sb << "// Slang HLSL compatibility library\n";
sb << "\n";
sb << "typedef uint UINT;\n";
sb << "\n";
sb << "__generic<T> __magic_type(HLSLAppendStructuredBufferType) struct AppendStructuredBuffer\n";
sb << "{\n";
sb << "    __intrinsic_op void Append(T value);\n";
sb << "\n";
sb << "    __intrinsic_op void GetDimensions(\n";
sb << "        out uint numStructs,\n";
sb << "        out uint stride);\n";
sb << "};\n";
sb << "\n";
sb << "__magic_type(HLSLByteAddressBufferType) struct ByteAddressBuffer\n";
sb << "{\n";
sb << "    __intrinsic_op void GetDimensions(\n";
sb << "        out uint dim);\n";
sb << "\n";
sb << "    __intrinsic_op uint Load(int location);\n";
sb << "    __intrinsic_op uint Load(int location, out uint status);\n";
sb << "\n";
sb << "    __intrinsic_op uint2 Load2(int location);\n";
sb << "    __intrinsic_op uint2 Load2(int location, out uint status);\n";
sb << "\n";
sb << "    __intrinsic_op uint3 Load3(int location);\n";
sb << "    __intrinsic_op uint3 Load3(int location, out uint status);\n";
sb << "\n";
sb << "    __intrinsic_op uint4 Load4(int location);\n";
sb << "    __intrinsic_op uint4 Load4(int location, out uint status);\n";
sb << "};\n";
sb << "\n";
sb << "__generic<T> __magic_type(HLSLStructuredBufferType) struct StructuredBuffer\n";
sb << "{\n";
sb << "    __intrinsic_op void GetDimensions(\n";
sb << "        out uint numStructs,\n";
sb << "        out uint stride);\n";
sb << "\n";
sb << "    __intrinsic_op T Load(int location);\n";
sb << "    __intrinsic_op T Load(int location, out uint status);\n";
sb << "\n";
sb << "    __intrinsic_op(bufferLoad)\n";
sb << "    __subscript(uint index) -> T;\n";
sb << "};\n";
sb << "\n";
sb << "__generic<T> __magic_type(HLSLConsumeStructuredBufferType) struct ConsumeStructuredBuffer\n";
sb << "{\n";
sb << "    __intrinsic_op T Consume();\n";
sb << "\n";
sb << "    __intrinsic_op void GetDimensions(\n";
sb << "        out uint numStructs,\n";
sb << "        out uint stride);\n";
sb << "};\n";
sb << "\n";
sb << "__generic<T, let N : int> __magic_type(HLSLInputPatchType) struct InputPatch\n";
sb << "{\n";
sb << "    __intrinsic_op __subscript(uint index) -> T;\n";
sb << "};\n";
sb << "\n";
sb << "__generic<T, let N : int> __magic_type(HLSLOutputPatchType) struct OutputPatch\n";
sb << "{\n";
sb << "    __intrinsic_op __subscript(uint index) -> T { set; }\n";
sb << "};\n";
sb << "\n";
sb << "__magic_type(HLSLRWByteAddressBufferType) struct RWByteAddressBuffer\n";
sb << "{\n";
sb << "    // Note(tfoley): supports alll operations from `ByteAddressBuffer`\n";
sb << "    // TODO(tfoley): can this be made a sub-type?\n";
sb << "\n";
sb << "    __intrinsic_op void GetDimensions(\n";
sb << "        out uint dim);\n";
sb << "\n";
sb << "    __intrinsic_op uint Load(int location);\n";
sb << "    __intrinsic_op uint Load(int location, out uint status);\n";
sb << "\n";
sb << "    __intrinsic_op uint2 Load2(int location);\n";
sb << "    __intrinsic_op uint2 Load2(int location, out uint status);\n";
sb << "\n";
sb << "    __intrinsic_op uint3 Load3(int location);\n";
sb << "    __intrinsic_op uint3 Load3(int location, out uint status);\n";
sb << "\n";
sb << "    __intrinsic_op uint4 Load4(int location);\n";
sb << "    __intrinsic_op uint4 Load4(int location, out uint status);\n";
sb << "\n";
sb << "    // Added operations:\n";
sb << "\n";
sb << "    __intrinsic_op void InterlockedAdd(\n";
sb << "        UINT dest,\n";
sb << "        UINT value,\n";
sb << "        out UINT original_value);\n";
sb << "    __intrinsic_op void InterlockedAdd(\n";
sb << "        UINT dest,\n";
sb << "        UINT value);\n";
sb << "\n";
sb << "    __intrinsic_op void InterlockedAnd(\n";
sb << "        UINT dest,\n";
sb << "        UINT value,\n";
sb << "        out UINT original_value);\n";
sb << "    __intrinsic_op void InterlockedAnd(\n";
sb << "        UINT dest,\n";
sb << "        UINT value);\n";
sb << "\n";
sb << "    __intrinsic_op void InterlockedCompareExchange(\n";
sb << "        UINT dest,\n";
sb << "        UINT compare_value,\n";
sb << "        UINT value,\n";
sb << "        out UINT original_value);\n";
sb << "    __intrinsic_op void InterlockedCompareExchange(\n";
sb << "        UINT dest,\n";
sb << "        UINT compare_value,\n";
sb << "        UINT value);\n";
sb << "\n";
sb << "    __intrinsic_op void InterlockedCompareStore(\n";
sb << "        UINT dest,\n";
sb << "        UINT compare_value,\n";
sb << "        UINT value);\n";
sb << "    __intrinsic_op void InterlockedCompareStore(\n";
sb << "        UINT dest,\n";
sb << "        UINT compare_value);\n";
sb << "\n";
sb << "    __intrinsic_op void InterlockedExchange(\n";
sb << "        UINT dest,\n";
sb << "        UINT value,\n";
sb << "        out UINT original_value);\n";
sb << "    __intrinsic_op void InterlockedExchange(\n";
sb << "        UINT dest,\n";
sb << "        UINT value);\n";
sb << "\n";
sb << "    __intrinsic_op void InterlockedMax(\n";
sb << "        UINT dest,\n";
sb << "        UINT value,\n";
sb << "        out UINT original_value);\n";
sb << "    __intrinsic_op void InterlockedMax(\n";
sb << "        UINT dest,\n";
sb << "        UINT value);\n";
sb << "\n";
sb << "    __intrinsic_op void InterlockedMin(\n";
sb << "        UINT dest,\n";
sb << "        UINT value,\n";
sb << "        out UINT original_value);\n";
sb << "    __intrinsic_op void InterlockedMin(\n";
sb << "        UINT dest,\n";
sb << "        UINT value);\n";
sb << "\n";
sb << "    __intrinsic_op void InterlockedOr(\n";
sb << "        UINT dest,\n";
sb << "        UINT value,\n";
sb << "        out UINT original_value);\n";
sb << "    __intrinsic_op void InterlockedOr(\n";
sb << "        UINT dest,\n";
sb << "        UINT value);\n";
sb << "\n";
sb << "    __intrinsic_op void InterlockedXor(\n";
sb << "        UINT dest,\n";
sb << "        UINT value,\n";
sb << "        out UINT original_value);\n";
sb << "    __intrinsic_op void InterlockedXor(\n";
sb << "        UINT dest,\n";
sb << "        UINT value);\n";
sb << "\n";
sb << "    __intrinsic_op void Store(\n";
sb << "        uint address,\n";
sb << "        uint value);\n";
sb << "\n";
sb << "    __intrinsic_op void Store2(\n";
sb << "        uint address,\n";
sb << "        uint2 value);\n";
sb << "\n";
sb << "    __intrinsic_op void Store3(\n";
sb << "        uint address,\n";
sb << "        uint3 value);\n";
sb << "\n";
sb << "    __intrinsic_op void Store4(\n";
sb << "        uint address,\n";
sb << "        uint4 value);\n";
sb << "};\n";
sb << "\n";
sb << "__generic<T>\n";
sb << "__magic_type(HLSLRWStructuredBufferType)\n";
sb << "__intrinsic_type(";

	// TODO: we really need a simple way to write an "expression splice"
	sb << kIROp_readWriteStructuredBufferType;
sb << ")\n";
sb << "struct RWStructuredBuffer\n";
sb << "{\n";
sb << "    __intrinsic_op uint DecrementCounter();\n";
sb << "\n";
sb << "    __intrinsic_op void GetDimensions(\n";
sb << "        out uint numStructs,\n";
sb << "        out uint stride);\n";
sb << "\n";
sb << "    __intrinsic_op void IncrementCounter();\n";
sb << "\n";
sb << "    __intrinsic_op T Load(int location);\n";
sb << "    __intrinsic_op T Load(int location, out uint status);\n";
sb << "\n";
sb << "\t__intrinsic_op\n";
sb << "\t__subscript(uint index) -> T\n";
sb << "\t{\n";
sb << "\t\t__intrinsic_op(bufferLoad)\n";
sb << "\t\tget;\n";
sb << "\n";
sb << "\t\t__intrinsic_op(bufferStore)\n";
sb << "\t\tset;\n";
sb << "\t}\n";
sb << "};\n";
sb << "\n";
sb << "__generic<T> __magic_type(HLSLPointStreamType) struct PointStream\n";
sb << "{\n";
sb << "    void Append(T value);\n";
sb << "    void RestartStrip();\n";
sb << "};\n";
sb << "\n";
sb << "__generic<T> __magic_type(HLSLLineStreamType) struct LineStream\n";
sb << "{\n";
sb << "    void Append(T value);\n";
sb << "    void RestartStrip();\n";
sb << "};\n";
sb << "\n";
sb << "__generic<T> __magic_type(HLSLTriangleStreamType) struct TriangleStream\n";
sb << "{\n";
sb << "    void Append(T value);\n";
sb << "    void RestartStrip();\n";
sb << "};\n";
sb << "\n";
sb << "// Note(tfoley): Trying to systematically add all the HLSL builtins\n";
sb << "\n";
sb << "// Try to terminate the current draw or dispatch call (HLSL SM 4.0)\n";
sb << "__intrinsic_op void abort();\n";
sb << "\n";
sb << "// Absolute value (HLSL SM 1.0)\n";
sb << "__generic<T : __BuiltinSignedArithmeticType> __intrinsic_op T abs(T x);\n";
sb << "__generic<T : __BuiltinSignedArithmeticType, let N : int> __intrinsic_op vector<T,N> abs(vector<T,N> x);\n";
sb << "__generic<T : __BuiltinSignedArithmeticType, let N : int, let M : int> __intrinsic_op matrix<T,N,M> abs(matrix<T,N,M> x);\n";
sb << "\n";
sb << "// Inverse cosine (HLSL SM 1.0)\n";
sb << "__generic<T : __BuiltinFloatingPointType> __intrinsic_op T acos(T x);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int> __intrinsic_op vector<T,N> acos(vector<T,N> x);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int, let M : int> __intrinsic_op matrix<T,N,M> acos(matrix<T,N,M> x);\n";
sb << "\n";
sb << "// Test if all components are non-zero (HLSL SM 1.0)\n";
sb << "__generic<T : __BuiltinType> __intrinsic_op T all(T x);\n";
sb << "__generic<T : __BuiltinType, let N : int> __intrinsic_op vector<T,N> all(vector<T,N> x);\n";
sb << "__generic<T : __BuiltinType, let N : int, let M : int> __intrinsic_op matrix<T,N,M> all(matrix<T,N,M> x);\n";
sb << "\n";
sb << "// Barrier for writes to all memory spaces (HLSL SM 5.0)\n";
sb << "__intrinsic_op void AllMemoryBarrier();\n";
sb << "\n";
sb << "// Thread-group sync and barrier for writes to all memory spaces (HLSL SM 5.0)\n";
sb << "__intrinsic_op void AllMemoryBarrierWithGroupSync();\n";
sb << "\n";
sb << "// Test if any components is non-zero (HLSL SM 1.0)\n";
sb << "__generic<T : __BuiltinType> __intrinsic_op T any(T x);\n";
sb << "__generic<T : __BuiltinType, let N : int> __intrinsic_op vector<T,N> any(vector<T,N> x);\n";
sb << "__generic<T : __BuiltinType, let N : int, let M : int> __intrinsic_op matrix<T,N,M> any(matrix<T,N,M> x);\n";
sb << "\n";
sb << "\n";
sb << "// Reinterpret bits as a double (HLSL SM 5.0)\n";
sb << "__intrinsic_op double asdouble(uint lowbits, uint highbits);\n";
sb << "\n";
sb << "// Reinterpret bits as a float (HLSL SM 4.0)\n";
sb << "__intrinsic_op float asfloat( int x);\n";
sb << "__intrinsic_op float asfloat(uint x);\n";
sb << "__generic<let N : int> __intrinsic_op vector<float,N> asfloat(vector< int,N> x);\n";
sb << "__generic<let N : int> __intrinsic_op vector<float,N> asfloat(vector<uint,N> x);\n";
sb << "__generic<let N : int, let M : int> __intrinsic_op matrix<float,N,M> asfloat(matrix< int,N,M> x);\n";
sb << "__generic<let N : int, let M : int> __intrinsic_op matrix<float,N,M> asfloat(matrix<uint,N,M> x);\n";
sb << "\n";
sb << "\n";
sb << "// Inverse sine (HLSL SM 1.0)\n";
sb << "__generic<T : __BuiltinFloatingPointType> __intrinsic_op T asin(T x);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int> __intrinsic_op vector<T,N> asin(vector<T,N> x);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int, let M : int> __intrinsic_op matrix<T,N,M> asin(matrix<T,N,M> x);\n";
sb << "\n";
sb << "// Reinterpret bits as an int (HLSL SM 4.0)\n";
sb << "__intrinsic_op int asint(float x);\n";
sb << "__intrinsic_op int asint(uint x);\n";
sb << "__generic<let N : int> __intrinsic_op vector<int,N> asint(vector<float,N> x);\n";
sb << "__generic<let N : int> __intrinsic_op vector<int,N> asint(vector<uint,N> x);\n";
sb << "__generic<let N : int, let M : int> __intrinsic_op matrix<int,N,M> asint(matrix<float,N,M> x);\n";
sb << "__generic<let N : int, let M : int> __intrinsic_op matrix<int,N,M> asint(matrix<uint,N,M> x);\n";
sb << "\n";
sb << "// Reinterpret bits of double as a uint (HLSL SM 5.0)\n";
sb << "__intrinsic_op void asuint(double value, out uint lowbits, out uint highbits);\n";
sb << "\n";
sb << "// Reinterpret bits as a uint (HLSL SM 4.0)\n";
sb << "__intrinsic_op uint asuint(float x);\n";
sb << "__intrinsic_op uint asuint(int x);\n";
sb << "__generic<let N : int> __intrinsic_op vector<uint,N> asuint(vector<float,N> x);\n";
sb << "__generic<let N : int> __intrinsic_op vector<uint,N> asuint(vector<int,N> x);\n";
sb << "__generic<let N : int, let M : int> __intrinsic_op matrix<uint,N,M> asuint(matrix<float,N,M> x);\n";
sb << "__generic<let N : int, let M : int> __intrinsic_op matrix<uint,N,M> asuint(matrix<int,N,M> x);\n";
sb << "\n";
sb << "// Inverse tangent (HLSL SM 1.0)\n";
sb << "__generic<T : __BuiltinFloatingPointType> __intrinsic_op T atan(T x);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int> __intrinsic_op vector<T,N> atan(vector<T,N> x);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int, let M : int> __intrinsic_op matrix<T,N,M> atan(matrix<T,N,M> x);\n";
sb << "\n";
sb << "__generic<T : __BuiltinFloatingPointType>\n";
sb << "__target_intrinsic(glsl,\"atan($0,$1)\")\n";
sb << "__intrinsic_op\n";
sb << "T atan2(T y, T x);\n";
sb << "\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int>\n";
sb << "__target_intrinsic(glsl,\"atan($0,$1)\")\n";
sb << "__intrinsic_op\n";
sb << "vector<T,N> atan2(vector<T,N> y, vector<T,N> x);\n";
sb << "\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int, let M : int>\n";
sb << "__target_intrinsic(glsl,\"atan($0,$1)\")\n";
sb << "__intrinsic_op\n";
sb << "matrix<T,N,M> atan2(matrix<T,N,M> y, matrix<T,N,M> x);\n";
sb << "\n";
sb << "// Ceiling (HLSL SM 1.0)\n";
sb << "__generic<T : __BuiltinFloatingPointType> __intrinsic_op T ceil(T x);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int> __intrinsic_op vector<T,N> ceil(vector<T,N> x);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int, let M : int> __intrinsic_op matrix<T,N,M> ceil(matrix<T,N,M> x);\n";
sb << "\n";
sb << "\n";
sb << "// Check access status to tiled resource\n";
sb << "__intrinsic_op bool CheckAccessFullyMapped(uint status);\n";
sb << "\n";
sb << "// Clamp (HLSL SM 1.0)\n";
sb << "__generic<T : __BuiltinArithmeticType> __intrinsic_op T clamp(T x, T min, T max);\n";
sb << "__generic<T : __BuiltinArithmeticType, let N : int> __intrinsic_op vector<T,N> clamp(vector<T,N> x, vector<T,N> min, vector<T,N> max);\n";
sb << "__generic<T : __BuiltinArithmeticType, let N : int, let M : int> __intrinsic_op matrix<T,N,M> clamp(matrix<T,N,M> x, matrix<T,N,M> min, matrix<T,N,M> max);\n";
sb << "\n";
sb << "// Clip (discard) fragment conditionally\n";
sb << "__generic<T : __BuiltinFloatingPointType> __intrinsic_op void clip(T x);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int> __intrinsic_op void clip(vector<T,N> x);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int, let M : int> __intrinsic_op void clip(matrix<T,N,M> x);\n";
sb << "\n";
sb << "// Cosine\n";
sb << "__generic<T : __BuiltinFloatingPointType> __intrinsic_op T cos(T x);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int> __intrinsic_op vector<T,N> cos(vector<T,N> x);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int, let M : int> __intrinsic_op matrix<T,N,M> cos(matrix<T,N,M> x);\n";
sb << "\n";
sb << "// Hyperbolic cosine\n";
sb << "__generic<T : __BuiltinFloatingPointType> __intrinsic_op T cosh(T x);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int> __intrinsic_op vector<T,N> cosh(vector<T,N> x);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int, let M : int> __intrinsic_op matrix<T,N,M> cosh(matrix<T,N,M> x);\n";
sb << "\n";
sb << "// Population count\n";
sb << "__intrinsic_op uint countbits(uint value);\n";
sb << "\n";
sb << "// Cross product\n";
sb << "__generic<T : __BuiltinArithmeticType> __intrinsic_op vector<T,3> cross(vector<T,3> x, vector<T,3> y);\n";
sb << "\n";
sb << "// Convert encoded color\n";
sb << "__intrinsic_op int4 D3DCOLORtoUBYTE4(float4 x);\n";
sb << "\n";
sb << "// Partial-difference derivatives\n";
sb << "__generic<T : __BuiltinFloatingPointType>\n";
sb << "__target_intrinsic(glsl, dFdx)\n";
sb << "__intrinsic_op\n";
sb << "T ddx(T x);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int>\n";
sb << "__target_intrinsic(glsl, dFdx)\n";
sb << "__intrinsic_op\n";
sb << "vector<T,N> ddx(vector<T,N> x);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int, let M : int>\n";
sb << "__target_intrinsic(glsl, dFdx)\n";
sb << "__intrinsic_op\n";
sb << "matrix<T,N,M> ddx(matrix<T,N,M> x);\n";
sb << "\n";
sb << "__generic<T : __BuiltinFloatingPointType>\n";
sb << "__glsl_extension(GL_ARB_derivative_control)\n";
sb << "__target_intrinsic(glsl, dFdxCoarse)\n";
sb << "__intrinsic_op\n";
sb << "T ddx_coarse(T x);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int>\n";
sb << "__glsl_extension(GL_ARB_derivative_control)\n";
sb << "__target_intrinsic(glsl, dFdxCoarse)\n";
sb << "__intrinsic_op\n";
sb << "vector<T,N> ddx_coarse(vector<T,N> x);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int, let M : int>\n";
sb << "__glsl_extension(GL_ARB_derivative_control)\n";
sb << "__target_intrinsic(glsl, dFdxCoarse)\n";
sb << "__intrinsic_op\n";
sb << "matrix<T,N,M> ddx_coarse(matrix<T,N,M> x);\n";
sb << "\n";
sb << "__generic<T : __BuiltinFloatingPointType>\n";
sb << "__glsl_extension(GL_ARB_derivative_control)\n";
sb << "__target_intrinsic(glsl, dFdxFine)\n";
sb << "__intrinsic_op\n";
sb << "T ddx_fine(T x);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int>\n";
sb << "__glsl_extension(GL_ARB_derivative_control)\n";
sb << "__target_intrinsic(glsl, dFdxFine)\n";
sb << "__intrinsic_op\n";
sb << "vector<T,N> ddx_fine(vector<T,N> x);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int, let M : int>\n";
sb << "__glsl_extension(GL_ARB_derivative_control)\n";
sb << "__target_intrinsic(glsl, dFdxFine)\n";
sb << "__intrinsic_op\n";
sb << "matrix<T,N,M> ddx_fine(matrix<T,N,M> x);\n";
sb << "\n";
sb << "__generic<T : __BuiltinFloatingPointType>\n";
sb << "__target_intrinsic(glsl, dFdy)\n";
sb << "__intrinsic_op\n";
sb << "T ddy(T x);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int>\n";
sb << "__target_intrinsic(glsl, dFdy)\n";
sb << "__intrinsic_op\n";
sb << "vector<T,N> ddy(vector<T,N> x);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int, let M : int>\n";
sb << "__target_intrinsic(glsl, dFdy)\n";
sb << "__intrinsic_op\n";
sb << " matrix<T,N,M> ddy(matrix<T,N,M> x);\n";
sb << "\n";
sb << "__generic<T : __BuiltinFloatingPointType>\n";
sb << "__glsl_extension(GL_ARB_derivative_control)\n";
sb << "__target_intrinsic(glsl, dFdyCoarse)\n";
sb << "__intrinsic_op\n";
sb << "T ddy_coarse(T x);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int>\n";
sb << "__glsl_extension(GL_ARB_derivative_control)\n";
sb << "__target_intrinsic(glsl, dFdyCoarse)\n";
sb << "__intrinsic_op\n";
sb << "vector<T,N> ddy_coarse(vector<T,N> x);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int, let M : int>\n";
sb << "__glsl_extension(GL_ARB_derivative_control)\n";
sb << "__target_intrinsic(glsl, dFdyCoarse)\n";
sb << "__intrinsic_op\n";
sb << "matrix<T,N,M> ddy_coarse(matrix<T,N,M> x);\n";
sb << "\n";
sb << "__generic<T : __BuiltinFloatingPointType>\n";
sb << "__glsl_extension(GL_ARB_derivative_control)\n";
sb << "__target_intrinsic(glsl, dFdyFine)\n";
sb << "__intrinsic_op\n";
sb << "T ddy_fine(T x);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int>\n";
sb << "__glsl_extension(GL_ARB_derivative_control)\n";
sb << "__target_intrinsic(glsl, dFdyFine)\n";
sb << "__intrinsic_op\n";
sb << "vector<T,N> ddy_fine(vector<T,N> x);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int, let M : int>\n";
sb << "__glsl_extension(GL_ARB_derivative_control)\n";
sb << "__target_intrinsic(glsl, dFdyFine)\n";
sb << "__intrinsic_op\n";
sb << "matrix<T,N,M> ddy_fine(matrix<T,N,M> x);\n";
sb << "\n";
sb << "\n";
sb << "// Radians to degrees\n";
sb << "__generic<T : __BuiltinFloatingPointType> __intrinsic_op T degrees(T x);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int> __intrinsic_op vector<T,N> degrees(vector<T,N> x);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int, let M : int> __intrinsic_op matrix<T,N,M> degrees(matrix<T,N,M> x);\n";
sb << "\n";
sb << "// Matrix determinant\n";
sb << "\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int> __intrinsic_op T determinant(matrix<T,N,N> m);\n";
sb << "\n";
sb << "// Barrier for device memory\n";
sb << "__intrinsic_op void DeviceMemoryBarrier();\n";
sb << "__intrinsic_op void DeviceMemoryBarrierWithGroupSync();\n";
sb << "\n";
sb << "// Vector distance\n";
sb << "\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int> __intrinsic_op T distance(vector<T,N> x, vector<T,N> y);\n";
sb << "\n";
sb << "// Vector dot product\n";
sb << "\n";
sb << "__generic<T : __BuiltinArithmeticType, let N : int> __intrinsic_op T dot(vector<T,N> x, vector<T,N> y);\n";
sb << "\n";
sb << "// Helper for computing distance terms for lighting (obsolete)\n";
sb << "\n";
sb << "__generic<T : __BuiltinFloatingPointType> __intrinsic_op vector<T,4> dst(vector<T,4> x, vector<T,4> y);\n";
sb << "\n";
sb << "// Error message\n";
sb << "\n";
sb << "// __intrinsic_op void errorf( string format, ... );\n";
sb << "\n";
sb << "// Attribute evaluation\n";
sb << "\n";
sb << "__generic<T : __BuiltinArithmeticType> __intrinsic_op T EvaluateAttributeAtCentroid(T x);\n";
sb << "__generic<T : __BuiltinArithmeticType, let N : int> __intrinsic_op vector<T,N> EvaluateAttributeAtCentroid(vector<T,N> x);\n";
sb << "__generic<T : __BuiltinArithmeticType, let N : int, let M : int> __intrinsic_op matrix<T,N,M> EvaluateAttributeAtCentroid(matrix<T,N,M> x);\n";
sb << "\n";
sb << "__generic<T : __BuiltinArithmeticType> __intrinsic_op T EvaluateAttributeAtSample(T x, uint sampleindex);\n";
sb << "__generic<T : __BuiltinArithmeticType, let N : int> __intrinsic_op vector<T,N> EvaluateAttributeAtSample(vector<T,N> x, uint sampleindex);\n";
sb << "__generic<T : __BuiltinArithmeticType, let N : int, let M : int> __intrinsic_op matrix<T,N,M> EvaluateAttributeAtSample(matrix<T,N,M> x, uint sampleindex);\n";
sb << "\n";
sb << "__generic<T : __BuiltinArithmeticType> __intrinsic_op T EvaluateAttributeSnapped(T x, int2 offset);\n";
sb << "__generic<T : __BuiltinArithmeticType, let N : int> __intrinsic_op vector<T,N> EvaluateAttributeSnapped(vector<T,N> x, int2 offset);\n";
sb << "__generic<T : __BuiltinArithmeticType, let N : int, let M : int> __intrinsic_op matrix<T,N,M> EvaluateAttributeSnapped(matrix<T,N,M> x, int2 offset);\n";
sb << "\n";
sb << "// Base-e exponent\n";
sb << "__generic<T : __BuiltinFloatingPointType> __intrinsic_op T exp(T x);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int> __intrinsic_op vector<T,N> exp(vector<T,N> x);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int, let M : int> __intrinsic_op matrix<T,N,M> exp(matrix<T,N,M> x);\n";
sb << "\n";
sb << "// Base-2 exponent\n";
sb << "__generic<T : __BuiltinFloatingPointType> __intrinsic_op T exp2(T x);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int> __intrinsic_op vector<T,N> exp2(vector<T,N> x);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int, let M : int> __intrinsic_op matrix<T,N,M> exp2(matrix<T,N,M> x);\n";
sb << "\n";
sb << "// Convert 16-bit float stored in low bits of integer\n";
sb << "__intrinsic_op float f16tof32(uint value);\n";
sb << "__generic<let N : int> __intrinsic_op vector<float,N> f16tof32(vector<uint,N> value);\n";
sb << "\n";
sb << "// Convert to 16-bit float stored in low bits of integer\n";
sb << "__intrinsic_op uint f32tof16(float value);\n";
sb << "__generic<let N : int> __intrinsic_op vector<uint,N> f32tof16(vector<float,N> value);\n";
sb << "\n";
sb << "// Flip surface normal to face forward, if needed\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int> __intrinsic_op vector<T,N> faceforward(vector<T,N> n, vector<T,N> i, vector<T,N> ng);\n";
sb << "\n";
sb << "// Find first set bit starting at high bit and working down\n";
sb << "__intrinsic_op int firstbithigh(int value);\n";
sb << "__generic<let N : int> __intrinsic_op vector<int,N> firstbithigh(vector<int,N> value);\n";
sb << "\n";
sb << "__intrinsic_op uint firstbithigh(uint value);\n";
sb << "__generic<let N : int> __intrinsic_op vector<uint,N> firstbithigh(vector<uint,N> value);\n";
sb << "\n";
sb << "// Find first set bit starting at low bit and working up\n";
sb << "__intrinsic_op int firstbitlow(int value);\n";
sb << "__generic<let N : int> __intrinsic_op vector<int,N> firstbitlow(vector<int,N> value);\n";
sb << "\n";
sb << "__intrinsic_op uint firstbitlow(uint value);\n";
sb << "__generic<let N : int> __intrinsic_op vector<uint,N> firstbitlow(vector<uint,N> value);\n";
sb << "\n";
sb << "// Floor (HLSL SM 1.0)\n";
sb << "__generic<T : __BuiltinFloatingPointType> __intrinsic_op T floor(T x);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int> __intrinsic_op vector<T,N> floor(vector<T,N> x);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int, let M : int> __intrinsic_op matrix<T,N,M> floor(matrix<T,N,M> x);\n";
sb << "\n";
sb << "// Fused multiply-add for doubles\n";
sb << "__intrinsic_op double fma(double a, double b, double c);\n";
sb << "__generic<let N : int> __intrinsic_op vector<double, N> fma(vector<double, N> a, vector<double, N> b, vector<double, N> c);\n";
sb << "__generic<let N : int, let M : int> __intrinsic_op matrix<double,N,M> fma(matrix<double,N,M> a, matrix<double,N,M> b, matrix<double,N,M> c);\n";
sb << "\n";
sb << "// Floating point remainder of x/y\n";
sb << "__generic<T : __BuiltinFloatingPointType> __intrinsic_op T fmod(T x, T y);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int> __intrinsic_op vector<T,N> fmod(vector<T,N> x, vector<T,N> y);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int, let M : int> __intrinsic_op matrix<T,N,M> fmod(matrix<T,N,M> x, matrix<T,N,M> y);\n";
sb << "\n";
sb << "// Fractional part\n";
sb << "__generic<T : __BuiltinFloatingPointType>\n";
sb << "__target_intrinsic(glsl, fract)\n";
sb << "__intrinsic_op\n";
sb << "T frac(T x);\n";
sb << "\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int>\n";
sb << "__target_intrinsic(glsl, fract)\n";
sb << "__intrinsic_op\n";
sb << "vector<T,N> frac(vector<T,N> x);\n";
sb << "\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int, let M : int>\n";
sb << "__target_intrinsic(glsl, fract)\n";
sb << "__intrinsic_op\n";
sb << "matrix<T,N,M> frac(matrix<T,N,M> x);\n";
sb << "\n";
sb << "// Split float into mantissa and exponent\n";
sb << "__generic<T : __BuiltinFloatingPointType> __intrinsic_op T frexp(T x, out T exp);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int> __intrinsic_op vector<T,N> frexp(vector<T,N> x, out vector<T,N> exp);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int, let M : int> __intrinsic_op matrix<T,N,M> frexp(matrix<T,N,M> x, out matrix<T,N,M> exp);\n";
sb << "\n";
sb << "// Texture filter width\n";
sb << "__generic<T : __BuiltinFloatingPointType> __intrinsic_op T fwidth(T x);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int> __intrinsic_op vector<T,N> fwidth(vector<T,N> x);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int, let M : int> __intrinsic_op matrix<T,N,M> fwidth(matrix<T,N,M> x);\n";
sb << "\n";
sb << "// Get number of samples in render target\n";
sb << "__intrinsic_op uint GetRenderTargetSampleCount();\n";
sb << "\n";
sb << "// Get position of given sample\n";
sb << "__intrinsic_op float2 GetRenderTargetSamplePosition(int Index);\n";
sb << "\n";
sb << "// Group memory barrier\n";
sb << "__intrinsic_op void GroupMemoryBarrier();\n";
sb << "__intrinsic_op void GroupMemoryBarrierWithGroupSync();\n";
sb << "\n";
sb << "// Atomics\n";
sb << "__intrinsic_op void InterlockedAdd(in out  int dest,  int value, out  int original_value);\n";
sb << "__intrinsic_op void InterlockedAdd(in out uint dest, uint value, out uint original_value);\n";
sb << "\n";
sb << "__intrinsic_op void InterlockedAnd(in out  int dest,  int value, out  int original_value);\n";
sb << "__intrinsic_op void InterlockedAnd(in out uint dest, uint value, out uint original_value);\n";
sb << "\n";
sb << "__intrinsic_op void InterlockedCompareExchange(in out  int dest,  int compare_value,  int value, out  int original_value);\n";
sb << "__intrinsic_op void InterlockedCompareExchange(in out uint dest, uint compare_value, uint value, out uint original_value);\n";
sb << "\n";
sb << "__intrinsic_op void InterlockedCompareStore(in out  int dest,  int compare_value,  int value);\n";
sb << "__intrinsic_op void InterlockedCompareStore(in out uint dest, uint compare_value, uint value);\n";
sb << "\n";
sb << "__intrinsic_op void InterlockedExchange(in out  int dest,  int value, out  int original_value);\n";
sb << "__intrinsic_op void InterlockedExchange(in out uint dest, uint value, out uint original_value);\n";
sb << "\n";
sb << "__intrinsic_op void InterlockedMax(in out  int dest,  int value, out  int original_value);\n";
sb << "__intrinsic_op void InterlockedMax(in out uint dest, uint value, out uint original_value);\n";
sb << "\n";
sb << "__intrinsic_op void InterlockedMin(in out  int dest,  int value, out  int original_value);\n";
sb << "__intrinsic_op void InterlockedMin(in out uint dest, uint value, out uint original_value);\n";
sb << "\n";
sb << "__intrinsic_op void InterlockedOr(in out  int dest,  int value, out  int original_value);\n";
sb << "__intrinsic_op void InterlockedOr(in out uint dest, uint value, out uint original_value);\n";
sb << "\n";
sb << "__intrinsic_op void InterlockedXor(in out  int dest,  int value, out  int original_value);\n";
sb << "__intrinsic_op void InterlockedXor(in out uint dest, uint value, out uint original_value);\n";
sb << "\n";
sb << "// Is floating-point value finite?\n";
sb << "__generic<T : __BuiltinFloatingPointType> __intrinsic_op bool isfinite(T x);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int> __intrinsic_op vector<bool,N> isfinite(vector<T,N> x);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int, let M : int> __intrinsic_op matrix<bool,N,M> isfinite(matrix<T,N,M> x);\n";
sb << "\n";
sb << "// Is floating-point value infinite?\n";
sb << "__generic<T : __BuiltinFloatingPointType> __intrinsic_op bool isinf(T x);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int> __intrinsic_op vector<bool,N> isinf(vector<T,N> x);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int, let M : int> __intrinsic_op matrix<bool,N,M> isinf(matrix<T,N,M> x);\n";
sb << "\n";
sb << "// Is floating-point value not-a-number?\n";
sb << "__generic<T : __BuiltinFloatingPointType> __intrinsic_op bool isnan(T x);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int> __intrinsic_op vector<bool,N> isnan(vector<T,N> x);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int, let M : int> __intrinsic_op matrix<bool,N,M> isnan(matrix<T,N,M> x);\n";
sb << "\n";
sb << "// Construct float from mantissa and exponent\n";
sb << "__generic<T : __BuiltinFloatingPointType> __intrinsic_op T ldexp(T x, T exp);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int> __intrinsic_op vector<T,N> ldexp(vector<T,N> x, vector<T,N> exp);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int, let M : int> __intrinsic_op matrix<T,N,M> ldexp(matrix<T,N,M> x, matrix<T,N,M> exp);\n";
sb << "\n";
sb << "// Vector length\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int> __intrinsic_op T length(vector<T,N> x);\n";
sb << "\n";
sb << "// Linear interpolation\n";
sb << "__generic<T : __BuiltinFloatingPointType>\n";
sb << "__target_intrinsic(glsl, mix)\n";
sb << "__intrinsic_op\n";
sb << "T lerp(T x, T y, T s);\n";
sb << "\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int>\n";
sb << "__target_intrinsic(glsl, mix)\n";
sb << "__intrinsic_op\n";
sb << "vector<T,N> lerp(vector<T,N> x, vector<T,N> y, vector<T,N> s);\n";
sb << "\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int, let M : int>\n";
sb << "__target_intrinsic(glsl, mix)\n";
sb << "__intrinsic_op\n";
sb << "matrix<T,N,M> lerp(matrix<T,N,M> x, matrix<T,N,M> y, matrix<T,N,M> s);\n";
sb << "\n";
sb << "// Legacy lighting function (obsolete)\n";
sb << "__intrinsic_op float4 lit(float n_dot_l, float n_dot_h, float m);\n";
sb << "\n";
sb << "// Base-e logarithm\n";
sb << "__generic<T : __BuiltinFloatingPointType> __intrinsic_op T log(T x);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int> __intrinsic_op vector<T,N> log(vector<T,N> x);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int, let M : int> __intrinsic_op matrix<T,N,M> log(matrix<T,N,M> x);\n";
sb << "\n";
sb << "// Base-10 logarithm\n";
sb << "__generic<T : __BuiltinFloatingPointType> __intrinsic_op T log10(T x);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int> __intrinsic_op vector<T,N> log10(vector<T,N> x);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int, let M : int> __intrinsic_op matrix<T,N,M> log10(matrix<T,N,M> x);\n";
sb << "\n";
sb << "// Base-2 logarithm\n";
sb << "__generic<T : __BuiltinFloatingPointType> __intrinsic_op T log2(T x);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int> __intrinsic_op vector<T,N> log2(vector<T,N> x);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int, let M : int> __intrinsic_op matrix<T,N,M> log2(matrix<T,N,M> x);\n";
sb << "\n";
sb << "// multiply-add\n";
sb << "__generic<T : __BuiltinArithmeticType> __intrinsic_op T mad(T mvalue, T avalue, T bvalue);\n";
sb << "__generic<T : __BuiltinArithmeticType, let N : int> __intrinsic_op vector<T,N> mad(vector<T,N> mvalue, vector<T,N> avalue, vector<T,N> bvalue);\n";
sb << "__generic<T : __BuiltinArithmeticType, let N : int, let M : int> __intrinsic_op matrix<T,N,M> mad(matrix<T,N,M> mvalue, matrix<T,N,M> avalue, matrix<T,N,M> bvalue);\n";
sb << "\n";
sb << "// maximum\n";
sb << "__generic<T : __BuiltinArithmeticType> __intrinsic_op T max(T x, T y);\n";
sb << "__generic<T : __BuiltinArithmeticType, let N : int> __intrinsic_op vector<T,N> max(vector<T,N> x, vector<T,N> y);\n";
sb << "__generic<T : __BuiltinArithmeticType, let N : int, let M : int> __intrinsic_op matrix<T,N,M> max(matrix<T,N,M> x, matrix<T,N,M> y);\n";
sb << "\n";
sb << "// minimum\n";
sb << "__generic<T : __BuiltinArithmeticType> __intrinsic_op T min(T x, T y);\n";
sb << "__generic<T : __BuiltinArithmeticType, let N : int> __intrinsic_op vector<T,N> min(vector<T,N> x, vector<T,N> y);\n";
sb << "__generic<T : __BuiltinArithmeticType, let N : int, let M : int> __intrinsic_op matrix<T,N,M> min(matrix<T,N,M> x, matrix<T,N,M> y);\n";
sb << "\n";
sb << "// split into integer and fractional parts (both with same sign)\n";
sb << "__generic<T : __BuiltinFloatingPointType> __intrinsic_op T modf(T x, out T ip);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int> __intrinsic_op vector<T,N> modf(vector<T,N> x, out vector<T,N> ip);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int, let M : int> __intrinsic_op matrix<T,N,M> modf(matrix<T,N,M> x, out matrix<T,N,M> ip);\n";
sb << "\n";
sb << "// msad4 (whatever that is)\n";
sb << "__intrinsic_op uint4 msad4(uint reference, uint2 source, uint4 accum);\n";
sb << "\n";
sb << "// General inner products\n";
sb << "\n";
sb << "// scalar-scalar\n";
sb << "__generic<T : __BuiltinArithmeticType> __intrinsic_op T mul(T x, T y);\n";
sb << "\n";
sb << "// scalar-vector and vector-scalar\n";
sb << "__generic<T : __BuiltinArithmeticType, let N : int> __intrinsic_op vector<T,N> mul(vector<T,N> x, T y);\n";
sb << "__generic<T : __BuiltinArithmeticType, let N : int> __intrinsic_op vector<T,N> mul(T x, vector<T,N> y);\n";
sb << "\n";
sb << "// scalar-matrix and matrix-scalar\n";
sb << "__generic<T : __BuiltinArithmeticType, let N : int, let M :int> __intrinsic_op matrix<T,N,M> mul(matrix<T,N,M> x, T y);\n";
sb << "__generic<T : __BuiltinArithmeticType, let N : int, let M :int> __intrinsic_op matrix<T,N,M> mul(T x, matrix<T,N,M> y);\n";
sb << "\n";
sb << "// vector-vector (dot product)\n";
sb << "__generic<T : __BuiltinArithmeticType, let N : int> __intrinsic_op(dot) T mul(vector<T,N> x, vector<T,N> y);\n";
sb << "\n";
sb << "// vector-matrix\n";
sb << "__generic<T : __BuiltinArithmeticType, let N : int, let M : int> __intrinsic_op(mulVectorMatrix) vector<T,M> mul(vector<T,N> x, matrix<T,N,M> y);\n";
sb << "\n";
sb << "// matrix-vector\n";
sb << "__generic<T : __BuiltinArithmeticType, let N : int, let M : int> __intrinsic_op(mulMatrixVector) vector<T,N> mul(matrix<T,N,M> x, vector<T,M> y);\n";
sb << "\n";
sb << "// matrix-matrix\n";
sb << "__generic<T : __BuiltinArithmeticType, let R : int, let N : int, let C : int> __intrinsic_op(mulMatrixMatrix) matrix<T,R,C> mul(matrix<T,R,N> x, matrix<T,N,C> y);\n";
sb << "\n";
sb << "// noise (deprecated)\n";
sb << "__intrinsic_op float noise(float x);\n";
sb << "__generic<let N : int> __intrinsic_op float noise(vector<float, N> x);\n";
sb << "\n";
sb << "// Normalize a vector\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int> __intrinsic_op vector<T,N> normalize(vector<T,N> x);\n";
sb << "\n";
sb << "// Raise to a power\n";
sb << "__generic<T : __BuiltinFloatingPointType> __intrinsic_op T pow(T x, T y);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int> __intrinsic_op vector<T,N> pow(vector<T,N> x, vector<T,N> y);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int, let M : int> __intrinsic_op matrix<T,N,M> pow(matrix<T,N,M> x, matrix<T,N,M> y);\n";
sb << "\n";
sb << "// Output message\n";
sb << "\n";
sb << "// __intrinsic_op void printf( string format, ... );\n";
sb << "\n";
sb << "// Tessellation factor fixup routines\n";
sb << "\n";
sb << "__intrinsic_op void Process2DQuadTessFactorsAvg(\n";
sb << "    in  float4 RawEdgeFactors,\n";
sb << "    in  float2 InsideScale,\n";
sb << "    out float4 RoundedEdgeTessFactors,\n";
sb << "    out float2 RoundedInsideTessFactors,\n";
sb << "    out float2 UnroundedInsideTessFactors);\n";
sb << "\n";
sb << "__intrinsic_op void Process2DQuadTessFactorsMax(\n";
sb << "    in  float4 RawEdgeFactors,\n";
sb << "    in  float2 InsideScale,\n";
sb << "    out float4 RoundedEdgeTessFactors,\n";
sb << "    out float2 RoundedInsideTessFactors,\n";
sb << "    out float2 UnroundedInsideTessFactors);\n";
sb << "\n";
sb << "__intrinsic_op void Process2DQuadTessFactorsMin(\n";
sb << "    in  float4 RawEdgeFactors,\n";
sb << "    in  float2 InsideScale,\n";
sb << "    out float4 RoundedEdgeTessFactors,\n";
sb << "    out float2 RoundedInsideTessFactors,\n";
sb << "    out float2 UnroundedInsideTessFactors);\n";
sb << "\n";
sb << "__intrinsic_op void ProcessIsolineTessFactors(\n";
sb << "    in  float RawDetailFactor,\n";
sb << "    in  float RawDensityFactor,\n";
sb << "    out float RoundedDetailFactor,\n";
sb << "    out float RoundedDensityFactor);\n";
sb << "\n";
sb << "__intrinsic_op void ProcessQuadTessFactorsAvg(\n";
sb << "    in  float4 RawEdgeFactors,\n";
sb << "    in  float InsideScale,\n";
sb << "    out float4 RoundedEdgeTessFactors,\n";
sb << "    out float2 RoundedInsideTessFactors,\n";
sb << "    out float2 UnroundedInsideTessFactors);\n";
sb << "\n";
sb << "__intrinsic_op void ProcessQuadTessFactorsMax(\n";
sb << "    in  float4 RawEdgeFactors,\n";
sb << "    in  float InsideScale,\n";
sb << "    out float4 RoundedEdgeTessFactors,\n";
sb << "    out float2 RoundedInsideTessFactors,\n";
sb << "    out float2 UnroundedInsideTessFactors);\n";
sb << "\n";
sb << "__intrinsic_op void ProcessQuadTessFactorsMin(\n";
sb << "    in  float4 RawEdgeFactors,\n";
sb << "    in  float InsideScale,\n";
sb << "    out float4 RoundedEdgeTessFactors,\n";
sb << "    out float2 RoundedInsideTessFactors,\n";
sb << "    out float2 UnroundedInsideTessFactors);\n";
sb << "\n";
sb << "__intrinsic_op void ProcessTriTessFactorsAvg(\n";
sb << "    in  float3 RawEdgeFactors,\n";
sb << "    in  float InsideScale,\n";
sb << "    out float3 RoundedEdgeTessFactors,\n";
sb << "    out float RoundedInsideTessFactor,\n";
sb << "    out float UnroundedInsideTessFactor);\n";
sb << "\n";
sb << "__intrinsic_op void ProcessTriTessFactorsMax(\n";
sb << "    in  float3 RawEdgeFactors,\n";
sb << "    in  float InsideScale,\n";
sb << "    out float3 RoundedEdgeTessFactors,\n";
sb << "    out float RoundedInsideTessFactor,\n";
sb << "    out float UnroundedInsideTessFactor);\n";
sb << "\n";
sb << "__intrinsic_op void ProcessTriTessFactorsMin(\n";
sb << "    in  float3 RawEdgeFactors,\n";
sb << "    in  float InsideScale,\n";
sb << "    out float3 RoundedEdgeTessFactors,\n";
sb << "    out float RoundedInsideTessFactors,\n";
sb << "    out float UnroundedInsideTessFactors);\n";
sb << "\n";
sb << "// Degrees to radians\n";
sb << "__generic<T : __BuiltinFloatingPointType> __intrinsic_op T radians(T x);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int> __intrinsic_op vector<T,N> radians(vector<T,N> x);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int, let M : int> __intrinsic_op matrix<T,N,M> radians(matrix<T,N,M> x);\n";
sb << "\n";
sb << "// Approximate reciprocal\n";
sb << "__generic<T : __BuiltinFloatingPointType> __intrinsic_op T rcp(T x);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int> __intrinsic_op vector<T,N> rcp(vector<T,N> x);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int, let M : int> __intrinsic_op matrix<T,N,M> rcp(matrix<T,N,M> x);\n";
sb << "\n";
sb << "// Reflect incident vector across plane with given normal\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int>\n";
sb << "__intrinsic_op\n";
sb << "vector<T,N> reflect(vector<T,N> i, vector<T,N> n);\n";
sb << "\n";
sb << "// Refract incident vector given surface normal and index of refraction\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int>\n";
sb << "__intrinsic_op\n";
sb << "vector<T,N> refract(vector<T,N> i, vector<T,N> n, float eta);\n";
sb << "\n";
sb << "// Reverse order of bits\n";
sb << "__intrinsic_op uint reversebits(uint value);\n";
sb << "__generic<let N : int> vector<uint,N> reversebits(vector<uint,N> value);\n";
sb << "\n";
sb << "// Round-to-nearest\n";
sb << "__generic<T : __BuiltinFloatingPointType> __intrinsic_op T round(T x);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int> __intrinsic_op vector<T,N> round(vector<T,N> x);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int, let M : int> __intrinsic_op matrix<T,N,M> round(matrix<T,N,M> x);\n";
sb << "\n";
sb << "// Reciprocal of square root\n";
sb << "__generic<T : __BuiltinFloatingPointType> __intrinsic_op T rsqrt(T x);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int> __intrinsic_op vector<T,N> rsqrt(vector<T,N> x);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int, let M : int> __intrinsic_op matrix<T,N,M> rsqrt(matrix<T,N,M> x);\n";
sb << "\n";
sb << "// Clamp value to [0,1] range\n";
sb << "__generic<T : __BuiltinFloatingPointType>\n";
sb << "__target_intrinsic(glsl, \"clamp($0, 0, 1)\") __intrinsic_op\n";
sb << "T saturate(T x);\n";
sb << "\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int>\n";
sb << "__target_intrinsic(glsl, \"clamp($0, 0, 1)\") __intrinsic_op\n";
sb << "vector<T,N> saturate(vector<T,N> x);\n";
sb << "\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int, let M : int>\n";
sb << "__target_intrinsic(glsl, \"clamp($0, 0, 1)\") __intrinsic_op\n";
sb << "matrix<T,N,M> saturate(matrix<T,N,M> x);\n";
sb << "\n";
sb << "\n";
sb << "// Extract sign of value\n";
sb << "__generic<T : __BuiltinSignedArithmeticType> __intrinsic_op int sign(T x);\n";
sb << "__generic<T : __BuiltinSignedArithmeticType, let N : int> __intrinsic_op vector<int,N> sign(vector<T,N> x);\n";
sb << "__generic<T : __BuiltinSignedArithmeticType, let N : int, let M : int> __intrinsic_op matrix<int,N,M> sign(matrix<T,N,M> x);\n";
sb << "\n";
sb << "\n";
sb << "// Sine\n";
sb << "__generic<T : __BuiltinFloatingPointType> __intrinsic_op T sin(T x);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int> __intrinsic_op vector<T,N> sin(vector<T,N> x);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int, let M : int> __intrinsic_op matrix<T,N,M> sin(matrix<T,N,M> x);\n";
sb << "\n";
sb << "// Sine and cosine\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int> __intrinsic_op void sincos(T x, out T s, out T c);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int> __intrinsic_op void sincos(vector<T,N> x, out vector<T,N> s, out vector<T,N> c);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int, let M : int> __intrinsic_op void sincos(matrix<T,N,M> x, out matrix<T,N,M> s, out matrix<T,N,M> c);\n";
sb << "\n";
sb << "// Hyperbolic Sine\n";
sb << "__generic<T : __BuiltinFloatingPointType> __intrinsic_op T sinh(T x);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int> __intrinsic_op vector<T,N> sinh(vector<T,N> x);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int, let M : int> __intrinsic_op matrix<T,N,M> sinh(matrix<T,N,M> x);\n";
sb << "\n";
sb << "// Smooth step (Hermite interpolation)\n";
sb << "__generic<T : __BuiltinFloatingPointType> __intrinsic_op T smoothstep(T min, T max, T x);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int> __intrinsic_op vector<T,N> smoothstep(vector<T,N> min, vector<T,N> max, vector<T,N> x);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int, let M : int> __intrinsic_op matrix<T,N,M> smoothstep(matrix<T,N,M> min, matrix<T,N,M> max, matrix<T,N,M> x);\n";
sb << "\n";
sb << "// Square root\n";
sb << "__generic<T : __BuiltinFloatingPointType> __intrinsic_op T sqrt(T x);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int> __intrinsic_op vector<T,N> sqrt(vector<T,N> x);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int, let M : int> __intrinsic_op matrix<T,N,M> sqrt(matrix<T,N,M> x);\n";
sb << "\n";
sb << "// Step function\n";
sb << "__generic<T : __BuiltinFloatingPointType> __intrinsic_op T step(T y, T x);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int> __intrinsic_op vector<T,N> step(vector<T,N> y, vector<T,N> x);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int, let M : int> __intrinsic_op matrix<T,N,M> step(matrix<T,N,M> y, matrix<T,N,M> x);\n";
sb << "\n";
sb << "// Tangent\n";
sb << "__generic<T : __BuiltinFloatingPointType> __intrinsic_op T tan(T x);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int> __intrinsic_op vector<T,N> tan(vector<T,N> x);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int, let M : int> __intrinsic_op matrix<T,N,M> tan(matrix<T,N,M> x);\n";
sb << "\n";
sb << "// Hyperbolic tangent\n";
sb << "__generic<T : __BuiltinFloatingPointType> __intrinsic_op T tanh(T x);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int> __intrinsic_op vector<T,N> tanh(vector<T,N> x);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int, let M : int> __intrinsic_op matrix<T,N,M> tanh(matrix<T,N,M> x);\n";
sb << "\n";
sb << "// Legacy texture-fetch operations\n";
sb << "\n";
sb << "/*\n";
sb << "__intrinsic_op float4 tex1D(sampler1D s, float t);\n";
sb << "__intrinsic_op float4 tex1D(sampler1D s, float t, float ddx, float ddy);\n";
sb << "__intrinsic_op float4 tex1Dbias(sampler1D s, float4 t);\n";
sb << "__intrinsic_op float4 tex1Dgrad(sampler1D s, float t, float ddx, float ddy);\n";
sb << "__intrinsic_op float4 tex1Dlod(sampler1D s, float4 t);\n";
sb << "__intrinsic_op float4 tex1Dproj(sampler1D s, float4 t);\n";
sb << "\n";
sb << "__intrinsic_op float4 tex2D(sampler2D s, float2 t);\n";
sb << "__intrinsic_op float4 tex2D(sampler2D s, float2 t, float2 ddx, float2 ddy);\n";
sb << "__intrinsic_op float4 tex2Dbias(sampler2D s, float4 t);\n";
sb << "__intrinsic_op float4 tex2Dgrad(sampler2D s, float2 t, float2 ddx, float2 ddy);\n";
sb << "__intrinsic_op float4 tex2Dlod(sampler2D s, float4 t);\n";
sb << "__intrinsic_op float4 tex2Dproj(sampler2D s, float4 t);\n";
sb << "\n";
sb << "__intrinsic_op float4 tex3D(sampler3D s, float3 t);\n";
sb << "__intrinsic_op float4 tex3D(sampler3D s, float3 t, float3 ddx, float3 ddy);\n";
sb << "__intrinsic_op float4 tex3Dbias(sampler3D s, float4 t);\n";
sb << "__intrinsic_op float4 tex3Dgrad(sampler3D s, float3 t, float3 ddx, float3 ddy);\n";
sb << "__intrinsic_op float4 tex3Dlod(sampler3D s, float4 t);\n";
sb << "__intrinsic_op float4 tex3Dproj(sampler3D s, float4 t);\n";
sb << "\n";
sb << "__intrinsic_op float4 texCUBE(samplerCUBE s, float3 t);\n";
sb << "__intrinsic_op float4 texCUBE(samplerCUBE s, float3 t, float3 ddx, float3 ddy);\n";
sb << "__intrinsic_op float4 texCUBEbias(samplerCUBE s, float4 t);\n";
sb << "__intrinsic_op float4 texCUBEgrad(samplerCUBE s, float3 t, float3 ddx, float3 ddy);\n";
sb << "__intrinsic_op float4 texCUBElod(samplerCUBE s, float4 t);\n";
sb << "__intrinsic_op float4 texCUBEproj(samplerCUBE s, float4 t);\n";
sb << "*/\n";
sb << "\n";
sb << "// Matrix transpose\n";
sb << "__generic<T : __BuiltinType, let N : int, let M : int> __intrinsic_op matrix<T,M,N> transpose(matrix<T,N,M> x);\n";
sb << "\n";
sb << "// Truncate to integer\n";
sb << "__generic<T : __BuiltinFloatingPointType> __intrinsic_op T trunc(T x);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int> __intrinsic_op vector<T,N> trunc(vector<T,N> x);\n";
sb << "__generic<T : __BuiltinFloatingPointType, let N : int, let M : int> __intrinsic_op matrix<T,N,M> trunc(matrix<T,N,M> x);\n";
sb << "\n";
sb << "// Shader model 6.0 stuff\n";
sb << "\n";
sb << "__intrinsic_op uint GlobalOrderedCountIncrement(uint countToAppendForThisLane);\n";
sb << "\n";
sb << "__generic<T : __BuiltinType> __intrinsic_op T QuadReadLaneAt(T sourceValue, int quadLaneID);\n";
sb << "__generic<T : __BuiltinType, let N : int> __intrinsic_op vector<T,N> QuadReadLaneAt(vector<T,N> sourceValue, int quadLaneID);\n";
sb << "__generic<T : __BuiltinType, let N : int, let M : int> __intrinsic_op matrix<T,N,M> QuadReadLaneAt(matrix<T,N,M> sourceValue, int quadLaneID);\n";
sb << "\n";
sb << "__generic<T : __BuiltinType> __intrinsic_op T QuadSwapX(T localValue);\n";
sb << "__generic<T : __BuiltinType, let N : int> __intrinsic_op vector<T,N> QuadSwapX(vector<T,N> localValue);\n";
sb << "__generic<T : __BuiltinType, let N : int, let M : int> __intrinsic_op matrix<T,N,M> QuadSwapX(matrix<T,N,M> localValue);\n";
sb << "\n";
sb << "__generic<T : __BuiltinType> __intrinsic_op T QuadSwapY(T localValue);\n";
sb << "__generic<T : __BuiltinType, let N : int> __intrinsic_op vector<T,N> QuadSwapY(vector<T,N> localValue);\n";
sb << "__generic<T : __BuiltinType, let N : int, let M : int> __intrinsic_op matrix<T,N,M> QuadSwapY(matrix<T,N,M> localValue);\n";
sb << "\n";
sb << "__generic<T : __BuiltinIntegerType> __intrinsic_op T WaveAllBitAnd(T expr);\n";
sb << "__generic<T : __BuiltinIntegerType, let N : int> __intrinsic_op vector<T,N> WaveAllBitAnd(vector<T,N> expr);\n";
sb << "__generic<T : __BuiltinIntegerType, let N : int, let M : int> __intrinsic_op matrix<T,N,M> WaveAllBitAnd(matrix<T,N,M> expr);\n";
sb << "\n";
sb << "__generic<T : __BuiltinIntegerType> __intrinsic_op T WaveAllBitOr(T expr);\n";
sb << "__generic<T : __BuiltinIntegerType, let N : int> __intrinsic_op vector<T,N> WaveAllBitOr(vector<T,N> expr);\n";
sb << "__generic<T : __BuiltinIntegerType, let N : int, let M : int> __intrinsic_op matrix<T,N,M> WaveAllBitOr(matrix<T,N,M> expr);\n";
sb << "\n";
sb << "__generic<T : __BuiltinIntegerType> __intrinsic_op T WaveAllBitXor(T expr);\n";
sb << "__generic<T : __BuiltinIntegerType, let N : int> __intrinsic_op vector<T,N> WaveAllBitXor(vector<T,N> expr);\n";
sb << "__generic<T : __BuiltinIntegerType, let N : int, let M : int> __intrinsic_op matrix<T,N,M> WaveAllBitXor(matrix<T,N,M> expr);\n";
sb << "\n";
sb << "__generic<T : __BuiltinArithmeticType> __intrinsic_op T WaveAllMax(T expr);\n";
sb << "__generic<T : __BuiltinArithmeticType, let N : int> __intrinsic_op vector<T,N> WaveAllMax(vector<T,N> expr);\n";
sb << "__generic<T : __BuiltinArithmeticType, let N : int, let M : int> __intrinsic_op matrix<T,N,M> WaveAllMax(matrix<T,N,M> expr);\n";
sb << "\n";
sb << "__generic<T : __BuiltinArithmeticType> __intrinsic_op T WaveAllMin(T expr);\n";
sb << "__generic<T : __BuiltinArithmeticType, let N : int> __intrinsic_op vector<T,N> WaveAllMin(vector<T,N> expr);\n";
sb << "__generic<T : __BuiltinArithmeticType, let N : int, let M : int> __intrinsic_op matrix<T,N,M> WaveAllMin(matrix<T,N,M> expr);\n";
sb << "\n";
sb << "__generic<T : __BuiltinArithmeticType> __intrinsic_op T WaveAllProduct(T expr);\n";
sb << "__generic<T : __BuiltinArithmeticType, let N : int> __intrinsic_op vector<T,N> WaveAllProduct(vector<T,N> expr);\n";
sb << "__generic<T : __BuiltinArithmeticType, let N : int, let M : int> __intrinsic_op matrix<T,N,M> WaveAllProduct(matrix<T,N,M> expr);\n";
sb << "\n";
sb << "__generic<T : __BuiltinArithmeticType> __intrinsic_op T WaveAllSum(T expr);\n";
sb << "__generic<T : __BuiltinArithmeticType, let N : int> __intrinsic_op vector<T,N> WaveAllSum(vector<T,N> expr);\n";
sb << "__generic<T : __BuiltinArithmeticType, let N : int, let M : int> __intrinsic_op matrix<T,N,M> WaveAllSum(matrix<T,N,M> expr);\n";
sb << "\n";
sb << "__intrinsic_op bool WaveAllEqual(bool expr);\n";
sb << "__intrinsic_op bool WaveAllTrue(bool expr);\n";
sb << "__intrinsic_op bool WaveAnyTrue(bool expr);\n";
sb << "\n";
sb << "uint64_t WaveBallot(bool expr);\n";
sb << "\n";
sb << "uint WaveGetLaneCount();\n";
sb << "uint WaveGetLaneIndex();\n";
sb << "uint WaveGetOrderedIndex();\n";
sb << "\n";
sb << "bool WaveIsHelperLane();\n";
sb << "\n";
sb << "bool WaveOnce();\n";
sb << "\n";
sb << "__generic<T : __BuiltinArithmeticType> __intrinsic_op T WavePrefixProduct(T expr);\n";
sb << "__generic<T : __BuiltinArithmeticType, let N : int> __intrinsic_op vector<T,N> WavePrefixProduct(vector<T,N> expr);\n";
sb << "__generic<T : __BuiltinArithmeticType, let N : int, let M : int> __intrinsic_op matrix<T,N,M> WavePrefixProduct(matrix<T,N,M> expr);\n";
sb << "\n";
sb << "__generic<T : __BuiltinArithmeticType> __intrinsic_op T WavePrefixSum(T expr);\n";
sb << "__generic<T : __BuiltinArithmeticType, let N : int> __intrinsic_op vector<T,N> WavePrefixSum(vector<T,N> expr);\n";
sb << "__generic<T : __BuiltinArithmeticType, let N : int, let M : int> __intrinsic_op matrix<T,N,M> WavePrefixSum(matrix<T,N,M> expr);\n";
sb << "\n";
sb << "__generic<T : __BuiltinType> __intrinsic_op T WaveReadFirstLane(T expr);\n";
sb << "__generic<T : __BuiltinType, let N : int> __intrinsic_op vector<T,N> WaveReadFirstLane(vector<T,N> expr);\n";
sb << "__generic<T : __BuiltinType, let N : int, let M : int> __intrinsic_op matrix<T,N,M> WaveReadFirstLane(matrix<T,N,M> expr);\n";
sb << "\n";
sb << "__generic<T : __BuiltinType> __intrinsic_op T WaveReadLaneAt(T expr, int laneIndex);\n";
sb << "__generic<T : __BuiltinType, let N : int> __intrinsic_op vector<T,N> WaveReadLaneAt(vector<T,N> expr, int laneIndex);\n";
sb << "__generic<T : __BuiltinType, let N : int, let M : int> __intrinsic_op matrix<T,N,M> WaveReadLaneAt(matrix<T,N,M> expr, int laneIndex);\n";
sb << "\n";
sb << "// `typedef`s to help with the fact that HLSL has been sorta-kinda case insensitive at various points\n";
sb << "typedef Texture2D texture2D;\n";
sb << "\n";
sb << "";


// Component-wise multiplication ops
for(auto op : binaryOps)
{
    switch (op.opCode)
    {
    default:
        continue;

    case kIROp_Mul:
    case kIRPseudoOp_MulAssign:
        break;
    }

    for (auto type : kBaseTypes)
    {
        if ((type.flags & op.flags) == 0)
            continue;

        char const* leftType = type.name;
        char const* rightType = leftType;
        char const* resultType = leftType;

        char const* leftQual = "";
        if(op.flags & ASSIGNMENT) leftQual = "in out ";

        sb << "__generic<let N : int, let M : int> ";
        sb << "__intrinsic_op(" << int(op.opCode) << ") matrix<" << resultType << ",N,M> operator" << op.opName << "(" << leftQual << "matrix<" << leftType << ",N,M> left, matrix<" << rightType << ",N,M> right);\n";
    }
}

//

// Buffer types

static const struct {
    char const*         name;
    SlangResourceAccess access;
} kBaseBufferAccessLevels[] = {
    { "",                   SLANG_RESOURCE_ACCESS_READ },
    { "RW",                 SLANG_RESOURCE_ACCESS_READ_WRITE },
    { "RasterizerOrdered",  SLANG_RESOURCE_ACCESS_RASTER_ORDERED },
};
static const int kBaseBufferAccessLevelCount = sizeof(kBaseBufferAccessLevels) / sizeof(kBaseBufferAccessLevels[0]);

for (int aa = 0; aa < kBaseBufferAccessLevelCount; ++aa)
{

    sb << "__generic<T> __magic_type(Texture, ";
    sb << ResourceType::makeFlavor(ResourceType::Shape::ShapeBuffer, kBaseBufferAccessLevels[aa].access);
    sb << ") struct ";
    sb << kBaseBufferAccessLevels[aa].name;
    sb << "Buffer {\n";

    sb << "__intrinsic_op void GetDimensions(out uint dim);\n";

    sb << "__target_intrinsic(glsl, \"texelFetch($P, $0)$z\")\n";
    sb << "__intrinsic_op T Load(int location);\n";

    sb << "__intrinsic_op T Load(int location, out uint status);\n";

    sb << "__target_intrinsic(glsl, \"texelFetch($P, int($0))$z\")\n";
    sb << "__intrinsic_op __subscript(uint index) -> T";

    if (kBaseBufferAccessLevels[aa].access != SLANG_RESOURCE_ACCESS_READ)
    {
        sb << " { get; set; }\n";
    }
    else
    {
        sb << ";\n";
    }

    sb << "};\n";
}

sb << "";