blob: ddfc2e9376f278043dd574ae092d09ca109266d0 [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/BindGroupLayout.h"
#include "common/BitSetIterator.h"
#include "dawn_native/ChainUtils_autogen.h"
#include "dawn_native/Device.h"
#include "dawn_native/ObjectContentHasher.h"
#include "dawn_native/PerStage.h"
#include "dawn_native/ValidationUtils_autogen.h"
#include <algorithm>
#include <functional>
#include <set>
namespace dawn_native {
namespace {
MaybeError ValidateStorageTextureFormat(DeviceBase* device,
wgpu::TextureFormat storageTextureFormat) {
const Format* format = nullptr;
DAWN_TRY_ASSIGN(format, device->GetInternalFormat(storageTextureFormat));
ASSERT(format != nullptr);
if (!format->supportsStorageUsage) {
return DAWN_VALIDATION_ERROR("Texture format does not support storage textures");
}
return {};
}
MaybeError ValidateStorageTextureViewDimension(wgpu::TextureViewDimension dimension) {
switch (dimension) {
case wgpu::TextureViewDimension::Cube:
case wgpu::TextureViewDimension::CubeArray:
return DAWN_VALIDATION_ERROR(
"Cube map and cube map texture views cannot be used as storage textures");
case wgpu::TextureViewDimension::e1D:
case wgpu::TextureViewDimension::e2D:
case wgpu::TextureViewDimension::e2DArray:
case wgpu::TextureViewDimension::e3D:
return {};
case wgpu::TextureViewDimension::Undefined:
UNREACHABLE();
}
}
} // anonymous namespace
MaybeError ValidateBindGroupLayoutDescriptor(DeviceBase* device,
const BindGroupLayoutDescriptor* descriptor,
bool allowInternalBinding) {
if (descriptor->nextInChain != nullptr) {
return DAWN_VALIDATION_ERROR("nextInChain must be nullptr");
}
std::set<BindingNumber> bindingsSet;
BindingCounts bindingCounts = {};
for (uint32_t i = 0; i < descriptor->entryCount; ++i) {
const BindGroupLayoutEntry& entry = descriptor->entries[i];
BindingNumber bindingNumber = BindingNumber(entry.binding);
if (bindingsSet.count(bindingNumber) != 0) {
return DAWN_VALIDATION_ERROR("some binding index was specified more than once");
}
DAWN_TRY(ValidateShaderStage(entry.visibility));
int bindingMemberCount = 0;
wgpu::ShaderStage allowedStages = kAllStages;
if (entry.buffer.type != wgpu::BufferBindingType::Undefined) {
bindingMemberCount++;
const BufferBindingLayout& buffer = entry.buffer;
// The kInternalStorageBufferBinding is used internally and not a value
// in wgpu::BufferBindingType.
if (buffer.type == kInternalStorageBufferBinding) {
if (!allowInternalBinding) {
return DAWN_VALIDATION_ERROR("Internal binding types are disallowed");
}
} else {
DAWN_TRY(ValidateBufferBindingType(buffer.type));
}
if (buffer.type == wgpu::BufferBindingType::Storage ||
buffer.type == kInternalStorageBufferBinding) {
allowedStages &= ~wgpu::ShaderStage::Vertex;
}
// Dynamic storage buffers aren't bounds checked properly in D3D12. Disallow them as
// unsafe until the bounds checks are implemented.
if (device->IsToggleEnabled(Toggle::DisallowUnsafeAPIs) &&
buffer.hasDynamicOffset &&
(buffer.type == wgpu::BufferBindingType::Storage ||
buffer.type == kInternalStorageBufferBinding ||
buffer.type == wgpu::BufferBindingType::ReadOnlyStorage)) {
return DAWN_VALIDATION_ERROR(
"Dynamic storage buffers are disallowed because they aren't secure yet. "
"See https://crbug.com/dawn/429");
}
}
if (entry.sampler.type != wgpu::SamplerBindingType::Undefined) {
bindingMemberCount++;
DAWN_TRY(ValidateSamplerBindingType(entry.sampler.type));
}
if (entry.texture.sampleType != wgpu::TextureSampleType::Undefined) {
bindingMemberCount++;
const TextureBindingLayout& texture = entry.texture;
DAWN_TRY(ValidateTextureSampleType(texture.sampleType));
// viewDimension defaults to 2D if left undefined, needs validation otherwise.
wgpu::TextureViewDimension viewDimension = wgpu::TextureViewDimension::e2D;
if (texture.viewDimension != wgpu::TextureViewDimension::Undefined) {
DAWN_TRY(ValidateTextureViewDimension(texture.viewDimension));
viewDimension = texture.viewDimension;
}
if (texture.multisampled) {
if (viewDimension != wgpu::TextureViewDimension::e2D) {
return DAWN_VALIDATION_ERROR("Multisampled texture bindings must be 2D.");
}
// TODO: This check should eventually become obsolete. According to the spec,
// depth can be used with both regular and comparison sampling. As such, during
// pipeline creation we have to check that if a comparison sampler is used
// with a texture, that texture must be both depth and not multisampled.
if (texture.sampleType == wgpu::TextureSampleType::Depth) {
return DAWN_VALIDATION_ERROR(
"Multisampled texture bindings must not be Depth.");
}
}
}
if (entry.storageTexture.access != wgpu::StorageTextureAccess::Undefined) {
bindingMemberCount++;
const StorageTextureBindingLayout& storageTexture = entry.storageTexture;
DAWN_TRY(ValidateStorageTextureAccess(storageTexture.access));
DAWN_TRY(ValidateStorageTextureFormat(device, storageTexture.format));
// viewDimension defaults to 2D if left undefined, needs validation otherwise.
if (storageTexture.viewDimension != wgpu::TextureViewDimension::Undefined) {
DAWN_TRY(ValidateTextureViewDimension(storageTexture.viewDimension));
DAWN_TRY(ValidateStorageTextureViewDimension(storageTexture.viewDimension));
}
if (storageTexture.access == wgpu::StorageTextureAccess::WriteOnly) {
allowedStages &= ~wgpu::ShaderStage::Vertex;
}
}
const ExternalTextureBindingLayout* externalTextureBindingLayout = nullptr;
FindInChain(entry.nextInChain, &externalTextureBindingLayout);
if (externalTextureBindingLayout != nullptr) {
bindingMemberCount++;
}
if (bindingMemberCount != 1) {
return DAWN_VALIDATION_ERROR(
"Exactly one of buffer, sampler, texture, storageTexture, or externalTexture "
"must be set for each BindGroupLayoutEntry");
}
if (!IsSubset(entry.visibility, allowedStages)) {
return DAWN_VALIDATION_ERROR("Binding type cannot be used with this visibility.");
}
IncrementBindingCounts(&bindingCounts, entry);
bindingsSet.insert(bindingNumber);
}
DAWN_TRY(ValidateBindingCounts(bindingCounts));
return {};
}
namespace {
bool operator!=(const BindingInfo& a, const BindingInfo& b) {
if (a.visibility != b.visibility || a.bindingType != b.bindingType) {
return true;
}
switch (a.bindingType) {
case BindingInfoType::Buffer:
return a.buffer.type != b.buffer.type ||
a.buffer.hasDynamicOffset != b.buffer.hasDynamicOffset ||
a.buffer.minBindingSize != b.buffer.minBindingSize;
case BindingInfoType::Sampler:
return a.sampler.type != b.sampler.type;
case BindingInfoType::Texture:
return a.texture.sampleType != b.texture.sampleType ||
a.texture.viewDimension != b.texture.viewDimension ||
a.texture.multisampled != b.texture.multisampled;
case BindingInfoType::StorageTexture:
return a.storageTexture.access != b.storageTexture.access ||
a.storageTexture.viewDimension != b.storageTexture.viewDimension ||
a.storageTexture.format != b.storageTexture.format;
case BindingInfoType::ExternalTexture:
return false;
}
}
bool IsBufferBinding(const BindGroupLayoutEntry& binding) {
return binding.buffer.type != wgpu::BufferBindingType::Undefined;
}
bool BindingHasDynamicOffset(const BindGroupLayoutEntry& binding) {
if (binding.buffer.type != wgpu::BufferBindingType::Undefined) {
return binding.buffer.hasDynamicOffset;
}
return false;
}
BindingInfo CreateBindGroupLayoutInfo(const BindGroupLayoutEntry& binding) {
BindingInfo bindingInfo;
bindingInfo.binding = BindingNumber(binding.binding);
bindingInfo.visibility = binding.visibility;
if (binding.buffer.type != wgpu::BufferBindingType::Undefined) {
bindingInfo.bindingType = BindingInfoType::Buffer;
bindingInfo.buffer = binding.buffer;
} else if (binding.sampler.type != wgpu::SamplerBindingType::Undefined) {
bindingInfo.bindingType = BindingInfoType::Sampler;
bindingInfo.sampler = binding.sampler;
} else if (binding.texture.sampleType != wgpu::TextureSampleType::Undefined) {
bindingInfo.bindingType = BindingInfoType::Texture;
bindingInfo.texture = binding.texture;
if (binding.texture.viewDimension == wgpu::TextureViewDimension::Undefined) {
bindingInfo.texture.viewDimension = wgpu::TextureViewDimension::e2D;
}
} else if (binding.storageTexture.access != wgpu::StorageTextureAccess::Undefined) {
bindingInfo.bindingType = BindingInfoType::StorageTexture;
bindingInfo.storageTexture = binding.storageTexture;
if (binding.storageTexture.viewDimension == wgpu::TextureViewDimension::Undefined) {
bindingInfo.storageTexture.viewDimension = wgpu::TextureViewDimension::e2D;
}
} else {
const ExternalTextureBindingLayout* externalTextureBindingLayout = nullptr;
FindInChain(binding.nextInChain, &externalTextureBindingLayout);
if (externalTextureBindingLayout != nullptr) {
bindingInfo.bindingType = BindingInfoType::ExternalTexture;
}
}
return bindingInfo;
}
bool SortBindingsCompare(const BindGroupLayoutEntry& a, const BindGroupLayoutEntry& b) {
const bool aIsBuffer = IsBufferBinding(a);
const bool bIsBuffer = IsBufferBinding(b);
if (aIsBuffer != bIsBuffer) {
// Always place buffers first.
return aIsBuffer;
}
if (aIsBuffer) {
bool aHasDynamicOffset = BindingHasDynamicOffset(a);
bool bHasDynamicOffset = BindingHasDynamicOffset(b);
ASSERT(bIsBuffer);
if (aHasDynamicOffset != bHasDynamicOffset) {
// Buffers with dynamic offsets should come before those without.
// This makes it easy to iterate over the dynamic buffer bindings
// [0, dynamicBufferCount) during validation.
return aHasDynamicOffset;
}
if (aHasDynamicOffset) {
ASSERT(bHasDynamicOffset);
ASSERT(a.binding != b.binding);
// Above, we ensured that dynamic buffers are first. Now, ensure that
// dynamic buffer bindings are in increasing order. This is because dynamic
// buffer offsets are applied in increasing order of binding number.
return a.binding < b.binding;
}
}
// This applies some defaults and gives us a single value to check for the binding type.
BindingInfo aInfo = CreateBindGroupLayoutInfo(a);
BindingInfo bInfo = CreateBindGroupLayoutInfo(b);
// Sort by type.
if (aInfo.bindingType != bInfo.bindingType) {
return aInfo.bindingType < bInfo.bindingType;
}
if (a.visibility != b.visibility) {
return a.visibility < b.visibility;
}
switch (aInfo.bindingType) {
case BindingInfoType::Buffer:
if (aInfo.buffer.minBindingSize != bInfo.buffer.minBindingSize) {
return aInfo.buffer.minBindingSize < bInfo.buffer.minBindingSize;
}
break;
case BindingInfoType::Sampler:
if (aInfo.sampler.type != bInfo.sampler.type) {
return aInfo.sampler.type < bInfo.sampler.type;
}
break;
case BindingInfoType::Texture:
if (aInfo.texture.multisampled != bInfo.texture.multisampled) {
return aInfo.texture.multisampled < bInfo.texture.multisampled;
}
if (aInfo.texture.viewDimension != bInfo.texture.viewDimension) {
return aInfo.texture.viewDimension < bInfo.texture.viewDimension;
}
if (aInfo.texture.sampleType != bInfo.texture.sampleType) {
return aInfo.texture.sampleType < bInfo.texture.sampleType;
}
break;
case BindingInfoType::StorageTexture:
if (aInfo.storageTexture.access != bInfo.storageTexture.access) {
return aInfo.storageTexture.access < bInfo.storageTexture.access;
}
if (aInfo.storageTexture.viewDimension != bInfo.storageTexture.viewDimension) {
return aInfo.storageTexture.viewDimension <
bInfo.storageTexture.viewDimension;
}
if (aInfo.storageTexture.format != bInfo.storageTexture.format) {
return aInfo.storageTexture.format < bInfo.storageTexture.format;
}
break;
case BindingInfoType::ExternalTexture:
break;
}
return false;
}
// This is a utility function to help ASSERT that the BGL-binding comparator places buffers
// first.
bool CheckBufferBindingsFirst(ityp::span<BindingIndex, const BindingInfo> bindings) {
BindingIndex lastBufferIndex{0};
BindingIndex firstNonBufferIndex = std::numeric_limits<BindingIndex>::max();
for (BindingIndex i{0}; i < bindings.size(); ++i) {
if (bindings[i].bindingType == BindingInfoType::Buffer) {
lastBufferIndex = std::max(i, lastBufferIndex);
} else {
firstNonBufferIndex = std::min(i, firstNonBufferIndex);
}
}
// If there are no buffers, then |lastBufferIndex| is initialized to 0 and
// |firstNonBufferIndex| gets set to 0.
return firstNonBufferIndex >= lastBufferIndex;
}
} // namespace
// BindGroupLayoutBase
BindGroupLayoutBase::BindGroupLayoutBase(DeviceBase* device,
const BindGroupLayoutDescriptor* descriptor)
: CachedObject(device), mBindingInfo(BindingIndex(descriptor->entryCount)) {
std::vector<BindGroupLayoutEntry> sortedBindings(
descriptor->entries, descriptor->entries + descriptor->entryCount);
std::sort(sortedBindings.begin(), sortedBindings.end(), SortBindingsCompare);
for (BindingIndex i{0}; i < mBindingInfo.size(); ++i) {
const BindGroupLayoutEntry& binding = sortedBindings[static_cast<uint32_t>(i)];
mBindingInfo[i] = CreateBindGroupLayoutInfo(binding);
if (IsBufferBinding(binding)) {
// Buffers must be contiguously packed at the start of the binding info.
ASSERT(GetBufferCount() == i);
}
IncrementBindingCounts(&mBindingCounts, binding);
const auto& it = mBindingMap.emplace(BindingNumber(binding.binding), i);
ASSERT(it.second);
}
ASSERT(CheckBufferBindingsFirst({mBindingInfo.data(), GetBindingCount()}));
ASSERT(mBindingInfo.size() <= kMaxBindingsPerPipelineLayoutTyped);
}
BindGroupLayoutBase::BindGroupLayoutBase(DeviceBase* device, ObjectBase::ErrorTag tag)
: CachedObject(device, tag) {
}
BindGroupLayoutBase::~BindGroupLayoutBase() {
// Do not uncache the actual cached object if we are a blueprint
if (IsCachedReference()) {
GetDevice()->UncacheBindGroupLayout(this);
}
}
// static
BindGroupLayoutBase* BindGroupLayoutBase::MakeError(DeviceBase* device) {
return new BindGroupLayoutBase(device, ObjectBase::kError);
}
const BindGroupLayoutBase::BindingMap& BindGroupLayoutBase::GetBindingMap() const {
ASSERT(!IsError());
return mBindingMap;
}
bool BindGroupLayoutBase::HasBinding(BindingNumber bindingNumber) const {
return mBindingMap.count(bindingNumber) != 0;
}
BindingIndex BindGroupLayoutBase::GetBindingIndex(BindingNumber bindingNumber) const {
ASSERT(!IsError());
const auto& it = mBindingMap.find(bindingNumber);
ASSERT(it != mBindingMap.end());
return it->second;
}
size_t BindGroupLayoutBase::ComputeContentHash() {
ObjectContentHasher recorder;
// std::map is sorted by key, so two BGLs constructed in different orders
// will still record the same.
for (const auto& it : mBindingMap) {
recorder.Record(it.first, it.second);
const BindingInfo& info = mBindingInfo[it.second];
recorder.Record(info.buffer.hasDynamicOffset, info.visibility, info.bindingType,
info.buffer.type, info.buffer.minBindingSize, info.sampler.type,
info.texture.sampleType, info.texture.viewDimension,
info.texture.multisampled, info.storageTexture.access,
info.storageTexture.format, info.storageTexture.viewDimension);
}
return recorder.GetContentHash();
}
bool BindGroupLayoutBase::EqualityFunc::operator()(const BindGroupLayoutBase* a,
const BindGroupLayoutBase* b) const {
if (a->GetBindingCount() != b->GetBindingCount()) {
return false;
}
for (BindingIndex i{0}; i < a->GetBindingCount(); ++i) {
if (a->mBindingInfo[i] != b->mBindingInfo[i]) {
return false;
}
}
return a->mBindingMap == b->mBindingMap;
}
BindingIndex BindGroupLayoutBase::GetBindingCount() const {
return mBindingInfo.size();
}
BindingIndex BindGroupLayoutBase::GetBufferCount() const {
return BindingIndex(mBindingCounts.bufferCount);
}
BindingIndex BindGroupLayoutBase::GetDynamicBufferCount() const {
// This is a binding index because dynamic buffers are packed at the front of the binding
// info.
return static_cast<BindingIndex>(mBindingCounts.dynamicStorageBufferCount +
mBindingCounts.dynamicUniformBufferCount);
}
uint32_t BindGroupLayoutBase::GetUnverifiedBufferCount() const {
return mBindingCounts.unverifiedBufferCount;
}
const BindingCounts& BindGroupLayoutBase::GetBindingCountInfo() const {
return mBindingCounts;
}
size_t BindGroupLayoutBase::GetBindingDataSize() const {
// | ------ buffer-specific ----------| ------------ object pointers -------------|
// | --- offsets + sizes -------------| --------------- Ref<ObjectBase> ----------|
// Followed by:
// |---------buffer size array--------|
// |-uint64_t[mUnverifiedBufferCount]-|
size_t objectPointerStart = mBindingCounts.bufferCount * sizeof(BufferBindingData);
ASSERT(IsAligned(objectPointerStart, alignof(Ref<ObjectBase>)));
size_t bufferSizeArrayStart =
Align(objectPointerStart + mBindingCounts.totalCount * sizeof(Ref<ObjectBase>),
sizeof(uint64_t));
ASSERT(IsAligned(bufferSizeArrayStart, alignof(uint64_t)));
return bufferSizeArrayStart + mBindingCounts.unverifiedBufferCount * sizeof(uint64_t);
}
BindGroupLayoutBase::BindingDataPointers BindGroupLayoutBase::ComputeBindingDataPointers(
void* dataStart) const {
BufferBindingData* bufferData = reinterpret_cast<BufferBindingData*>(dataStart);
auto bindings = reinterpret_cast<Ref<ObjectBase>*>(bufferData + mBindingCounts.bufferCount);
uint64_t* unverifiedBufferSizes = AlignPtr(
reinterpret_cast<uint64_t*>(bindings + mBindingCounts.totalCount), sizeof(uint64_t));
ASSERT(IsPtrAligned(bufferData, alignof(BufferBindingData)));
ASSERT(IsPtrAligned(bindings, alignof(Ref<ObjectBase>)));
ASSERT(IsPtrAligned(unverifiedBufferSizes, alignof(uint64_t)));
return {{bufferData, GetBufferCount()},
{bindings, GetBindingCount()},
{unverifiedBufferSizes, mBindingCounts.unverifiedBufferCount}};
}
} // namespace dawn_native