[HLSL][DXIL] Implement WaveGetLaneIndex Intrinsic #111576
Merged
+60
−9
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
inbelic
changed the title
- add additional lowering for directx backend in CGBuiltin.cpp
- add directx intrinsic to IntrinscsDirectX.td
- add semantic check of arguments in SemaHLSL.cpp
- add mapping to DXIL op in DXIL.td
- add testing of semantics in WaveGetLaneIndex-errors.hlsl
- add testing of dxil lowering in WaveGetLaneIndex.ll
inbelic
force-pushed
the
inbelic/get-lane-index
branch
from
0911b56 to
167718e
Compare
llvmbot
added
clang
|
@llvm/pr-subscribers-hlsl @llvm/pr-subscribers-clang Author: Finn Plummer (inbelic) ChangesResolves #70105 Full diff: https://github.com/llvm/llvm-project/pull/111576.diff 7 Files Affected:
diff --git a/clang/lib/CodeGen/CGBuiltin.cpp b/clang/lib/CodeGen/CGBuiltin.cpp
index da3eca73bfb575..803bdbaa0e7e01 100644
--- a/clang/lib/CodeGen/CGBuiltin.cpp
+++ b/clang/lib/CodeGen/CGBuiltin.cpp
@@ -18827,9 +18827,21 @@ case Builtin::BI__builtin_hlsl_elementwise_isinf: {
ArrayRef<Value *>{Op0, Op1}, nullptr, "hlsl.step");
}
case Builtin::BI__builtin_hlsl_wave_get_lane_index: {
- return EmitRuntimeCall(CGM.CreateRuntimeFunction(
- llvm::FunctionType::get(IntTy, {}, false), "__hlsl_wave_get_lane_index",
- {}, false, true));
+ // Since we don't define a SPIR-V intrinsic for the SPIR-V built-in from
+ // SPIRVBuiltins.td, manually get the matching name for the DirectX
+ // intrinsic and the demangled builtin name
+ switch (CGM.getTarget().getTriple().getArch()) {
+ case llvm::Triple::dxil:
+ return EmitRuntimeCall(Intrinsic::getDeclaration(
+ &CGM.getModule(), Intrinsic::dx_waveGetLaneIndex));
+ case llvm::Triple::spirv:
+ return EmitRuntimeCall(CGM.CreateRuntimeFunction(
+ llvm::FunctionType::get(IntTy, {}, false),
+ "__hlsl_wave_get_lane_index", {}, false, true));
+ default:
+ llvm_unreachable(
+ "Intrinsic WaveGetLaneIndex not supported by target architecture");
+ }
}
case Builtin::BI__builtin_hlsl_wave_is_first_lane: {
Intrinsic::ID ID = CGM.getHLSLRuntime().getWaveIsFirstLaneIntrinsic();
diff --git a/clang/lib/Sema/SemaHLSL.cpp b/clang/lib/Sema/SemaHLSL.cpp
index 43cc6c81ae5cb0..7d93f41bb2d7a0 100644
--- a/clang/lib/Sema/SemaHLSL.cpp
+++ b/clang/lib/Sema/SemaHLSL.cpp
@@ -1956,6 +1956,11 @@ bool SemaHLSL::CheckBuiltinFunctionCall(unsigned BuiltinID, CallExpr *TheCall) {
return true;
break;
}
+ case Builtin::BI__builtin_hlsl_wave_get_lane_index: {
+ if (SemaRef.checkArgCount(TheCall, 0))
+ return true;
+ break;
+ }
case Builtin::BI__builtin_elementwise_acos:
case Builtin::BI__builtin_elementwise_asin:
case Builtin::BI__builtin_elementwise_atan:
diff --git a/clang/test/CodeGenHLSL/builtins/wave_get_lane_index_simple.hlsl b/clang/test/CodeGenHLSL/builtins/wave_get_lane_index_simple.hlsl
index 8f52d81091c180..e76d35a86e512a 100644
--- a/clang/test/CodeGenHLSL/builtins/wave_get_lane_index_simple.hlsl
+++ b/clang/test/CodeGenHLSL/builtins/wave_get_lane_index_simple.hlsl
@@ -1,14 +1,22 @@
// RUN: %clang_cc1 -std=hlsl2021 -finclude-default-header -x hlsl -triple \
-// RUN: spirv-pc-vulkan-library %s -emit-llvm -disable-llvm-passes -o - | FileCheck %s
+// RUN: spirv-pc-vulkan-library %s -emit-llvm -disable-llvm-passes -o - | FileCheck %s \
+// RUN: --check-prefixes=CHECK,CHECK-SPIRV
+// RUN: %clang_cc1 -finclude-default-header \
+// RUN: -triple dxil-pc-shadermodel6.3-library %s \
+// RUN: -emit-llvm -disable-llvm-passes -o - | FileCheck %s \
+// RUN: --check-prefixes=CHECK,CHECK-DXIL
-// CHECK: define spir_func noundef i32 @_Z6test_1v() [[A0:#[0-9]+]] {
-// CHECK: %[[CI:[0-9]+]] = call token @llvm.experimental.convergence.entry()
-// CHECK: call i32 @__hlsl_wave_get_lane_index() [ "convergencectrl"(token %[[CI]]) ]
-uint test_1() {
+// CHECK-SPIRV: define spir_func noundef i32 @{{.*test_1.*}}() [[A0:#[0-9]+]] {
+// CHECK-DXIL: define noundef i32 @{{.*test_1.*}}() [[A0:#[0-9]+]] {
+// CHECK-SPIRV: %[[CI:[0-9]+]] = call token @llvm.experimental.convergence.entry()
+// CHECK-SPIRV: call i32 @__hlsl_wave_get_lane_index() [ "convergencectrl"(token %[[CI]]) ]
+// CHECK-DXIL: call i32 @llvm.dx.waveGetLaneIndex()
+int test_1() {
return WaveGetLaneIndex();
}
-// CHECK: declare i32 @__hlsl_wave_get_lane_index() [[A1:#[0-9]+]]
+// CHECK-SPIRV: declare i32 @__hlsl_wave_get_lane_index() [[A1:#[0-9]+]]
+// CHECK-DXIL: declare i32 @llvm.dx.waveGetLaneIndex() [[A1:#[0-9]+]]
// CHECK-DAG: attributes [[A0]] = { {{.*}}convergent{{.*}} }
// CHECK-DAG: attributes [[A1]] = { {{.*}}convergent{{.*}} }
diff --git a/clang/test/SemaHLSL/BuiltIns/WaveGetLaneIndex-errors.hlsl b/clang/test/SemaHLSL/BuiltIns/WaveGetLaneIndex-errors.hlsl
new file mode 100644
index 00000000000000..94cfd0662b5fc3
--- /dev/null
+++ b/clang/test/SemaHLSL/BuiltIns/WaveGetLaneIndex-errors.hlsl
@@ -0,0 +1,6 @@
+// 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
+
+int test_too_many_arg(int x) {
+ return __builtin_hlsl_wave_get_lane_index(x);
+ // expected-error@-1 {{too many arguments to function call, expected 0, have 1}}
+}
diff --git a/llvm/include/llvm/IR/IntrinsicsDirectX.td b/llvm/include/llvm/IR/IntrinsicsDirectX.td
index 555877e7aaf0e5..585cb2b7b300df 100644
--- a/llvm/include/llvm/IR/IntrinsicsDirectX.td
+++ b/llvm/include/llvm/IR/IntrinsicsDirectX.td
@@ -82,6 +82,7 @@ def int_dx_imad : DefaultAttrsIntrinsic<[llvm_anyint_ty], [LLVMMatchType<0>, LLV
def int_dx_umad : DefaultAttrsIntrinsic<[llvm_anyint_ty], [LLVMMatchType<0>, LLVMMatchType<0>, LLVMMatchType<0>], [IntrNoMem]>;
def int_dx_normalize : DefaultAttrsIntrinsic<[LLVMMatchType<0>], [llvm_anyfloat_ty], [IntrNoMem]>;
def int_dx_rsqrt : DefaultAttrsIntrinsic<[llvm_anyfloat_ty], [LLVMMatchType<0>], [IntrNoMem]>;
+def int_dx_waveGetLaneIndex : DefaultAttrsIntrinsic<[llvm_i32_ty], [], [IntrConvergent, IntrNoMem]>;
def int_dx_wave_is_first_lane : DefaultAttrsIntrinsic<[llvm_i1_ty], [], [IntrConvergent]>;
def int_dx_sign : DefaultAttrsIntrinsic<[LLVMScalarOrSameVectorWidth<0, llvm_i32_ty>], [llvm_any_ty], [IntrNoMem]>;
def int_dx_step : DefaultAttrsIntrinsic<[LLVMMatchType<0>], [llvm_anyfloat_ty, LLVMMatchType<0>], [IntrNoMem]>;
diff --git a/llvm/lib/Target/DirectX/DXIL.td b/llvm/lib/Target/DirectX/DXIL.td
index 9aa0af3e3a6b17..f0db3b4c165e41 100644
--- a/llvm/lib/Target/DirectX/DXIL.td
+++ b/llvm/lib/Target/DirectX/DXIL.td
@@ -801,3 +801,12 @@ def WaveIsFirstLane : DXILOp<110, waveIsFirstLane> {
let stages = [Stages<DXIL1_0, [all_stages]>];
let attributes = [Attributes<DXIL1_0, [ReadNone]>];
}
+
+def WaveGetLaneIndex : DXILOp<111, waveGetLaneIndex> {
+ let Doc = "returns the index of the current lane in the wave";
+ let LLVMIntrinsic = int_dx_waveGetLaneIndex;
+ let arguments = [];
+ let result = Int32Ty;
+ let stages = [Stages<DXIL1_0, [all_stages]>];
+ let attributes = [Attributes<DXIL1_0, [ReadNone]>];
+}
diff --git a/llvm/test/CodeGen/DirectX/WaveGetLaneIndex.ll b/llvm/test/CodeGen/DirectX/WaveGetLaneIndex.ll
new file mode 100644
index 00000000000000..0b1c96976f594a
--- /dev/null
+++ b/llvm/test/CodeGen/DirectX/WaveGetLaneIndex.ll
@@ -0,0 +1,10 @@
+; RUN: opt -S -dxil-op-lower -mtriple=dxil-pc-shadermodel6.3-compute %s | FileCheck %s
+
+define void @main() {
+entry:
+; CHECK: call i32 @dx.op.waveGetLaneIndex(i32 111)
+ %0 = call i32 @llvm.dx.waveGetLaneIndex()
+ ret void
+}
+
+declare i32 @llvm.dx.waveGetLaneIndex()
|
|
@llvm/pr-subscribers-clang-codegen Author: Finn Plummer (inbelic) ChangesResolves #70105 Full diff: https://github.com/llvm/llvm-project/pull/111576.diff 7 Files Affected:
diff --git a/clang/lib/CodeGen/CGBuiltin.cpp b/clang/lib/CodeGen/CGBuiltin.cpp
index da3eca73bfb575..803bdbaa0e7e01 100644
--- a/clang/lib/CodeGen/CGBuiltin.cpp
+++ b/clang/lib/CodeGen/CGBuiltin.cpp
@@ -18827,9 +18827,21 @@ case Builtin::BI__builtin_hlsl_elementwise_isinf: {
ArrayRef<Value *>{Op0, Op1}, nullptr, "hlsl.step");
}
case Builtin::BI__builtin_hlsl_wave_get_lane_index: {
- return EmitRuntimeCall(CGM.CreateRuntimeFunction(
- llvm::FunctionType::get(IntTy, {}, false), "__hlsl_wave_get_lane_index",
- {}, false, true));
+ // Since we don't define a SPIR-V intrinsic for the SPIR-V built-in from
+ // SPIRVBuiltins.td, manually get the matching name for the DirectX
+ // intrinsic and the demangled builtin name
+ switch (CGM.getTarget().getTriple().getArch()) {
+ case llvm::Triple::dxil:
+ return EmitRuntimeCall(Intrinsic::getDeclaration(
+ &CGM.getModule(), Intrinsic::dx_waveGetLaneIndex));
+ case llvm::Triple::spirv:
+ return EmitRuntimeCall(CGM.CreateRuntimeFunction(
+ llvm::FunctionType::get(IntTy, {}, false),
+ "__hlsl_wave_get_lane_index", {}, false, true));
+ default:
+ llvm_unreachable(
+ "Intrinsic WaveGetLaneIndex not supported by target architecture");
+ }
}
case Builtin::BI__builtin_hlsl_wave_is_first_lane: {
Intrinsic::ID ID = CGM.getHLSLRuntime().getWaveIsFirstLaneIntrinsic();
diff --git a/clang/lib/Sema/SemaHLSL.cpp b/clang/lib/Sema/SemaHLSL.cpp
index 43cc6c81ae5cb0..7d93f41bb2d7a0 100644
--- a/clang/lib/Sema/SemaHLSL.cpp
+++ b/clang/lib/Sema/SemaHLSL.cpp
@@ -1956,6 +1956,11 @@ bool SemaHLSL::CheckBuiltinFunctionCall(unsigned BuiltinID, CallExpr *TheCall) {
return true;
break;
}
+ case Builtin::BI__builtin_hlsl_wave_get_lane_index: {
+ if (SemaRef.checkArgCount(TheCall, 0))
+ return true;
+ break;
+ }
case Builtin::BI__builtin_elementwise_acos:
case Builtin::BI__builtin_elementwise_asin:
case Builtin::BI__builtin_elementwise_atan:
diff --git a/clang/test/CodeGenHLSL/builtins/wave_get_lane_index_simple.hlsl b/clang/test/CodeGenHLSL/builtins/wave_get_lane_index_simple.hlsl
index 8f52d81091c180..e76d35a86e512a 100644
--- a/clang/test/CodeGenHLSL/builtins/wave_get_lane_index_simple.hlsl
+++ b/clang/test/CodeGenHLSL/builtins/wave_get_lane_index_simple.hlsl
@@ -1,14 +1,22 @@
// RUN: %clang_cc1 -std=hlsl2021 -finclude-default-header -x hlsl -triple \
-// RUN: spirv-pc-vulkan-library %s -emit-llvm -disable-llvm-passes -o - | FileCheck %s
+// RUN: spirv-pc-vulkan-library %s -emit-llvm -disable-llvm-passes -o - | FileCheck %s \
+// RUN: --check-prefixes=CHECK,CHECK-SPIRV
+// RUN: %clang_cc1 -finclude-default-header \
+// RUN: -triple dxil-pc-shadermodel6.3-library %s \
+// RUN: -emit-llvm -disable-llvm-passes -o - | FileCheck %s \
+// RUN: --check-prefixes=CHECK,CHECK-DXIL
-// CHECK: define spir_func noundef i32 @_Z6test_1v() [[A0:#[0-9]+]] {
-// CHECK: %[[CI:[0-9]+]] = call token @llvm.experimental.convergence.entry()
-// CHECK: call i32 @__hlsl_wave_get_lane_index() [ "convergencectrl"(token %[[CI]]) ]
-uint test_1() {
+// CHECK-SPIRV: define spir_func noundef i32 @{{.*test_1.*}}() [[A0:#[0-9]+]] {
+// CHECK-DXIL: define noundef i32 @{{.*test_1.*}}() [[A0:#[0-9]+]] {
+// CHECK-SPIRV: %[[CI:[0-9]+]] = call token @llvm.experimental.convergence.entry()
+// CHECK-SPIRV: call i32 @__hlsl_wave_get_lane_index() [ "convergencectrl"(token %[[CI]]) ]
+// CHECK-DXIL: call i32 @llvm.dx.waveGetLaneIndex()
+int test_1() {
return WaveGetLaneIndex();
}
-// CHECK: declare i32 @__hlsl_wave_get_lane_index() [[A1:#[0-9]+]]
+// CHECK-SPIRV: declare i32 @__hlsl_wave_get_lane_index() [[A1:#[0-9]+]]
+// CHECK-DXIL: declare i32 @llvm.dx.waveGetLaneIndex() [[A1:#[0-9]+]]
// CHECK-DAG: attributes [[A0]] = { {{.*}}convergent{{.*}} }
// CHECK-DAG: attributes [[A1]] = { {{.*}}convergent{{.*}} }
diff --git a/clang/test/SemaHLSL/BuiltIns/WaveGetLaneIndex-errors.hlsl b/clang/test/SemaHLSL/BuiltIns/WaveGetLaneIndex-errors.hlsl
new file mode 100644
index 00000000000000..94cfd0662b5fc3
--- /dev/null
+++ b/clang/test/SemaHLSL/BuiltIns/WaveGetLaneIndex-errors.hlsl
@@ -0,0 +1,6 @@
+// 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
+
+int test_too_many_arg(int x) {
+ return __builtin_hlsl_wave_get_lane_index(x);
+ // expected-error@-1 {{too many arguments to function call, expected 0, have 1}}
+}
diff --git a/llvm/include/llvm/IR/IntrinsicsDirectX.td b/llvm/include/llvm/IR/IntrinsicsDirectX.td
index 555877e7aaf0e5..585cb2b7b300df 100644
--- a/llvm/include/llvm/IR/IntrinsicsDirectX.td
+++ b/llvm/include/llvm/IR/IntrinsicsDirectX.td
@@ -82,6 +82,7 @@ def int_dx_imad : DefaultAttrsIntrinsic<[llvm_anyint_ty], [LLVMMatchType<0>, LLV
def int_dx_umad : DefaultAttrsIntrinsic<[llvm_anyint_ty], [LLVMMatchType<0>, LLVMMatchType<0>, LLVMMatchType<0>], [IntrNoMem]>;
def int_dx_normalize : DefaultAttrsIntrinsic<[LLVMMatchType<0>], [llvm_anyfloat_ty], [IntrNoMem]>;
def int_dx_rsqrt : DefaultAttrsIntrinsic<[llvm_anyfloat_ty], [LLVMMatchType<0>], [IntrNoMem]>;
+def int_dx_waveGetLaneIndex : DefaultAttrsIntrinsic<[llvm_i32_ty], [], [IntrConvergent, IntrNoMem]>;
def int_dx_wave_is_first_lane : DefaultAttrsIntrinsic<[llvm_i1_ty], [], [IntrConvergent]>;
def int_dx_sign : DefaultAttrsIntrinsic<[LLVMScalarOrSameVectorWidth<0, llvm_i32_ty>], [llvm_any_ty], [IntrNoMem]>;
def int_dx_step : DefaultAttrsIntrinsic<[LLVMMatchType<0>], [llvm_anyfloat_ty, LLVMMatchType<0>], [IntrNoMem]>;
diff --git a/llvm/lib/Target/DirectX/DXIL.td b/llvm/lib/Target/DirectX/DXIL.td
index 9aa0af3e3a6b17..f0db3b4c165e41 100644
--- a/llvm/lib/Target/DirectX/DXIL.td
+++ b/llvm/lib/Target/DirectX/DXIL.td
@@ -801,3 +801,12 @@ def WaveIsFirstLane : DXILOp<110, waveIsFirstLane> {
let stages = [Stages<DXIL1_0, [all_stages]>];
let attributes = [Attributes<DXIL1_0, [ReadNone]>];
}
+
+def WaveGetLaneIndex : DXILOp<111, waveGetLaneIndex> {
+ let Doc = "returns the index of the current lane in the wave";
+ let LLVMIntrinsic = int_dx_waveGetLaneIndex;
+ let arguments = [];
+ let result = Int32Ty;
+ let stages = [Stages<DXIL1_0, [all_stages]>];
+ let attributes = [Attributes<DXIL1_0, [ReadNone]>];
+}
diff --git a/llvm/test/CodeGen/DirectX/WaveGetLaneIndex.ll b/llvm/test/CodeGen/DirectX/WaveGetLaneIndex.ll
new file mode 100644
index 00000000000000..0b1c96976f594a
--- /dev/null
+++ b/llvm/test/CodeGen/DirectX/WaveGetLaneIndex.ll
@@ -0,0 +1,10 @@
+; RUN: opt -S -dxil-op-lower -mtriple=dxil-pc-shadermodel6.3-compute %s | FileCheck %s
+
+define void @main() {
+entry:
+; CHECK: call i32 @dx.op.waveGetLaneIndex(i32 111)
+ %0 = call i32 @llvm.dx.waveGetLaneIndex()
+ ret void
+}
+
+declare i32 @llvm.dx.waveGetLaneIndex()
|
|
@llvm/pr-subscribers-backend-directx Author: Finn Plummer (inbelic) ChangesResolves #70105 Full diff: https://github.com/llvm/llvm-project/pull/111576.diff 7 Files Affected:
diff --git a/clang/lib/CodeGen/CGBuiltin.cpp b/clang/lib/CodeGen/CGBuiltin.cpp
index da3eca73bfb575..803bdbaa0e7e01 100644
--- a/clang/lib/CodeGen/CGBuiltin.cpp
+++ b/clang/lib/CodeGen/CGBuiltin.cpp
@@ -18827,9 +18827,21 @@ case Builtin::BI__builtin_hlsl_elementwise_isinf: {
ArrayRef<Value *>{Op0, Op1}, nullptr, "hlsl.step");
}
case Builtin::BI__builtin_hlsl_wave_get_lane_index: {
- return EmitRuntimeCall(CGM.CreateRuntimeFunction(
- llvm::FunctionType::get(IntTy, {}, false), "__hlsl_wave_get_lane_index",
- {}, false, true));
+ // Since we don't define a SPIR-V intrinsic for the SPIR-V built-in from
+ // SPIRVBuiltins.td, manually get the matching name for the DirectX
+ // intrinsic and the demangled builtin name
+ switch (CGM.getTarget().getTriple().getArch()) {
+ case llvm::Triple::dxil:
+ return EmitRuntimeCall(Intrinsic::getDeclaration(
+ &CGM.getModule(), Intrinsic::dx_waveGetLaneIndex));
+ case llvm::Triple::spirv:
+ return EmitRuntimeCall(CGM.CreateRuntimeFunction(
+ llvm::FunctionType::get(IntTy, {}, false),
+ "__hlsl_wave_get_lane_index", {}, false, true));
+ default:
+ llvm_unreachable(
+ "Intrinsic WaveGetLaneIndex not supported by target architecture");
+ }
}
case Builtin::BI__builtin_hlsl_wave_is_first_lane: {
Intrinsic::ID ID = CGM.getHLSLRuntime().getWaveIsFirstLaneIntrinsic();
diff --git a/clang/lib/Sema/SemaHLSL.cpp b/clang/lib/Sema/SemaHLSL.cpp
index 43cc6c81ae5cb0..7d93f41bb2d7a0 100644
--- a/clang/lib/Sema/SemaHLSL.cpp
+++ b/clang/lib/Sema/SemaHLSL.cpp
@@ -1956,6 +1956,11 @@ bool SemaHLSL::CheckBuiltinFunctionCall(unsigned BuiltinID, CallExpr *TheCall) {
return true;
break;
}
+ case Builtin::BI__builtin_hlsl_wave_get_lane_index: {
+ if (SemaRef.checkArgCount(TheCall, 0))
+ return true;
+ break;
+ }
case Builtin::BI__builtin_elementwise_acos:
case Builtin::BI__builtin_elementwise_asin:
case Builtin::BI__builtin_elementwise_atan:
diff --git a/clang/test/CodeGenHLSL/builtins/wave_get_lane_index_simple.hlsl b/clang/test/CodeGenHLSL/builtins/wave_get_lane_index_simple.hlsl
index 8f52d81091c180..e76d35a86e512a 100644
--- a/clang/test/CodeGenHLSL/builtins/wave_get_lane_index_simple.hlsl
+++ b/clang/test/CodeGenHLSL/builtins/wave_get_lane_index_simple.hlsl
@@ -1,14 +1,22 @@
// RUN: %clang_cc1 -std=hlsl2021 -finclude-default-header -x hlsl -triple \
-// RUN: spirv-pc-vulkan-library %s -emit-llvm -disable-llvm-passes -o - | FileCheck %s
+// RUN: spirv-pc-vulkan-library %s -emit-llvm -disable-llvm-passes -o - | FileCheck %s \
+// RUN: --check-prefixes=CHECK,CHECK-SPIRV
+// RUN: %clang_cc1 -finclude-default-header \
+// RUN: -triple dxil-pc-shadermodel6.3-library %s \
+// RUN: -emit-llvm -disable-llvm-passes -o - | FileCheck %s \
+// RUN: --check-prefixes=CHECK,CHECK-DXIL
-// CHECK: define spir_func noundef i32 @_Z6test_1v() [[A0:#[0-9]+]] {
-// CHECK: %[[CI:[0-9]+]] = call token @llvm.experimental.convergence.entry()
-// CHECK: call i32 @__hlsl_wave_get_lane_index() [ "convergencectrl"(token %[[CI]]) ]
-uint test_1() {
+// CHECK-SPIRV: define spir_func noundef i32 @{{.*test_1.*}}() [[A0:#[0-9]+]] {
+// CHECK-DXIL: define noundef i32 @{{.*test_1.*}}() [[A0:#[0-9]+]] {
+// CHECK-SPIRV: %[[CI:[0-9]+]] = call token @llvm.experimental.convergence.entry()
+// CHECK-SPIRV: call i32 @__hlsl_wave_get_lane_index() [ "convergencectrl"(token %[[CI]]) ]
+// CHECK-DXIL: call i32 @llvm.dx.waveGetLaneIndex()
+int test_1() {
return WaveGetLaneIndex();
}
-// CHECK: declare i32 @__hlsl_wave_get_lane_index() [[A1:#[0-9]+]]
+// CHECK-SPIRV: declare i32 @__hlsl_wave_get_lane_index() [[A1:#[0-9]+]]
+// CHECK-DXIL: declare i32 @llvm.dx.waveGetLaneIndex() [[A1:#[0-9]+]]
// CHECK-DAG: attributes [[A0]] = { {{.*}}convergent{{.*}} }
// CHECK-DAG: attributes [[A1]] = { {{.*}}convergent{{.*}} }
diff --git a/clang/test/SemaHLSL/BuiltIns/WaveGetLaneIndex-errors.hlsl b/clang/test/SemaHLSL/BuiltIns/WaveGetLaneIndex-errors.hlsl
new file mode 100644
index 00000000000000..94cfd0662b5fc3
--- /dev/null
+++ b/clang/test/SemaHLSL/BuiltIns/WaveGetLaneIndex-errors.hlsl
@@ -0,0 +1,6 @@
+// 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
+
+int test_too_many_arg(int x) {
+ return __builtin_hlsl_wave_get_lane_index(x);
+ // expected-error@-1 {{too many arguments to function call, expected 0, have 1}}
+}
diff --git a/llvm/include/llvm/IR/IntrinsicsDirectX.td b/llvm/include/llvm/IR/IntrinsicsDirectX.td
index 555877e7aaf0e5..585cb2b7b300df 100644
--- a/llvm/include/llvm/IR/IntrinsicsDirectX.td
+++ b/llvm/include/llvm/IR/IntrinsicsDirectX.td
@@ -82,6 +82,7 @@ def int_dx_imad : DefaultAttrsIntrinsic<[llvm_anyint_ty], [LLVMMatchType<0>, LLV
def int_dx_umad : DefaultAttrsIntrinsic<[llvm_anyint_ty], [LLVMMatchType<0>, LLVMMatchType<0>, LLVMMatchType<0>], [IntrNoMem]>;
def int_dx_normalize : DefaultAttrsIntrinsic<[LLVMMatchType<0>], [llvm_anyfloat_ty], [IntrNoMem]>;
def int_dx_rsqrt : DefaultAttrsIntrinsic<[llvm_anyfloat_ty], [LLVMMatchType<0>], [IntrNoMem]>;
+def int_dx_waveGetLaneIndex : DefaultAttrsIntrinsic<[llvm_i32_ty], [], [IntrConvergent, IntrNoMem]>;
def int_dx_wave_is_first_lane : DefaultAttrsIntrinsic<[llvm_i1_ty], [], [IntrConvergent]>;
def int_dx_sign : DefaultAttrsIntrinsic<[LLVMScalarOrSameVectorWidth<0, llvm_i32_ty>], [llvm_any_ty], [IntrNoMem]>;
def int_dx_step : DefaultAttrsIntrinsic<[LLVMMatchType<0>], [llvm_anyfloat_ty, LLVMMatchType<0>], [IntrNoMem]>;
diff --git a/llvm/lib/Target/DirectX/DXIL.td b/llvm/lib/Target/DirectX/DXIL.td
index 9aa0af3e3a6b17..f0db3b4c165e41 100644
--- a/llvm/lib/Target/DirectX/DXIL.td
+++ b/llvm/lib/Target/DirectX/DXIL.td
@@ -801,3 +801,12 @@ def WaveIsFirstLane : DXILOp<110, waveIsFirstLane> {
let stages = [Stages<DXIL1_0, [all_stages]>];
let attributes = [Attributes<DXIL1_0, [ReadNone]>];
}
+
+def WaveGetLaneIndex : DXILOp<111, waveGetLaneIndex> {
+ let Doc = "returns the index of the current lane in the wave";
+ let LLVMIntrinsic = int_dx_waveGetLaneIndex;
+ let arguments = [];
+ let result = Int32Ty;
+ let stages = [Stages<DXIL1_0, [all_stages]>];
+ let attributes = [Attributes<DXIL1_0, [ReadNone]>];
+}
diff --git a/llvm/test/CodeGen/DirectX/WaveGetLaneIndex.ll b/llvm/test/CodeGen/DirectX/WaveGetLaneIndex.ll
new file mode 100644
index 00000000000000..0b1c96976f594a
--- /dev/null
+++ b/llvm/test/CodeGen/DirectX/WaveGetLaneIndex.ll
@@ -0,0 +1,10 @@
+; RUN: opt -S -dxil-op-lower -mtriple=dxil-pc-shadermodel6.3-compute %s | FileCheck %s
+
+define void @main() {
+entry:
+; CHECK: call i32 @dx.op.waveGetLaneIndex(i32 111)
+ %0 = call i32 @llvm.dx.waveGetLaneIndex()
+ ret void
+}
+
+declare i32 @llvm.dx.waveGetLaneIndex()
|
pow2clk
reviewed
Oct 9, 2024
- remove flags that are not needed
- update intrinsic naming to wave_getlaneindex to follow llvm conventions
farzonl
approved these changes
Oct 9, 2024
pow2clk
approved these changes
Oct 10, 2024
DanielCChen
pushed a commit
to DanielCChen/llvm-project
that referenced
this pull request
Oct 16, 2024
- add additional lowering for directx backend in CGBuiltin.cpp
- add directx intrinsic to IntrinsicsDirectX.td
- add semantic check of arguments in SemaHLSL.cpp
- add mapping to DXIL op in DXIL.td
- add testing of semantics in WaveGetLaneIndex-errors.hlsl
- add testing of dxil lowering in WaveGetLaneIndex.ll
Resolves llvm#70105
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Resolves #70105