// 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++ -emit-llvm %s -triple x86_64-unknown-linux -fexceptions -fcxx-exceptions -o - | FileCheck %s --check-prefix=CHECK1
// RUN: %clang_cc1 -fopenmp -x c++ -std=c++11 -triple x86_64-unknown-unknown -fexceptions -fcxx-exceptions -emit-pch -o %t %s
// RUN: %clang_cc1 -fopenmp -x c++ -triple x86_64-unknown-unknown -fexceptions -fcxx-exceptions -debug-info-kind=limited -std=c++11 -include-pch %t -verify %s -emit-llvm -o - | FileCheck %s --check-prefix=CHECK2
// RUN: %clang_cc1 -verify -fopenmp -fopenmp-enable-irbuilder -DIRBUILDER -x c++ -emit-llvm %s -triple x86_64-unknown-linux -fexceptions -fcxx-exceptions -o - | FileCheck %s --check-prefix=CHECK3
// RUN: %clang_cc1 -fopenmp -fopenmp-enable-irbuilder -DIRBUILDER -x c++ -std=c++11 -triple x86_64-unknown-unknown -fexceptions -fcxx-exceptions -emit-pch -o %t %s
// RUN: %clang_cc1 -fopenmp -fopenmp-enable-irbuilder -DIRBUILDER -x c++ -triple x86_64-unknown-unknown -fexceptions -fcxx-exceptions -debug-info-kind=limited -gno-column-info -std=c++11 -include-pch %t -verify %s -emit-llvm -o - | FileCheck %s --check-prefix=CHECK4

// RUN: %clang_cc1 -verify -fopenmp-simd -x c++ -emit-llvm %s -triple x86_64-unknown-linux -fexceptions -fcxx-exceptions -o - | FileCheck %s --implicit-check-not="{{__kmpc|__tgt}}"
// RUN: %clang_cc1 -fopenmp-simd -x c++ -std=c++11 -triple x86_64-unknown-unknown -fexceptions -fcxx-exceptions -emit-pch -o %t %s
// RUN: %clang_cc1 -fopenmp-simd -x c++ -triple x86_64-unknown-unknown -fexceptions -fcxx-exceptions -debug-info-kind=limited -std=c++11 -include-pch %t -verify %s -emit-llvm -o - | FileCheck %s --implicit-check-not="{{__kmpc|__tgt}}"
// RUN: %clang_cc1 -verify -fopenmp-simd -fopenmp-enable-irbuilder -x c++ -emit-llvm %s -triple x86_64-unknown-linux -fexceptions -fcxx-exceptions -o - | FileCheck %s --implicit-check-not="{{__kmpc|__tgt}}"
// RUN: %clang_cc1 -fopenmp-simd -fopenmp-enable-irbuilder -x c++ -std=c++11 -triple x86_64-unknown-unknown -fexceptions -fcxx-exceptions -emit-pch -o %t %s
// RUN: %clang_cc1 -fopenmp-simd -fopenmp-enable-irbuilder -x c++ -triple x86_64-unknown-unknown -fexceptions -fcxx-exceptions -debug-info-kind=limited -std=c++11 -include-pch %t -verify %s -emit-llvm -o - | FileCheck %s --implicit-check-not="{{__kmpc|__tgt}}"
// expected-no-diagnostics
#ifndef HEADER
#define HEADER


template <class T>
void foo(T argc) {}

template <typename T>
int tmain(T argc) {
  typedef double (*chunk_t)[argc[0][0]];
#pragma omp parallel
  {
  foo(argc);
  chunk_t var;(void)var[0][0];
  }
  return 0;
}

int global;
int main (int argc, char **argv) {
  int a[argc];
#pragma omp parallel shared(global, a) default(none)
  foo(a[1]), a[1] = global;
#ifndef IRBUILDER
// TODO: Support for privates in IRBuilder.
#pragma omp parallel private(global, a) default(none)
#pragma omp parallel shared(global, a) default(none)
  foo(a[1]), a[1] = global;
// FIXME: IRBuilder crashes in void llvm::OpenMPIRBuilder::finalize()
// Assertion `Extractor.isEligible() && "Expected OpenMP outlining to be possible!"' failed.
#pragma omp parallel shared(global, a) default(none)
#pragma omp parallel shared(global, a) default(none)
  foo(a[1]), a[1] = global;
#endif // IRBUILDER
  return tmain(argv);
}


// Note that OpenMPIRBuilder puts the trailing arguments in a different order:
// arguments that are wrapped into additional pointers precede the other
// arguments. This is expected and not problematic because both the call and the
// function are generated from the same place, and the function is internal.


#endif
// CHECK1-LABEL: define {{[^@]+}}@main
// CHECK1-SAME: (i32 [[ARGC:%.*]], i8** [[ARGV:%.*]]) #[[ATTR0:[0-9]+]] {
// CHECK1-NEXT:  entry:
// CHECK1-NEXT:    [[RETVAL:%.*]] = alloca i32, align 4
// CHECK1-NEXT:    [[ARGC_ADDR:%.*]] = alloca i32, align 4
// CHECK1-NEXT:    [[ARGV_ADDR:%.*]] = alloca i8**, align 8
// CHECK1-NEXT:    [[SAVED_STACK:%.*]] = alloca i8*, align 8
// CHECK1-NEXT:    [[__VLA_EXPR0:%.*]] = alloca i64, align 8
// CHECK1-NEXT:    store i32 0, i32* [[RETVAL]], align 4
// CHECK1-NEXT:    store i32 [[ARGC]], i32* [[ARGC_ADDR]], align 4
// CHECK1-NEXT:    store i8** [[ARGV]], i8*** [[ARGV_ADDR]], align 8
// CHECK1-NEXT:    [[TMP0:%.*]] = load i32, i32* [[ARGC_ADDR]], align 4
// CHECK1-NEXT:    [[TMP1:%.*]] = zext i32 [[TMP0]] to i64
// CHECK1-NEXT:    [[TMP2:%.*]] = call i8* @llvm.stacksave()
// CHECK1-NEXT:    store i8* [[TMP2]], i8** [[SAVED_STACK]], align 8
// CHECK1-NEXT:    [[VLA:%.*]] = alloca i32, i64 [[TMP1]], align 16
// CHECK1-NEXT:    store i64 [[TMP1]], i64* [[__VLA_EXPR0]], align 8
// CHECK1-NEXT:    call void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%struct.ident_t* @[[GLOB1:[0-9]+]], i32 2, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, i64, i32*)* @.omp_outlined. to void (i32*, i32*, ...)*), i64 [[TMP1]], i32* [[VLA]])
// CHECK1-NEXT:    call void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%struct.ident_t* @[[GLOB1]], i32 1, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, i64)* @.omp_outlined..1 to void (i32*, i32*, ...)*), i64 [[TMP1]])
// CHECK1-NEXT:    call void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%struct.ident_t* @[[GLOB1]], i32 2, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, i64, i32*)* @.omp_outlined..3 to void (i32*, i32*, ...)*), i64 [[TMP1]], i32* [[VLA]])
// CHECK1-NEXT:    [[TMP3:%.*]] = load i8**, i8*** [[ARGV_ADDR]], align 8
// CHECK1-NEXT:    [[CALL:%.*]] = call i32 @_Z5tmainIPPcEiT_(i8** [[TMP3]])
// CHECK1-NEXT:    store i32 [[CALL]], i32* [[RETVAL]], align 4
// CHECK1-NEXT:    [[TMP4:%.*]] = load i8*, i8** [[SAVED_STACK]], align 8
// CHECK1-NEXT:    call void @llvm.stackrestore(i8* [[TMP4]])
// CHECK1-NEXT:    [[TMP5:%.*]] = load i32, i32* [[RETVAL]], align 4
// CHECK1-NEXT:    ret i32 [[TMP5]]
//
//
// CHECK1-LABEL: define {{[^@]+}}@.omp_outlined.
// CHECK1-SAME: (i32* noalias [[DOTGLOBAL_TID_:%.*]], i32* noalias [[DOTBOUND_TID_:%.*]], i64 [[VLA:%.*]], i32* nonnull align 4 dereferenceable(4) [[A:%.*]]) #[[ATTR2:[0-9]+]] personality i8* bitcast (i32 (...)* @__gxx_personality_v0 to i8*) {
// 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:    [[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* [[A]], i32** [[A_ADDR]], align 8
// CHECK1-NEXT:    [[TMP0:%.*]] = load i64, i64* [[VLA_ADDR]], align 8
// CHECK1-NEXT:    [[TMP1:%.*]] = load i32*, i32** [[A_ADDR]], align 8
// CHECK1-NEXT:    [[ARRAYIDX:%.*]] = getelementptr inbounds i32, i32* [[TMP1]], i64 1
// CHECK1-NEXT:    [[TMP2:%.*]] = load i32, i32* [[ARRAYIDX]], align 4
// CHECK1-NEXT:    invoke void @_Z3fooIiEvT_(i32 [[TMP2]])
// CHECK1-NEXT:    to label [[INVOKE_CONT:%.*]] unwind label [[TERMINATE_LPAD:%.*]]
// CHECK1:       invoke.cont:
// CHECK1-NEXT:    [[TMP3:%.*]] = load i32, i32* @global, align 4
// CHECK1-NEXT:    [[ARRAYIDX1:%.*]] = getelementptr inbounds i32, i32* [[TMP1]], i64 1
// CHECK1-NEXT:    store i32 [[TMP3]], i32* [[ARRAYIDX1]], align 4
// CHECK1-NEXT:    ret void
// CHECK1:       terminate.lpad:
// CHECK1-NEXT:    [[TMP4:%.*]] = landingpad { i8*, i32 }
// CHECK1-NEXT:    catch i8* null
// CHECK1-NEXT:    [[TMP5:%.*]] = extractvalue { i8*, i32 } [[TMP4]], 0
// CHECK1-NEXT:    call void @__clang_call_terminate(i8* [[TMP5]]) #[[ATTR6:[0-9]+]]
// CHECK1-NEXT:    unreachable
//
//
// CHECK1-LABEL: define {{[^@]+}}@_Z3fooIiEvT_
// CHECK1-SAME: (i32 [[ARGC:%.*]]) #[[ATTR3:[0-9]+]] comdat {
// CHECK1-NEXT:  entry:
// CHECK1-NEXT:    [[ARGC_ADDR:%.*]] = alloca i32, align 4
// CHECK1-NEXT:    store i32 [[ARGC]], i32* [[ARGC_ADDR]], align 4
// CHECK1-NEXT:    ret void
//
//
// CHECK1-LABEL: define {{[^@]+}}@__clang_call_terminate
// CHECK1-SAME: (i8* [[TMP0:%.*]]) #[[ATTR4:[0-9]+]] comdat {
// CHECK1-NEXT:    [[TMP2:%.*]] = call i8* @__cxa_begin_catch(i8* [[TMP0]]) #[[ATTR5:[0-9]+]]
// CHECK1-NEXT:    call void @_ZSt9terminatev() #[[ATTR6]]
// CHECK1-NEXT:    unreachable
//
//
// CHECK1-LABEL: define {{[^@]+}}@.omp_outlined..1
// CHECK1-SAME: (i32* noalias [[DOTGLOBAL_TID_:%.*]], i32* noalias [[DOTBOUND_TID_:%.*]], i64 [[VLA:%.*]]) #[[ATTR2]] {
// 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:    [[GLOBAL:%.*]] = alloca i32, align 4
// CHECK1-NEXT:    [[SAVED_STACK:%.*]] = alloca i8*, align 8
// CHECK1-NEXT:    [[__VLA_EXPR0:%.*]] = alloca i64, 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:    [[TMP0:%.*]] = load i64, i64* [[VLA_ADDR]], align 8
// CHECK1-NEXT:    [[TMP1:%.*]] = call i8* @llvm.stacksave()
// CHECK1-NEXT:    store i8* [[TMP1]], i8** [[SAVED_STACK]], align 8
// CHECK1-NEXT:    [[VLA1:%.*]] = alloca i32, i64 [[TMP0]], align 16
// CHECK1-NEXT:    store i64 [[TMP0]], i64* [[__VLA_EXPR0]], align 8
// CHECK1-NEXT:    call void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%struct.ident_t* @[[GLOB1]], i32 3, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, i64, i32*, i32*)* @.omp_outlined..2 to void (i32*, i32*, ...)*), i64 [[TMP0]], i32* [[VLA1]], i32* [[GLOBAL]])
// CHECK1-NEXT:    [[TMP2:%.*]] = load i8*, i8** [[SAVED_STACK]], align 8
// CHECK1-NEXT:    call void @llvm.stackrestore(i8* [[TMP2]])
// CHECK1-NEXT:    ret void
//
//
// CHECK1-LABEL: define {{[^@]+}}@.omp_outlined..2
// CHECK1-SAME: (i32* noalias [[DOTGLOBAL_TID_:%.*]], i32* noalias [[DOTBOUND_TID_:%.*]], i64 [[VLA:%.*]], i32* nonnull align 4 dereferenceable(4) [[A:%.*]], i32* nonnull align 4 dereferenceable(4) [[GLOBAL:%.*]]) #[[ATTR2]] personality i8* bitcast (i32 (...)* @__gxx_personality_v0 to i8*) {
// 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:    [[A_ADDR:%.*]] = alloca i32*, align 8
// CHECK1-NEXT:    [[GLOBAL_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* [[A]], i32** [[A_ADDR]], align 8
// CHECK1-NEXT:    store i32* [[GLOBAL]], i32** [[GLOBAL_ADDR]], align 8
// CHECK1-NEXT:    [[TMP0:%.*]] = load i64, i64* [[VLA_ADDR]], align 8
// CHECK1-NEXT:    [[TMP1:%.*]] = load i32*, i32** [[A_ADDR]], align 8
// CHECK1-NEXT:    [[TMP2:%.*]] = load i32*, i32** [[GLOBAL_ADDR]], align 8
// CHECK1-NEXT:    [[ARRAYIDX:%.*]] = getelementptr inbounds i32, i32* [[TMP1]], i64 1
// CHECK1-NEXT:    [[TMP3:%.*]] = load i32, i32* [[ARRAYIDX]], align 4
// CHECK1-NEXT:    invoke void @_Z3fooIiEvT_(i32 [[TMP3]])
// CHECK1-NEXT:    to label [[INVOKE_CONT:%.*]] unwind label [[TERMINATE_LPAD:%.*]]
// CHECK1:       invoke.cont:
// CHECK1-NEXT:    [[TMP4:%.*]] = load i32, i32* [[TMP2]], align 4
// CHECK1-NEXT:    [[ARRAYIDX1:%.*]] = getelementptr inbounds i32, i32* [[TMP1]], i64 1
// CHECK1-NEXT:    store i32 [[TMP4]], i32* [[ARRAYIDX1]], align 4
// CHECK1-NEXT:    ret void
// CHECK1:       terminate.lpad:
// CHECK1-NEXT:    [[TMP5:%.*]] = landingpad { i8*, i32 }
// CHECK1-NEXT:    catch i8* null
// CHECK1-NEXT:    [[TMP6:%.*]] = extractvalue { i8*, i32 } [[TMP5]], 0
// CHECK1-NEXT:    call void @__clang_call_terminate(i8* [[TMP6]]) #[[ATTR6]]
// CHECK1-NEXT:    unreachable
//
//
// CHECK1-LABEL: define {{[^@]+}}@.omp_outlined..3
// CHECK1-SAME: (i32* noalias [[DOTGLOBAL_TID_:%.*]], i32* noalias [[DOTBOUND_TID_:%.*]], i64 [[VLA:%.*]], i32* nonnull align 4 dereferenceable(4) [[A:%.*]]) #[[ATTR2]] {
// 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:    [[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* [[A]], i32** [[A_ADDR]], align 8
// CHECK1-NEXT:    [[TMP0:%.*]] = load i64, i64* [[VLA_ADDR]], align 8
// CHECK1-NEXT:    [[TMP1:%.*]] = load i32*, i32** [[A_ADDR]], align 8
// CHECK1-NEXT:    call void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%struct.ident_t* @[[GLOB1]], i32 2, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, i64, i32*)* @.omp_outlined..4 to void (i32*, i32*, ...)*), i64 [[TMP0]], i32* [[TMP1]])
// CHECK1-NEXT:    ret void
//
//
// CHECK1-LABEL: define {{[^@]+}}@.omp_outlined..4
// CHECK1-SAME: (i32* noalias [[DOTGLOBAL_TID_:%.*]], i32* noalias [[DOTBOUND_TID_:%.*]], i64 [[VLA:%.*]], i32* nonnull align 4 dereferenceable(4) [[A:%.*]]) #[[ATTR2]] personality i8* bitcast (i32 (...)* @__gxx_personality_v0 to i8*) {
// 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:    [[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* [[A]], i32** [[A_ADDR]], align 8
// CHECK1-NEXT:    [[TMP0:%.*]] = load i64, i64* [[VLA_ADDR]], align 8
// CHECK1-NEXT:    [[TMP1:%.*]] = load i32*, i32** [[A_ADDR]], align 8
// CHECK1-NEXT:    [[ARRAYIDX:%.*]] = getelementptr inbounds i32, i32* [[TMP1]], i64 1
// CHECK1-NEXT:    [[TMP2:%.*]] = load i32, i32* [[ARRAYIDX]], align 4
// CHECK1-NEXT:    invoke void @_Z3fooIiEvT_(i32 [[TMP2]])
// CHECK1-NEXT:    to label [[INVOKE_CONT:%.*]] unwind label [[TERMINATE_LPAD:%.*]]
// CHECK1:       invoke.cont:
// CHECK1-NEXT:    [[TMP3:%.*]] = load i32, i32* @global, align 4
// CHECK1-NEXT:    [[ARRAYIDX1:%.*]] = getelementptr inbounds i32, i32* [[TMP1]], i64 1
// CHECK1-NEXT:    store i32 [[TMP3]], i32* [[ARRAYIDX1]], align 4
// CHECK1-NEXT:    ret void
// CHECK1:       terminate.lpad:
// CHECK1-NEXT:    [[TMP4:%.*]] = landingpad { i8*, i32 }
// CHECK1-NEXT:    catch i8* null
// CHECK1-NEXT:    [[TMP5:%.*]] = extractvalue { i8*, i32 } [[TMP4]], 0
// CHECK1-NEXT:    call void @__clang_call_terminate(i8* [[TMP5]]) #[[ATTR6]]
// CHECK1-NEXT:    unreachable
//
//
// CHECK1-LABEL: define {{[^@]+}}@_Z5tmainIPPcEiT_
// CHECK1-SAME: (i8** [[ARGC:%.*]]) #[[ATTR3]] comdat {
// CHECK1-NEXT:  entry:
// CHECK1-NEXT:    [[ARGC_ADDR:%.*]] = alloca i8**, align 8
// CHECK1-NEXT:    store i8** [[ARGC]], i8*** [[ARGC_ADDR]], align 8
// CHECK1-NEXT:    [[TMP0:%.*]] = load i8**, i8*** [[ARGC_ADDR]], align 8
// CHECK1-NEXT:    [[ARRAYIDX:%.*]] = getelementptr inbounds i8*, i8** [[TMP0]], i64 0
// CHECK1-NEXT:    [[TMP1:%.*]] = load i8*, i8** [[ARRAYIDX]], align 8
// CHECK1-NEXT:    [[ARRAYIDX1:%.*]] = getelementptr inbounds i8, i8* [[TMP1]], i64 0
// CHECK1-NEXT:    [[TMP2:%.*]] = load i8, i8* [[ARRAYIDX1]], align 1
// CHECK1-NEXT:    [[TMP3:%.*]] = zext i8 [[TMP2]] to i64
// CHECK1-NEXT:    call void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%struct.ident_t* @[[GLOB1]], i32 2, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, i8***, i64)* @.omp_outlined..5 to void (i32*, i32*, ...)*), i8*** [[ARGC_ADDR]], i64 [[TMP3]])
// CHECK1-NEXT:    ret i32 0
//
//
// CHECK1-LABEL: define {{[^@]+}}@.omp_outlined..5
// CHECK1-SAME: (i32* noalias [[DOTGLOBAL_TID_:%.*]], i32* noalias [[DOTBOUND_TID_:%.*]], i8*** nonnull align 8 dereferenceable(8) [[ARGC:%.*]], i64 [[VLA:%.*]]) #[[ATTR2]] personality i8* bitcast (i32 (...)* @__gxx_personality_v0 to i8*) {
// CHECK1-NEXT:  entry:
// CHECK1-NEXT:    [[DOTGLOBAL_TID__ADDR:%.*]] = alloca i32*, align 8
// CHECK1-NEXT:    [[DOTBOUND_TID__ADDR:%.*]] = alloca i32*, align 8
// CHECK1-NEXT:    [[ARGC_ADDR:%.*]] = alloca i8***, align 8
// CHECK1-NEXT:    [[VLA_ADDR:%.*]] = alloca i64, align 8
// CHECK1-NEXT:    [[VAR:%.*]] = alloca double*, 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 i8*** [[ARGC]], i8**** [[ARGC_ADDR]], align 8
// CHECK1-NEXT:    store i64 [[VLA]], i64* [[VLA_ADDR]], align 8
// CHECK1-NEXT:    [[TMP0:%.*]] = load i8***, i8**** [[ARGC_ADDR]], align 8
// CHECK1-NEXT:    [[TMP1:%.*]] = load i64, i64* [[VLA_ADDR]], align 8
// CHECK1-NEXT:    [[TMP2:%.*]] = load i8**, i8*** [[TMP0]], align 8
// CHECK1-NEXT:    invoke void @_Z3fooIPPcEvT_(i8** [[TMP2]])
// CHECK1-NEXT:    to label [[INVOKE_CONT:%.*]] unwind label [[TERMINATE_LPAD:%.*]]
// CHECK1:       invoke.cont:
// CHECK1-NEXT:    [[TMP3:%.*]] = load double*, double** [[VAR]], align 8
// CHECK1-NEXT:    [[TMP4:%.*]] = mul nsw i64 0, [[TMP1]]
// CHECK1-NEXT:    [[ARRAYIDX:%.*]] = getelementptr inbounds double, double* [[TMP3]], i64 [[TMP4]]
// CHECK1-NEXT:    [[ARRAYIDX1:%.*]] = getelementptr inbounds double, double* [[ARRAYIDX]], i64 0
// CHECK1-NEXT:    ret void
// CHECK1:       terminate.lpad:
// CHECK1-NEXT:    [[TMP5:%.*]] = landingpad { i8*, i32 }
// CHECK1-NEXT:    catch i8* null
// CHECK1-NEXT:    [[TMP6:%.*]] = extractvalue { i8*, i32 } [[TMP5]], 0
// CHECK1-NEXT:    call void @__clang_call_terminate(i8* [[TMP6]]) #[[ATTR6]]
// CHECK1-NEXT:    unreachable
//
//
// CHECK1-LABEL: define {{[^@]+}}@_Z3fooIPPcEvT_
// CHECK1-SAME: (i8** [[ARGC:%.*]]) #[[ATTR3]] comdat {
// CHECK1-NEXT:  entry:
// CHECK1-NEXT:    [[ARGC_ADDR:%.*]] = alloca i8**, align 8
// CHECK1-NEXT:    store i8** [[ARGC]], i8*** [[ARGC_ADDR]], align 8
// CHECK1-NEXT:    ret void
//
//
// CHECK2-LABEL: define {{[^@]+}}@main
// CHECK2-SAME: (i32 [[ARGC:%.*]], i8** [[ARGV:%.*]]) #[[ATTR0:[0-9]+]] !dbg [[DBG11:![0-9]+]] {
// CHECK2-NEXT:  entry:
// CHECK2-NEXT:    [[RETVAL:%.*]] = alloca i32, align 4
// CHECK2-NEXT:    [[ARGC_ADDR:%.*]] = alloca i32, align 4
// CHECK2-NEXT:    [[ARGV_ADDR:%.*]] = alloca i8**, align 8
// CHECK2-NEXT:    [[SAVED_STACK:%.*]] = alloca i8*, align 8
// CHECK2-NEXT:    [[__VLA_EXPR0:%.*]] = alloca i64, align 8
// CHECK2-NEXT:    store i32 0, i32* [[RETVAL]], align 4
// CHECK2-NEXT:    store i32 [[ARGC]], i32* [[ARGC_ADDR]], align 4
// CHECK2-NEXT:    call void @llvm.dbg.declare(metadata i32* [[ARGC_ADDR]], metadata [[META17:![0-9]+]], metadata !DIExpression()), !dbg [[DBG18:![0-9]+]]
// CHECK2-NEXT:    store i8** [[ARGV]], i8*** [[ARGV_ADDR]], align 8
// CHECK2-NEXT:    call void @llvm.dbg.declare(metadata i8*** [[ARGV_ADDR]], metadata [[META19:![0-9]+]], metadata !DIExpression()), !dbg [[DBG20:![0-9]+]]
// CHECK2-NEXT:    [[TMP0:%.*]] = load i32, i32* [[ARGC_ADDR]], align 4, !dbg [[DBG21:![0-9]+]]
// CHECK2-NEXT:    [[TMP1:%.*]] = zext i32 [[TMP0]] to i64, !dbg [[DBG22:![0-9]+]]
// CHECK2-NEXT:    [[TMP2:%.*]] = call i8* @llvm.stacksave(), !dbg [[DBG22]]
// CHECK2-NEXT:    store i8* [[TMP2]], i8** [[SAVED_STACK]], align 8, !dbg [[DBG22]]
// CHECK2-NEXT:    [[VLA:%.*]] = alloca i32, i64 [[TMP1]], align 16, !dbg [[DBG22]]
// CHECK2-NEXT:    store i64 [[TMP1]], i64* [[__VLA_EXPR0]], align 8, !dbg [[DBG22]]
// CHECK2-NEXT:    call void @llvm.dbg.declare(metadata i64* [[__VLA_EXPR0]], metadata [[META23:![0-9]+]], metadata !DIExpression()), !dbg [[DBG25:![0-9]+]]
// CHECK2-NEXT:    call void @llvm.dbg.declare(metadata i32* [[VLA]], metadata [[META26:![0-9]+]], metadata !DIExpression()), !dbg [[DBG30:![0-9]+]]
// CHECK2-NEXT:    call void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%struct.ident_t* @[[GLOB1:[0-9]+]], i32 2, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, i64, i32*)* @.omp_outlined. to void (i32*, i32*, ...)*), i64 [[TMP1]], i32* [[VLA]]), !dbg [[DBG31:![0-9]+]]
// CHECK2-NEXT:    call void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%struct.ident_t* @[[GLOB5:[0-9]+]], i32 1, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, i64)* @.omp_outlined..4 to void (i32*, i32*, ...)*), i64 [[TMP1]]), !dbg [[DBG32:![0-9]+]]
// CHECK2-NEXT:    call void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%struct.ident_t* @[[GLOB9:[0-9]+]], i32 2, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, i64, i32*)* @.omp_outlined..8 to void (i32*, i32*, ...)*), i64 [[TMP1]], i32* [[VLA]]), !dbg [[DBG33:![0-9]+]]
// CHECK2-NEXT:    [[TMP3:%.*]] = load i8**, i8*** [[ARGV_ADDR]], align 8, !dbg [[DBG34:![0-9]+]]
// CHECK2-NEXT:    [[CALL:%.*]] = call i32 @_Z5tmainIPPcEiT_(i8** [[TMP3]]), !dbg [[DBG35:![0-9]+]]
// CHECK2-NEXT:    store i32 [[CALL]], i32* [[RETVAL]], align 4, !dbg [[DBG36:![0-9]+]]
// CHECK2-NEXT:    [[TMP4:%.*]] = load i8*, i8** [[SAVED_STACK]], align 8, !dbg [[DBG37:![0-9]+]]
// CHECK2-NEXT:    call void @llvm.stackrestore(i8* [[TMP4]]), !dbg [[DBG37]]
// CHECK2-NEXT:    [[TMP5:%.*]] = load i32, i32* [[RETVAL]], align 4, !dbg [[DBG37]]
// CHECK2-NEXT:    ret i32 [[TMP5]], !dbg [[DBG37]]
//
//
// CHECK2-LABEL: define {{[^@]+}}@.omp_outlined._debug__
// CHECK2-SAME: (i32* noalias [[DOTGLOBAL_TID_:%.*]], i32* noalias [[DOTBOUND_TID_:%.*]], i64 [[VLA:%.*]], i32* nonnull align 4 dereferenceable(4) [[A:%.*]]) #[[ATTR3:[0-9]+]] personality i8* bitcast (i32 (...)* @__gxx_personality_v0 to i8*) !dbg [[DBG38:![0-9]+]] {
// CHECK2-NEXT:  entry:
// CHECK2-NEXT:    [[DOTGLOBAL_TID__ADDR:%.*]] = alloca i32*, align 8
// CHECK2-NEXT:    [[DOTBOUND_TID__ADDR:%.*]] = alloca i32*, align 8
// CHECK2-NEXT:    [[VLA_ADDR:%.*]] = alloca i64, align 8
// CHECK2-NEXT:    [[A_ADDR:%.*]] = alloca i32*, align 8
// CHECK2-NEXT:    store i32* [[DOTGLOBAL_TID_]], i32** [[DOTGLOBAL_TID__ADDR]], align 8
// CHECK2-NEXT:    call void @llvm.dbg.declare(metadata i32** [[DOTGLOBAL_TID__ADDR]], metadata [[META46:![0-9]+]], metadata !DIExpression()), !dbg [[DBG47:![0-9]+]]
// CHECK2-NEXT:    store i32* [[DOTBOUND_TID_]], i32** [[DOTBOUND_TID__ADDR]], align 8
// CHECK2-NEXT:    call void @llvm.dbg.declare(metadata i32** [[DOTBOUND_TID__ADDR]], metadata [[META48:![0-9]+]], metadata !DIExpression()), !dbg [[DBG47]]
// CHECK2-NEXT:    store i64 [[VLA]], i64* [[VLA_ADDR]], align 8
// CHECK2-NEXT:    call void @llvm.dbg.declare(metadata i64* [[VLA_ADDR]], metadata [[META49:![0-9]+]], metadata !DIExpression()), !dbg [[DBG47]]
// CHECK2-NEXT:    store i32* [[A]], i32** [[A_ADDR]], align 8
// CHECK2-NEXT:    call void @llvm.dbg.declare(metadata i32** [[A_ADDR]], metadata [[META50:![0-9]+]], metadata !DIExpression()), !dbg [[DBG51:![0-9]+]]
// CHECK2-NEXT:    [[TMP0:%.*]] = load i64, i64* [[VLA_ADDR]], align 8, !dbg [[DBG52:![0-9]+]]
// CHECK2-NEXT:    [[TMP1:%.*]] = load i32*, i32** [[A_ADDR]], align 8, !dbg [[DBG52]]
// CHECK2-NEXT:    [[ARRAYIDX:%.*]] = getelementptr inbounds i32, i32* [[TMP1]], i64 1, !dbg [[DBG53:![0-9]+]]
// CHECK2-NEXT:    [[TMP2:%.*]] = load i32, i32* [[ARRAYIDX]], align 4, !dbg [[DBG53]]
// CHECK2-NEXT:    invoke void @_Z3fooIiEvT_(i32 [[TMP2]])
// CHECK2-NEXT:    to label [[INVOKE_CONT:%.*]] unwind label [[TERMINATE_LPAD:%.*]], !dbg [[DBG52]]
// CHECK2:       invoke.cont:
// CHECK2-NEXT:    [[TMP3:%.*]] = load i32, i32* @global, align 4, !dbg [[DBG54:![0-9]+]]
// CHECK2-NEXT:    [[ARRAYIDX1:%.*]] = getelementptr inbounds i32, i32* [[TMP1]], i64 1, !dbg [[DBG55:![0-9]+]]
// CHECK2-NEXT:    store i32 [[TMP3]], i32* [[ARRAYIDX1]], align 4, !dbg [[DBG56:![0-9]+]]
// CHECK2-NEXT:    ret void, !dbg [[DBG54]]
// CHECK2:       terminate.lpad:
// CHECK2-NEXT:    [[TMP4:%.*]] = landingpad { i8*, i32 }
// CHECK2-NEXT:    catch i8* null, !dbg [[DBG52]]
// CHECK2-NEXT:    [[TMP5:%.*]] = extractvalue { i8*, i32 } [[TMP4]], 0, !dbg [[DBG52]]
// CHECK2-NEXT:    call void @__clang_call_terminate(i8* [[TMP5]]) #[[ATTR7:[0-9]+]], !dbg [[DBG52]]
// CHECK2-NEXT:    unreachable, !dbg [[DBG52]]
//
//
// CHECK2-LABEL: define {{[^@]+}}@_Z3fooIiEvT_
// CHECK2-SAME: (i32 [[ARGC:%.*]]) #[[ATTR4:[0-9]+]] comdat !dbg [[DBG57:![0-9]+]] {
// CHECK2-NEXT:  entry:
// CHECK2-NEXT:    [[ARGC_ADDR:%.*]] = alloca i32, align 4
// CHECK2-NEXT:    store i32 [[ARGC]], i32* [[ARGC_ADDR]], align 4
// CHECK2-NEXT:    call void @llvm.dbg.declare(metadata i32* [[ARGC_ADDR]], metadata [[META62:![0-9]+]], metadata !DIExpression()), !dbg [[DBG63:![0-9]+]]
// CHECK2-NEXT:    ret void, !dbg [[DBG64:![0-9]+]]
//
//
// CHECK2-LABEL: define {{[^@]+}}@__clang_call_terminate
// CHECK2-SAME: (i8* [[TMP0:%.*]]) #[[ATTR5:[0-9]+]] comdat {
// CHECK2-NEXT:    [[TMP2:%.*]] = call i8* @__cxa_begin_catch(i8* [[TMP0]]) #[[ATTR6:[0-9]+]]
// CHECK2-NEXT:    call void @_ZSt9terminatev() #[[ATTR7]]
// CHECK2-NEXT:    unreachable
//
//
// CHECK2-LABEL: define {{[^@]+}}@.omp_outlined.
// CHECK2-SAME: (i32* noalias [[DOTGLOBAL_TID_:%.*]], i32* noalias [[DOTBOUND_TID_:%.*]], i64 [[VLA:%.*]], i32* nonnull align 4 dereferenceable(4) [[A:%.*]]) #[[ATTR3]] !dbg [[DBG65:![0-9]+]] {
// CHECK2-NEXT:  entry:
// CHECK2-NEXT:    [[DOTGLOBAL_TID__ADDR:%.*]] = alloca i32*, align 8
// CHECK2-NEXT:    [[DOTBOUND_TID__ADDR:%.*]] = alloca i32*, align 8
// CHECK2-NEXT:    [[VLA_ADDR:%.*]] = alloca i64, align 8
// CHECK2-NEXT:    [[A_ADDR:%.*]] = alloca i32*, align 8
// CHECK2-NEXT:    store i32* [[DOTGLOBAL_TID_]], i32** [[DOTGLOBAL_TID__ADDR]], align 8
// CHECK2-NEXT:    call void @llvm.dbg.declare(metadata i32** [[DOTGLOBAL_TID__ADDR]], metadata [[META66:![0-9]+]], metadata !DIExpression()), !dbg [[DBG67:![0-9]+]]
// CHECK2-NEXT:    store i32* [[DOTBOUND_TID_]], i32** [[DOTBOUND_TID__ADDR]], align 8
// CHECK2-NEXT:    call void @llvm.dbg.declare(metadata i32** [[DOTBOUND_TID__ADDR]], metadata [[META68:![0-9]+]], metadata !DIExpression()), !dbg [[DBG67]]
// CHECK2-NEXT:    store i64 [[VLA]], i64* [[VLA_ADDR]], align 8
// CHECK2-NEXT:    call void @llvm.dbg.declare(metadata i64* [[VLA_ADDR]], metadata [[META69:![0-9]+]], metadata !DIExpression()), !dbg [[DBG67]]
// CHECK2-NEXT:    store i32* [[A]], i32** [[A_ADDR]], align 8
// CHECK2-NEXT:    call void @llvm.dbg.declare(metadata i32** [[A_ADDR]], metadata [[META70:![0-9]+]], metadata !DIExpression()), !dbg [[DBG67]]
// CHECK2-NEXT:    [[TMP0:%.*]] = load i64, i64* [[VLA_ADDR]], align 8, !dbg [[DBG71:![0-9]+]]
// CHECK2-NEXT:    [[TMP1:%.*]] = load i32*, i32** [[A_ADDR]], align 8, !dbg [[DBG71]]
// CHECK2-NEXT:    [[TMP2:%.*]] = load i32*, i32** [[DOTGLOBAL_TID__ADDR]], align 8, !dbg [[DBG71]]
// CHECK2-NEXT:    [[TMP3:%.*]] = load i32*, i32** [[DOTBOUND_TID__ADDR]], align 8, !dbg [[DBG71]]
// CHECK2-NEXT:    [[TMP4:%.*]] = load i32*, i32** [[A_ADDR]], align 8, !dbg [[DBG71]]
// CHECK2-NEXT:    call void @.omp_outlined._debug__(i32* [[TMP2]], i32* [[TMP3]], i64 [[TMP0]], i32* [[TMP4]]) #[[ATTR6]], !dbg [[DBG71]]
// CHECK2-NEXT:    ret void, !dbg [[DBG71]]
//
//
// CHECK2-LABEL: define {{[^@]+}}@.omp_outlined._debug__.1
// CHECK2-SAME: (i32* noalias [[DOTGLOBAL_TID_:%.*]], i32* noalias [[DOTBOUND_TID_:%.*]], i64 [[VLA:%.*]]) #[[ATTR3]] !dbg [[DBG74:![0-9]+]] {
// CHECK2-NEXT:  entry:
// CHECK2-NEXT:    [[DOTGLOBAL_TID__ADDR:%.*]] = alloca i32*, align 8
// CHECK2-NEXT:    [[DOTBOUND_TID__ADDR:%.*]] = alloca i32*, align 8
// CHECK2-NEXT:    [[VLA_ADDR:%.*]] = alloca i64, align 8
// CHECK2-NEXT:    [[GLOBAL:%.*]] = alloca i32, align 4
// CHECK2-NEXT:    [[SAVED_STACK:%.*]] = alloca i8*, align 8
// CHECK2-NEXT:    [[__VLA_EXPR0:%.*]] = alloca i64, align 8
// CHECK2-NEXT:    store i32* [[DOTGLOBAL_TID_]], i32** [[DOTGLOBAL_TID__ADDR]], align 8
// CHECK2-NEXT:    call void @llvm.dbg.declare(metadata i32** [[DOTGLOBAL_TID__ADDR]], metadata [[META77:![0-9]+]], metadata !DIExpression()), !dbg [[DBG78:![0-9]+]]
// CHECK2-NEXT:    store i32* [[DOTBOUND_TID_]], i32** [[DOTBOUND_TID__ADDR]], align 8
// CHECK2-NEXT:    call void @llvm.dbg.declare(metadata i32** [[DOTBOUND_TID__ADDR]], metadata [[META79:![0-9]+]], metadata !DIExpression()), !dbg [[DBG78]]
// CHECK2-NEXT:    store i64 [[VLA]], i64* [[VLA_ADDR]], align 8
// CHECK2-NEXT:    call void @llvm.dbg.declare(metadata i64* [[VLA_ADDR]], metadata [[META80:![0-9]+]], metadata !DIExpression()), !dbg [[DBG78]]
// CHECK2-NEXT:    [[TMP0:%.*]] = load i64, i64* [[VLA_ADDR]], align 8, !dbg [[DBG81:![0-9]+]]
// CHECK2-NEXT:    call void @llvm.dbg.declare(metadata i32* [[GLOBAL]], metadata [[META82:![0-9]+]], metadata !DIExpression()), !dbg [[DBG78]]
// CHECK2-NEXT:    [[TMP1:%.*]] = call i8* @llvm.stacksave(), !dbg [[DBG81]]
// CHECK2-NEXT:    store i8* [[TMP1]], i8** [[SAVED_STACK]], align 8, !dbg [[DBG81]]
// CHECK2-NEXT:    [[VLA1:%.*]] = alloca i32, i64 [[TMP0]], align 16, !dbg [[DBG81]]
// CHECK2-NEXT:    store i64 [[TMP0]], i64* [[__VLA_EXPR0]], align 8, !dbg [[DBG81]]
// CHECK2-NEXT:    call void @llvm.dbg.declare(metadata i64* [[__VLA_EXPR0]], metadata [[META83:![0-9]+]], metadata !DIExpression()), !dbg [[DBG78]]
// CHECK2-NEXT:    call void @llvm.dbg.declare(metadata i32* [[VLA1]], metadata [[META84:![0-9]+]], metadata !DIExpression()), !dbg [[DBG78]]
// CHECK2-NEXT:    call void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%struct.ident_t* @[[GLOB3:[0-9]+]], i32 3, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, i64, i32*, i32*)* @.omp_outlined..3 to void (i32*, i32*, ...)*), i64 [[TMP0]], i32* [[VLA1]], i32* [[GLOBAL]]), !dbg [[DBG81]]
// CHECK2-NEXT:    [[TMP2:%.*]] = load i8*, i8** [[SAVED_STACK]], align 8, !dbg [[DBG85:![0-9]+]]
// CHECK2-NEXT:    call void @llvm.stackrestore(i8* [[TMP2]]), !dbg [[DBG85]]
// CHECK2-NEXT:    ret void, !dbg [[DBG87:![0-9]+]]
//
//
// CHECK2-LABEL: define {{[^@]+}}@.omp_outlined._debug__.2
// CHECK2-SAME: (i32* noalias [[DOTGLOBAL_TID_:%.*]], i32* noalias [[DOTBOUND_TID_:%.*]], i64 [[VLA:%.*]], i32* nonnull align 4 dereferenceable(4) [[A:%.*]], i32* nonnull align 4 dereferenceable(4) [[GLOBAL:%.*]]) #[[ATTR3]] personality i8* bitcast (i32 (...)* @__gxx_personality_v0 to i8*) !dbg [[DBG88:![0-9]+]] {
// CHECK2-NEXT:  entry:
// CHECK2-NEXT:    [[DOTGLOBAL_TID__ADDR:%.*]] = alloca i32*, align 8
// CHECK2-NEXT:    [[DOTBOUND_TID__ADDR:%.*]] = alloca i32*, align 8
// CHECK2-NEXT:    [[VLA_ADDR:%.*]] = alloca i64, align 8
// CHECK2-NEXT:    [[A_ADDR:%.*]] = alloca i32*, align 8
// CHECK2-NEXT:    [[GLOBAL_ADDR:%.*]] = alloca i32*, align 8
// CHECK2-NEXT:    store i32* [[DOTGLOBAL_TID_]], i32** [[DOTGLOBAL_TID__ADDR]], align 8
// CHECK2-NEXT:    call void @llvm.dbg.declare(metadata i32** [[DOTGLOBAL_TID__ADDR]], metadata [[META91:![0-9]+]], metadata !DIExpression()), !dbg [[DBG92:![0-9]+]]
// CHECK2-NEXT:    store i32* [[DOTBOUND_TID_]], i32** [[DOTBOUND_TID__ADDR]], align 8
// CHECK2-NEXT:    call void @llvm.dbg.declare(metadata i32** [[DOTBOUND_TID__ADDR]], metadata [[META93:![0-9]+]], metadata !DIExpression()), !dbg [[DBG92]]
// CHECK2-NEXT:    store i64 [[VLA]], i64* [[VLA_ADDR]], align 8
// CHECK2-NEXT:    call void @llvm.dbg.declare(metadata i64* [[VLA_ADDR]], metadata [[META94:![0-9]+]], metadata !DIExpression()), !dbg [[DBG92]]
// CHECK2-NEXT:    store i32* [[A]], i32** [[A_ADDR]], align 8
// CHECK2-NEXT:    call void @llvm.dbg.declare(metadata i32** [[A_ADDR]], metadata [[META95:![0-9]+]], metadata !DIExpression()), !dbg [[DBG96:![0-9]+]]
// CHECK2-NEXT:    store i32* [[GLOBAL]], i32** [[GLOBAL_ADDR]], align 8
// CHECK2-NEXT:    call void @llvm.dbg.declare(metadata i32** [[GLOBAL_ADDR]], metadata [[META97:![0-9]+]], metadata !DIExpression()), !dbg [[DBG98:![0-9]+]]
// CHECK2-NEXT:    [[TMP0:%.*]] = load i64, i64* [[VLA_ADDR]], align 8, !dbg [[DBG99:![0-9]+]]
// CHECK2-NEXT:    [[TMP1:%.*]] = load i32*, i32** [[A_ADDR]], align 8, !dbg [[DBG99]]
// CHECK2-NEXT:    [[TMP2:%.*]] = load i32*, i32** [[GLOBAL_ADDR]], align 8, !dbg [[DBG99]]
// CHECK2-NEXT:    [[ARRAYIDX:%.*]] = getelementptr inbounds i32, i32* [[TMP1]], i64 1, !dbg [[DBG100:![0-9]+]]
// CHECK2-NEXT:    [[TMP3:%.*]] = load i32, i32* [[ARRAYIDX]], align 4, !dbg [[DBG100]]
// CHECK2-NEXT:    invoke void @_Z3fooIiEvT_(i32 [[TMP3]])
// CHECK2-NEXT:    to label [[INVOKE_CONT:%.*]] unwind label [[TERMINATE_LPAD:%.*]], !dbg [[DBG99]]
// CHECK2:       invoke.cont:
// CHECK2-NEXT:    [[TMP4:%.*]] = load i32, i32* [[TMP2]], align 4, !dbg [[DBG101:![0-9]+]]
// CHECK2-NEXT:    [[ARRAYIDX1:%.*]] = getelementptr inbounds i32, i32* [[TMP1]], i64 1, !dbg [[DBG102:![0-9]+]]
// CHECK2-NEXT:    store i32 [[TMP4]], i32* [[ARRAYIDX1]], align 4, !dbg [[DBG103:![0-9]+]]
// CHECK2-NEXT:    ret void, !dbg [[DBG101]]
// CHECK2:       terminate.lpad:
// CHECK2-NEXT:    [[TMP5:%.*]] = landingpad { i8*, i32 }
// CHECK2-NEXT:    catch i8* null, !dbg [[DBG99]]
// CHECK2-NEXT:    [[TMP6:%.*]] = extractvalue { i8*, i32 } [[TMP5]], 0, !dbg [[DBG99]]
// CHECK2-NEXT:    call void @__clang_call_terminate(i8* [[TMP6]]) #[[ATTR7]], !dbg [[DBG99]]
// CHECK2-NEXT:    unreachable, !dbg [[DBG99]]
//
//
// CHECK2-LABEL: define {{[^@]+}}@.omp_outlined..3
// CHECK2-SAME: (i32* noalias [[DOTGLOBAL_TID_:%.*]], i32* noalias [[DOTBOUND_TID_:%.*]], i64 [[VLA:%.*]], i32* nonnull align 4 dereferenceable(4) [[A:%.*]], i32* nonnull align 4 dereferenceable(4) [[GLOBAL:%.*]]) #[[ATTR3]] !dbg [[DBG104:![0-9]+]] {
// CHECK2-NEXT:  entry:
// CHECK2-NEXT:    [[DOTGLOBAL_TID__ADDR:%.*]] = alloca i32*, align 8
// CHECK2-NEXT:    [[DOTBOUND_TID__ADDR:%.*]] = alloca i32*, align 8
// CHECK2-NEXT:    [[VLA_ADDR:%.*]] = alloca i64, align 8
// CHECK2-NEXT:    [[A_ADDR:%.*]] = alloca i32*, align 8
// CHECK2-NEXT:    [[GLOBAL_ADDR:%.*]] = alloca i32*, align 8
// CHECK2-NEXT:    store i32* [[DOTGLOBAL_TID_]], i32** [[DOTGLOBAL_TID__ADDR]], align 8
// CHECK2-NEXT:    call void @llvm.dbg.declare(metadata i32** [[DOTGLOBAL_TID__ADDR]], metadata [[META105:![0-9]+]], metadata !DIExpression()), !dbg [[DBG106:![0-9]+]]
// CHECK2-NEXT:    store i32* [[DOTBOUND_TID_]], i32** [[DOTBOUND_TID__ADDR]], align 8
// CHECK2-NEXT:    call void @llvm.dbg.declare(metadata i32** [[DOTBOUND_TID__ADDR]], metadata [[META107:![0-9]+]], metadata !DIExpression()), !dbg [[DBG106]]
// CHECK2-NEXT:    store i64 [[VLA]], i64* [[VLA_ADDR]], align 8
// CHECK2-NEXT:    call void @llvm.dbg.declare(metadata i64* [[VLA_ADDR]], metadata [[META108:![0-9]+]], metadata !DIExpression()), !dbg [[DBG106]]
// CHECK2-NEXT:    store i32* [[A]], i32** [[A_ADDR]], align 8
// CHECK2-NEXT:    call void @llvm.dbg.declare(metadata i32** [[A_ADDR]], metadata [[META109:![0-9]+]], metadata !DIExpression()), !dbg [[DBG106]]
// CHECK2-NEXT:    store i32* [[GLOBAL]], i32** [[GLOBAL_ADDR]], align 8
// CHECK2-NEXT:    call void @llvm.dbg.declare(metadata i32** [[GLOBAL_ADDR]], metadata [[META110:![0-9]+]], metadata !DIExpression()), !dbg [[DBG106]]
// CHECK2-NEXT:    [[TMP0:%.*]] = load i64, i64* [[VLA_ADDR]], align 8, !dbg [[DBG111:![0-9]+]]
// CHECK2-NEXT:    [[TMP1:%.*]] = load i32*, i32** [[A_ADDR]], align 8, !dbg [[DBG111]]
// CHECK2-NEXT:    [[TMP2:%.*]] = load i32*, i32** [[GLOBAL_ADDR]], align 8, !dbg [[DBG111]]
// CHECK2-NEXT:    [[TMP3:%.*]] = load i32*, i32** [[DOTGLOBAL_TID__ADDR]], align 8, !dbg [[DBG111]]
// CHECK2-NEXT:    [[TMP4:%.*]] = load i32*, i32** [[DOTBOUND_TID__ADDR]], align 8, !dbg [[DBG111]]
// CHECK2-NEXT:    [[TMP5:%.*]] = load i32*, i32** [[A_ADDR]], align 8, !dbg [[DBG111]]
// CHECK2-NEXT:    [[TMP6:%.*]] = load i32*, i32** [[GLOBAL_ADDR]], align 8, !dbg [[DBG111]]
// CHECK2-NEXT:    call void @.omp_outlined._debug__.2(i32* [[TMP3]], i32* [[TMP4]], i64 [[TMP0]], i32* [[TMP5]], i32* [[TMP6]]) #[[ATTR6]], !dbg [[DBG111]]
// CHECK2-NEXT:    ret void, !dbg [[DBG111]]
//
//
// CHECK2-LABEL: define {{[^@]+}}@.omp_outlined..4
// CHECK2-SAME: (i32* noalias [[DOTGLOBAL_TID_:%.*]], i32* noalias [[DOTBOUND_TID_:%.*]], i64 [[VLA:%.*]]) #[[ATTR3]] !dbg [[DBG112:![0-9]+]] {
// CHECK2-NEXT:  entry:
// CHECK2-NEXT:    [[DOTGLOBAL_TID__ADDR:%.*]] = alloca i32*, align 8
// CHECK2-NEXT:    [[DOTBOUND_TID__ADDR:%.*]] = alloca i32*, align 8
// CHECK2-NEXT:    [[VLA_ADDR:%.*]] = alloca i64, align 8
// CHECK2-NEXT:    store i32* [[DOTGLOBAL_TID_]], i32** [[DOTGLOBAL_TID__ADDR]], align 8
// CHECK2-NEXT:    call void @llvm.dbg.declare(metadata i32** [[DOTGLOBAL_TID__ADDR]], metadata [[META113:![0-9]+]], metadata !DIExpression()), !dbg [[DBG114:![0-9]+]]
// CHECK2-NEXT:    store i32* [[DOTBOUND_TID_]], i32** [[DOTBOUND_TID__ADDR]], align 8
// CHECK2-NEXT:    call void @llvm.dbg.declare(metadata i32** [[DOTBOUND_TID__ADDR]], metadata [[META115:![0-9]+]], metadata !DIExpression()), !dbg [[DBG114]]
// CHECK2-NEXT:    store i64 [[VLA]], i64* [[VLA_ADDR]], align 8
// CHECK2-NEXT:    call void @llvm.dbg.declare(metadata i64* [[VLA_ADDR]], metadata [[META116:![0-9]+]], metadata !DIExpression()), !dbg [[DBG114]]
// CHECK2-NEXT:    [[TMP0:%.*]] = load i64, i64* [[VLA_ADDR]], align 8, !dbg [[DBG117:![0-9]+]]
// CHECK2-NEXT:    [[TMP1:%.*]] = load i32*, i32** [[DOTGLOBAL_TID__ADDR]], align 8, !dbg [[DBG117]]
// CHECK2-NEXT:    [[TMP2:%.*]] = load i32*, i32** [[DOTBOUND_TID__ADDR]], align 8, !dbg [[DBG117]]
// CHECK2-NEXT:    call void @.omp_outlined._debug__.1(i32* [[TMP1]], i32* [[TMP2]], i64 [[TMP0]]) #[[ATTR6]], !dbg [[DBG117]]
// CHECK2-NEXT:    ret void, !dbg [[DBG117]]
//
//
// CHECK2-LABEL: define {{[^@]+}}@.omp_outlined._debug__.5
// CHECK2-SAME: (i32* noalias [[DOTGLOBAL_TID_:%.*]], i32* noalias [[DOTBOUND_TID_:%.*]], i64 [[VLA:%.*]], i32* nonnull align 4 dereferenceable(4) [[A:%.*]]) #[[ATTR3]] !dbg [[DBG118:![0-9]+]] {
// CHECK2-NEXT:  entry:
// CHECK2-NEXT:    [[DOTGLOBAL_TID__ADDR:%.*]] = alloca i32*, align 8
// CHECK2-NEXT:    [[DOTBOUND_TID__ADDR:%.*]] = alloca i32*, align 8
// CHECK2-NEXT:    [[VLA_ADDR:%.*]] = alloca i64, align 8
// CHECK2-NEXT:    [[A_ADDR:%.*]] = alloca i32*, align 8
// CHECK2-NEXT:    store i32* [[DOTGLOBAL_TID_]], i32** [[DOTGLOBAL_TID__ADDR]], align 8
// CHECK2-NEXT:    call void @llvm.dbg.declare(metadata i32** [[DOTGLOBAL_TID__ADDR]], metadata [[META119:![0-9]+]], metadata !DIExpression()), !dbg [[DBG120:![0-9]+]]
// CHECK2-NEXT:    store i32* [[DOTBOUND_TID_]], i32** [[DOTBOUND_TID__ADDR]], align 8
// CHECK2-NEXT:    call void @llvm.dbg.declare(metadata i32** [[DOTBOUND_TID__ADDR]], metadata [[META121:![0-9]+]], metadata !DIExpression()), !dbg [[DBG120]]
// CHECK2-NEXT:    store i64 [[VLA]], i64* [[VLA_ADDR]], align 8
// CHECK2-NEXT:    call void @llvm.dbg.declare(metadata i64* [[VLA_ADDR]], metadata [[META122:![0-9]+]], metadata !DIExpression()), !dbg [[DBG120]]
// CHECK2-NEXT:    store i32* [[A]], i32** [[A_ADDR]], align 8
// CHECK2-NEXT:    call void @llvm.dbg.declare(metadata i32** [[A_ADDR]], metadata [[META123:![0-9]+]], metadata !DIExpression()), !dbg [[DBG124:![0-9]+]]
// CHECK2-NEXT:    [[TMP0:%.*]] = load i64, i64* [[VLA_ADDR]], align 8, !dbg [[DBG125:![0-9]+]]
// CHECK2-NEXT:    [[TMP1:%.*]] = load i32*, i32** [[A_ADDR]], align 8, !dbg [[DBG125]]
// CHECK2-NEXT:    call void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%struct.ident_t* @[[GLOB7:[0-9]+]], i32 2, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, i64, i32*)* @.omp_outlined..7 to void (i32*, i32*, ...)*), i64 [[TMP0]], i32* [[TMP1]]), !dbg [[DBG125]]
// CHECK2-NEXT:    ret void, !dbg [[DBG126:![0-9]+]]
//
//
// CHECK2-LABEL: define {{[^@]+}}@.omp_outlined._debug__.6
// CHECK2-SAME: (i32* noalias [[DOTGLOBAL_TID_:%.*]], i32* noalias [[DOTBOUND_TID_:%.*]], i64 [[VLA:%.*]], i32* nonnull align 4 dereferenceable(4) [[A:%.*]]) #[[ATTR3]] personality i8* bitcast (i32 (...)* @__gxx_personality_v0 to i8*) !dbg [[DBG127:![0-9]+]] {
// CHECK2-NEXT:  entry:
// CHECK2-NEXT:    [[DOTGLOBAL_TID__ADDR:%.*]] = alloca i32*, align 8
// CHECK2-NEXT:    [[DOTBOUND_TID__ADDR:%.*]] = alloca i32*, align 8
// CHECK2-NEXT:    [[VLA_ADDR:%.*]] = alloca i64, align 8
// CHECK2-NEXT:    [[A_ADDR:%.*]] = alloca i32*, align 8
// CHECK2-NEXT:    store i32* [[DOTGLOBAL_TID_]], i32** [[DOTGLOBAL_TID__ADDR]], align 8
// CHECK2-NEXT:    call void @llvm.dbg.declare(metadata i32** [[DOTGLOBAL_TID__ADDR]], metadata [[META128:![0-9]+]], metadata !DIExpression()), !dbg [[DBG129:![0-9]+]]
// CHECK2-NEXT:    store i32* [[DOTBOUND_TID_]], i32** [[DOTBOUND_TID__ADDR]], align 8
// CHECK2-NEXT:    call void @llvm.dbg.declare(metadata i32** [[DOTBOUND_TID__ADDR]], metadata [[META130:![0-9]+]], metadata !DIExpression()), !dbg [[DBG129]]
// CHECK2-NEXT:    store i64 [[VLA]], i64* [[VLA_ADDR]], align 8
// CHECK2-NEXT:    call void @llvm.dbg.declare(metadata i64* [[VLA_ADDR]], metadata [[META131:![0-9]+]], metadata !DIExpression()), !dbg [[DBG129]]
// CHECK2-NEXT:    store i32* [[A]], i32** [[A_ADDR]], align 8
// CHECK2-NEXT:    call void @llvm.dbg.declare(metadata i32** [[A_ADDR]], metadata [[META132:![0-9]+]], metadata !DIExpression()), !dbg [[DBG133:![0-9]+]]
// CHECK2-NEXT:    [[TMP0:%.*]] = load i64, i64* [[VLA_ADDR]], align 8, !dbg [[DBG134:![0-9]+]]
// CHECK2-NEXT:    [[TMP1:%.*]] = load i32*, i32** [[A_ADDR]], align 8, !dbg [[DBG134]]
// CHECK2-NEXT:    [[ARRAYIDX:%.*]] = getelementptr inbounds i32, i32* [[TMP1]], i64 1, !dbg [[DBG135:![0-9]+]]
// CHECK2-NEXT:    [[TMP2:%.*]] = load i32, i32* [[ARRAYIDX]], align 4, !dbg [[DBG135]]
// CHECK2-NEXT:    invoke void @_Z3fooIiEvT_(i32 [[TMP2]])
// CHECK2-NEXT:    to label [[INVOKE_CONT:%.*]] unwind label [[TERMINATE_LPAD:%.*]], !dbg [[DBG134]]
// CHECK2:       invoke.cont:
// CHECK2-NEXT:    [[TMP3:%.*]] = load i32, i32* @global, align 4, !dbg [[DBG136:![0-9]+]]
// CHECK2-NEXT:    [[ARRAYIDX1:%.*]] = getelementptr inbounds i32, i32* [[TMP1]], i64 1, !dbg [[DBG137:![0-9]+]]
// CHECK2-NEXT:    store i32 [[TMP3]], i32* [[ARRAYIDX1]], align 4, !dbg [[DBG138:![0-9]+]]
// CHECK2-NEXT:    ret void, !dbg [[DBG136]]
// CHECK2:       terminate.lpad:
// CHECK2-NEXT:    [[TMP4:%.*]] = landingpad { i8*, i32 }
// CHECK2-NEXT:    catch i8* null, !dbg [[DBG134]]
// CHECK2-NEXT:    [[TMP5:%.*]] = extractvalue { i8*, i32 } [[TMP4]], 0, !dbg [[DBG134]]
// CHECK2-NEXT:    call void @__clang_call_terminate(i8* [[TMP5]]) #[[ATTR7]], !dbg [[DBG134]]
// CHECK2-NEXT:    unreachable, !dbg [[DBG134]]
//
//
// CHECK2-LABEL: define {{[^@]+}}@.omp_outlined..7
// CHECK2-SAME: (i32* noalias [[DOTGLOBAL_TID_:%.*]], i32* noalias [[DOTBOUND_TID_:%.*]], i64 [[VLA:%.*]], i32* nonnull align 4 dereferenceable(4) [[A:%.*]]) #[[ATTR3]] !dbg [[DBG139:![0-9]+]] {
// CHECK2-NEXT:  entry:
// CHECK2-NEXT:    [[DOTGLOBAL_TID__ADDR:%.*]] = alloca i32*, align 8
// CHECK2-NEXT:    [[DOTBOUND_TID__ADDR:%.*]] = alloca i32*, align 8
// CHECK2-NEXT:    [[VLA_ADDR:%.*]] = alloca i64, align 8
// CHECK2-NEXT:    [[A_ADDR:%.*]] = alloca i32*, align 8
// CHECK2-NEXT:    store i32* [[DOTGLOBAL_TID_]], i32** [[DOTGLOBAL_TID__ADDR]], align 8
// CHECK2-NEXT:    call void @llvm.dbg.declare(metadata i32** [[DOTGLOBAL_TID__ADDR]], metadata [[META140:![0-9]+]], metadata !DIExpression()), !dbg [[DBG141:![0-9]+]]
// CHECK2-NEXT:    store i32* [[DOTBOUND_TID_]], i32** [[DOTBOUND_TID__ADDR]], align 8
// CHECK2-NEXT:    call void @llvm.dbg.declare(metadata i32** [[DOTBOUND_TID__ADDR]], metadata [[META142:![0-9]+]], metadata !DIExpression()), !dbg [[DBG141]]
// CHECK2-NEXT:    store i64 [[VLA]], i64* [[VLA_ADDR]], align 8
// CHECK2-NEXT:    call void @llvm.dbg.declare(metadata i64* [[VLA_ADDR]], metadata [[META143:![0-9]+]], metadata !DIExpression()), !dbg [[DBG141]]
// CHECK2-NEXT:    store i32* [[A]], i32** [[A_ADDR]], align 8
// CHECK2-NEXT:    call void @llvm.dbg.declare(metadata i32** [[A_ADDR]], metadata [[META144:![0-9]+]], metadata !DIExpression()), !dbg [[DBG141]]
// CHECK2-NEXT:    [[TMP0:%.*]] = load i64, i64* [[VLA_ADDR]], align 8, !dbg [[DBG145:![0-9]+]]
// CHECK2-NEXT:    [[TMP1:%.*]] = load i32*, i32** [[A_ADDR]], align 8, !dbg [[DBG145]]
// CHECK2-NEXT:    [[TMP2:%.*]] = load i32*, i32** [[DOTGLOBAL_TID__ADDR]], align 8, !dbg [[DBG145]]
// CHECK2-NEXT:    [[TMP3:%.*]] = load i32*, i32** [[DOTBOUND_TID__ADDR]], align 8, !dbg [[DBG145]]
// CHECK2-NEXT:    [[TMP4:%.*]] = load i32*, i32** [[A_ADDR]], align 8, !dbg [[DBG145]]
// CHECK2-NEXT:    call void @.omp_outlined._debug__.6(i32* [[TMP2]], i32* [[TMP3]], i64 [[TMP0]], i32* [[TMP4]]) #[[ATTR6]], !dbg [[DBG145]]
// CHECK2-NEXT:    ret void, !dbg [[DBG145]]
//
//
// CHECK2-LABEL: define {{[^@]+}}@.omp_outlined..8
// CHECK2-SAME: (i32* noalias [[DOTGLOBAL_TID_:%.*]], i32* noalias [[DOTBOUND_TID_:%.*]], i64 [[VLA:%.*]], i32* nonnull align 4 dereferenceable(4) [[A:%.*]]) #[[ATTR3]] !dbg [[DBG146:![0-9]+]] {
// CHECK2-NEXT:  entry:
// CHECK2-NEXT:    [[DOTGLOBAL_TID__ADDR:%.*]] = alloca i32*, align 8
// CHECK2-NEXT:    [[DOTBOUND_TID__ADDR:%.*]] = alloca i32*, align 8
// CHECK2-NEXT:    [[VLA_ADDR:%.*]] = alloca i64, align 8
// CHECK2-NEXT:    [[A_ADDR:%.*]] = alloca i32*, align 8
// CHECK2-NEXT:    store i32* [[DOTGLOBAL_TID_]], i32** [[DOTGLOBAL_TID__ADDR]], align 8
// CHECK2-NEXT:    call void @llvm.dbg.declare(metadata i32** [[DOTGLOBAL_TID__ADDR]], metadata [[META147:![0-9]+]], metadata !DIExpression()), !dbg [[DBG148:![0-9]+]]
// CHECK2-NEXT:    store i32* [[DOTBOUND_TID_]], i32** [[DOTBOUND_TID__ADDR]], align 8
// CHECK2-NEXT:    call void @llvm.dbg.declare(metadata i32** [[DOTBOUND_TID__ADDR]], metadata [[META149:![0-9]+]], metadata !DIExpression()), !dbg [[DBG148]]
// CHECK2-NEXT:    store i64 [[VLA]], i64* [[VLA_ADDR]], align 8
// CHECK2-NEXT:    call void @llvm.dbg.declare(metadata i64* [[VLA_ADDR]], metadata [[META150:![0-9]+]], metadata !DIExpression()), !dbg [[DBG148]]
// CHECK2-NEXT:    store i32* [[A]], i32** [[A_ADDR]], align 8
// CHECK2-NEXT:    call void @llvm.dbg.declare(metadata i32** [[A_ADDR]], metadata [[META151:![0-9]+]], metadata !DIExpression()), !dbg [[DBG148]]
// CHECK2-NEXT:    [[TMP0:%.*]] = load i64, i64* [[VLA_ADDR]], align 8, !dbg [[DBG152:![0-9]+]]
// CHECK2-NEXT:    [[TMP1:%.*]] = load i32*, i32** [[A_ADDR]], align 8, !dbg [[DBG152]]
// CHECK2-NEXT:    [[TMP2:%.*]] = load i32*, i32** [[DOTGLOBAL_TID__ADDR]], align 8, !dbg [[DBG152]]
// CHECK2-NEXT:    [[TMP3:%.*]] = load i32*, i32** [[DOTBOUND_TID__ADDR]], align 8, !dbg [[DBG152]]
// CHECK2-NEXT:    [[TMP4:%.*]] = load i32*, i32** [[A_ADDR]], align 8, !dbg [[DBG152]]
// CHECK2-NEXT:    call void @.omp_outlined._debug__.5(i32* [[TMP2]], i32* [[TMP3]], i64 [[TMP0]], i32* [[TMP4]]) #[[ATTR6]], !dbg [[DBG152]]
// CHECK2-NEXT:    ret void, !dbg [[DBG152]]
//
//
// CHECK2-LABEL: define {{[^@]+}}@_Z5tmainIPPcEiT_
// CHECK2-SAME: (i8** [[ARGC:%.*]]) #[[ATTR4]] comdat !dbg [[DBG153:![0-9]+]] {
// CHECK2-NEXT:  entry:
// CHECK2-NEXT:    [[ARGC_ADDR:%.*]] = alloca i8**, align 8
// CHECK2-NEXT:    store i8** [[ARGC]], i8*** [[ARGC_ADDR]], align 8
// CHECK2-NEXT:    call void @llvm.dbg.declare(metadata i8*** [[ARGC_ADDR]], metadata [[META158:![0-9]+]], metadata !DIExpression()), !dbg [[DBG159:![0-9]+]]
// CHECK2-NEXT:    [[TMP0:%.*]] = load i8**, i8*** [[ARGC_ADDR]], align 8, !dbg [[DBG160:![0-9]+]]
// CHECK2-NEXT:    [[ARRAYIDX:%.*]] = getelementptr inbounds i8*, i8** [[TMP0]], i64 0, !dbg [[DBG160]]
// CHECK2-NEXT:    [[TMP1:%.*]] = load i8*, i8** [[ARRAYIDX]], align 8, !dbg [[DBG160]]
// CHECK2-NEXT:    [[ARRAYIDX1:%.*]] = getelementptr inbounds i8, i8* [[TMP1]], i64 0, !dbg [[DBG160]]
// CHECK2-NEXT:    [[TMP2:%.*]] = load i8, i8* [[ARRAYIDX1]], align 1, !dbg [[DBG160]]
// CHECK2-NEXT:    [[TMP3:%.*]] = zext i8 [[TMP2]] to i64, !dbg [[DBG161:![0-9]+]]
// CHECK2-NEXT:    call void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%struct.ident_t* @[[GLOB11:[0-9]+]], i32 2, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, i8***, i64)* @.omp_outlined..10 to void (i32*, i32*, ...)*), i8*** [[ARGC_ADDR]], i64 [[TMP3]]), !dbg [[DBG162:![0-9]+]]
// CHECK2-NEXT:    ret i32 0, !dbg [[DBG163:![0-9]+]]
//
//
// CHECK2-LABEL: define {{[^@]+}}@.omp_outlined._debug__.9
// CHECK2-SAME: (i32* noalias [[DOTGLOBAL_TID_:%.*]], i32* noalias [[DOTBOUND_TID_:%.*]], i8*** nonnull align 8 dereferenceable(8) [[ARGC:%.*]], i64 [[VLA:%.*]]) #[[ATTR3]] personality i8* bitcast (i32 (...)* @__gxx_personality_v0 to i8*) !dbg [[DBG164:![0-9]+]] {
// CHECK2-NEXT:  entry:
// CHECK2-NEXT:    [[DOTGLOBAL_TID__ADDR:%.*]] = alloca i32*, align 8
// CHECK2-NEXT:    [[DOTBOUND_TID__ADDR:%.*]] = alloca i32*, align 8
// CHECK2-NEXT:    [[ARGC_ADDR:%.*]] = alloca i8***, align 8
// CHECK2-NEXT:    [[VLA_ADDR:%.*]] = alloca i64, align 8
// CHECK2-NEXT:    [[VAR:%.*]] = alloca double*, align 8
// CHECK2-NEXT:    store i32* [[DOTGLOBAL_TID_]], i32** [[DOTGLOBAL_TID__ADDR]], align 8
// CHECK2-NEXT:    call void @llvm.dbg.declare(metadata i32** [[DOTGLOBAL_TID__ADDR]], metadata [[META168:![0-9]+]], metadata !DIExpression()), !dbg [[DBG169:![0-9]+]]
// CHECK2-NEXT:    store i32* [[DOTBOUND_TID_]], i32** [[DOTBOUND_TID__ADDR]], align 8
// CHECK2-NEXT:    call void @llvm.dbg.declare(metadata i32** [[DOTBOUND_TID__ADDR]], metadata [[META170:![0-9]+]], metadata !DIExpression()), !dbg [[DBG169]]
// CHECK2-NEXT:    store i8*** [[ARGC]], i8**** [[ARGC_ADDR]], align 8
// CHECK2-NEXT:    call void @llvm.dbg.declare(metadata i8**** [[ARGC_ADDR]], metadata [[META171:![0-9]+]], metadata !DIExpression()), !dbg [[DBG172:![0-9]+]]
// CHECK2-NEXT:    store i64 [[VLA]], i64* [[VLA_ADDR]], align 8
// CHECK2-NEXT:    call void @llvm.dbg.declare(metadata i64* [[VLA_ADDR]], metadata [[META173:![0-9]+]], metadata !DIExpression()), !dbg [[DBG169]]
// CHECK2-NEXT:    [[TMP0:%.*]] = load i8***, i8**** [[ARGC_ADDR]], align 8, !dbg [[DBG174:![0-9]+]]
// CHECK2-NEXT:    [[TMP1:%.*]] = load i64, i64* [[VLA_ADDR]], align 8, !dbg [[DBG174]]
// CHECK2-NEXT:    [[TMP2:%.*]] = load i8**, i8*** [[TMP0]], align 8, !dbg [[DBG175:![0-9]+]]
// CHECK2-NEXT:    invoke void @_Z3fooIPPcEvT_(i8** [[TMP2]])
// CHECK2-NEXT:    to label [[INVOKE_CONT:%.*]] unwind label [[TERMINATE_LPAD:%.*]], !dbg [[DBG177:![0-9]+]]
// CHECK2:       invoke.cont:
// CHECK2-NEXT:    call void @llvm.dbg.declare(metadata double** [[VAR]], metadata [[META178:![0-9]+]], metadata !DIExpression()), !dbg [[DBG185:![0-9]+]]
// CHECK2-NEXT:    [[TMP3:%.*]] = load double*, double** [[VAR]], align 8, !dbg [[DBG186:![0-9]+]]
// CHECK2-NEXT:    [[TMP4:%.*]] = mul nsw i64 0, [[TMP1]], !dbg [[DBG186]]
// CHECK2-NEXT:    [[ARRAYIDX:%.*]] = getelementptr inbounds double, double* [[TMP3]], i64 [[TMP4]], !dbg [[DBG186]]
// CHECK2-NEXT:    [[ARRAYIDX1:%.*]] = getelementptr inbounds double, double* [[ARRAYIDX]], i64 0, !dbg [[DBG186]]
// CHECK2-NEXT:    ret void, !dbg [[DBG187:![0-9]+]]
// CHECK2:       terminate.lpad:
// CHECK2-NEXT:    [[TMP5:%.*]] = landingpad { i8*, i32 }
// CHECK2-NEXT:    catch i8* null, !dbg [[DBG177]]
// CHECK2-NEXT:    [[TMP6:%.*]] = extractvalue { i8*, i32 } [[TMP5]], 0, !dbg [[DBG177]]
// CHECK2-NEXT:    call void @__clang_call_terminate(i8* [[TMP6]]) #[[ATTR7]], !dbg [[DBG177]]
// CHECK2-NEXT:    unreachable, !dbg [[DBG177]]
//
//
// CHECK2-LABEL: define {{[^@]+}}@_Z3fooIPPcEvT_
// CHECK2-SAME: (i8** [[ARGC:%.*]]) #[[ATTR4]] comdat !dbg [[DBG188:![0-9]+]] {
// CHECK2-NEXT:  entry:
// CHECK2-NEXT:    [[ARGC_ADDR:%.*]] = alloca i8**, align 8
// CHECK2-NEXT:    store i8** [[ARGC]], i8*** [[ARGC_ADDR]], align 8
// CHECK2-NEXT:    call void @llvm.dbg.declare(metadata i8*** [[ARGC_ADDR]], metadata [[META191:![0-9]+]], metadata !DIExpression()), !dbg [[DBG192:![0-9]+]]
// CHECK2-NEXT:    ret void, !dbg [[DBG193:![0-9]+]]
//
//
// CHECK2-LABEL: define {{[^@]+}}@.omp_outlined..10
// CHECK2-SAME: (i32* noalias [[DOTGLOBAL_TID_:%.*]], i32* noalias [[DOTBOUND_TID_:%.*]], i8*** nonnull align 8 dereferenceable(8) [[ARGC:%.*]], i64 [[VLA:%.*]]) #[[ATTR3]] !dbg [[DBG194:![0-9]+]] {
// CHECK2-NEXT:  entry:
// CHECK2-NEXT:    [[DOTGLOBAL_TID__ADDR:%.*]] = alloca i32*, align 8
// CHECK2-NEXT:    [[DOTBOUND_TID__ADDR:%.*]] = alloca i32*, align 8
// CHECK2-NEXT:    [[ARGC_ADDR:%.*]] = alloca i8***, align 8
// CHECK2-NEXT:    [[VLA_ADDR:%.*]] = alloca i64, align 8
// CHECK2-NEXT:    store i32* [[DOTGLOBAL_TID_]], i32** [[DOTGLOBAL_TID__ADDR]], align 8
// CHECK2-NEXT:    call void @llvm.dbg.declare(metadata i32** [[DOTGLOBAL_TID__ADDR]], metadata [[META195:![0-9]+]], metadata !DIExpression()), !dbg [[DBG196:![0-9]+]]
// CHECK2-NEXT:    store i32* [[DOTBOUND_TID_]], i32** [[DOTBOUND_TID__ADDR]], align 8
// CHECK2-NEXT:    call void @llvm.dbg.declare(metadata i32** [[DOTBOUND_TID__ADDR]], metadata [[META197:![0-9]+]], metadata !DIExpression()), !dbg [[DBG196]]
// CHECK2-NEXT:    store i8*** [[ARGC]], i8**** [[ARGC_ADDR]], align 8
// CHECK2-NEXT:    call void @llvm.dbg.declare(metadata i8**** [[ARGC_ADDR]], metadata [[META198:![0-9]+]], metadata !DIExpression()), !dbg [[DBG196]]
// CHECK2-NEXT:    store i64 [[VLA]], i64* [[VLA_ADDR]], align 8
// CHECK2-NEXT:    call void @llvm.dbg.declare(metadata i64* [[VLA_ADDR]], metadata [[META199:![0-9]+]], metadata !DIExpression()), !dbg [[DBG196]]
// CHECK2-NEXT:    [[TMP0:%.*]] = load i8***, i8**** [[ARGC_ADDR]], align 8, !dbg [[DBG200:![0-9]+]]
// CHECK2-NEXT:    [[TMP1:%.*]] = load i64, i64* [[VLA_ADDR]], align 8, !dbg [[DBG200]]
// CHECK2-NEXT:    [[TMP2:%.*]] = load i32*, i32** [[DOTGLOBAL_TID__ADDR]], align 8, !dbg [[DBG200]]
// CHECK2-NEXT:    [[TMP3:%.*]] = load i32*, i32** [[DOTBOUND_TID__ADDR]], align 8, !dbg [[DBG200]]
// CHECK2-NEXT:    [[TMP4:%.*]] = load i8***, i8**** [[ARGC_ADDR]], align 8, !dbg [[DBG200]]
// CHECK2-NEXT:    call void @.omp_outlined._debug__.9(i32* [[TMP2]], i32* [[TMP3]], i8*** [[TMP4]], i64 [[TMP1]]) #[[ATTR6]], !dbg [[DBG200]]
// CHECK2-NEXT:    ret void, !dbg [[DBG200]]
//
//
// CHECK3-LABEL: define {{[^@]+}}@main
// CHECK3-SAME: (i32 [[ARGC:%.*]], i8** [[ARGV:%.*]]) #[[ATTR0:[0-9]+]] {
// CHECK3-NEXT:  entry:
// CHECK3-NEXT:    [[RETVAL:%.*]] = alloca i32, align 4
// CHECK3-NEXT:    [[ARGC_ADDR:%.*]] = alloca i32, align 4
// CHECK3-NEXT:    [[ARGV_ADDR:%.*]] = alloca i8**, align 8
// CHECK3-NEXT:    [[SAVED_STACK:%.*]] = alloca i8*, align 8
// CHECK3-NEXT:    [[__VLA_EXPR0:%.*]] = alloca i64, align 8
// CHECK3-NEXT:    store i32 0, i32* [[RETVAL]], align 4
// CHECK3-NEXT:    store i32 [[ARGC]], i32* [[ARGC_ADDR]], align 4
// CHECK3-NEXT:    store i8** [[ARGV]], i8*** [[ARGV_ADDR]], align 8
// CHECK3-NEXT:    [[TMP0:%.*]] = load i32, i32* [[ARGC_ADDR]], align 4
// CHECK3-NEXT:    [[TMP1:%.*]] = zext i32 [[TMP0]] to i64
// CHECK3-NEXT:    [[TMP2:%.*]] = call i8* @llvm.stacksave()
// CHECK3-NEXT:    store i8* [[TMP2]], i8** [[SAVED_STACK]], align 8
// CHECK3-NEXT:    [[VLA:%.*]] = alloca i32, i64 [[TMP1]], align 16
// CHECK3-NEXT:    store i64 [[TMP1]], i64* [[__VLA_EXPR0]], align 8
// CHECK3-NEXT:    [[OMP_GLOBAL_THREAD_NUM:%.*]] = call i32 @__kmpc_global_thread_num(%struct.ident_t* @[[GLOB1:[0-9]+]])
// CHECK3-NEXT:    br label [[OMP_PARALLEL:%.*]]
// CHECK3:       omp_parallel:
// CHECK3-NEXT:    call void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%struct.ident_t* @[[GLOB1]], i32 1, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, i32*)* @main..omp_par to void (i32*, i32*, ...)*), i32* [[VLA]])
// CHECK3-NEXT:    br label [[OMP_PAR_OUTLINED_EXIT:%.*]]
// CHECK3:       omp.par.outlined.exit:
// CHECK3-NEXT:    br label [[OMP_PAR_EXIT_SPLIT:%.*]]
// CHECK3:       omp.par.exit.split:
// CHECK3-NEXT:    [[TMP3:%.*]] = load i8**, i8*** [[ARGV_ADDR]], align 8
// CHECK3-NEXT:    [[CALL:%.*]] = call i32 @_Z5tmainIPPcEiT_(i8** [[TMP3]])
// CHECK3-NEXT:    store i32 [[CALL]], i32* [[RETVAL]], align 4
// CHECK3-NEXT:    [[TMP4:%.*]] = load i8*, i8** [[SAVED_STACK]], align 8
// CHECK3-NEXT:    call void @llvm.stackrestore(i8* [[TMP4]])
// CHECK3-NEXT:    [[TMP5:%.*]] = load i32, i32* [[RETVAL]], align 4
// CHECK3-NEXT:    ret i32 [[TMP5]]
//
//
// CHECK3-LABEL: define {{[^@]+}}@main..omp_par
// CHECK3-SAME: (i32* noalias [[TID_ADDR:%.*]], i32* noalias [[ZERO_ADDR:%.*]], i32* [[VLA:%.*]]) #[[ATTR1:[0-9]+]] {
// CHECK3-NEXT:  omp.par.entry:
// CHECK3-NEXT:    [[TID_ADDR_LOCAL:%.*]] = alloca i32, align 4
// CHECK3-NEXT:    [[TMP0:%.*]] = load i32, i32* [[TID_ADDR]], align 4
// CHECK3-NEXT:    store i32 [[TMP0]], i32* [[TID_ADDR_LOCAL]], align 4
// CHECK3-NEXT:    [[TID:%.*]] = load i32, i32* [[TID_ADDR_LOCAL]], align 4
// CHECK3-NEXT:    br label [[OMP_PAR_REGION:%.*]]
// CHECK3:       omp.par.region:
// CHECK3-NEXT:    [[ARRAYIDX:%.*]] = getelementptr inbounds i32, i32* [[VLA]], i64 1
// CHECK3-NEXT:    [[TMP1:%.*]] = load i32, i32* [[ARRAYIDX]], align 4
// CHECK3-NEXT:    call void @_Z3fooIiEvT_(i32 [[TMP1]])
// CHECK3-NEXT:    [[TMP2:%.*]] = load i32, i32* @global, align 4
// CHECK3-NEXT:    [[ARRAYIDX1:%.*]] = getelementptr inbounds i32, i32* [[VLA]], i64 1
// CHECK3-NEXT:    store i32 [[TMP2]], i32* [[ARRAYIDX1]], align 4
// CHECK3-NEXT:    br label [[OMP_PAR_PRE_FINALIZE:%.*]]
// CHECK3:       omp.par.pre_finalize:
// CHECK3-NEXT:    br label [[OMP_PAR_OUTLINED_EXIT_EXITSTUB:%.*]]
// CHECK3:       omp.par.outlined.exit.exitStub:
// CHECK3-NEXT:    ret void
//
//
// CHECK3-LABEL: define {{[^@]+}}@_Z3fooIiEvT_
// CHECK3-SAME: (i32 [[ARGC:%.*]]) #[[ATTR4:[0-9]+]] comdat {
// CHECK3-NEXT:  entry:
// CHECK3-NEXT:    [[ARGC_ADDR:%.*]] = alloca i32, align 4
// CHECK3-NEXT:    store i32 [[ARGC]], i32* [[ARGC_ADDR]], align 4
// CHECK3-NEXT:    ret void
//
//
// CHECK3-LABEL: define {{[^@]+}}@_Z5tmainIPPcEiT_
// CHECK3-SAME: (i8** [[ARGC:%.*]]) #[[ATTR5:[0-9]+]] comdat {
// CHECK3-NEXT:  entry:
// CHECK3-NEXT:    [[DOTRELOADED:%.*]] = alloca i64, align 8
// CHECK3-NEXT:    [[ARGC_ADDR:%.*]] = alloca i8**, align 8
// CHECK3-NEXT:    store i8** [[ARGC]], i8*** [[ARGC_ADDR]], align 8
// CHECK3-NEXT:    [[TMP0:%.*]] = load i8**, i8*** [[ARGC_ADDR]], align 8
// CHECK3-NEXT:    [[ARRAYIDX:%.*]] = getelementptr inbounds i8*, i8** [[TMP0]], i64 0
// CHECK3-NEXT:    [[TMP1:%.*]] = load i8*, i8** [[ARRAYIDX]], align 8
// CHECK3-NEXT:    [[ARRAYIDX1:%.*]] = getelementptr inbounds i8, i8* [[TMP1]], i64 0
// CHECK3-NEXT:    [[TMP2:%.*]] = load i8, i8* [[ARRAYIDX1]], align 1
// CHECK3-NEXT:    [[TMP3:%.*]] = zext i8 [[TMP2]] to i64
// CHECK3-NEXT:    [[OMP_GLOBAL_THREAD_NUM:%.*]] = call i32 @__kmpc_global_thread_num(%struct.ident_t* @[[GLOB1]])
// CHECK3-NEXT:    store i64 [[TMP3]], i64* [[DOTRELOADED]], align 8
// CHECK3-NEXT:    br label [[OMP_PARALLEL:%.*]]
// CHECK3:       omp_parallel:
// CHECK3-NEXT:    call void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%struct.ident_t* @[[GLOB1]], i32 2, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, i64*, i8***)* @_Z5tmainIPPcEiT_..omp_par to void (i32*, i32*, ...)*), i64* [[DOTRELOADED]], i8*** [[ARGC_ADDR]])
// CHECK3-NEXT:    br label [[OMP_PAR_OUTLINED_EXIT:%.*]]
// CHECK3:       omp.par.outlined.exit:
// CHECK3-NEXT:    br label [[OMP_PAR_EXIT_SPLIT:%.*]]
// CHECK3:       omp.par.exit.split:
// CHECK3-NEXT:    ret i32 0
//
//
// CHECK3-LABEL: define {{[^@]+}}@_Z5tmainIPPcEiT_..omp_par
// CHECK3-SAME: (i32* noalias [[TID_ADDR:%.*]], i32* noalias [[ZERO_ADDR:%.*]], i64* [[DOTRELOADED:%.*]], i8*** [[ARGC_ADDR:%.*]]) #[[ATTR1]] {
// CHECK3-NEXT:  omp.par.entry:
// CHECK3-NEXT:    [[TID_ADDR_LOCAL:%.*]] = alloca i32, align 4
// CHECK3-NEXT:    [[TMP0:%.*]] = load i32, i32* [[TID_ADDR]], align 4
// CHECK3-NEXT:    store i32 [[TMP0]], i32* [[TID_ADDR_LOCAL]], align 4
// CHECK3-NEXT:    [[TID:%.*]] = load i32, i32* [[TID_ADDR_LOCAL]], align 4
// CHECK3-NEXT:    [[TMP1:%.*]] = load i64, i64* [[DOTRELOADED]], align 8
// CHECK3-NEXT:    [[VAR:%.*]] = alloca double*, align 8
// CHECK3-NEXT:    br label [[OMP_PAR_REGION:%.*]]
// CHECK3:       omp.par.region:
// CHECK3-NEXT:    [[TMP2:%.*]] = load i8**, i8*** [[ARGC_ADDR]], align 8
// CHECK3-NEXT:    call void @_Z3fooIPPcEvT_(i8** [[TMP2]])
// CHECK3-NEXT:    [[TMP3:%.*]] = load double*, double** [[VAR]], align 8
// CHECK3-NEXT:    [[TMP4:%.*]] = mul nsw i64 0, [[TMP1]]
// CHECK3-NEXT:    [[ARRAYIDX2:%.*]] = getelementptr inbounds double, double* [[TMP3]], i64 [[TMP4]]
// CHECK3-NEXT:    [[ARRAYIDX3:%.*]] = getelementptr inbounds double, double* [[ARRAYIDX2]], i64 0
// CHECK3-NEXT:    br label [[OMP_PAR_PRE_FINALIZE:%.*]]
// CHECK3:       omp.par.pre_finalize:
// CHECK3-NEXT:    br label [[OMP_PAR_OUTLINED_EXIT_EXITSTUB:%.*]]
// CHECK3:       omp.par.outlined.exit.exitStub:
// CHECK3-NEXT:    ret void
//
//
// CHECK3-LABEL: define {{[^@]+}}@_Z3fooIPPcEvT_
// CHECK3-SAME: (i8** [[ARGC:%.*]]) #[[ATTR4]] comdat {
// CHECK3-NEXT:  entry:
// CHECK3-NEXT:    [[ARGC_ADDR:%.*]] = alloca i8**, align 8
// CHECK3-NEXT:    store i8** [[ARGC]], i8*** [[ARGC_ADDR]], align 8
// CHECK3-NEXT:    ret void
//
//
// CHECK4-LABEL: define {{[^@]+}}@main
// CHECK4-SAME: (i32 [[ARGC:%.*]], i8** [[ARGV:%.*]]) #[[ATTR0:[0-9]+]] !dbg [[DBG11:![0-9]+]] {
// CHECK4-NEXT:  entry:
// CHECK4-NEXT:    [[RETVAL:%.*]] = alloca i32, align 4
// CHECK4-NEXT:    [[ARGC_ADDR:%.*]] = alloca i32, align 4
// CHECK4-NEXT:    [[ARGV_ADDR:%.*]] = alloca i8**, align 8
// CHECK4-NEXT:    [[SAVED_STACK:%.*]] = alloca i8*, align 8
// CHECK4-NEXT:    [[__VLA_EXPR0:%.*]] = alloca i64, align 8
// CHECK4-NEXT:    store i32 0, i32* [[RETVAL]], align 4
// CHECK4-NEXT:    store i32 [[ARGC]], i32* [[ARGC_ADDR]], align 4
// CHECK4-NEXT:    call void @llvm.dbg.declare(metadata i32* [[ARGC_ADDR]], metadata [[META17:![0-9]+]], metadata !DIExpression()), !dbg [[DBG18:![0-9]+]]
// CHECK4-NEXT:    store i8** [[ARGV]], i8*** [[ARGV_ADDR]], align 8
// CHECK4-NEXT:    call void @llvm.dbg.declare(metadata i8*** [[ARGV_ADDR]], metadata [[META19:![0-9]+]], metadata !DIExpression()), !dbg [[DBG18]]
// CHECK4-NEXT:    [[TMP0:%.*]] = load i32, i32* [[ARGC_ADDR]], align 4, !dbg [[DBG20:![0-9]+]]
// CHECK4-NEXT:    [[TMP1:%.*]] = zext i32 [[TMP0]] to i64, !dbg [[DBG20]]
// CHECK4-NEXT:    [[TMP2:%.*]] = call i8* @llvm.stacksave(), !dbg [[DBG20]]
// CHECK4-NEXT:    store i8* [[TMP2]], i8** [[SAVED_STACK]], align 8, !dbg [[DBG20]]
// CHECK4-NEXT:    [[VLA:%.*]] = alloca i32, i64 [[TMP1]], align 16, !dbg [[DBG20]]
// CHECK4-NEXT:    store i64 [[TMP1]], i64* [[__VLA_EXPR0]], align 8, !dbg [[DBG20]]
// CHECK4-NEXT:    call void @llvm.dbg.declare(metadata i64* [[__VLA_EXPR0]], metadata [[META21:![0-9]+]], metadata !DIExpression()), !dbg [[DBG23:![0-9]+]]
// CHECK4-NEXT:    call void @llvm.dbg.declare(metadata i32* [[VLA]], metadata [[META24:![0-9]+]], metadata !DIExpression()), !dbg [[DBG20]]
// CHECK4-NEXT:    [[OMP_GLOBAL_THREAD_NUM:%.*]] = call i32 @__kmpc_global_thread_num(%struct.ident_t* @[[GLOB1:[0-9]+]]), !dbg [[DBG28:![0-9]+]]
// CHECK4-NEXT:    br label [[OMP_PARALLEL:%.*]]
// CHECK4:       omp_parallel:
// CHECK4-NEXT:    call void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%struct.ident_t* @[[GLOB1]], i32 1, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, i32*)* @main..omp_par to void (i32*, i32*, ...)*), i32* [[VLA]]), !dbg [[DBG29:![0-9]+]]
// CHECK4-NEXT:    br label [[OMP_PAR_OUTLINED_EXIT:%.*]]
// CHECK4:       omp.par.outlined.exit:
// CHECK4-NEXT:    br label [[OMP_PAR_EXIT_SPLIT:%.*]]
// CHECK4:       omp.par.exit.split:
// CHECK4-NEXT:    [[TMP3:%.*]] = load i8**, i8*** [[ARGV_ADDR]], align 8, !dbg [[DBG30:![0-9]+]]
// CHECK4-NEXT:    [[CALL:%.*]] = call i32 @_Z5tmainIPPcEiT_(i8** [[TMP3]]), !dbg [[DBG30]]
// CHECK4-NEXT:    store i32 [[CALL]], i32* [[RETVAL]], align 4, !dbg [[DBG30]]
// CHECK4-NEXT:    [[TMP4:%.*]] = load i8*, i8** [[SAVED_STACK]], align 8, !dbg [[DBG31:![0-9]+]]
// CHECK4-NEXT:    call void @llvm.stackrestore(i8* [[TMP4]]), !dbg [[DBG31]]
// CHECK4-NEXT:    [[TMP5:%.*]] = load i32, i32* [[RETVAL]], align 4, !dbg [[DBG31]]
// CHECK4-NEXT:    ret i32 [[TMP5]], !dbg [[DBG31]]
//
//
// CHECK4-LABEL: define {{[^@]+}}@main..omp_par
// CHECK4-SAME: (i32* noalias [[TID_ADDR:%.*]], i32* noalias [[ZERO_ADDR:%.*]], i32* [[VLA:%.*]]) #[[ATTR1:[0-9]+]] !dbg [[DBG32:![0-9]+]] {
// CHECK4-NEXT:  omp.par.entry:
// CHECK4-NEXT:    [[TID_ADDR_LOCAL:%.*]] = alloca i32, align 4
// CHECK4-NEXT:    [[TMP0:%.*]] = load i32, i32* [[TID_ADDR]], align 4
// CHECK4-NEXT:    store i32 [[TMP0]], i32* [[TID_ADDR_LOCAL]], align 4
// CHECK4-NEXT:    [[TID:%.*]] = load i32, i32* [[TID_ADDR_LOCAL]], align 4
// CHECK4-NEXT:    br label [[OMP_PAR_REGION:%.*]]
// CHECK4:       omp.par.region:
// CHECK4-NEXT:    [[ARRAYIDX:%.*]] = getelementptr inbounds i32, i32* [[VLA]], i64 1, !dbg [[DBG34:![0-9]+]]
// CHECK4-NEXT:    [[TMP1:%.*]] = load i32, i32* [[ARRAYIDX]], align 4, !dbg [[DBG34]]
// CHECK4-NEXT:    call void @_Z3fooIiEvT_(i32 [[TMP1]]), !dbg [[DBG34]]
// CHECK4-NEXT:    [[TMP2:%.*]] = load i32, i32* @global, align 4, !dbg [[DBG34]]
// CHECK4-NEXT:    [[ARRAYIDX1:%.*]] = getelementptr inbounds i32, i32* [[VLA]], i64 1, !dbg [[DBG34]]
// CHECK4-NEXT:    store i32 [[TMP2]], i32* [[ARRAYIDX1]], align 4, !dbg [[DBG34]]
// CHECK4-NEXT:    br label [[OMP_PAR_PRE_FINALIZE:%.*]], !dbg [[DBG34]]
// CHECK4:       omp.par.pre_finalize:
// CHECK4-NEXT:    br label [[OMP_PAR_OUTLINED_EXIT_EXITSTUB:%.*]], !dbg [[DBG34]]
// CHECK4:       omp.par.outlined.exit.exitStub:
// CHECK4-NEXT:    ret void
//
//
// CHECK4-LABEL: define {{[^@]+}}@_Z3fooIiEvT_
// CHECK4-SAME: (i32 [[ARGC:%.*]]) #[[ATTR5:[0-9]+]] comdat !dbg [[DBG35:![0-9]+]] {
// CHECK4-NEXT:  entry:
// CHECK4-NEXT:    [[ARGC_ADDR:%.*]] = alloca i32, align 4
// CHECK4-NEXT:    store i32 [[ARGC]], i32* [[ARGC_ADDR]], align 4
// CHECK4-NEXT:    call void @llvm.dbg.declare(metadata i32* [[ARGC_ADDR]], metadata [[META40:![0-9]+]], metadata !DIExpression()), !dbg [[DBG41:![0-9]+]]
// CHECK4-NEXT:    ret void, !dbg [[DBG41]]
//
//
// CHECK4-LABEL: define {{[^@]+}}@_Z5tmainIPPcEiT_
// CHECK4-SAME: (i8** [[ARGC:%.*]]) #[[ATTR6:[0-9]+]] comdat !dbg [[DBG44:![0-9]+]] {
// CHECK4-NEXT:  entry:
// CHECK4-NEXT:    [[DOTRELOADED:%.*]] = alloca i64, align 8
// CHECK4-NEXT:    [[ARGC_ADDR:%.*]] = alloca i8**, align 8
// CHECK4-NEXT:    store i8** [[ARGC]], i8*** [[ARGC_ADDR]], align 8
// CHECK4-NEXT:    call void @llvm.dbg.declare(metadata i8*** [[ARGC_ADDR]], metadata [[META49:![0-9]+]], metadata !DIExpression()), !dbg [[DBG50:![0-9]+]]
// CHECK4-NEXT:    [[TMP0:%.*]] = load i8**, i8*** [[ARGC_ADDR]], align 8, !dbg [[DBG51:![0-9]+]]
// CHECK4-NEXT:    [[ARRAYIDX:%.*]] = getelementptr inbounds i8*, i8** [[TMP0]], i64 0, !dbg [[DBG51]]
// CHECK4-NEXT:    [[TMP1:%.*]] = load i8*, i8** [[ARRAYIDX]], align 8, !dbg [[DBG51]]
// CHECK4-NEXT:    [[ARRAYIDX1:%.*]] = getelementptr inbounds i8, i8* [[TMP1]], i64 0, !dbg [[DBG51]]
// CHECK4-NEXT:    [[TMP2:%.*]] = load i8, i8* [[ARRAYIDX1]], align 1, !dbg [[DBG51]]
// CHECK4-NEXT:    [[TMP3:%.*]] = zext i8 [[TMP2]] to i64, !dbg [[DBG51]]
// CHECK4-NEXT:    [[OMP_GLOBAL_THREAD_NUM:%.*]] = call i32 @__kmpc_global_thread_num(%struct.ident_t* @[[GLOB3:[0-9]+]]), !dbg [[DBG52:![0-9]+]]
// CHECK4-NEXT:    store i64 [[TMP3]], i64* [[DOTRELOADED]], align 8
// CHECK4-NEXT:    br label [[OMP_PARALLEL:%.*]]
// CHECK4:       omp_parallel:
// CHECK4-NEXT:    call void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%struct.ident_t* @[[GLOB3]], i32 2, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, i64*, i8***)* @_Z5tmainIPPcEiT_..omp_par to void (i32*, i32*, ...)*), i64* [[DOTRELOADED]], i8*** [[ARGC_ADDR]]), !dbg [[DBG53:![0-9]+]]
// CHECK4-NEXT:    br label [[OMP_PAR_OUTLINED_EXIT:%.*]]
// CHECK4:       omp.par.outlined.exit:
// CHECK4-NEXT:    br label [[OMP_PAR_EXIT_SPLIT:%.*]]
// CHECK4:       omp.par.exit.split:
// CHECK4-NEXT:    ret i32 0, !dbg [[DBG55:![0-9]+]]
//
//
// CHECK4-LABEL: define {{[^@]+}}@_Z5tmainIPPcEiT_..omp_par
// CHECK4-SAME: (i32* noalias [[TID_ADDR:%.*]], i32* noalias [[ZERO_ADDR:%.*]], i64* [[DOTRELOADED:%.*]], i8*** [[ARGC_ADDR:%.*]]) #[[ATTR1]] !dbg [[DBG56:![0-9]+]] {
// CHECK4-NEXT:  omp.par.entry:
// CHECK4-NEXT:    [[TID_ADDR_LOCAL:%.*]] = alloca i32, align 4
// CHECK4-NEXT:    [[TMP0:%.*]] = load i32, i32* [[TID_ADDR]], align 4
// CHECK4-NEXT:    store i32 [[TMP0]], i32* [[TID_ADDR_LOCAL]], align 4
// CHECK4-NEXT:    [[TID:%.*]] = load i32, i32* [[TID_ADDR_LOCAL]], align 4
// CHECK4-NEXT:    [[TMP1:%.*]] = load i64, i64* [[DOTRELOADED]], align 8
// CHECK4-NEXT:    [[VAR:%.*]] = alloca double*, align 8
// CHECK4-NEXT:    br label [[OMP_PAR_REGION:%.*]]
// CHECK4:       omp.par.region:
// CHECK4-NEXT:    [[TMP2:%.*]] = load i8**, i8*** [[ARGC_ADDR]], align 8, !dbg [[DBG57:![0-9]+]]
// CHECK4-NEXT:    call void @_Z3fooIPPcEvT_(i8** [[TMP2]]), !dbg [[DBG57]]
// CHECK4-NEXT:    call void @llvm.dbg.declare(metadata double** [[VAR]], metadata [[META58:![0-9]+]], metadata !DIExpression()), !dbg [[DBG65:![0-9]+]]
// CHECK4-NEXT:    [[TMP3:%.*]] = load double*, double** [[VAR]], align 8, !dbg [[DBG65]]
// CHECK4-NEXT:    [[TMP4:%.*]] = mul nsw i64 0, [[TMP1]], !dbg [[DBG65]]
// CHECK4-NEXT:    [[ARRAYIDX2:%.*]] = getelementptr inbounds double, double* [[TMP3]], i64 [[TMP4]], !dbg [[DBG65]]
// CHECK4-NEXT:    [[ARRAYIDX3:%.*]] = getelementptr inbounds double, double* [[ARRAYIDX2]], i64 0, !dbg [[DBG65]]
// CHECK4-NEXT:    br label [[OMP_PAR_PRE_FINALIZE:%.*]], !dbg [[DBG66:![0-9]+]]
// CHECK4:       omp.par.pre_finalize:
// CHECK4-NEXT:    br label [[OMP_PAR_OUTLINED_EXIT_EXITSTUB:%.*]], !dbg [[DBG66]]
// CHECK4:       omp.par.outlined.exit.exitStub:
// CHECK4-NEXT:    ret void
//
//
// CHECK4-LABEL: define {{[^@]+}}@_Z3fooIPPcEvT_
// CHECK4-SAME: (i8** [[ARGC:%.*]]) #[[ATTR5]] comdat !dbg [[DBG67:![0-9]+]] {
// CHECK4-NEXT:  entry:
// CHECK4-NEXT:    [[ARGC_ADDR:%.*]] = alloca i8**, align 8
// CHECK4-NEXT:    store i8** [[ARGC]], i8*** [[ARGC_ADDR]], align 8
// CHECK4-NEXT:    call void @llvm.dbg.declare(metadata i8*** [[ARGC_ADDR]], metadata [[META70:![0-9]+]], metadata !DIExpression()), !dbg [[DBG71:![0-9]+]]
// CHECK4-NEXT:    ret void, !dbg [[DBG71]]
//
//