Add D3D12 Functionality For Multiple Bind Groups
SPIRV-Cross outputs HLSL utilizing register spaces for seperate bind groups.
This changes the D3D12 backend to also use them.
Bug: dawn:66
Change-Id: I0590ae59fa3d369b57cdb32e4c9808c137fa88dc
Reviewed-on: https://dawn-review.googlesource.com/c/3360
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Kai Ninomiya <kainino@chromium.org>
Commit-Queue: Brandon Jones <brandon1.jones@intel.com>
diff --git a/src/dawn_native/d3d12/PipelineLayoutD3D12.cpp b/src/dawn_native/d3d12/PipelineLayoutD3D12.cpp
index 1ee21e1..e63d9ea 100644
--- a/src/dawn_native/d3d12/PipelineLayoutD3D12.cpp
+++ b/src/dawn_native/d3d12/PipelineLayoutD3D12.cpp
@@ -65,7 +65,7 @@
for (uint32_t i = 0; i < rangeCount; ++i) {
ranges[rangeIndex] = descriptorRanges[i];
- ranges[rangeIndex].BaseShaderRegister += group * kMaxBindingsPerGroup;
+ ranges[rangeIndex].RegisterSpace = group;
rangeIndex++;
}
diff --git a/src/dawn_native/d3d12/ShaderModuleD3D12.cpp b/src/dawn_native/d3d12/ShaderModuleD3D12.cpp
index ad1c506..9911d51 100644
--- a/src/dawn_native/d3d12/ShaderModuleD3D12.cpp
+++ b/src/dawn_native/d3d12/ShaderModuleD3D12.cpp
@@ -52,10 +52,8 @@
for (uint32_t binding = 0; binding < groupBindingInfo.size(); ++binding) {
const BindingInfo& bindingInfo = groupBindingInfo[binding];
if (bindingInfo.used) {
- uint32_t bindGroupOffset = group * kMaxBindingsPerGroup;
uint32_t bindingOffset = bindingOffsets[binding];
- compiler.set_decoration(bindingInfo.id, spv::DecorationBinding,
- bindGroupOffset + bindingOffset);
+ compiler.set_decoration(bindingInfo.id, spv::DecorationBinding, bindingOffset);
}
}
}
diff --git a/src/tests/end2end/BindGroupTests.cpp b/src/tests/end2end/BindGroupTests.cpp
index f86646d..9a7e1f1 100644
--- a/src/tests/end2end/BindGroupTests.cpp
+++ b/src/tests/end2end/BindGroupTests.cpp
@@ -32,6 +32,17 @@
pass.EndPass();
return builder.GetResult();
}
+
+ dawn::PipelineLayout MakeBasicPipelineLayout(
+ dawn::Device device,
+ std::vector<dawn::BindGroupLayout> bindingInitializer) const {
+ dawn::PipelineLayoutDescriptor descriptor;
+
+ descriptor.numBindGroupLayouts = bindingInitializer.size();
+ descriptor.bindGroupLayouts = bindingInitializer.data();
+
+ return device.CreatePipelineLayout(&descriptor);
+ }
};
// Test a bindgroup reused in two command buffers in the same call to queue.Submit().
@@ -277,4 +288,100 @@
EXPECT_PIXEL_RGBA8_EQ(notFilled, renderPass.color, max, max);
}
+TEST_P(BindGroupTests, MultipleBindLayouts) {
+ // Test fails on Metal.
+ // https://bugs.chromium.org/p/dawn/issues/detail?id=33
+ DAWN_SKIP_TEST_IF(IsMetal());
+
+ utils::BasicRenderPass renderPass = utils::CreateBasicRenderPass(device, kRTSize, kRTSize);
+
+ dawn::ShaderModule vsModule = utils::CreateShaderModule(device, dawn::ShaderStage::Vertex, R"(
+ #version 450
+ layout (set = 0, binding = 0) uniform vertexUniformBuffer1 {
+ mat2 transform1;
+ };
+ layout (set = 1, binding = 0) uniform vertexUniformBuffer2 {
+ mat2 transform2;
+ };
+ void main() {
+ const vec2 pos[3] = vec2[3](vec2(-1.f, -1.f), vec2(1.f, -1.f), vec2(-1.f, 1.f));
+ gl_Position = vec4((transform1 + transform2) * pos[gl_VertexIndex], 0.f, 1.f);
+ })");
+
+ dawn::ShaderModule fsModule =
+ utils::CreateShaderModule(device, dawn::ShaderStage::Fragment, R"(
+ #version 450
+ layout (set = 0, binding = 1) uniform fragmentUniformBuffer1 {
+ vec4 color1;
+ };
+ layout (set = 1, binding = 1) uniform fragmentUniformBuffer2 {
+ vec4 color2;
+ };
+ layout(location = 0) out vec4 fragColor;
+ void main() {
+ fragColor = color1 + color2;
+ })");
+
+ dawn::BindGroupLayout layout = utils::MakeBindGroupLayout(
+ device, {
+ {0, dawn::ShaderStageBit::Vertex, dawn::BindingType::UniformBuffer},
+ {1, dawn::ShaderStageBit::Fragment, dawn::BindingType::UniformBuffer},
+ });
+
+ dawn::PipelineLayout pipelineLayout = MakeBasicPipelineLayout(device, {layout, layout});
+
+ utils::ComboRenderPipelineDescriptor textureDescriptor(device);
+ textureDescriptor.layout = pipelineLayout;
+ textureDescriptor.cVertexStage.module = vsModule;
+ textureDescriptor.cFragmentStage.module = fsModule;
+ textureDescriptor.cColorAttachments[0].format = renderPass.colorFormat;
+
+ dawn::RenderPipeline pipeline = device.CreateRenderPipeline(&textureDescriptor);
+
+ struct Data {
+ float transform[8];
+ char padding[256 - 8 * sizeof(float)];
+ float color[4];
+ };
+ ASSERT(offsetof(Data, color) == 256);
+
+ std::vector<Data> data;
+ std::vector<dawn::Buffer> buffers;
+ std::vector<dawn::BindGroup> bindGroups;
+
+ data.push_back(
+ {{1.0f, 0.0f, 0.0f, 0.0f, 0.0f, 0.0f, 0.0f, 0.0f}, {0}, {0.0f, 1.0f, 0.0f, 1.0f}});
+
+ data.push_back(
+ {{0.0f, 0.0f, 0.0f, 0.0f, 0.0f, 1.0f, 0.0f, 0.0f}, {0}, {1.0f, 0.0f, 0.0f, 1.0f}});
+
+ for (int i = 0; i < 2; i++) {
+ dawn::Buffer buffer = utils::CreateBufferFromData(device, &data[i], sizeof(Data),
+ dawn::BufferUsageBit::Uniform);
+ buffers.push_back(buffer);
+ bindGroups.push_back(utils::MakeBindGroup(device, layout,
+ {{0, buffers[i], 0, sizeof(Data::transform)},
+ {1, buffers[i], 256, sizeof(Data::color)}}));
+ }
+
+ dawn::CommandBufferBuilder builder = device.CreateCommandBufferBuilder();
+ dawn::RenderPassEncoder pass = builder.BeginRenderPass(renderPass.renderPassInfo);
+ pass.SetRenderPipeline(pipeline);
+ pass.SetBindGroup(0, bindGroups[0]);
+ pass.SetBindGroup(1, bindGroups[1]);
+ pass.Draw(3, 1, 0, 0);
+ pass.EndPass();
+
+ dawn::CommandBuffer commands = builder.GetResult();
+ queue.Submit(1, &commands);
+
+ RGBA8 filled(255, 255, 0, 255);
+ RGBA8 notFilled(0, 0, 0, 0);
+ int min = 1, max = kRTSize - 3;
+ EXPECT_PIXEL_RGBA8_EQ(filled, renderPass.color, min, min);
+ EXPECT_PIXEL_RGBA8_EQ(filled, renderPass.color, max, min);
+ EXPECT_PIXEL_RGBA8_EQ(filled, renderPass.color, min, max);
+ EXPECT_PIXEL_RGBA8_EQ(notFilled, renderPass.color, max, max);
+}
+
DAWN_INSTANTIATE_TEST(BindGroupTests, D3D12Backend, MetalBackend, OpenGLBackend, VulkanBackend);
