Implement the `and` HLSL Function

[farzonl]

• Implement the and api in hlsl_intrinsics.h
  - [ ] Investigate to see if we can use __builtin_reduce_and
• If a clang builtin is needed add sema checks for and to CheckHLSLBuiltinFunctionCall in SemaHLSL.cpp
• If codegen is needed, add codegen for and to EmitHLSLBuiltinExpr in CGBuiltin.cpp
• Add codegen tests to clang/test/CodeGenHLSL/builtins/and.hlsl
• Add sema tests to clang/test/SemaHLSL/BuiltIns/and-errors.hlsl

DirectX

There were no DXIL opcodes found for and.

SPIR-V

There were no SPIR-V opcodes found for and.

Test Case(s)

Example 1

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

export bool4 fn(bool4 p1, bool4 p2) {
    return and(p1, p2);
}

HLSL:

Logically ands a vector and produces a bool vector output

Syntax

bool<> and(bool<> x, bool<> y);

Type Description

Name	Template Type	Component Type	Size
ret	scalar, vector, or matrix	bool	any
x	scalar, vector, or matrix	bool	any
y	scalar, vector, or matrix	bool	any

Minimum Shader Model

This function is supported in the following shader models.

Shader Model	Supported
All Shader Models with HLSL 2021 Language Mode	yes

Shader Stages

All Shader Stages

Remarks

In HLSL 2021 and(X, Y); is a replacement for

int3 X = {1, 1, 1};
int3 Y = {0, 0, 0};
bool3 Cond = X && Y;

See also

• Intrinsic Functions (DirectX HLSL)
• See HLSL 2021 Logical operation short-circuiting for scalars

[Icohedron]

I'll implement it!