// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --function-signature --include-generated-funcs --replace-value-regex "__omp_offloading_[0-9a-z]+_[0-9a-z]+" "reduction_size[.].+[.]" "pl_cond[.].+[.|,]" --prefix-filecheck-ir-name _
// RUN: %clang_cc1 -no-opaque-pointers -verify -triple powerpc64le-unknown-linux-gnu -fopenmp -x c -emit-llvm %s -o - | FileCheck %s --check-prefix=CHECK1
// RUN: %clang_cc1 -no-opaque-pointers -verify -triple powerpc64le-unknown-linux-gnu -fopenmp-simd -x c -emit-llvm %s -o - | FileCheck %s --implicit-check-not="{{__kmpc|__tgt}}"
int a;
void
void
// CHECK1-LABEL: define {{[^@]+}}@foo
// CHECK1-SAME: () #[[ATTR0:[0-9]+]] {
// CHECK1-NEXT: entry:
// CHECK1-NEXT: [[B:%.*]] = alloca i32*, align 8
// CHECK1-NEXT: [[C:%.*]] = alloca i32***, align 8
// CHECK1-NEXT: [[DOTTHREADID_TEMP_:%.*]] = alloca i32, align 4
// CHECK1-NEXT: [[DOTBOUND_ZERO_ADDR:%.*]] = alloca i32, align 4
// CHECK1-NEXT: [[TMP0:%.*]] = call i32 @__kmpc_global_thread_num(%struct.ident_t* @[[GLOB1:[0-9]+]])
// CHECK1-NEXT: [[TMP1:%.*]] = load i32, i32* @a, align 4
// CHECK1-NEXT: [[TMP2:%.*]] = zext i32 [[TMP1]] to i64
// CHECK1-NEXT: [[TMP3:%.*]] = load i32, i32* @a, align 4
// CHECK1-NEXT: [[TMP4:%.*]] = zext i32 [[TMP3]] to i64
// CHECK1-NEXT: call void @__kmpc_serialized_parallel(%struct.ident_t* @[[GLOB1]], i32 [[TMP0]])
// CHECK1-NEXT: store i32 [[TMP0]], i32* [[DOTTHREADID_TEMP_]], align 4
// CHECK1-NEXT: store i32 0, i32* [[DOTBOUND_ZERO_ADDR]], align 4
// CHECK1-NEXT: call void @.omp_outlined.(i32* [[DOTTHREADID_TEMP_]], i32* [[DOTBOUND_ZERO_ADDR]], i64 [[TMP2]], i32** [[B]], i64 [[TMP4]], i32**** [[C]]) #[[ATTR2:[0-9]+]]
// CHECK1-NEXT: call void @__kmpc_end_serialized_parallel(%struct.ident_t* @[[GLOB1]], i32 [[TMP0]])
// CHECK1-NEXT: ret void
//
//
// CHECK1-LABEL: define {{[^@]+}}@.omp_outlined.
// CHECK1-SAME: (i32* noalias noundef [[DOTGLOBAL_TID_:%.*]], i32* noalias noundef [[DOTBOUND_TID_:%.*]], i64 noundef [[VLA:%.*]], i32** noundef nonnull align 8 dereferenceable(8) [[B:%.*]], i64 noundef [[VLA1:%.*]], i32**** noundef nonnull align 8 dereferenceable(8) [[C:%.*]]) #[[ATTR1:[0-9]+]] {
// CHECK1-NEXT: entry:
// CHECK1-NEXT: [[DOTGLOBAL_TID__ADDR:%.*]] = alloca i32*, align 8
// CHECK1-NEXT: [[DOTBOUND_TID__ADDR:%.*]] = alloca i32*, align 8
// CHECK1-NEXT: [[VLA_ADDR:%.*]] = alloca i64, align 8
// CHECK1-NEXT: [[B_ADDR:%.*]] = alloca i32**, align 8
// CHECK1-NEXT: [[VLA_ADDR2:%.*]] = alloca i64, align 8
// CHECK1-NEXT: [[C_ADDR:%.*]] = alloca i32****, align 8
// CHECK1-NEXT: store i32* [[DOTGLOBAL_TID_]], i32** [[DOTGLOBAL_TID__ADDR]], align 8
// CHECK1-NEXT: store i32* [[DOTBOUND_TID_]], i32** [[DOTBOUND_TID__ADDR]], align 8
// CHECK1-NEXT: store i64 [[VLA]], i64* [[VLA_ADDR]], align 8
// CHECK1-NEXT: store i32** [[B]], i32*** [[B_ADDR]], align 8
// CHECK1-NEXT: store i64 [[VLA1]], i64* [[VLA_ADDR2]], align 8
// CHECK1-NEXT: store i32**** [[C]], i32***** [[C_ADDR]], align 8
// CHECK1-NEXT: [[TMP0:%.*]] = load i64, i64* [[VLA_ADDR]], align 8
// CHECK1-NEXT: [[TMP1:%.*]] = load i32**, i32*** [[B_ADDR]], align 8
// CHECK1-NEXT: [[TMP2:%.*]] = load i64, i64* [[VLA_ADDR2]], align 8
// CHECK1-NEXT: [[TMP3:%.*]] = load i32****, i32***** [[C_ADDR]], align 8
// CHECK1-NEXT: [[TMP4:%.*]] = load i32***, i32**** [[TMP3]], align 8
// CHECK1-NEXT: [[ARRAYIDX:%.*]] = getelementptr inbounds i32**, i32*** [[TMP4]], i64 0
// CHECK1-NEXT: [[TMP5:%.*]] = load i32**, i32*** [[ARRAYIDX]], align 8
// CHECK1-NEXT: [[TMP6:%.*]] = load i32, i32* @a, align 4
// CHECK1-NEXT: [[IDXPROM:%.*]] = sext i32 [[TMP6]] to i64
// CHECK1-NEXT: [[TMP7:%.*]] = mul nsw i64 [[IDXPROM]], [[TMP2]]
// CHECK1-NEXT: [[ARRAYIDX3:%.*]] = getelementptr inbounds i32*, i32** [[TMP5]], i64 [[TMP7]]
// CHECK1-NEXT: [[ARRAYIDX4:%.*]] = getelementptr inbounds i32*, i32** [[ARRAYIDX3]], i64 0
// CHECK1-NEXT: [[TMP8:%.*]] = load i32*, i32** [[ARRAYIDX4]], align 8
// CHECK1-NEXT: [[TMP9:%.*]] = load i32, i32* @a, align 4
// CHECK1-NEXT: [[IDXPROM5:%.*]] = sext i32 [[TMP9]] to i64
// CHECK1-NEXT: [[ARRAYIDX6:%.*]] = getelementptr inbounds i32, i32* [[TMP8]], i64 [[IDXPROM5]]
// CHECK1-NEXT: [[TMP10:%.*]] = load i32, i32* [[ARRAYIDX6]], align 4
// CHECK1-NEXT: [[TMP11:%.*]] = load i32*, i32** [[TMP1]], align 8
// CHECK1-NEXT: [[TMP12:%.*]] = mul nsw i64 0, [[TMP0]]
// CHECK1-NEXT: [[ARRAYIDX7:%.*]] = getelementptr inbounds i32, i32* [[TMP11]], i64 [[TMP12]]
// CHECK1-NEXT: [[ARRAYIDX8:%.*]] = getelementptr inbounds i32, i32* [[ARRAYIDX7]], i64 0
// CHECK1-NEXT: store i32 [[TMP10]], i32* [[ARRAYIDX8]], align 4
// CHECK1-NEXT: ret void
//
//
// CHECK1-LABEL: define {{[^@]+}}@bar
// CHECK1-SAME: (i32 noundef signext [[N:%.*]], i32* noundef [[A:%.*]]) #[[ATTR0]] {
// CHECK1-NEXT: entry:
// CHECK1-NEXT: [[N_ADDR:%.*]] = alloca i32, align 4
// CHECK1-NEXT: [[A_ADDR:%.*]] = alloca i32*, align 8
// CHECK1-NEXT: [[P:%.*]] = alloca i32*, align 8
// CHECK1-NEXT: [[DOTTHREADID_TEMP_:%.*]] = alloca i32, align 4
// CHECK1-NEXT: [[DOTBOUND_ZERO_ADDR:%.*]] = alloca i32, align 4
// CHECK1-NEXT: [[TMP0:%.*]] = call i32 @__kmpc_global_thread_num(%struct.ident_t* @[[GLOB1]])
// CHECK1-NEXT: store i32 [[N]], i32* [[N_ADDR]], align 4
// CHECK1-NEXT: store i32* [[A]], i32** [[A_ADDR]], align 8
// CHECK1-NEXT: [[TMP1:%.*]] = load i32, i32* [[N_ADDR]], align 4
// CHECK1-NEXT: [[TMP2:%.*]] = zext i32 [[TMP1]] to i64
// CHECK1-NEXT: [[TMP3:%.*]] = bitcast i32** [[A_ADDR]] to i32*
// CHECK1-NEXT: store i32* [[TMP3]], i32** [[P]], align 8
// CHECK1-NEXT: call void @__kmpc_serialized_parallel(%struct.ident_t* @[[GLOB1]], i32 [[TMP0]])
// CHECK1-NEXT: store i32 [[TMP0]], i32* [[DOTTHREADID_TEMP_]], align 4
// CHECK1-NEXT: store i32 0, i32* [[DOTBOUND_ZERO_ADDR]], align 4
// CHECK1-NEXT: call void @.omp_outlined..1(i32* [[DOTTHREADID_TEMP_]], i32* [[DOTBOUND_ZERO_ADDR]], i64 [[TMP2]], i32** [[P]], i32** [[A_ADDR]]) #[[ATTR2]]
// CHECK1-NEXT: call void @__kmpc_end_serialized_parallel(%struct.ident_t* @[[GLOB1]], i32 [[TMP0]])
// CHECK1-NEXT: ret void
//
//
// CHECK1-LABEL: define {{[^@]+}}@.omp_outlined..1
// CHECK1-SAME: (i32* noalias noundef [[DOTGLOBAL_TID_:%.*]], i32* noalias noundef [[DOTBOUND_TID_:%.*]], i64 noundef [[VLA:%.*]], i32** noundef nonnull align 8 dereferenceable(8) [[P:%.*]], i32** noundef nonnull align 8 dereferenceable(8) [[A:%.*]]) #[[ATTR1]] {
// CHECK1-NEXT: entry:
// CHECK1-NEXT: [[DOTGLOBAL_TID__ADDR:%.*]] = alloca i32*, align 8
// CHECK1-NEXT: [[DOTBOUND_TID__ADDR:%.*]] = alloca i32*, align 8
// CHECK1-NEXT: [[VLA_ADDR:%.*]] = alloca i64, align 8
// CHECK1-NEXT: [[P_ADDR:%.*]] = alloca i32**, align 8
// CHECK1-NEXT: [[A_ADDR:%.*]] = alloca i32**, align 8
// CHECK1-NEXT: store i32* [[DOTGLOBAL_TID_]], i32** [[DOTGLOBAL_TID__ADDR]], align 8
// CHECK1-NEXT: store i32* [[DOTBOUND_TID_]], i32** [[DOTBOUND_TID__ADDR]], align 8
// CHECK1-NEXT: store i64 [[VLA]], i64* [[VLA_ADDR]], align 8
// CHECK1-NEXT: store i32** [[P]], i32*** [[P_ADDR]], align 8
// CHECK1-NEXT: store i32** [[A]], i32*** [[A_ADDR]], align 8
// CHECK1-NEXT: [[TMP0:%.*]] = load i64, i64* [[VLA_ADDR]], align 8
// CHECK1-NEXT: [[TMP1:%.*]] = load i32**, i32*** [[P_ADDR]], align 8
// CHECK1-NEXT: [[TMP2:%.*]] = load i32**, i32*** [[A_ADDR]], align 8
// CHECK1-NEXT: [[TMP3:%.*]] = load i32*, i32** [[TMP1]], align 8
// CHECK1-NEXT: [[TMP4:%.*]] = bitcast i32** [[TMP2]] to i32*
// CHECK1-NEXT: [[CMP:%.*]] = icmp eq i32* [[TMP3]], [[TMP4]]
// CHECK1-NEXT: br i1 [[CMP]], label [[IF_THEN:%.*]], label [[IF_END:%.*]]
// CHECK1: if.then:
// CHECK1-NEXT: br label [[IF_END]]
// CHECK1: if.end:
// CHECK1-NEXT: ret void
//