llvm/clang/test/OpenMP/for_firstprivate_codegen.cpp

// 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 -verify -fopenmp -x c++ -triple x86_64-apple-darwin10 -emit-llvm %s -o - | FileCheck %s --check-prefix=CHECK1
// RUN: %clang_cc1 -fopenmp -x c++ -std=c++11 -triple x86_64-apple-darwin10 -emit-pch -o %t %s
// RUN: %clang_cc1 -fopenmp -x c++ -triple x86_64-apple-darwin10 -std=c++11 -include-pch %t -verify %s -emit-llvm -o - | FileCheck %s --check-prefix=CHECK1
// RUN: %clang_cc1 -verify -fopenmp -x c++ -std=c++11 -DLAMBDA -triple x86_64-apple-darwin10 -emit-llvm %s -o - | FileCheck %s --check-prefix=CHECK3
// RUN: %clang_cc1 -verify -fopenmp -x c++ -fblocks -DBLOCKS -triple x86_64-apple-darwin10 -emit-llvm %s -o - | FileCheck %s --check-prefix=CHECK4

// RUN: %clang_cc1 -verify -fopenmp-simd -x c++ -triple x86_64-apple-darwin10 -emit-llvm %s -o - | FileCheck %s --implicit-check-not="{{__kmpc|__tgt}}"
// RUN: %clang_cc1 -fopenmp-simd -x c++ -std=c++11 -triple x86_64-apple-darwin10 -emit-pch -o %t %s
// RUN: %clang_cc1 -fopenmp-simd -x c++ -triple x86_64-apple-darwin10 -std=c++11 -include-pch %t -verify %s -emit-llvm -o - | FileCheck %s --implicit-check-not="{{__kmpc|__tgt}}"
// RUN: %clang_cc1 -verify -fopenmp-simd -x c++ -std=c++11 -DLAMBDA -triple x86_64-apple-darwin10 -emit-llvm %s -o - | FileCheck %s --implicit-check-not="{{__kmpc|__tgt}}"
// RUN: %clang_cc1 -verify -fopenmp-simd -x c++ -fblocks -DBLOCKS -triple x86_64-apple-darwin10 -emit-llvm %s -o - | FileCheck %s --implicit-check-not="{{__kmpc|__tgt}}"
// expected-no-diagnostics
#ifndef HEADER
#define HEADER

struct St {
  int a, b;
  St() : a(0), b(0) {}
  St(const St &st) : a(st.a + st.b), b(0) {}
  ~St() {}
};

volatile int g = 1212;
volatile int &g1 = g;

template <class T>
struct S {
  T f;
  S(T a) : f(a + g) {}
  S() : f(g) {}
  S(const S &s, St t = St()) : f(s.f + t.a) {}
  operator T() { return T(); }
  ~S() {}
};


template <typename T>
T tmain() {
  S<T> test;
  T t_var = T();
  T vec[] = {1, 2};
  S<T> s_arr[] = {1, 2};
  S<T> &var = test;
#pragma omp parallel
#pragma omp for firstprivate(t_var, vec, s_arr, var)
  for (int i = 0; i < 2; ++i) {
    vec[i] = t_var;
    s_arr[i] = var;
  }
  return T();
}

S<float> test;
int t_var = 333;
int vec[] = {1, 2};
S<float> s_arr[] = {1, 2};
S<float> var(3);

int main() {
  static int sivar;
#ifdef LAMBDA
  [&]() {
#pragma omp parallel
#pragma omp for firstprivate(g, g1, sivar)
  for (int i = 0; i < 2; ++i) {
    // Skip temp vars for loop




    g = 1;
    g1 = 2;
    sivar = 3;

    [&]() {
      g = 4;
      g1 = 5;
      sivar = 6;

    }();
  }
  }();
  return 0;
#elif defined(BLOCKS)
  ^{
#pragma omp parallel
#pragma omp for firstprivate(g, g1, sivar)
  for (int i = 0; i < 2; ++i) {
    // Skip temp vars for loop




    g = 1;
    g1 =1;
    sivar = 2;
    ^{
      g = 2;
      g1 = 2;
      sivar = 4;
    }();
  }
  }();
  return 0;
#else
#pragma omp for firstprivate(t_var, vec, s_arr, var, sivar)
  for (int i = 0; i < 2; ++i) {
    vec[i] = t_var;
    s_arr[i] = var;
    sivar += i;
  }
  return tmain<int>();
#endif
}

// Skip temp vars for loop

// firstprivate t_var(t_var)

// firstprivate vec(vec)

// firstprivate s_arr(s_arr)

// firstprivate var(var)

// firstprivate (sivar)

// Synchronization for initialization.


// ~(firstprivate var), ~(firstprivate s_arr)



// Skip temp vars for loop


// firstprivate vec(vec)

// firstprivate s_arr(s_arr)

// firstprivate var(var)

// No synchronization for initialization.


// ~(firstprivate var), ~(firstprivate s_arr)
#endif

// CHECK1-LABEL: define {{[^@]+}}@__cxx_global_var_init
// CHECK1-SAME: () #[[ATTR0:[0-9]+]] section "__TEXT,__StaticInit,regular,pure_instructions" {
// CHECK1-NEXT:  entry:
// CHECK1-NEXT:    call void @_ZN1SIfEC1Ev(ptr noundef nonnull align 4 dereferenceable(4) @test)
// CHECK1-NEXT:    [[TMP0:%.*]] = call i32 @__cxa_atexit(ptr @_ZN1SIfED1Ev, ptr @test, ptr @__dso_handle) #[[ATTR2:[0-9]+]]
// CHECK1-NEXT:    ret void
//
//
// CHECK1-LABEL: define {{[^@]+}}@_ZN1SIfEC1Ev
// CHECK1-SAME: (ptr noundef nonnull align 4 dereferenceable(4) [[THIS:%.*]]) unnamed_addr #[[ATTR1:[0-9]+]] align 2 {
// CHECK1-NEXT:  entry:
// CHECK1-NEXT:    [[THIS_ADDR:%.*]] = alloca ptr, align 8
// CHECK1-NEXT:    store ptr [[THIS]], ptr [[THIS_ADDR]], align 8
// CHECK1-NEXT:    [[THIS1:%.*]] = load ptr, ptr [[THIS_ADDR]], align 8
// CHECK1-NEXT:    call void @_ZN1SIfEC2Ev(ptr noundef nonnull align 4 dereferenceable(4) [[THIS1]])
// CHECK1-NEXT:    ret void
//
//
// CHECK1-LABEL: define {{[^@]+}}@_ZN1SIfED1Ev
// CHECK1-SAME: (ptr noundef nonnull align 4 dereferenceable(4) [[THIS:%.*]]) unnamed_addr #[[ATTR1]] align 2 {
// CHECK1-NEXT:  entry:
// CHECK1-NEXT:    [[THIS_ADDR:%.*]] = alloca ptr, align 8
// CHECK1-NEXT:    store ptr [[THIS]], ptr [[THIS_ADDR]], align 8
// CHECK1-NEXT:    [[THIS1:%.*]] = load ptr, ptr [[THIS_ADDR]], align 8
// CHECK1-NEXT:    call void @_ZN1SIfED2Ev(ptr noundef nonnull align 4 dereferenceable(4) [[THIS1]]) #[[ATTR2]]
// CHECK1-NEXT:    ret void
//
//
// CHECK1-LABEL: define {{[^@]+}}@_ZN1SIfEC2Ev
// CHECK1-SAME: (ptr noundef nonnull align 4 dereferenceable(4) [[THIS:%.*]]) unnamed_addr #[[ATTR1]] align 2 {
// CHECK1-NEXT:  entry:
// CHECK1-NEXT:    [[THIS_ADDR:%.*]] = alloca ptr, align 8
// CHECK1-NEXT:    store ptr [[THIS]], ptr [[THIS_ADDR]], align 8
// CHECK1-NEXT:    [[THIS1:%.*]] = load ptr, ptr [[THIS_ADDR]], align 8
// CHECK1-NEXT:    [[F:%.*]] = getelementptr inbounds nuw [[STRUCT_S:%.*]], ptr [[THIS1]], i32 0, i32 0
// CHECK1-NEXT:    [[TMP0:%.*]] = load volatile i32, ptr @g, align 4
// CHECK1-NEXT:    [[CONV:%.*]] = sitofp i32 [[TMP0]] to float
// CHECK1-NEXT:    store float [[CONV]], ptr [[F]], align 4
// CHECK1-NEXT:    ret void
//
//
// CHECK1-LABEL: define {{[^@]+}}@_ZN1SIfED2Ev
// CHECK1-SAME: (ptr noundef nonnull align 4 dereferenceable(4) [[THIS:%.*]]) unnamed_addr #[[ATTR1]] align 2 {
// CHECK1-NEXT:  entry:
// CHECK1-NEXT:    [[THIS_ADDR:%.*]] = alloca ptr, align 8
// CHECK1-NEXT:    store ptr [[THIS]], ptr [[THIS_ADDR]], align 8
// CHECK1-NEXT:    [[THIS1:%.*]] = load ptr, ptr [[THIS_ADDR]], align 8
// CHECK1-NEXT:    ret void
//
//
// CHECK1-LABEL: define {{[^@]+}}@__cxx_global_var_init.1
// CHECK1-SAME: () #[[ATTR0]] section "__TEXT,__StaticInit,regular,pure_instructions" {
// CHECK1-NEXT:  entry:
// CHECK1-NEXT:    call void @_ZN1SIfEC1Ef(ptr noundef nonnull align 4 dereferenceable(4) @s_arr, float noundef 1.000000e+00)
// CHECK1-NEXT:    call void @_ZN1SIfEC1Ef(ptr noundef nonnull align 4 dereferenceable(4) getelementptr inbounds ([[STRUCT_S:%.*]], ptr @s_arr, i64 1), float noundef 2.000000e+00)
// CHECK1-NEXT:    [[TMP0:%.*]] = call i32 @__cxa_atexit(ptr @__cxx_global_array_dtor, ptr null, ptr @__dso_handle) #[[ATTR2]]
// CHECK1-NEXT:    ret void
//
//
// CHECK1-LABEL: define {{[^@]+}}@_ZN1SIfEC1Ef
// CHECK1-SAME: (ptr noundef nonnull align 4 dereferenceable(4) [[THIS:%.*]], float noundef [[A:%.*]]) unnamed_addr #[[ATTR1]] align 2 {
// CHECK1-NEXT:  entry:
// CHECK1-NEXT:    [[THIS_ADDR:%.*]] = alloca ptr, align 8
// CHECK1-NEXT:    [[A_ADDR:%.*]] = alloca float, align 4
// CHECK1-NEXT:    store ptr [[THIS]], ptr [[THIS_ADDR]], align 8
// CHECK1-NEXT:    store float [[A]], ptr [[A_ADDR]], align 4
// CHECK1-NEXT:    [[THIS1:%.*]] = load ptr, ptr [[THIS_ADDR]], align 8
// CHECK1-NEXT:    [[TMP0:%.*]] = load float, ptr [[A_ADDR]], align 4
// CHECK1-NEXT:    call void @_ZN1SIfEC2Ef(ptr noundef nonnull align 4 dereferenceable(4) [[THIS1]], float noundef [[TMP0]])
// CHECK1-NEXT:    ret void
//
//
// CHECK1-LABEL: define {{[^@]+}}@__cxx_global_array_dtor
// CHECK1-SAME: (ptr noundef [[TMP0:%.*]]) #[[ATTR0]] section "__TEXT,__StaticInit,regular,pure_instructions" {
// CHECK1-NEXT:  entry:
// CHECK1-NEXT:    [[DOTADDR:%.*]] = alloca ptr, align 8
// CHECK1-NEXT:    store ptr [[TMP0]], ptr [[DOTADDR]], align 8
// CHECK1-NEXT:    br label [[ARRAYDESTROY_BODY:%.*]]
// CHECK1:       arraydestroy.body:
// CHECK1-NEXT:    [[ARRAYDESTROY_ELEMENTPAST:%.*]] = phi ptr [ getelementptr inbounds ([[STRUCT_S:%.*]], ptr @s_arr, i64 2), [[ENTRY:%.*]] ], [ [[ARRAYDESTROY_ELEMENT:%.*]], [[ARRAYDESTROY_BODY]] ]
// CHECK1-NEXT:    [[ARRAYDESTROY_ELEMENT]] = getelementptr inbounds [[STRUCT_S]], ptr [[ARRAYDESTROY_ELEMENTPAST]], i64 -1
// CHECK1-NEXT:    call void @_ZN1SIfED1Ev(ptr noundef nonnull align 4 dereferenceable(4) [[ARRAYDESTROY_ELEMENT]]) #[[ATTR2]]
// CHECK1-NEXT:    [[ARRAYDESTROY_DONE:%.*]] = icmp eq ptr [[ARRAYDESTROY_ELEMENT]], @s_arr
// CHECK1-NEXT:    br i1 [[ARRAYDESTROY_DONE]], label [[ARRAYDESTROY_DONE1:%.*]], label [[ARRAYDESTROY_BODY]]
// CHECK1:       arraydestroy.done1:
// CHECK1-NEXT:    ret void
//
//
// CHECK1-LABEL: define {{[^@]+}}@_ZN1SIfEC2Ef
// CHECK1-SAME: (ptr noundef nonnull align 4 dereferenceable(4) [[THIS:%.*]], float noundef [[A:%.*]]) unnamed_addr #[[ATTR1]] align 2 {
// CHECK1-NEXT:  entry:
// CHECK1-NEXT:    [[THIS_ADDR:%.*]] = alloca ptr, align 8
// CHECK1-NEXT:    [[A_ADDR:%.*]] = alloca float, align 4
// CHECK1-NEXT:    store ptr [[THIS]], ptr [[THIS_ADDR]], align 8
// CHECK1-NEXT:    store float [[A]], ptr [[A_ADDR]], align 4
// CHECK1-NEXT:    [[THIS1:%.*]] = load ptr, ptr [[THIS_ADDR]], align 8
// CHECK1-NEXT:    [[F:%.*]] = getelementptr inbounds nuw [[STRUCT_S:%.*]], ptr [[THIS1]], i32 0, i32 0
// CHECK1-NEXT:    [[TMP0:%.*]] = load float, ptr [[A_ADDR]], align 4
// CHECK1-NEXT:    [[TMP1:%.*]] = load volatile i32, ptr @g, align 4
// CHECK1-NEXT:    [[CONV:%.*]] = sitofp i32 [[TMP1]] to float
// CHECK1-NEXT:    [[ADD:%.*]] = fadd float [[TMP0]], [[CONV]]
// CHECK1-NEXT:    store float [[ADD]], ptr [[F]], align 4
// CHECK1-NEXT:    ret void
//
//
// CHECK1-LABEL: define {{[^@]+}}@__cxx_global_var_init.2
// CHECK1-SAME: () #[[ATTR0]] section "__TEXT,__StaticInit,regular,pure_instructions" {
// CHECK1-NEXT:  entry:
// CHECK1-NEXT:    call void @_ZN1SIfEC1Ef(ptr noundef nonnull align 4 dereferenceable(4) @var, float noundef 3.000000e+00)
// CHECK1-NEXT:    [[TMP0:%.*]] = call i32 @__cxa_atexit(ptr @_ZN1SIfED1Ev, ptr @var, ptr @__dso_handle) #[[ATTR2]]
// CHECK1-NEXT:    ret void
//
//
// CHECK1-LABEL: define {{[^@]+}}@main
// CHECK1-SAME: () #[[ATTR3:[0-9]+]] {
// CHECK1-NEXT:  entry:
// CHECK1-NEXT:    [[RETVAL:%.*]] = alloca i32, align 4
// CHECK1-NEXT:    [[DOTOMP_IV:%.*]] = alloca i32, align 4
// CHECK1-NEXT:    [[TMP:%.*]] = alloca i32, align 4
// CHECK1-NEXT:    [[DOTOMP_LB:%.*]] = alloca i32, align 4
// CHECK1-NEXT:    [[DOTOMP_UB:%.*]] = alloca i32, align 4
// CHECK1-NEXT:    [[DOTOMP_STRIDE:%.*]] = alloca i32, align 4
// CHECK1-NEXT:    [[DOTOMP_IS_LAST:%.*]] = alloca i32, align 4
// CHECK1-NEXT:    [[T_VAR:%.*]] = alloca i32, align 4
// CHECK1-NEXT:    [[VEC:%.*]] = alloca [2 x i32], align 4
// CHECK1-NEXT:    [[S_ARR:%.*]] = alloca [2 x %struct.S], align 4
// CHECK1-NEXT:    [[AGG_TMP:%.*]] = alloca [[STRUCT_ST:%.*]], align 4
// CHECK1-NEXT:    [[VAR:%.*]] = alloca [[STRUCT_S:%.*]], align 4
// CHECK1-NEXT:    [[AGG_TMP2:%.*]] = alloca [[STRUCT_ST]], align 4
// CHECK1-NEXT:    [[SIVAR:%.*]] = alloca i32, align 4
// CHECK1-NEXT:    [[I:%.*]] = alloca i32, align 4
// CHECK1-NEXT:    [[TMP0:%.*]] = call i32 @__kmpc_global_thread_num(ptr @[[GLOB2:[0-9]+]])
// CHECK1-NEXT:    store i32 0, ptr [[RETVAL]], align 4
// CHECK1-NEXT:    store i32 0, ptr [[DOTOMP_LB]], align 4
// CHECK1-NEXT:    store i32 1, ptr [[DOTOMP_UB]], align 4
// CHECK1-NEXT:    store i32 1, ptr [[DOTOMP_STRIDE]], align 4
// CHECK1-NEXT:    store i32 0, ptr [[DOTOMP_IS_LAST]], align 4
// CHECK1-NEXT:    [[TMP1:%.*]] = load i32, ptr @t_var, align 4
// CHECK1-NEXT:    store i32 [[TMP1]], ptr [[T_VAR]], align 4
// CHECK1-NEXT:    call void @llvm.memcpy.p0.p0.i64(ptr align 4 [[VEC]], ptr align 4 @vec, i64 8, i1 false)
// CHECK1-NEXT:    [[ARRAY_BEGIN:%.*]] = getelementptr inbounds [2 x %struct.S], ptr [[S_ARR]], i32 0, i32 0
// CHECK1-NEXT:    [[TMP2:%.*]] = getelementptr inbounds [[STRUCT_S]], ptr [[ARRAY_BEGIN]], i64 2
// CHECK1-NEXT:    [[OMP_ARRAYCPY_ISEMPTY:%.*]] = icmp eq ptr [[ARRAY_BEGIN]], [[TMP2]]
// CHECK1-NEXT:    br i1 [[OMP_ARRAYCPY_ISEMPTY]], label [[OMP_ARRAYCPY_DONE1:%.*]], label [[OMP_ARRAYCPY_BODY:%.*]]
// CHECK1:       omp.arraycpy.body:
// CHECK1-NEXT:    [[OMP_ARRAYCPY_SRCELEMENTPAST:%.*]] = phi ptr [ @s_arr, [[ENTRY:%.*]] ], [ [[OMP_ARRAYCPY_SRC_ELEMENT:%.*]], [[OMP_ARRAYCPY_BODY]] ]
// CHECK1-NEXT:    [[OMP_ARRAYCPY_DESTELEMENTPAST:%.*]] = phi ptr [ [[ARRAY_BEGIN]], [[ENTRY]] ], [ [[OMP_ARRAYCPY_DEST_ELEMENT:%.*]], [[OMP_ARRAYCPY_BODY]] ]
// CHECK1-NEXT:    call void @_ZN2StC1Ev(ptr noundef nonnull align 4 dereferenceable(8) [[AGG_TMP]])
// CHECK1-NEXT:    call void @_ZN1SIfEC1ERKS0_2St(ptr noundef nonnull align 4 dereferenceable(4) [[OMP_ARRAYCPY_DESTELEMENTPAST]], ptr noundef nonnull align 4 dereferenceable(4) [[OMP_ARRAYCPY_SRCELEMENTPAST]], ptr noundef [[AGG_TMP]])
// CHECK1-NEXT:    call void @_ZN2StD1Ev(ptr noundef nonnull align 4 dereferenceable(8) [[AGG_TMP]]) #[[ATTR2]]
// CHECK1-NEXT:    [[OMP_ARRAYCPY_DEST_ELEMENT]] = getelementptr [[STRUCT_S]], ptr [[OMP_ARRAYCPY_DESTELEMENTPAST]], i32 1
// CHECK1-NEXT:    [[OMP_ARRAYCPY_SRC_ELEMENT]] = getelementptr [[STRUCT_S]], ptr [[OMP_ARRAYCPY_SRCELEMENTPAST]], i32 1
// CHECK1-NEXT:    [[OMP_ARRAYCPY_DONE:%.*]] = icmp eq ptr [[OMP_ARRAYCPY_DEST_ELEMENT]], [[TMP2]]
// CHECK1-NEXT:    br i1 [[OMP_ARRAYCPY_DONE]], label [[OMP_ARRAYCPY_DONE1]], label [[OMP_ARRAYCPY_BODY]]
// CHECK1:       omp.arraycpy.done1:
// CHECK1-NEXT:    call void @_ZN2StC1Ev(ptr noundef nonnull align 4 dereferenceable(8) [[AGG_TMP2]])
// CHECK1-NEXT:    call void @_ZN1SIfEC1ERKS0_2St(ptr noundef nonnull align 4 dereferenceable(4) [[VAR]], ptr noundef nonnull align 4 dereferenceable(4) @var, ptr noundef [[AGG_TMP2]])
// CHECK1-NEXT:    call void @_ZN2StD1Ev(ptr noundef nonnull align 4 dereferenceable(8) [[AGG_TMP2]]) #[[ATTR2]]
// CHECK1-NEXT:    [[TMP3:%.*]] = load i32, ptr @_ZZ4mainE5sivar, align 4
// CHECK1-NEXT:    store i32 [[TMP3]], ptr [[SIVAR]], align 4
// CHECK1-NEXT:    call void @__kmpc_for_static_init_4(ptr @[[GLOB1:[0-9]+]], i32 [[TMP0]], i32 34, ptr [[DOTOMP_IS_LAST]], ptr [[DOTOMP_LB]], ptr [[DOTOMP_UB]], ptr [[DOTOMP_STRIDE]], i32 1, i32 1)
// CHECK1-NEXT:    [[TMP4:%.*]] = load i32, ptr [[DOTOMP_UB]], align 4
// CHECK1-NEXT:    [[CMP:%.*]] = icmp sgt i32 [[TMP4]], 1
// CHECK1-NEXT:    br i1 [[CMP]], label [[COND_TRUE:%.*]], label [[COND_FALSE:%.*]]
// CHECK1:       cond.true:
// CHECK1-NEXT:    br label [[COND_END:%.*]]
// CHECK1:       cond.false:
// CHECK1-NEXT:    [[TMP5:%.*]] = load i32, ptr [[DOTOMP_UB]], align 4
// CHECK1-NEXT:    br label [[COND_END]]
// CHECK1:       cond.end:
// CHECK1-NEXT:    [[COND:%.*]] = phi i32 [ 1, [[COND_TRUE]] ], [ [[TMP5]], [[COND_FALSE]] ]
// CHECK1-NEXT:    store i32 [[COND]], ptr [[DOTOMP_UB]], align 4
// CHECK1-NEXT:    [[TMP6:%.*]] = load i32, ptr [[DOTOMP_LB]], align 4
// CHECK1-NEXT:    store i32 [[TMP6]], ptr [[DOTOMP_IV]], align 4
// CHECK1-NEXT:    br label [[OMP_INNER_FOR_COND:%.*]]
// CHECK1:       omp.inner.for.cond:
// CHECK1-NEXT:    [[TMP7:%.*]] = load i32, ptr [[DOTOMP_IV]], align 4
// CHECK1-NEXT:    [[TMP8:%.*]] = load i32, ptr [[DOTOMP_UB]], align 4
// CHECK1-NEXT:    [[CMP3:%.*]] = icmp sle i32 [[TMP7]], [[TMP8]]
// CHECK1-NEXT:    br i1 [[CMP3]], label [[OMP_INNER_FOR_BODY:%.*]], label [[OMP_INNER_FOR_COND_CLEANUP:%.*]]
// CHECK1:       omp.inner.for.cond.cleanup:
// CHECK1-NEXT:    br label [[OMP_INNER_FOR_END:%.*]]
// CHECK1:       omp.inner.for.body:
// CHECK1-NEXT:    [[TMP9:%.*]] = load i32, ptr [[DOTOMP_IV]], align 4
// CHECK1-NEXT:    [[MUL:%.*]] = mul nsw i32 [[TMP9]], 1
// CHECK1-NEXT:    [[ADD:%.*]] = add nsw i32 0, [[MUL]]
// CHECK1-NEXT:    store i32 [[ADD]], ptr [[I]], align 4
// CHECK1-NEXT:    [[TMP10:%.*]] = load i32, ptr [[T_VAR]], align 4
// CHECK1-NEXT:    [[TMP11:%.*]] = load i32, ptr [[I]], align 4
// CHECK1-NEXT:    [[IDXPROM:%.*]] = sext i32 [[TMP11]] to i64
// CHECK1-NEXT:    [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x i32], ptr [[VEC]], i64 0, i64 [[IDXPROM]]
// CHECK1-NEXT:    store i32 [[TMP10]], ptr [[ARRAYIDX]], align 4
// CHECK1-NEXT:    [[TMP12:%.*]] = load i32, ptr [[I]], align 4
// CHECK1-NEXT:    [[IDXPROM4:%.*]] = sext i32 [[TMP12]] to i64
// CHECK1-NEXT:    [[ARRAYIDX5:%.*]] = getelementptr inbounds [2 x %struct.S], ptr [[S_ARR]], i64 0, i64 [[IDXPROM4]]
// CHECK1-NEXT:    call void @llvm.memcpy.p0.p0.i64(ptr align 4 [[ARRAYIDX5]], ptr align 4 [[VAR]], i64 4, i1 false)
// CHECK1-NEXT:    [[TMP13:%.*]] = load i32, ptr [[I]], align 4
// CHECK1-NEXT:    [[TMP14:%.*]] = load i32, ptr [[SIVAR]], align 4
// CHECK1-NEXT:    [[ADD6:%.*]] = add nsw i32 [[TMP14]], [[TMP13]]
// CHECK1-NEXT:    store i32 [[ADD6]], ptr [[SIVAR]], align 4
// CHECK1-NEXT:    br label [[OMP_BODY_CONTINUE:%.*]]
// CHECK1:       omp.body.continue:
// CHECK1-NEXT:    br label [[OMP_INNER_FOR_INC:%.*]]
// CHECK1:       omp.inner.for.inc:
// CHECK1-NEXT:    [[TMP15:%.*]] = load i32, ptr [[DOTOMP_IV]], align 4
// CHECK1-NEXT:    [[ADD7:%.*]] = add nsw i32 [[TMP15]], 1
// CHECK1-NEXT:    store i32 [[ADD7]], ptr [[DOTOMP_IV]], align 4
// CHECK1-NEXT:    br label [[OMP_INNER_FOR_COND]]
// CHECK1:       omp.inner.for.end:
// CHECK1-NEXT:    br label [[OMP_LOOP_EXIT:%.*]]
// CHECK1:       omp.loop.exit:
// CHECK1-NEXT:    call void @__kmpc_for_static_fini(ptr @[[GLOB1]], i32 [[TMP0]])
// CHECK1-NEXT:    call void @_ZN1SIfED1Ev(ptr noundef nonnull align 4 dereferenceable(4) [[VAR]]) #[[ATTR2]]
// CHECK1-NEXT:    [[ARRAY_BEGIN8:%.*]] = getelementptr inbounds [2 x %struct.S], ptr [[S_ARR]], i32 0, i32 0
// CHECK1-NEXT:    [[TMP16:%.*]] = getelementptr inbounds [[STRUCT_S]], ptr [[ARRAY_BEGIN8]], i64 2
// CHECK1-NEXT:    br label [[ARRAYDESTROY_BODY:%.*]]
// CHECK1:       arraydestroy.body:
// CHECK1-NEXT:    [[ARRAYDESTROY_ELEMENTPAST:%.*]] = phi ptr [ [[TMP16]], [[OMP_LOOP_EXIT]] ], [ [[ARRAYDESTROY_ELEMENT:%.*]], [[ARRAYDESTROY_BODY]] ]
// CHECK1-NEXT:    [[ARRAYDESTROY_ELEMENT]] = getelementptr inbounds [[STRUCT_S]], ptr [[ARRAYDESTROY_ELEMENTPAST]], i64 -1
// CHECK1-NEXT:    call void @_ZN1SIfED1Ev(ptr noundef nonnull align 4 dereferenceable(4) [[ARRAYDESTROY_ELEMENT]]) #[[ATTR2]]
// CHECK1-NEXT:    [[ARRAYDESTROY_DONE:%.*]] = icmp eq ptr [[ARRAYDESTROY_ELEMENT]], [[ARRAY_BEGIN8]]
// CHECK1-NEXT:    br i1 [[ARRAYDESTROY_DONE]], label [[ARRAYDESTROY_DONE9:%.*]], label [[ARRAYDESTROY_BODY]]
// CHECK1:       arraydestroy.done9:
// CHECK1-NEXT:    call void @__kmpc_barrier(ptr @[[GLOB3:[0-9]+]], i32 [[TMP0]])
// CHECK1-NEXT:    [[CALL:%.*]] = call noundef i32 @_Z5tmainIiET_v()
// CHECK1-NEXT:    ret i32 [[CALL]]
//
//
// CHECK1-LABEL: define {{[^@]+}}@_ZN2StC1Ev
// CHECK1-SAME: (ptr noundef nonnull align 4 dereferenceable(8) [[THIS:%.*]]) unnamed_addr #[[ATTR1]] align 2 {
// CHECK1-NEXT:  entry:
// CHECK1-NEXT:    [[THIS_ADDR:%.*]] = alloca ptr, align 8
// CHECK1-NEXT:    store ptr [[THIS]], ptr [[THIS_ADDR]], align 8
// CHECK1-NEXT:    [[THIS1:%.*]] = load ptr, ptr [[THIS_ADDR]], align 8
// CHECK1-NEXT:    call void @_ZN2StC2Ev(ptr noundef nonnull align 4 dereferenceable(8) [[THIS1]])
// CHECK1-NEXT:    ret void
//
//
// CHECK1-LABEL: define {{[^@]+}}@_ZN1SIfEC1ERKS0_2St
// CHECK1-SAME: (ptr noundef nonnull align 4 dereferenceable(4) [[THIS:%.*]], ptr noundef nonnull align 4 dereferenceable(4) [[S:%.*]], ptr noundef [[T:%.*]]) unnamed_addr #[[ATTR1]] align 2 {
// CHECK1-NEXT:  entry:
// CHECK1-NEXT:    [[THIS_ADDR:%.*]] = alloca ptr, align 8
// CHECK1-NEXT:    [[S_ADDR:%.*]] = alloca ptr, align 8
// CHECK1-NEXT:    [[T_INDIRECT_ADDR:%.*]] = alloca ptr, align 8
// CHECK1-NEXT:    store ptr [[THIS]], ptr [[THIS_ADDR]], align 8
// CHECK1-NEXT:    store ptr [[S]], ptr [[S_ADDR]], align 8
// CHECK1-NEXT:    store ptr [[T]], ptr [[T_INDIRECT_ADDR]], align 8
// CHECK1-NEXT:    [[THIS1:%.*]] = load ptr, ptr [[THIS_ADDR]], align 8
// CHECK1-NEXT:    [[TMP0:%.*]] = load ptr, ptr [[S_ADDR]], align 8
// CHECK1-NEXT:    call void @_ZN1SIfEC2ERKS0_2St(ptr noundef nonnull align 4 dereferenceable(4) [[THIS1]], ptr noundef nonnull align 4 dereferenceable(4) [[TMP0]], ptr noundef [[T]])
// CHECK1-NEXT:    ret void
//
//
// CHECK1-LABEL: define {{[^@]+}}@_ZN2StD1Ev
// CHECK1-SAME: (ptr noundef nonnull align 4 dereferenceable(8) [[THIS:%.*]]) unnamed_addr #[[ATTR1]] align 2 {
// CHECK1-NEXT:  entry:
// CHECK1-NEXT:    [[THIS_ADDR:%.*]] = alloca ptr, align 8
// CHECK1-NEXT:    store ptr [[THIS]], ptr [[THIS_ADDR]], align 8
// CHECK1-NEXT:    [[THIS1:%.*]] = load ptr, ptr [[THIS_ADDR]], align 8
// CHECK1-NEXT:    call void @_ZN2StD2Ev(ptr noundef nonnull align 4 dereferenceable(8) [[THIS1]]) #[[ATTR2]]
// CHECK1-NEXT:    ret void
//
//
// CHECK1-LABEL: define {{[^@]+}}@_Z5tmainIiET_v
// CHECK1-SAME: () #[[ATTR1]] {
// CHECK1-NEXT:  entry:
// CHECK1-NEXT:    [[RETVAL:%.*]] = alloca i32, align 4
// CHECK1-NEXT:    [[TEST:%.*]] = alloca [[STRUCT_S_0:%.*]], align 4
// CHECK1-NEXT:    [[T_VAR:%.*]] = alloca i32, align 4
// CHECK1-NEXT:    [[VEC:%.*]] = alloca [2 x i32], align 4
// CHECK1-NEXT:    [[S_ARR:%.*]] = alloca [2 x %struct.S.0], align 4
// CHECK1-NEXT:    [[VAR:%.*]] = alloca ptr, align 8
// CHECK1-NEXT:    call void @_ZN1SIiEC1Ev(ptr noundef nonnull align 4 dereferenceable(4) [[TEST]])
// CHECK1-NEXT:    store i32 0, ptr [[T_VAR]], align 4
// CHECK1-NEXT:    call void @llvm.memcpy.p0.p0.i64(ptr align 4 [[VEC]], ptr align 4 @__const._Z5tmainIiET_v.vec, i64 8, i1 false)
// CHECK1-NEXT:    call void @_ZN1SIiEC1Ei(ptr noundef nonnull align 4 dereferenceable(4) [[S_ARR]], i32 noundef 1)
// CHECK1-NEXT:    [[ARRAYINIT_ELEMENT:%.*]] = getelementptr inbounds [[STRUCT_S_0]], ptr [[S_ARR]], i64 1
// CHECK1-NEXT:    call void @_ZN1SIiEC1Ei(ptr noundef nonnull align 4 dereferenceable(4) [[ARRAYINIT_ELEMENT]], i32 noundef 2)
// CHECK1-NEXT:    store ptr [[TEST]], ptr [[VAR]], align 8
// CHECK1-NEXT:    [[TMP0:%.*]] = load ptr, ptr [[VAR]], align 8
// CHECK1-NEXT:    call void (ptr, i32, ptr, ...) @__kmpc_fork_call(ptr @[[GLOB2]], i32 4, ptr @_Z5tmainIiET_v.omp_outlined, ptr [[T_VAR]], ptr [[VEC]], ptr [[S_ARR]], ptr [[TMP0]])
// CHECK1-NEXT:    store i32 0, ptr [[RETVAL]], align 4
// CHECK1-NEXT:    [[ARRAY_BEGIN:%.*]] = getelementptr inbounds [2 x %struct.S.0], ptr [[S_ARR]], i32 0, i32 0
// CHECK1-NEXT:    [[TMP1:%.*]] = getelementptr inbounds [[STRUCT_S_0]], ptr [[ARRAY_BEGIN]], i64 2
// CHECK1-NEXT:    br label [[ARRAYDESTROY_BODY:%.*]]
// CHECK1:       arraydestroy.body:
// CHECK1-NEXT:    [[ARRAYDESTROY_ELEMENTPAST:%.*]] = phi ptr [ [[TMP1]], [[ENTRY:%.*]] ], [ [[ARRAYDESTROY_ELEMENT:%.*]], [[ARRAYDESTROY_BODY]] ]
// CHECK1-NEXT:    [[ARRAYDESTROY_ELEMENT]] = getelementptr inbounds [[STRUCT_S_0]], ptr [[ARRAYDESTROY_ELEMENTPAST]], i64 -1
// CHECK1-NEXT:    call void @_ZN1SIiED1Ev(ptr noundef nonnull align 4 dereferenceable(4) [[ARRAYDESTROY_ELEMENT]]) #[[ATTR2]]
// CHECK1-NEXT:    [[ARRAYDESTROY_DONE:%.*]] = icmp eq ptr [[ARRAYDESTROY_ELEMENT]], [[ARRAY_BEGIN]]
// CHECK1-NEXT:    br i1 [[ARRAYDESTROY_DONE]], label [[ARRAYDESTROY_DONE1:%.*]], label [[ARRAYDESTROY_BODY]]
// CHECK1:       arraydestroy.done1:
// CHECK1-NEXT:    call void @_ZN1SIiED1Ev(ptr noundef nonnull align 4 dereferenceable(4) [[TEST]]) #[[ATTR2]]
// CHECK1-NEXT:    [[TMP2:%.*]] = load i32, ptr [[RETVAL]], align 4
// CHECK1-NEXT:    ret i32 [[TMP2]]
//
//
// CHECK1-LABEL: define {{[^@]+}}@_ZN2StC2Ev
// CHECK1-SAME: (ptr noundef nonnull align 4 dereferenceable(8) [[THIS:%.*]]) unnamed_addr #[[ATTR1]] align 2 {
// CHECK1-NEXT:  entry:
// CHECK1-NEXT:    [[THIS_ADDR:%.*]] = alloca ptr, align 8
// CHECK1-NEXT:    store ptr [[THIS]], ptr [[THIS_ADDR]], align 8
// CHECK1-NEXT:    [[THIS1:%.*]] = load ptr, ptr [[THIS_ADDR]], align 8
// CHECK1-NEXT:    [[A:%.*]] = getelementptr inbounds nuw [[STRUCT_ST:%.*]], ptr [[THIS1]], i32 0, i32 0
// CHECK1-NEXT:    store i32 0, ptr [[A]], align 4
// CHECK1-NEXT:    [[B:%.*]] = getelementptr inbounds nuw [[STRUCT_ST]], ptr [[THIS1]], i32 0, i32 1
// CHECK1-NEXT:    store i32 0, ptr [[B]], align 4
// CHECK1-NEXT:    ret void
//
//
// CHECK1-LABEL: define {{[^@]+}}@_ZN1SIfEC2ERKS0_2St
// CHECK1-SAME: (ptr noundef nonnull align 4 dereferenceable(4) [[THIS:%.*]], ptr noundef nonnull align 4 dereferenceable(4) [[S:%.*]], ptr noundef [[T:%.*]]) unnamed_addr #[[ATTR1]] align 2 {
// CHECK1-NEXT:  entry:
// CHECK1-NEXT:    [[THIS_ADDR:%.*]] = alloca ptr, align 8
// CHECK1-NEXT:    [[S_ADDR:%.*]] = alloca ptr, align 8
// CHECK1-NEXT:    [[T_INDIRECT_ADDR:%.*]] = alloca ptr, align 8
// CHECK1-NEXT:    store ptr [[THIS]], ptr [[THIS_ADDR]], align 8
// CHECK1-NEXT:    store ptr [[S]], ptr [[S_ADDR]], align 8
// CHECK1-NEXT:    store ptr [[T]], ptr [[T_INDIRECT_ADDR]], align 8
// CHECK1-NEXT:    [[THIS1:%.*]] = load ptr, ptr [[THIS_ADDR]], align 8
// CHECK1-NEXT:    [[F:%.*]] = getelementptr inbounds nuw [[STRUCT_S:%.*]], ptr [[THIS1]], i32 0, i32 0
// CHECK1-NEXT:    [[TMP0:%.*]] = load ptr, ptr [[S_ADDR]], align 8
// CHECK1-NEXT:    [[F2:%.*]] = getelementptr inbounds nuw [[STRUCT_S]], ptr [[TMP0]], i32 0, i32 0
// CHECK1-NEXT:    [[TMP1:%.*]] = load float, ptr [[F2]], align 4
// CHECK1-NEXT:    [[A:%.*]] = getelementptr inbounds nuw [[STRUCT_ST:%.*]], ptr [[T]], i32 0, i32 0
// CHECK1-NEXT:    [[TMP2:%.*]] = load i32, ptr [[A]], align 4
// CHECK1-NEXT:    [[CONV:%.*]] = sitofp i32 [[TMP2]] to float
// CHECK1-NEXT:    [[ADD:%.*]] = fadd float [[TMP1]], [[CONV]]
// CHECK1-NEXT:    store float [[ADD]], ptr [[F]], align 4
// CHECK1-NEXT:    ret void
//
//
// CHECK1-LABEL: define {{[^@]+}}@_ZN2StD2Ev
// CHECK1-SAME: (ptr noundef nonnull align 4 dereferenceable(8) [[THIS:%.*]]) unnamed_addr #[[ATTR1]] align 2 {
// CHECK1-NEXT:  entry:
// CHECK1-NEXT:    [[THIS_ADDR:%.*]] = alloca ptr, align 8
// CHECK1-NEXT:    store ptr [[THIS]], ptr [[THIS_ADDR]], align 8
// CHECK1-NEXT:    [[THIS1:%.*]] = load ptr, ptr [[THIS_ADDR]], align 8
// CHECK1-NEXT:    ret void
//
//
// CHECK1-LABEL: define {{[^@]+}}@_ZN1SIiEC1Ev
// CHECK1-SAME: (ptr noundef nonnull align 4 dereferenceable(4) [[THIS:%.*]]) unnamed_addr #[[ATTR1]] align 2 {
// CHECK1-NEXT:  entry:
// CHECK1-NEXT:    [[THIS_ADDR:%.*]] = alloca ptr, align 8
// CHECK1-NEXT:    store ptr [[THIS]], ptr [[THIS_ADDR]], align 8
// CHECK1-NEXT:    [[THIS1:%.*]] = load ptr, ptr [[THIS_ADDR]], align 8
// CHECK1-NEXT:    call void @_ZN1SIiEC2Ev(ptr noundef nonnull align 4 dereferenceable(4) [[THIS1]])
// CHECK1-NEXT:    ret void
//
//
// CHECK1-LABEL: define {{[^@]+}}@_ZN1SIiEC1Ei
// CHECK1-SAME: (ptr noundef nonnull align 4 dereferenceable(4) [[THIS:%.*]], i32 noundef [[A:%.*]]) unnamed_addr #[[ATTR1]] align 2 {
// CHECK1-NEXT:  entry:
// CHECK1-NEXT:    [[THIS_ADDR:%.*]] = alloca ptr, align 8
// CHECK1-NEXT:    [[A_ADDR:%.*]] = alloca i32, align 4
// CHECK1-NEXT:    store ptr [[THIS]], ptr [[THIS_ADDR]], align 8
// CHECK1-NEXT:    store i32 [[A]], ptr [[A_ADDR]], align 4
// CHECK1-NEXT:    [[THIS1:%.*]] = load ptr, ptr [[THIS_ADDR]], align 8
// CHECK1-NEXT:    [[TMP0:%.*]] = load i32, ptr [[A_ADDR]], align 4
// CHECK1-NEXT:    call void @_ZN1SIiEC2Ei(ptr noundef nonnull align 4 dereferenceable(4) [[THIS1]], i32 noundef [[TMP0]])
// CHECK1-NEXT:    ret void
//
//
// CHECK1-LABEL: define {{[^@]+}}@_Z5tmainIiET_v.omp_outlined
// CHECK1-SAME: (ptr noalias noundef [[DOTGLOBAL_TID_:%.*]], ptr noalias noundef [[DOTBOUND_TID_:%.*]], ptr noundef nonnull align 4 dereferenceable(4) [[T_VAR:%.*]], ptr noundef nonnull align 4 dereferenceable(8) [[VEC:%.*]], ptr noundef nonnull align 4 dereferenceable(8) [[S_ARR:%.*]], ptr noundef nonnull align 4 dereferenceable(4) [[VAR:%.*]]) #[[ATTR6:[0-9]+]] {
// CHECK1-NEXT:  entry:
// CHECK1-NEXT:    [[DOTGLOBAL_TID__ADDR:%.*]] = alloca ptr, align 8
// CHECK1-NEXT:    [[DOTBOUND_TID__ADDR:%.*]] = alloca ptr, align 8
// CHECK1-NEXT:    [[T_VAR_ADDR:%.*]] = alloca ptr, align 8
// CHECK1-NEXT:    [[VEC_ADDR:%.*]] = alloca ptr, align 8
// CHECK1-NEXT:    [[S_ARR_ADDR:%.*]] = alloca ptr, align 8
// CHECK1-NEXT:    [[VAR_ADDR:%.*]] = alloca ptr, align 8
// CHECK1-NEXT:    [[TMP:%.*]] = alloca ptr, align 8
// CHECK1-NEXT:    [[_TMP1:%.*]] = alloca ptr, align 8
// CHECK1-NEXT:    [[DOTOMP_IV:%.*]] = alloca i32, align 4
// CHECK1-NEXT:    [[_TMP2:%.*]] = alloca i32, align 4
// CHECK1-NEXT:    [[DOTOMP_LB:%.*]] = alloca i32, align 4
// CHECK1-NEXT:    [[DOTOMP_UB:%.*]] = alloca i32, align 4
// CHECK1-NEXT:    [[DOTOMP_STRIDE:%.*]] = alloca i32, align 4
// CHECK1-NEXT:    [[DOTOMP_IS_LAST:%.*]] = alloca i32, align 4
// CHECK1-NEXT:    [[T_VAR3:%.*]] = alloca i32, align 4
// CHECK1-NEXT:    [[VEC4:%.*]] = alloca [2 x i32], align 4
// CHECK1-NEXT:    [[S_ARR5:%.*]] = alloca [2 x %struct.S.0], align 4
// CHECK1-NEXT:    [[AGG_TMP:%.*]] = alloca [[STRUCT_ST:%.*]], align 4
// CHECK1-NEXT:    [[VAR7:%.*]] = alloca [[STRUCT_S_0:%.*]], align 4
// CHECK1-NEXT:    [[AGG_TMP8:%.*]] = alloca [[STRUCT_ST]], align 4
// CHECK1-NEXT:    [[_TMP9:%.*]] = alloca ptr, align 8
// CHECK1-NEXT:    [[I:%.*]] = alloca i32, align 4
// CHECK1-NEXT:    store ptr [[DOTGLOBAL_TID_]], ptr [[DOTGLOBAL_TID__ADDR]], align 8
// CHECK1-NEXT:    store ptr [[DOTBOUND_TID_]], ptr [[DOTBOUND_TID__ADDR]], align 8
// CHECK1-NEXT:    store ptr [[T_VAR]], ptr [[T_VAR_ADDR]], align 8
// CHECK1-NEXT:    store ptr [[VEC]], ptr [[VEC_ADDR]], align 8
// CHECK1-NEXT:    store ptr [[S_ARR]], ptr [[S_ARR_ADDR]], align 8
// CHECK1-NEXT:    store ptr [[VAR]], ptr [[VAR_ADDR]], align 8
// CHECK1-NEXT:    [[TMP0:%.*]] = load ptr, ptr [[T_VAR_ADDR]], align 8
// CHECK1-NEXT:    [[TMP1:%.*]] = load ptr, ptr [[VEC_ADDR]], align 8
// CHECK1-NEXT:    [[TMP2:%.*]] = load ptr, ptr [[S_ARR_ADDR]], align 8
// CHECK1-NEXT:    [[TMP3:%.*]] = load ptr, ptr [[VAR_ADDR]], align 8
// CHECK1-NEXT:    store ptr [[TMP3]], ptr [[TMP]], align 8
// CHECK1-NEXT:    [[TMP4:%.*]] = load ptr, ptr [[TMP]], align 8
// CHECK1-NEXT:    store ptr [[TMP4]], ptr [[_TMP1]], align 8
// CHECK1-NEXT:    store i32 0, ptr [[DOTOMP_LB]], align 4
// CHECK1-NEXT:    store i32 1, ptr [[DOTOMP_UB]], align 4
// CHECK1-NEXT:    store i32 1, ptr [[DOTOMP_STRIDE]], align 4
// CHECK1-NEXT:    store i32 0, ptr [[DOTOMP_IS_LAST]], align 4
// CHECK1-NEXT:    [[TMP5:%.*]] = load i32, ptr [[TMP0]], align 4
// CHECK1-NEXT:    store i32 [[TMP5]], ptr [[T_VAR3]], align 4
// CHECK1-NEXT:    call void @llvm.memcpy.p0.p0.i64(ptr align 4 [[VEC4]], ptr align 4 [[TMP1]], i64 8, i1 false)
// CHECK1-NEXT:    [[ARRAY_BEGIN:%.*]] = getelementptr inbounds [2 x %struct.S.0], ptr [[S_ARR5]], i32 0, i32 0
// CHECK1-NEXT:    [[TMP6:%.*]] = getelementptr inbounds [[STRUCT_S_0]], ptr [[ARRAY_BEGIN]], i64 2
// CHECK1-NEXT:    [[OMP_ARRAYCPY_ISEMPTY:%.*]] = icmp eq ptr [[ARRAY_BEGIN]], [[TMP6]]
// CHECK1-NEXT:    br i1 [[OMP_ARRAYCPY_ISEMPTY]], label [[OMP_ARRAYCPY_DONE6:%.*]], label [[OMP_ARRAYCPY_BODY:%.*]]
// CHECK1:       omp.arraycpy.body:
// CHECK1-NEXT:    [[OMP_ARRAYCPY_SRCELEMENTPAST:%.*]] = phi ptr [ [[TMP2]], [[ENTRY:%.*]] ], [ [[OMP_ARRAYCPY_SRC_ELEMENT:%.*]], [[OMP_ARRAYCPY_BODY]] ]
// CHECK1-NEXT:    [[OMP_ARRAYCPY_DESTELEMENTPAST:%.*]] = phi ptr [ [[ARRAY_BEGIN]], [[ENTRY]] ], [ [[OMP_ARRAYCPY_DEST_ELEMENT:%.*]], [[OMP_ARRAYCPY_BODY]] ]
// CHECK1-NEXT:    call void @_ZN2StC1Ev(ptr noundef nonnull align 4 dereferenceable(8) [[AGG_TMP]])
// CHECK1-NEXT:    call void @_ZN1SIiEC1ERKS0_2St(ptr noundef nonnull align 4 dereferenceable(4) [[OMP_ARRAYCPY_DESTELEMENTPAST]], ptr noundef nonnull align 4 dereferenceable(4) [[OMP_ARRAYCPY_SRCELEMENTPAST]], ptr noundef [[AGG_TMP]])
// CHECK1-NEXT:    call void @_ZN2StD1Ev(ptr noundef nonnull align 4 dereferenceable(8) [[AGG_TMP]]) #[[ATTR2]]
// CHECK1-NEXT:    [[OMP_ARRAYCPY_DEST_ELEMENT]] = getelementptr [[STRUCT_S_0]], ptr [[OMP_ARRAYCPY_DESTELEMENTPAST]], i32 1
// CHECK1-NEXT:    [[OMP_ARRAYCPY_SRC_ELEMENT]] = getelementptr [[STRUCT_S_0]], ptr [[OMP_ARRAYCPY_SRCELEMENTPAST]], i32 1
// CHECK1-NEXT:    [[OMP_ARRAYCPY_DONE:%.*]] = icmp eq ptr [[OMP_ARRAYCPY_DEST_ELEMENT]], [[TMP6]]
// CHECK1-NEXT:    br i1 [[OMP_ARRAYCPY_DONE]], label [[OMP_ARRAYCPY_DONE6]], label [[OMP_ARRAYCPY_BODY]]
// CHECK1:       omp.arraycpy.done6:
// CHECK1-NEXT:    [[TMP7:%.*]] = load ptr, ptr [[_TMP1]], align 8
// CHECK1-NEXT:    call void @_ZN2StC1Ev(ptr noundef nonnull align 4 dereferenceable(8) [[AGG_TMP8]])
// CHECK1-NEXT:    call void @_ZN1SIiEC1ERKS0_2St(ptr noundef nonnull align 4 dereferenceable(4) [[VAR7]], ptr noundef nonnull align 4 dereferenceable(4) [[TMP7]], ptr noundef [[AGG_TMP8]])
// CHECK1-NEXT:    call void @_ZN2StD1Ev(ptr noundef nonnull align 4 dereferenceable(8) [[AGG_TMP8]]) #[[ATTR2]]
// CHECK1-NEXT:    store ptr [[VAR7]], ptr [[_TMP9]], align 8
// CHECK1-NEXT:    [[TMP8:%.*]] = load ptr, ptr [[DOTGLOBAL_TID__ADDR]], align 8
// CHECK1-NEXT:    [[TMP9:%.*]] = load i32, ptr [[TMP8]], align 4
// CHECK1-NEXT:    call void @__kmpc_for_static_init_4(ptr @[[GLOB1]], i32 [[TMP9]], i32 34, ptr [[DOTOMP_IS_LAST]], ptr [[DOTOMP_LB]], ptr [[DOTOMP_UB]], ptr [[DOTOMP_STRIDE]], i32 1, i32 1)
// CHECK1-NEXT:    [[TMP10:%.*]] = load i32, ptr [[DOTOMP_UB]], align 4
// CHECK1-NEXT:    [[CMP:%.*]] = icmp sgt i32 [[TMP10]], 1
// CHECK1-NEXT:    br i1 [[CMP]], label [[COND_TRUE:%.*]], label [[COND_FALSE:%.*]]
// CHECK1:       cond.true:
// CHECK1-NEXT:    br label [[COND_END:%.*]]
// CHECK1:       cond.false:
// CHECK1-NEXT:    [[TMP11:%.*]] = load i32, ptr [[DOTOMP_UB]], align 4
// CHECK1-NEXT:    br label [[COND_END]]
// CHECK1:       cond.end:
// CHECK1-NEXT:    [[COND:%.*]] = phi i32 [ 1, [[COND_TRUE]] ], [ [[TMP11]], [[COND_FALSE]] ]
// CHECK1-NEXT:    store i32 [[COND]], ptr [[DOTOMP_UB]], align 4
// CHECK1-NEXT:    [[TMP12:%.*]] = load i32, ptr [[DOTOMP_LB]], align 4
// CHECK1-NEXT:    store i32 [[TMP12]], ptr [[DOTOMP_IV]], align 4
// CHECK1-NEXT:    br label [[OMP_INNER_FOR_COND:%.*]]
// CHECK1:       omp.inner.for.cond:
// CHECK1-NEXT:    [[TMP13:%.*]] = load i32, ptr [[DOTOMP_IV]], align 4
// CHECK1-NEXT:    [[TMP14:%.*]] = load i32, ptr [[DOTOMP_UB]], align 4
// CHECK1-NEXT:    [[CMP10:%.*]] = icmp sle i32 [[TMP13]], [[TMP14]]
// CHECK1-NEXT:    br i1 [[CMP10]], label [[OMP_INNER_FOR_BODY:%.*]], label [[OMP_INNER_FOR_COND_CLEANUP:%.*]]
// CHECK1:       omp.inner.for.cond.cleanup:
// CHECK1-NEXT:    br label [[OMP_INNER_FOR_END:%.*]]
// CHECK1:       omp.inner.for.body:
// CHECK1-NEXT:    [[TMP15:%.*]] = load i32, ptr [[DOTOMP_IV]], align 4
// CHECK1-NEXT:    [[MUL:%.*]] = mul nsw i32 [[TMP15]], 1
// CHECK1-NEXT:    [[ADD:%.*]] = add nsw i32 0, [[MUL]]
// CHECK1-NEXT:    store i32 [[ADD]], ptr [[I]], align 4
// CHECK1-NEXT:    [[TMP16:%.*]] = load i32, ptr [[T_VAR3]], align 4
// CHECK1-NEXT:    [[TMP17:%.*]] = load i32, ptr [[I]], align 4
// CHECK1-NEXT:    [[IDXPROM:%.*]] = sext i32 [[TMP17]] to i64
// CHECK1-NEXT:    [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x i32], ptr [[VEC4]], i64 0, i64 [[IDXPROM]]
// CHECK1-NEXT:    store i32 [[TMP16]], ptr [[ARRAYIDX]], align 4
// CHECK1-NEXT:    [[TMP18:%.*]] = load ptr, ptr [[_TMP9]], align 8
// CHECK1-NEXT:    [[TMP19:%.*]] = load i32, ptr [[I]], align 4
// CHECK1-NEXT:    [[IDXPROM11:%.*]] = sext i32 [[TMP19]] to i64
// CHECK1-NEXT:    [[ARRAYIDX12:%.*]] = getelementptr inbounds [2 x %struct.S.0], ptr [[S_ARR5]], i64 0, i64 [[IDXPROM11]]
// CHECK1-NEXT:    call void @llvm.memcpy.p0.p0.i64(ptr align 4 [[ARRAYIDX12]], ptr align 4 [[TMP18]], i64 4, i1 false)
// CHECK1-NEXT:    br label [[OMP_BODY_CONTINUE:%.*]]
// CHECK1:       omp.body.continue:
// CHECK1-NEXT:    br label [[OMP_INNER_FOR_INC:%.*]]
// CHECK1:       omp.inner.for.inc:
// CHECK1-NEXT:    [[TMP20:%.*]] = load i32, ptr [[DOTOMP_IV]], align 4
// CHECK1-NEXT:    [[ADD13:%.*]] = add nsw i32 [[TMP20]], 1
// CHECK1-NEXT:    store i32 [[ADD13]], ptr [[DOTOMP_IV]], align 4
// CHECK1-NEXT:    br label [[OMP_INNER_FOR_COND]]
// CHECK1:       omp.inner.for.end:
// CHECK1-NEXT:    br label [[OMP_LOOP_EXIT:%.*]]
// CHECK1:       omp.loop.exit:
// CHECK1-NEXT:    [[TMP21:%.*]] = load ptr, ptr [[DOTGLOBAL_TID__ADDR]], align 8
// CHECK1-NEXT:    [[TMP22:%.*]] = load i32, ptr [[TMP21]], align 4
// CHECK1-NEXT:    call void @__kmpc_for_static_fini(ptr @[[GLOB1]], i32 [[TMP22]])
// CHECK1-NEXT:    call void @_ZN1SIiED1Ev(ptr noundef nonnull align 4 dereferenceable(4) [[VAR7]]) #[[ATTR2]]
// CHECK1-NEXT:    [[ARRAY_BEGIN14:%.*]] = getelementptr inbounds [2 x %struct.S.0], ptr [[S_ARR5]], i32 0, i32 0
// CHECK1-NEXT:    [[TMP23:%.*]] = getelementptr inbounds [[STRUCT_S_0]], ptr [[ARRAY_BEGIN14]], i64 2
// CHECK1-NEXT:    br label [[ARRAYDESTROY_BODY:%.*]]
// CHECK1:       arraydestroy.body:
// CHECK1-NEXT:    [[ARRAYDESTROY_ELEMENTPAST:%.*]] = phi ptr [ [[TMP23]], [[OMP_LOOP_EXIT]] ], [ [[ARRAYDESTROY_ELEMENT:%.*]], [[ARRAYDESTROY_BODY]] ]
// CHECK1-NEXT:    [[ARRAYDESTROY_ELEMENT]] = getelementptr inbounds [[STRUCT_S_0]], ptr [[ARRAYDESTROY_ELEMENTPAST]], i64 -1
// CHECK1-NEXT:    call void @_ZN1SIiED1Ev(ptr noundef nonnull align 4 dereferenceable(4) [[ARRAYDESTROY_ELEMENT]]) #[[ATTR2]]
// CHECK1-NEXT:    [[ARRAYDESTROY_DONE:%.*]] = icmp eq ptr [[ARRAYDESTROY_ELEMENT]], [[ARRAY_BEGIN14]]
// CHECK1-NEXT:    br i1 [[ARRAYDESTROY_DONE]], label [[ARRAYDESTROY_DONE15:%.*]], label [[ARRAYDESTROY_BODY]]
// CHECK1:       arraydestroy.done15:
// CHECK1-NEXT:    [[TMP24:%.*]] = load ptr, ptr [[DOTGLOBAL_TID__ADDR]], align 8
// CHECK1-NEXT:    [[TMP25:%.*]] = load i32, ptr [[TMP24]], align 4
// CHECK1-NEXT:    call void @__kmpc_barrier(ptr @[[GLOB3]], i32 [[TMP25]])
// CHECK1-NEXT:    ret void
//
//
// CHECK1-LABEL: define {{[^@]+}}@_ZN1SIiEC1ERKS0_2St
// CHECK1-SAME: (ptr noundef nonnull align 4 dereferenceable(4) [[THIS:%.*]], ptr noundef nonnull align 4 dereferenceable(4) [[S:%.*]], ptr noundef [[T:%.*]]) unnamed_addr #[[ATTR1]] align 2 {
// CHECK1-NEXT:  entry:
// CHECK1-NEXT:    [[THIS_ADDR:%.*]] = alloca ptr, align 8
// CHECK1-NEXT:    [[S_ADDR:%.*]] = alloca ptr, align 8
// CHECK1-NEXT:    [[T_INDIRECT_ADDR:%.*]] = alloca ptr, align 8
// CHECK1-NEXT:    store ptr [[THIS]], ptr [[THIS_ADDR]], align 8
// CHECK1-NEXT:    store ptr [[S]], ptr [[S_ADDR]], align 8
// CHECK1-NEXT:    store ptr [[T]], ptr [[T_INDIRECT_ADDR]], align 8
// CHECK1-NEXT:    [[THIS1:%.*]] = load ptr, ptr [[THIS_ADDR]], align 8
// CHECK1-NEXT:    [[TMP0:%.*]] = load ptr, ptr [[S_ADDR]], align 8
// CHECK1-NEXT:    call void @_ZN1SIiEC2ERKS0_2St(ptr noundef nonnull align 4 dereferenceable(4) [[THIS1]], ptr noundef nonnull align 4 dereferenceable(4) [[TMP0]], ptr noundef [[T]])
// CHECK1-NEXT:    ret void
//
//
// CHECK1-LABEL: define {{[^@]+}}@_ZN1SIiED1Ev
// CHECK1-SAME: (ptr noundef nonnull align 4 dereferenceable(4) [[THIS:%.*]]) unnamed_addr #[[ATTR1]] align 2 {
// CHECK1-NEXT:  entry:
// CHECK1-NEXT:    [[THIS_ADDR:%.*]] = alloca ptr, align 8
// CHECK1-NEXT:    store ptr [[THIS]], ptr [[THIS_ADDR]], align 8
// CHECK1-NEXT:    [[THIS1:%.*]] = load ptr, ptr [[THIS_ADDR]], align 8
// CHECK1-NEXT:    call void @_ZN1SIiED2Ev(ptr noundef nonnull align 4 dereferenceable(4) [[THIS1]]) #[[ATTR2]]
// CHECK1-NEXT:    ret void
//
//
// CHECK1-LABEL: define {{[^@]+}}@_ZN1SIiEC2Ev
// CHECK1-SAME: (ptr noundef nonnull align 4 dereferenceable(4) [[THIS:%.*]]) unnamed_addr #[[ATTR1]] align 2 {
// CHECK1-NEXT:  entry:
// CHECK1-NEXT:    [[THIS_ADDR:%.*]] = alloca ptr, align 8
// CHECK1-NEXT:    store ptr [[THIS]], ptr [[THIS_ADDR]], align 8
// CHECK1-NEXT:    [[THIS1:%.*]] = load ptr, ptr [[THIS_ADDR]], align 8
// CHECK1-NEXT:    [[F:%.*]] = getelementptr inbounds nuw [[STRUCT_S_0:%.*]], ptr [[THIS1]], i32 0, i32 0
// CHECK1-NEXT:    [[TMP0:%.*]] = load volatile i32, ptr @g, align 4
// CHECK1-NEXT:    store i32 [[TMP0]], ptr [[F]], align 4
// CHECK1-NEXT:    ret void
//
//
// CHECK1-LABEL: define {{[^@]+}}@_ZN1SIiEC2Ei
// CHECK1-SAME: (ptr noundef nonnull align 4 dereferenceable(4) [[THIS:%.*]], i32 noundef [[A:%.*]]) unnamed_addr #[[ATTR1]] align 2 {
// CHECK1-NEXT:  entry:
// CHECK1-NEXT:    [[THIS_ADDR:%.*]] = alloca ptr, align 8
// CHECK1-NEXT:    [[A_ADDR:%.*]] = alloca i32, align 4
// CHECK1-NEXT:    store ptr [[THIS]], ptr [[THIS_ADDR]], align 8
// CHECK1-NEXT:    store i32 [[A]], ptr [[A_ADDR]], align 4
// CHECK1-NEXT:    [[THIS1:%.*]] = load ptr, ptr [[THIS_ADDR]], align 8
// CHECK1-NEXT:    [[F:%.*]] = getelementptr inbounds nuw [[STRUCT_S_0:%.*]], ptr [[THIS1]], i32 0, i32 0
// CHECK1-NEXT:    [[TMP0:%.*]] = load i32, ptr [[A_ADDR]], align 4
// CHECK1-NEXT:    [[TMP1:%.*]] = load volatile i32, ptr @g, align 4
// CHECK1-NEXT:    [[ADD:%.*]] = add nsw i32 [[TMP0]], [[TMP1]]
// CHECK1-NEXT:    store i32 [[ADD]], ptr [[F]], align 4
// CHECK1-NEXT:    ret void
//
//
// CHECK1-LABEL: define {{[^@]+}}@_ZN1SIiEC2ERKS0_2St
// CHECK1-SAME: (ptr noundef nonnull align 4 dereferenceable(4) [[THIS:%.*]], ptr noundef nonnull align 4 dereferenceable(4) [[S:%.*]], ptr noundef [[T:%.*]]) unnamed_addr #[[ATTR1]] align 2 {
// CHECK1-NEXT:  entry:
// CHECK1-NEXT:    [[THIS_ADDR:%.*]] = alloca ptr, align 8
// CHECK1-NEXT:    [[S_ADDR:%.*]] = alloca ptr, align 8
// CHECK1-NEXT:    [[T_INDIRECT_ADDR:%.*]] = alloca ptr, align 8
// CHECK1-NEXT:    store ptr [[THIS]], ptr [[THIS_ADDR]], align 8
// CHECK1-NEXT:    store ptr [[S]], ptr [[S_ADDR]], align 8
// CHECK1-NEXT:    store ptr [[T]], ptr [[T_INDIRECT_ADDR]], align 8
// CHECK1-NEXT:    [[THIS1:%.*]] = load ptr, ptr [[THIS_ADDR]], align 8
// CHECK1-NEXT:    [[F:%.*]] = getelementptr inbounds nuw [[STRUCT_S_0:%.*]], ptr [[THIS1]], i32 0, i32 0
// CHECK1-NEXT:    [[TMP0:%.*]] = load ptr, ptr [[S_ADDR]], align 8
// CHECK1-NEXT:    [[F2:%.*]] = getelementptr inbounds nuw [[STRUCT_S_0]], ptr [[TMP0]], i32 0, i32 0
// CHECK1-NEXT:    [[TMP1:%.*]] = load i32, ptr [[F2]], align 4
// CHECK1-NEXT:    [[A:%.*]] = getelementptr inbounds nuw [[STRUCT_ST:%.*]], ptr [[T]], i32 0, i32 0
// CHECK1-NEXT:    [[TMP2:%.*]] = load i32, ptr [[A]], align 4
// CHECK1-NEXT:    [[ADD:%.*]] = add nsw i32 [[TMP1]], [[TMP2]]
// CHECK1-NEXT:    store i32 [[ADD]], ptr [[F]], align 4
// CHECK1-NEXT:    ret void
//
//
// CHECK1-LABEL: define {{[^@]+}}@_ZN1SIiED2Ev
// CHECK1-SAME: (ptr noundef nonnull align 4 dereferenceable(4) [[THIS:%.*]]) unnamed_addr #[[ATTR1]] align 2 {
// CHECK1-NEXT:  entry:
// CHECK1-NEXT:    [[THIS_ADDR:%.*]] = alloca ptr, align 8
// CHECK1-NEXT:    store ptr [[THIS]], ptr [[THIS_ADDR]], align 8
// CHECK1-NEXT:    [[THIS1:%.*]] = load ptr, ptr [[THIS_ADDR]], align 8
// CHECK1-NEXT:    ret void
//
//
// CHECK1-LABEL: define {{[^@]+}}@_GLOBAL__sub_I_for_firstprivate_codegen.cpp
// CHECK1-SAME: () #[[ATTR0]] section "__TEXT,__StaticInit,regular,pure_instructions" {
// CHECK1-NEXT:  entry:
// CHECK1-NEXT:    call void @__cxx_global_var_init()
// CHECK1-NEXT:    call void @__cxx_global_var_init.1()
// CHECK1-NEXT:    call void @__cxx_global_var_init.2()
// CHECK1-NEXT:    ret void
//
//
// CHECK3-LABEL: define {{[^@]+}}@__cxx_global_var_init
// CHECK3-SAME: () #[[ATTR0:[0-9]+]] section "__TEXT,__StaticInit,regular,pure_instructions" {
// CHECK3-NEXT:  entry:
// CHECK3-NEXT:    call void @_ZN1SIfEC1Ev(ptr noundef nonnull align 4 dereferenceable(4) @test)
// CHECK3-NEXT:    [[TMP0:%.*]] = call i32 @__cxa_atexit(ptr @_ZN1SIfED1Ev, ptr @test, ptr @__dso_handle) #[[ATTR2:[0-9]+]]
// CHECK3-NEXT:    ret void
//
//
// CHECK3-LABEL: define {{[^@]+}}@_ZN1SIfEC1Ev
// CHECK3-SAME: (ptr noundef nonnull align 4 dereferenceable(4) [[THIS:%.*]]) unnamed_addr #[[ATTR1:[0-9]+]] align 2 {
// CHECK3-NEXT:  entry:
// CHECK3-NEXT:    [[THIS_ADDR:%.*]] = alloca ptr, align 8
// CHECK3-NEXT:    store ptr [[THIS]], ptr [[THIS_ADDR]], align 8
// CHECK3-NEXT:    [[THIS1:%.*]] = load ptr, ptr [[THIS_ADDR]], align 8
// CHECK3-NEXT:    call void @_ZN1SIfEC2Ev(ptr noundef nonnull align 4 dereferenceable(4) [[THIS1]])
// CHECK3-NEXT:    ret void
//
//
// CHECK3-LABEL: define {{[^@]+}}@_ZN1SIfED1Ev
// CHECK3-SAME: (ptr noundef nonnull align 4 dereferenceable(4) [[THIS:%.*]]) unnamed_addr #[[ATTR1]] align 2 {
// CHECK3-NEXT:  entry:
// CHECK3-NEXT:    [[THIS_ADDR:%.*]] = alloca ptr, align 8
// CHECK3-NEXT:    store ptr [[THIS]], ptr [[THIS_ADDR]], align 8
// CHECK3-NEXT:    [[THIS1:%.*]] = load ptr, ptr [[THIS_ADDR]], align 8
// CHECK3-NEXT:    call void @_ZN1SIfED2Ev(ptr noundef nonnull align 4 dereferenceable(4) [[THIS1]]) #[[ATTR2]]
// CHECK3-NEXT:    ret void
//
//
// CHECK3-LABEL: define {{[^@]+}}@_ZN1SIfEC2Ev
// CHECK3-SAME: (ptr noundef nonnull align 4 dereferenceable(4) [[THIS:%.*]]) unnamed_addr #[[ATTR1]] align 2 {
// CHECK3-NEXT:  entry:
// CHECK3-NEXT:    [[THIS_ADDR:%.*]] = alloca ptr, align 8
// CHECK3-NEXT:    store ptr [[THIS]], ptr [[THIS_ADDR]], align 8
// CHECK3-NEXT:    [[THIS1:%.*]] = load ptr, ptr [[THIS_ADDR]], align 8
// CHECK3-NEXT:    [[F:%.*]] = getelementptr inbounds nuw [[STRUCT_S:%.*]], ptr [[THIS1]], i32 0, i32 0
// CHECK3-NEXT:    [[TMP0:%.*]] = load volatile i32, ptr @g, align 4
// CHECK3-NEXT:    [[CONV:%.*]] = sitofp i32 [[TMP0]] to float
// CHECK3-NEXT:    store float [[CONV]], ptr [[F]], align 4
// CHECK3-NEXT:    ret void
//
//
// CHECK3-LABEL: define {{[^@]+}}@_ZN1SIfED2Ev
// CHECK3-SAME: (ptr noundef nonnull align 4 dereferenceable(4) [[THIS:%.*]]) unnamed_addr #[[ATTR1]] align 2 {
// CHECK3-NEXT:  entry:
// CHECK3-NEXT:    [[THIS_ADDR:%.*]] = alloca ptr, align 8
// CHECK3-NEXT:    store ptr [[THIS]], ptr [[THIS_ADDR]], align 8
// CHECK3-NEXT:    [[THIS1:%.*]] = load ptr, ptr [[THIS_ADDR]], align 8
// CHECK3-NEXT:    ret void
//
//
// CHECK3-LABEL: define {{[^@]+}}@__cxx_global_var_init.1
// CHECK3-SAME: () #[[ATTR0]] section "__TEXT,__StaticInit,regular,pure_instructions" {
// CHECK3-NEXT:  entry:
// CHECK3-NEXT:    call void @_ZN1SIfEC1Ef(ptr noundef nonnull align 4 dereferenceable(4) @s_arr, float noundef 1.000000e+00)
// CHECK3-NEXT:    call void @_ZN1SIfEC1Ef(ptr noundef nonnull align 4 dereferenceable(4) getelementptr inbounds ([[STRUCT_S:%.*]], ptr @s_arr, i64 1), float noundef 2.000000e+00)
// CHECK3-NEXT:    [[TMP0:%.*]] = call i32 @__cxa_atexit(ptr @__cxx_global_array_dtor, ptr null, ptr @__dso_handle) #[[ATTR2]]
// CHECK3-NEXT:    ret void
//
//
// CHECK3-LABEL: define {{[^@]+}}@_ZN1SIfEC1Ef
// CHECK3-SAME: (ptr noundef nonnull align 4 dereferenceable(4) [[THIS:%.*]], float noundef [[A:%.*]]) unnamed_addr #[[ATTR1]] align 2 {
// CHECK3-NEXT:  entry:
// CHECK3-NEXT:    [[THIS_ADDR:%.*]] = alloca ptr, align 8
// CHECK3-NEXT:    [[A_ADDR:%.*]] = alloca float, align 4
// CHECK3-NEXT:    store ptr [[THIS]], ptr [[THIS_ADDR]], align 8
// CHECK3-NEXT:    store float [[A]], ptr [[A_ADDR]], align 4
// CHECK3-NEXT:    [[THIS1:%.*]] = load ptr, ptr [[THIS_ADDR]], align 8
// CHECK3-NEXT:    [[TMP0:%.*]] = load float, ptr [[A_ADDR]], align 4
// CHECK3-NEXT:    call void @_ZN1SIfEC2Ef(ptr noundef nonnull align 4 dereferenceable(4) [[THIS1]], float noundef [[TMP0]])
// CHECK3-NEXT:    ret void
//
//
// CHECK3-LABEL: define {{[^@]+}}@__cxx_global_array_dtor
// CHECK3-SAME: (ptr noundef [[TMP0:%.*]]) #[[ATTR0]] section "__TEXT,__StaticInit,regular,pure_instructions" {
// CHECK3-NEXT:  entry:
// CHECK3-NEXT:    [[DOTADDR:%.*]] = alloca ptr, align 8
// CHECK3-NEXT:    store ptr [[TMP0]], ptr [[DOTADDR]], align 8
// CHECK3-NEXT:    br label [[ARRAYDESTROY_BODY:%.*]]
// CHECK3:       arraydestroy.body:
// CHECK3-NEXT:    [[ARRAYDESTROY_ELEMENTPAST:%.*]] = phi ptr [ getelementptr inbounds ([[STRUCT_S:%.*]], ptr @s_arr, i64 2), [[ENTRY:%.*]] ], [ [[ARRAYDESTROY_ELEMENT:%.*]], [[ARRAYDESTROY_BODY]] ]
// CHECK3-NEXT:    [[ARRAYDESTROY_ELEMENT]] = getelementptr inbounds [[STRUCT_S]], ptr [[ARRAYDESTROY_ELEMENTPAST]], i64 -1
// CHECK3-NEXT:    call void @_ZN1SIfED1Ev(ptr noundef nonnull align 4 dereferenceable(4) [[ARRAYDESTROY_ELEMENT]]) #[[ATTR2]]
// CHECK3-NEXT:    [[ARRAYDESTROY_DONE:%.*]] = icmp eq ptr [[ARRAYDESTROY_ELEMENT]], @s_arr
// CHECK3-NEXT:    br i1 [[ARRAYDESTROY_DONE]], label [[ARRAYDESTROY_DONE1:%.*]], label [[ARRAYDESTROY_BODY]]
// CHECK3:       arraydestroy.done1:
// CHECK3-NEXT:    ret void
//
//
// CHECK3-LABEL: define {{[^@]+}}@_ZN1SIfEC2Ef
// CHECK3-SAME: (ptr noundef nonnull align 4 dereferenceable(4) [[THIS:%.*]], float noundef [[A:%.*]]) unnamed_addr #[[ATTR1]] align 2 {
// CHECK3-NEXT:  entry:
// CHECK3-NEXT:    [[THIS_ADDR:%.*]] = alloca ptr, align 8
// CHECK3-NEXT:    [[A_ADDR:%.*]] = alloca float, align 4
// CHECK3-NEXT:    store ptr [[THIS]], ptr [[THIS_ADDR]], align 8
// CHECK3-NEXT:    store float [[A]], ptr [[A_ADDR]], align 4
// CHECK3-NEXT:    [[THIS1:%.*]] = load ptr, ptr [[THIS_ADDR]], align 8
// CHECK3-NEXT:    [[F:%.*]] = getelementptr inbounds nuw [[STRUCT_S:%.*]], ptr [[THIS1]], i32 0, i32 0
// CHECK3-NEXT:    [[TMP0:%.*]] = load float, ptr [[A_ADDR]], align 4
// CHECK3-NEXT:    [[TMP1:%.*]] = load volatile i32, ptr @g, align 4
// CHECK3-NEXT:    [[CONV:%.*]] = sitofp i32 [[TMP1]] to float
// CHECK3-NEXT:    [[ADD:%.*]] = fadd float [[TMP0]], [[CONV]]
// CHECK3-NEXT:    store float [[ADD]], ptr [[F]], align 4
// CHECK3-NEXT:    ret void
//
//
// CHECK3-LABEL: define {{[^@]+}}@__cxx_global_var_init.2
// CHECK3-SAME: () #[[ATTR0]] section "__TEXT,__StaticInit,regular,pure_instructions" {
// CHECK3-NEXT:  entry:
// CHECK3-NEXT:    call void @_ZN1SIfEC1Ef(ptr noundef nonnull align 4 dereferenceable(4) @var, float noundef 3.000000e+00)
// CHECK3-NEXT:    [[TMP0:%.*]] = call i32 @__cxa_atexit(ptr @_ZN1SIfED1Ev, ptr @var, ptr @__dso_handle) #[[ATTR2]]
// CHECK3-NEXT:    ret void
//
//
// CHECK3-LABEL: define {{[^@]+}}@main
// CHECK3-SAME: () #[[ATTR3:[0-9]+]] {
// CHECK3-NEXT:  entry:
// CHECK3-NEXT:    [[RETVAL:%.*]] = alloca i32, align 4
// CHECK3-NEXT:    [[REF_TMP:%.*]] = alloca [[CLASS_ANON:%.*]], align 8
// CHECK3-NEXT:    store i32 0, ptr [[RETVAL]], align 4
// CHECK3-NEXT:    [[TMP0:%.*]] = getelementptr inbounds nuw [[CLASS_ANON]], ptr [[REF_TMP]], i32 0, i32 0
// CHECK3-NEXT:    store ptr @_ZZ4mainE5sivar, ptr [[TMP0]], align 8
// CHECK3-NEXT:    call void @"_ZZ4mainENK3$_0clEv"(ptr noundef nonnull align 8 dereferenceable(8) [[REF_TMP]])
// CHECK3-NEXT:    ret i32 0
//
//
// CHECK3-LABEL: define {{[^@]+}}@_GLOBAL__sub_I_for_firstprivate_codegen.cpp
// CHECK3-SAME: () #[[ATTR0]] section "__TEXT,__StaticInit,regular,pure_instructions" {
// CHECK3-NEXT:  entry:
// CHECK3-NEXT:    call void @__cxx_global_var_init()
// CHECK3-NEXT:    call void @__cxx_global_var_init.1()
// CHECK3-NEXT:    call void @__cxx_global_var_init.2()
// CHECK3-NEXT:    ret void
//
//
// CHECK4-LABEL: define {{[^@]+}}@__cxx_global_var_init
// CHECK4-SAME: () #[[ATTR0:[0-9]+]] section "__TEXT,__StaticInit,regular,pure_instructions" {
// CHECK4-NEXT:  entry:
// CHECK4-NEXT:    call void @_ZN1SIfEC1Ev(ptr noundef nonnull align 4 dereferenceable(4) @test)
// CHECK4-NEXT:    [[TMP0:%.*]] = call i32 @__cxa_atexit(ptr @_ZN1SIfED1Ev, ptr @test, ptr @__dso_handle) #[[ATTR2:[0-9]+]]
// CHECK4-NEXT:    ret void
//
//
// CHECK4-LABEL: define {{[^@]+}}@_ZN1SIfEC1Ev
// CHECK4-SAME: (ptr noundef nonnull align 4 dereferenceable(4) [[THIS:%.*]]) unnamed_addr #[[ATTR1:[0-9]+]] align 2 {
// CHECK4-NEXT:  entry:
// CHECK4-NEXT:    [[THIS_ADDR:%.*]] = alloca ptr, align 8
// CHECK4-NEXT:    store ptr [[THIS]], ptr [[THIS_ADDR]], align 8
// CHECK4-NEXT:    [[THIS1:%.*]] = load ptr, ptr [[THIS_ADDR]], align 8
// CHECK4-NEXT:    call void @_ZN1SIfEC2Ev(ptr noundef nonnull align 4 dereferenceable(4) [[THIS1]])
// CHECK4-NEXT:    ret void
//
//
// CHECK4-LABEL: define {{[^@]+}}@_ZN1SIfED1Ev
// CHECK4-SAME: (ptr noundef nonnull align 4 dereferenceable(4) [[THIS:%.*]]) unnamed_addr #[[ATTR1]] align 2 {
// CHECK4-NEXT:  entry:
// CHECK4-NEXT:    [[THIS_ADDR:%.*]] = alloca ptr, align 8
// CHECK4-NEXT:    store ptr [[THIS]], ptr [[THIS_ADDR]], align 8
// CHECK4-NEXT:    [[THIS1:%.*]] = load ptr, ptr [[THIS_ADDR]], align 8
// CHECK4-NEXT:    call void @_ZN1SIfED2Ev(ptr noundef nonnull align 4 dereferenceable(4) [[THIS1]]) #[[ATTR2]]
// CHECK4-NEXT:    ret void
//
//
// CHECK4-LABEL: define {{[^@]+}}@_ZN1SIfEC2Ev
// CHECK4-SAME: (ptr noundef nonnull align 4 dereferenceable(4) [[THIS:%.*]]) unnamed_addr #[[ATTR1]] align 2 {
// CHECK4-NEXT:  entry:
// CHECK4-NEXT:    [[THIS_ADDR:%.*]] = alloca ptr, align 8
// CHECK4-NEXT:    store ptr [[THIS]], ptr [[THIS_ADDR]], align 8
// CHECK4-NEXT:    [[THIS1:%.*]] = load ptr, ptr [[THIS_ADDR]], align 8
// CHECK4-NEXT:    [[F:%.*]] = getelementptr inbounds nuw [[STRUCT_S:%.*]], ptr [[THIS1]], i32 0, i32 0
// CHECK4-NEXT:    [[TMP0:%.*]] = load volatile i32, ptr @g, align 4
// CHECK4-NEXT:    [[CONV:%.*]] = sitofp i32 [[TMP0]] to float
// CHECK4-NEXT:    store float [[CONV]], ptr [[F]], align 4
// CHECK4-NEXT:    ret void
//
//
// CHECK4-LABEL: define {{[^@]+}}@_ZN1SIfED2Ev
// CHECK4-SAME: (ptr noundef nonnull align 4 dereferenceable(4) [[THIS:%.*]]) unnamed_addr #[[ATTR1]] align 2 {
// CHECK4-NEXT:  entry:
// CHECK4-NEXT:    [[THIS_ADDR:%.*]] = alloca ptr, align 8
// CHECK4-NEXT:    store ptr [[THIS]], ptr [[THIS_ADDR]], align 8
// CHECK4-NEXT:    [[THIS1:%.*]] = load ptr, ptr [[THIS_ADDR]], align 8
// CHECK4-NEXT:    ret void
//
//
// CHECK4-LABEL: define {{[^@]+}}@__cxx_global_var_init.1
// CHECK4-SAME: () #[[ATTR0]] section "__TEXT,__StaticInit,regular,pure_instructions" {
// CHECK4-NEXT:  entry:
// CHECK4-NEXT:    call void @_ZN1SIfEC1Ef(ptr noundef nonnull align 4 dereferenceable(4) @s_arr, float noundef 1.000000e+00)
// CHECK4-NEXT:    call void @_ZN1SIfEC1Ef(ptr noundef nonnull align 4 dereferenceable(4) getelementptr inbounds ([[STRUCT_S:%.*]], ptr @s_arr, i64 1), float noundef 2.000000e+00)
// CHECK4-NEXT:    [[TMP0:%.*]] = call i32 @__cxa_atexit(ptr @__cxx_global_array_dtor, ptr null, ptr @__dso_handle) #[[ATTR2]]
// CHECK4-NEXT:    ret void
//
//
// CHECK4-LABEL: define {{[^@]+}}@_ZN1SIfEC1Ef
// CHECK4-SAME: (ptr noundef nonnull align 4 dereferenceable(4) [[THIS:%.*]], float noundef [[A:%.*]]) unnamed_addr #[[ATTR1]] align 2 {
// CHECK4-NEXT:  entry:
// CHECK4-NEXT:    [[THIS_ADDR:%.*]] = alloca ptr, align 8
// CHECK4-NEXT:    [[A_ADDR:%.*]] = alloca float, align 4
// CHECK4-NEXT:    store ptr [[THIS]], ptr [[THIS_ADDR]], align 8
// CHECK4-NEXT:    store float [[A]], ptr [[A_ADDR]], align 4
// CHECK4-NEXT:    [[THIS1:%.*]] = load ptr, ptr [[THIS_ADDR]], align 8
// CHECK4-NEXT:    [[TMP0:%.*]] = load float, ptr [[A_ADDR]], align 4
// CHECK4-NEXT:    call void @_ZN1SIfEC2Ef(ptr noundef nonnull align 4 dereferenceable(4) [[THIS1]], float noundef [[TMP0]])
// CHECK4-NEXT:    ret void
//
//
// CHECK4-LABEL: define {{[^@]+}}@__cxx_global_array_dtor
// CHECK4-SAME: (ptr noundef [[TMP0:%.*]]) #[[ATTR0]] section "__TEXT,__StaticInit,regular,pure_instructions" {
// CHECK4-NEXT:  entry:
// CHECK4-NEXT:    [[DOTADDR:%.*]] = alloca ptr, align 8
// CHECK4-NEXT:    store ptr [[TMP0]], ptr [[DOTADDR]], align 8
// CHECK4-NEXT:    br label [[ARRAYDESTROY_BODY:%.*]]
// CHECK4:       arraydestroy.body:
// CHECK4-NEXT:    [[ARRAYDESTROY_ELEMENTPAST:%.*]] = phi ptr [ getelementptr inbounds ([[STRUCT_S:%.*]], ptr @s_arr, i64 2), [[ENTRY:%.*]] ], [ [[ARRAYDESTROY_ELEMENT:%.*]], [[ARRAYDESTROY_BODY]] ]
// CHECK4-NEXT:    [[ARRAYDESTROY_ELEMENT]] = getelementptr inbounds [[STRUCT_S]], ptr [[ARRAYDESTROY_ELEMENTPAST]], i64 -1
// CHECK4-NEXT:    call void @_ZN1SIfED1Ev(ptr noundef nonnull align 4 dereferenceable(4) [[ARRAYDESTROY_ELEMENT]]) #[[ATTR2]]
// CHECK4-NEXT:    [[ARRAYDESTROY_DONE:%.*]] = icmp eq ptr [[ARRAYDESTROY_ELEMENT]], @s_arr
// CHECK4-NEXT:    br i1 [[ARRAYDESTROY_DONE]], label [[ARRAYDESTROY_DONE1:%.*]], label [[ARRAYDESTROY_BODY]]
// CHECK4:       arraydestroy.done1:
// CHECK4-NEXT:    ret void
//
//
// CHECK4-LABEL: define {{[^@]+}}@_ZN1SIfEC2Ef
// CHECK4-SAME: (ptr noundef nonnull align 4 dereferenceable(4) [[THIS:%.*]], float noundef [[A:%.*]]) unnamed_addr #[[ATTR1]] align 2 {
// CHECK4-NEXT:  entry:
// CHECK4-NEXT:    [[THIS_ADDR:%.*]] = alloca ptr, align 8
// CHECK4-NEXT:    [[A_ADDR:%.*]] = alloca float, align 4
// CHECK4-NEXT:    store ptr [[THIS]], ptr [[THIS_ADDR]], align 8
// CHECK4-NEXT:    store float [[A]], ptr [[A_ADDR]], align 4
// CHECK4-NEXT:    [[THIS1:%.*]] = load ptr, ptr [[THIS_ADDR]], align 8
// CHECK4-NEXT:    [[F:%.*]] = getelementptr inbounds nuw [[STRUCT_S:%.*]], ptr [[THIS1]], i32 0, i32 0
// CHECK4-NEXT:    [[TMP0:%.*]] = load float, ptr [[A_ADDR]], align 4
// CHECK4-NEXT:    [[TMP1:%.*]] = load volatile i32, ptr @g, align 4
// CHECK4-NEXT:    [[CONV:%.*]] = sitofp i32 [[TMP1]] to float
// CHECK4-NEXT:    [[ADD:%.*]] = fadd float [[TMP0]], [[CONV]]
// CHECK4-NEXT:    store float [[ADD]], ptr [[F]], align 4
// CHECK4-NEXT:    ret void
//
//
// CHECK4-LABEL: define {{[^@]+}}@__cxx_global_var_init.2
// CHECK4-SAME: () #[[ATTR0]] section "__TEXT,__StaticInit,regular,pure_instructions" {
// CHECK4-NEXT:  entry:
// CHECK4-NEXT:    call void @_ZN1SIfEC1Ef(ptr noundef nonnull align 4 dereferenceable(4) @var, float noundef 3.000000e+00)
// CHECK4-NEXT:    [[TMP0:%.*]] = call i32 @__cxa_atexit(ptr @_ZN1SIfED1Ev, ptr @var, ptr @__dso_handle) #[[ATTR2]]
// CHECK4-NEXT:    ret void
//
//
// CHECK4-LABEL: define {{[^@]+}}@main
// CHECK4-SAME: () #[[ATTR3:[0-9]+]] {
// CHECK4-NEXT:  entry:
// CHECK4-NEXT:    [[RETVAL:%.*]] = alloca i32, align 4
// CHECK4-NEXT:    [[BLOCK:%.*]] = alloca <{ ptr, i32, i32, ptr, ptr, i32 }>, align 8
// CHECK4-NEXT:    store i32 0, ptr [[RETVAL]], align 4
// CHECK4-NEXT:    [[BLOCK_ISA:%.*]] = getelementptr inbounds nuw <{ ptr, i32, i32, ptr, ptr, i32 }>, ptr [[BLOCK]], i32 0, i32 0
// CHECK4-NEXT:    store ptr @_NSConcreteStackBlock, ptr [[BLOCK_ISA]], align 8
// CHECK4-NEXT:    [[BLOCK_FLAGS:%.*]] = getelementptr inbounds nuw <{ ptr, i32, i32, ptr, ptr, i32 }>, ptr [[BLOCK]], i32 0, i32 1
// CHECK4-NEXT:    store i32 1073741824, ptr [[BLOCK_FLAGS]], align 8
// CHECK4-NEXT:    [[BLOCK_RESERVED:%.*]] = getelementptr inbounds nuw <{ ptr, i32, i32, ptr, ptr, i32 }>, ptr [[BLOCK]], i32 0, i32 2
// CHECK4-NEXT:    store i32 0, ptr [[BLOCK_RESERVED]], align 4
// CHECK4-NEXT:    [[BLOCK_INVOKE:%.*]] = getelementptr inbounds nuw <{ ptr, i32, i32, ptr, ptr, i32 }>, ptr [[BLOCK]], i32 0, i32 3
// CHECK4-NEXT:    store ptr @__main_block_invoke, ptr [[BLOCK_INVOKE]], align 8
// CHECK4-NEXT:    [[BLOCK_DESCRIPTOR:%.*]] = getelementptr inbounds nuw <{ ptr, i32, i32, ptr, ptr, i32 }>, ptr [[BLOCK]], i32 0, i32 4
// CHECK4-NEXT:    store ptr @__block_descriptor_tmp.3, ptr [[BLOCK_DESCRIPTOR]], align 8
// CHECK4-NEXT:    [[BLOCK_CAPTURED:%.*]] = getelementptr inbounds nuw <{ ptr, i32, i32, ptr, ptr, i32 }>, ptr [[BLOCK]], i32 0, i32 5
// CHECK4-NEXT:    [[TMP0:%.*]] = load i32, ptr @_ZZ4mainE5sivar, align 4
// CHECK4-NEXT:    store i32 [[TMP0]], ptr [[BLOCK_CAPTURED]], align 8
// CHECK4-NEXT:    [[TMP1:%.*]] = getelementptr inbounds nuw [[STRUCT___BLOCK_LITERAL_GENERIC:%.*]], ptr [[BLOCK]], i32 0, i32 3
// CHECK4-NEXT:    [[TMP2:%.*]] = load ptr, ptr [[TMP1]], align 8
// CHECK4-NEXT:    call void [[TMP2]](ptr noundef [[BLOCK]])
// CHECK4-NEXT:    ret i32 0
//
//
// CHECK4-LABEL: define {{[^@]+}}@__main_block_invoke
// CHECK4-SAME: (ptr noundef [[DOTBLOCK_DESCRIPTOR:%.*]]) #[[ATTR4:[0-9]+]] {
// CHECK4-NEXT:  entry:
// CHECK4-NEXT:    [[DOTBLOCK_DESCRIPTOR_ADDR:%.*]] = alloca ptr, align 8
// CHECK4-NEXT:    [[BLOCK_ADDR:%.*]] = alloca ptr, align 8
// CHECK4-NEXT:    store ptr [[DOTBLOCK_DESCRIPTOR]], ptr [[DOTBLOCK_DESCRIPTOR_ADDR]], align 8
// CHECK4-NEXT:    store ptr [[DOTBLOCK_DESCRIPTOR]], ptr [[BLOCK_ADDR]], align 8
// CHECK4-NEXT:    call void (ptr, i32, ptr, ...) @__kmpc_fork_call(ptr @[[GLOB3:[0-9]+]], i32 1, ptr @__main_block_invoke.omp_outlined, ptr @_ZZ4mainE5sivar)
// CHECK4-NEXT:    ret void
//
//
// CHECK4-LABEL: define {{[^@]+}}@__main_block_invoke.omp_outlined
// CHECK4-SAME: (ptr noalias noundef [[DOTGLOBAL_TID_:%.*]], ptr noalias noundef [[DOTBOUND_TID_:%.*]], ptr noundef nonnull align 4 dereferenceable(4) [[SIVAR:%.*]]) #[[ATTR5:[0-9]+]] {
// CHECK4-NEXT:  entry:
// CHECK4-NEXT:    [[DOTGLOBAL_TID__ADDR:%.*]] = alloca ptr, align 8
// CHECK4-NEXT:    [[DOTBOUND_TID__ADDR:%.*]] = alloca ptr, align 8
// CHECK4-NEXT:    [[SIVAR_ADDR:%.*]] = alloca ptr, align 8
// CHECK4-NEXT:    [[TMP:%.*]] = alloca ptr, align 8
// CHECK4-NEXT:    [[DOTOMP_IV:%.*]] = alloca i32, align 4
// CHECK4-NEXT:    [[_TMP1:%.*]] = alloca i32, align 4
// CHECK4-NEXT:    [[DOTOMP_LB:%.*]] = alloca i32, align 4
// CHECK4-NEXT:    [[DOTOMP_UB:%.*]] = alloca i32, align 4
// CHECK4-NEXT:    [[DOTOMP_STRIDE:%.*]] = alloca i32, align 4
// CHECK4-NEXT:    [[DOTOMP_IS_LAST:%.*]] = alloca i32, align 4
// CHECK4-NEXT:    [[G:%.*]] = alloca i32, align 4
// CHECK4-NEXT:    [[G1:%.*]] = alloca i32, align 4
// CHECK4-NEXT:    [[_TMP2:%.*]] = alloca ptr, align 8
// CHECK4-NEXT:    [[SIVAR3:%.*]] = alloca i32, align 4
// CHECK4-NEXT:    [[I:%.*]] = alloca i32, align 4
// CHECK4-NEXT:    [[BLOCK:%.*]] = alloca <{ ptr, i32, i32, ptr, ptr, ptr, i32, i32 }>, align 8
// CHECK4-NEXT:    store ptr [[DOTGLOBAL_TID_]], ptr [[DOTGLOBAL_TID__ADDR]], align 8
// CHECK4-NEXT:    store ptr [[DOTBOUND_TID_]], ptr [[DOTBOUND_TID__ADDR]], align 8
// CHECK4-NEXT:    store ptr [[SIVAR]], ptr [[SIVAR_ADDR]], align 8
// CHECK4-NEXT:    [[TMP0:%.*]] = load ptr, ptr [[SIVAR_ADDR]], align 8
// CHECK4-NEXT:    [[TMP1:%.*]] = load ptr, ptr @g1, align 8
// CHECK4-NEXT:    store ptr [[TMP1]], ptr [[TMP]], align 8
// CHECK4-NEXT:    store i32 0, ptr [[DOTOMP_LB]], align 4
// CHECK4-NEXT:    store i32 1, ptr [[DOTOMP_UB]], align 4
// CHECK4-NEXT:    store i32 1, ptr [[DOTOMP_STRIDE]], align 4
// CHECK4-NEXT:    store i32 0, ptr [[DOTOMP_IS_LAST]], align 4
// CHECK4-NEXT:    [[TMP2:%.*]] = load volatile i32, ptr @g, align 4
// CHECK4-NEXT:    store i32 [[TMP2]], ptr [[G]], align 4
// CHECK4-NEXT:    [[TMP3:%.*]] = load volatile i32, ptr @g, align 4
// CHECK4-NEXT:    store i32 [[TMP3]], ptr [[G1]], align 4
// CHECK4-NEXT:    store ptr [[G1]], ptr [[_TMP2]], align 8
// CHECK4-NEXT:    [[TMP4:%.*]] = load i32, ptr [[TMP0]], align 4
// CHECK4-NEXT:    store i32 [[TMP4]], ptr [[SIVAR3]], align 4
// CHECK4-NEXT:    [[TMP5:%.*]] = load ptr, ptr [[DOTGLOBAL_TID__ADDR]], align 8
// CHECK4-NEXT:    [[TMP6:%.*]] = load i32, ptr [[TMP5]], align 4
// CHECK4-NEXT:    call void @__kmpc_for_static_init_4(ptr @[[GLOB1:[0-9]+]], i32 [[TMP6]], i32 34, ptr [[DOTOMP_IS_LAST]], ptr [[DOTOMP_LB]], ptr [[DOTOMP_UB]], ptr [[DOTOMP_STRIDE]], i32 1, i32 1)
// CHECK4-NEXT:    [[TMP7:%.*]] = load i32, ptr [[DOTOMP_UB]], align 4
// CHECK4-NEXT:    [[CMP:%.*]] = icmp sgt i32 [[TMP7]], 1
// CHECK4-NEXT:    br i1 [[CMP]], label [[COND_TRUE:%.*]], label [[COND_FALSE:%.*]]
// CHECK4:       cond.true:
// CHECK4-NEXT:    br label [[COND_END:%.*]]
// CHECK4:       cond.false:
// CHECK4-NEXT:    [[TMP8:%.*]] = load i32, ptr [[DOTOMP_UB]], align 4
// CHECK4-NEXT:    br label [[COND_END]]
// CHECK4:       cond.end:
// CHECK4-NEXT:    [[COND:%.*]] = phi i32 [ 1, [[COND_TRUE]] ], [ [[TMP8]], [[COND_FALSE]] ]
// CHECK4-NEXT:    store i32 [[COND]], ptr [[DOTOMP_UB]], align 4
// CHECK4-NEXT:    [[TMP9:%.*]] = load i32, ptr [[DOTOMP_LB]], align 4
// CHECK4-NEXT:    store i32 [[TMP9]], ptr [[DOTOMP_IV]], align 4
// CHECK4-NEXT:    br label [[OMP_INNER_FOR_COND:%.*]]
// CHECK4:       omp.inner.for.cond:
// CHECK4-NEXT:    [[TMP10:%.*]] = load i32, ptr [[DOTOMP_IV]], align 4
// CHECK4-NEXT:    [[TMP11:%.*]] = load i32, ptr [[DOTOMP_UB]], align 4
// CHECK4-NEXT:    [[CMP4:%.*]] = icmp sle i32 [[TMP10]], [[TMP11]]
// CHECK4-NEXT:    br i1 [[CMP4]], label [[OMP_INNER_FOR_BODY:%.*]], label [[OMP_INNER_FOR_END:%.*]]
// CHECK4:       omp.inner.for.body:
// CHECK4-NEXT:    [[TMP12:%.*]] = load i32, ptr [[DOTOMP_IV]], align 4
// CHECK4-NEXT:    [[MUL:%.*]] = mul nsw i32 [[TMP12]], 1
// CHECK4-NEXT:    [[ADD:%.*]] = add nsw i32 0, [[MUL]]
// CHECK4-NEXT:    store i32 [[ADD]], ptr [[I]], align 4
// CHECK4-NEXT:    store i32 1, ptr [[G]], align 4
// CHECK4-NEXT:    [[TMP13:%.*]] = load ptr, ptr [[_TMP2]], align 8
// CHECK4-NEXT:    store volatile i32 1, ptr [[TMP13]], align 4
// CHECK4-NEXT:    store i32 2, ptr [[SIVAR3]], align 4
// CHECK4-NEXT:    [[BLOCK_ISA:%.*]] = getelementptr inbounds nuw <{ ptr, i32, i32, ptr, ptr, ptr, i32, i32 }>, ptr [[BLOCK]], i32 0, i32 0
// CHECK4-NEXT:    store ptr @_NSConcreteStackBlock, ptr [[BLOCK_ISA]], align 8
// CHECK4-NEXT:    [[BLOCK_FLAGS:%.*]] = getelementptr inbounds nuw <{ ptr, i32, i32, ptr, ptr, ptr, i32, i32 }>, ptr [[BLOCK]], i32 0, i32 1
// CHECK4-NEXT:    store i32 1073741824, ptr [[BLOCK_FLAGS]], align 8
// CHECK4-NEXT:    [[BLOCK_RESERVED:%.*]] = getelementptr inbounds nuw <{ ptr, i32, i32, ptr, ptr, ptr, i32, i32 }>, ptr [[BLOCK]], i32 0, i32 2
// CHECK4-NEXT:    store i32 0, ptr [[BLOCK_RESERVED]], align 4
// CHECK4-NEXT:    [[BLOCK_INVOKE:%.*]] = getelementptr inbounds nuw <{ ptr, i32, i32, ptr, ptr, ptr, i32, i32 }>, ptr [[BLOCK]], i32 0, i32 3
// CHECK4-NEXT:    store ptr @var_block_invoke, ptr [[BLOCK_INVOKE]], align 8
// CHECK4-NEXT:    [[BLOCK_DESCRIPTOR:%.*]] = getelementptr inbounds nuw <{ ptr, i32, i32, ptr, ptr, ptr, i32, i32 }>, ptr [[BLOCK]], i32 0, i32 4
// CHECK4-NEXT:    store ptr @__block_descriptor_tmp, ptr [[BLOCK_DESCRIPTOR]], align 8
// CHECK4-NEXT:    [[BLOCK_CAPTURED:%.*]] = getelementptr inbounds nuw <{ ptr, i32, i32, ptr, ptr, ptr, i32, i32 }>, ptr [[BLOCK]], i32 0, i32 6
// CHECK4-NEXT:    [[TMP14:%.*]] = load volatile i32, ptr [[G]], align 4
// CHECK4-NEXT:    store volatile i32 [[TMP14]], ptr [[BLOCK_CAPTURED]], align 8
// CHECK4-NEXT:    [[BLOCK_CAPTURED5:%.*]] = getelementptr inbounds nuw <{ ptr, i32, i32, ptr, ptr, ptr, i32, i32 }>, ptr [[BLOCK]], i32 0, i32 5
// CHECK4-NEXT:    [[TMP15:%.*]] = load ptr, ptr [[_TMP2]], align 8
// CHECK4-NEXT:    store ptr [[TMP15]], ptr [[BLOCK_CAPTURED5]], align 8
// CHECK4-NEXT:    [[BLOCK_CAPTURED6:%.*]] = getelementptr inbounds nuw <{ ptr, i32, i32, ptr, ptr, ptr, i32, i32 }>, ptr [[BLOCK]], i32 0, i32 7
// CHECK4-NEXT:    [[TMP16:%.*]] = load i32, ptr [[SIVAR3]], align 4
// CHECK4-NEXT:    store i32 [[TMP16]], ptr [[BLOCK_CAPTURED6]], align 4
// CHECK4-NEXT:    [[TMP17:%.*]] = getelementptr inbounds nuw [[STRUCT___BLOCK_LITERAL_GENERIC:%.*]], ptr [[BLOCK]], i32 0, i32 3
// CHECK4-NEXT:    [[TMP18:%.*]] = load ptr, ptr [[TMP17]], align 8
// CHECK4-NEXT:    call void [[TMP18]](ptr noundef [[BLOCK]])
// CHECK4-NEXT:    br label [[OMP_BODY_CONTINUE:%.*]]
// CHECK4:       omp.body.continue:
// CHECK4-NEXT:    br label [[OMP_INNER_FOR_INC:%.*]]
// CHECK4:       omp.inner.for.inc:
// CHECK4-NEXT:    [[TMP19:%.*]] = load i32, ptr [[DOTOMP_IV]], align 4
// CHECK4-NEXT:    [[ADD7:%.*]] = add nsw i32 [[TMP19]], 1
// CHECK4-NEXT:    store i32 [[ADD7]], ptr [[DOTOMP_IV]], align 4
// CHECK4-NEXT:    br label [[OMP_INNER_FOR_COND]]
// CHECK4:       omp.inner.for.end:
// CHECK4-NEXT:    br label [[OMP_LOOP_EXIT:%.*]]
// CHECK4:       omp.loop.exit:
// CHECK4-NEXT:    call void @__kmpc_for_static_fini(ptr @[[GLOB1]], i32 [[TMP6]])
// CHECK4-NEXT:    call void @__kmpc_barrier(ptr @[[GLOB2:[0-9]+]], i32 [[TMP6]])
// CHECK4-NEXT:    ret void
//
//
// CHECK4-LABEL: define {{[^@]+}}@var_block_invoke
// CHECK4-SAME: (ptr noundef [[DOTBLOCK_DESCRIPTOR:%.*]]) #[[ATTR4]] {
// CHECK4-NEXT:  entry:
// CHECK4-NEXT:    [[DOTBLOCK_DESCRIPTOR_ADDR:%.*]] = alloca ptr, align 8
// CHECK4-NEXT:    [[BLOCK_ADDR:%.*]] = alloca ptr, align 8
// CHECK4-NEXT:    store ptr [[DOTBLOCK_DESCRIPTOR]], ptr [[DOTBLOCK_DESCRIPTOR_ADDR]], align 8
// CHECK4-NEXT:    store ptr [[DOTBLOCK_DESCRIPTOR]], ptr [[BLOCK_ADDR]], align 8
// CHECK4-NEXT:    [[BLOCK_CAPTURE_ADDR:%.*]] = getelementptr inbounds nuw <{ ptr, i32, i32, ptr, ptr, ptr, i32, i32 }>, ptr [[DOTBLOCK_DESCRIPTOR]], i32 0, i32 6
// CHECK4-NEXT:    store i32 2, ptr [[BLOCK_CAPTURE_ADDR]], align 8
// CHECK4-NEXT:    [[BLOCK_CAPTURE_ADDR1:%.*]] = getelementptr inbounds nuw <{ ptr, i32, i32, ptr, ptr, ptr, i32, i32 }>, ptr [[DOTBLOCK_DESCRIPTOR]], i32 0, i32 5
// CHECK4-NEXT:    [[TMP0:%.*]] = load ptr, ptr [[BLOCK_CAPTURE_ADDR1]], align 8
// CHECK4-NEXT:    store i32 2, ptr [[TMP0]], align 4
// CHECK4-NEXT:    [[BLOCK_CAPTURE_ADDR2:%.*]] = getelementptr inbounds nuw <{ ptr, i32, i32, ptr, ptr, ptr, i32, i32 }>, ptr [[DOTBLOCK_DESCRIPTOR]], i32 0, i32 7
// CHECK4-NEXT:    store i32 4, ptr [[BLOCK_CAPTURE_ADDR2]], align 4
// CHECK4-NEXT:    ret void
//
//
// CHECK4-LABEL: define {{[^@]+}}@_GLOBAL__sub_I_for_firstprivate_codegen.cpp
// CHECK4-SAME: () #[[ATTR0]] section "__TEXT,__StaticInit,regular,pure_instructions" {
// CHECK4-NEXT:  entry:
// CHECK4-NEXT:    call void @__cxx_global_var_init()
// CHECK4-NEXT:    call void @__cxx_global_var_init.1()
// CHECK4-NEXT:    call void @__cxx_global_var_init.2()
// CHECK4-NEXT:    ret void
//