[clang] [HLSL] Use icmp ne instead of trunc on boolean load (PR #221105)
Dan Brown via cfe-commits
cfe-commits at lists.llvm.org
Wed Oct 14 23:06:02 PDT 2026
https://github.com/danbrown-amd updated https://github.com/llvm/llvm-project/pull/221105
>From fd03dab0bfc9b9c3acbe02f83850bfa70284c08a Mon Sep 17 00:00:00 2001
From: danbrown-amd <danbrown at amd.com>
Date: Thu, 3 Sep 2026 18:46:23 -0600
Subject: [PATCH 1/4] Use icmp-ne instead of truncation on boolean load
Fixes #140824
Assisted-by: Claude Sonnet 4
---
clang/lib/CodeGen/CGExpr.cpp | 5 +++--
clang/lib/Frontend/CompilerInvocation.cpp | 6 ++++++
clang/test/CodeGenHLSL/strict-bool.hlsl | 2 +-
3 files changed, 10 insertions(+), 3 deletions(-)
diff --git a/clang/lib/CodeGen/CGExpr.cpp b/clang/lib/CodeGen/CGExpr.cpp
index 019159dd620fda..f5d053e6b67130 100644
--- a/clang/lib/CodeGen/CGExpr.cpp
+++ b/clang/lib/CodeGen/CGExpr.cpp
@@ -5681,9 +5681,10 @@ EmitExtVectorElementExpr(const ExtVectorElementExpr *E) {
Address VecMem = CreateMemTemp(E->getBase()->getType());
// need to zero extend an hlsl boolean vector to store it back to memory
QualType Ty = E->getBase()->getType();
- llvm::Type *LTy = convertTypeForLoadStore(Ty, Vec->getType());
- if (LTy->getScalarSizeInBits() > Vec->getType()->getScalarSizeInBits())
+ if (getLangOpts().HLSL && Ty->isExtVectorBoolType()) {
+ llvm::Type *LTy = convertTypeForLoadStore(Ty, Vec->getType());
Vec = Builder.CreateZExt(Vec, LTy);
+ }
Builder.CreateStore(Vec, VecMem);
Base = MakeAddrLValue(VecMem, Ty, AlignmentSource::Decl);
}
diff --git a/clang/lib/Frontend/CompilerInvocation.cpp b/clang/lib/Frontend/CompilerInvocation.cpp
index c9281c2a6d6b41..0a3a235fbac33c 100644
--- a/clang/lib/Frontend/CompilerInvocation.cpp
+++ b/clang/lib/Frontend/CompilerInvocation.cpp
@@ -657,6 +657,12 @@ static bool FixupInvocation(CompilerInvocation &Invocation,
Warnings.insert(Warnings.begin(), "vector-conversion");
if (!llvm::is_contained(Warnings, "matrix-conversion"))
Warnings.insert(Warnings.begin(), "matrix-conversion");
+
+ // HLSL booleans in memory are i32; any non-zero value is true.
+ // Use icmp-ne rather than truncation to avoid InstCombine rewriting
+ // trunc i32 to i1 into an illegal wide <N x i1> bitcast (issue #140824).
+ if (!Args.hasArg(OPT_load_bool_from_mem))
+ CodeGenOpts.setLoadBoolFromMem(CodeGenOptions::BoolFromMem::NonZero);
}
// When these options are used, the compiler is allowed to apply
diff --git a/clang/test/CodeGenHLSL/strict-bool.hlsl b/clang/test/CodeGenHLSL/strict-bool.hlsl
index 31867065e77a3b..30bdc02231fa5a 100644
--- a/clang/test/CodeGenHLSL/strict-bool.hlsl
+++ b/clang/test/CodeGenHLSL/strict-bool.hlsl
@@ -1,6 +1,6 @@
// RUN: %clang_cc1 -triple dxil-pc-shadermodel6.3-library -O1 -load-bool-from-mem=truncate -emit-llvm -o - %s | FileCheck %s -check-prefixes=CHECK-TRUNCATE
// RUN: %clang_cc1 -triple dxil-pc-shadermodel6.3-library -O1 -load-bool-from-mem=nonzero -emit-llvm -o - %s | FileCheck %s -check-prefixes=CHECK-NONZERO
-// RUN: %clang_cc1 -triple dxil-pc-shadermodel6.3-library -O0 -emit-llvm -o - %s | FileCheck %s -check-prefixes=CHECK-NONZERO
+// RUN: %clang_cc1 -triple dxil-pc-shadermodel6.3-library -O1 -emit-llvm -o - %s | FileCheck %s -check-prefixes=CHECK-NONZERO
typedef bool bool8_t __attribute__((ext_vector_type(8)));
extern bool8_t vec;
>From a54e001a8ec8dc38aaaed2eed75b54af43a15fa8 Mon Sep 17 00:00:00 2001
From: danbrown-amd <danbrown at amd.com>
Date: Thu, 8 Oct 2026 16:30:54 -0600
Subject: [PATCH 2/4] Revert test and one source file as per review comments
---
clang/lib/CodeGen/CGExpr.cpp | 5 ++---
clang/test/CodeGenHLSL/strict-bool.hlsl | 2 +-
2 files changed, 3 insertions(+), 4 deletions(-)
diff --git a/clang/lib/CodeGen/CGExpr.cpp b/clang/lib/CodeGen/CGExpr.cpp
index f5d053e6b67130..019159dd620fda 100644
--- a/clang/lib/CodeGen/CGExpr.cpp
+++ b/clang/lib/CodeGen/CGExpr.cpp
@@ -5681,10 +5681,9 @@ EmitExtVectorElementExpr(const ExtVectorElementExpr *E) {
Address VecMem = CreateMemTemp(E->getBase()->getType());
// need to zero extend an hlsl boolean vector to store it back to memory
QualType Ty = E->getBase()->getType();
- if (getLangOpts().HLSL && Ty->isExtVectorBoolType()) {
- llvm::Type *LTy = convertTypeForLoadStore(Ty, Vec->getType());
+ llvm::Type *LTy = convertTypeForLoadStore(Ty, Vec->getType());
+ if (LTy->getScalarSizeInBits() > Vec->getType()->getScalarSizeInBits())
Vec = Builder.CreateZExt(Vec, LTy);
- }
Builder.CreateStore(Vec, VecMem);
Base = MakeAddrLValue(VecMem, Ty, AlignmentSource::Decl);
}
diff --git a/clang/test/CodeGenHLSL/strict-bool.hlsl b/clang/test/CodeGenHLSL/strict-bool.hlsl
index 30bdc02231fa5a..31867065e77a3b 100644
--- a/clang/test/CodeGenHLSL/strict-bool.hlsl
+++ b/clang/test/CodeGenHLSL/strict-bool.hlsl
@@ -1,6 +1,6 @@
// RUN: %clang_cc1 -triple dxil-pc-shadermodel6.3-library -O1 -load-bool-from-mem=truncate -emit-llvm -o - %s | FileCheck %s -check-prefixes=CHECK-TRUNCATE
// RUN: %clang_cc1 -triple dxil-pc-shadermodel6.3-library -O1 -load-bool-from-mem=nonzero -emit-llvm -o - %s | FileCheck %s -check-prefixes=CHECK-NONZERO
-// RUN: %clang_cc1 -triple dxil-pc-shadermodel6.3-library -O1 -emit-llvm -o - %s | FileCheck %s -check-prefixes=CHECK-NONZERO
+// RUN: %clang_cc1 -triple dxil-pc-shadermodel6.3-library -O0 -emit-llvm -o - %s | FileCheck %s -check-prefixes=CHECK-NONZERO
typedef bool bool8_t __attribute__((ext_vector_type(8)));
extern bool8_t vec;
>From 3591f213ca47a3e79e4f86de7a5d8e561250d358 Mon Sep 17 00:00:00 2001
From: danbrown-amd <danbrown at amd.com>
Date: Tue, 13 Oct 2026 23:10:48 -0600
Subject: [PATCH 3/4] Address additional review comments
---
clang/include/clang/Basic/DiagnosticFrontendKinds.td | 3 +++
clang/lib/Driver/ToolChains/HLSL.cpp | 1 +
clang/lib/Frontend/CompilerInvocation.cpp | 4 ++++
clang/test/Driver/HLSL/load-bool-from-mem.hlsl | 3 +++
4 files changed, 11 insertions(+)
create mode 100644 clang/test/Driver/HLSL/load-bool-from-mem.hlsl
diff --git a/clang/include/clang/Basic/DiagnosticFrontendKinds.td b/clang/include/clang/Basic/DiagnosticFrontendKinds.td
index 66eeb08034bf4b..2281836406e63e 100644
--- a/clang/include/clang/Basic/DiagnosticFrontendKinds.td
+++ b/clang/include/clang/Basic/DiagnosticFrontendKinds.td
@@ -514,6 +514,9 @@ def warn_hlsl_signature_packing_ignored : Warning<
def warn_hlsl_failed_to_embed_source
: Warning<"failed to embed source for '%0' into dx.source.contents">,
InGroup<HLSLDXCCompat>;
+def warn_hlsl_bool_from_mem
+ : Warning<"-load-bool-from-mem=%0 ignored for HLSL (always nonzero)">,
+ InGroup<OptionIgnored>, DefaultWarn;
def err_data_layout_mismatch : Error<"backend data layout '%0' does not match "
"expected target description '%1'">;
diff --git a/clang/lib/Driver/ToolChains/HLSL.cpp b/clang/lib/Driver/ToolChains/HLSL.cpp
index b5356f1ec45d33..7b0be0ab1f4020 100644
--- a/clang/lib/Driver/ToolChains/HLSL.cpp
+++ b/clang/lib/Driver/ToolChains/HLSL.cpp
@@ -689,4 +689,5 @@ void HLSLToolChain::addClangWarningOptions(ArgStringList &CC1Args) const {
CC1Args.push_back("-Wconversion");
CC1Args.push_back("-Wvector-conversion");
CC1Args.push_back("-Wmatrix-conversion");
+ CC1Args.push_back("-load-bool-from-mem=nonzero");
}
diff --git a/clang/lib/Frontend/CompilerInvocation.cpp b/clang/lib/Frontend/CompilerInvocation.cpp
index 0a3a235fbac33c..49246f7b6d4d04 100644
--- a/clang/lib/Frontend/CompilerInvocation.cpp
+++ b/clang/lib/Frontend/CompilerInvocation.cpp
@@ -663,6 +663,10 @@ static bool FixupInvocation(CompilerInvocation &Invocation,
// trunc i32 to i1 into an illegal wide <N x i1> bitcast (issue #140824).
if (!Args.hasArg(OPT_load_bool_from_mem))
CodeGenOpts.setLoadBoolFromMem(CodeGenOptions::BoolFromMem::NonZero);
+ else if (CodeGenOpts.getLoadBoolFromMem() !=
+ CodeGenOptions::BoolFromMem::NonZero)
+ Diags.Report(diag::warn_hlsl_bool_from_mem)
+ << Args.getLastArg(OPT_load_bool_from_mem)->getValue();
}
// When these options are used, the compiler is allowed to apply
diff --git a/clang/test/Driver/HLSL/load-bool-from-mem.hlsl b/clang/test/Driver/HLSL/load-bool-from-mem.hlsl
new file mode 100644
index 00000000000000..23ee8cca8b6def
--- /dev/null
+++ b/clang/test/Driver/HLSL/load-bool-from-mem.hlsl
@@ -0,0 +1,3 @@
+// RUN: %clang_dxc -T cs_6_3 -### %s 2>&1 | FileCheck %s
+
+// CHECK: "-load-bool-from-mem=nonzero"
>From 172a354f0fac1c64949f611fdb9489e00213ca5b Mon Sep 17 00:00:00 2001
From: danbrown-amd <danbrown at amd.com>
Date: Wed, 14 Oct 2026 21:55:05 -0600
Subject: [PATCH 4/4] Address further review comments
---
clang/lib/Frontend/CompilerInvocation.cpp | 8 ++--
clang/test/CodeGenHLSL/bool-vector-load.hlsl | 16 +++++++
.../CodeGenHLSL/builtins/BoolSwizzles.hlsl | 45 +++++++------------
clang/test/CodeGenHLSL/strict-bool.hlsl | 15 +++----
4 files changed, 42 insertions(+), 42 deletions(-)
create mode 100644 clang/test/CodeGenHLSL/bool-vector-load.hlsl
diff --git a/clang/lib/Frontend/CompilerInvocation.cpp b/clang/lib/Frontend/CompilerInvocation.cpp
index 49246f7b6d4d04..57bad35e51f610 100644
--- a/clang/lib/Frontend/CompilerInvocation.cpp
+++ b/clang/lib/Frontend/CompilerInvocation.cpp
@@ -661,12 +661,12 @@ static bool FixupInvocation(CompilerInvocation &Invocation,
// HLSL booleans in memory are i32; any non-zero value is true.
// Use icmp-ne rather than truncation to avoid InstCombine rewriting
// trunc i32 to i1 into an illegal wide <N x i1> bitcast (issue #140824).
- if (!Args.hasArg(OPT_load_bool_from_mem))
- CodeGenOpts.setLoadBoolFromMem(CodeGenOptions::BoolFromMem::NonZero);
- else if (CodeGenOpts.getLoadBoolFromMem() !=
- CodeGenOptions::BoolFromMem::NonZero)
+ if (Args.hasArg(OPT_load_bool_from_mem) &&
+ CodeGenOpts.getLoadBoolFromMem() !=
+ CodeGenOptions::BoolFromMem::NonZero)
Diags.Report(diag::warn_hlsl_bool_from_mem)
<< Args.getLastArg(OPT_load_bool_from_mem)->getValue();
+ CodeGenOpts.setLoadBoolFromMem(CodeGenOptions::BoolFromMem::NonZero);
}
// When these options are used, the compiler is allowed to apply
diff --git a/clang/test/CodeGenHLSL/bool-vector-load.hlsl b/clang/test/CodeGenHLSL/bool-vector-load.hlsl
new file mode 100644
index 00000000000000..d78db854310d83
--- /dev/null
+++ b/clang/test/CodeGenHLSL/bool-vector-load.hlsl
@@ -0,0 +1,16 @@
+// RUN: %clang_dxc -T cs_6_6 %s | FileCheck %s
+
+StructuredBuffer<bool4> In : register(t0);
+RWStructuredBuffer<bool4> Out : register(u1);
+
+// CHECK-LABEL: define {{.*}}main
+// CHECK: [[Load:%.*]] = call {{[^ ]*}} @dx.op.rawBufferLoad.i32(i32 139, %dx.types.Handle %{{[^,]*}}, i32 0, i32 12, i8 1, i32 4)
+// CHECK: [[Elem:%.*]] = extractvalue {{.*}} [[Load]], 0
+// CHECK-NEXT: [[Bool:%.*]] = icmp ne i32 [[Elem]], 0
+// CHECK: [[Wide:%.*]] = zext i1 [[Bool]] to i32
+// CHECK: call void @dx.op.rawBufferStore.i32(i32 140, %dx.types.Handle %{{[^,]*}},{{[^,]*}},{{[^,]*}}, i32 [[Wide]]
+[numthreads(1,1,1)]
+void main(void) {
+ Out[1].w = In[0].w;
+}
+
diff --git a/clang/test/CodeGenHLSL/builtins/BoolSwizzles.hlsl b/clang/test/CodeGenHLSL/builtins/BoolSwizzles.hlsl
index 88058255e94c3d..191a6ad1d604c5 100644
--- a/clang/test/CodeGenHLSL/builtins/BoolSwizzles.hlsl
+++ b/clang/test/CodeGenHLSL/builtins/BoolSwizzles.hlsl
@@ -1,12 +1,10 @@
-// RUN: %clang_cc1 -std=hlsl2021 -finclude-default-header -x hlsl -load-bool-from-mem=truncate -triple dxil-pc-shadermodel6.3-library %s -emit-llvm -disable-llvm-passes -o - | FileCheck -check-prefixes=CHECK,CHECK-TR %s
-// RUN: %clang_cc1 -std=hlsl2021 -finclude-default-header -x hlsl -load-bool-from-mem=nonzero -triple dxil-pc-shadermodel6.3-library %s -emit-llvm -disable-llvm-passes -o - | FileCheck -check-prefixes=CHECK,CHECK-NZ %s
+// RUN: %clang_cc1 -std=hlsl2021 -finclude-default-header -x hlsl -load-bool-from-mem=nonzero -triple dxil-pc-shadermodel6.3-library %s -emit-llvm -disable-llvm-passes -o - | FileCheck %s
// CHECK-LABEL: ToFourBools
// CHECK: {{%.*}} = zext i1 {{.*}} to i32
// CHECK: [[splat:%.*]] = insertelement <1 x i32> poison, i32 {{.*}}, i64 0
// CHECK-NEXT: [[vec4:%.*]] = shufflevector <1 x i32> [[splat]], <1 x i32> poison, <4 x i32> zeroinitializer
-// CHECK-NZ-NEXT: [[vec2Ret:%.*]] = icmp ne <4 x i32> [[vec4]], zeroinitializer
-// CHECK-TR-NEXT: [[vec2Ret:%.*]] = trunc <4 x i32> [[vec4]] to <4 x i1>
+// CHECK-NEXT: [[vec2Ret:%.*]] = icmp ne <4 x i32> [[vec4]], zeroinitializer
// CHECK-NEXT: ret <4 x i1> [[vec2Ret]]
bool4 ToFourBools(bool V) {
return V.rrrr;
@@ -17,8 +15,7 @@ bool4 ToFourBools(bool V) {
// CHECK-NEXT: store <1 x i32> splat (i32 1), ptr [[Tmp]], align 4
// CHECK-NEXT: [[Vec1:%.*]] = load <1 x i32>, ptr [[Tmp]], align 4
// CHECK-NEXT: [[Vec2:%.*]] = shufflevector <1 x i32> [[Vec1]], <1 x i32> poison, <2 x i32> zeroinitializer
-// CHECK-NZ-NEXT: [[Vec2Ret:%.*]] = icmp ne <2 x i32> [[Vec2]], zeroinitializer
-// CHECK-TR-NEXT: [[Vec2Ret:%.*]] = trunc <2 x i32> [[Vec2]] to <2 x i1>
+// CHECK-NEXT: [[Vec2Ret:%.*]] = icmp ne <2 x i32> [[Vec2]], zeroinitializer
// CHECK-NEXT: ret <2 x i1> [[Vec2Ret]]
bool2 FillTrue() {
return true.xx;
@@ -32,14 +29,12 @@ bool2 FillTrue() {
// CHECK-NEXT: [[VVal:%.*]] = load i32, ptr [[VAddr]], align 4
// CHECK-NEXT: [[Splat:%.*]] = insertelement <1 x i32> poison, i32 [[VVal]], i64 0
// CHECK-NEXT: [[Vec2:%.*]] = shufflevector <1 x i32> [[Splat]], <1 x i32> poison, <2 x i32> zeroinitializer
-// CHECK-NZ-NEXT: [[Trunc:%.*]] = icmp ne <2 x i32> [[Vec2]], zeroinitializer
-// CHECK-TR-NEXT: [[Trunc:%.*]] = trunc <2 x i32> [[Vec2]] to <2 x i1>
+// CHECK-NEXT: [[Trunc:%.*]] = icmp ne <2 x i32> [[Vec2]], zeroinitializer
// CHECK-NEXT: [[Ext:%.*]] = zext <2 x i1> [[Trunc]] to <2 x i32>
// CHECK-NEXT: store <2 x i32> [[Ext]], ptr [[Vec2Ptr]], align 4
// CHECK-NEXT: [[V2:%.*]] = load <2 x i32>, ptr [[Vec2Ptr]], align 4
// CHECK-NEXT: [[V3:%.*]] = shufflevector <2 x i32> [[V2]], <2 x i32> poison, <2 x i32> zeroinitializer
-// CHECK-NZ-NEXT: [[LV1:%.*]] = icmp ne <2 x i32> [[V3]], zeroinitializer
-// CHECK-TR-NEXT: [[LV1:%.*]] = trunc <2 x i32> [[V3]] to <2 x i1>
+// CHECK-NEXT: [[LV1:%.*]] = icmp ne <2 x i32> [[V3]], zeroinitializer
// CHECK-NEXT: ret <2 x i1> [[LV1]]
bool2 HowManyBools(bool V) {
return V.rr.rr;
@@ -53,23 +48,20 @@ bool2 HowManyBools(bool V) {
// CHECK-NEXT: [[X:%.*]] = load i32, ptr [[VAddr]], align 4
// CHECK-NEXT: [[Splat:%.*]] = insertelement <1 x i32> poison, i32 [[X]], i64 0
// CHECK-NEXT: [[Y:%.*]] = extractelement <1 x i32> [[Splat]], i32 0
-// CHECK-NZ-NEXT: [[Z:%.*]] = icmp ne i32 [[Y]], 0
-// CHECK-TR-NEXT: [[Z:%.*]] = trunc i32 [[Y]] to i1
+// CHECK-NEXT: [[Z:%.*]] = icmp ne i32 [[Y]], 0
// CHECK-NEXT: [[A:%.*]] = zext i1 [[Z]] to i32
// CHECK-NEXT: store i32 [[A]], ptr [[XAddr]], align 4
// CHECK-NEXT: [[B:%.*]] = load i32, ptr [[VAddr]], align 4
// CHECK-NEXT: [[Splat2:%.*]] = insertelement <1 x i32> poison, i32 [[B]], i64 0
// CHECK-NEXT: [[C:%.*]] = extractelement <1 x i32> [[Splat2]], i32 0
-// CHECK-NZ-NEXT: [[D:%.*]] = icmp ne i32 [[C]], 0
-// CHECK-TR-NEXT: [[D:%.*]] = trunc i32 [[C]] to i1
+// CHECK-NEXT: [[D:%.*]] = icmp ne i32 [[C]], 0
// CHECK-NEXT: br i1 [[D]], label %lor.end, label %lor.rhs
// CHECK: lor.rhs:
// CHECK-NEXT: [[E:%.*]] = load i32, ptr [[VAddr]], align 4
// CHECK-NEXT: [[Splat3:%.*]] = insertelement <1 x i32> poison, i32 [[E]], i64 0
// CHECK-NEXT: [[F:%.*]] = extractelement <1 x i32> [[Splat3]], i32 0
-// CHECK-NZ-NEXT: [[G:%.*]] = icmp ne i32 [[F]], 0
-// CHECK-TR-NEXT: [[G:%.*]] = trunc i32 [[F]] to i1
+// CHECK-NEXT: [[G:%.*]] = icmp ne i32 [[F]], 0
// CHECK-NEXT: br label %lor.end
// CHECK: lor.end:
@@ -77,8 +69,7 @@ bool2 HowManyBools(bool V) {
// CHECK-NEXT: [[J:%.*]] = zext i1 [[H]] to i32
// CHECK-NEXT: store i32 [[J]], ptr [[XAddr]], align 4
// CHECK-NEXT: [[I:%.*]] = load i32, ptr [[XAddr]], align 4
-// CHECK-NZ-NEXT: [[LoadV:%.*]] = icmp ne i32 [[I]], 0
-// CHECK-TR-NEXT: [[LoadV:%.*]] = trunc i32 [[I]] to i1
+// CHECK-NEXT: [[LoadV:%.*]] = icmp ne i32 [[I]], 0
// CHECK-NEXT: ret i1 [[LoadV]]
bool AssignBool(bool V) {
bool X = V.x;
@@ -95,13 +86,11 @@ bool AssignBool(bool V) {
// CHECK-NEXT: store <1 x i32> splat (i32 1), ptr [[Tmp]], align 4
// CHECK-NEXT: [[Y:%.*]] = load <1 x i32>, ptr [[Tmp]], align 4
// CHECK-NEXT: [[Z:%.*]] = shufflevector <1 x i32> [[Y]], <1 x i32> poison, <2 x i32> zeroinitializer
-// CHECK-NZ-NEXT: [[LV:%.*]] = icmp ne <2 x i32> [[Z]], zeroinitializer
-// CHECK-TR-NEXT: [[LV:%.*]] = trunc <2 x i32> [[Z]] to <2 x i1>
+// CHECK-NEXT: [[LV:%.*]] = icmp ne <2 x i32> [[Z]], zeroinitializer
// CHECK-NEXT: [[A:%.*]] = zext <2 x i1> [[LV]] to <2 x i32>
// CHECK-NEXT: store <2 x i32> [[A]], ptr [[X]], align 4
// CHECK-NEXT: [[B:%.*]] = load i32, ptr [[VAddr]], align 4
-// CHECK-NZ-NEXT: [[LV1:%.*]] = icmp ne i32 [[B]], 0
-// CHECK-TR-NEXT: [[LV1:%.*]] = trunc i32 [[B]] to i1
+// CHECK-NEXT: [[LV1:%.*]] = icmp ne i32 [[B]], 0
// CHECK-NEXT: [[D:%.*]] = zext i1 [[LV1]] to i32
// CHECK-NEXT: [[C:%.*]] = getelementptr <2 x i32>, ptr [[X]], i32 0, i32 1
// CHECK-NEXT: store i32 [[D]], ptr [[C]], align 4
@@ -118,8 +107,7 @@ void AssignBool2(bool V) {
// CHECK-NEXT: store <2 x i32> [[Y]], ptr [[VAddr]], align 4
// CHECK-NEXT: store <2 x i32> splat (i32 1), ptr [[X]], align 4
// CHECK-NEXT: [[Z:%.*]] = load <2 x i32>, ptr [[VAddr]], align 4
-// CHECK-NZ-NEXT: [[LV:%.*]] = icmp ne <2 x i32> [[Z]], zeroinitializer
-// CHECK-TR-NEXT: [[LV:%.*]] = trunc <2 x i32> [[Z]] to <2 x i1>
+// CHECK-NEXT: [[LV:%.*]] = icmp ne <2 x i32> [[Z]], zeroinitializer
// CHECK-NEXT: [[B:%.*]] = zext <2 x i1> [[LV]] to <2 x i32>
// CHECK-NEXT: [[V1:%.*]] = extractelement <2 x i32> [[B]], i32 0
// CHECK-NEXT: store i32 [[V1]], ptr [[X]], align 4
@@ -139,14 +127,12 @@ void AssignBool3(bool2 V) {
// CHECK-NEXT: store <1 x i32> splat (i32 1), ptr [[Tmp]], align 4
// CHECK-NEXT: [[Y:%.*]] = load <1 x i32>, ptr [[Tmp]], align 4
// CHECK-NEXT: [[Z:%.*]] = shufflevector <1 x i32> [[Y]], <1 x i32> poison, <4 x i32> zeroinitializer
-// CHECK-NZ-NEXT: [[LV:%.*]] = icmp ne <4 x i32> [[Z]], zeroinitializer
-// CHECK-TR-NEXT: [[LV:%.*]] = trunc <4 x i32> [[Vec2]] to <4 x i1>
+// CHECK-NEXT: [[LV:%.*]] = icmp ne <4 x i32> [[Z]], zeroinitializer
// CHECK-NEXT: [[A:%.*]] = zext <4 x i1> [[LV]] to <4 x i32>
// CHECK-NEXT: store <4 x i32> [[A]], ptr [[X]], align 4
// CHECK-NEXT: [[B:%.*]] = load <4 x i32>, ptr [[X]], align 4
// CHECK-NEXT: [[C:%.*]] = shufflevector <4 x i32> [[B]], <4 x i32> poison, <2 x i32> <i32 2, i32 3>
-// CHECK-NZ-NEXT: [[LV1:%.*]] = icmp ne <2 x i32> [[C]], zeroinitializer
-// CHECK-TR-NEXT: [[LV1:%.*]] = trunc <2 x i32> [[C]] to <2 x i1>
+// CHECK-NEXT: [[LV1:%.*]] = icmp ne <2 x i32> [[C]], zeroinitializer
// CHECK-NEXT: ret <2 x i1> [[LV1]]
bool2 AccessBools() {
bool4 X = true.xxxx;
@@ -160,8 +146,7 @@ bool2 AccessBools() {
// CHECK-NEXT: store <1 x i32> zeroinitializer, ptr [[Tmp]], align 4
// CHECK-NEXT: [[L0:%.*]] = load <1 x i32>, ptr [[Tmp]], align 4
// CHECK-NEXT: [[L1:%.*]] = shufflevector <1 x i32> [[L0]], <1 x i32> poison, <3 x i32> zeroinitializer
-// CHECK-NZ-NEXT: [[TruncV:%.*]] = icmp ne <3 x i32> [[L1]], zeroinitializer
-// CHECK-TR-NEXT: [[TruncV:%.*]] = trunc <3 x i32> [[Vec2]] to <3 x i1>
+// CHECK-NEXT: [[TruncV:%.*]] = icmp ne <3 x i32> [[L1]], zeroinitializer
// CHECK-NEXT: [[L2:%.*]] = zext <3 x i1> [[TruncV]] to <3 x i32>
// CHECK-NEXT: [[V1:%.*]] = extractelement <3 x i32> [[L2]], i32 0
// CHECK-NEXT: store i32 [[V1]], ptr %B, align 4
diff --git a/clang/test/CodeGenHLSL/strict-bool.hlsl b/clang/test/CodeGenHLSL/strict-bool.hlsl
index 31867065e77a3b..c9bacffc04469c 100644
--- a/clang/test/CodeGenHLSL/strict-bool.hlsl
+++ b/clang/test/CodeGenHLSL/strict-bool.hlsl
@@ -1,15 +1,14 @@
-// RUN: %clang_cc1 -triple dxil-pc-shadermodel6.3-library -O1 -load-bool-from-mem=truncate -emit-llvm -o - %s | FileCheck %s -check-prefixes=CHECK-TRUNCATE
-// RUN: %clang_cc1 -triple dxil-pc-shadermodel6.3-library -O1 -load-bool-from-mem=nonzero -emit-llvm -o - %s | FileCheck %s -check-prefixes=CHECK-NONZERO
-// RUN: %clang_cc1 -triple dxil-pc-shadermodel6.3-library -O0 -emit-llvm -o - %s | FileCheck %s -check-prefixes=CHECK-NONZERO
+// RUN: %clang_cc1 -triple dxil-pc-shadermodel6.3-library -O1 -load-bool-from-mem=truncate -emit-llvm -o - %s 2>&1 | FileCheck %s --check-prefix=CHECK,TRUNCATE
+// RUN: %clang_cc1 -triple dxil-pc-shadermodel6.3-library -O1 -load-bool-from-mem=nonzero -emit-llvm -o - %s | FileCheck %s
+// RUN: %clang_cc1 -triple dxil-pc-shadermodel6.3-library -O0 -emit-llvm -o - %s | FileCheck %s
+
+// TRUNCATE: -load-bool-from-mem=truncate ignored for HLSL (always nonzero)
typedef bool bool8_t __attribute__((ext_vector_type(8)));
extern bool8_t vec;
bool8_t getvec(void) {
- // CHECK-TRUNCATE: [[BOOL:%.+]] = load <8 x i32>
- // CHECK-TRUNCATE: trunc <8 x i32> [[BOOL]] to <8 x i1>
-
- // CHECK-NONZERO: [[BOOL:%.+]] = load <8 x i32>
- // CHECK-NONZERO: icmp ne <8 x i32> [[BOOL]], zeroinitializer
+ // CHECK: [[BOOL:%.+]] = load <8 x i32>
+ // CHECK: icmp ne <8 x i32> [[BOOL]], zeroinitializer
return vec;
}
More information about the cfe-commits
mailing list