[clang] [llvm] [HLSL][DXIL][SPRIV] Added WaveActiveMin intrinsic (PR #164385)
Sietze Riemersma via cfe-commits
cfe-commits at lists.llvm.org
Wed Oct 22 00:50:28 PDT 2025
https://github.com/KungFuDonkey updated https://github.com/llvm/llvm-project/pull/164385
>From 830e8f0c62522bbc70186d019d4d18da46690089 Mon Sep 17 00:00:00 2001
From: KungFuDonkey <sietze.riemersma at gmail.com>
Date: Tue, 21 Oct 2025 12:09:48 +0200
Subject: [PATCH 1/2] Added WaveActiveMin intrinsic
---
clang/include/clang/Basic/Builtins.td | 6 +
clang/lib/CodeGen/CGHLSLBuiltins.cpp | 32 +++-
.../lib/Headers/hlsl/hlsl_alias_intrinsics.h | 123 +++++++++++++++
clang/lib/Sema/SemaHLSL.cpp | 1 +
.../CodeGenHLSL/builtins/WaveActiveMin.hlsl | 46 ++++++
.../test/SemaHLSL/BuiltIns/WaveActiveMin.hlsl | 29 ++++
llvm/include/llvm/IR/IntrinsicsDirectX.td | 2 +
llvm/include/llvm/IR/IntrinsicsSPIRV.td | 4 +-
llvm/lib/Target/DirectX/DXIL.td | 10 ++
.../DirectX/DirectXTargetTransformInfo.cpp | 1 +
.../Target/SPIRV/SPIRVInstructionSelector.cpp | 36 +++++
llvm/test/CodeGen/DirectX/WaveActiveMin.ll | 143 ++++++++++++++++++
.../SPIRV/hlsl-intrinsics/WaveActiveMin.ll | 57 +++++++
13 files changed, 488 insertions(+), 2 deletions(-)
create mode 100644 clang/test/CodeGenHLSL/builtins/WaveActiveMin.hlsl
create mode 100644 clang/test/SemaHLSL/BuiltIns/WaveActiveMin.hlsl
create mode 100644 llvm/test/CodeGen/DirectX/WaveActiveMin.ll
create mode 100644 llvm/test/CodeGen/SPIRV/hlsl-intrinsics/WaveActiveMin.ll
diff --git a/clang/include/clang/Basic/Builtins.td b/clang/include/clang/Basic/Builtins.td
index 792e2e07ec594..5c81dc4159801 100644
--- a/clang/include/clang/Basic/Builtins.td
+++ b/clang/include/clang/Basic/Builtins.td
@@ -5011,6 +5011,12 @@ def HLSLWaveActiveMax : LangBuiltin<"HLSL_LANG"> {
let Prototype = "void (...)";
}
+def HLSLWaveActiveMin : LangBuiltin<"HLSL_LANG"> {
+ let Spellings = ["__builtin_hlsl_wave_active_min"];
+ let Attributes = [NoThrow, Const];
+ let Prototype = "void (...)";
+}
+
def HLSLWaveActiveSum : LangBuiltin<"HLSL_LANG"> {
let Spellings = ["__builtin_hlsl_wave_active_sum"];
let Attributes = [NoThrow, Const];
diff --git a/clang/lib/CodeGen/CGHLSLBuiltins.cpp b/clang/lib/CodeGen/CGHLSLBuiltins.cpp
index 4f2f5a761f197..700c506200390 100644
--- a/clang/lib/CodeGen/CGHLSLBuiltins.cpp
+++ b/clang/lib/CodeGen/CGHLSLBuiltins.cpp
@@ -196,7 +196,7 @@ static Intrinsic::ID getWaveActiveSumIntrinsic(llvm::Triple::ArchType Arch,
}
}
-// Return wave active sum that corresponds to the QT scalar type
+// Return wave active max that corresponds to the QT scalar type
static Intrinsic::ID getWaveActiveMaxIntrinsic(llvm::Triple::ArchType Arch,
CGHLSLRuntime &RT, QualType QT) {
switch (Arch) {
@@ -215,6 +215,25 @@ static Intrinsic::ID getWaveActiveMaxIntrinsic(llvm::Triple::ArchType Arch,
}
}
+// Return wave active min that corresponds to the QT scalar type
+static Intrinsic::ID getWaveActiveMinIntrinsic(llvm::Triple::ArchType Arch,
+ CGHLSLRuntime &RT, QualType QT) {
+ switch (Arch) {
+ case llvm::Triple::spirv:
+ if (QT->isUnsignedIntegerType())
+ return Intrinsic::spv_wave_reduce_umin;
+ return Intrinsic::spv_wave_reduce_min;
+ case llvm::Triple::dxil: {
+ if (QT->isUnsignedIntegerType())
+ return Intrinsic::dx_wave_reduce_umin;
+ return Intrinsic::dx_wave_reduce_min;
+ }
+ default:
+ llvm_unreachable("Intrinsic WaveActiveMin"
+ " not supported by target architecture");
+ }
+}
+
// Returns the mangled name for a builtin function that the SPIR-V backend
// will expand into a spec Constant.
static std::string getSpecConstantFunctionName(clang::QualType SpecConstantType,
@@ -719,6 +738,17 @@ Value *CodeGenFunction::EmitHLSLBuiltinExpr(unsigned BuiltinID,
&CGM.getModule(), IID, {OpExpr->getType()}),
ArrayRef{OpExpr}, "hlsl.wave.active.max");
}
+ case Builtin::BI__builtin_hlsl_wave_active_min: {
+ // Due to the use of variadic arguments, explicitly retreive argument
+ Value *OpExpr = EmitScalarExpr(E->getArg(0));
+ Intrinsic::ID IID = getWaveActiveMinIntrinsic(
+ getTarget().getTriple().getArch(), CGM.getHLSLRuntime(),
+ E->getArg(0)->getType());
+
+ return EmitRuntimeCall(Intrinsic::getOrInsertDeclaration(
+ &CGM.getModule(), IID, {OpExpr->getType()}),
+ ArrayRef{OpExpr}, "hlsl.wave.active.min");
+ }
case Builtin::BI__builtin_hlsl_wave_get_lane_index: {
// We don't define a SPIR-V intrinsic, instead it is a SPIR-V built-in
// defined in SPIRVBuiltins.td. So instead we manually get the matching name
diff --git a/clang/lib/Headers/hlsl/hlsl_alias_intrinsics.h b/clang/lib/Headers/hlsl/hlsl_alias_intrinsics.h
index d973371312701..1d440f49af6c4 100644
--- a/clang/lib/Headers/hlsl/hlsl_alias_intrinsics.h
+++ b/clang/lib/Headers/hlsl/hlsl_alias_intrinsics.h
@@ -2597,6 +2597,129 @@ __attribute__((convergent)) double3 WaveActiveMax(double3);
_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
__attribute__((convergent)) double4 WaveActiveMax(double4);
+//===----------------------------------------------------------------------===//
+// WaveActiveMax builtins
+//===----------------------------------------------------------------------===//
+
+_HLSL_16BIT_AVAILABILITY(shadermodel, 6.0)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
+__attribute__((convergent)) half WaveActiveMin(half);
+_HLSL_16BIT_AVAILABILITY(shadermodel, 6.0)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
+__attribute__((convergent)) half2 WaveActiveMin(half2);
+_HLSL_16BIT_AVAILABILITY(shadermodel, 6.0)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
+__attribute__((convergent)) half3 WaveActiveMin(half3);
+_HLSL_16BIT_AVAILABILITY(shadermodel, 6.0)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
+__attribute__((convergent)) half4 WaveActiveMin(half4);
+
+#ifdef __HLSL_ENABLE_16_BIT
+_HLSL_AVAILABILITY(shadermodel, 6.0)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
+__attribute__((convergent)) int16_t WaveActiveMin(int16_t);
+_HLSL_AVAILABILITY(shadermodel, 6.0)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
+__attribute__((convergent)) int16_t2 WaveActiveMin(int16_t2);
+_HLSL_AVAILABILITY(shadermodel, 6.0)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
+__attribute__((convergent)) int16_t3 WaveActiveMin(int16_t3);
+_HLSL_AVAILABILITY(shadermodel, 6.0)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
+__attribute__((convergent)) int16_t4 WaveActiveMin(int16_t4);
+
+_HLSL_AVAILABILITY(shadermodel, 6.0)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
+__attribute__((convergent)) uint16_t WaveActiveMin(uint16_t);
+_HLSL_AVAILABILITY(shadermodel, 6.0)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
+__attribute__((convergent)) uint16_t2 WaveActiveMin(uint16_t2);
+_HLSL_AVAILABILITY(shadermodel, 6.0)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
+__attribute__((convergent)) uint16_t3 WaveActiveMin(uint16_t3);
+_HLSL_AVAILABILITY(shadermodel, 6.0)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
+__attribute__((convergent)) uint16_t4 WaveActiveMin(uint16_t4);
+#endif
+
+_HLSL_AVAILABILITY(shadermodel, 6.0)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
+__attribute__((convergent)) int WaveActiveMin(int);
+_HLSL_AVAILABILITY(shadermodel, 6.0)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
+__attribute__((convergent)) int2 WaveActiveMin(int2);
+_HLSL_AVAILABILITY(shadermodel, 6.0)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
+__attribute__((convergent)) int3 WaveActiveMin(int3);
+_HLSL_AVAILABILITY(shadermodel, 6.0)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
+__attribute__((convergent)) int4 WaveActiveMin(int4);
+
+_HLSL_AVAILABILITY(shadermodel, 6.0)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
+__attribute__((convergent)) uint WaveActiveMin(uint);
+_HLSL_AVAILABILITY(shadermodel, 6.0)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
+__attribute__((convergent)) uint2 WaveActiveMin(uint2);
+_HLSL_AVAILABILITY(shadermodel, 6.0)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
+__attribute__((convergent)) uint3 WaveActiveMin(uint3);
+_HLSL_AVAILABILITY(shadermodel, 6.0)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
+__attribute__((convergent)) uint4 WaveActiveMin(uint4);
+
+_HLSL_AVAILABILITY(shadermodel, 6.0)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
+__attribute__((convergent)) int64_t WaveActiveMin(int64_t);
+_HLSL_AVAILABILITY(shadermodel, 6.0)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
+__attribute__((convergent)) int64_t2 WaveActiveMin(int64_t2);
+_HLSL_AVAILABILITY(shadermodel, 6.0)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
+__attribute__((convergent)) int64_t3 WaveActiveMin(int64_t3);
+_HLSL_AVAILABILITY(shadermodel, 6.0)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
+__attribute__((convergent)) int64_t4 WaveActiveMin(int64_t4);
+
+_HLSL_AVAILABILITY(shadermodel, 6.0)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
+__attribute__((convergent)) uint64_t WaveActiveMin(uint64_t);
+_HLSL_AVAILABILITY(shadermodel, 6.0)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
+__attribute__((convergent)) uint64_t2 WaveActiveMin(uint64_t2);
+_HLSL_AVAILABILITY(shadermodel, 6.0)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
+__attribute__((convergent)) uint64_t3 WaveActiveMin(uint64_t3);
+_HLSL_AVAILABILITY(shadermodel, 6.0)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
+__attribute__((convergent)) uint64_t4 WaveActiveMin(uint64_t4);
+
+_HLSL_AVAILABILITY(shadermodel, 6.0)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
+__attribute__((convergent)) float WaveActiveMin(float);
+_HLSL_AVAILABILITY(shadermodel, 6.0)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
+__attribute__((convergent)) float2 WaveActiveMin(float2);
+_HLSL_AVAILABILITY(shadermodel, 6.0)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
+__attribute__((convergent)) float3 WaveActiveMin(float3);
+_HLSL_AVAILABILITY(shadermodel, 6.0)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
+__attribute__((convergent)) float4 WaveActiveMin(float4);
+
+_HLSL_AVAILABILITY(shadermodel, 6.0)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
+__attribute__((convergent)) double WaveActiveMin(double);
+_HLSL_AVAILABILITY(shadermodel, 6.0)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
+__attribute__((convergent)) double2 WaveActiveMin(double2);
+_HLSL_AVAILABILITY(shadermodel, 6.0)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
+__attribute__((convergent)) double3 WaveActiveMin(double3);
+_HLSL_AVAILABILITY(shadermodel, 6.0)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
+__attribute__((convergent)) double4 WaveActiveMin(double4);
+
//===----------------------------------------------------------------------===//
// WaveActiveSum builtins
//===----------------------------------------------------------------------===//
diff --git a/clang/lib/Sema/SemaHLSL.cpp b/clang/lib/Sema/SemaHLSL.cpp
index f34706677b59f..680036b8ca241 100644
--- a/clang/lib/Sema/SemaHLSL.cpp
+++ b/clang/lib/Sema/SemaHLSL.cpp
@@ -3197,6 +3197,7 @@ bool SemaHLSL::CheckBuiltinFunctionCall(unsigned BuiltinID, CallExpr *TheCall) {
break;
}
case Builtin::BI__builtin_hlsl_wave_active_max:
+ case Builtin::BI__builtin_hlsl_wave_active_min:
case Builtin::BI__builtin_hlsl_wave_active_sum: {
if (SemaRef.checkArgCount(TheCall, 1))
return true;
diff --git a/clang/test/CodeGenHLSL/builtins/WaveActiveMin.hlsl b/clang/test/CodeGenHLSL/builtins/WaveActiveMin.hlsl
new file mode 100644
index 0000000000000..1194f842deed6
--- /dev/null
+++ b/clang/test/CodeGenHLSL/builtins/WaveActiveMin.hlsl
@@ -0,0 +1,46 @@
+// RUN: %clang_cc1 -std=hlsl2021 -finclude-default-header -triple \
+// RUN: dxil-pc-shadermodel6.3-compute %s -emit-llvm -disable-llvm-passes -o - | \
+// RUN: FileCheck %s --check-prefixes=CHECK,CHECK-DXIL
+// RUN: %clang_cc1 -std=hlsl2021 -finclude-default-header -triple \
+// RUN: spirv-pc-vulkan-compute %s -emit-llvm -disable-llvm-passes -o - | \
+// RUN: FileCheck %s --check-prefixes=CHECK,CHECK-SPIRV
+
+// Test basic lowering to runtime function call.
+
+// CHECK-LABEL: test_int
+int test_int(int expr) {
+ // CHECK-SPIRV: %[[RET:.*]] = call spir_func [[TY:.*]] @llvm.spv.wave.reduce.min.i32([[TY]] %[[#]])
+ // CHECK-DXIL: %[[RET:.*]] = call [[TY:.*]] @llvm.dx.wave.reduce.min.i32([[TY]] %[[#]])
+ // CHECK: ret [[TY]] %[[RET]]
+ return WaveActiveMin(expr);
+}
+
+// CHECK-DXIL: declare [[TY]] @llvm.dx.wave.reduce.min.i32([[TY]]) #[[#attr:]]
+// CHECK-SPIRV: declare [[TY]] @llvm.spv.wave.reduce.min.i32([[TY]]) #[[#attr:]]
+
+// CHECK-LABEL: test_uint64_t
+uint64_t test_uint64_t(uint64_t expr) {
+ // CHECK-SPIRV: %[[RET:.*]] = call spir_func [[TY:.*]] @llvm.spv.wave.reduce.umin.i64([[TY]] %[[#]])
+ // CHECK-DXIL: %[[RET:.*]] = call [[TY:.*]] @llvm.dx.wave.reduce.umin.i64([[TY]] %[[#]])
+ // CHECK: ret [[TY]] %[[RET]]
+ return WaveActiveMin(expr);
+}
+
+// CHECK-DXIL: declare [[TY]] @llvm.dx.wave.reduce.umin.i64([[TY]]) #[[#attr:]]
+// CHECK-SPIRV: declare [[TY]] @llvm.spv.wave.reduce.umin.i64([[TY]]) #[[#attr:]]
+
+// Test basic lowering to runtime function call with array and float value.
+
+// CHECK-LABEL: test_floatv4
+float4 test_floatv4(float4 expr) {
+ // CHECK-SPIRV: %[[RET1:.*]] = call reassoc nnan ninf nsz arcp afn spir_func [[TY1:.*]] @llvm.spv.wave.reduce.min.v4f32([[TY1]] %[[#]]
+ // CHECK-DXIL: %[[RET1:.*]] = call reassoc nnan ninf nsz arcp afn [[TY1:.*]] @llvm.dx.wave.reduce.min.v4f32([[TY1]] %[[#]])
+ // CHECK: ret [[TY1]] %[[RET1]]
+ return WaveActiveMin(expr);
+}
+
+// CHECK-DXIL: declare [[TY1]] @llvm.dx.wave.reduce.min.v4f32([[TY1]]) #[[#attr]]
+// CHECK-SPIRV: declare [[TY1]] @llvm.spv.wave.reduce.min.v4f32([[TY1]]) #[[#attr]]
+
+// CHECK: attributes #[[#attr]] = {{{.*}} convergent {{.*}}}
+
diff --git a/clang/test/SemaHLSL/BuiltIns/WaveActiveMin.hlsl b/clang/test/SemaHLSL/BuiltIns/WaveActiveMin.hlsl
new file mode 100644
index 0000000000000..3b12faf8d9978
--- /dev/null
+++ b/clang/test/SemaHLSL/BuiltIns/WaveActiveMin.hlsl
@@ -0,0 +1,29 @@
+// RUN: %clang_cc1 -finclude-default-header -triple dxil-pc-shadermodel6.6-library %s -emit-llvm-only -disable-llvm-passes -verify
+
+int test_too_few_arg() {
+ return __builtin_hlsl_wave_active_min();
+ // expected-error at -1 {{too few arguments to function call, expected 1, have 0}}
+}
+
+float2 test_too_many_arg(float2 p0) {
+ return __builtin_hlsl_wave_active_min(p0, p0);
+ // expected-error at -1 {{too many arguments to function call, expected 1, have 2}}
+}
+
+bool test_expr_bool_type_check(bool p0) {
+ return __builtin_hlsl_wave_active_min(p0);
+ // expected-error at -1 {{invalid operand of type 'bool'}}
+}
+
+bool2 test_expr_bool_vec_type_check(bool2 p0) {
+ return __builtin_hlsl_wave_active_min(p0);
+ // expected-error at -1 {{invalid operand of type 'bool2' (aka 'vector<bool, 2>')}}
+}
+
+struct S { float f; };
+
+S test_expr_struct_type_check(S p0) {
+ return __builtin_hlsl_wave_active_min(p0);
+ // expected-error at -1 {{invalid operand of type 'S' where a scalar or vector is required}}
+}
+
diff --git a/llvm/include/llvm/IR/IntrinsicsDirectX.td b/llvm/include/llvm/IR/IntrinsicsDirectX.td
index 3b7077c52db21..dbec8e72fad19 100644
--- a/llvm/include/llvm/IR/IntrinsicsDirectX.td
+++ b/llvm/include/llvm/IR/IntrinsicsDirectX.td
@@ -153,6 +153,8 @@ def int_dx_wave_any : DefaultAttrsIntrinsic<[llvm_i1_ty], [llvm_i1_ty], [IntrCon
def int_dx_wave_getlaneindex : DefaultAttrsIntrinsic<[llvm_i32_ty], [], [IntrConvergent, IntrNoMem]>;
def int_dx_wave_reduce_max : DefaultAttrsIntrinsic<[llvm_any_ty], [LLVMMatchType<0>], [IntrConvergent, IntrNoMem]>;
def int_dx_wave_reduce_umax : DefaultAttrsIntrinsic<[llvm_anyint_ty], [LLVMMatchType<0>], [IntrConvergent, IntrNoMem]>;
+def int_dx_wave_reduce_min : DefaultAttrsIntrinsic<[llvm_anyint_ty], [LLVMMatchType<0>], [IntrConvergent, IntrNoMem]>;
+def int_dx_wave_reduce_umin : DefaultAttrsIntrinsic<[llvm_anyint_ty], [LLVMMatchType<0>], [IntrConvergent, IntrNoMem]>;
def int_dx_wave_reduce_sum : DefaultAttrsIntrinsic<[llvm_any_ty], [LLVMMatchType<0>], [IntrConvergent, IntrNoMem]>;
def int_dx_wave_reduce_usum : DefaultAttrsIntrinsic<[llvm_anyint_ty], [LLVMMatchType<0>], [IntrConvergent, IntrNoMem]>;
def int_dx_wave_is_first_lane : DefaultAttrsIntrinsic<[llvm_i1_ty], [], [IntrConvergent]>;
diff --git a/llvm/include/llvm/IR/IntrinsicsSPIRV.td b/llvm/include/llvm/IR/IntrinsicsSPIRV.td
index 49a182be98acd..bc51fb639fd75 100644
--- a/llvm/include/llvm/IR/IntrinsicsSPIRV.td
+++ b/llvm/include/llvm/IR/IntrinsicsSPIRV.td
@@ -122,6 +122,8 @@ def int_spv_rsqrt : DefaultAttrsIntrinsic<[LLVMMatchType<0>], [llvm_anyfloat_ty]
def int_spv_wave_any : DefaultAttrsIntrinsic<[llvm_i1_ty], [llvm_i1_ty], [IntrConvergent, IntrNoMem]>;
def int_spv_wave_reduce_umax : DefaultAttrsIntrinsic<[llvm_any_ty], [LLVMMatchType<0>], [IntrConvergent, IntrNoMem]>;
def int_spv_wave_reduce_max : DefaultAttrsIntrinsic<[llvm_any_ty], [LLVMMatchType<0>], [IntrConvergent, IntrNoMem]>;
+ def int_spv_wave_reduce_min : DefaultAttrsIntrinsic<[llvm_any_ty], [LLVMMatchType<0>], [IntrConvergent, IntrNoMem]>;
+ def int_spv_wave_reduce_umin : DefaultAttrsIntrinsic<[llvm_any_ty], [LLVMMatchType<0>], [IntrConvergent, IntrNoMem]>;
def int_spv_wave_reduce_sum : DefaultAttrsIntrinsic<[llvm_any_ty], [LLVMMatchType<0>], [IntrConvergent, IntrNoMem]>;
def int_spv_wave_is_first_lane : DefaultAttrsIntrinsic<[llvm_i1_ty], [], [IntrConvergent]>;
def int_spv_wave_readlane : DefaultAttrsIntrinsic<[llvm_any_ty], [LLVMMatchType<0>, llvm_i32_ty], [IntrConvergent, IntrNoMem]>;
@@ -136,7 +138,7 @@ def int_spv_rsqrt : DefaultAttrsIntrinsic<[LLVMMatchType<0>], [llvm_anyfloat_ty]
def int_spv_sclamp : DefaultAttrsIntrinsic<[llvm_anyint_ty], [LLVMMatchType<0>, LLVMMatchType<0>, LLVMMatchType<0>], [IntrNoMem]>;
def int_spv_nclamp : DefaultAttrsIntrinsic<[llvm_anyfloat_ty], [LLVMMatchType<0>, LLVMMatchType<0>, LLVMMatchType<0>], [IntrNoMem]>;
- // Create resource handle given the binding information. Returns a
+ // Create resource handle given the binding information. Returns a
// type appropriate for the kind of resource given the set id, binding id,
// array size of the binding, as well as an index and an indicator
// whether that index may be non-uniform.
diff --git a/llvm/lib/Target/DirectX/DXIL.td b/llvm/lib/Target/DirectX/DXIL.td
index 44c48305f2832..3986f62d47cb8 100644
--- a/llvm/lib/Target/DirectX/DXIL.td
+++ b/llvm/lib/Target/DirectX/DXIL.td
@@ -1058,6 +1058,16 @@ def WaveActiveOp : DXILOp<119, waveActiveOp> {
IntrinArgIndex<0>, IntrinArgI8<WaveOpKind_Max>,
IntrinArgI8<SignedOpKind_Unsigned>
]>,
+ IntrinSelect<int_dx_wave_reduce_min,
+ [
+ IntrinArgIndex<0>, IntrinArgI8<WaveOpKind_Max>,
+ IntrinArgI8<SignedOpKind_Signed>
+ ]>,
+ IntrinSelect<int_dx_wave_reduce_umin,
+ [
+ IntrinArgIndex<0>, IntrinArgI8<WaveOpKind_Max>,
+ IntrinArgI8<SignedOpKind_Unsigned>
+ ]>,
];
let arguments = [OverloadTy, Int8Ty, Int8Ty];
diff --git a/llvm/lib/Target/DirectX/DirectXTargetTransformInfo.cpp b/llvm/lib/Target/DirectX/DirectXTargetTransformInfo.cpp
index 68fd3e0bc74c7..96a9285767aa6 100644
--- a/llvm/lib/Target/DirectX/DirectXTargetTransformInfo.cpp
+++ b/llvm/lib/Target/DirectX/DirectXTargetTransformInfo.cpp
@@ -55,6 +55,7 @@ bool DirectXTTIImpl::isTargetIntrinsicTriviallyScalarizable(
case Intrinsic::dx_splitdouble:
case Intrinsic::dx_wave_readlane:
case Intrinsic::dx_wave_reduce_max:
+ case Intrinsic::dx_wave_reduce_min:
case Intrinsic::dx_wave_reduce_sum:
case Intrinsic::dx_wave_reduce_umax:
case Intrinsic::dx_wave_reduce_usum:
diff --git a/llvm/lib/Target/SPIRV/SPIRVInstructionSelector.cpp b/llvm/lib/Target/SPIRV/SPIRVInstructionSelector.cpp
index a0cff4d82b500..0719efed5e6a8 100644
--- a/llvm/lib/Target/SPIRV/SPIRVInstructionSelector.cpp
+++ b/llvm/lib/Target/SPIRV/SPIRVInstructionSelector.cpp
@@ -222,6 +222,9 @@ class SPIRVInstructionSelector : public InstructionSelector {
bool selectWaveReduceMax(Register ResVReg, const SPIRVType *ResType,
MachineInstr &I, bool IsUnsigned) const;
+ bool selectWaveActiveMin(Register ResVReg, const SPIRVType *ResType,
+ MachineInstr &I, bool IsUnsigned) const;
+
bool selectWaveReduceSum(Register ResVReg, const SPIRVType *ResType,
MachineInstr &I) const;
@@ -2456,6 +2459,35 @@ bool SPIRVInstructionSelector::selectWaveReduceMax(Register ResVReg,
.constrainAllUses(TII, TRI, RBI);
}
+bool SPIRVInstructionSelector::selectWaveActiveMin(Register ResVReg,
+ const SPIRVType *ResType,
+ MachineInstr &I,
+ bool IsUnsigned) const {
+ assert(I.getNumOperands() == 3);
+ assert(I.getOperand(2).isReg());
+ MachineBasicBlock &BB = *I.getParent();
+ Register InputRegister = I.getOperand(2).getReg();
+ SPIRVType *InputType = GR.getSPIRVTypeForVReg(InputRegister);
+
+ if (!InputType)
+ report_fatal_error("Input Type could not be determined.");
+
+ SPIRVType *IntTy = GR.getOrCreateSPIRVIntegerType(32, I, TII);
+ // Retreive the operation to use based on input type
+ bool IsFloatTy = GR.isScalarOrVectorOfType(InputRegister, SPIRV::OpTypeFloat);
+ auto IntegerOpcodeType =
+ IsUnsigned ? SPIRV::OpGroupNonUniformUMin : SPIRV::OpGroupNonUniformSMin;
+ auto Opcode = IsFloatTy ? SPIRV::OpGroupNonUniformFMin : IntegerOpcodeType;
+ return BuildMI(BB, I, I.getDebugLoc(), TII.get(Opcode))
+ .addDef(ResVReg)
+ .addUse(GR.getSPIRVTypeID(ResType))
+ .addUse(GR.getOrCreateConstInt(SPIRV::Scope::Subgroup, I, IntTy, TII,
+ !STI.isShader()))
+ .addImm(SPIRV::GroupOperation::Reduce)
+ .addUse(I.getOperand(2).getReg())
+ .constrainAllUses(TII, TRI, RBI);
+}
+
bool SPIRVInstructionSelector::selectWaveReduceSum(Register ResVReg,
const SPIRVType *ResType,
MachineInstr &I) const {
@@ -3431,6 +3463,10 @@ bool SPIRVInstructionSelector::selectIntrinsic(Register ResVReg,
return selectWaveReduceMax(ResVReg, ResType, I, /*IsUnsigned*/ true);
case Intrinsic::spv_wave_reduce_max:
return selectWaveReduceMax(ResVReg, ResType, I, /*IsUnsigned*/ false);
+ case Intrinsic::spv_wave_reduce_umin:
+ return selectWaveReduceMax(ResVReg, ResType, I, /*IsUnsigned*/ true);
+ case Intrinsic::spv_wave_reduce_min:
+ return selectWaveReduceMax(ResVReg, ResType, I, /*IsUnsigned*/ false);
case Intrinsic::spv_wave_reduce_sum:
return selectWaveReduceSum(ResVReg, ResType, I);
case Intrinsic::spv_wave_readlane:
diff --git a/llvm/test/CodeGen/DirectX/WaveActiveMin.ll b/llvm/test/CodeGen/DirectX/WaveActiveMin.ll
new file mode 100644
index 0000000000000..24fde48fadfeb
--- /dev/null
+++ b/llvm/test/CodeGen/DirectX/WaveActiveMin.ll
@@ -0,0 +1,143 @@
+; RUN: opt -S -scalarizer -dxil-op-lower -mtriple=dxil-pc-shadermodel6.3-library < %s | FileCheck %s
+
+; Test that for scalar values, WaveActiveMin maps down to the DirectX op
+
+define noundef half @wave_active_min_half(half noundef %expr) {
+entry:
+; CHECK: call half @dx.op.waveActiveOp.f16(i32 119, half %expr, i8 2, i8 0){{$}}
+ %ret = call half @llvm.dx.wave.reduce.min.f16(half %expr)
+ ret half %ret
+}
+
+define noundef float @wave_active_min_float(float noundef %expr) {
+entry:
+; CHECK: call float @dx.op.waveActiveOp.f32(i32 119, float %expr, i8 2, i8 0){{$}}
+ %ret = call float @llvm.dx.wave.reduce.min.f32(float %expr)
+ ret float %ret
+}
+
+define noundef double @wave_active_min_double(double noundef %expr) {
+entry:
+; CHECK: call double @dx.op.waveActiveOp.f64(i32 119, double %expr, i8 2, i8 0){{$}}
+ %ret = call double @llvm.dx.wave.reduce.min.f64(double %expr)
+ ret double %ret
+}
+
+define noundef i16 @wave_active_min_i16(i16 noundef %expr) {
+entry:
+; CHECK: call i16 @dx.op.waveActiveOp.i16(i32 119, i16 %expr, i8 2, i8 0){{$}}
+ %ret = call i16 @llvm.dx.wave.reduce.min.i16(i16 %expr)
+ ret i16 %ret
+}
+
+define noundef i32 @wave_active_min_i32(i32 noundef %expr) {
+entry:
+; CHECK: call i32 @dx.op.waveActiveOp.i32(i32 119, i32 %expr, i8 2, i8 0){{$}}
+ %ret = call i32 @llvm.dx.wave.reduce.min.i32(i32 %expr)
+ ret i32 %ret
+}
+
+define noundef i64 @wave_active_min_i64(i64 noundef %expr) {
+entry:
+; CHECK: call i64 @dx.op.waveActiveOp.i64(i32 119, i64 %expr, i8 2, i8 0){{$}}
+ %ret = call i64 @llvm.dx.wave.reduce.min.i64(i64 %expr)
+ ret i64 %ret
+}
+
+define noundef i16 @wave_active_umin_i16(i16 noundef %expr) {
+entry:
+; CHECK: call i16 @dx.op.waveActiveOp.i16(i32 119, i16 %expr, i8 2, i8 1){{$}}
+ %ret = call i16 @llvm.dx.wave.reduce.umin.i16(i16 %expr)
+ ret i16 %ret
+}
+
+define noundef i32 @wave_active_umin_i32(i32 noundef %expr) {
+entry:
+; CHECK: call i32 @dx.op.waveActiveOp.i32(i32 119, i32 %expr, i8 2, i8 1){{$}}
+ %ret = call i32 @llvm.dx.wave.reduce.umin.i32(i32 %expr)
+ ret i32 %ret
+}
+
+define noundef i64 @wave_active_umin_i64(i64 noundef %expr) {
+entry:
+; CHECK: call i64 @dx.op.waveActiveOp.i64(i32 119, i64 %expr, i8 2, i8 1){{$}}
+ %ret = call i64 @llvm.dx.wave.reduce.umin.i64(i64 %expr)
+ ret i64 %ret
+}
+
+declare half @llvm.dx.wave.reduce.min.f16(half)
+declare float @llvm.dx.wave.reduce.min.f32(float)
+declare double @llvm.dx.wave.reduce.min.f64(double)
+
+declare i16 @llvm.dx.wave.reduce.min.i16(i16)
+declare i32 @llvm.dx.wave.reduce.min.i32(i32)
+declare i64 @llvm.dx.wave.reduce.min.i64(i64)
+
+declare i16 @llvm.dx.wave.reduce.umin.i16(i16)
+declare i32 @llvm.dx.wave.reduce.umin.i32(i32)
+declare i64 @llvm.dx.wave.reduce.umin.i64(i64)
+
+; Test that for vector values, WaveActiveMin scalarizes and maps down to the
+; DirectX op
+
+define noundef <2 x half> @wave_active_min_v2half(<2 x half> noundef %expr) {
+entry:
+; CHECK: call half @dx.op.waveActiveOp.f16(i32 119, half %expr.i0, i8 2, i8 0){{$}}
+; CHECK: call half @dx.op.waveActiveOp.f16(i32 119, half %expr.i1, i8 2, i8 0){{$}}
+ %ret = call <2 x half> @llvm.dx.wave.reduce.min.v2f16(<2 x half> %expr)
+ ret <2 x half> %ret
+}
+
+define noundef <3 x i32> @wave_active_min_v3i32(<3 x i32> noundef %expr) {
+entry:
+; CHECK: call i32 @dx.op.waveActiveOp.i32(i32 119, i32 %expr.i0, i8 2, i8 0){{$}}
+; CHECK: call i32 @dx.op.waveActiveOp.i32(i32 119, i32 %expr.i1, i8 2, i8 0){{$}}
+; CHECK: call i32 @dx.op.waveActiveOp.i32(i32 119, i32 %expr.i2, i8 2, i8 0){{$}}
+ %ret = call <3 x i32> @llvm.dx.wave.reduce.min.v3i32(<3 x i32> %expr)
+ ret <3 x i32> %ret
+}
+
+define noundef <4 x double> @wave_active_min_v4f64(<4 x double> noundef %expr) {
+entry:
+; CHECK: call double @dx.op.waveActiveOp.f64(i32 119, double %expr.i0, i8 2, i8 0){{$}}
+; CHECK: call double @dx.op.waveActiveOp.f64(i32 119, double %expr.i1, i8 2, i8 0){{$}}
+; CHECK: call double @dx.op.waveActiveOp.f64(i32 119, double %expr.i2, i8 2, i8 0){{$}}
+; CHECK: call double @dx.op.waveActiveOp.f64(i32 119, double %expr.i3, i8 2, i8 0){{$}}
+ %ret = call <4 x double> @llvm.dx.wave.reduce.min.v4f64(<4 x double> %expr)
+ ret <4 x double> %ret
+}
+
+declare <2 x half> @llvm.dx.wave.reduce.min.v2f16(<2 x half>)
+declare <3 x i32> @llvm.dx.wave.reduce.min.v3i32(<3 x i32>)
+declare <4 x double> @llvm.dx.wave.reduce.min.v4f64(<4 x double>)
+
+define noundef <2 x i16> @wave_active_umin_v2i16(<2 x i16> noundef %expr) {
+entry:
+; CHECK: call i16 @dx.op.waveActiveOp.i16(i32 119, i16 %expr.i0, i8 2, i8 1){{$}}
+; CHECK: call i16 @dx.op.waveActiveOp.i16(i32 119, i16 %expr.i1, i8 2, i8 1){{$}}
+ %ret = call <2 x i16> @llvm.dx.wave.reduce.umin.v2f16(<2 x i16> %expr)
+ ret <2 x i16> %ret
+}
+
+define noundef <3 x i32> @wave_active_umin_v3i32(<3 x i32> noundef %expr) {
+entry:
+; CHECK: call i32 @dx.op.waveActiveOp.i32(i32 119, i32 %expr.i0, i8 2, i8 1){{$}}
+; CHECK: call i32 @dx.op.waveActiveOp.i32(i32 119, i32 %expr.i1, i8 2, i8 1){{$}}
+; CHECK: call i32 @dx.op.waveActiveOp.i32(i32 119, i32 %expr.i2, i8 2, i8 1){{$}}
+ %ret = call <3 x i32> @llvm.dx.wave.reduce.umin.v3i32(<3 x i32> %expr)
+ ret <3 x i32> %ret
+}
+
+define noundef <4 x i64> @wave_active_umin_v4f64(<4 x i64> noundef %expr) {
+entry:
+; CHECK: call i64 @dx.op.waveActiveOp.i64(i32 119, i64 %expr.i0, i8 2, i8 1){{$}}
+; CHECK: call i64 @dx.op.waveActiveOp.i64(i32 119, i64 %expr.i1, i8 2, i8 1){{$}}
+; CHECK: call i64 @dx.op.waveActiveOp.i64(i32 119, i64 %expr.i2, i8 2, i8 1){{$}}
+; CHECK: call i64 @dx.op.waveActiveOp.i64(i32 119, i64 %expr.i3, i8 2, i8 1){{$}}
+ %ret = call <4 x i64> @llvm.dx.wave.reduce.umin.v4f64(<4 x i64> %expr)
+ ret <4 x i64> %ret
+}
+
+declare <2 x i16> @llvm.dx.wave.reduce.umin.v2f16(<2 x i16>)
+declare <3 x i32> @llvm.dx.wave.reduce.umin.v3i32(<3 x i32>)
+declare <4 x i64> @llvm.dx.wave.reduce.umin.v4f64(<4 x i64>)
diff --git a/llvm/test/CodeGen/SPIRV/hlsl-intrinsics/WaveActiveMin.ll b/llvm/test/CodeGen/SPIRV/hlsl-intrinsics/WaveActiveMin.ll
new file mode 100644
index 0000000000000..d121c1a937a9b
--- /dev/null
+++ b/llvm/test/CodeGen/SPIRV/hlsl-intrinsics/WaveActiveMin.ll
@@ -0,0 +1,57 @@
+; RUN: llc -verify-machineinstrs -O0 -mtriple=spirv-vulkan-unknown %s -o - | FileCheck %s
+; RUN: %if spirv-tools %{ llc -O0 -mtriple=spirv-vulkan-unknown %s -o - -filetype=obj | spirv-val %}
+
+; Test lowering to spir-v backend for various types and scalar/vector
+
+; CHECK: OpCapability GroupNonUniformArithmetic
+
+; CHECK-DAG: %[[#f16:]] = OpTypeFloat 16
+; CHECK-DAG: %[[#f32:]] = OpTypeFloat 32
+; CHECK-DAG: %[[#uint:]] = OpTypeInt 32 0
+; CHECK-DAG: %[[#v4_half:]] = OpTypeVector %[[#f16]] 4
+; CHECK-DAG: %[[#scope:]] = OpConstant %[[#uint]] 3
+
+; CHECK-LABEL: Begin function test_float
+; CHECK: %[[#fexpr:]] = OpFunctionParameter %[[#f32]]
+define float @test_float(float %fexpr) {
+entry:
+; CHECK: %[[#fret:]] = OpGroupNonUniformFMin %[[#f32]] %[[#scope]] Reduce %[[#fexpr]]
+ %0 = call float @llvm.spv.wave.reduce.min.f32(float %fexpr)
+ ret float %0
+}
+
+; CHECK-LABEL: Begin function test_int_signed
+; CHECK: %[[#iexpr:]] = OpFunctionParameter %[[#uint]]
+define i32 @test_int_signed(i32 %iexpr) {
+entry:
+; CHECK: %[[#iret:]] = OpGroupNonUniformSMin %[[#uint]] %[[#scope]] Reduce %[[#iexpr]]
+ %0 = call i32 @llvm.spv.wave.reduce.min.i32(i32 %iexpr)
+ ret i32 %0
+}
+
+; CHECK-LABEL: Begin function test_int_unsigned
+; CHECK: %[[#iexpr:]] = OpFunctionParameter %[[#uint]]
+define i32 @test_int_unsigned(i32 %iexpr) {
+entry:
+; CHECK: %[[#iret:]] = OpGroupNonUniformUMin %[[#uint]] %[[#scope]] Reduce %[[#iexpr]]
+ %0 = call i32 @llvm.spv.wave.reduce.umin.i32(i32 %iexpr)
+ ret i32 %0
+}
+
+; CHECK-LABEL: Begin function test_vhalf
+; CHECK: %[[#vbexpr:]] = OpFunctionParameter %[[#v4_half]]
+define <4 x half> @test_vhalf(<4 x half> %vbexpr) {
+entry:
+; CHECK: %[[#vhalfret:]] = OpGroupNonUniformFMin %[[#v4_half]] %[[#scope]] Reduce %[[#vbexpr]]
+ %0 = call <4 x half> @llvm.spv.wave.reduce.min.v4half(<4 x half> %vbexpr)
+ ret <4 x half> %0
+}
+
+declare float @llvm.spv.wave.reduce.min.f32(float)
+declare i32 @llvm.spv.wave.reduce.min.i32(i32)
+declare <4 x half> @llvm.spv.wave.reduce.min.v4half(<4 x half>)
+
+declare float @llvm.spv.wave.reduce.umin.f32(float)
+declare i32 @llvm.spv.wave.reduce.umin.i32(i32)
+declare <4 x half> @llvm.spv.wave.reduce.umin.v4half(<4 x half>)
+
>From 3be953b2416797f3932ace28e2961f855dea5a5c Mon Sep 17 00:00:00 2001
From: KungFuDonkey <sietze.riemersma at gmail.com>
Date: Wed, 22 Oct 2025 09:50:15 +0200
Subject: [PATCH 2/2] Fixed max/min copy typos and added umin to dx
---
.../lib/Headers/hlsl/hlsl_alias_intrinsics.h | 74 +++++++++----------
.../DirectX/DirectXTargetTransformInfo.cpp | 1 +
2 files changed, 38 insertions(+), 37 deletions(-)
diff --git a/clang/lib/Headers/hlsl/hlsl_alias_intrinsics.h b/clang/lib/Headers/hlsl/hlsl_alias_intrinsics.h
index 1d440f49af6c4..a918af39e4074 100644
--- a/clang/lib/Headers/hlsl/hlsl_alias_intrinsics.h
+++ b/clang/lib/Headers/hlsl/hlsl_alias_intrinsics.h
@@ -2598,126 +2598,126 @@ _HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
__attribute__((convergent)) double4 WaveActiveMax(double4);
//===----------------------------------------------------------------------===//
-// WaveActiveMax builtins
+// WaveActiveMin builtins
//===----------------------------------------------------------------------===//
_HLSL_16BIT_AVAILABILITY(shadermodel, 6.0)
-_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_min)
__attribute__((convergent)) half WaveActiveMin(half);
_HLSL_16BIT_AVAILABILITY(shadermodel, 6.0)
-_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_min)
__attribute__((convergent)) half2 WaveActiveMin(half2);
_HLSL_16BIT_AVAILABILITY(shadermodel, 6.0)
-_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_min)
__attribute__((convergent)) half3 WaveActiveMin(half3);
_HLSL_16BIT_AVAILABILITY(shadermodel, 6.0)
-_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_min)
__attribute__((convergent)) half4 WaveActiveMin(half4);
#ifdef __HLSL_ENABLE_16_BIT
_HLSL_AVAILABILITY(shadermodel, 6.0)
-_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_min)
__attribute__((convergent)) int16_t WaveActiveMin(int16_t);
_HLSL_AVAILABILITY(shadermodel, 6.0)
-_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_min)
__attribute__((convergent)) int16_t2 WaveActiveMin(int16_t2);
_HLSL_AVAILABILITY(shadermodel, 6.0)
-_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_min)
__attribute__((convergent)) int16_t3 WaveActiveMin(int16_t3);
_HLSL_AVAILABILITY(shadermodel, 6.0)
-_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_min)
__attribute__((convergent)) int16_t4 WaveActiveMin(int16_t4);
_HLSL_AVAILABILITY(shadermodel, 6.0)
-_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_min)
__attribute__((convergent)) uint16_t WaveActiveMin(uint16_t);
_HLSL_AVAILABILITY(shadermodel, 6.0)
-_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_min)
__attribute__((convergent)) uint16_t2 WaveActiveMin(uint16_t2);
_HLSL_AVAILABILITY(shadermodel, 6.0)
-_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_min)
__attribute__((convergent)) uint16_t3 WaveActiveMin(uint16_t3);
_HLSL_AVAILABILITY(shadermodel, 6.0)
-_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_min)
__attribute__((convergent)) uint16_t4 WaveActiveMin(uint16_t4);
#endif
_HLSL_AVAILABILITY(shadermodel, 6.0)
-_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_min)
__attribute__((convergent)) int WaveActiveMin(int);
_HLSL_AVAILABILITY(shadermodel, 6.0)
-_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_min)
__attribute__((convergent)) int2 WaveActiveMin(int2);
_HLSL_AVAILABILITY(shadermodel, 6.0)
-_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_min)
__attribute__((convergent)) int3 WaveActiveMin(int3);
_HLSL_AVAILABILITY(shadermodel, 6.0)
-_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_min)
__attribute__((convergent)) int4 WaveActiveMin(int4);
_HLSL_AVAILABILITY(shadermodel, 6.0)
-_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_min)
__attribute__((convergent)) uint WaveActiveMin(uint);
_HLSL_AVAILABILITY(shadermodel, 6.0)
-_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_min)
__attribute__((convergent)) uint2 WaveActiveMin(uint2);
_HLSL_AVAILABILITY(shadermodel, 6.0)
-_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_min)
__attribute__((convergent)) uint3 WaveActiveMin(uint3);
_HLSL_AVAILABILITY(shadermodel, 6.0)
-_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_min)
__attribute__((convergent)) uint4 WaveActiveMin(uint4);
_HLSL_AVAILABILITY(shadermodel, 6.0)
-_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_min)
__attribute__((convergent)) int64_t WaveActiveMin(int64_t);
_HLSL_AVAILABILITY(shadermodel, 6.0)
-_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_min)
__attribute__((convergent)) int64_t2 WaveActiveMin(int64_t2);
_HLSL_AVAILABILITY(shadermodel, 6.0)
-_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_min)
__attribute__((convergent)) int64_t3 WaveActiveMin(int64_t3);
_HLSL_AVAILABILITY(shadermodel, 6.0)
-_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_min)
__attribute__((convergent)) int64_t4 WaveActiveMin(int64_t4);
_HLSL_AVAILABILITY(shadermodel, 6.0)
-_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_min)
__attribute__((convergent)) uint64_t WaveActiveMin(uint64_t);
_HLSL_AVAILABILITY(shadermodel, 6.0)
-_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_min)
__attribute__((convergent)) uint64_t2 WaveActiveMin(uint64_t2);
_HLSL_AVAILABILITY(shadermodel, 6.0)
-_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_min)
__attribute__((convergent)) uint64_t3 WaveActiveMin(uint64_t3);
_HLSL_AVAILABILITY(shadermodel, 6.0)
-_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_min)
__attribute__((convergent)) uint64_t4 WaveActiveMin(uint64_t4);
_HLSL_AVAILABILITY(shadermodel, 6.0)
-_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_min)
__attribute__((convergent)) float WaveActiveMin(float);
_HLSL_AVAILABILITY(shadermodel, 6.0)
-_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_min)
__attribute__((convergent)) float2 WaveActiveMin(float2);
_HLSL_AVAILABILITY(shadermodel, 6.0)
-_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_min)
__attribute__((convergent)) float3 WaveActiveMin(float3);
_HLSL_AVAILABILITY(shadermodel, 6.0)
-_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_min)
__attribute__((convergent)) float4 WaveActiveMin(float4);
_HLSL_AVAILABILITY(shadermodel, 6.0)
-_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_min)
__attribute__((convergent)) double WaveActiveMin(double);
_HLSL_AVAILABILITY(shadermodel, 6.0)
-_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_min)
__attribute__((convergent)) double2 WaveActiveMin(double2);
_HLSL_AVAILABILITY(shadermodel, 6.0)
-_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_min)
__attribute__((convergent)) double3 WaveActiveMin(double3);
_HLSL_AVAILABILITY(shadermodel, 6.0)
-_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_max)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_wave_active_min)
__attribute__((convergent)) double4 WaveActiveMin(double4);
//===----------------------------------------------------------------------===//
diff --git a/llvm/lib/Target/DirectX/DirectXTargetTransformInfo.cpp b/llvm/lib/Target/DirectX/DirectXTargetTransformInfo.cpp
index 96a9285767aa6..60dfd9650937c 100644
--- a/llvm/lib/Target/DirectX/DirectXTargetTransformInfo.cpp
+++ b/llvm/lib/Target/DirectX/DirectXTargetTransformInfo.cpp
@@ -58,6 +58,7 @@ bool DirectXTTIImpl::isTargetIntrinsicTriviallyScalarizable(
case Intrinsic::dx_wave_reduce_min:
case Intrinsic::dx_wave_reduce_sum:
case Intrinsic::dx_wave_reduce_umax:
+ case Intrinsic::dx_wave_reduce_umin:
case Intrinsic::dx_wave_reduce_usum:
case Intrinsic::dx_imad:
case Intrinsic::dx_umad:
More information about the cfe-commits
mailing list