2 changes: 1 addition & 1 deletion clang/test/CXX/drs/cwg722.cpp
View file
Original file line number Diff line number Diff line change
Expand Up @@ -14,7 +14,7 @@ namespace std {
using nullptr_t = decltype(nullptr);
}

void f(std::nullptr_t...);
void f(std::nullptr_t, ...);
std::nullptr_t g();
void h() {
std::nullptr_t np;
Expand Down
10 changes: 1 addition & 9 deletions clang/test/CXX/temp/temp.deduct.guide/p3.cpp
View file
Original file line number Diff line number Diff line change
Expand Up @@ -33,7 +33,7 @@ template<template<typename> typename TT> struct E { // expected-note 2{{template
};

A(int) -> int; // expected-error {{deduced type 'int' of deduction guide is not a specialization of template 'A'}}
template <typename T> A(T)->B<T>;
template <typename T> A(T)->B<T>; // expected-error {{deduced type 'B<T>' (aka 'A<T>') of deduction guide is not written as a specialization of template 'A'}}
template<typename T> A(T*) -> const A<T>; // expected-error {{deduced type 'const A<T>' of deduction guide is not a specialization of template 'A'}}

// A deduction-guide shall be declared in the same scope as the corresponding
Expand Down Expand Up @@ -71,11 +71,3 @@ namespace WrongScope {
Local(int) -> Local<int>; // expected-error {{expected}}
}
}

namespace GH54909 {
template <class T> struct A {};
struct B {};

template <typename T> using C = B;
template <typename T> A() -> C<T>; // expected-error {{deduced type 'C<T>' (aka 'GH54909::B') of deduction guide is not a specialization of template 'A'}}
}
20 changes: 10 additions & 10 deletions clang/test/CodeGen/AArch64/pure-scalable-args.c
View file
Original file line number Diff line number Diff line change
Expand Up @@ -70,23 +70,23 @@ void test_argpass_simple(PST *p) {
// CHECK-AAPCS: define dso_local void @test_argpass_simple(ptr nocapture noundef readonly %p)
// CHECK-AAPCS-NEXT: entry:
// CHECK-AAPCS-NEXT: %0 = load <2 x i8>, ptr %p, align 16
// CHECK-AAPCS-NEXT: %cast.scalable = tail call <vscale x 2 x i8> @llvm.vector.insert.nxv2i8.v2i8(<vscale x 2 x i8> undef, <2 x i8> %0, i64 0)
// CHECK-AAPCS-NEXT: %cast.scalable = tail call <vscale x 2 x i8> @llvm.vector.insert.nxv2i8.v2i8(<vscale x 2 x i8> poison, <2 x i8> %0, i64 0)
// CHECK-AAPCS-NEXT: %1 = bitcast <vscale x 2 x i8> %cast.scalable to <vscale x 16 x i1>
// CHECK-AAPCS-NEXT: %2 = getelementptr inbounds nuw i8, ptr %p, i64 16
// CHECK-AAPCS-NEXT: %3 = load <2 x double>, ptr %2, align 16
// CHECK-AAPCS-NEXT: %cast.scalable1 = tail call <vscale x 2 x double> @llvm.vector.insert.nxv2f64.v2f64(<vscale x 2 x double> undef, <2 x double> %3, i64 0)
// CHECK-AAPCS-NEXT: %cast.scalable1 = tail call <vscale x 2 x double> @llvm.vector.insert.nxv2f64.v2f64(<vscale x 2 x double> poison, <2 x double> %3, i64 0)
// CHECK-AAPCS-NEXT: %4 = getelementptr inbounds nuw i8, ptr %p, i64 32
// CHECK-AAPCS-NEXT: %5 = load <4 x float>, ptr %4, align 16
// CHECK-AAPCS-NEXT: %cast.scalable2 = tail call <vscale x 4 x float> @llvm.vector.insert.nxv4f32.v4f32(<vscale x 4 x float> undef, <4 x float> %5, i64 0)
// CHECK-AAPCS-NEXT: %cast.scalable2 = tail call <vscale x 4 x float> @llvm.vector.insert.nxv4f32.v4f32(<vscale x 4 x float> poison, <4 x float> %5, i64 0)
// CHECK-AAPCS-NEXT: %6 = getelementptr inbounds nuw i8, ptr %p, i64 48
// CHECK-AAPCS-NEXT: %7 = load <4 x float>, ptr %6, align 16
// CHECK-AAPCS-NEXT: %cast.scalable3 = tail call <vscale x 4 x float> @llvm.vector.insert.nxv4f32.v4f32(<vscale x 4 x float> undef, <4 x float> %7, i64 0)
// CHECK-AAPCS-NEXT: %cast.scalable3 = tail call <vscale x 4 x float> @llvm.vector.insert.nxv4f32.v4f32(<vscale x 4 x float> poison, <4 x float> %7, i64 0)
// CHECK-AAPCS-NEXT: %8 = getelementptr inbounds nuw i8, ptr %p, i64 64
// CHECK-AAPCS-NEXT: %9 = load <16 x i8>, ptr %8, align 16
// CHECK-AAPCS-NEXT: %cast.scalable4 = tail call <vscale x 16 x i8> @llvm.vector.insert.nxv16i8.v16i8(<vscale x 16 x i8> undef, <16 x i8> %9, i64 0)
// CHECK-AAPCS-NEXT: %cast.scalable4 = tail call <vscale x 16 x i8> @llvm.vector.insert.nxv16i8.v16i8(<vscale x 16 x i8> poison, <16 x i8> %9, i64 0)
// CHECK-AAPCS-NEXT: %10 = getelementptr inbounds nuw i8, ptr %p, i64 80
// CHECK-AAPCS-NEXT: %11 = load <2 x i8>, ptr %10, align 16
// CHECK-AAPCS-NEXT: %cast.scalable5 = tail call <vscale x 2 x i8> @llvm.vector.insert.nxv2i8.v2i8(<vscale x 2 x i8> undef, <2 x i8> %11, i64 0)
// CHECK-AAPCS-NEXT: %cast.scalable5 = tail call <vscale x 2 x i8> @llvm.vector.insert.nxv2i8.v2i8(<vscale x 2 x i8> poison, <2 x i8> %11, i64 0)
// CHECK-AAPCS-NEXT: %12 = bitcast <vscale x 2 x i8> %cast.scalable5 to <vscale x 16 x i1>
// CHECK-AAPCS-NEXT: tail call void @argpass_simple_callee(<vscale x 16 x i1> %1, <vscale x 2 x double> %cast.scalable1, <vscale x 4 x float> %cast.scalable2, <vscale x 4 x float> %cast.scalable3, <vscale x 16 x i8> %cast.scalable4, <vscale x 16 x i1> %12)
// CHECK-AAPCS-NEXT: ret void
Expand Down Expand Up @@ -431,9 +431,9 @@ void test_va_arg(int n, ...) {
// CHECK-AAPCS-NEXT: %v.sroa.43.0.vaarg.addr.sroa_idx = getelementptr inbounds i8, ptr %vaarg.addr, i64 48
// CHECK-AAPCS-NEXT: %v.sroa.43.0.copyload = load <4 x float>, ptr %v.sroa.43.0.vaarg.addr.sroa_idx, align 16
// CHECK-AAPCS-NEXT: call void @llvm.va_end.p0(ptr nonnull %ap)
// CHECK-AAPCS-NEXT: %cast.scalable = call <vscale x 2 x i8> @llvm.vector.insert.nxv2i8.v2i8(<vscale x 2 x i8> undef, <2 x i8> %v.sroa.0.0.copyload, i64 0)
// CHECK-AAPCS-NEXT: %cast.scalable = call <vscale x 2 x i8> @llvm.vector.insert.nxv2i8.v2i8(<vscale x 2 x i8> poison, <2 x i8> %v.sroa.0.0.copyload, i64 0)
// CHECK-AAPCS-NEXT: %3 = bitcast <vscale x 2 x i8> %cast.scalable to <vscale x 16 x i1>
// CHECK-AAPCS-NEXT: %cast.scalable2 = call <vscale x 4 x float> @llvm.vector.insert.nxv4f32.v4f32(<vscale x 4 x float> undef, <4 x float> %v.sroa.43.0.copyload, i64 0)
// CHECK-AAPCS-NEXT: %cast.scalable2 = call <vscale x 4 x float> @llvm.vector.insert.nxv4f32.v4f32(<vscale x 4 x float> poison, <4 x float> %v.sroa.43.0.copyload, i64 0)
// CHECK-AAPCS-NEXT: call void @use1(<vscale x 16 x i1> noundef %3, <vscale x 4 x float> noundef %cast.scalable2)
// CHECK-AAPCS-NEXT: call void @llvm.lifetime.end.p0(i64 32, ptr nonnull %ap)
// CHECK-AAPCS-NEXT: ret void
Expand All @@ -452,9 +452,9 @@ void test_va_arg(int n, ...) {
// CHECK-DARWIN-NEXT: %v.sroa.43.0..sroa_idx = getelementptr inbounds i8, ptr %0, i64 48
// CHECK-DARWIN-NEXT: %v.sroa.43.0.copyload = load <4 x float>, ptr %v.sroa.43.0..sroa_idx, align 16
// CHECK-DARWIN-NEXT: call void @llvm.va_end.p0(ptr nonnull %ap)
// CHECK-DARWIN-NEXT: %cast.scalable = call <vscale x 2 x i8> @llvm.vector.insert.nxv2i8.v2i8(<vscale x 2 x i8> undef, <2 x i8> %v.sroa.0.0.copyload, i64 0)
// CHECK-DARWIN-NEXT: %cast.scalable = call <vscale x 2 x i8> @llvm.vector.insert.nxv2i8.v2i8(<vscale x 2 x i8> poison, <2 x i8> %v.sroa.0.0.copyload, i64 0)
// CHECK-DARWIN-NEXT: %1 = bitcast <vscale x 2 x i8> %cast.scalable to <vscale x 16 x i1>
// CHECK-DARWIN-NEXT: %cast.scalable2 = call <vscale x 4 x float> @llvm.vector.insert.nxv4f32.v4f32(<vscale x 4 x float> undef, <4 x float> %v.sroa.43.0.copyload, i64 0)
// CHECK-DARWIN-NEXT: %cast.scalable2 = call <vscale x 4 x float> @llvm.vector.insert.nxv4f32.v4f32(<vscale x 4 x float> poison, <4 x float> %v.sroa.43.0.copyload, i64 0)
// CHECK-DARWIN-NEXT: call void @use1(<vscale x 16 x i1> noundef %1, <vscale x 4 x float> noundef %cast.scalable2)
// CHECK-DARWIN-NEXT: call void @llvm.lifetime.end.p0(i64 8, ptr nonnull %ap)
// CHECK-DARWIN-NEXT: ret void
Expand Down
2 changes: 1 addition & 1 deletion clang/test/CodeGen/AArch64/sve-acle-__ARM_FEATURE_SVE_VECTOR_OPERATORS.cpp
View file
Original file line number Diff line number Diff line change
Expand Up @@ -52,7 +52,7 @@ void test02() {
// CHECK-NEXT: [[X:%.*]] = tail call <[[#div(VBITS, 32)]] x i32> @llvm.vector.extract.v[[#div(VBITS, 32)]]i32.nxv4i32(<vscale x 4 x i32> [[X_COERCE:%.*]], i64 0)
// CHECK-NEXT: [[Y:%.*]] = tail call <[[#div(VBITS, 32)]] x i32> @llvm.vector.extract.v[[#div(VBITS, 32)]]i32.nxv4i32(<vscale x 4 x i32> [[X_COERCE1:%.*]], i64 0)
// CHECK-NEXT: [[ADD:%.*]] = add <[[#div(VBITS, 32)]] x i32> [[Y]], [[X]]
// CHECK-NEXT: [[CASTSCALABLESVE:%.*]] = tail call <vscale x 4 x i32> @llvm.vector.insert.nxv4i32.v[[#div(VBITS, 32)]]i32(<vscale x 4 x i32> undef, <[[#div(VBITS, 32)]] x i32> [[ADD]], i64 0)
// CHECK-NEXT: [[CASTSCALABLESVE:%.*]] = tail call <vscale x 4 x i32> @llvm.vector.insert.nxv4i32.v[[#div(VBITS, 32)]]i32(<vscale x 4 x i32> poison, <[[#div(VBITS, 32)]] x i32> [[ADD]], i64 0)
// CHECK-NEXT: ret <vscale x 4 x i32> [[CASTSCALABLESVE]]
typedef svint32_t vec __attribute__((arm_sve_vector_bits(N)));
auto f(vec x, vec y) { return x + y; } // Returns a vec.
Expand Down
312 changes: 156 additions & 156 deletions clang/test/CodeGen/AArch64/sve-vls-arith-ops.c
View file

Large diffs are not rendered by default.

72 changes: 36 additions & 36 deletions clang/test/CodeGen/AArch64/sve-vls-bitwise-ops.c
View file

Large diffs are not rendered by default.

144 changes: 72 additions & 72 deletions clang/test/CodeGen/AArch64/sve-vls-compare-ops.c
View file

Large diffs are not rendered by default.

96 changes: 48 additions & 48 deletions clang/test/CodeGen/AArch64/sve-vls-shift-ops.c
View file

Large diffs are not rendered by default.

2 changes: 1 addition & 1 deletion clang/test/CodeGen/RISCV/attr-rvv-vector-bits-cast.c
View file
Original file line number Diff line number Diff line change
Expand Up @@ -140,7 +140,7 @@ gnu_int32m1_t from_vint32m1_t__to_gnu_int32m1_t(vint32m1_t type) {
// CHECK-LABEL: @to_fixed_int32m1_t__from_gnu_int32m1_t(
// CHECK-NEXT: entry:
// CHECK-NEXT: [[TYPE:%.*]] = load <8 x i32>, ptr [[TMP0:%.*]], align 32, !tbaa [[TBAA8]]
// CHECK-NEXT: [[CAST_SCALABLE:%.*]] = tail call <vscale x 2 x i32> @llvm.vector.insert.nxv2i32.v8i32(<vscale x 2 x i32> undef, <8 x i32> [[TYPE]], i64 0)
// CHECK-NEXT: [[CAST_SCALABLE:%.*]] = tail call <vscale x 2 x i32> @llvm.vector.insert.nxv2i32.v8i32(<vscale x 2 x i32> poison, <8 x i32> [[TYPE]], i64 0)
// CHECK-NEXT: ret <vscale x 2 x i32> [[CAST_SCALABLE]]
//
fixed_int32m1_t to_fixed_int32m1_t__from_gnu_int32m1_t(gnu_int32m1_t type) {
Expand Down
22 changes: 11 additions & 11 deletions clang/test/CodeGen/RISCV/attr-rvv-vector-bits-codegen.c
View file
Original file line number Diff line number Diff line change
Expand Up @@ -69,7 +69,7 @@ fixed_bool32_t global_bool32;
// CHECK-NEXT: [[CAST_FIXED:%.*]] = call <256 x i8> @llvm.vector.extract.v256i8.nxv64i8(<vscale x 64 x i8> [[TMP7]], i64 0)
// CHECK-NEXT: store <256 x i8> [[CAST_FIXED]], ptr [[RETVAL]], align 8
// CHECK-NEXT: [[TMP8:%.*]] = load <256 x i8>, ptr [[RETVAL]], align 8
// CHECK-NEXT: [[CAST_SCALABLE2:%.*]] = call <vscale x 64 x i8> @llvm.vector.insert.nxv64i8.v256i8(<vscale x 64 x i8> undef, <256 x i8> [[TMP8]], i64 0)
// CHECK-NEXT: [[CAST_SCALABLE2:%.*]] = call <vscale x 64 x i8> @llvm.vector.insert.nxv64i8.v256i8(<vscale x 64 x i8> poison, <256 x i8> [[TMP8]], i64 0)
// CHECK-NEXT: ret <vscale x 64 x i8> [[CAST_SCALABLE2]]
//
fixed_int8m8_t test_bool1(vbool1_t m, vint8m8_t vec) {
Expand Down Expand Up @@ -99,7 +99,7 @@ fixed_int8m8_t test_bool1(vbool1_t m, vint8m8_t vec) {
// CHECK-NEXT: [[CAST_FIXED:%.*]] = call <64 x i16> @llvm.vector.extract.v64i16.nxv16i16(<vscale x 16 x i16> [[TMP7]], i64 0)
// CHECK-NEXT: store <64 x i16> [[CAST_FIXED]], ptr [[RETVAL]], align 8
// CHECK-NEXT: [[TMP8:%.*]] = load <64 x i16>, ptr [[RETVAL]], align 8
// CHECK-NEXT: [[CAST_SCALABLE2:%.*]] = call <vscale x 16 x i16> @llvm.vector.insert.nxv16i16.v64i16(<vscale x 16 x i16> undef, <64 x i16> [[TMP8]], i64 0)
// CHECK-NEXT: [[CAST_SCALABLE2:%.*]] = call <vscale x 16 x i16> @llvm.vector.insert.nxv16i16.v64i16(<vscale x 16 x i16> poison, <64 x i16> [[TMP8]], i64 0)
// CHECK-NEXT: ret <vscale x 16 x i16> [[CAST_SCALABLE2]]
//
fixed_int16m4_t test_bool4(vbool4_t m, vint16m4_t vec) {
Expand Down Expand Up @@ -130,7 +130,7 @@ fixed_int16m4_t test_bool4(vbool4_t m, vint16m4_t vec) {
// CHECK-NEXT: [[CAST_FIXED:%.*]] = call <8 x i32> @llvm.vector.extract.v8i32.nxv2i32(<vscale x 2 x i32> [[TMP7]], i64 0)
// CHECK-NEXT: store <8 x i32> [[CAST_FIXED]], ptr [[RETVAL]], align 8
// CHECK-NEXT: [[TMP8:%.*]] = load <8 x i32>, ptr [[RETVAL]], align 8
// CHECK-NEXT: [[CAST_SCALABLE1:%.*]] = call <vscale x 2 x i32> @llvm.vector.insert.nxv2i32.v8i32(<vscale x 2 x i32> undef, <8 x i32> [[TMP8]], i64 0)
// CHECK-NEXT: [[CAST_SCALABLE1:%.*]] = call <vscale x 2 x i32> @llvm.vector.insert.nxv2i32.v8i32(<vscale x 2 x i32> poison, <8 x i32> [[TMP8]], i64 0)
// CHECK-NEXT: ret <vscale x 2 x i32> [[CAST_SCALABLE1]]
//
fixed_int32m1_t test_bool32(vbool32_t m, vint32m1_t vec) {
Expand All @@ -147,7 +147,7 @@ fixed_int32m1_t test_bool32(vbool32_t m, vint32m1_t vec) {
// CHECK-NEXT: [[TMP1:%.*]] = load <8 x i32>, ptr [[TMP0]], align 8
// CHECK-NEXT: store <8 x i32> [[TMP1]], ptr [[RETVAL]], align 8
// CHECK-NEXT: [[TMP2:%.*]] = load <8 x i32>, ptr [[RETVAL]], align 8
// CHECK-NEXT: [[CAST_SCALABLE:%.*]] = call <vscale x 2 x i32> @llvm.vector.insert.nxv2i32.v8i32(<vscale x 2 x i32> undef, <8 x i32> [[TMP2]], i64 0)
// CHECK-NEXT: [[CAST_SCALABLE:%.*]] = call <vscale x 2 x i32> @llvm.vector.insert.nxv2i32.v8i32(<vscale x 2 x i32> poison, <8 x i32> [[TMP2]], i64 0)
// CHECK-NEXT: ret <vscale x 2 x i32> [[CAST_SCALABLE]]
//
fixed_int32m1_t test_ptr_to_global() {
Expand All @@ -168,7 +168,7 @@ fixed_int32m1_t test_ptr_to_global() {
// CHECK-NEXT: [[TMP1:%.*]] = load <8 x i32>, ptr [[ARRAYIDX]], align 8
// CHECK-NEXT: store <8 x i32> [[TMP1]], ptr [[RETVAL]], align 8
// CHECK-NEXT: [[TMP2:%.*]] = load <8 x i32>, ptr [[RETVAL]], align 8
// CHECK-NEXT: [[CAST_SCALABLE:%.*]] = call <vscale x 2 x i32> @llvm.vector.insert.nxv2i32.v8i32(<vscale x 2 x i32> undef, <8 x i32> [[TMP2]], i64 0)
// CHECK-NEXT: [[CAST_SCALABLE:%.*]] = call <vscale x 2 x i32> @llvm.vector.insert.nxv2i32.v8i32(<vscale x 2 x i32> poison, <8 x i32> [[TMP2]], i64 0)
// CHECK-NEXT: ret <vscale x 2 x i32> [[CAST_SCALABLE]]
//
fixed_int32m1_t array_arg(fixed_int32m1_t arr[]) {
Expand All @@ -186,7 +186,7 @@ fixed_int32m1_t array_arg(fixed_int32m1_t arr[]) {
// CHECK-NEXT: [[TMP1:%.*]] = load <32 x i8>, ptr [[TMP0]], align 8
// CHECK-NEXT: store <32 x i8> [[TMP1]], ptr [[RETVAL]], align 8
// CHECK-NEXT: [[TMP2:%.*]] = load <32 x i8>, ptr [[RETVAL]], align 8
// CHECK-NEXT: [[CAST_SCALABLE:%.*]] = call <vscale x 8 x i8> @llvm.vector.insert.nxv8i8.v32i8(<vscale x 8 x i8> undef, <32 x i8> [[TMP2]], i64 0)
// CHECK-NEXT: [[CAST_SCALABLE:%.*]] = call <vscale x 8 x i8> @llvm.vector.insert.nxv8i8.v32i8(<vscale x 8 x i8> poison, <32 x i8> [[TMP2]], i64 0)
// CHECK-NEXT: [[TMP3:%.*]] = bitcast <vscale x 8 x i8> [[CAST_SCALABLE]] to <vscale x 64 x i1>
// CHECK-NEXT: ret <vscale x 64 x i1> [[TMP3]]
//
Expand All @@ -208,7 +208,7 @@ fixed_bool1_t address_of_array_idx_bool1() {
// CHECK-NEXT: [[TMP1:%.*]] = load <8 x i8>, ptr [[TMP0]], align 8
// CHECK-NEXT: store <8 x i8> [[TMP1]], ptr [[RETVAL]], align 8
// CHECK-NEXT: [[TMP2:%.*]] = load <8 x i8>, ptr [[RETVAL]], align 8
// CHECK-NEXT: [[CAST_SCALABLE:%.*]] = call <vscale x 2 x i8> @llvm.vector.insert.nxv2i8.v8i8(<vscale x 2 x i8> undef, <8 x i8> [[TMP2]], i64 0)
// CHECK-NEXT: [[CAST_SCALABLE:%.*]] = call <vscale x 2 x i8> @llvm.vector.insert.nxv2i8.v8i8(<vscale x 2 x i8> poison, <8 x i8> [[TMP2]], i64 0)
// CHECK-NEXT: [[TMP3:%.*]] = bitcast <vscale x 2 x i8> [[CAST_SCALABLE]] to <vscale x 16 x i1>
// CHECK-NEXT: ret <vscale x 16 x i1> [[TMP3]]
//
Expand Down Expand Up @@ -253,7 +253,7 @@ fixed_bool32_t address_of_array_idx_bool32() {
// CHECK-NEXT: [[CAST_FIXED:%.*]] = call <8 x i32> @llvm.vector.extract.v8i32.nxv2i32(<vscale x 2 x i32> [[TMP2]], i64 0)
// CHECK-NEXT: store <8 x i32> [[CAST_FIXED]], ptr [[RETVAL]], align 8
// CHECK-NEXT: [[TMP3:%.*]] = load <8 x i32>, ptr [[RETVAL]], align 8
// CHECK-NEXT: [[CAST_SCALABLE1:%.*]] = call <vscale x 2 x i32> @llvm.vector.insert.nxv2i32.v8i32(<vscale x 2 x i32> undef, <8 x i32> [[TMP3]], i64 0)
// CHECK-NEXT: [[CAST_SCALABLE1:%.*]] = call <vscale x 2 x i32> @llvm.vector.insert.nxv2i32.v8i32(<vscale x 2 x i32> poison, <8 x i32> [[TMP3]], i64 0)
// CHECK-NEXT: ret <vscale x 2 x i32> [[CAST_SCALABLE1]]
//
fixed_int32m1_t test_cast(vint32m1_t vec) {
Expand All @@ -269,7 +269,7 @@ fixed_int32m1_t test_cast(vint32m1_t vec) {
// CHECK-NEXT: [[TMP1:%.*]] = load <16 x i32>, ptr [[TMP0]], align 8
// CHECK-NEXT: store <16 x i32> [[TMP1]], ptr [[RETVAL]], align 8
// CHECK-NEXT: [[TMP2:%.*]] = load <16 x i32>, ptr [[RETVAL]], align 8
// CHECK-NEXT: [[CAST_SCALABLE:%.*]] = call <vscale x 4 x i32> @llvm.vector.insert.nxv4i32.v16i32(<vscale x 4 x i32> undef, <16 x i32> [[TMP2]], i64 0)
// CHECK-NEXT: [[CAST_SCALABLE:%.*]] = call <vscale x 4 x i32> @llvm.vector.insert.nxv4i32.v16i32(<vscale x 4 x i32> poison, <16 x i32> [[TMP2]], i64 0)
// CHECK-NEXT: ret <vscale x 4 x i32> [[CAST_SCALABLE]]
//
fixed_int32m2_t test_ptr_to_global_m2() {
Expand All @@ -290,7 +290,7 @@ fixed_int32m2_t test_ptr_to_global_m2() {
// CHECK-NEXT: [[TMP1:%.*]] = load <16 x i32>, ptr [[ARRAYIDX]], align 8
// CHECK-NEXT: store <16 x i32> [[TMP1]], ptr [[RETVAL]], align 8
// CHECK-NEXT: [[TMP2:%.*]] = load <16 x i32>, ptr [[RETVAL]], align 8
// CHECK-NEXT: [[CAST_SCALABLE:%.*]] = call <vscale x 4 x i32> @llvm.vector.insert.nxv4i32.v16i32(<vscale x 4 x i32> undef, <16 x i32> [[TMP2]], i64 0)
// CHECK-NEXT: [[CAST_SCALABLE:%.*]] = call <vscale x 4 x i32> @llvm.vector.insert.nxv4i32.v16i32(<vscale x 4 x i32> poison, <16 x i32> [[TMP2]], i64 0)
// CHECK-NEXT: ret <vscale x 4 x i32> [[CAST_SCALABLE]]
//
fixed_int32m2_t array_arg_m2(fixed_int32m2_t arr[]) {
Expand All @@ -309,7 +309,7 @@ fixed_int32m2_t array_arg_m2(fixed_int32m2_t arr[]) {
// CHECK-NEXT: [[CAST_FIXED:%.*]] = call <16 x i32> @llvm.vector.extract.v16i32.nxv4i32(<vscale x 4 x i32> [[TMP2]], i64 0)
// CHECK-NEXT: store <16 x i32> [[CAST_FIXED]], ptr [[RETVAL]], align 8
// CHECK-NEXT: [[TMP3:%.*]] = load <16 x i32>, ptr [[RETVAL]], align 8
// CHECK-NEXT: [[CAST_SCALABLE1:%.*]] = call <vscale x 4 x i32> @llvm.vector.insert.nxv4i32.v16i32(<vscale x 4 x i32> undef, <16 x i32> [[TMP3]], i64 0)
// CHECK-NEXT: [[CAST_SCALABLE1:%.*]] = call <vscale x 4 x i32> @llvm.vector.insert.nxv4i32.v16i32(<vscale x 4 x i32> poison, <16 x i32> [[TMP3]], i64 0)
// CHECK-NEXT: ret <vscale x 4 x i32> [[CAST_SCALABLE1]]
//
fixed_int32m2_t test_cast_m2(vint32m2_t vec) {
Expand Down
288 changes: 144 additions & 144 deletions clang/test/CodeGen/RISCV/rvv-vls-arith-ops.c
View file

Large diffs are not rendered by default.

64 changes: 32 additions & 32 deletions clang/test/CodeGen/RISCV/rvv-vls-bitwise-ops.c
View file

Large diffs are not rendered by default.

120 changes: 60 additions & 60 deletions clang/test/CodeGen/RISCV/rvv-vls-compare-ops.c
View file

Large diffs are not rendered by default.

96 changes: 48 additions & 48 deletions clang/test/CodeGen/RISCV/rvv-vls-shift-ops.c
View file

Large diffs are not rendered by default.

3 changes: 3 additions & 0 deletions clang/test/CodeGen/alias.cpp
View file
Original file line number Diff line number Diff line change
Expand Up @@ -26,6 +26,9 @@ __attribute__((unused, alias("resolver"), deprecated("hahahaha, isn't C great?")
void func();
// expected-error@-2 {{alias must point to a defined variable or function}}
// expected-note@-3 {{must refer to its mangled name}}

void *operator new(unsigned long) __attribute__((alias("A"))); // expected-error {{alias must point to a defined variable or function}} \
// expected-note {{the function or variable specified in an alias must refer to its mangled name}}
#endif

// CHECK: @_ZN4libc4log2Ed ={{.*}} alias double (double), ptr @log2
Expand Down
2 changes: 1 addition & 1 deletion clang/test/CodeGen/attr-arm-sve-vector-bits-cast.c
View file
Original file line number Diff line number Diff line change
Expand Up @@ -95,7 +95,7 @@ gnu_int32_t from_svint32_t__to_gnu_int32_t(svint32_t type) {
// CHECK-LABEL: @to_fixed_int32_t__from_gnu_int32_t(
// CHECK-NEXT: entry:
// CHECK-NEXT: [[TYPE:%.*]] = load <16 x i32>, ptr [[TMP0:%.*]], align 16, !tbaa [[TBAA6]]
// CHECK-NEXT: [[CASTSCALABLESVE:%.*]] = tail call <vscale x 4 x i32> @llvm.vector.insert.nxv4i32.v16i32(<vscale x 4 x i32> undef, <16 x i32> [[TYPE]], i64 0)
// CHECK-NEXT: [[CASTSCALABLESVE:%.*]] = tail call <vscale x 4 x i32> @llvm.vector.insert.nxv4i32.v16i32(<vscale x 4 x i32> poison, <16 x i32> [[TYPE]], i64 0)
// CHECK-NEXT: ret <vscale x 4 x i32> [[CASTSCALABLESVE]]
//
fixed_int32_t to_fixed_int32_t__from_gnu_int32_t(gnu_int32_t type) {
Expand Down
10 changes: 5 additions & 5 deletions clang/test/CodeGen/attr-arm-sve-vector-bits-codegen.c
View file
Original file line number Diff line number Diff line change
Expand Up @@ -40,7 +40,7 @@ fixed_int32_t global_vec;
// CHECK-NEXT: [[CASTFIXEDSVE3:%.*]] = call <16 x i32> @llvm.vector.extract.v16i32.nxv4i32(<vscale x 4 x i32> [[TMP10]], i64 0)
// CHECK-NEXT: store <16 x i32> [[CASTFIXEDSVE3]], ptr [[RETVAL]], align 16
// CHECK-NEXT: [[TMP11:%.*]] = load <16 x i32>, ptr [[RETVAL]], align 16
// CHECK-NEXT: [[CASTSCALABLESVE4:%.*]] = call <vscale x 4 x i32> @llvm.vector.insert.nxv4i32.v16i32(<vscale x 4 x i32> undef, <16 x i32> [[TMP11]], i64 0)
// CHECK-NEXT: [[CASTSCALABLESVE4:%.*]] = call <vscale x 4 x i32> @llvm.vector.insert.nxv4i32.v16i32(<vscale x 4 x i32> poison, <16 x i32> [[TMP11]], i64 0)
// CHECK-NEXT: ret <vscale x 4 x i32> [[CASTSCALABLESVE4]]
//
fixed_int32_t foo(svbool_t pred, svint32_t vec) {
Expand All @@ -57,7 +57,7 @@ fixed_int32_t foo(svbool_t pred, svint32_t vec) {
// CHECK-NEXT: [[TMP1:%.*]] = load <16 x i32>, ptr [[TMP0]], align 16
// CHECK-NEXT: store <16 x i32> [[TMP1]], ptr [[RETVAL]], align 16
// CHECK-NEXT: [[TMP2:%.*]] = load <16 x i32>, ptr [[RETVAL]], align 16
// CHECK-NEXT: [[CASTSCALABLESVE:%.*]] = call <vscale x 4 x i32> @llvm.vector.insert.nxv4i32.v16i32(<vscale x 4 x i32> undef, <16 x i32> [[TMP2]], i64 0)
// CHECK-NEXT: [[CASTSCALABLESVE:%.*]] = call <vscale x 4 x i32> @llvm.vector.insert.nxv4i32.v16i32(<vscale x 4 x i32> poison, <16 x i32> [[TMP2]], i64 0)
// CHECK-NEXT: ret <vscale x 4 x i32> [[CASTSCALABLESVE]]
//
fixed_int32_t test_ptr_to_global() {
Expand All @@ -78,7 +78,7 @@ fixed_int32_t test_ptr_to_global() {
// CHECK-NEXT: [[TMP1:%.*]] = load <16 x i32>, ptr [[ARRAYIDX]], align 16
// CHECK-NEXT: store <16 x i32> [[TMP1]], ptr [[RETVAL]], align 16
// CHECK-NEXT: [[TMP2:%.*]] = load <16 x i32>, ptr [[RETVAL]], align 16
// CHECK-NEXT: [[CASTSCALABLESVE:%.*]] = call <vscale x 4 x i32> @llvm.vector.insert.nxv4i32.v16i32(<vscale x 4 x i32> undef, <16 x i32> [[TMP2]], i64 0)
// CHECK-NEXT: [[CASTSCALABLESVE:%.*]] = call <vscale x 4 x i32> @llvm.vector.insert.nxv4i32.v16i32(<vscale x 4 x i32> poison, <16 x i32> [[TMP2]], i64 0)
// CHECK-NEXT: ret <vscale x 4 x i32> [[CASTSCALABLESVE]]
//
fixed_int32_t array_arg(fixed_int32_t arr[]) {
Expand All @@ -96,7 +96,7 @@ fixed_int32_t array_arg(fixed_int32_t arr[]) {
// CHECK-NEXT: [[TMP1:%.*]] = load <8 x i8>, ptr [[TMP0]], align 2
// CHECK-NEXT: store <8 x i8> [[TMP1]], ptr [[RETVAL]], align 2
// CHECK-NEXT: [[TMP2:%.*]] = load <8 x i8>, ptr [[RETVAL]], align 2
// CHECK-NEXT: [[CASTFIXEDSVE:%.*]] = call <vscale x 2 x i8> @llvm.vector.insert.nxv2i8.v8i8(<vscale x 2 x i8> undef, <8 x i8> [[TMP2]], i64 0)
// CHECK-NEXT: [[CASTFIXEDSVE:%.*]] = call <vscale x 2 x i8> @llvm.vector.insert.nxv2i8.v8i8(<vscale x 2 x i8> poison, <8 x i8> [[TMP2]], i64 0)
// CHECK-NEXT: [[TMP3:%.*]] = bitcast <vscale x 2 x i8> [[CASTFIXEDSVE]] to <vscale x 16 x i1>
// CHECK-NEXT: ret <vscale x 16 x i1> [[TMP3]]
//
Expand Down Expand Up @@ -139,7 +139,7 @@ fixed_bool_t address_of_array_idx() {
// CHECK-NEXT: [[CASTFIXEDSVE3:%.*]] = call <16 x i32> @llvm.vector.extract.v16i32.nxv4i32(<vscale x 4 x i32> [[TMP11]], i64 0)
// CHECK-NEXT: store <16 x i32> [[CASTFIXEDSVE3]], ptr [[RETVAL]], align 16
// CHECK-NEXT: [[TMP12:%.*]] = load <16 x i32>, ptr [[RETVAL]], align 16
// CHECK-NEXT: [[CASTSCALABLESVE4:%.*]] = call <vscale x 4 x i32> @llvm.vector.insert.nxv4i32.v16i32(<vscale x 4 x i32> undef, <16 x i32> [[TMP12]], i64 0)
// CHECK-NEXT: [[CASTSCALABLESVE4:%.*]] = call <vscale x 4 x i32> @llvm.vector.insert.nxv4i32.v16i32(<vscale x 4 x i32> poison, <16 x i32> [[TMP12]], i64 0)
// CHECK-NEXT: ret <vscale x 4 x i32> [[CASTSCALABLESVE4]]
//
fixed_int32_t test_cast(svbool_t pred, svint32_t vec) {
Expand Down
44 changes: 18 additions & 26 deletions clang/test/CodeGen/attr-counted-by.c
View file
Original file line number Diff line number Diff line change
Expand Up @@ -119,9 +119,8 @@ void test1(struct annotated *p, int index, int val) {
// SANITIZE-WITH-ATTR: cont3:
// SANITIZE-WITH-ATTR-NEXT: [[ARRAY:%.*]] = getelementptr inbounds nuw i8, ptr [[P]], i64 12
// SANITIZE-WITH-ATTR-NEXT: [[ARRAYIDX:%.*]] = getelementptr inbounds nuw [0 x i32], ptr [[ARRAY]], i64 0, i64 [[INDEX]]
// SANITIZE-WITH-ATTR-NEXT: [[TMP2:%.*]] = shl i32 [[DOT_COUNTED_BY_LOAD]], 2
// SANITIZE-WITH-ATTR-NEXT: [[DOTINV:%.*]] = icmp slt i32 [[DOT_COUNTED_BY_LOAD]], 0
// SANITIZE-WITH-ATTR-NEXT: [[CONV:%.*]] = select i1 [[DOTINV]], i32 0, i32 [[TMP2]]
// SANITIZE-WITH-ATTR-NEXT: [[TMP2:%.*]] = tail call i32 @llvm.smax.i32(i32 [[DOT_COUNTED_BY_LOAD]], i32 0)
// SANITIZE-WITH-ATTR-NEXT: [[CONV:%.*]] = shl i32 [[TMP2]], 2
// SANITIZE-WITH-ATTR-NEXT: store i32 [[CONV]], ptr [[ARRAYIDX]], align 4, !tbaa [[TBAA4]]
// SANITIZE-WITH-ATTR-NEXT: ret void
//
Expand All @@ -130,9 +129,8 @@ void test1(struct annotated *p, int index, int val) {
// NO-SANITIZE-WITH-ATTR-NEXT: entry:
// NO-SANITIZE-WITH-ATTR-NEXT: [[DOT_COUNTED_BY_GEP:%.*]] = getelementptr inbounds i8, ptr [[P]], i64 8
// NO-SANITIZE-WITH-ATTR-NEXT: [[DOT_COUNTED_BY_LOAD:%.*]] = load i32, ptr [[DOT_COUNTED_BY_GEP]], align 4
// NO-SANITIZE-WITH-ATTR-NEXT: [[TMP0:%.*]] = shl i32 [[DOT_COUNTED_BY_LOAD]], 2
// NO-SANITIZE-WITH-ATTR-NEXT: [[DOTINV:%.*]] = icmp slt i32 [[DOT_COUNTED_BY_LOAD]], 0
// NO-SANITIZE-WITH-ATTR-NEXT: [[CONV:%.*]] = select i1 [[DOTINV]], i32 0, i32 [[TMP0]]
// NO-SANITIZE-WITH-ATTR-NEXT: [[TMP0:%.*]] = tail call i32 @llvm.smax.i32(i32 [[DOT_COUNTED_BY_LOAD]], i32 0)
// NO-SANITIZE-WITH-ATTR-NEXT: [[CONV:%.*]] = shl i32 [[TMP0]], 2
// NO-SANITIZE-WITH-ATTR-NEXT: [[ARRAY:%.*]] = getelementptr inbounds nuw i8, ptr [[P]], i64 12
// NO-SANITIZE-WITH-ATTR-NEXT: [[ARRAYIDX:%.*]] = getelementptr inbounds nuw [0 x i32], ptr [[ARRAY]], i64 0, i64 [[INDEX]]
// NO-SANITIZE-WITH-ATTR-NEXT: store i32 [[CONV]], ptr [[ARRAYIDX]], align 4, !tbaa [[TBAA2]]
Expand Down Expand Up @@ -539,10 +537,9 @@ size_t test5_bdos(struct anon_struct *p) {
// SANITIZE-WITH-ATTR: cont3:
// SANITIZE-WITH-ATTR-NEXT: [[TMP1:%.*]] = getelementptr inbounds nuw i8, ptr [[P]], i64 16
// SANITIZE-WITH-ATTR-NEXT: [[ARRAYIDX:%.*]] = getelementptr inbounds [0 x i32], ptr [[TMP1]], i64 0, i64 [[IDXPROM]]
// SANITIZE-WITH-ATTR-NEXT: [[DOTINV:%.*]] = icmp slt i64 [[DOT_COUNTED_BY_LOAD]], 0
// SANITIZE-WITH-ATTR-NEXT: [[DOT_COUNTED_BY_LOAD_TR:%.*]] = trunc i64 [[DOT_COUNTED_BY_LOAD]] to i32
// SANITIZE-WITH-ATTR-NEXT: [[TMP2:%.*]] = shl i32 [[DOT_COUNTED_BY_LOAD_TR]], 2
// SANITIZE-WITH-ATTR-NEXT: [[CONV:%.*]] = select i1 [[DOTINV]], i32 0, i32 [[TMP2]]
// SANITIZE-WITH-ATTR-NEXT: [[TMP2:%.*]] = tail call i64 @llvm.smax.i64(i64 [[DOT_COUNTED_BY_LOAD]], i64 0)
// SANITIZE-WITH-ATTR-NEXT: [[DOTTR:%.*]] = trunc i64 [[TMP2]] to i32
// SANITIZE-WITH-ATTR-NEXT: [[CONV:%.*]] = shl i32 [[DOTTR]], 2
// SANITIZE-WITH-ATTR-NEXT: store i32 [[CONV]], ptr [[ARRAYIDX]], align 4, !tbaa [[TBAA4]]
// SANITIZE-WITH-ATTR-NEXT: ret void
//
Expand All @@ -551,10 +548,9 @@ size_t test5_bdos(struct anon_struct *p) {
// NO-SANITIZE-WITH-ATTR-NEXT: entry:
// NO-SANITIZE-WITH-ATTR-NEXT: [[DOT_COUNTED_BY_GEP:%.*]] = getelementptr inbounds i8, ptr [[P]], i64 8
// NO-SANITIZE-WITH-ATTR-NEXT: [[DOT_COUNTED_BY_LOAD:%.*]] = load i64, ptr [[DOT_COUNTED_BY_GEP]], align 4
// NO-SANITIZE-WITH-ATTR-NEXT: [[DOTINV:%.*]] = icmp slt i64 [[DOT_COUNTED_BY_LOAD]], 0
// NO-SANITIZE-WITH-ATTR-NEXT: [[DOT_COUNTED_BY_LOAD_TR:%.*]] = trunc i64 [[DOT_COUNTED_BY_LOAD]] to i32
// NO-SANITIZE-WITH-ATTR-NEXT: [[TMP0:%.*]] = shl i32 [[DOT_COUNTED_BY_LOAD_TR]], 2
// NO-SANITIZE-WITH-ATTR-NEXT: [[CONV:%.*]] = select i1 [[DOTINV]], i32 0, i32 [[TMP0]]
// NO-SANITIZE-WITH-ATTR-NEXT: [[TMP0:%.*]] = tail call i64 @llvm.smax.i64(i64 [[DOT_COUNTED_BY_LOAD]], i64 0)
// NO-SANITIZE-WITH-ATTR-NEXT: [[DOTTR:%.*]] = trunc i64 [[TMP0]] to i32
// NO-SANITIZE-WITH-ATTR-NEXT: [[CONV:%.*]] = shl i32 [[DOTTR]], 2
// NO-SANITIZE-WITH-ATTR-NEXT: [[TMP1:%.*]] = getelementptr inbounds nuw i8, ptr [[P]], i64 16
// NO-SANITIZE-WITH-ATTR-NEXT: [[IDXPROM:%.*]] = sext i32 [[INDEX]] to i64
// NO-SANITIZE-WITH-ATTR-NEXT: [[ARRAYIDX:%.*]] = getelementptr inbounds [0 x i32], ptr [[TMP1]], i64 0, i64 [[IDXPROM]]
Expand Down Expand Up @@ -588,19 +584,17 @@ void test6(struct anon_struct *p, int index) {
// SANITIZE-WITH-ATTR-NEXT: entry:
// SANITIZE-WITH-ATTR-NEXT: [[DOT_COUNTED_BY_GEP:%.*]] = getelementptr inbounds i8, ptr [[P]], i64 8
// SANITIZE-WITH-ATTR-NEXT: [[DOT_COUNTED_BY_LOAD:%.*]] = load i64, ptr [[DOT_COUNTED_BY_GEP]], align 4
// SANITIZE-WITH-ATTR-NEXT: [[TMP0:%.*]] = shl nuw i64 [[DOT_COUNTED_BY_LOAD]], 2
// SANITIZE-WITH-ATTR-NEXT: [[DOTINV:%.*]] = icmp slt i64 [[DOT_COUNTED_BY_LOAD]], 0
// SANITIZE-WITH-ATTR-NEXT: [[TMP1:%.*]] = select i1 [[DOTINV]], i64 0, i64 [[TMP0]]
// SANITIZE-WITH-ATTR-NEXT: [[TMP0:%.*]] = tail call i64 @llvm.smax.i64(i64 [[DOT_COUNTED_BY_LOAD]], i64 0)
// SANITIZE-WITH-ATTR-NEXT: [[TMP1:%.*]] = shl i64 [[TMP0]], 2
// SANITIZE-WITH-ATTR-NEXT: ret i64 [[TMP1]]
//
// NO-SANITIZE-WITH-ATTR-LABEL: define dso_local range(i64 0, -3) i64 @test6_bdos(
// NO-SANITIZE-WITH-ATTR-SAME: ptr nocapture noundef readonly [[P:%.*]]) local_unnamed_addr #[[ATTR2]] {
// NO-SANITIZE-WITH-ATTR-NEXT: entry:
// NO-SANITIZE-WITH-ATTR-NEXT: [[DOT_COUNTED_BY_GEP:%.*]] = getelementptr inbounds i8, ptr [[P]], i64 8
// NO-SANITIZE-WITH-ATTR-NEXT: [[DOT_COUNTED_BY_LOAD:%.*]] = load i64, ptr [[DOT_COUNTED_BY_GEP]], align 4
// NO-SANITIZE-WITH-ATTR-NEXT: [[TMP0:%.*]] = shl nuw i64 [[DOT_COUNTED_BY_LOAD]], 2
// NO-SANITIZE-WITH-ATTR-NEXT: [[DOTINV:%.*]] = icmp slt i64 [[DOT_COUNTED_BY_LOAD]], 0
// NO-SANITIZE-WITH-ATTR-NEXT: [[TMP1:%.*]] = select i1 [[DOTINV]], i64 0, i64 [[TMP0]]
// NO-SANITIZE-WITH-ATTR-NEXT: [[TMP0:%.*]] = tail call i64 @llvm.smax.i64(i64 [[DOT_COUNTED_BY_LOAD]], i64 0)
// NO-SANITIZE-WITH-ATTR-NEXT: [[TMP1:%.*]] = shl i64 [[TMP0]], 2
// NO-SANITIZE-WITH-ATTR-NEXT: ret i64 [[TMP1]]
//
// SANITIZE-WITHOUT-ATTR-LABEL: define dso_local i64 @test6_bdos(
Expand Down Expand Up @@ -1740,9 +1734,8 @@ struct annotated_struct_array {
// SANITIZE-WITH-ATTR: cont20:
// SANITIZE-WITH-ATTR-NEXT: [[ARRAY:%.*]] = getelementptr inbounds nuw i8, ptr [[TMP2]], i64 12
// SANITIZE-WITH-ATTR-NEXT: [[ARRAYIDX18:%.*]] = getelementptr inbounds [0 x i32], ptr [[ARRAY]], i64 0, i64 [[IDXPROM15]]
// SANITIZE-WITH-ATTR-NEXT: [[TMP5:%.*]] = shl i32 [[DOT_COUNTED_BY_LOAD]], 2
// SANITIZE-WITH-ATTR-NEXT: [[DOTINV:%.*]] = icmp slt i32 [[DOT_COUNTED_BY_LOAD]], 0
// SANITIZE-WITH-ATTR-NEXT: [[CONV:%.*]] = select i1 [[DOTINV]], i32 0, i32 [[TMP5]]
// SANITIZE-WITH-ATTR-NEXT: [[TMP5:%.*]] = tail call i32 @llvm.smax.i32(i32 [[DOT_COUNTED_BY_LOAD]], i32 0)
// SANITIZE-WITH-ATTR-NEXT: [[CONV:%.*]] = shl i32 [[TMP5]], 2
// SANITIZE-WITH-ATTR-NEXT: store i32 [[CONV]], ptr [[ARRAYIDX18]], align 4, !tbaa [[TBAA4]]
// SANITIZE-WITH-ATTR-NEXT: ret void
//
Expand All @@ -1754,9 +1747,8 @@ struct annotated_struct_array {
// NO-SANITIZE-WITH-ATTR-NEXT: [[TMP0:%.*]] = load ptr, ptr [[ARRAYIDX]], align 8, !tbaa [[TBAA11]]
// NO-SANITIZE-WITH-ATTR-NEXT: [[DOT_COUNTED_BY_GEP:%.*]] = getelementptr inbounds i8, ptr [[TMP0]], i64 8
// NO-SANITIZE-WITH-ATTR-NEXT: [[DOT_COUNTED_BY_LOAD:%.*]] = load i32, ptr [[DOT_COUNTED_BY_GEP]], align 4
// NO-SANITIZE-WITH-ATTR-NEXT: [[TMP1:%.*]] = shl i32 [[DOT_COUNTED_BY_LOAD]], 2
// NO-SANITIZE-WITH-ATTR-NEXT: [[DOTINV:%.*]] = icmp slt i32 [[DOT_COUNTED_BY_LOAD]], 0
// NO-SANITIZE-WITH-ATTR-NEXT: [[CONV:%.*]] = select i1 [[DOTINV]], i32 0, i32 [[TMP1]]
// NO-SANITIZE-WITH-ATTR-NEXT: [[TMP1:%.*]] = tail call i32 @llvm.smax.i32(i32 [[DOT_COUNTED_BY_LOAD]], i32 0)
// NO-SANITIZE-WITH-ATTR-NEXT: [[CONV:%.*]] = shl i32 [[TMP1]], 2
// NO-SANITIZE-WITH-ATTR-NEXT: [[ARRAY:%.*]] = getelementptr inbounds nuw i8, ptr [[TMP0]], i64 12
// NO-SANITIZE-WITH-ATTR-NEXT: [[IDXPROM4:%.*]] = sext i32 [[IDX2]] to i64
// NO-SANITIZE-WITH-ATTR-NEXT: [[ARRAYIDX5:%.*]] = getelementptr inbounds [0 x i32], ptr [[ARRAY]], i64 0, i64 [[IDXPROM4]]
Expand Down
2 changes: 1 addition & 1 deletion clang/test/CodeGen/builtins-elementwise-math.c
View file
Original file line number Diff line number Diff line change
Expand Up @@ -443,7 +443,7 @@ void test_builtin_elementwise_bitreverse(si8 vi1, si8 vi2,
// CHECK-NEXT: call i32 @llvm.bitreverse.i32(i32 [[IA1]])
b = __builtin_elementwise_bitreverse(int_as_one);

// CHECK: call i32 @llvm.bitreverse.i32(i32 -10)
// CHECK: store i32 1879048191, ptr @b, align 4
b = __builtin_elementwise_bitreverse(-10);

// CHECK: [[SI:%.+]] = load i16, ptr %si.addr, align 2
Expand Down
74 changes: 74 additions & 0 deletions clang/test/CodeGenCXX/aarch64-mangle-sme-atts.cpp
View file
Original file line number Diff line number Diff line change
@@ -0,0 +1,74 @@
// RUN: %clang_cc1 -triple aarch64-none-linux-gnu -target-feature +sve -target-feature +sme -target-feature +sme2 %s -emit-llvm -o - | FileCheck %s

typedef __attribute__((neon_vector_type(2))) int int32x2_t;

//
// Streaming-Mode Attributes
//

// CHECK: define dso_local void @_Z12fn_streamingP11__SME_ATTRSIFvvELj1EE
void fn_streaming(void (*foo)() __arm_streaming) { foo(); }

// CHECK: define dso_local void @_Z23fn_streaming_compatibleP11__SME_ATTRSIFivELj2EE(
void fn_streaming_compatible(int (*foo)() __arm_streaming_compatible) { foo(); }

//
// ZA Attributes
//

// CHECK: define dso_local void @_Z15fn_za_preservedP11__SME_ATTRSIF11__Int32x2_tvELj32EE(
__arm_new("za") void fn_za_preserved(int32x2_t (*foo)() __arm_preserves("za")) { foo(); }

// CHECK: define dso_local void @_Z8fn_za_inP11__SME_ATTRSIFvu13__SVFloat64_tELj8EES_(
__arm_new("za") void fn_za_in(void (*foo)(__SVFloat64_t) __arm_in("za"), __SVFloat64_t x) { foo(x); }

// CHECK: define dso_local noundef i32 @_Z9fn_za_outP11__SME_ATTRSIFivELj16EE(
__arm_new("za") int fn_za_out(int (*foo)() __arm_out("za")) { return foo(); }

// CHECK: define dso_local void @_Z11fn_za_inoutP11__SME_ATTRSIFvvELj24EE(
__arm_new("za") void fn_za_inout(void (*foo)() __arm_inout("za")) { foo(); }


//
// ZT0 Attributes
//

// CHECK: define dso_local void @_Z16fn_zt0_preservedP11__SME_ATTRSIFivELj256EE(
__arm_new("zt0") void fn_zt0_preserved(int (*foo)() __arm_preserves("zt0")) { foo(); }

// CHECK: define dso_local void @_Z9fn_zt0_inP11__SME_ATTRSIFivELj64EE(
__arm_new("zt0") void fn_zt0_in(int (*foo)() __arm_in("zt0")) { foo(); }

// CHECK: define dso_local void @_Z10fn_zt0_outP11__SME_ATTRSIFivELj128EE(
__arm_new("zt0") void fn_zt0_out(int (*foo)() __arm_out("zt0")) { foo(); }

// CHECK: define dso_local void @_Z12fn_zt0_inoutP11__SME_ATTRSIFivELj192EE(
__arm_new("zt0") void fn_zt0_inout(int (*foo)() __arm_inout("zt0")) { foo(); }

//
// Streaming-mode, ZA & ZT0 Attributes
//

// CHECK: define dso_local void @_Z17fn_all_attr_typesP11__SME_ATTRSIFivELj282EE(
__arm_new("za") __arm_new("zt0")
void fn_all_attr_types(int (*foo)() __arm_streaming_compatible __arm_inout("za") __arm_preserves("zt0"))
{ foo(); }

//
// No SME Attributes
//

// CHECK: define dso_local void @_Z12no_sme_attrsPFvvE(
void no_sme_attrs(void (*foo)()) { foo(); }

// CHECK: define dso_local void @_Z24locally_streaming_callerPFvvE(
__arm_locally_streaming void locally_streaming_caller(void (*foo)()) { foo(); }

// CHECK: define dso_local void @_Z16streaming_callerv(
void streaming_caller() __arm_streaming {}

// CHECK: define dso_local void @_Z16za_shared_callerv(
void za_shared_caller() __arm_in("za") {}

// CHECK: define dso_local void @_Z17zt0_shared_callerv(
void zt0_shared_caller() __arm_out("zt0") {}
4 changes: 2 additions & 2 deletions clang/test/CodeGenHLSL/builtins/RWBuffer-constructor.hlsl
View file
Original file line number Diff line number Diff line change
Expand Up @@ -6,10 +6,10 @@

RWBuffer<float> Buf : register(u5, space3);

// CHECK: %"class.hlsl::RWBuffer" = type { target("dx.TypedBuffer", float, 1, 0, 0), float }
// CHECK: %"class.hlsl::RWBuffer" = type { target("dx.TypedBuffer", float, 1, 0, 0) }
// CHECK: @Buf = global %"class.hlsl::RWBuffer" zeroinitializer, align 4

// CHECK: define linkonce_odr void @_ZN4hlsl8RWBufferIfEC2Ev(ptr noundef nonnull align 4 dereferenceable(8) %this)
// CHECK: define linkonce_odr void @_ZN4hlsl8RWBufferIfEC2Ev(ptr noundef nonnull align 4 dereferenceable(4) %this)
// CHECK-NEXT: entry:

// CHECK: define internal void @_GLOBAL__sub_I_RWBuffer_constructor.hlsl()
Expand Down
56 changes: 26 additions & 30 deletions clang/test/CodeGenHLSL/builtins/RWBuffer-elementtype.hlsl
View file
Original file line number Diff line number Diff line change
@@ -1,37 +1,33 @@
// RUN: %clang_cc1 -triple dxil-pc-shadermodel6.2-compute -finclude-default-header -fnative-half-type -emit-llvm -o - %s | FileCheck %s -check-prefixes=DXIL
// RUN: %clang_cc1 -triple spirv-pc-vulkan-compute -finclude-default-header -fnative-half-type -emit-llvm -o - %s | FileCheck %s -check-prefixes=SPIRV

// NOTE: The type name number and whether the struct is packed or not will mostly
// likely change once subscript operators are properly implemented (llvm/llvm-project#95956)
// and theinterim field of the contained type is removed.
// DXIL: %"class.hlsl::RWBuffer" = type { target("dx.TypedBuffer", i16, 1, 0, 1) }
// DXIL: %"class.hlsl::RWBuffer.0" = type { target("dx.TypedBuffer", i16, 1, 0, 0) }
// DXIL: %"class.hlsl::RWBuffer.1" = type { target("dx.TypedBuffer", i32, 1, 0, 1) }
// DXIL: %"class.hlsl::RWBuffer.2" = type { target("dx.TypedBuffer", i32, 1, 0, 0) }
// DXIL: %"class.hlsl::RWBuffer.3" = type { target("dx.TypedBuffer", i64, 1, 0, 1) }
// DXIL: %"class.hlsl::RWBuffer.4" = type { target("dx.TypedBuffer", i64, 1, 0, 0) }
// DXIL: %"class.hlsl::RWBuffer.5" = type { target("dx.TypedBuffer", half, 1, 0, 0) }
// DXIL: %"class.hlsl::RWBuffer.6" = type { target("dx.TypedBuffer", float, 1, 0, 0) }
// DXIL: %"class.hlsl::RWBuffer.7" = type { target("dx.TypedBuffer", double, 1, 0, 0) }
// DXIL: %"class.hlsl::RWBuffer.8" = type { target("dx.TypedBuffer", <4 x i16>, 1, 0, 0) }
// DXIL: %"class.hlsl::RWBuffer.9" = type { target("dx.TypedBuffer", <3 x i32>, 1, 0, 0) }
// DXIL: %"class.hlsl::RWBuffer.10" = type { target("dx.TypedBuffer", <2 x half>, 1, 0, 0) }
// DXIL: %"class.hlsl::RWBuffer.11" = type { target("dx.TypedBuffer", <3 x float>, 1, 0, 0) }

// DXIL: %"class.hlsl::RWBuffer" = type <{ target("dx.TypedBuffer", i16, 1, 0, 1)
// DXIL: %"class.hlsl::RWBuffer.0" = type <{ target("dx.TypedBuffer", i16, 1, 0, 0)
// DXIL: %"class.hlsl::RWBuffer.2" = type { target("dx.TypedBuffer", i32, 1, 0, 1)
// DXIL: %"class.hlsl::RWBuffer.3" = type { target("dx.TypedBuffer", i32, 1, 0, 0)
// DXIL: %"class.hlsl::RWBuffer.4" = type { target("dx.TypedBuffer", i64, 1, 0, 1)
// DXIL: %"class.hlsl::RWBuffer.5" = type { target("dx.TypedBuffer", i64, 1, 0, 0)
// DXIL: %"class.hlsl::RWBuffer.6" = type <{ target("dx.TypedBuffer", half, 1, 0, 0)
// DXIL: %"class.hlsl::RWBuffer.8" = type { target("dx.TypedBuffer", float, 1, 0, 0)
// DXIL: %"class.hlsl::RWBuffer.9" = type { target("dx.TypedBuffer", double, 1, 0, 0)
// DXIL: %"class.hlsl::RWBuffer.10" = type { target("dx.TypedBuffer", <4 x i16>, 1, 0, 0)
// DXIL: %"class.hlsl::RWBuffer.11" = type { target("dx.TypedBuffer", <3 x i32>, 1, 0, 0)
// DXIL: %"class.hlsl::RWBuffer.12" = type { target("dx.TypedBuffer", <2 x half>, 1, 0, 0)
// DXIL: %"class.hlsl::RWBuffer.13" = type { target("dx.TypedBuffer", <3 x float>, 1, 0, 0)

// SPIRV: %"class.hlsl::RWBuffer" = type <{ target("spirv.Image", i16, 5, 2, 0, 0, 2, 0), i16, [6 x i8] }>
// SPIRV: %"class.hlsl::RWBuffer.0" = type <{ target("spirv.Image", i16, 5, 2, 0, 0, 2, 0), i16, [6 x i8] }>
// SPIRV: %"class.hlsl::RWBuffer.2" = type <{ target("spirv.Image", i32, 5, 2, 0, 0, 2, 0), i32, [4 x i8] }>
// SPIRV: %"class.hlsl::RWBuffer.4" = type <{ target("spirv.Image", i32, 5, 2, 0, 0, 2, 0), i32, [4 x i8] }>
// SPIRV: %"class.hlsl::RWBuffer.6" = type { target("spirv.Image", i64, 5, 2, 0, 0, 2, 0), i64 }
// SPIRV: %"class.hlsl::RWBuffer.7" = type { target("spirv.Image", i64, 5, 2, 0, 0, 2, 0), i64 }
// SPIRV: %"class.hlsl::RWBuffer.8" = type <{ target("spirv.Image", half, 5, 2, 0, 0, 2, 0), half, [6 x i8] }>
// SPIRV: %"class.hlsl::RWBuffer.10" = type <{ target("spirv.Image", float, 5, 2, 0, 0, 2, 0), float, [4 x i8] }>
// SPIRV: %"class.hlsl::RWBuffer.12" = type { target("spirv.Image", double, 5, 2, 0, 0, 2, 0), double }
// SPIRV: %"class.hlsl::RWBuffer.13" = type { target("spirv.Image", i16, 5, 2, 0, 0, 2, 0), <4 x i16> }
// SPIRV: %"class.hlsl::RWBuffer.14" = type { target("spirv.Image", i32, 5, 2, 0, 0, 2, 0), <3 x i32> }
// SPIRV: %"class.hlsl::RWBuffer.15" = type <{ target("spirv.Image", half, 5, 2, 0, 0, 2, 0), <2 x half>, [4 x i8] }>
// SPIRV: %"class.hlsl::RWBuffer.17" = type { target("spirv.Image", float, 5, 2, 0, 0, 2, 0), <3 x float> }
// SPIRV: %"class.hlsl::RWBuffer" = type { target("spirv.Image", i16, 5, 2, 0, 0, 2, 0) }
// SPIRV: %"class.hlsl::RWBuffer.0" = type { target("spirv.Image", i16, 5, 2, 0, 0, 2, 0) }
// SPIRV: %"class.hlsl::RWBuffer.1" = type { target("spirv.Image", i32, 5, 2, 0, 0, 2, 0) }
// SPIRV: %"class.hlsl::RWBuffer.2" = type { target("spirv.Image", i32, 5, 2, 0, 0, 2, 0) }
// SPIRV: %"class.hlsl::RWBuffer.3" = type { target("spirv.Image", i64, 5, 2, 0, 0, 2, 0) }
// SPIRV: %"class.hlsl::RWBuffer.4" = type { target("spirv.Image", i64, 5, 2, 0, 0, 2, 0) }
// SPIRV: %"class.hlsl::RWBuffer.5" = type { target("spirv.Image", half, 5, 2, 0, 0, 2, 0) }
// SPIRV: %"class.hlsl::RWBuffer.6" = type { target("spirv.Image", float, 5, 2, 0, 0, 2, 0) }
// SPIRV: %"class.hlsl::RWBuffer.7" = type { target("spirv.Image", double, 5, 2, 0, 0, 2, 0) }
// SPIRV: %"class.hlsl::RWBuffer.8" = type { target("spirv.Image", i16, 5, 2, 0, 0, 2, 0) }
// SPIRV: %"class.hlsl::RWBuffer.9" = type { target("spirv.Image", i32, 5, 2, 0, 0, 2, 0) }
// SPIRV: %"class.hlsl::RWBuffer.10" = type { target("spirv.Image", half, 5, 2, 0, 0, 2, 0) }
// SPIRV: %"class.hlsl::RWBuffer.11" = type { target("spirv.Image", float, 5, 2, 0, 0, 2, 0) }



Expand Down
13 changes: 5 additions & 8 deletions clang/test/CodeGenHLSL/builtins/RWBuffer-subscript.hlsl
View file
Original file line number Diff line number Diff line change
Expand Up @@ -5,13 +5,10 @@ RWBuffer<int> Out;

[numthreads(1,1,1)]
void main(unsigned GI : SV_GroupIndex) {
// CHECK: define void @main()
// CHECK: %[[INPTR:.*]] = call noundef nonnull align 4 dereferenceable(4) ptr @llvm.dx.resource.getpointer.p0.tdx.TypedBuffer_i32_1_0_1t(target("dx.TypedBuffer", i32, 1, 0, 1) %{{.*}}, i32 %{{.*}})
// CHECK: %[[LOAD:.*]] = load i32, ptr %[[INPTR]]
// CHECK: %[[OUTPTR:.*]] = call noundef nonnull align 4 dereferenceable(4) ptr @llvm.dx.resource.getpointer.p0.tdx.TypedBuffer_i32_1_0_1t(target("dx.TypedBuffer", i32, 1, 0, 1) %{{.*}}, i32 %{{.*}})
// CHECK: store i32 %[[LOAD]], ptr %[[OUTPTR]]
Out[GI] = In[GI];
}

// Even at -O0 the subscript operators get inlined. The -O0 IR is a bit messy
// and confusing to follow so the match here is pretty weak.

// CHECK: define void @main()
// Verify inlining leaves only calls to "llvm." intrinsics
// CHECK-NOT: call {{[^@]*}} @{{[^l][^l][^v][^m][^\.]}}
// CHECK: ret void
30 changes: 13 additions & 17 deletions clang/test/CodeGenHLSL/builtins/RWStructuredBuffer-elementtype.hlsl
View file
Original file line number Diff line number Diff line change
@@ -1,22 +1,18 @@
// RUN: %clang_cc1 -triple dxil-pc-shadermodel6.2-compute -finclude-default-header -fnative-half-type -emit-llvm -o - %s | FileCheck %s

// NOTE: The number in type name and whether the struct is packed or not will mostly
// likely change once subscript operators are properly implemented (llvm/llvm-project#95956)
// and theinterim field of the contained type is removed.

// CHECK: %"class.hlsl::RWStructuredBuffer" = type <{ target("dx.RawBuffer", i16, 1, 0)
// CHECK: %"class.hlsl::RWStructuredBuffer.0" = type <{ target("dx.RawBuffer", i16, 1, 0)
// CHECK: %"class.hlsl::RWStructuredBuffer.2" = type { target("dx.RawBuffer", i32, 1, 0)
// CHECK: %"class.hlsl::RWStructuredBuffer.3" = type { target("dx.RawBuffer", i32, 1, 0)
// CHECK: %"class.hlsl::RWStructuredBuffer.4" = type { target("dx.RawBuffer", i64, 1, 0)
// CHECK: %"class.hlsl::RWStructuredBuffer.5" = type { target("dx.RawBuffer", i64, 1, 0)
// CHECK: %"class.hlsl::RWStructuredBuffer.6" = type <{ target("dx.RawBuffer", half, 1, 0)
// CHECK: %"class.hlsl::RWStructuredBuffer.8" = type { target("dx.RawBuffer", float, 1, 0)
// CHECK: %"class.hlsl::RWStructuredBuffer.9" = type { target("dx.RawBuffer", double, 1, 0)
// CHECK: %"class.hlsl::RWStructuredBuffer.10" = type { target("dx.RawBuffer", <4 x i16>, 1, 0)
// CHECK: %"class.hlsl::RWStructuredBuffer.11" = type { target("dx.RawBuffer", <3 x i32>, 1, 0)
// CHECK: %"class.hlsl::RWStructuredBuffer.12" = type { target("dx.RawBuffer", <2 x half>, 1, 0)
// CHECK: %"class.hlsl::RWStructuredBuffer.13" = type { target("dx.RawBuffer", <3 x float>, 1, 0)
// CHECK: %"class.hlsl::RWStructuredBuffer" = type { target("dx.RawBuffer", i16, 1, 0) }
// CHECK: %"class.hlsl::RWStructuredBuffer.0" = type { target("dx.RawBuffer", i16, 1, 0) }
// CHECK: %"class.hlsl::RWStructuredBuffer.1" = type { target("dx.RawBuffer", i32, 1, 0) }
// CHECK: %"class.hlsl::RWStructuredBuffer.2" = type { target("dx.RawBuffer", i32, 1, 0) }
// CHECK: %"class.hlsl::RWStructuredBuffer.3" = type { target("dx.RawBuffer", i64, 1, 0) }
// CHECK: %"class.hlsl::RWStructuredBuffer.4" = type { target("dx.RawBuffer", i64, 1, 0) }
// CHECK: %"class.hlsl::RWStructuredBuffer.5" = type { target("dx.RawBuffer", half, 1, 0) }
// CHECK: %"class.hlsl::RWStructuredBuffer.6" = type { target("dx.RawBuffer", float, 1, 0) }
// CHECK: %"class.hlsl::RWStructuredBuffer.7" = type { target("dx.RawBuffer", double, 1, 0) }
// CHECK: %"class.hlsl::RWStructuredBuffer.8" = type { target("dx.RawBuffer", <4 x i16>, 1, 0) }
// CHECK: %"class.hlsl::RWStructuredBuffer.9" = type { target("dx.RawBuffer", <3 x i32>, 1, 0) }
// CHECK: %"class.hlsl::RWStructuredBuffer.10" = type { target("dx.RawBuffer", <2 x half>, 1, 0) }
// CHECK: %"class.hlsl::RWStructuredBuffer.11" = type { target("dx.RawBuffer", <3 x float>, 1, 0) }

RWStructuredBuffer<int16_t> BufI16;
RWStructuredBuffer<uint16_t> BufU16;
Expand Down
32 changes: 14 additions & 18 deletions clang/test/CodeGenHLSL/builtins/RasterizerOrderedStructuredBuffer-elementtype.hlsl
View file
Original file line number Diff line number Diff line change
@@ -1,28 +1,24 @@
// RUN: %clang_cc1 -triple dxil-pc-shadermodel6.2-compute -finclude-default-header -fnative-half-type -emit-llvm -o - %s | FileCheck %s -check-prefixes=DXIL

// NOTE: The number in type name and whether the struct is packed or not will mostly
// likely change once subscript operators are properly implemented (llvm/llvm-project#95956)
// and theinterim field of the contained type is removed.

struct MyStruct {
float4 a;
int2 b;
};

// DXIL: %"class.hlsl::RasterizerOrderedStructuredBuffer" = type <{ target("dx.RawBuffer", i16, 1, 1)
// DXIL: %"class.hlsl::RasterizerOrderedStructuredBuffer.0" = type <{ target("dx.RawBuffer", i16, 1, 1)
// DXIL: %"class.hlsl::RasterizerOrderedStructuredBuffer.2" = type { target("dx.RawBuffer", i32, 1, 1)
// DXIL: %"class.hlsl::RasterizerOrderedStructuredBuffer.3" = type { target("dx.RawBuffer", i32, 1, 1)
// DXIL: %"class.hlsl::RasterizerOrderedStructuredBuffer.4" = type { target("dx.RawBuffer", i64, 1, 1)
// DXIL: %"class.hlsl::RasterizerOrderedStructuredBuffer.5" = type { target("dx.RawBuffer", i64, 1, 1)
// DXIL: %"class.hlsl::RasterizerOrderedStructuredBuffer.6" = type <{ target("dx.RawBuffer", half, 1, 1)
// DXIL: %"class.hlsl::RasterizerOrderedStructuredBuffer.8" = type { target("dx.RawBuffer", float, 1, 1)
// DXIL: %"class.hlsl::RasterizerOrderedStructuredBuffer.9" = type { target("dx.RawBuffer", double, 1, 1)
// DXIL: %"class.hlsl::RasterizerOrderedStructuredBuffer.10" = type { target("dx.RawBuffer", <4 x i16>, 1, 1)
// DXIL: %"class.hlsl::RasterizerOrderedStructuredBuffer.11" = type { target("dx.RawBuffer", <3 x i32>, 1, 1)
// DXIL: %"class.hlsl::RasterizerOrderedStructuredBuffer.12" = type { target("dx.RawBuffer", <2 x half>, 1, 1)
// DXIL: %"class.hlsl::RasterizerOrderedStructuredBuffer.13" = type { target("dx.RawBuffer", <3 x float>, 1, 1)
// DXIL: %"class.hlsl::RasterizerOrderedStructuredBuffer.14" = type { target("dx.RawBuffer", %struct.MyStruct = type { <4 x float>, <2 x i32>, [8 x i8] }, 1, 1)
// DXIL: %"class.hlsl::RasterizerOrderedStructuredBuffer" = type { target("dx.RawBuffer", i16, 1, 1) }
// DXIL: %"class.hlsl::RasterizerOrderedStructuredBuffer.0" = type { target("dx.RawBuffer", i16, 1, 1) }
// DXIL: %"class.hlsl::RasterizerOrderedStructuredBuffer.1" = type { target("dx.RawBuffer", i32, 1, 1) }
// DXIL: %"class.hlsl::RasterizerOrderedStructuredBuffer.2" = type { target("dx.RawBuffer", i32, 1, 1) }
// DXIL: %"class.hlsl::RasterizerOrderedStructuredBuffer.3" = type { target("dx.RawBuffer", i64, 1, 1) }
// DXIL: %"class.hlsl::RasterizerOrderedStructuredBuffer.4" = type { target("dx.RawBuffer", i64, 1, 1) }
// DXIL: %"class.hlsl::RasterizerOrderedStructuredBuffer.5" = type { target("dx.RawBuffer", half, 1, 1) }
// DXIL: %"class.hlsl::RasterizerOrderedStructuredBuffer.6" = type { target("dx.RawBuffer", float, 1, 1) }
// DXIL: %"class.hlsl::RasterizerOrderedStructuredBuffer.7" = type { target("dx.RawBuffer", double, 1, 1) }
// DXIL: %"class.hlsl::RasterizerOrderedStructuredBuffer.8" = type { target("dx.RawBuffer", <4 x i16>, 1, 1) }
// DXIL: %"class.hlsl::RasterizerOrderedStructuredBuffer.9" = type { target("dx.RawBuffer", <3 x i32>, 1, 1) }
// DXIL: %"class.hlsl::RasterizerOrderedStructuredBuffer.10" = type { target("dx.RawBuffer", <2 x half>, 1, 1) }
// DXIL: %"class.hlsl::RasterizerOrderedStructuredBuffer.11" = type { target("dx.RawBuffer", <3 x float>, 1, 1) }
// DXIL: %"class.hlsl::RasterizerOrderedStructuredBuffer.12" = type { target("dx.RawBuffer", %struct.MyStruct = type { <4 x float>, <2 x i32>, [8 x i8] }, 1, 1) }

RasterizerOrderedStructuredBuffer<int16_t> BufI16;
RasterizerOrderedStructuredBuffer<uint16_t> BufU16;
Expand Down
30 changes: 13 additions & 17 deletions clang/test/CodeGenHLSL/builtins/StructuredBuffer-elementtype.hlsl
View file
Original file line number Diff line number Diff line change
@@ -1,22 +1,18 @@
// RUN: %clang_cc1 -triple dxil-pc-shadermodel6.2-compute -finclude-default-header -fnative-half-type -emit-llvm -o - %s | FileCheck %s

// NOTE: The number in type name and whether the struct is packed or not will mostly
// likely change once subscript operators are properly implemented (llvm/llvm-project#95956)
// and theinterim field of the contained type is removed.

// CHECK: %"class.hlsl::StructuredBuffer" = type <{ target("dx.RawBuffer", i16, 0, 0)
// CHECK: %"class.hlsl::StructuredBuffer.0" = type <{ target("dx.RawBuffer", i16, 0, 0)
// CHECK: %"class.hlsl::StructuredBuffer.2" = type { target("dx.RawBuffer", i32, 0, 0)
// CHECK: %"class.hlsl::StructuredBuffer.3" = type { target("dx.RawBuffer", i32, 0, 0)
// CHECK: %"class.hlsl::StructuredBuffer.4" = type { target("dx.RawBuffer", i64, 0, 0)
// CHECK: %"class.hlsl::StructuredBuffer.5" = type { target("dx.RawBuffer", i64, 0, 0)
// CHECK: %"class.hlsl::StructuredBuffer.6" = type <{ target("dx.RawBuffer", half, 0, 0)
// CHECK: %"class.hlsl::StructuredBuffer.8" = type { target("dx.RawBuffer", float, 0, 0)
// CHECK: %"class.hlsl::StructuredBuffer.9" = type { target("dx.RawBuffer", double, 0, 0)
// CHECK: %"class.hlsl::StructuredBuffer.10" = type { target("dx.RawBuffer", <4 x i16>, 0, 0)
// CHECK: %"class.hlsl::StructuredBuffer.11" = type { target("dx.RawBuffer", <3 x i32>, 0, 0)
// CHECK: %"class.hlsl::StructuredBuffer.12" = type { target("dx.RawBuffer", <2 x half>, 0, 0)
// CHECK: %"class.hlsl::StructuredBuffer.13" = type { target("dx.RawBuffer", <3 x float>, 0, 0)
// CHECK: %"class.hlsl::StructuredBuffer" = type { target("dx.RawBuffer", i16, 0, 0) }
// CHECK: %"class.hlsl::StructuredBuffer.0" = type { target("dx.RawBuffer", i16, 0, 0) }
// CHECK: %"class.hlsl::StructuredBuffer.1" = type { target("dx.RawBuffer", i32, 0, 0) }
// CHECK: %"class.hlsl::StructuredBuffer.2" = type { target("dx.RawBuffer", i32, 0, 0) }
// CHECK: %"class.hlsl::StructuredBuffer.3" = type { target("dx.RawBuffer", i64, 0, 0) }
// CHECK: %"class.hlsl::StructuredBuffer.4" = type { target("dx.RawBuffer", i64, 0, 0) }
// CHECK: %"class.hlsl::StructuredBuffer.5" = type { target("dx.RawBuffer", half, 0, 0) }
// CHECK: %"class.hlsl::StructuredBuffer.6" = type { target("dx.RawBuffer", float, 0, 0) }
// CHECK: %"class.hlsl::StructuredBuffer.7" = type { target("dx.RawBuffer", double, 0, 0) }
// CHECK: %"class.hlsl::StructuredBuffer.8" = type { target("dx.RawBuffer", <4 x i16>, 0, 0) }
// CHECK: %"class.hlsl::StructuredBuffer.9" = type { target("dx.RawBuffer", <3 x i32>, 0, 0) }
// CHECK: %"class.hlsl::StructuredBuffer.10" = type { target("dx.RawBuffer", <2 x half>, 0, 0) }
// CHECK: %"class.hlsl::StructuredBuffer.11" = type { target("dx.RawBuffer", <3 x float>, 0, 0) }

StructuredBuffer<int16_t> BufI16;
StructuredBuffer<uint16_t> BufU16;
Expand Down
16 changes: 8 additions & 8 deletions clang/test/CodeGenHLSL/builtins/StructuredBuffers-constructors.hlsl
View file
Original file line number Diff line number Diff line change
Expand Up @@ -9,26 +9,26 @@ AppendStructuredBuffer<float> Buf3 : register(u3);
ConsumeStructuredBuffer<float> Buf4 : register(u4);
RasterizerOrderedStructuredBuffer<float> Buf5 : register(u1, space2);

// CHECK: %"class.hlsl::StructuredBuffer" = type { target("dx.RawBuffer", float, 0, 0)
// CHECK: %"class.hlsl::RWStructuredBuffer" = type { target("dx.RawBuffer", float, 1, 0)
// CHECK: %"class.hlsl::AppendStructuredBuffer" = type { target("dx.RawBuffer", float, 1, 0)
// CHECK: %"class.hlsl::ConsumeStructuredBuffer" = type { target("dx.RawBuffer", float, 1, 0)
// CHECK: %"class.hlsl::RasterizerOrderedStructuredBuffer" = type { target("dx.RawBuffer", float, 1, 1), float }
// CHECK: %"class.hlsl::StructuredBuffer" = type { target("dx.RawBuffer", float, 0, 0) }
// CHECK: %"class.hlsl::RWStructuredBuffer" = type { target("dx.RawBuffer", float, 1, 0) }
// CHECK: %"class.hlsl::AppendStructuredBuffer" = type { target("dx.RawBuffer", float, 1, 0) }
// CHECK: %"class.hlsl::ConsumeStructuredBuffer" = type { target("dx.RawBuffer", float, 1, 0) }
// CHECK: %"class.hlsl::RasterizerOrderedStructuredBuffer" = type { target("dx.RawBuffer", float, 1, 1) }

// CHECK: @Buf = global %"class.hlsl::StructuredBuffer" zeroinitializer, align 4
// CHECK: @Buf2 = global %"class.hlsl::RWStructuredBuffer" zeroinitializer, align 4
// CHECK: @Buf3 = global %"class.hlsl::AppendStructuredBuffer" zeroinitializer, align 4
// CHECK: @Buf4 = global %"class.hlsl::ConsumeStructuredBuffer" zeroinitializer, align 4
// CHECK: @Buf5 = global %"class.hlsl::RasterizerOrderedStructuredBuffer" zeroinitializer, align 4

// CHECK: define linkonce_odr void @_ZN4hlsl16StructuredBufferIfEC2Ev(ptr noundef nonnull align 4 dereferenceable(8) %this)
// CHECK: define linkonce_odr void @_ZN4hlsl16StructuredBufferIfEC2Ev(ptr noundef nonnull align 4 dereferenceable(4) %this)
// CHECK-NEXT: entry:
// CHECK: define linkonce_odr void @_ZN4hlsl18RWStructuredBufferIfEC2Ev(ptr noundef nonnull align 4 dereferenceable(8) %this)
// CHECK: define linkonce_odr void @_ZN4hlsl18RWStructuredBufferIfEC2Ev(ptr noundef nonnull align 4 dereferenceable(4) %this)
// CHECK-NEXT: entry:
// CHECK: define linkonce_odr void @_ZN4hlsl22AppendStructuredBufferIfEC2Ev(ptr noundef nonnull align 4 dereferenceable(4) %this)
// CHECK-NEXT: entry:
// CHECK: define linkonce_odr void @_ZN4hlsl23ConsumeStructuredBufferIfEC2Ev(ptr noundef nonnull align 4 dereferenceable(4) %this)
// CHECK: define linkonce_odr void @_ZN4hlsl33RasterizerOrderedStructuredBufferIfEC2Ev(ptr noundef nonnull align 4 dereferenceable(8) %this)
// CHECK: define linkonce_odr void @_ZN4hlsl33RasterizerOrderedStructuredBufferIfEC2Ev(ptr noundef nonnull align 4 dereferenceable(4) %this)
// CHECK-NEXT: entry:

// CHECK: define internal void @_GLOBAL__sub_I_StructuredBuffers_constructors.hlsl()
Expand Down
2 changes: 1 addition & 1 deletion clang/test/CodeGenHLSL/builtins/StructuredBuffers-methods-lib.hlsl
View file
Original file line number Diff line number Diff line change
Expand Up @@ -6,7 +6,7 @@
RWStructuredBuffer<float> RWSB1 : register(u0);
RWStructuredBuffer<float> RWSB2 : register(u1);

// CHECK: %"class.hlsl::RWStructuredBuffer" = type { target("dx.RawBuffer", float, 1, 0), float }
// CHECK: %"class.hlsl::RWStructuredBuffer" = type { target("dx.RawBuffer", float, 1, 0) }

export void TestIncrementCounter() {
RWSB1.IncrementCounter();
Expand Down
2 changes: 1 addition & 1 deletion clang/test/CodeGenHLSL/builtins/StructuredBuffers-methods-ps.hlsl
View file
Original file line number Diff line number Diff line change
Expand Up @@ -6,7 +6,7 @@
RWStructuredBuffer<float> RWSB1, RWSB2;
RasterizerOrderedStructuredBuffer<float> ROSB1, ROSB2;

// CHECK: %"class.hlsl::RWStructuredBuffer" = type { target("dx.RawBuffer", float, 1, 0), float }
// CHECK: %"class.hlsl::RWStructuredBuffer" = type { target("dx.RawBuffer", float, 1, 0) }

export void TestIncrementCounter() {
// CHECK: define void @_Z20TestIncrementCounterv()
Expand Down
19 changes: 11 additions & 8 deletions clang/test/CodeGenHLSL/builtins/StructuredBuffers-subscripts.hlsl
View file
Original file line number Diff line number Diff line change
Expand Up @@ -6,14 +6,17 @@ RasterizerOrderedStructuredBuffer<int> Out2;

[numthreads(1,1,1)]
void main(unsigned GI : SV_GroupIndex) {
// CHECK: define void @main()

// CHECK: %[[INPTR:.*]] = call noundef nonnull align 4 dereferenceable(4) ptr @llvm.dx.resource.getpointer.p0.tdx.RawBuffer_i32_0_0t(target("dx.RawBuffer", i32, 0, 0) %{{.*}}, i32 %{{.*}})
// CHECK: %[[LOAD:.*]] = load i32, ptr %[[INPTR]]
// CHECK: %[[OUT1PTR:.*]] = call noundef nonnull align 4 dereferenceable(4) ptr @llvm.dx.resource.getpointer.p0.tdx.RawBuffer_i32_1_0t(target("dx.RawBuffer", i32, 1, 0) %{{.*}}, i32 %{{.*}})
// CHECK: store i32 %[[LOAD]], ptr %[[OUT1PTR]]
Out1[GI] = In[GI];

// CHECK: %[[INPTR:.*]] = call noundef nonnull align 4 dereferenceable(4) ptr @llvm.dx.resource.getpointer.p0.tdx.RawBuffer_i32_0_0t(target("dx.RawBuffer", i32, 0, 0) %{{.*}}, i32 %{{.*}})
// CHECK: %[[LOAD:.*]] = load i32, ptr %[[INPTR]]
// CHECK: %[[OUT2PTR:.*]] = call noundef nonnull align 4 dereferenceable(4) ptr @llvm.dx.resource.getpointer.p0.tdx.RawBuffer_i32_1_1t(target("dx.RawBuffer", i32, 1, 1) %{{.*}}, i32 %{{.*}})
// CHECK: store i32 %[[LOAD]], ptr %[[OUT2PTR]]
Out2[GI] = In[GI];
}

// Even at -O0 the subscript operators get inlined. The -O0 IR is a bit messy
// and confusing to follow so the match here is pretty weak.

// CHECK: define void @main()
// Verify inlining leaves only calls to "llvm." intrinsics
// CHECK-NOT: call {{[^@]*}} @{{[^l][^l][^v][^m][^\.]}}
// CHECK: ret void
13 changes: 6 additions & 7 deletions clang/test/CodeGenHLSL/builtins/hlsl_resource_t.hlsl
View file
Original file line number Diff line number Diff line change
Expand Up @@ -20,9 +20,9 @@ void fb(handle_float_t a) {
handle_float_t b = a;
}

// CHECK: define void @_Z2fcN4hlsl8RWBufferIDv4_fEE(ptr noundef byval(%"class.hlsl::RWBuffer") align 16 %a)
// CHECK: call void @_Z4foo2N4hlsl8RWBufferIDv4_fEE(ptr noundef byval(%"class.hlsl::RWBuffer") align 16 %agg.tmp)
// CHECK: declare void @_Z4foo2N4hlsl8RWBufferIDv4_fEE(ptr noundef byval(%"class.hlsl::RWBuffer") align 16)
// CHECK: define void @_Z2fcN4hlsl8RWBufferIDv4_fEE(ptr noundef byval(%"class.hlsl::RWBuffer") align 4 %a)
// CHECK: call void @_Z4foo2N4hlsl8RWBufferIDv4_fEE(ptr noundef byval(%"class.hlsl::RWBuffer") align 4 %agg.tmp)
// CHECK: declare void @_Z4foo2N4hlsl8RWBufferIDv4_fEE(ptr noundef byval(%"class.hlsl::RWBuffer") align 4)
void foo2(RWBuffer<float4> buf);

void fc(RWBuffer<float4> a) {
Expand All @@ -38,9 +38,9 @@ struct MyStruct {
int2 i;
};

// CHECK: define void @_Z2feN4hlsl16StructuredBufferI8MyStructEE(ptr noundef byval(%"class.hlsl::StructuredBuffer") align 16 %a)
// CHECK: call void @_Z4foo3N4hlsl16StructuredBufferI8MyStructEE(ptr noundef byval(%"class.hlsl::StructuredBuffer") align 16 %agg.tmp)
// CHECK: declare void @_Z4foo3N4hlsl16StructuredBufferI8MyStructEE(ptr noundef byval(%"class.hlsl::StructuredBuffer") align 16)
// CHECK: define void @_Z2feN4hlsl16StructuredBufferI8MyStructEE(ptr noundef byval(%"class.hlsl::StructuredBuffer") align 4 %a)
// CHECK: call void @_Z4foo3N4hlsl16StructuredBufferI8MyStructEE(ptr noundef byval(%"class.hlsl::StructuredBuffer") align 4 %agg.tmp)
// CHECK: declare void @_Z4foo3N4hlsl16StructuredBufferI8MyStructEE(ptr noundef byval(%"class.hlsl::StructuredBuffer") align 4)
void foo3(StructuredBuffer<MyStruct> buf);

void fe(StructuredBuffer<MyStruct> a) {
Expand All @@ -50,4 +50,3 @@ void fe(StructuredBuffer<MyStruct> a) {
void ff(StructuredBuffer<MyStruct> a) {
StructuredBuffer<MyStruct> b = a;
}

2 changes: 1 addition & 1 deletion clang/test/CodeGenHLSL/implicit-norecurse-attrib.hlsl
View file
Original file line number Diff line number Diff line change
Expand Up @@ -31,7 +31,7 @@ uint Find(Node SortedTree[MAX], uint key) {
}

// CHECK: Function Attrs:{{.*}}norecurse
// CHECK: define noundef i1 @_Z8InitTreeA100_4NodeN4hlsl8RWBufferIDv4_jEEj(ptr noundef byval([100 x %struct.Node]) align 4 %tree, ptr noundef byval(%"class.hlsl::RWBuffer") align 16 %encodedTree, i32 noundef %maxDepth) [[ExtAttr:\#[0-9]+]]
// CHECK: define noundef i1 @_Z8InitTreeA100_4NodeN4hlsl8RWBufferIDv4_jEEj(ptr noundef byval([100 x %struct.Node]) align 4 %tree, ptr noundef byval(%"class.hlsl::RWBuffer") align 4 %encodedTree, i32 noundef %maxDepth) [[ExtAttr:\#[0-9]+]]
// CHECK: ret i1
// Initialize tree with given buffer
// Imagine the inout works
Expand Down
23 changes: 23 additions & 0 deletions clang/test/CodeGenOpenCL/builtins-amdgcn-gfx950-read-tr.cl
View file
Original file line number Diff line number Diff line change
Expand Up @@ -4,6 +4,8 @@
typedef int v2i __attribute__((ext_vector_type(2)));
typedef int v3i __attribute__((ext_vector_type(3)));
typedef short v4s __attribute__((ext_vector_type(4)));
typedef half v4h __attribute__((ext_vector_type(4)));
typedef __bf16 v4y __attribute__((ext_vector_type(4)));

// GFX950-LABEL: define dso_local <2 x i32> @test_amdgcn_ds_read_b64_tr_b4_v2i32(
// GFX950-SAME: ptr addrspace(3) nocapture noundef readonly [[INPTR:%.*]]) local_unnamed_addr #[[ATTR0:[0-9]+]] {
Expand Down Expand Up @@ -48,3 +50,24 @@ v4s test_amdgcn_ds_read_b64_tr_b16_v2i16(local v4s* inptr)
{
return __builtin_amdgcn_ds_read_tr16_b64_v4i16(inptr);
}

// GFX950-LABEL: define dso_local <4 x half> @test_amdgcn_ds_read_b64_tr_b16_v2f16(
// GFX950-SAME: ptr addrspace(3) nocapture noundef readonly [[INPTR:%.*]]) local_unnamed_addr #[[ATTR0:[0-9]+]] {
// GFX950-NEXT: entry:
// GFX950-NEXT: [[TMP0:%.*]] = tail call <4 x half> @llvm.amdgcn.ds.read.tr16.b64.v4f16(ptr addrspace(3) [[INPTR]])
// GFX950-NEXT: ret <4 x half> [[TMP0]]
//
v4h test_amdgcn_ds_read_b64_tr_b16_v2f16(local v4h* inptr)
{
return __builtin_amdgcn_ds_read_tr16_b64_v4f16(inptr);
}

// GFX950-LABEL: define dso_local <4 x bfloat> @test_amdgcn_ds_read_b64_tr_b16_v2bf16(
// GFX950-SAME: ptr addrspace(3) nocapture noundef readonly [[INPTR:%.*]]) local_unnamed_addr #[[ATTR0:[0-9]+]] {
// GFX950-NEXT: entry:
// GFX950-NEXT: [[TMP0:%.*]] = tail call <4 x bfloat> @llvm.amdgcn.ds.read.tr16.b64.v4bf16(ptr addrspace(3) [[INPTR]])
// GFX950-NEXT: ret <4 x bfloat> [[TMP0]]
v4y test_amdgcn_ds_read_b64_tr_b16_v2bf16(local v4y* inptr)
{
return __builtin_amdgcn_ds_read_tr16_b64_v4bf16(inptr);
}
4 changes: 4 additions & 0 deletions clang/test/Driver/xtensa-char.c
View file
Original file line number Diff line number Diff line change
@@ -0,0 +1,4 @@
/// Check that char is unsigned by default.
// RUN: %clang -### %s --target=xtensa -c 2>&1 | FileCheck %s
// CHECK: "-cc1" "-triple" "xtensa"
// CHECK-SAME: "-fno-signed-char"
54 changes: 54 additions & 0 deletions clang/test/Parser/cxx2c-oxford-variadic-comma.cpp
View file
Original file line number Diff line number Diff line change
@@ -0,0 +1,54 @@
// RUN: %clang_cc1 -std=c++2c -fsyntax-only -fblocks -verify %s

void a(...);

void b(auto...);
void c(auto, ...);

void d(auto......); // expected-warning {{declaration of a variadic function without a comma before '...' is deprecated}} \
// expected-warning {{'...' in this location creates a C-style varargs function}} \
// expected-note {{preceding '...' declares a function parameter pack}} \
// expected-note {{insert ',' before '...' to silence this warning}}
void e(auto..., ...);

void f(auto x...); // expected-warning {{declaration of a variadic function without a comma before '...' is deprecated}}
void g(auto x, ...);

void h(auto... x...); // expected-warning {{declaration of a variadic function without a comma before '...' is deprecated}} \
// expected-warning {{'...' in this location creates a C-style varargs function}} \
// expected-note {{preceding '...' declares a function parameter pack}} \
// expected-note {{insert ',' before '...' to silence this warning}}
void i(auto... x, ...);

template<class ...Ts>
void j(Ts... t...) {}; // expected-warning {{declaration of a variadic function without a comma before '...' is deprecated}} \
// expected-warning {{'...' in this location creates a C-style varargs function}} \
// expected-note {{preceding '...' declares a function parameter pack}} \
// expected-note {{insert ',' before '...' to silence this warning}}
template<class ...Ts>
void k(Ts... t, ...) {}

void l(int...); // expected-warning {{declaration of a variadic function without a comma before '...' is deprecated}}
void m(int, ...);

void n(int x...); // expected-warning {{declaration of a variadic function without a comma before '...' is deprecated}}
void o(int x, ...);

struct S {
void p(this S...) {} // expected-warning {{declaration of a variadic function without a comma before '...' is deprecated}}
};

template<class ...Ts>
void q(Ts......) {} // expected-warning {{declaration of a variadic function without a comma before '...' is deprecated}} \
// expected-warning {{'...' in this location creates a C-style varargs function}} \
// expected-note {{preceding '...' declares a function parameter pack}} \
// expected-note {{insert ',' before '...' to silence this warning}}

template<class T>
void r(T...) {} // expected-warning {{declaration of a variadic function without a comma before '...' is deprecated}}

auto type_specifier = (void (*)(int...)) nullptr; // expected-warning {{declaration of a variadic function without a comma before '...' is deprecated}}

auto lambda = [](int...) {}; // expected-warning {{declaration of a variadic function without a comma before '...' is deprecated}}

auto block = ^(int...){}; // expected-warning {{declaration of a variadic function without a comma before '...' is deprecated}}
5 changes: 5 additions & 0 deletions clang/test/Sema/constant_builtins_vector.cpp
View file
Original file line number Diff line number Diff line change
Expand Up @@ -817,3 +817,8 @@ static_assert(__builtin_elementwise_popcount(~0U) == 8 * sizeof(int));
static_assert(__builtin_elementwise_popcount(0L) == 0);
static_assert(__builtin_elementwise_popcount(0xF0F0L) == 8);
static_assert(__builtin_elementwise_popcount(~0LL) == 8 * sizeof(long long));

static_assert(__builtin_elementwise_bitreverse(0x12345678) == 0x1E6A2C48);
static_assert(__builtin_elementwise_bitreverse(0x0123456789ABCDEFULL) == 0xF7B3D591E6A2C480);
static_assert(__builtin_bit_cast(unsigned, __builtin_elementwise_bitreverse((vector4char){1, 2, 4, 8})) == (LITTLE_END ? 0x10204080 : 0x80402010));
static_assert(__builtin_bit_cast(unsigned long long, __builtin_elementwise_bitreverse((vector4short){1, 2, 4, 8})) == (LITTLE_END ? 0x1000200040008000 : 0x8000400020001000));
16 changes: 16 additions & 0 deletions clang/test/SemaCXX/lambda-capture-type-deduction.cpp
View file
Original file line number Diff line number Diff line change
Expand Up @@ -298,6 +298,22 @@ void __trans_tmp_1() {

}

namespace GH115931 {

struct Range {};

template <Range>
struct LengthPercentage {};

void reflectSum() {
Range resultR;
[&] (auto) -> LengthPercentage<resultR> {
return {};
}(0);
}

} // namespace GH115931

namespace GH47400 {

struct Foo {};
Expand Down
27 changes: 27 additions & 0 deletions clang/test/SemaHLSL/BuiltIns/resource_getpointer-errors.hlsl
View file
Original file line number Diff line number Diff line change
@@ -0,0 +1,27 @@
// RUN: %clang_cc1 -finclude-default-header -triple dxil-pc-shadermodel6.6-library %s -emit-llvm-only -disable-llvm-passes -verify

// RWBuffer<int>
using handle_t = __hlsl_resource_t
[[hlsl::resource_class(UAV)]] [[hlsl::contained_type(int)]];

void test_args(unsigned int x) {
// expected-error@+1 {{too few arguments to function call, expected 2, have 1}}
__builtin_hlsl_resource_getpointer(x);

// expected-error@+1 {{too many arguments to function call, expected 2, have 3}}
__builtin_hlsl_resource_getpointer(x, x, x);

// expected-error@+1 {{used type 'unsigned int' where __hlsl_resource_t is required}}
__builtin_hlsl_resource_getpointer(x, x);

handle_t res;

// expected-error@+1 {{passing 'const char *' to parameter of incompatible type 'unsigned int'}}
__builtin_hlsl_resource_getpointer(res, "1");

// no error
__builtin_hlsl_resource_getpointer(res, 0u);

// no error
__builtin_hlsl_resource_getpointer(res, x);
}
252 changes: 48 additions & 204 deletions clang/test/SemaOpenACC/combined-construct-auto_seq_independent-clauses.c
View file

Large diffs are not rendered by default.

219 changes: 219 additions & 0 deletions clang/test/SemaOpenACC/combined-construct-collapse-ast.cpp
View file
Original file line number Diff line number Diff line change
@@ -0,0 +1,219 @@

// RUN: %clang_cc1 %s -fopenacc -ast-dump | FileCheck %s

// Test this with PCH.
// RUN: %clang_cc1 %s -fopenacc -emit-pch -o %t %s
// RUN: %clang_cc1 %s -fopenacc -include-pch %t -ast-dump-all | FileCheck %s
#ifndef PCH_HELPER
#define PCH_HELPER

struct S {
constexpr S(){};
constexpr operator auto() {return 1;}
};

void NormalUses() {
// CHECK: FunctionDecl{{.*}}NormalUses
// CHECK-NEXT: CompoundStmt

#pragma acc parallel loop collapse(1)
for(int i = 0; i < 5; ++i);
// CHECK-NEXT: OpenACCCombinedConstruct{{.*}} parallel loop
// CHECK-NEXT: collapse clause
// CHECK-NEXT: ConstantExpr{{.*}}'int'
// CHECK-NEXT: value: Int 1
// CHECK-NEXT: IntegerLiteral{{.*}} 'int' 1
// CHECK-NEXT: ForStmt
// CHECK-NEXT: DeclStmt
// CHECK-NEXT: VarDecl{{.*}} i 'int'
// CHECK-NEXT: IntegerLiteral{{.*}} 'int' 0
// CHECK-NEXT: <<<NULL>>>
// CHECK-NEXT: BinaryOperator{{.*}}'<'
// CHECK-NEXT: ImplicitCastExpr
// CHECK-NEXT: DeclRefExpr{{.*}}'i' 'int'
// CHECK-NEXT: IntegerLiteral{{.*}} 'int' 5
// CHECK-NEXT: UnaryOperator{{.*}}++
// CHECK-NEXT: DeclRefExpr{{.*}}'i' 'int'
// CHECK-NEXT: NullStmt

#pragma acc serial loop collapse(force:S{})
for(int i = 0; i < 5; ++i);
// CHECK-NEXT: OpenACCCombinedConstruct{{.*}} serial loop
// CHECK-NEXT: collapse clause
// CHECK-NEXT: ConstantExpr{{.*}}'int'
// CHECK-NEXT: value: Int 1
// CHECK-NEXT: ImplicitCastExpr{{.*}} 'int' <UserDefinedConversion>
// CHECK-NEXT: CXXMemberCallExpr{{.*}} 'int'
// CHECK-NEXT: MemberExpr{{.*}} .operator auto
// CHECK-NEXT: MaterializeTemporaryExpr{{.*}}'S' lvalue
// CHECK-NEXT: CXXTemporaryObjectExpr{{.*}}'S' 'void ()' list
// CHECK-NEXT: ForStmt
// CHECK-NEXT: DeclStmt
// CHECK-NEXT: VarDecl{{.*}} i 'int'
// CHECK-NEXT: IntegerLiteral{{.*}} 'int' 0
// CHECK-NEXT: <<<NULL>>>
// CHECK-NEXT: BinaryOperator{{.*}}'<'
// CHECK-NEXT: ImplicitCastExpr
// CHECK-NEXT: DeclRefExpr{{.*}}'i' 'int'
// CHECK-NEXT: IntegerLiteral{{.*}} 'int' 5
// CHECK-NEXT: UnaryOperator{{.*}}++
// CHECK-NEXT: DeclRefExpr{{.*}}'i' 'int'
// CHECK-NEXT: NullStmt
}

template<typename T, unsigned Value>
void TemplUses() {
// CHECK: FunctionTemplateDecl{{.*}}TemplUses
// CHECK-NEXT: TemplateTypeParmDecl {{.*}} referenced typename depth 0 index 0 T
// CHECK-NEXT: NonTypeTemplateParmDecl{{.*}} referenced 'unsigned int' depth 0 index 1 Value
// CHECK-NEXT: FunctionDecl{{.*}} TemplUses 'void ()'
// CHECK-NEXT: CompoundStmt

#pragma acc parallel loop collapse(Value)
for(int i = 0; i < 5; ++i)
for(int j = 0; j < 5; ++j);
// CHECK-NEXT: OpenACCCombinedConstruct{{.*}} parallel loop
// CHECK-NEXT: collapse clause
// CHECK-NEXT: DeclRefExpr{{.*}} 'unsigned int' NonTypeTemplateParm{{.*}} 'Value'
// CHECK-NEXT: ForStmt
// CHECK-NEXT: DeclStmt
// CHECK-NEXT: VarDecl{{.*}} i 'int'
// CHECK-NEXT: IntegerLiteral{{.*}} 'int' 0
// CHECK-NEXT: <<<NULL>>>
// CHECK-NEXT: BinaryOperator{{.*}}'<'
// CHECK-NEXT: ImplicitCastExpr
// CHECK-NEXT: DeclRefExpr{{.*}}'i' 'int'
// CHECK-NEXT: IntegerLiteral{{.*}} 'int' 5
// CHECK-NEXT: UnaryOperator{{.*}}++
// CHECK-NEXT: DeclRefExpr{{.*}}'i' 'int'
// CHECK-NEXT: ForStmt
// CHECK-NEXT: DeclStmt
// CHECK-NEXT: VarDecl{{.*}} j 'int'
// CHECK-NEXT: IntegerLiteral{{.*}} 'int' 0
// CHECK-NEXT: <<<NULL>>>
// CHECK-NEXT: BinaryOperator{{.*}}'<'
// CHECK-NEXT: ImplicitCastExpr
// CHECK-NEXT: DeclRefExpr{{.*}}'j' 'int'
// CHECK-NEXT: IntegerLiteral{{.*}} 'int' 5
// CHECK-NEXT: UnaryOperator{{.*}}++
// CHECK-NEXT: DeclRefExpr{{.*}}'j' 'int'
// CHECK-NEXT: NullStmt

#pragma acc kernels loop collapse(force:T{} + S{})
for(int i = 0; i < 5; ++i)
for(int j = 0; j < 5; ++j);
// CHECK-NEXT: OpenACCCombinedConstruct{{.*}} kernels loop
// CHECK-NEXT: collapse clause
// CHECK-NEXT: BinaryOperator {{.*}}'+'
// CHECK-NEXT: CXXUnresolvedConstructExpr{{.*}}'T' 'T' list
// CHECK-NEXT: InitListExpr
// CHECK-NEXT: CXXTemporaryObjectExpr{{.*}}'S' 'void ()' list
// CHECK-NEXT: ForStmt
// CHECK-NEXT: DeclStmt
// CHECK-NEXT: VarDecl{{.*}} i 'int'
// CHECK-NEXT: IntegerLiteral{{.*}} 'int' 0
// CHECK-NEXT: <<<NULL>>>
// CHECK-NEXT: BinaryOperator{{.*}}'<'
// CHECK-NEXT: ImplicitCastExpr
// CHECK-NEXT: DeclRefExpr{{.*}}'i' 'int'
// CHECK-NEXT: IntegerLiteral{{.*}} 'int' 5
// CHECK-NEXT: UnaryOperator{{.*}}++
// CHECK-NEXT: DeclRefExpr{{.*}}'i' 'int'
// CHECK-NEXT: ForStmt
// CHECK-NEXT: DeclStmt
// CHECK-NEXT: VarDecl{{.*}} j 'int'
// CHECK-NEXT: IntegerLiteral{{.*}} 'int' 0
// CHECK-NEXT: <<<NULL>>>
// CHECK-NEXT: BinaryOperator{{.*}}'<'
// CHECK-NEXT: ImplicitCastExpr
// CHECK-NEXT: DeclRefExpr{{.*}}'j' 'int'
// CHECK-NEXT: IntegerLiteral{{.*}} 'int' 5
// CHECK-NEXT: UnaryOperator{{.*}}++
// CHECK-NEXT: DeclRefExpr{{.*}}'j' 'int'
// CHECK-NEXT: NullStmt

// Instantiation:
// CHECK-NEXT: FunctionDecl{{.*}} used TemplUses 'void ()' implicit_instantiation
// CHECK-NEXT: TemplateArgument type 'S'
// CHECK-NEXT: RecordType{{.*}} 'S'
// CHECK-NEXT: CXXRecord{{.*}} 'S'
// CHECK-NEXT: TemplateArgument integral '2U'
// CHECK-NEXT: CompoundStmt

// CHECK-NEXT: OpenACCCombinedConstruct{{.*}} parallel loop
// CHECK-NEXT: collapse clause
// CHECK-NEXT: ConstantExpr{{.*}}'unsigned int'
// CHECK-NEXT: value: Int 2
// CHECK-NEXT: SubstNonTypeTemplateParmExpr{{.*}} 'unsigned int'
// CHECK-NEXT: NonTypeTemplateParmDecl
// CHECK-NEXT: IntegerLiteral {{.*}} 'unsigned int' 2
// CHECK-NEXT: ForStmt
// CHECK-NEXT: DeclStmt
// CHECK-NEXT: VarDecl{{.*}} i 'int'
// CHECK-NEXT: IntegerLiteral{{.*}} 'int' 0
// CHECK-NEXT: <<<NULL>>>
// CHECK-NEXT: BinaryOperator{{.*}}'<'
// CHECK-NEXT: ImplicitCastExpr
// CHECK-NEXT: DeclRefExpr{{.*}}'i' 'int'
// CHECK-NEXT: IntegerLiteral{{.*}} 'int' 5
// CHECK-NEXT: UnaryOperator{{.*}}++
// CHECK-NEXT: DeclRefExpr{{.*}}'i' 'int'
// CHECK-NEXT: ForStmt
// CHECK-NEXT: DeclStmt
// CHECK-NEXT: VarDecl{{.*}} j 'int'
// CHECK-NEXT: IntegerLiteral{{.*}} 'int' 0
// CHECK-NEXT: <<<NULL>>>
// CHECK-NEXT: BinaryOperator{{.*}}'<'
// CHECK-NEXT: ImplicitCastExpr
// CHECK-NEXT: DeclRefExpr{{.*}}'j' 'int'
// CHECK-NEXT: IntegerLiteral{{.*}} 'int' 5
// CHECK-NEXT: UnaryOperator{{.*}}++
// CHECK-NEXT: DeclRefExpr{{.*}}'j' 'int'
// CHECK-NEXT: NullStmt

// CHECK-NEXT: OpenACCCombinedConstruct{{.*}} kernels loop
// CHECK-NEXT: collapse clause
// CHECK-NEXT: ConstantExpr{{.*}}'int'
// CHECK-NEXT: value: Int 2
// CHECK-NEXT: BinaryOperator {{.*}}'+'
// CHECK-NEXT: ImplicitCastExpr{{.*}} 'int' <UserDefinedConversion>
// CHECK-NEXT: CXXMemberCallExpr{{.*}} 'int'
// CHECK-NEXT: MemberExpr{{.*}} .operator auto
// CHECK-NEXT: MaterializeTemporaryExpr{{.*}}'S' lvalue
// CHECK-NEXT: CXXTemporaryObjectExpr{{.*}}'S' 'void ()' list
// CHECK-NEXT: ImplicitCastExpr{{.*}} 'int' <UserDefinedConversion>
// CHECK-NEXT: CXXMemberCallExpr{{.*}} 'int'
// CHECK-NEXT: MemberExpr{{.*}} .operator auto
// CHECK-NEXT: MaterializeTemporaryExpr{{.*}}'S' lvalue
// CHECK-NEXT: CXXTemporaryObjectExpr{{.*}}'S' 'void ()' list
// CHECK-NEXT: ForStmt
// CHECK-NEXT: DeclStmt
// CHECK-NEXT: VarDecl{{.*}} i 'int'
// CHECK-NEXT: IntegerLiteral{{.*}} 'int' 0
// CHECK-NEXT: <<<NULL>>>
// CHECK-NEXT: BinaryOperator{{.*}}'<'
// CHECK-NEXT: ImplicitCastExpr
// CHECK-NEXT: DeclRefExpr{{.*}}'i' 'int'
// CHECK-NEXT: IntegerLiteral{{.*}} 'int' 5
// CHECK-NEXT: UnaryOperator{{.*}}++
// CHECK-NEXT: DeclRefExpr{{.*}}'i' 'int'
// CHECK-NEXT: ForStmt
// CHECK-NEXT: DeclStmt
// CHECK-NEXT: VarDecl{{.*}} j 'int'
// CHECK-NEXT: IntegerLiteral{{.*}} 'int' 0
// CHECK-NEXT: <<<NULL>>>
// CHECK-NEXT: BinaryOperator{{.*}}'<'
// CHECK-NEXT: ImplicitCastExpr
// CHECK-NEXT: DeclRefExpr{{.*}}'j' 'int'
// CHECK-NEXT: IntegerLiteral{{.*}} 'int' 5
// CHECK-NEXT: UnaryOperator{{.*}}++
// CHECK-NEXT: DeclRefExpr{{.*}}'j' 'int'
// CHECK-NEXT: NullStmt

}

void Inst() {
TemplUses<S, 2>();
}

#endif // PCH_HELPER
390 changes: 390 additions & 0 deletions clang/test/SemaOpenACC/combined-construct-collapse-clause.cpp
View file
Original file line number Diff line number Diff line change
@@ -0,0 +1,390 @@
// RUN: %clang_cc1 %s -fopenacc -verify

template<unsigned Val>
void depth_too_high_templ() {
// expected-error@+2{{'collapse' clause specifies a loop count greater than the number of available loops}}
// expected-note@+1{{active 'collapse' clause defined here}}
#pragma acc parallel loop collapse(Val)
for(unsigned i = 0; i < 5; ++i)
for(unsigned j = 0; j < 5; ++j);
}
constexpr int three() { return 3; }
struct ConvertsThree{
constexpr ConvertsThree(){};

constexpr operator int(){ return 3; }
};

void depth_too_high() {
depth_too_high_templ<3>(); // expected-note{{in instantiation of function template specialization}}

// expected-error@+2{{'collapse' clause specifies a loop count greater than the number of available loops}}
// expected-note@+1{{active 'collapse' clause defined here}}
#pragma acc serial loop collapse(3)
for(unsigned i = 0; i < 5; ++i)
for(unsigned j = 0; j < 5; ++j);

// expected-error@+2{{'collapse' clause specifies a loop count greater than the number of available loops}}
// expected-note@+1{{active 'collapse' clause defined here}}
#pragma acc kernels loop collapse(three())
for(unsigned i = 0; i < 5; ++i)
for(unsigned j = 0; j < 5; ++j);

// expected-error@+2{{'collapse' clause specifies a loop count greater than the number of available loops}}
// expected-note@+1{{active 'collapse' clause defined here}}
#pragma acc parallel loop collapse(ConvertsThree{})
for(unsigned i = 0; i < 5; ++i)
for(unsigned j = 0; j < 5; ++j);
}

template<typename T, unsigned Three>
void not_single_loop_templ() {
T Arr[5];
// expected-error@+2{{'collapse' clause specifies a loop count greater than the number of available loops}}
// expected-note@+1 2{{active 'collapse' clause defined here}}
#pragma acc parallel loop collapse(3)
for(auto x : Arr) {
for(auto y : Arr){
do{}while(true); // expected-error{{do loop cannot appear in intervening code of a 'parallel loop' with a 'collapse' clause}}
}
}

// expected-error@+2{{'collapse' clause specifies a loop count greater than the number of available loops}}
// expected-note@+1 2{{active 'collapse' clause defined here}}
#pragma acc serial loop collapse(Three)
for(unsigned i = 0; i < 5; ++i) {
for(unsigned j = 0; j < 5; ++j) {
do{}while(true); // expected-error{{do loop cannot appear in intervening code of a 'serial loop' with a 'collapse' clause}}
}
}

#pragma acc kernels loop collapse(Three)
for(unsigned i = 0; i < 5; ++i) {
for(unsigned j = 0; j < 5; ++j) {
for(unsigned k = 0; k < 5;++k) {
do{}while(true);
}
}
}
// expected-error@+2{{'collapse' clause specifies a loop count greater than the number of available loops}}
// expected-note@+1 2{{active 'collapse' clause defined here}}
#pragma acc parallel loop collapse(Three)
for(auto x : Arr) {
for(auto y: Arr) {
do{}while(true); // expected-error{{do loop cannot appear in intervening code of a 'parallel loop' with a 'collapse' clause}}
}
}

#pragma acc serial loop collapse(Three)
for(auto x : Arr) {
for(auto y: Arr) {
for(auto z: Arr) {
do{}while(true);
}
}
}
}

void not_single_loop() {
not_single_loop_templ<int, 3>(); // expected-note{{in instantiation of function template}}

// expected-note@+1{{active 'collapse' clause defined here}}
#pragma acc kernels loop collapse(3)
for(unsigned i = 0; i < 5; ++i) {
for(unsigned j = 0; j < 5; ++j) {
for(unsigned k = 0; k < 5;++k);
}
while(true); // expected-error{{while loop cannot appear in intervening code of a 'kernels loop' with a 'collapse' clause}}
}

// expected-note@+1{{active 'collapse' clause defined here}}
#pragma acc parallel loop collapse(3)
for(unsigned i = 0; i < 5; ++i) {
for(unsigned j = 0; j < 5; ++j) {
for(unsigned k = 0; k < 5;++k);
}
do{}while(true); // expected-error{{do loop cannot appear in intervening code of a 'parallel loop' with a 'collapse' clause}}
}

// expected-error@+2{{'collapse' clause specifies a loop count greater than the number of available loops}}
// expected-note@+1 2{{active 'collapse' clause defined here}}
#pragma acc serial loop collapse(3)
for(unsigned i = 0; i < 5; ++i) {
for(unsigned j = 0; j < 5; ++j) {
while(true); // expected-error{{while loop cannot appear in intervening code of a 'serial loop' with a 'collapse' clause}}
}
}
// expected-error@+2{{'collapse' clause specifies a loop count greater than the number of available loops}}
// expected-note@+1 2{{active 'collapse' clause defined here}}
#pragma acc kernels loop collapse(3)
for(unsigned i = 0; i < 5; ++i) {
for(unsigned j = 0; j < 5; ++j) {
do{}while(true); // expected-error{{do loop cannot appear in intervening code of a 'kernels loop' with a 'collapse' clause}}
}
}

#pragma acc parallel loop collapse(2)
for(unsigned i = 0; i < 5; ++i) {
for(unsigned j = 0; j < 5; ++j) {
do{}while(true);
}
}
#pragma acc serial loop collapse(2)
for(unsigned i = 0; i < 5; ++i) {
for(unsigned j = 0; j < 5; ++j) {
while(true);
}
}

int Arr[5];
// expected-error@+2{{'collapse' clause specifies a loop count greater than the number of available loops}}
// expected-note@+1 2{{active 'collapse' clause defined here}}
#pragma acc kernels loop collapse(3)
for(auto x : Arr) {
for(auto y : Arr){
do{}while(true); // expected-error{{do loop cannot appear in intervening code of a 'kernels loop' with a 'collapse' clause}}
}
}

// expected-note@+1 {{active 'collapse' clause defined here}}
#pragma acc parallel loop collapse(3)
for(unsigned i = 0; i < 5; ++i) {
for(unsigned j = 0; j < 5; ++j) {
for(unsigned k = 0; k < 5;++k);
}
// expected-error@+1{{more than one for-loop in a loop associated with OpenACC 'parallel loop' construct with a 'collapse' clause}}
for(unsigned k = 0; k < 5;++k);
}

// expected-note@+1 {{active 'collapse' clause defined here}}
#pragma acc serial loop collapse(3)
for(unsigned i = 0; i < 5; ++i) {
for(unsigned j = 0; j < 5; ++j) {
for(unsigned k = 0; k < 5;++k);
// expected-error@+1{{more than one for-loop in a loop associated with OpenACC 'serial loop' construct with a 'collapse' clause}}
for(unsigned k = 0; k < 5;++k);
}
}

for(unsigned k = 0; k < 5;++k);
#pragma acc kernels loop collapse(3)
for(unsigned i = 0; i < 5; ++i) {
for(unsigned j = 0; j < 5; ++j) {
for(unsigned k = 0; k < 5;++k);
}
}
}

template<unsigned Two, unsigned Three>
void no_other_directives() {
#pragma acc parallel loop collapse(Two)
for(unsigned i = 0; i < 5; ++i) {
for(unsigned j = 0; j < 5; ++j) {// last loop associated with the top level.
// expected-error@+1{{'collapse' clause specifies a loop count greater than the number of available loops}}
#pragma acc serial loop collapse(Three) // expected-note 2{{active 'collapse' clause defined here}}
for(unsigned k = 0; k < 6;++k) {
for(unsigned l = 0; l < 5; ++l) {
// expected-error@+1{{OpenACC 'serial' construct cannot appear in intervening code of a 'serial loop' with a 'collapse' clause}}
#pragma acc serial
;
}
}
}
}
#pragma acc kernels loop collapse(Two)// expected-note{{active 'collapse' clause defined here}}
for(unsigned i = 0; i < 5; ++i) {
for(unsigned j = 0; j < 5; ++j) {// last loop associated with the top level.
#pragma acc parallel loop collapse(Three)
for(unsigned k = 0; k < 6;++k) {
for(unsigned l = 0; l < 5; ++l) {
for(unsigned m = 0; m < 5; ++m);
}
}
}
// expected-error@+1{{OpenACC 'serial' construct cannot appear in intervening code of a 'kernels loop' with a 'collapse' clause}}
#pragma acc serial
;
}
}

void no_other_directives() {
no_other_directives<2,3>(); // expected-note{{in instantiation of function template specialization}}

// Ok, not inside the intervening list
#pragma acc serial loop collapse(2)
for(unsigned i = 0; i < 5; ++i) {
for(unsigned j = 0; j < 5; ++j) {
#pragma acc data // expected-warning{{OpenACC construct 'data' not yet implemented}}
}
}
// expected-note@+1{{active 'collapse' clause defined here}}
#pragma acc kernels loop collapse(2)
for(unsigned i = 0; i < 5; ++i) {
// expected-error@+1{{OpenACC 'data' construct cannot appear in intervening code of a 'kernels loop' with a 'collapse' clause}}
#pragma acc data // expected-warning{{OpenACC construct 'data' not yet implemented}}
for(unsigned j = 0; j < 5; ++j) {
}
}
}

void call();

template<unsigned Two>
void intervening_without_force_templ() {
// expected-note@+1{{active 'collapse' clause defined here}}
#pragma acc parallel loop collapse(2)
for(unsigned i = 0; i < 5; ++i) {
// expected-error@+1{{inner loops must be tightly nested inside a 'collapse' clause on a 'parallel loop' construct}}
call();
for(unsigned j = 0; j < 5; ++j);
}

// expected-note@+1{{active 'collapse' clause defined here}}
#pragma acc serial loop collapse(Two)
for(unsigned i = 0; i < 5; ++i) {
// expected-error@+1{{inner loops must be tightly nested inside a 'collapse' clause on a 'serial loop' construct}}
call();
for(unsigned j = 0; j < 5; ++j);
}

// expected-note@+1{{active 'collapse' clause defined here}}
#pragma acc kernels loop collapse(2)
for(unsigned i = 0; i < 5; ++i) {
for(unsigned j = 0; j < 5; ++j);
// expected-error@+1{{inner loops must be tightly nested inside a 'collapse' clause on a 'kernels loop' construct}}
call();
}

#pragma acc parallel loop collapse(force:2)
for(unsigned i = 0; i < 5; ++i) {
call();
for(unsigned j = 0; j < 5; ++j);
}

#pragma acc parallel loop collapse(force:Two)
for(unsigned i = 0; i < 5; ++i) {
call();
for(unsigned j = 0; j < 5; ++j);
}


#pragma acc parallel loop collapse(force:2)
for(unsigned i = 0; i < 5; ++i) {
for(unsigned j = 0; j < 5; ++j);
call();
}

#pragma acc parallel loop collapse(force:Two)
for(unsigned i = 0; i < 5; ++i) {
for(unsigned j = 0; j < 5; ++j);
call();
}

#pragma acc parallel loop collapse(Two)
for(unsigned i = 0; i < 5; ++i) {
for(unsigned j = 0; j < 5; ++j) {
call();
}
}

#pragma acc parallel loop collapse(Two)
for(unsigned i = 0; i < 5; ++i) {
{
{
for(unsigned j = 0; j < 5; ++j) {
call();
}
}
}
}

#pragma acc parallel loop collapse(force:Two)
for(unsigned i = 0; i < 5; ++i) {
for(unsigned j = 0; j < 5; ++j) {
call();
}
}

// expected-note@+1{{active 'collapse' clause defined here}}
#pragma acc parallel loop collapse(Two)
for(unsigned i = 0; i < 5; ++i) {
for(unsigned j = 0; j < 5; ++j);
// expected-error@+1{{inner loops must be tightly nested inside a 'collapse' clause on a 'parallel loop' construct}}
call();
}

#pragma acc parallel loop collapse(2)
// expected-error@+2{{OpenACC 'parallel loop' construct must have a terminating condition}}
// expected-note@-2{{'parallel loop' construct is here}}
for(int i = 0;;++i)
// expected-error@+2{{OpenACC 'parallel loop' construct must have a terminating condition}}
// expected-note@-5{{'parallel loop' construct is here}}
for(int j = 0;;++j)
for(;;);
}

void intervening_without_force() {
intervening_without_force_templ<2>(); // expected-note{{in instantiation of function template specialization}}
// expected-note@+1{{active 'collapse' clause defined here}}
#pragma acc parallel loop collapse(2)
for(unsigned i = 0; i < 5; ++i) {
// expected-error@+1{{inner loops must be tightly nested inside a 'collapse' clause on a 'parallel loop' construct}}
call();
for(unsigned j = 0; j < 5; ++j);
}

// expected-note@+1{{active 'collapse' clause defined here}}
#pragma acc parallel loop collapse(2)
for(unsigned i = 0; i < 5; ++i) {
for(unsigned j = 0; j < 5; ++j);
// expected-error@+1{{inner loops must be tightly nested inside a 'collapse' clause on a 'parallel loop' construct}}
call();
}

// The below two are fine, as they use the 'force' tag.
#pragma acc parallel loop collapse(force:2)
for(unsigned i = 0; i < 5; ++i) {
call();
for(unsigned j = 0; j < 5; ++j);
}

#pragma acc parallel loop collapse(force:2)
for(unsigned i = 0; i < 5; ++i) {
for(unsigned j = 0; j < 5; ++j);
call();
}

#pragma acc parallel loop collapse(2)
for(unsigned i = 0; i < 5; ++i) {
for(unsigned j = 0; j < 5; ++j) {
call();
}
}
#pragma acc parallel loop collapse(2)
for(unsigned i = 0; i < 5; ++i) {
{
{
for(unsigned j = 0; j < 5; ++j) {
call();
}
}
}
}

#pragma acc parallel loop collapse(force:2)
for(unsigned i = 0; i < 5; ++i) {
for(unsigned j = 0; j < 5; ++j) {
call();
}
}

#pragma acc parallel loop collapse(2)
// expected-error@+2{{OpenACC 'parallel loop' construct must have a terminating condition}}
// expected-note@-2{{'parallel loop' construct is here}}
for(int i = 0;;++i)
// expected-error@+2{{OpenACC 'parallel loop' construct must have a terminating condition}}
// expected-note@-5{{'parallel loop' construct is here}}
for(int j = 0;;++j)
for(;;);
}

99 changes: 99 additions & 0 deletions clang/test/SemaOpenACC/combined-construct-copy-ast.cpp
View file
Original file line number Diff line number Diff line change
@@ -0,0 +1,99 @@
// RUN: %clang_cc1 %s -fopenacc -Wno-openacc-deprecated-clause-alias -ast-dump | FileCheck %s

// Test this with PCH.
// RUN: %clang_cc1 %s -fopenacc -Wno-openacc-deprecated-clause-alias -emit-pch -o %t %s
// RUN: %clang_cc1 %s -fopenacc -Wno-openacc-deprecated-clause-alias -include-pch %t -ast-dump-all | FileCheck %s

#ifndef PCH_HELPER
#define PCH_HELPER

int Global;
short GlobalArray[5];
void NormalUses(float *PointerParam) {
// CHECK: FunctionDecl{{.*}}NormalUses
// CHECK: ParmVarDecl
// CHECK-NEXT: CompoundStmt

#pragma acc parallel loop copy(GlobalArray) pcopy(PointerParam[Global]) present_or_copy(Global)
for (unsigned i = 0; i < 5; ++i);
// CHECK-NEXT: OpenACCCombinedConstruct{{.*}} parallel loop
// CHECK-NEXT: copy clause
// CHECK-NEXT: DeclRefExpr{{.*}}'short[5]' lvalue Var{{.*}}'GlobalArray' 'short[5]'
// CHECK-NEXT: pcopy clause
// CHECK-NEXT: ArraySubscriptExpr{{.*}}'float' lvalue
// CHECK-NEXT: ImplicitCastExpr{{.*}} 'float *' <LValueToRValue>
// CHECK-NEXT: DeclRefExpr{{.*}}'float *' lvalue ParmVar{{.*}}'PointerParam' 'float *'
// CHECK-NEXT: ImplicitCastExpr{{.*}} 'int' <LValueToRValue>
// CHECK-NEXT: DeclRefExpr{{.*}}'int' lvalue Var{{.*}}'Global' 'int'
// CHECK-NEXT: present_or_copy clause
// CHECK-NEXT: DeclRefExpr{{.*}}'int' lvalue Var{{.*}}'Global' 'int'
// CHECK-NEXT: ForStmt
// CHECK:NullStmt
}

template<auto &NTTP, typename T, typename U>
void TemplUses(T t, U u) {
// CHECK-NEXT: FunctionTemplateDecl
// CHECK-NEXT: NonTypeTemplateParmDecl {{.*}}referenced 'auto &' depth 0 index 0 NTTP
// CHECK-NEXT: TemplateTypeParmDecl{{.*}}typename depth 0 index 1 T
// CHECK-NEXT: TemplateTypeParmDecl{{.*}}typename depth 0 index 2 U
// CHECK-NEXT: FunctionDecl{{.*}} TemplUses 'void (T, U)'
// CHECK-NEXT: ParmVarDecl{{.*}} referenced t 'T'
// CHECK-NEXT: ParmVarDecl{{.*}} referenced u 'U'
// CHECK-NEXT: CompoundStmt

#pragma acc parallel loop copy(t) pcopy(NTTP, u) present_or_copy(u[0:t])
for (unsigned i = 0; i < 5; ++i);
// CHECK-NEXT: OpenACCCombinedConstruct{{.*}} parallel loop
// CHECK-NEXT: copy clause
// CHECK-NEXT: DeclRefExpr{{.*}}'T' lvalue ParmVar{{.*}} 't' 'T'
// CHECK-NEXT: pcopy clause
// CHECK-NEXT: DeclRefExpr{{.*}}'auto' lvalue NonTypeTemplateParm{{.*}} 'NTTP' 'auto &'
// CHECK-NEXT: DeclRefExpr{{.*}}'U' lvalue ParmVar{{.*}} 'u' 'U'
// CHECK-NEXT: present_or_copy clause
// CHECK-NEXT: ArraySectionExpr
// CHECK-NEXT: DeclRefExpr{{.*}}'U' lvalue ParmVar{{.*}} 'u' 'U'
// CHECK-NEXT: IntegerLiteral{{.*}} 'int' 0
// CHECK-NEXT: DeclRefExpr{{.*}}'T' lvalue ParmVar{{.*}} 't' 'T'
// CHECK-NEXT: ForStmt
// CHECK:NullStmt

// Check the instantiated versions of the above.
// CHECK-NEXT: FunctionDecl{{.*}} used TemplUses 'void (int, int *)' implicit_instantiation
// CHECK-NEXT: TemplateArgument decl
// CHECK-NEXT: Var{{.*}} 'CEVar' 'const unsigned int'
// CHECK-NEXT: TemplateArgument type 'int'
// CHECK-NEXT: BuiltinType{{.*}} 'int'
// CHECK-NEXT: TemplateArgument type 'int *'
// CHECK-NEXT: PointerType{{.*}} 'int *'
// CHECK-NEXT: BuiltinType{{.*}} 'int'
// CHECK-NEXT: ParmVarDecl{{.*}} used t 'int'
// CHECK-NEXT: ParmVarDecl{{.*}} used u 'int *'
// CHECK-NEXT: CompoundStmt

// #pragma acc parallel copy(t) pcopy(NTTP, u) present_or_copy(u[0:t])
// CHECK-NEXT: OpenACCCombinedConstruct{{.*}} parallel loop
// CHECK-NEXT: copy clause
// CHECK-NEXT: DeclRefExpr{{.*}}'int' lvalue ParmVar{{.*}} 't' 'int'
// CHECK-NEXT: pcopy clause
// CHECK-NEXT: SubstNonTypeTemplateParmExpr{{.*}}'const unsigned int' lvalue
// CHECK-NEXT: NonTypeTemplateParmDecl{{.*}} referenced 'auto &' depth 0 index 0 NTTP
// CHECK-NEXT: DeclRefExpr{{.*}}'const unsigned int' lvalue Var{{.*}} 'CEVar' 'const unsigned int'
// CHECK-NEXT: DeclRefExpr{{.*}}'int *' lvalue ParmVar{{.*}} 'u' 'int *'
// CHECK-NEXT: present_or_copy clause
// CHECK-NEXT: ArraySectionExpr
// CHECK-NEXT: ImplicitCastExpr{{.*}} 'int *' <LValueToRValue>
// CHECK-NEXT: DeclRefExpr{{.*}}'int *' lvalue ParmVar{{.*}} 'u' 'int *'
// CHECK-NEXT: IntegerLiteral{{.*}} 'int' 0
// CHECK-NEXT: ImplicitCastExpr{{.*}} 'int' <LValueToRValue>
// CHECK-NEXT: DeclRefExpr{{.*}}'int' lvalue ParmVar{{.*}} 't' 'int'
// CHECK-NEXT: ForStmt
// CHECK:NullStmt
}

void Inst() {
static constexpr unsigned CEVar = 1;
int i;
TemplUses<CEVar>(i, &i);
}
#endif
72 changes: 72 additions & 0 deletions clang/test/SemaOpenACC/combined-construct-copy-clause.c
View file
Original file line number Diff line number Diff line change
@@ -0,0 +1,72 @@
// RUN: %clang_cc1 %s -fopenacc -verify

typedef struct IsComplete {
struct S { int A; } CompositeMember;
int ScalarMember;
float ArrayMember[5];
void *PointerMember;
} Complete;
void uses(int IntParam, short *PointerParam, float ArrayParam[5], Complete CompositeParam) {
int LocalInt;
short *LocalPointer;
float LocalArray[5];
Complete LocalComposite;
// Check Appertainment:
#pragma acc parallel loop copy(LocalInt)
for(int i = 0; i < 5; ++i);
#pragma acc serial loop copy(LocalInt)
for(int i = 0; i < 5; ++i);
#pragma acc kernels loop copy(LocalInt)
for(int i = 0; i < 5; ++i);

// expected-warning@+1{{OpenACC clause name 'pcopy' is a deprecated clause name and is now an alias for 'copy'}}
#pragma acc parallel loop pcopy(LocalInt)
for(int i = 0; i < 5; ++i);

// expected-warning@+1{{OpenACC clause name 'present_or_copy' is a deprecated clause name and is now an alias for 'copy'}}
#pragma acc parallel loop present_or_copy(LocalInt)
for(int i = 0; i < 5; ++i);

// Valid cases:
#pragma acc parallel loop copy(LocalInt, LocalPointer, LocalArray)
for(int i = 0; i < 5; ++i);
#pragma acc parallel loop copy(LocalArray[2:1])
for(int i = 0; i < 5; ++i);

#pragma acc parallel loop copy(LocalComposite.ScalarMember, LocalComposite.ScalarMember)
for(int i = 0; i < 5; ++i);

// expected-error@+1{{OpenACC variable is not a valid variable name, sub-array, array element, member of a composite variable, or composite variable member}}
#pragma acc parallel loop copy(1 + IntParam)
for(int i = 0; i < 5; ++i);

// expected-error@+1{{OpenACC variable is not a valid variable name, sub-array, array element, member of a composite variable, or composite variable member}}
#pragma acc parallel loop copy(+IntParam)
for(int i = 0; i < 5; ++i);

// expected-error@+1{{OpenACC sub-array length is unspecified and cannot be inferred because the subscripted value is not an array}}
#pragma acc parallel loop copy(PointerParam[2:])
for(int i = 0; i < 5; ++i);

// expected-error@+1{{OpenACC sub-array specified range [2:5] would be out of the range of the subscripted array size of 5}}
#pragma acc parallel loop copy(ArrayParam[2:5])
for(int i = 0; i < 5; ++i);

// expected-error@+2{{OpenACC sub-array specified range [2:5] would be out of the range of the subscripted array size of 5}}
// expected-error@+1{{OpenACC variable is not a valid variable name, sub-array, array element, member of a composite variable, or composite variable member}}
#pragma acc parallel loop copy((float*)ArrayParam[2:5])
for(int i = 0; i < 5; ++i);
// expected-error@+1{{OpenACC variable is not a valid variable name, sub-array, array element, member of a composite variable, or composite variable member}}
#pragma acc parallel loop copy((float)ArrayParam[2])
for(int i = 0; i < 5; ++i);

// expected-error@+1{{OpenACC 'copy' clause is not valid on 'loop' directive}}
#pragma acc loop copy(LocalInt)
for(int i = 5; i < 10;++i);
// expected-error@+1{{OpenACC 'pcopy' clause is not valid on 'loop' directive}}
#pragma acc loop pcopy(LocalInt)
for(int i = 5; i < 10;++i);
// expected-error@+1{{OpenACC 'present_or_copy' clause is not valid on 'loop' directive}}
#pragma acc loop present_or_copy(LocalInt)
for(int i = 5; i < 10;++i);
}
112 changes: 112 additions & 0 deletions clang/test/SemaOpenACC/combined-construct-copy-clause.cpp
View file
Original file line number Diff line number Diff line change
@@ -0,0 +1,112 @@
// RUN: %clang_cc1 %s -fopenacc -verify

enum SomeE{};
typedef struct IsComplete {
struct S { int A; } CompositeMember;
int ScalarMember;
float ArrayMember[5];
SomeE EnumMember;
char *PointerMember;
} Complete;

void uses(int IntParam, char *PointerParam, float ArrayParam[5], Complete CompositeParam, int &IntParamRef) {
int LocalInt;
char *LocalPointer;
float LocalArray[5];
// Check Appertainment:
#pragma acc parallel loop copy(LocalInt)
for(int i = 0; i < 5; ++i);
#pragma acc serial loop copy(LocalInt)
for(int i = 0; i < 5; ++i);
#pragma acc kernels loop copy(LocalInt)
for(int i = 0; i < 5; ++i);

// Valid cases:
#pragma acc parallel loop copy(LocalInt, LocalPointer, LocalArray)
for(int i = 0; i < 5; ++i);
#pragma acc parallel loop copy(LocalArray[2:1])
for(int i = 0; i < 5; ++i);

Complete LocalComposite2;
#pragma acc parallel loop copy(LocalComposite2.ScalarMember, LocalComposite2.ScalarMember)
for(int i = 0; i < 5; ++i);

// expected-error@+1{{OpenACC variable is not a valid variable name, sub-array, array element, member of a composite variable, or composite variable member}}
#pragma acc parallel loop copy(1 + IntParam)
for(int i = 0; i < 5; ++i);

// expected-error@+1{{OpenACC variable is not a valid variable name, sub-array, array element, member of a composite variable, or composite variable member}}
#pragma acc parallel loop copy(+IntParam)
for(int i = 0; i < 5; ++i);

// expected-error@+1{{OpenACC sub-array length is unspecified and cannot be inferred because the subscripted value is not an array}}
#pragma acc parallel loop copy(PointerParam[2:])
for(int i = 0; i < 5; ++i);

// expected-error@+1{{OpenACC sub-array specified range [2:5] would be out of the range of the subscripted array size of 5}}
#pragma acc parallel loop copy(ArrayParam[2:5])
for(int i = 0; i < 5; ++i);

// expected-error@+2{{OpenACC sub-array specified range [2:5] would be out of the range of the subscripted array size of 5}}
// expected-error@+1{{OpenACC variable is not a valid variable name, sub-array, array element, member of a composite variable, or composite variable member}}
#pragma acc parallel loop copy((float*)ArrayParam[2:5])
for(int i = 0; i < 5; ++i);
// expected-error@+1{{OpenACC variable is not a valid variable name, sub-array, array element, member of a composite variable, or composite variable member}}
#pragma acc parallel loop copy((float)ArrayParam[2])
for(int i = 0; i < 5; ++i);
}

template<typename T, unsigned I, typename V>
void TemplUses(T t, T (&arrayT)[I], V TemplComp) {
// expected-error@+1{{OpenACC variable is not a valid variable name, sub-array, array element, member of a composite variable, or composite variable member}}
#pragma acc parallel loop copy(+t)
for(int i = 0; i < 5; ++i);

// NTTP's are only valid if it is a reference to something.
// expected-error@+2{{OpenACC variable is not a valid variable name, sub-array, array element, member of a composite variable, or composite variable member}}
// expected-note@#TEMPL_USES_INST{{in instantiation of}}
#pragma acc parallel loop copy(I)
for(int i = 0; i < 5; ++i);

// expected-error@+1{{OpenACC variable is not a valid variable name, sub-array, array element, member of a composite variable, or composite variable member}}
#pragma acc parallel loop copy(t, I)
for(int i = 0; i < 5; ++i);

#pragma acc parallel loop copy(arrayT)
for(int i = 0; i < 5; ++i);

#pragma acc parallel loop copy(TemplComp)
for(int i = 0; i < 5; ++i);

#pragma acc parallel loop copy(TemplComp.PointerMember[5])
for(int i = 0; i < 5; ++i);
int *Pointer;
#pragma acc parallel loop copy(Pointer[:I])
for(int i = 0; i < 5; ++i);
#pragma acc parallel loop copy(Pointer[:t])
for(int i = 0; i < 5; ++i);
// expected-error@+1{{OpenACC sub-array length is unspecified and cannot be inferred because the subscripted value is not an array}}
#pragma acc parallel loop copy(Pointer[1:])
for(int i = 0; i < 5; ++i);
}

template<unsigned I, auto &NTTP_REF>
void NTTP() {
// NTTP's are only valid if it is a reference to something.
// expected-error@+2{{OpenACC variable is not a valid variable name, sub-array, array element, member of a composite variable, or composite variable member}}
// expected-note@#NTTP_INST{{in instantiation of}}
#pragma acc parallel loop copy(I)
for(int i = 0; i < 5; ++i);

#pragma acc parallel loop copy(NTTP_REF)
for(int i = 0; i < 5; ++i);
}

void Inst() {
static constexpr int NTTP_REFed = 1;
int i;
int Arr[5];
Complete C;
TemplUses(i, Arr, C); // #TEMPL_USES_INST
NTTP<5, NTTP_REFed>(); // #NTTP_INST
}
Loading