Implement the step HLSL Function

[farzonl]

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

DirectX

There were no DXIL opcodes found for step.

SPIR-V

Step:

Description:

Step

Result is 0.0 if x < edge; otherwise result is 1.0.

The operands must all be a scalar or vector whose component type is
floating-point.

Result Type and the type of all operands must be the same type.
Results are computed per component.

Number	Operand 1	Operand 2	Operand 3	Operand 4
48	<id> edge	<id> x

Test Case(s)

Example 1

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

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

HLSL:

Compares two values, returning 0 or 1 based on which value is greater.

ret step(y, x)

Parameters

Item	Description
y	[in] The first floating-point value to compare.
x	[in] The second floating-point value to compare.

Return Value

1 if the x parameter is greater than or equal to the y parameter; otherwise, 0.

Remarks

This function uses the following formula: (x >= y) ? 1 : 0. The function returns either 0 or 1 depending on whether the x parameter is greater than the y parameter. To compute a smooth interpolation between 0 and 1, use the smoothstep HLSL intrinsic function.

Type Description

Name	Template Type	Component Type	Size
y	scalar, vector, or matrix	float	any
x	same as input y	float	same dimension(s) as input y
ret	same as input y	float	same dimension(s) as input y

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 and ps_1_4)

See also

Intrinsic Functions (DirectX HLSL)

[farzonl]

@damyanp Adding refinment priority. This is used in DirectML shader Lighting.fxh

[llvmbot]

@llvm/issue-subscribers-clang-codegen

Author: Farzon Lotfi (farzonl)

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

DirectX

There were no DXIL opcodes found for step.

SPIR-V

Step:

Description:

Step

Result is 0.0 if x &lt; edge; otherwise result is 1.0.

The operands must all be a scalar or vector whose component type is
floating-point.

Result Type and the type of all operands must be the same type.
Results are computed per component.

<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>48</p></td>
<td
class="tableblock halign-left valign-top"><p><em>&lt;id&gt;</em><br />
<em>edge</em></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>
</tr>
</tbody>
</table>

Test Case(s)

Example 1

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

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

HLSL:

Compares two values, returning 0 or 1 based on which value is greater.

ret step(y, x)

Parameters

Item	Description
<span id="y"></span><span id="Y"></span>y<br/>	[in] The first floating-point value to compare.<br/>
<span id="x"></span><span id="X"></span>x<br/>	[in] The second floating-point value to compare.<br/>

Return Value

1 if the x parameter is greater than or equal to the y parameter; otherwise, 0.

Remarks

This function uses the following formula: (x >= y) ? 1 : 0. The function returns either 0 or 1 depending on whether the x parameter is greater than the y parameter. To compute a smooth interpolation between 0 and 1, use the smoothstep HLSL intrinsic function.

Type Description

Name	Template Type	Component Type	Size
y	scalar, vector, or matrix	float	any
x	same as input y	float	same dimension(s) as input y
ret	same as input y	float	same dimension(s) as input y

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 and ps_1_4)

See also

<dl> <dt>

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

[llvmbot]

@llvm/issue-subscribers-clang-frontend

Author: Farzon Lotfi (farzonl)

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

DirectX

There were no DXIL opcodes found for step.

SPIR-V

Step:

Description:

Step

Result is 0.0 if x &lt; edge; otherwise result is 1.0.

The operands must all be a scalar or vector whose component type is
floating-point.

Result Type and the type of all operands must be the same type.
Results are computed per component.

<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>48</p></td>
<td
class="tableblock halign-left valign-top"><p><em>&lt;id&gt;</em><br />
<em>edge</em></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>
</tr>
</tbody>
</table>

Test Case(s)

Example 1

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

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

HLSL:

Compares two values, returning 0 or 1 based on which value is greater.

ret step(y, x)

Parameters

Item	Description
<span id="y"></span><span id="Y"></span>y<br/>	[in] The first floating-point value to compare.<br/>
<span id="x"></span><span id="X"></span>x<br/>	[in] The second floating-point value to compare.<br/>

Return Value

1 if the x parameter is greater than or equal to the y parameter; otherwise, 0.

Remarks

This function uses the following formula: (x >= y) ? 1 : 0. The function returns either 0 or 1 depending on whether the x parameter is greater than the y parameter. To compute a smooth interpolation between 0 and 1, use the smoothstep HLSL intrinsic function.

Type Description

Name	Template Type	Component Type	Size
y	scalar, vector, or matrix	float	any
x	same as input y	float	same dimension(s) as input y
ret	same as input y	float	same dimension(s) as input y

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 and ps_1_4)

See also

<dl> <dt>

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