Implement the WaveMultiPrefixBitAnd HLSL Function

[farzonl]

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

OpGroupNonUniformBitwiseAnd:

Description:

A bitwise and group operation of all Value
operands contributed by active invocations in the
group.

Result Type must be a scalar or vector of integer
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.

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	359	<id> Result Type	Result <id>	Scope <id> Execution	Group Operation Operation	<id> Value	Optional <id> ClusterSize

Test Case(s)

Example 1

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

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

HLSL:

Syntax

any_int<> WaveMultiPrefixBitAnd(any_int<> value, uint<4> mask);

Type Description

Name	Template Type	Component Type	Size
ret	scalar, vector, or matrix	int, int16_t, int64_t, uint, uint16_t, or uint64_t	any
value	scalar, vector, or matrix	int, int16_t, int64_t, uint, uint16_t, or uint64_t	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