Adding tests for the FFT compute shaders

64_FFT/CMakeLists.txt

@@ -0,0 +1,24 @@
+include(common RESULT_VARIABLE RES)
+if(NOT RES)
+	message(FATAL_ERROR "common.cmake not found. Should be in {repo_root}/cmake directory")
+endif()
+
+nbl_create_executable_project("" "" "" "" "${NBL_EXECUTABLE_PROJECT_CREATION_PCH_TARGET}")
+
+if(NBL_EMBED_BUILTIN_RESOURCES)
+	set(_BR_TARGET_ ${EXECUTABLE_NAME}_builtinResourceData)
+	set(RESOURCE_DIR "app_resources")
+
+	get_filename_component(_SEARCH_DIRECTORIES_ "${CMAKE_CURRENT_SOURCE_DIR}" ABSOLUTE)
+	get_filename_component(_OUTPUT_DIRECTORY_SOURCE_ "${CMAKE_CURRENT_BINARY_DIR}/src" ABSOLUTE)
+	get_filename_component(_OUTPUT_DIRECTORY_HEADER_ "${CMAKE_CURRENT_BINARY_DIR}/include" ABSOLUTE)
+
+    file(GLOB_RECURSE BUILTIN_RESOURCE_FILES RELATIVE "${CMAKE_CURRENT_SOURCE_DIR}/${RESOURCE_DIR}" CONFIGURE_DEPENDS "${CMAKE_CURRENT_SOURCE_DIR}/${RESOURCE_DIR}/*")
+    foreach(RES_FILE ${BUILTIN_RESOURCE_FILES})
+      LIST_BUILTIN_RESOURCE(RESOURCES_TO_EMBED "${RES_FILE}")
+    endforeach()
+
+	ADD_CUSTOM_BUILTIN_RESOURCES(${_BR_TARGET_} RESOURCES_TO_EMBED "${_SEARCH_DIRECTORIES_}" "${RESOURCE_DIR}" "nbl::this_example::builtin" "${_OUTPUT_DIRECTORY_HEADER_}" "${_OUTPUT_DIRECTORY_SOURCE_}")
+
+	LINK_BUILTIN_RESOURCES_TO_TARGET(${EXECUTABLE_NAME} ${_BR_TARGET_})
+endif()

64_FFT/app_resources/common.hlsl

@@ -0,0 +1,19 @@
+#include "nbl/builtin/hlsl/cpp_compat.hlsl"
+
+// Unfortunately not every piece of C++14 metaprogramming syntax is available in HLSL 202x
+// https://github.com/microsoft/DirectXShaderCompiler/issues/5751#issuecomment-1800847954
+using input_t = nbl::hlsl::float32_t;
+
+struct PushConstantData
+{
+	uint64_t inputAddress;
+	uint64_t outputAddress;
+	uint32_t dataElementCount;
+};
+
+#define _NBL_HLSL_WORKGROUP_SIZE_ 128
+NBL_CONSTEXPR uint32_t WorkgroupSize = _NBL_HLSL_WORKGROUP_SIZE_;
+NBL_CONSTEXPR uint32_t ElementsPerThread = 4;
+NBL_CONSTEXPR uint32_t complexElementCount = WorkgroupSize * ElementsPerThread;
+
+#include "nbl/builtin/hlsl/random/xoroshiro.hlsl"

64_FFT/app_resources/shader.comp.hlsl

@@ -0,0 +1,72 @@
+#include "common.hlsl"
+#include "nbl/builtin/hlsl/workgroup/fft.hlsl"
+//#include "nbl/builtin/hlsl/subgroup/fft.hlsl"
+#include "nbl/builtin/hlsl/glsl_compat/subgroup_basic.hlsl"
+
+[[vk::push_constant]] PushConstantData pushConstants;
+
+// careful: change size according to Scalar type
+groupshared uint32_t sharedmem[4 * WorkgroupSize];
+
+// Users MUST define this method for FFT to work
+namespace nbl { namespace hlsl { namespace glsl
+{
+uint32_t3 gl_WorkGroupSize() { return uint32_t3(WorkgroupSize, 1, 1); }
+} } }
+
+struct SharedMemoryAccessor {
+	void set(uint32_t idx, uint32_t value) 
+	{
+		sharedmem[idx] = value;
+	}
+
+	void get(uint32_t idx, NBL_REF_ARG(uint32_t) value) 
+	{
+		value = sharedmem[idx];
+	}
+
+	void workgroupExecutionAndMemoryBarrier() 
+	{
+		AllMemoryBarrierWithGroupSync();
+    }
+
+};
+
+struct Accessor {
+	void set(uint32_t idx, nbl::hlsl::complex_t<input_t> value) 
+	{
+		vk::RawBufferStore< vector<input_t, 2> >(pushConstants.outputAddress + sizeof(vector<input_t, 2>) * idx, vector<input_t, 2>(value.real(), value.imag()));
+	}
+
+	void get(uint32_t idx, NBL_REF_ARG(nbl::hlsl::complex_t<input_t>) value) 
+	{
+		vector<input_t, 2> aux = vk::RawBufferLoad< vector<input_t, 2> >(pushConstants.inputAddress + sizeof(vector<input_t, 2>) * idx);
+		value.real(aux.x);
+		value.imag(aux.y);
+	}
+
+	void workgroupExecutionAndMemoryBarrier() 
+	{
+		AllMemoryBarrierWithGroupSync();
+    }
+
+	void memoryBarrier() 
+	{
+		AllMemoryBarrier();
+	}
+};
+
+[numthreads(WorkgroupSize,1,1)]
+void main(uint32_t3 ID : SV_DispatchThreadID)
+{
+
+	Accessor accessor;
+	SharedMemoryAccessor sharedmemAccessor;
+
+	// Workgroup	
+
+	nbl::hlsl::workgroup::FFT<ElementsPerThread, true, input_t>::template __call<Accessor, SharedMemoryAccessor>(accessor, sharedmemAccessor);
+	accessor.workgroupExecutionAndMemoryBarrier();
+	nbl::hlsl::workgroup::FFT<ElementsPerThread, false, input_t>::template __call<Accessor, SharedMemoryAccessor>(accessor, sharedmemAccessor);	
+
+}

64_FFT/config.json.template

@@ -0,0 +1,28 @@
+{
+  "enableParallelBuild": true,
+  "threadsPerBuildProcess" : 2,
+  "isExecuted": false,
+  "scriptPath": "",
+  "cmake": {
+    "configurations": [ "Release", "Debug", "RelWithDebInfo" ],
+    "buildModes": [],
+    "requiredOptions": []
+  }, 
+  "profiles": [
+    {
+      "backend": "vulkan", // should be none
+      "platform": "windows",
+      "buildModes": [],
+      "runConfiguration": "Release", // we also need to run in Debug nad RWDI because foundational example
+      "gpuArchitectures": []
+    }
+  ],
+  "dependencies": [],
+  "data": [
+    {
+      "dependencies": [],
+      "command": [""],
+      "outputs": []
+    }
+  ]
+}