blob: e041f8c7744124c762c75e790266f3ee0c9b4b41 [file] [log] [blame]
// Copyright 2017 The Dawn Authors
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
#include "dawn_native/d3d12/CommandBufferD3D12.h"
#include "common/Assert.h"
#include "dawn_native/CommandEncoder.h"
#include "dawn_native/Commands.h"
#include "dawn_native/d3d12/BindGroupD3D12.h"
#include "dawn_native/d3d12/BindGroupLayoutD3D12.h"
#include "dawn_native/d3d12/BufferD3D12.h"
#include "dawn_native/d3d12/ComputePipelineD3D12.h"
#include "dawn_native/d3d12/DescriptorHeapAllocator.h"
#include "dawn_native/d3d12/DeviceD3D12.h"
#include "dawn_native/d3d12/PipelineLayoutD3D12.h"
#include "dawn_native/d3d12/PlatformFunctions.h"
#include "dawn_native/d3d12/RenderPipelineD3D12.h"
#include "dawn_native/d3d12/ResourceAllocator.h"
#include "dawn_native/d3d12/SamplerD3D12.h"
#include "dawn_native/d3d12/TextureCopySplitter.h"
#include "dawn_native/d3d12/TextureD3D12.h"
#include <deque>
namespace dawn_native { namespace d3d12 {
namespace {
DXGI_FORMAT DXGIIndexFormat(dawn::IndexFormat format) {
switch (format) {
case dawn::IndexFormat::Uint16:
return DXGI_FORMAT_R16_UINT;
case dawn::IndexFormat::Uint32:
return DXGI_FORMAT_R32_UINT;
default:
UNREACHABLE();
}
}
D3D12_TEXTURE_COPY_LOCATION CreateTextureCopyLocationForTexture(const Texture& texture,
uint32_t level,
uint32_t slice) {
D3D12_TEXTURE_COPY_LOCATION copyLocation;
copyLocation.pResource = texture.GetD3D12Resource();
copyLocation.Type = D3D12_TEXTURE_COPY_TYPE_SUBRESOURCE_INDEX;
copyLocation.SubresourceIndex = texture.GetSubresourceIndex(level, slice);
return copyLocation;
}
bool CanUseCopyResource(const uint32_t sourceNumMipLevels,
const Extent3D& srcSize,
const Extent3D& dstSize,
const Extent3D& copySize) {
if (sourceNumMipLevels == 1 && srcSize.width == dstSize.width &&
srcSize.height == dstSize.height && srcSize.depth == dstSize.depth &&
srcSize.width == copySize.width && srcSize.height == copySize.height &&
srcSize.depth == copySize.depth) {
return true;
}
return false;
}
} // anonymous namespace
class BindGroupStateTracker {
public:
BindGroupStateTracker(Device* device) : mDevice(device) {
}
void SetInComputePass(bool inCompute_) {
mInCompute = inCompute_;
}
void AllocateDescriptorHeaps(Device* device) {
// This function should only be called once.
ASSERT(mCbvSrvUavGPUDescriptorHeap.Get() == nullptr &&
mSamplerGPUDescriptorHeap.Get() == nullptr);
DescriptorHeapAllocator* descriptorHeapAllocator = device->GetDescriptorHeapAllocator();
if (mCbvSrvUavDescriptorHeapSize > 0) {
mCbvSrvUavGPUDescriptorHeap = descriptorHeapAllocator->AllocateGPUHeap(
D3D12_DESCRIPTOR_HEAP_TYPE_CBV_SRV_UAV, mCbvSrvUavDescriptorHeapSize);
}
if (mSamplerDescriptorHeapSize > 0) {
mSamplerGPUDescriptorHeap = descriptorHeapAllocator->AllocateGPUHeap(
D3D12_DESCRIPTOR_HEAP_TYPE_SAMPLER, mSamplerDescriptorHeapSize);
}
uint32_t cbvSrvUavDescriptorIndex = 0;
uint32_t samplerDescriptorIndex = 0;
for (BindGroup* group : mBindGroupsList) {
ASSERT(group);
ASSERT(cbvSrvUavDescriptorIndex +
ToBackend(group->GetLayout())->GetCbvUavSrvDescriptorCount() <=
mCbvSrvUavDescriptorHeapSize);
ASSERT(samplerDescriptorIndex +
ToBackend(group->GetLayout())->GetSamplerDescriptorCount() <=
mSamplerDescriptorHeapSize);
group->AllocateDescriptors(mCbvSrvUavGPUDescriptorHeap, &cbvSrvUavDescriptorIndex,
mSamplerGPUDescriptorHeap, &samplerDescriptorIndex);
}
ASSERT(cbvSrvUavDescriptorIndex == mCbvSrvUavDescriptorHeapSize);
ASSERT(samplerDescriptorIndex == mSamplerDescriptorHeapSize);
}
// This function must only be called before calling AllocateDescriptorHeaps().
void TrackSetBindGroup(BindGroup* group, uint32_t index, uint32_t indexInSubmit) {
if (mBindGroups[index] != group) {
mBindGroups[index] = group;
if (!group->TestAndSetCounted(mDevice->GetPendingCommandSerial(), indexInSubmit)) {
const BindGroupLayout* layout = ToBackend(group->GetLayout());
mCbvSrvUavDescriptorHeapSize += layout->GetCbvUavSrvDescriptorCount();
mSamplerDescriptorHeapSize += layout->GetSamplerDescriptorCount();
mBindGroupsList.push_back(group);
}
}
}
// This function must only be called before calling AllocateDescriptorHeaps().
void TrackInheritedGroups(PipelineLayout* oldLayout,
PipelineLayout* newLayout,
uint32_t indexInSubmit) {
if (oldLayout == nullptr) {
return;
}
uint32_t inheritUntil = oldLayout->GroupsInheritUpTo(newLayout);
for (uint32_t i = 0; i < inheritUntil; ++i) {
TrackSetBindGroup(mBindGroups[i], i, indexInSubmit);
}
}
void SetBindGroup(ComPtr<ID3D12GraphicsCommandList> commandList,
PipelineLayout* pipelineLayout,
BindGroup* group,
uint32_t index,
uint32_t dynamicOffsetCount,
uint64_t* dynamicOffsets,
bool force = false) {
if (mBindGroups[index] != group || force) {
mBindGroups[index] = group;
uint32_t currentDynamicBufferIndex = 0;
const BindGroupLayout::LayoutBindingInfo& layout =
group->GetLayout()->GetBindingInfo();
for (uint32_t bindingIndex : IterateBitSet(layout.dynamic)) {
ASSERT(dynamicOffsetCount > 0);
uint32_t parameterIndex =
pipelineLayout->GetDynamicRootParameterIndex(index, bindingIndex);
BufferBinding binding = group->GetBindingAsBufferBinding(bindingIndex);
// Calculate buffer locations that root descriptors links to. The location is
// (base buffer location + initial offset + dynamic offset)
uint64_t offset = dynamicOffsets[currentDynamicBufferIndex++] + binding.offset;
D3D12_GPU_VIRTUAL_ADDRESS bufferLocation =
ToBackend(binding.buffer)->GetVA() + offset;
switch (layout.types[bindingIndex]) {
case dawn::BindingType::UniformBuffer:
if (mInCompute) {
commandList->SetComputeRootConstantBufferView(parameterIndex,
bufferLocation);
} else {
commandList->SetGraphicsRootConstantBufferView(parameterIndex,
bufferLocation);
}
break;
case dawn::BindingType::StorageBuffer:
if (mInCompute) {
commandList->SetComputeRootUnorderedAccessView(parameterIndex,
bufferLocation);
} else {
commandList->SetGraphicsRootUnorderedAccessView(parameterIndex,
bufferLocation);
}
break;
case dawn::BindingType::SampledTexture:
case dawn::BindingType::Sampler:
case dawn::BindingType::StorageTexture:
case dawn::BindingType::ReadonlyStorageBuffer:
UNREACHABLE();
break;
}
}
uint32_t cbvUavSrvCount =
ToBackend(group->GetLayout())->GetCbvUavSrvDescriptorCount();
uint32_t samplerCount = ToBackend(group->GetLayout())->GetSamplerDescriptorCount();
if (cbvUavSrvCount > 0) {
uint32_t parameterIndex = pipelineLayout->GetCbvUavSrvRootParameterIndex(index);
if (mInCompute) {
commandList->SetComputeRootDescriptorTable(
parameterIndex, mCbvSrvUavGPUDescriptorHeap.GetGPUHandle(
group->GetCbvUavSrvHeapOffset()));
} else {
commandList->SetGraphicsRootDescriptorTable(
parameterIndex, mCbvSrvUavGPUDescriptorHeap.GetGPUHandle(
group->GetCbvUavSrvHeapOffset()));
}
}
if (samplerCount > 0) {
uint32_t parameterIndex = pipelineLayout->GetSamplerRootParameterIndex(index);
if (mInCompute) {
commandList->SetComputeRootDescriptorTable(
parameterIndex,
mSamplerGPUDescriptorHeap.GetGPUHandle(group->GetSamplerHeapOffset()));
} else {
commandList->SetGraphicsRootDescriptorTable(
parameterIndex,
mSamplerGPUDescriptorHeap.GetGPUHandle(group->GetSamplerHeapOffset()));
}
}
}
}
void SetInheritedBindGroups(ComPtr<ID3D12GraphicsCommandList> commandList,
PipelineLayout* oldLayout,
PipelineLayout* newLayout) {
if (oldLayout == nullptr) {
return;
}
uint32_t inheritUntil = oldLayout->GroupsInheritUpTo(newLayout);
for (uint32_t i = 0; i < inheritUntil; ++i) {
const BindGroupLayout* layout = ToBackend(mBindGroups[i]->GetLayout());
const uint32_t dynamicBufferCount = layout->GetDynamicBufferCount();
// TODO(shaobo.yan@intel.com) : Need to handle dynamic resources inherited with last
// dynamic offsets.
if (dynamicBufferCount > 0) {
std::vector<uint64_t> zeroOffsets(dynamicBufferCount, 0);
SetBindGroup(commandList, newLayout, mBindGroups[i], i, dynamicBufferCount,
zeroOffsets.data(), true);
} else {
SetBindGroup(commandList, newLayout, mBindGroups[i], i, 0, nullptr, true);
}
}
}
void Reset() {
for (uint32_t i = 0; i < kMaxBindGroups; ++i) {
mBindGroups[i] = nullptr;
}
}
void SetID3D12DescriptorHeaps(ComPtr<ID3D12GraphicsCommandList> commandList) {
ASSERT(commandList != nullptr);
ID3D12DescriptorHeap* descriptorHeaps[2] = {mCbvSrvUavGPUDescriptorHeap.Get(),
mSamplerGPUDescriptorHeap.Get()};
if (descriptorHeaps[0] && descriptorHeaps[1]) {
commandList->SetDescriptorHeaps(2, descriptorHeaps);
} else if (descriptorHeaps[0]) {
commandList->SetDescriptorHeaps(1, descriptorHeaps);
} else if (descriptorHeaps[1]) {
commandList->SetDescriptorHeaps(1, &descriptorHeaps[1]);
}
}
private:
uint32_t mCbvSrvUavDescriptorHeapSize = 0;
uint32_t mSamplerDescriptorHeapSize = 0;
std::array<BindGroup*, kMaxBindGroups> mBindGroups = {};
std::deque<BindGroup*> mBindGroupsList = {};
bool mInCompute = false;
DescriptorHeapHandle mCbvSrvUavGPUDescriptorHeap = {};
DescriptorHeapHandle mSamplerGPUDescriptorHeap = {};
Device* mDevice;
};
struct OMSetRenderTargetArgs {
unsigned int numRTVs = 0;
std::array<D3D12_CPU_DESCRIPTOR_HANDLE, kMaxColorAttachments> RTVs = {};
D3D12_CPU_DESCRIPTOR_HANDLE dsv = {};
};
class RenderPassDescriptorHeapTracker {
public:
RenderPassDescriptorHeapTracker(Device* device) : mDevice(device) {
}
// This function must only be called before calling AllocateRTVAndDSVHeaps().
void TrackRenderPass(const BeginRenderPassCmd* renderPass) {
DAWN_ASSERT(mRTVHeap.Get() == nullptr && mDSVHeap.Get() == nullptr);
mNumRTVs += static_cast<uint32_t>(
renderPass->attachmentState->GetColorAttachmentsMask().count());
if (renderPass->attachmentState->HasDepthStencilAttachment()) {
++mNumDSVs;
}
}
void AllocateRTVAndDSVHeaps() {
// This function should only be called once.
DAWN_ASSERT(mRTVHeap.Get() == nullptr && mDSVHeap.Get() == nullptr);
DescriptorHeapAllocator* allocator = mDevice->GetDescriptorHeapAllocator();
if (mNumRTVs > 0) {
mRTVHeap = allocator->AllocateCPUHeap(D3D12_DESCRIPTOR_HEAP_TYPE_RTV, mNumRTVs);
}
if (mNumDSVs > 0) {
mDSVHeap = allocator->AllocateCPUHeap(D3D12_DESCRIPTOR_HEAP_TYPE_DSV, mNumDSVs);
}
}
// TODO(jiawei.shao@intel.com): use hash map <RenderPass, OMSetRenderTargetArgs> as cache to
// avoid redundant RTV and DSV memory allocations.
OMSetRenderTargetArgs GetSubpassOMSetRenderTargetArgs(BeginRenderPassCmd* renderPass) {
OMSetRenderTargetArgs args = {};
unsigned int rtvIndex = 0;
uint32_t rtvCount = static_cast<uint32_t>(
renderPass->attachmentState->GetColorAttachmentsMask().count());
DAWN_ASSERT(mAllocatedRTVs + rtvCount <= mNumRTVs);
for (uint32_t i :
IterateBitSet(renderPass->attachmentState->GetColorAttachmentsMask())) {
TextureView* view = ToBackend(renderPass->colorAttachments[i].view).Get();
D3D12_CPU_DESCRIPTOR_HANDLE rtvHandle = mRTVHeap.GetCPUHandle(mAllocatedRTVs);
D3D12_RENDER_TARGET_VIEW_DESC rtvDesc = view->GetRTVDescriptor();
mDevice->GetD3D12Device()->CreateRenderTargetView(
ToBackend(view->GetTexture())->GetD3D12Resource(), &rtvDesc, rtvHandle);
args.RTVs[i] = rtvHandle;
++rtvIndex;
++mAllocatedRTVs;
}
args.numRTVs = rtvIndex;
if (renderPass->attachmentState->HasDepthStencilAttachment()) {
DAWN_ASSERT(mAllocatedDSVs < mNumDSVs);
TextureView* view = ToBackend(renderPass->depthStencilAttachment.view).Get();
D3D12_CPU_DESCRIPTOR_HANDLE dsvHandle = mDSVHeap.GetCPUHandle(mAllocatedDSVs);
D3D12_DEPTH_STENCIL_VIEW_DESC dsvDesc = view->GetDSVDescriptor();
mDevice->GetD3D12Device()->CreateDepthStencilView(
ToBackend(view->GetTexture())->GetD3D12Resource(), &dsvDesc, dsvHandle);
args.dsv = dsvHandle;
++mAllocatedDSVs;
}
return args;
}
bool IsHeapAllocationCompleted() const {
return mNumRTVs == mAllocatedRTVs && mNumDSVs == mAllocatedDSVs;
}
private:
Device* mDevice;
DescriptorHeapHandle mRTVHeap = {};
DescriptorHeapHandle mDSVHeap = {};
uint32_t mNumRTVs = 0;
uint32_t mNumDSVs = 0;
uint32_t mAllocatedRTVs = 0;
uint32_t mAllocatedDSVs = 0;
};
namespace {
void AllocateAndSetDescriptorHeaps(Device* device,
BindGroupStateTracker* bindingTracker,
RenderPassDescriptorHeapTracker* renderPassTracker,
CommandIterator* commands,
uint32_t indexInSubmit) {
{
Command type;
PipelineLayout* lastLayout = nullptr;
while (commands->NextCommandId(&type)) {
switch (type) {
case Command::SetComputePipeline: {
SetComputePipelineCmd* cmd =
commands->NextCommand<SetComputePipelineCmd>();
PipelineLayout* layout = ToBackend(cmd->pipeline->GetLayout());
bindingTracker->TrackInheritedGroups(lastLayout, layout, indexInSubmit);
lastLayout = layout;
} break;
case Command::SetRenderPipeline: {
SetRenderPipelineCmd* cmd =
commands->NextCommand<SetRenderPipelineCmd>();
PipelineLayout* layout = ToBackend(cmd->pipeline->GetLayout());
bindingTracker->TrackInheritedGroups(lastLayout, layout, indexInSubmit);
lastLayout = layout;
} break;
case Command::SetBindGroup: {
SetBindGroupCmd* cmd = commands->NextCommand<SetBindGroupCmd>();
BindGroup* group = ToBackend(cmd->group.Get());
if (cmd->dynamicOffsetCount) {
commands->NextData<uint64_t>(cmd->dynamicOffsetCount);
}
bindingTracker->TrackSetBindGroup(group, cmd->index, indexInSubmit);
} break;
case Command::BeginRenderPass: {
BeginRenderPassCmd* cmd = commands->NextCommand<BeginRenderPassCmd>();
renderPassTracker->TrackRenderPass(cmd);
} break;
default:
SkipCommand(commands, type);
}
}
commands->Reset();
}
renderPassTracker->AllocateRTVAndDSVHeaps();
bindingTracker->AllocateDescriptorHeaps(device);
}
void ResolveMultisampledRenderPass(ComPtr<ID3D12GraphicsCommandList> commandList,
BeginRenderPassCmd* renderPass) {
ASSERT(renderPass != nullptr);
for (uint32_t i :
IterateBitSet(renderPass->attachmentState->GetColorAttachmentsMask())) {
TextureViewBase* resolveTarget =
renderPass->colorAttachments[i].resolveTarget.Get();
if (resolveTarget == nullptr) {
continue;
}
Texture* colorTexture =
ToBackend(renderPass->colorAttachments[i].view->GetTexture());
Texture* resolveTexture = ToBackend(resolveTarget->GetTexture());
// Transition the usages of the color attachment and resolve target.
colorTexture->TransitionUsageNow(commandList, D3D12_RESOURCE_STATE_RESOLVE_SOURCE);
resolveTexture->TransitionUsageNow(commandList, D3D12_RESOURCE_STATE_RESOLVE_DEST);
// Do MSAA resolve with ResolveSubResource().
ID3D12Resource* colorTextureHandle = colorTexture->GetD3D12Resource();
ID3D12Resource* resolveTextureHandle = resolveTexture->GetD3D12Resource();
const uint32_t resolveTextureSubresourceIndex = resolveTexture->GetSubresourceIndex(
resolveTarget->GetBaseMipLevel(), resolveTarget->GetBaseArrayLayer());
constexpr uint32_t kColorTextureSubresourceIndex = 0;
commandList->ResolveSubresource(
resolveTextureHandle, resolveTextureSubresourceIndex, colorTextureHandle,
kColorTextureSubresourceIndex, colorTexture->GetD3D12Format());
}
}
} // anonymous namespace
CommandBuffer::CommandBuffer(CommandEncoderBase* encoder,
const CommandBufferDescriptor* descriptor)
: CommandBufferBase(encoder, descriptor), mCommands(encoder->AcquireCommands()) {
}
CommandBuffer::~CommandBuffer() {
FreeCommands(&mCommands);
}
void CommandBuffer::RecordCommands(ComPtr<ID3D12GraphicsCommandList> commandList,
uint32_t indexInSubmit) {
Device* device = ToBackend(GetDevice());
BindGroupStateTracker bindingTracker(device);
RenderPassDescriptorHeapTracker renderPassTracker(device);
// Precompute the allocation of bindgroups in descriptor heaps
// TODO(cwallez@chromium.org): Iterating over all the commands here is inefficient. We
// should have a system where commands and descriptors are recorded in parallel then the
// heaps set using a small CommandList inserted just before the main CommandList.
{
AllocateAndSetDescriptorHeaps(device, &bindingTracker, &renderPassTracker, &mCommands,
indexInSubmit);
bindingTracker.Reset();
bindingTracker.SetID3D12DescriptorHeaps(commandList);
}
// Records the necessary barriers for the resource usage pre-computed by the frontend
auto TransitionForPass = [](ComPtr<ID3D12GraphicsCommandList> commandList,
const PassResourceUsage& usages) {
std::vector<D3D12_RESOURCE_BARRIER> barriers;
for (size_t i = 0; i < usages.buffers.size(); ++i) {
D3D12_RESOURCE_BARRIER barrier;
if (ToBackend(usages.buffers[i])
->TransitionUsageAndGetResourceBarrier(&barrier, usages.bufferUsages[i])) {
barriers.push_back(barrier);
}
}
for (size_t i = 0; i < usages.textures.size(); ++i) {
Texture* texture = ToBackend(usages.textures[i]);
// TODO(natlee@microsoft.com): Update clearing here when subresource tracking is
// implemented
texture->EnsureSubresourceContentInitialized(
commandList, 0, texture->GetNumMipLevels(), 0, texture->GetArrayLayers());
}
for (size_t i = 0; i < usages.textures.size(); ++i) {
D3D12_RESOURCE_BARRIER barrier;
if (ToBackend(usages.textures[i])
->TransitionUsageAndGetResourceBarrier(&barrier, usages.textureUsages[i])) {
barriers.push_back(barrier);
}
}
if (barriers.size()) {
commandList->ResourceBarrier(barriers.size(), barriers.data());
}
};
const std::vector<PassResourceUsage>& passResourceUsages = GetResourceUsages().perPass;
uint32_t nextPassNumber = 0;
Command type;
while (mCommands.NextCommandId(&type)) {
switch (type) {
case Command::BeginComputePass: {
mCommands.NextCommand<BeginComputePassCmd>();
TransitionForPass(commandList, passResourceUsages[nextPassNumber]);
bindingTracker.SetInComputePass(true);
RecordComputePass(commandList, &bindingTracker);
nextPassNumber++;
} break;
case Command::BeginRenderPass: {
BeginRenderPassCmd* beginRenderPassCmd =
mCommands.NextCommand<BeginRenderPassCmd>();
TransitionForPass(commandList, passResourceUsages[nextPassNumber]);
bindingTracker.SetInComputePass(false);
RecordRenderPass(commandList, &bindingTracker, &renderPassTracker,
beginRenderPassCmd);
nextPassNumber++;
} break;
case Command::CopyBufferToBuffer: {
CopyBufferToBufferCmd* copy = mCommands.NextCommand<CopyBufferToBufferCmd>();
Buffer* srcBuffer = ToBackend(copy->source.Get());
Buffer* dstBuffer = ToBackend(copy->destination.Get());
srcBuffer->TransitionUsageNow(commandList, dawn::BufferUsageBit::CopySrc);
dstBuffer->TransitionUsageNow(commandList, dawn::BufferUsageBit::CopyDst);
commandList->CopyBufferRegion(
dstBuffer->GetD3D12Resource().Get(), copy->destinationOffset,
srcBuffer->GetD3D12Resource().Get(), copy->sourceOffset, copy->size);
} break;
case Command::CopyBufferToTexture: {
CopyBufferToTextureCmd* copy = mCommands.NextCommand<CopyBufferToTextureCmd>();
Buffer* buffer = ToBackend(copy->source.buffer.Get());
Texture* texture = ToBackend(copy->destination.texture.Get());
if (IsCompleteSubresourceCopiedTo(texture, copy->copySize,
copy->destination.mipLevel)) {
texture->SetIsSubresourceContentInitialized(
copy->destination.mipLevel, 1, copy->destination.arrayLayer, 1);
} else {
texture->EnsureSubresourceContentInitialized(
commandList, copy->destination.mipLevel, 1,
copy->destination.arrayLayer, 1);
}
buffer->TransitionUsageNow(commandList, dawn::BufferUsageBit::CopySrc);
texture->TransitionUsageNow(commandList, dawn::TextureUsageBit::CopyDst);
auto copySplit = ComputeTextureCopySplit(
copy->destination.origin, copy->copySize, texture->GetFormat(),
copy->source.offset, copy->source.rowPitch, copy->source.imageHeight);
D3D12_TEXTURE_COPY_LOCATION textureLocation =
CreateTextureCopyLocationForTexture(*texture, copy->destination.mipLevel,
copy->destination.arrayLayer);
for (uint32_t i = 0; i < copySplit.count; ++i) {
auto& info = copySplit.copies[i];
D3D12_TEXTURE_COPY_LOCATION bufferLocation;
bufferLocation.pResource = buffer->GetD3D12Resource().Get();
bufferLocation.Type = D3D12_TEXTURE_COPY_TYPE_PLACED_FOOTPRINT;
bufferLocation.PlacedFootprint.Offset = copySplit.offset;
bufferLocation.PlacedFootprint.Footprint.Format = texture->GetD3D12Format();
bufferLocation.PlacedFootprint.Footprint.Width = info.bufferSize.width;
bufferLocation.PlacedFootprint.Footprint.Height = info.bufferSize.height;
bufferLocation.PlacedFootprint.Footprint.Depth = info.bufferSize.depth;
bufferLocation.PlacedFootprint.Footprint.RowPitch = copy->source.rowPitch;
D3D12_BOX sourceRegion;
sourceRegion.left = info.bufferOffset.x;
sourceRegion.top = info.bufferOffset.y;
sourceRegion.front = info.bufferOffset.z;
sourceRegion.right = info.bufferOffset.x + info.copySize.width;
sourceRegion.bottom = info.bufferOffset.y + info.copySize.height;
sourceRegion.back = info.bufferOffset.z + info.copySize.depth;
commandList->CopyTextureRegion(&textureLocation, info.textureOffset.x,
info.textureOffset.y, info.textureOffset.z,
&bufferLocation, &sourceRegion);
}
} break;
case Command::CopyTextureToBuffer: {
CopyTextureToBufferCmd* copy = mCommands.NextCommand<CopyTextureToBufferCmd>();
Texture* texture = ToBackend(copy->source.texture.Get());
Buffer* buffer = ToBackend(copy->destination.buffer.Get());
texture->EnsureSubresourceContentInitialized(commandList, copy->source.mipLevel,
1, copy->source.arrayLayer, 1);
texture->TransitionUsageNow(commandList, dawn::TextureUsageBit::CopySrc);
buffer->TransitionUsageNow(commandList, dawn::BufferUsageBit::CopyDst);
auto copySplit = ComputeTextureCopySplit(
copy->source.origin, copy->copySize, texture->GetFormat(),
copy->destination.offset, copy->destination.rowPitch,
copy->destination.imageHeight);
D3D12_TEXTURE_COPY_LOCATION textureLocation =
CreateTextureCopyLocationForTexture(*texture, copy->source.mipLevel,
copy->source.arrayLayer);
for (uint32_t i = 0; i < copySplit.count; ++i) {
auto& info = copySplit.copies[i];
D3D12_TEXTURE_COPY_LOCATION bufferLocation;
bufferLocation.pResource = buffer->GetD3D12Resource().Get();
bufferLocation.Type = D3D12_TEXTURE_COPY_TYPE_PLACED_FOOTPRINT;
bufferLocation.PlacedFootprint.Offset = copySplit.offset;
bufferLocation.PlacedFootprint.Footprint.Format = texture->GetD3D12Format();
bufferLocation.PlacedFootprint.Footprint.Width = info.bufferSize.width;
bufferLocation.PlacedFootprint.Footprint.Height = info.bufferSize.height;
bufferLocation.PlacedFootprint.Footprint.Depth = info.bufferSize.depth;
bufferLocation.PlacedFootprint.Footprint.RowPitch =
copy->destination.rowPitch;
D3D12_BOX sourceRegion;
sourceRegion.left = info.textureOffset.x;
sourceRegion.top = info.textureOffset.y;
sourceRegion.front = info.textureOffset.z;
sourceRegion.right = info.textureOffset.x + info.copySize.width;
sourceRegion.bottom = info.textureOffset.y + info.copySize.height;
sourceRegion.back = info.textureOffset.z + info.copySize.depth;
commandList->CopyTextureRegion(&bufferLocation, info.bufferOffset.x,
info.bufferOffset.y, info.bufferOffset.z,
&textureLocation, &sourceRegion);
}
} break;
case Command::CopyTextureToTexture: {
CopyTextureToTextureCmd* copy =
mCommands.NextCommand<CopyTextureToTextureCmd>();
Texture* source = ToBackend(copy->source.texture.Get());
Texture* destination = ToBackend(copy->destination.texture.Get());
source->EnsureSubresourceContentInitialized(commandList, copy->source.mipLevel,
1, copy->source.arrayLayer, 1);
if (IsCompleteSubresourceCopiedTo(destination, copy->copySize,
copy->destination.mipLevel)) {
destination->SetIsSubresourceContentInitialized(
copy->destination.mipLevel, 1, copy->destination.arrayLayer, 1);
} else {
destination->EnsureSubresourceContentInitialized(
commandList, copy->destination.mipLevel, 1,
copy->destination.arrayLayer, 1);
}
source->TransitionUsageNow(commandList, dawn::TextureUsageBit::CopySrc);
destination->TransitionUsageNow(commandList, dawn::TextureUsageBit::CopyDst);
if (CanUseCopyResource(source->GetNumMipLevels(), source->GetSize(),
destination->GetSize(), copy->copySize)) {
commandList->CopyResource(destination->GetD3D12Resource(),
source->GetD3D12Resource());
} else {
D3D12_TEXTURE_COPY_LOCATION srcLocation =
CreateTextureCopyLocationForTexture(*source, copy->source.mipLevel,
copy->source.arrayLayer);
D3D12_TEXTURE_COPY_LOCATION dstLocation =
CreateTextureCopyLocationForTexture(*destination,
copy->destination.mipLevel,
copy->destination.arrayLayer);
D3D12_BOX sourceRegion;
sourceRegion.left = copy->source.origin.x;
sourceRegion.top = copy->source.origin.y;
sourceRegion.front = copy->source.origin.z;
sourceRegion.right = copy->source.origin.x + copy->copySize.width;
sourceRegion.bottom = copy->source.origin.y + copy->copySize.height;
sourceRegion.back = copy->source.origin.z + copy->copySize.depth;
commandList->CopyTextureRegion(
&dstLocation, copy->destination.origin.x, copy->destination.origin.y,
copy->destination.origin.z, &srcLocation, &sourceRegion);
}
} break;
default: { UNREACHABLE(); } break;
}
}
DAWN_ASSERT(renderPassTracker.IsHeapAllocationCompleted());
}
void CommandBuffer::FlushSetVertexBuffers(ComPtr<ID3D12GraphicsCommandList> commandList,
VertexBuffersInfo* vertexBuffersInfo,
const RenderPipeline* renderPipeline) {
DAWN_ASSERT(vertexBuffersInfo != nullptr);
DAWN_ASSERT(renderPipeline != nullptr);
auto inputsMask = renderPipeline->GetInputsSetMask();
uint32_t startSlot = vertexBuffersInfo->startSlot;
uint32_t endSlot = vertexBuffersInfo->endSlot;
// If the input state has changed, we need to update the StrideInBytes
// for the D3D12 buffer views. We also need to extend the dirty range to
// touch all these slots because the stride may have changed.
if (vertexBuffersInfo->lastRenderPipeline != renderPipeline) {
vertexBuffersInfo->lastRenderPipeline = renderPipeline;
for (uint32_t slot : IterateBitSet(inputsMask)) {
startSlot = std::min(startSlot, slot);
endSlot = std::max(endSlot, slot + 1);
vertexBuffersInfo->d3d12BufferViews[slot].StrideInBytes =
renderPipeline->GetInput(slot).stride;
}
}
if (endSlot <= startSlot) {
return;
}
// d3d12BufferViews is kept up to date with the most recent data passed
// to SetVertexBuffers. This makes it correct to only track the start
// and end of the dirty range. When FlushSetVertexBuffers is called,
// we will at worst set non-dirty vertex buffers in duplicate.
uint32_t count = endSlot - startSlot;
commandList->IASetVertexBuffers(startSlot, count,
&vertexBuffersInfo->d3d12BufferViews[startSlot]);
vertexBuffersInfo->startSlot = kMaxVertexBuffers;
vertexBuffersInfo->endSlot = 0;
}
void CommandBuffer::RecordComputePass(ComPtr<ID3D12GraphicsCommandList> commandList,
BindGroupStateTracker* bindingTracker) {
PipelineLayout* lastLayout = nullptr;
Command type;
while (mCommands.NextCommandId(&type)) {
switch (type) {
case Command::Dispatch: {
DispatchCmd* dispatch = mCommands.NextCommand<DispatchCmd>();
commandList->Dispatch(dispatch->x, dispatch->y, dispatch->z);
} break;
case Command::DispatchIndirect: {
DispatchIndirectCmd* dispatch = mCommands.NextCommand<DispatchIndirectCmd>();
Buffer* buffer = ToBackend(dispatch->indirectBuffer.Get());
ComPtr<ID3D12CommandSignature> signature =
ToBackend(GetDevice())->GetDispatchIndirectSignature();
commandList->ExecuteIndirect(signature.Get(), 1,
buffer->GetD3D12Resource().Get(),
dispatch->indirectOffset, nullptr, 0);
} break;
case Command::EndComputePass: {
mCommands.NextCommand<EndComputePassCmd>();
return;
} break;
case Command::SetComputePipeline: {
SetComputePipelineCmd* cmd = mCommands.NextCommand<SetComputePipelineCmd>();
ComputePipeline* pipeline = ToBackend(cmd->pipeline).Get();
PipelineLayout* layout = ToBackend(pipeline->GetLayout());
commandList->SetComputeRootSignature(layout->GetRootSignature().Get());
commandList->SetPipelineState(pipeline->GetPipelineState().Get());
bindingTracker->SetInheritedBindGroups(commandList, lastLayout, layout);
lastLayout = layout;
} break;
case Command::SetBindGroup: {
SetBindGroupCmd* cmd = mCommands.NextCommand<SetBindGroupCmd>();
BindGroup* group = ToBackend(cmd->group.Get());
uint64_t* dynamicOffsets = nullptr;
if (cmd->dynamicOffsetCount > 0) {
dynamicOffsets = mCommands.NextData<uint64_t>(cmd->dynamicOffsetCount);
}
bindingTracker->SetBindGroup(commandList, lastLayout, group, cmd->index,
cmd->dynamicOffsetCount, dynamicOffsets);
} break;
case Command::InsertDebugMarker: {
InsertDebugMarkerCmd* cmd = mCommands.NextCommand<InsertDebugMarkerCmd>();
const char* label = mCommands.NextData<char>(cmd->length + 1);
if (ToBackend(GetDevice())->GetFunctions()->IsPIXEventRuntimeLoaded()) {
// PIX color is 1 byte per channel in ARGB format
constexpr uint64_t kPIXBlackColor = 0xff000000;
ToBackend(GetDevice())
->GetFunctions()
->pixSetMarkerOnCommandList(commandList.Get(), kPIXBlackColor, label);
}
} break;
case Command::PopDebugGroup: {
mCommands.NextCommand<PopDebugGroupCmd>();
if (ToBackend(GetDevice())->GetFunctions()->IsPIXEventRuntimeLoaded()) {
ToBackend(GetDevice())
->GetFunctions()
->pixEndEventOnCommandList(commandList.Get());
}
} break;
case Command::PushDebugGroup: {
PushDebugGroupCmd* cmd = mCommands.NextCommand<PushDebugGroupCmd>();
const char* label = mCommands.NextData<char>(cmd->length + 1);
if (ToBackend(GetDevice())->GetFunctions()->IsPIXEventRuntimeLoaded()) {
// PIX color is 1 byte per channel in ARGB format
constexpr uint64_t kPIXBlackColor = 0xff000000;
ToBackend(GetDevice())
->GetFunctions()
->pixBeginEventOnCommandList(commandList.Get(), kPIXBlackColor, label);
}
} break;
default: { UNREACHABLE(); } break;
}
}
}
void CommandBuffer::RecordRenderPass(ComPtr<ID3D12GraphicsCommandList> commandList,
BindGroupStateTracker* bindingTracker,
RenderPassDescriptorHeapTracker* renderPassTracker,
BeginRenderPassCmd* renderPass) {
OMSetRenderTargetArgs args = renderPassTracker->GetSubpassOMSetRenderTargetArgs(renderPass);
// Clear framebuffer attachments as needed and transition to render target
{
for (uint32_t i :
IterateBitSet(renderPass->attachmentState->GetColorAttachmentsMask())) {
auto& attachmentInfo = renderPass->colorAttachments[i];
TextureView* view = ToBackend(attachmentInfo.view.Get());
// Load op - color
ASSERT(view->GetLevelCount() == 1);
ASSERT(view->GetLayerCount() == 1);
if (attachmentInfo.loadOp == dawn::LoadOp::Clear) {
D3D12_CPU_DESCRIPTOR_HANDLE handle = args.RTVs[i];
commandList->ClearRenderTargetView(handle, &attachmentInfo.clearColor.r, 0,
nullptr);
} else if (attachmentInfo.loadOp == dawn::LoadOp::Load && view->GetTexture()) {
ToBackend(view->GetTexture())
->EnsureSubresourceContentInitialized(commandList, view->GetBaseMipLevel(),
1, view->GetBaseArrayLayer(), 1);
}
switch (attachmentInfo.storeOp) {
case dawn::StoreOp::Store: {
view->GetTexture()->SetIsSubresourceContentInitialized(
view->GetBaseMipLevel(), 1, view->GetBaseArrayLayer(), 1);
} break;
default: { UNREACHABLE(); } break;
}
}
if (renderPass->attachmentState->HasDepthStencilAttachment()) {
auto& attachmentInfo = renderPass->depthStencilAttachment;
Texture* texture = ToBackend(renderPass->depthStencilAttachment.view->GetTexture());
if ((texture->GetFormat().HasDepth() &&
attachmentInfo.depthLoadOp == dawn::LoadOp::Load) ||
(texture->GetFormat().HasStencil() &&
attachmentInfo.stencilLoadOp == dawn::LoadOp::Load)) {
texture->EnsureSubresourceContentInitialized(
commandList, attachmentInfo.view->GetBaseMipLevel(),
attachmentInfo.view->GetLevelCount(),
attachmentInfo.view->GetBaseArrayLayer(),
attachmentInfo.view->GetLayerCount());
}
// Load op - depth/stencil
bool doDepthClear = texture->GetFormat().HasDepth() &&
(attachmentInfo.depthLoadOp == dawn::LoadOp::Clear);
bool doStencilClear = texture->GetFormat().HasStencil() &&
(attachmentInfo.stencilLoadOp == dawn::LoadOp::Clear);
D3D12_CLEAR_FLAGS clearFlags = {};
if (doDepthClear) {
clearFlags |= D3D12_CLEAR_FLAG_DEPTH;
}
if (doStencilClear) {
clearFlags |= D3D12_CLEAR_FLAG_STENCIL;
}
if (clearFlags) {
D3D12_CPU_DESCRIPTOR_HANDLE handle = args.dsv;
// TODO(kainino@chromium.org): investigate: should the Dawn clear
// stencil type be uint8_t?
uint8_t clearStencil = static_cast<uint8_t>(attachmentInfo.clearStencil);
commandList->ClearDepthStencilView(
handle, clearFlags, attachmentInfo.clearDepth, clearStencil, 0, nullptr);
texture->SetIsSubresourceContentInitialized(
attachmentInfo.view->GetBaseMipLevel(),
attachmentInfo.view->GetLevelCount(),
attachmentInfo.view->GetBaseArrayLayer(),
attachmentInfo.view->GetLayerCount());
}
}
}
// Set up render targets
{
if (args.dsv.ptr) {
commandList->OMSetRenderTargets(args.numRTVs, args.RTVs.data(), FALSE, &args.dsv);
} else {
commandList->OMSetRenderTargets(args.numRTVs, args.RTVs.data(), FALSE, nullptr);
}
}
// Set up default dynamic state
{
uint32_t width = renderPass->width;
uint32_t height = renderPass->height;
D3D12_VIEWPORT viewport = {
0.f, 0.f, static_cast<float>(width), static_cast<float>(height), 0.f, 1.f};
D3D12_RECT scissorRect = {0, 0, static_cast<long>(width), static_cast<long>(height)};
commandList->RSSetViewports(1, &viewport);
commandList->RSSetScissorRects(1, &scissorRect);
static constexpr std::array<float, 4> defaultBlendFactor = {0, 0, 0, 0};
commandList->OMSetBlendFactor(&defaultBlendFactor[0]);
}
RenderPipeline* lastPipeline = nullptr;
PipelineLayout* lastLayout = nullptr;
VertexBuffersInfo vertexBuffersInfo = {};
Command type;
while (mCommands.NextCommandId(&type)) {
switch (type) {
case Command::EndRenderPass: {
mCommands.NextCommand<EndRenderPassCmd>();
// TODO(brandon1.jones@intel.com): avoid calling this function and enable MSAA
// resolve in D3D12 render pass on the platforms that support this feature.
if (renderPass->attachmentState->GetSampleCount() > 1) {
ResolveMultisampledRenderPass(commandList, renderPass);
}
return;
} break;
case Command::Draw: {
DrawCmd* draw = mCommands.NextCommand<DrawCmd>();
FlushSetVertexBuffers(commandList, &vertexBuffersInfo, lastPipeline);
commandList->DrawInstanced(draw->vertexCount, draw->instanceCount,
draw->firstVertex, draw->firstInstance);
} break;
case Command::DrawIndexed: {
DrawIndexedCmd* draw = mCommands.NextCommand<DrawIndexedCmd>();
FlushSetVertexBuffers(commandList, &vertexBuffersInfo, lastPipeline);
commandList->DrawIndexedInstanced(draw->indexCount, draw->instanceCount,
draw->firstIndex, draw->baseVertex,
draw->firstInstance);
} break;
case Command::DrawIndirect: {
DrawIndirectCmd* draw = mCommands.NextCommand<DrawIndirectCmd>();
FlushSetVertexBuffers(commandList, &vertexBuffersInfo, lastPipeline);
Buffer* buffer = ToBackend(draw->indirectBuffer.Get());
ComPtr<ID3D12CommandSignature> signature =
ToBackend(GetDevice())->GetDrawIndirectSignature();
commandList->ExecuteIndirect(signature.Get(), 1,
buffer->GetD3D12Resource().Get(),
draw->indirectOffset, nullptr, 0);
} break;
case Command::DrawIndexedIndirect: {
DrawIndexedIndirectCmd* draw = mCommands.NextCommand<DrawIndexedIndirectCmd>();
FlushSetVertexBuffers(commandList, &vertexBuffersInfo, lastPipeline);
Buffer* buffer = ToBackend(draw->indirectBuffer.Get());
ComPtr<ID3D12CommandSignature> signature =
ToBackend(GetDevice())->GetDrawIndexedIndirectSignature();
commandList->ExecuteIndirect(signature.Get(), 1,
buffer->GetD3D12Resource().Get(),
draw->indirectOffset, nullptr, 0);
} break;
case Command::InsertDebugMarker: {
InsertDebugMarkerCmd* cmd = mCommands.NextCommand<InsertDebugMarkerCmd>();
const char* label = mCommands.NextData<char>(cmd->length + 1);
if (ToBackend(GetDevice())->GetFunctions()->IsPIXEventRuntimeLoaded()) {
// PIX color is 1 byte per channel in ARGB format
constexpr uint64_t kPIXBlackColor = 0xff000000;
ToBackend(GetDevice())
->GetFunctions()
->pixSetMarkerOnCommandList(commandList.Get(), kPIXBlackColor, label);
}
} break;
case Command::PopDebugGroup: {
mCommands.NextCommand<PopDebugGroupCmd>();
if (ToBackend(GetDevice())->GetFunctions()->IsPIXEventRuntimeLoaded()) {
ToBackend(GetDevice())
->GetFunctions()
->pixEndEventOnCommandList(commandList.Get());
}
} break;
case Command::PushDebugGroup: {
PushDebugGroupCmd* cmd = mCommands.NextCommand<PushDebugGroupCmd>();
const char* label = mCommands.NextData<char>(cmd->length + 1);
if (ToBackend(GetDevice())->GetFunctions()->IsPIXEventRuntimeLoaded()) {
// PIX color is 1 byte per channel in ARGB format
constexpr uint64_t kPIXBlackColor = 0xff000000;
ToBackend(GetDevice())
->GetFunctions()
->pixBeginEventOnCommandList(commandList.Get(), kPIXBlackColor, label);
}
} break;
case Command::SetRenderPipeline: {
SetRenderPipelineCmd* cmd = mCommands.NextCommand<SetRenderPipelineCmd>();
RenderPipeline* pipeline = ToBackend(cmd->pipeline).Get();
PipelineLayout* layout = ToBackend(pipeline->GetLayout());
commandList->SetGraphicsRootSignature(layout->GetRootSignature().Get());
commandList->SetPipelineState(pipeline->GetPipelineState().Get());
commandList->IASetPrimitiveTopology(pipeline->GetD3D12PrimitiveTopology());
bindingTracker->SetInheritedBindGroups(commandList, lastLayout, layout);
lastPipeline = pipeline;
lastLayout = layout;
} break;
case Command::SetStencilReference: {
SetStencilReferenceCmd* cmd = mCommands.NextCommand<SetStencilReferenceCmd>();
commandList->OMSetStencilRef(cmd->reference);
} break;
case Command::SetViewport: {
SetViewportCmd* cmd = mCommands.NextCommand<SetViewportCmd>();
D3D12_VIEWPORT viewport;
viewport.TopLeftX = cmd->x;
viewport.TopLeftY = cmd->y;
viewport.Width = cmd->width;
viewport.Height = cmd->height;
viewport.MinDepth = cmd->minDepth;
viewport.MaxDepth = cmd->maxDepth;
commandList->RSSetViewports(1, &viewport);
} break;
case Command::SetScissorRect: {
SetScissorRectCmd* cmd = mCommands.NextCommand<SetScissorRectCmd>();
D3D12_RECT rect;
rect.left = cmd->x;
rect.top = cmd->y;
rect.right = cmd->x + cmd->width;
rect.bottom = cmd->y + cmd->height;
commandList->RSSetScissorRects(1, &rect);
} break;
case Command::SetBlendColor: {
SetBlendColorCmd* cmd = mCommands.NextCommand<SetBlendColorCmd>();
commandList->OMSetBlendFactor(static_cast<const FLOAT*>(&cmd->color.r));
} break;
case Command::SetBindGroup: {
SetBindGroupCmd* cmd = mCommands.NextCommand<SetBindGroupCmd>();
BindGroup* group = ToBackend(cmd->group.Get());
uint64_t* dynamicOffsets = nullptr;
if (cmd->dynamicOffsetCount > 0) {
dynamicOffsets = mCommands.NextData<uint64_t>(cmd->dynamicOffsetCount);
}
bindingTracker->SetBindGroup(commandList, lastLayout, group, cmd->index,
cmd->dynamicOffsetCount, dynamicOffsets);
} break;
case Command::SetIndexBuffer: {
SetIndexBufferCmd* cmd = mCommands.NextCommand<SetIndexBufferCmd>();
Buffer* buffer = ToBackend(cmd->buffer.Get());
D3D12_INDEX_BUFFER_VIEW bufferView;
bufferView.BufferLocation = buffer->GetVA() + cmd->offset;
bufferView.SizeInBytes = buffer->GetSize() - cmd->offset;
// TODO(cwallez@chromium.org): Make index buffers lazily applied, right now
// this will break if the pipeline is changed for one with a different index
// format after SetIndexBuffer
bufferView.Format =
DXGIIndexFormat(lastPipeline->GetVertexInputDescriptor()->indexFormat);
commandList->IASetIndexBuffer(&bufferView);
} break;
case Command::SetVertexBuffers: {
SetVertexBuffersCmd* cmd = mCommands.NextCommand<SetVertexBuffersCmd>();
auto buffers = mCommands.NextData<Ref<BufferBase>>(cmd->count);
auto offsets = mCommands.NextData<uint64_t>(cmd->count);
vertexBuffersInfo.startSlot =
std::min(vertexBuffersInfo.startSlot, cmd->startSlot);
vertexBuffersInfo.endSlot =
std::max(vertexBuffersInfo.endSlot, cmd->startSlot + cmd->count);
for (uint32_t i = 0; i < cmd->count; ++i) {
Buffer* buffer = ToBackend(buffers[i].Get());
auto* d3d12BufferView =
&vertexBuffersInfo.d3d12BufferViews[cmd->startSlot + i];
d3d12BufferView->BufferLocation = buffer->GetVA() + offsets[i];
d3d12BufferView->SizeInBytes = buffer->GetSize() - offsets[i];
// The bufferView stride is set based on the input state before a draw.
}
} break;
default: { UNREACHABLE(); } break;
}
}
}
}} // namespace dawn_native::d3d12