineLayoutCreateInfo;
pipeline_layout_info.setLayoutCount = 1;
pipeline_layout_info.pSetLayouts = &m_fault_process_desc_layout;
RequireVulkanSuccess(
@@ -66,12 +63,10 @@ FaultManager::FaultManager(GraphicContext& graphics, CommandScheduler& scheduler
"create fault-buffer pipeline layout");
vk::PipelineShaderStageCreateInfo stage {};
stage.sType = vk::StructureType::ePipelineShaderStageCreateInfo;
stage.stage = vk::ShaderStageFlagBits::eCompute;
stage.module = module;
stage.pName = "main";
vk::ComputePipelineCreateInfo pipeline_info {};
pipeline_info.sType = vk::StructureType::eComputePipelineCreateInfo;
pipeline_info.stage = stage;
pipeline_info.layout = m_fault_process_pipeline_layout;
const auto result = m_graphics.device.createComputePipelines(
@@ -99,7 +94,6 @@ void FaultManager::ProcessFaultBuffer() {
m_download_buffer.Flush(offset, PageFaultAreaSize);
vk::BufferMemoryBarrier2 pre_barrier {};
pre_barrier.sType = vk::StructureType::eBufferMemoryBarrier2;
pre_barrier.srcStageMask = vk::PipelineStageFlagBits2::eAllCommands;
pre_barrier.srcAccessMask = vk::AccessFlagBits2::eShaderWrite;
pre_barrier.dstStageMask = vk::PipelineStageFlagBits2::eComputeShader;
@@ -119,7 +113,6 @@ void FaultManager::ProcessFaultBuffer() {
};
std::array<vk::WriteDescriptorSet, 2> writes {};
for (uint32_t index = 0; index < writes.size(); ++index) {
writes[index].sType = vk::StructureType::eWriteDescriptorSet;
writes[index].dstBinding = index;
writes[index].descriptorCount = 1;
writes[index].descriptorType = vk::DescriptorType::eStorageBuffer;
@@ -129,7 +122,6 @@ void FaultManager::ProcessFaultBuffer() {
m_scheduler.EndRendering();
auto command = m_scheduler.Current().Handle();
vk::DependencyInfo dependency {};
dependency.sType = vk::StructureType::eDependencyInfo;
dependency.dependencyFlags = vk::DependencyFlagBits::eByRegion;
dependency.bufferMemoryBarrierCount = 1;
dependency.pBufferMemoryBarriers = &pre_barrier;
src/graphics/host_gpu/renderer/cache/samplerCache.cpp
Vendored
-3
View File
@@ -116,9 +116,6 @@ vk::Sampler SamplerCache::GetSampler(const ShaderSamplerResource& r) {
default: EXIT("unknown border color: %d", static_cast<int>(r.BorderColorType()));
}
sampler_info.sType = vk::StructureType::eSamplerCreateInfo;
sampler_info.pNext = nullptr;
sampler_info.flags = {};
sampler_info.magFilter = to_vk_filter(mag_filter);
sampler_info.minFilter = to_vk_filter(min_filter);
sampler_info.mipmapMode =
src/graphics/host_gpu/renderer/cache/streamBuffer.cpp
Vendored
-3
View File
@@ -71,7 +71,6 @@ Buffer::Buffer(GraphicContext& graphics, CommandScheduler& scheduler, MemoryUsag
vk::BufferCreateInfo buffer_info {};
buffer_info.size = size;
buffer_info.usage = flags;
buffer_info.sharingMode = vk::SharingMode::eExclusive;
const bool with_bda = bool(flags & vk::BufferUsageFlagBits::eShaderDeviceAddress);
const VmaAllocationCreateFlags bda_flag =
@@ -105,7 +104,6 @@ Buffer::Buffer(GraphicContext& graphics, CommandScheduler& scheduler, MemoryUsag
graphics.device.getBufferMemoryRequirements(m_buffer->buffer, &m_buffer->memory.requirements);
if (static_cast<bool>(flags & vk::BufferUsageFlagBits::eShaderDeviceAddress)) {
vk::BufferDeviceAddressInfo address_info {};
address_info.sType = vk::StructureType::eBufferDeviceAddressInfo;
address_info.buffer = m_buffer->buffer;
m_device_address = graphics.device.getBufferAddress(address_info);
EXIT_IF(m_device_address == 0);
@@ -174,7 +172,6 @@ vk::BufferMemoryBarrier Buffer::Barrier(uint64_t offset, uint64_t size, vk::Acce
static_cast<const void*>(Handle()), offset, size, m_size);
}
vk::BufferMemoryBarrier barrier {};
barrier.sType = vk::StructureType::eBufferMemoryBarrier;
barrier.srcAccessMask = source;
barrier.dstAccessMask = destination;
barrier.srcQueueFamilyIndex = VK_QUEUE_FAMILY_IGNORED;
src/graphics/host_gpu/renderer/cache/textureCache.cpp
Vendored
+3 -10
View File
@@ -116,10 +116,8 @@ bool TextureCache::SameBacking(const ImageInfo& cached, const ImageInfo& request
if (cached.tile_mode != requested.tile_mode) {
return false;
}
if (!ImageViewOps::FormatsCompatible(cached.pixel_format, requested.pixel_format)) {
return false;
}
if (cached.type != requested.type && requested.extent != vk::Extent3D {1, 1, 1}) {
if (!ImageViewOps::FormatsCompatible(cached.pixel_format, requested.pixel_format) ||
(cached.type != requested.type && requested.extent != vk::Extent3D {1, 1, 1})) {
return false;
}
if (exact_format && cached.pixel_format != requested.pixel_format) {
@@ -1036,18 +1034,14 @@ void TextureCache::InitializeImage(ImageId id) {
if (image.info.samples > 1) {
return;
}
bool data_imported = false;
const bool upload = image.IsBufferModified() || image.IsCpuDirty();
const bool upload = image.IsBufferModified() || image.IsCpuDirty();
if (upload) {
const auto [source, source_offset] =
m_buffer_cache.ObtainBufferForImage(image.info.data.address, image.info.data.size);
if (source == nullptr) {
EXIT("TextureCache: failed to obtain image upload source\n");
}
data_imported = true;
UploadImage(image, *source, source_offset);
}
if (data_imported) {
image.ClearBufferModified();
}
if (image.IsCpuDirty()) {
@@ -1694,7 +1688,6 @@ bool TextureCache::TryDownloadImage(ImageId id) {
DownloadImageData(image, download, offset, range.size, std::move(plan));
vk::BufferMemoryBarrier barrier {};
barrier.sType = vk::StructureType::eBufferMemoryBarrier;
barrier.srcAccessMask = vk::AccessFlagBits::eMemoryWrite | vk::AccessFlagBits::eTransferWrite |
vk::AccessFlagBits::eShaderWrite;
barrier.dstAccessMask = vk::AccessFlagBits::eHostRead;