Implement the WaveMultiPrefixSum HLSL Function

[farzonl]

• Implement WaveMultiPrefixSum clang builtin,
• Link WaveMultiPrefixSum clang builtin with hlsl_intrinsics.h
• Add sema checks for WaveMultiPrefixSum to CheckHLSLBuiltinFunctionCall in SemaChecking.cpp
• Add codegen for WaveMultiPrefixSum to EmitHLSLBuiltinExpr in CGBuiltin.cpp
• Add codegen tests to clang/test/CodeGenHLSL/builtins/WaveMultiPrefixSum.hlsl
• Add sema tests to clang/test/SemaHLSL/BuiltIns/WaveMultiPrefixSum-errors.hlsl
• Create the int_dx_WaveMultiPrefixSum intrinsic in IntrinsicsDirectX.td
• Create the DXILOpMapping of int_dx_WaveMultiPrefixSum to 166 in DXIL.td
• Create the WaveMultiPrefixSum.ll and WaveMultiPrefixSum_errors.ll tests in llvm/test/CodeGen/DirectX/
• Create the int_spv_WaveMultiPrefixSum intrinsic in IntrinsicsSPIRV.td
• In SPIRVInstructionSelector.cpp create the WaveMultiPrefixSum lowering and map it to int_spv_WaveMultiPrefixSum in SPIRVInstructionSelector::selectIntrinsic.
• Create SPIR-V backend test case in llvm/test/CodeGen/SPIRV/hlsl-intrinsics/WaveMultiPrefixSum.ll

DirectX

DXIL Opcode	DXIL OpName	Shader Model	Shader Stages
166	WaveMultiPrefixOp	6.5	('library', 'compute', 'amplification', 'mesh', 'pixel', 'vertex', 'hull', 'domain', 'geometry', 'raygeneration', 'intersection', 'anyhit', 'closesthit', 'miss', 'callable', 'node')

SPIR-V

OpGroupNonUniformFAdd:

Description:

A floating point add group operation of all Value
operands contributed by active invocations in the
group.

Result Type must be a scalar or vector of floating-point
type.

Execution is a Scope that identifies the group of
invocations affected by this command. It must be Subgroup.

The identity I for Operation is 0. If Operation is
ClusteredReduce, ClusterSize must be present.

The type of Value must be the same as Result Type. The method used
to perform the group operation on the contributed Value(s) from active
invocations is implementation defined.

ClusterSize is the size of cluster to use. ClusterSize must be a
scalar of integer type, whose Signedness operand is 0.
ClusterSize must come from a constant
instruction. Behavior is undefined unless
ClusterSize is at least 1 and a power of 2. If ClusterSize is
greater than the size of the group, executing this instruction
results in undefined behavior.

Capability:
GroupNonUniformArithmetic, GroupNonUniformClustered,
GroupNonUniformPartitionedNV

Missing before version 1.3.

Word Count	Opcode	Results	Operands
6 + variable	350	<id> Result Type	Result <id>	Scope <id> Execution	Group Operation Operation	<id> Value	Optional <id> ClusterSize

Test Case(s)

Example 1

//dxc WaveMultiPrefixSum_test.hlsl -T lib_6_8 -enable-16bit-types -O0

export float4 fn(float4 p1, uint4 p2) {
    return WaveMultiPrefixSum(p1, p2);
}

Example 2

//dxc WaveMultiPrefixSum_1_test.hlsl -T lib_6_8 -enable-16bit-types -O0

export uint4 fn(uint4 p1, uint4 p2) {
    return WaveMultiPrefixSum(p1, p2);
}

Example 3

//dxc WaveMultiPrefixSum_2_test.hlsl -T lib_6_8 -enable-16bit-types -O0

export int4 fn(int4 p1, uint4 p2) {
    return WaveMultiPrefixSum(p1, p2);
}

HLSL:

Syntax

numeric<> WaveMultiPrefixSum(numeric<> value, uint<4> mask);

Type Description

Name	Template Type	Component Type	Size
ret	scalar, vector, or matrix	float or int	any
value	scalar, vector, or matrix	float or int	any
mask	vector	uint	4

Minimum Shader Model

This function is supported in the following shader models.

Shader Model	Supported
Shader Model 6.5 and higher shader models	yes

Shader Stages

• Library Shader
• Compute Shader
• Amplification Shader
• Mesh Shader
• Pixel Shader
• Vertex Shader
• Hull Shader
• Domain Shader
• Geometry Shader
• Raygeneration Shader
• Intersection Shader
• Anyhit Shader
• Closesthit Shader
• Miss Shader
• Callable Shader
• Node Shader

See also

• Intrinsic Functions (DirectX HLSL)
• See WaveMultiPrefix*() Functions