[HLSL] Implementation of the fmod intrinsic

This change implements the frontend for llvm#99118
Builtins.td           - add the fmod builtin
CGBuiltin.cpp         - lower the builtin to llvm FRem instruction
hlsl_intrinsics.h     - add the fmod api
SemaHLSL.cpp          - add type checks for builtin

clang/docs/LanguageExtensions.rst  - add the builtin in *Elementwise Builtins*

clang/docs/LanguageExtensions.rst

@@ -700,6 +700,8 @@ Unless specified otherwise operation(±0) = ±0 and operation(±infinity) = ±in
  T __builtin_elementwise_canonicalize(T x)   return the platform specific canonical encoding                  floating point types
                                              of a floating-point number
  T __builtin_elementwise_copysign(T x, T y)  return the magnitude of x with the sign of y.                    floating point types
+ T __builtin_elementwise_fmod(T x, T y)      return The floating-point remainder of (x/y) whose sign          floating point types
+                                             matches the sign of x.
  T __builtin_elementwise_max(T x, T y)       return x or y, whichever is larger                               integer and floating point types
  T __builtin_elementwise_min(T x, T y)       return x or y, whichever is smaller                              integer and floating point types
  T __builtin_elementwise_add_sat(T x, T y)   return the sum of x and y, clamped to the range of               integer types

clang/include/clang/Basic/Builtins.td

@@ -4788,6 +4788,12 @@ def HLSLStep: LangBuiltin<"HLSL_LANG"> {
   let Prototype = "void(...)";
 }
 
+def HLSLFmod : LangBuiltin<"HLSL_LANG"> {
+  let Spellings = ["__builtin_elementwise_fmod"];
+  let Attributes = [NoThrow, Const];
+  let Prototype = "void(...)";
+}
+
 // Builtins for XRay.
 def XRayCustomEvent : Builtin {
   let Spellings = ["__xray_customevent"];

clang/lib/CodeGen/CGBuiltin.cpp

@@ -2844,7 +2844,8 @@ RValue CodeGenFunction::EmitBuiltinExpr(const GlobalDecl GD, unsigned BuiltinID,
     case Builtin::BI__builtin_fmodf:
     case Builtin::BI__builtin_fmodf16:
     case Builtin::BI__builtin_fmodl:
-    case Builtin::BI__builtin_fmodf128: {
+    case Builtin::BI__builtin_fmodf128:
+    case Builtin::BI__builtin_elementwise_fmod: {
       CodeGenFunction::CGFPOptionsRAII FPOptsRAII(*this, E);
       Value *Arg1 = EmitScalarExpr(E->getArg(0));
       Value *Arg2 = EmitScalarExpr(E->getArg(1));

clang/lib/Headers/hlsl/hlsl_intrinsics.h

@@ -913,6 +913,39 @@ float3 floor(float3);
 _HLSL_BUILTIN_ALIAS(__builtin_elementwise_floor)
 float4 floor(float4);
 
+//===----------------------------------------------------------------------===//
+// fmod builtins
+//===----------------------------------------------------------------------===//
+
+/// \fn T fmod(T x, T y)
+/// \brief Returns the linear interpolation of x to y.
+/// \param x [in] The dividend.
+/// \param y [in] The divisor.
+///
+/// Return the floating-point remainder of the x parameter divided by the y parameter.
+
+_HLSL_16BIT_AVAILABILITY(shadermodel, 6.2)
+_HLSL_BUILTIN_ALIAS(__builtin_elementwise_fmod)
+half fmod(half, half);
+_HLSL_16BIT_AVAILABILITY(shadermodel, 6.2)
+_HLSL_BUILTIN_ALIAS(__builtin_elementwise_fmod)
+half2 fmod(half2, half2);
+_HLSL_16BIT_AVAILABILITY(shadermodel, 6.2)
+_HLSL_BUILTIN_ALIAS(__builtin_elementwise_fmod)
+half3 fmod(half3, half3);
+_HLSL_16BIT_AVAILABILITY(shadermodel, 6.2)
+_HLSL_BUILTIN_ALIAS(__builtin_elementwise_fmod)
+half4 fmod(half4, half4);
+
+_HLSL_BUILTIN_ALIAS(__builtin_elementwise_fmod)
+float fmod(float, float);
+_HLSL_BUILTIN_ALIAS(__builtin_elementwise_fmod)
+float2 fmod(float2, float2);
+_HLSL_BUILTIN_ALIAS(__builtin_elementwise_fmod)
+float3 fmod(float3, float3);
+_HLSL_BUILTIN_ALIAS(__builtin_elementwise_fmod)
+float4 fmod(float4, float4);
+
 //===----------------------------------------------------------------------===//
 // frac builtins
 //===----------------------------------------------------------------------===//

clang/lib/Sema/SemaHLSL.cpp

@@ -1645,6 +1645,18 @@ bool SemaHLSL::CheckBuiltinFunctionCall(unsigned BuiltinID, CallExpr *TheCall) {
       return true;
     break;
   }
+  case Builtin::BI__builtin_elementwise_fmod: {
+    if (SemaRef.checkArgCount(TheCall, 2))
+      return true;
+    if (CheckFloatOrHalfRepresentations(&SemaRef, TheCall))
+      return true;
+
+    ExprResult A = TheCall->getArg(0);
+    QualType ArgTyA = A.get()->getType();
+    // return type is the same as the input type
+    TheCall->setType(ArgTyA);
+    break;
+  }
   case Builtin::BI__builtin_hlsl_select: {
     if (SemaRef.checkArgCount(TheCall, 3))
       return true;

clang/test/CodeGenHLSL/builtins/fmod.hlsl

@@ -0,0 +1,79 @@
+// DirectX target:
+//
+// ---------- Native Half support test -----------
+//
+// RUN: %clang_cc1 -finclude-default-header -x hlsl -triple \
+// RUN:   dxil-pc-shadermodel6.3-library %s -fnative-half-type \
+// RUN:   -emit-llvm -disable-llvm-passes -o - | FileCheck %s \
+// RUN:   --check-prefixes=CHECK \
+// RUN:   -DFNATTRS=noundef -DTYPE=half
+
+//
+// ---------- No Native Half support test -----------
+//
+// RUN: %clang_cc1 -finclude-default-header -x hlsl -triple \
+// RUN:   dxil-pc-shadermodel6.3-library %s -emit-llvm -disable-llvm-passes \
+// RUN:   -o - | FileCheck %s --check-prefixes=CHECK \
+// RUN:   -DFNATTRS=noundef -DTYPE=float
+
+
+// Spirv target:
+//
+// ---------- Native Half support test -----------
+//
+// RUN: %clang_cc1 -finclude-default-header -x hlsl -triple \
+// RUN:   spirv-unknown-vulkan-compute %s -fnative-half-type \
+// RUN:   -emit-llvm -disable-llvm-passes -o - | FileCheck %s \
+// RUN:   --check-prefixes=CHECK \
+// RUN:   -DFNATTRS="spir_func noundef" -DTYPE=half
+
+//
+// ---------- No Native Half support test -----------
+//
+// RUN: %clang_cc1 -finclude-default-header -x hlsl -triple \
+// RUN:   spirv-unknown-vulkan-compute %s -emit-llvm -disable-llvm-passes \
+// RUN:   -o - | FileCheck %s --check-prefixes=CHECK \
+// RUN:   -DFNATTRS="spir_func noundef" -DTYPE=float
+
+
+
+// CHECK: define [[FNATTRS]] [[TYPE]] @
+// CHECK: %fmod = frem [[TYPE]]
+// CHECK: ret [[TYPE]] %fmod
+half test_fmod_half(half p0, half p1) { return fmod(p0, p1); }
+
+// CHECK: define [[FNATTRS]] <2 x [[TYPE]]> @
+// CHECK: %fmod = frem <2 x [[TYPE]]>
+// CHECK: ret <2 x [[TYPE]]> %fmod
+half2 test_fmod_half2(half2 p0, half2 p1) { return fmod(p0, p1); }
+
+// CHECK: define [[FNATTRS]] <3 x [[TYPE]]> @
+// CHECK: %fmod = frem <3 x [[TYPE]]>
+// CHECK: ret <3 x [[TYPE]]> %fmod
+half3 test_fmod_half3(half3 p0, half3 p1) { return fmod(p0, p1); }
+
+// CHECK: define [[FNATTRS]] <4 x [[TYPE]]> @
+// CHECK: %fmod = frem <4 x [[TYPE]]>
+// CHECK: ret <4 x [[TYPE]]> %fmod
+half4 test_fmod_half4(half4 p0, half4 p1) { return fmod(p0, p1); }
+
+// CHECK: define [[FNATTRS]] float @
+// CHECK: %fmod = frem float
+// CHECK: ret float %fmod
+float test_fmod_float(float p0, float p1) { return fmod(p0, p1); }
+
+// CHECK: define [[FNATTRS]] <2 x float> @
+// CHECK: %fmod = frem <2 x float>
+// CHECK: ret <2 x float> %fmod
+float2 test_fmod_float2(float2 p0, float2 p1) { return fmod(p0, p1); }
+
+// CHECK: define [[FNATTRS]] <3 x float> @
+// CHECK: %fmod = frem <3 x float>
+// CHECK: ret <3 x float> %fmod
+float3 test_fmod_float3(float3 p0, float3 p1) { return fmod(p0, p1); }
+
+// DX-CHECK: define [[FNATTRS]] <4 x float> @
+// DX-CHECK: %fmod = frem <4 x float>
+// DX-CHECK: ret <4 x float> %fmod
+float4 test_fmod_float4(float4 p0, float4 p1) { return fmod(p0, p1); }
+

clang/test/SemaHLSL/BuiltIns/fmod-errors.hlsl

@@ -0,0 +1,38 @@
+
+// RUN: %clang_cc1 -finclude-default-header -triple dxil-pc-shadermodel6.6-library %s -fnative-half-type -emit-llvm-only -disable-llvm-passes -verify -verify-ignore-unexpected
+
+float test_too_few_arg() {
+  return __builtin_elementwise_fmod();
+  // expected-error@-1 {{too few arguments to function call, expected 2, have 0}}
+}
+
+float2 test_too_many_arg(float2 p0, float2 p1, float2 p3) {
+  return __builtin_elementwise_fmod(p0, p1, p3);
+  // expected-error@-1 {{too many arguments to function call, expected 2, have 3}}
+}
+
+float builtin_bool_to_float_type_promotion(bool p1, bool p2) {
+  return __builtin_elementwise_fmod(p1, p2);
+  // expected-error@-1 {{passing 'bool' to parameter of incompatible type 'float'}}
+}
+
+float builtin_fmod_int_to_float_promotion(int p1, int p2) {
+  return __builtin_elementwise_fmod(p1, p2);
+  // expected-error@-1 {{passing 'int' to parameter of incompatible type 'float'}}
+}
+
+float2 builtin_fmod_int2_to_float2_promotion(int2 p1, int2 p2) {
+  return __builtin_elementwise_fmod(p1, p2);
+  // expected-error@-1 {{passing 'int2' (aka 'vector<int, 2>') to parameter of incompatible type '__attribute__((__vector_size__(2 * sizeof(float)))) float' (vector of 2 'float' values)}}
+}
+
+// builtins are variadic functions and so are subject to DefaultVariadicArgumentPromotion
+half builtin_fmod_half_scalar (half p0, half p1) {
+  return __builtin_elementwise_fmod(p0, p1);
+  // expected-error@-1 {{passing 'double' to parameter of incompatible type 'float'}}
+}
+
+float builtin_fmod_float_scalar (float p0, float p1) {
+  return __builtin_elementwise_fmod (p0, p1);
+  // expected-error@-1 {{passing 'double' to parameter of incompatible type 'float'}}
+}