Implement the `normalize` HLSL Function

[farzonl]

• Implement normalize clang builtin,
• Link normalize clang builtin with hlsl_intrinsics.h
• Add sema checks for normalize to CheckHLSLBuiltinFunctionCall in SemaChecking.cpp
• Add codegen for normalize to EmitHLSLBuiltinExpr in CGBuiltin.cpp
• Add codegen tests to clang/test/CodeGenHLSL/builtins/normalize.hlsl
• Add sema tests to clang/test/SemaHLSL/BuiltIns/normalize-errors.hlsl
• Create the int_dx_normalize intrinsic in IntrinsicsDirectX.td
• Create an intrinsic expansion int_dx_normalize in llvm/lib/Target/DirectX/DXILIntrinsicExpansion.cpp
• Create the normalize.ll and normalize_errors.ll tests in llvm/test/CodeGen/DirectX/
• Create the int_spv_normalize intrinsic in IntrinsicsSPIRV.td
• In SPIRVInstructionSelector.cpp create the normalize lowering and map it to int_spv_normalize in SPIRVInstructionSelector::selectIntrinsic.
• Create SPIR-V backend test case in llvm/test/CodeGen/SPIRV/hlsl-intrinsics/normalize.ll

DirectX

DXIL Opcode	DXIL OpName	Shader Model	Shader Stages
56	Dot4	6.0	()

SPIR-V

Normalize:

Description:

Normalize

Result is the vector in the same direction as x but with a length of
1.

The operand x must be a scalar or vector whose component type is
floating-point.

Result Type and the type of x must be the same type.

Number	Operand 1	Operand 2	Operand 3	Operand 4
69	<id> x

Test Case(s)

Example 1

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

export float4 fn(float4 p1) {
    return normalize(p1);
}

HLSL:

Normalizes the specified floating-point vector according to x / length(x).

ret normalize(x)

Parameters

Item	Description
x	[in] The specified floating-point vector.

Return Value

The normalized x parameter. If the length of the x parameter is 0, the result is indefinite.

Remarks

The normalize HLSL intrinsic function uses the following formula: x / length(x).

Type Description

Name	Template Type	Component Type	Size
x	vector	float	any
ret	same as input x	float	same dimension(s) as input x

Minimum Shader Model

This function is supported in the following shader models.

Shader Model	Supported
Shader Model 2 (DirectX HLSL) and higher shader models	yes
Shader Model 1 (DirectX HLSL)	yes (vs_1_1 only)

See also

Intrinsic Functions (DirectX HLSL)

[llvmbot]

@llvm/issue-subscribers-clang-codegen

Author: Farzon Lotfi (farzonl)

Details

- [ ] Implement `normalize` clang builtin, - [ ] Link `normalize` clang builtin with `hlsl_intrinsics.h` - [ ] Add sema checks for `normalize` to `CheckHLSLBuiltinFunctionCall` in `SemaChecking.cpp` - [ ] Add codegen for `normalize` to `EmitHLSLBuiltinExpr` in `CGBuiltin.cpp` - [ ] Add codegen tests to `clang/test/CodeGenHLSL/builtins/normalize.hlsl` - [ ] Add sema tests to `clang/test/SemaHLSL/BuiltIns/normalize-errors.hlsl` - [ ] Create the `int_dx_normalize` intrinsic in `IntrinsicsDirectX.td` - [ ] Create an intrinsic expansion `int_dx_normalize` in `llvm/lib/Target/DirectX/DXILIntrinsicExpansion.cpp` - [ ] Create the `normalize.ll` and `normalize_errors.ll` tests in `llvm/test/CodeGen/DirectX/` - [ ] Create the `int_spv_normalize` intrinsic in `IntrinsicsSPIRV.td` - [ ] In SPIRVInstructionSelector.cpp create the `normalize` lowering and map it to `int_spv_normalize` in `SPIRVInstructionSelector::selectIntrinsic`. - [ ] Create SPIR-V backend test case in `llvm/test/CodeGen/SPIRV/hlsl-intrinsics/normalize.ll`

DirectX

DXIL Opcode	DXIL OpName	Shader Model	Shader Stages
56	Dot4	6.0	()

SPIR-V

Normalize:

Description:

Normalize

Result is the vector in the same direction as x but with a length of
1.

The operand x must be a scalar or vector whose component type is
floating-point.

Result Type and the type of x must be the same type.

<table>
<colgroup>
<col style="width: 20%" />
<col style="width: 20%" />
<col style="width: 20%" />
<col style="width: 20%" />
<col style="width: 20%" />
</colgroup>
<thead>
<tr>
<th>Number</th>
<th>Operand 1</th>
<th>Operand 2</th>
<th>Operand 3</th>
<th>Operand 4</th>
</tr>
</thead>
<tbody>
<tr>
<td class="tableblock halign-left valign-top"><p>69</p></td>
<td
class="tableblock halign-left valign-top"><p><em>&lt;id&gt;</em><br />
<em>x</em></p></td>
<td></td>
<td></td>
<td></td>
</tr>
</tbody>
</table>

Test Case(s)

Example 1

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

export float4 fn(float4 p1) {
    return normalize(p1);
}

HLSL:

Normalizes the specified floating-point vector according to x / length(x).

ret normalize(x)

Parameters

Item	Description
<span id="x"></span><span id="X"></span>x<br/>	[in] The specified floating-point vector.<br/>

Return Value

The normalized x parameter. If the length of the x parameter is 0, the result is indefinite.

Remarks

The normalize HLSL intrinsic function uses the following formula: x / length(x).

Type Description

Name	Template Type	Component Type	Size
x	vector	float	any
ret	same as input x	float	same dimension(s) as input x

Minimum Shader Model

This function is supported in the following shader models.

Shader Model	Supported
Shader Model 2 (DirectX HLSL) and higher shader models	yes
Shader Model 1 (DirectX HLSL)	yes (vs_1_1 only)

See also

<dl> <dt>

Intrinsic Functions (DirectX HLSL)
</dt> </dl>