[OpenMP][FIX] Consistently use OpenMPIRBuilder if requested

[jdoerfert]

Summary

When we use the OpenMPIRBuilder for the parallel region we need to also
use it to get the thread ID (among other things) in the body. This is
because CGOpenMPRuntime::getThreadID() and
CGOpenMPRuntime::emitUpdateLocation implicitly assumes that if they are
called from within a parallel region there is a certain structure to the
code and certain members of the OMPRegionInfo are initialized. It might
make sense to initialize them even if we use the OpenMPIRBuilder but we
would preferably get rid of such state instead.

Bug reported by Anchu Rajendran Sudhakumari.

Depends on D82470.

[anchu-rajendran]

LGTM :). Thanks!

clang/lib/CodeGen/CGOpenMPRuntime.cpp

1547

Nit: I think its better to add comments on the implicit assumptions in clang code / on why initializations here is not required if OMPIRBuilder is not used.

[fghanim]

LGTM

[jdoerfert]

Rebase

[davezarzycki]

Hello, I have a twice-daily auto-bisecting multi-stage cron job running on Fedora 32 (x86-64) that has identified this commit as failing my first stage test (release + no asserts). Is this expected? Can we get a quick fix or revert this?

FAIL: Clang :: OpenMP/irbuilder_nested_parallel_for.c (13233 of 67864)

• TEST 'Clang :: OpenMP/irbuilder_nested_parallel_for.c' FAILED ****

Script:

: 'RUN: at line 2'; /tmp/_update_lc/r/bin/clang -cc1 -internal-isystem /tmp/_update_lc/r/lib/clang/12.0.0/include -nostdsysteminc -verify -fopenmp -fopenmp-enable-irbuilder -x c++ -emit-llvm /home/dave/ro_s/lp/clang/test/OpenMP/irbuilder_nested_parallel_for.c -triple x86_64-unknown-unknown -fexceptions -fcxx-exceptions -o - | /tmp/_update_lc/r/bin/FileCheck --check-prefixes=CHECK /home/dave/ro_s/lp/clang/test/OpenMP/irbuilder_nested_parallel_for.c

: 'RUN: at line 3'; /tmp/_update_lc/r/bin/clang -cc1 -internal-isystem /tmp/_update_lc/r/lib/clang/12.0.0/include -nostdsysteminc -fopenmp -fopenmp-enable-irbuilder -x c++ -triple x86_64-unknown-unknown -fexceptions -fcxx-exceptions -debug-info-kind=limited -std=c++11 -verify /home/dave/ro_s/lp/clang/test/OpenMP/irbuilder_nested_parallel_for.c -emit-llvm -o - | /tmp/_update_lc/r/bin/FileCheck --check-prefixes=CHECK-DEBUG /home/dave/ro_s/lp/clang/test/OpenMP/irbuilder_nested_parallel_for.c

Exit Code: 1

Command Output (stderr):

/home/dave/ro_s/lp/clang/test/OpenMP/irbuilder_nested_parallel_for.c:74:22: error: CHECK-DEBUG-NEXT: is not on the line after the previous match
// CHECK-DEBUG-NEXT: [[OMP_GLOBAL_THREAD_NUM:%.*]] = call i32 @__kmpc_global_thread_num(%struct.ident_t* @12), !dbg !{{[0-9]*}}

^

<stdin>:252:2: note: 'next' match was here
%omp_global_thread_num1 = call i32 @__kmpc_global_thread_num(%struct.ident_t* @12), !dbg !52
^
<stdin>:224:105: note: previous match ended here
call void @llvm.dbg.declare(metadata double* %b.addr, metadata !45, metadata !DIExpression()), !dbg !46

                        ^

<stdin>:225:1: note: non-matching line after previous match is here
%omp_global_thread_num = call i32 @__kmpc_global_thread_num(%struct.ident_t* @10), !dbg !47
^
/home/dave/ro_s/lp/clang/test/OpenMP/irbuilder_nested_parallel_for.c:192:22: error: CHECK-DEBUG-NEXT: expected string not found in input
// CHECK-DEBUG-NEXT: [[OMP_GLOBAL_THREAD_NUM:%.*]] = call i32 @__kmpc_global_thread_num(%struct.ident_t* @25), !dbg !{{[0-9]*}}

^

<stdin>:381:2: note: scanning from here
%omp_global_thread_num = call i32 @__kmpc_global_thread_num(%struct.ident_t* @21), !dbg !83
^
<stdin>:381:19: note: possible intended match here
%omp_global_thread_num = call i32 @__kmpc_global_thread_num(%struct.ident_t* @21), !dbg !83

^

Input file: <stdin>
Check file: /home/dave/ro_s/lp/clang/test/OpenMP/irbuilder_nested_parallel_for.c

-dump-input=help explains the following input dump.

Input was:
<<<<<<

  .
  .
  .
247:
248: omp.par.outlined.exit19.exitStub: ; preds = %omp.par.pre_finalize
249:  ret void
250:
251: omp.par.region: ; preds = %omp.par.entry
252:  %omp_global_thread_num1 = call i32 @__kmpc_global_thread_num(%struct.ident_t* @12), !dbg !52

next:74 !~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ error: match on wrong line

253:  br label %omp_parallel
254:
255: omp_parallel: ; preds = %omp.par.region
256:  call void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%struct.ident_t* @12, i32 3, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, i32*, double*, float**)* @_Z14parallel_for_1Pfid..omp_par to void (i32*, i32*, ...)*), i32* %a.addr, double* %b.addr, float** %r.addr), !dbg !53
257:  br label %omp.par.outlined.exit
  .
  .
  .
376:  call void @llvm.dbg.declare(metadata float** %r.addr, metadata !77, metadata !DIExpression()), !dbg !78
377:  store i32 %a, i32* %a.addr, align 4
378:  call void @llvm.dbg.declare(metadata i32* %a.addr, metadata !79, metadata !DIExpression()), !dbg !80
379:  store double %b, double* %b.addr, align 8
380:  call void @llvm.dbg.declare(metadata double* %b.addr, metadata !81, metadata !DIExpression()), !dbg !82
381:  %omp_global_thread_num = call i32 @__kmpc_global_thread_num(%struct.ident_t* @21), !dbg !83

next:192'0 X~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ error: no match found
next:192'1 ? possible intended match

382:  br label %omp_parallel

next:192'0 ~~~~~~~~~~~~~~~~~~~~~~~

383:

next:192'0 ~

384: omp_parallel: ; preds = %entry

next:192'0 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

385:  call void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%struct.ident_t* @21, i32 3, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, i32*, double*, float**)* @_Z14parallel_for_2Pfid..omp_par.4 to void (i32*, i32*, ...)*), i32* %a.addr, double* %b.addr, float** %r.addr), !dbg !84

next:192'0 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

386:  br label %omp.par.outlined.exit211

next:192'0 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

.
.
.

---

Testing: 0.. 10.. 20.. 30.. 40.. 50.. 60.. 70.. 80.. 90..

---

Failed Tests (1):

Clang :: OpenMP/irbuilder_nested_parallel_for.c

Testing Time: 124.68s

Unsupported      : 10711
Passed           : 57049
Expectedly Failed:   103
Failed           :     1

[jdoerfert]

@davezarzycki The bots reported this as well, didn't build MLIR locally, this should have been fixed by 4d83aa4771d84940626d86c883193af390812281

[hoyFB]

In D82822#2187576, @jdoerfert wrote:

@davezarzycki The bots reported this as well, didn't build MLIR locally, this should have been fixed by 4d83aa4771d84940626d86c883193af390812281

I was seeing the same issue with what @davezarzycki has encountered. Not sure why this is related to MLIR. The failure is cause by this line:

74: // CHECK-DEBUG-NEXT:    [[OMP_GLOBAL_THREAD_NUM:%.*]] = call i32 @__kmpc_global_thread_num(%struct.ident_t* **@12**), !dbg !{{[0-9]*}}

where @12 is supposed to be defined as

@11 = private unnamed_addr constant [90 x i8] c";llvm-project/clang/test/OpenMP/irbuilder_nested_parallel_for.c;parallel_for_1;85;1;;\00", align 1
@12 = private unnamed_addr global %struct.ident_t { i32 0, i32 2, i32 0, i32 0, i8* getelementptr inbounds ([90 x i8], [90 x i8]* @11, i32 0, i32 0) }, align 8

The exact match of @12 seems a bit fragile to me since global data names could be changed from time to time. Can you please make it bit more robust? Thanks.

[jdoerfert]

In D82822#2189366, @hoyFB wrote:

In D82822#2187576, @jdoerfert wrote:

@davezarzycki The bots reported this as well, didn't build MLIR locally, this should have been fixed by 4d83aa4771d84940626d86c883193af390812281

I was seeing the same issue with what @davezarzycki has encountered. Not sure why this is related to MLIR. The failure is cause by this line:

74: // CHECK-DEBUG-NEXT:    [[OMP_GLOBAL_THREAD_NUM:%.*]] = call i32 @__kmpc_global_thread_num(%struct.ident_t* **@12**), !dbg !{{[0-9]*}}

where @12 is supposed to be defined as

@11 = private unnamed_addr constant [90 x i8] c";llvm-project/clang/test/OpenMP/irbuilder_nested_parallel_for.c;parallel_for_1;85;1;;\00", align 1
@12 = private unnamed_addr global %struct.ident_t { i32 0, i32 2, i32 0, i32 0, i8* getelementptr inbounds ([90 x i8], [90 x i8]* @11, i32 0, i32 0) }, align 8

The exact match of @12 seems a bit fragile to me since global data names could be changed from time to time. Can you please make it bit more robust? Thank

I just created D85099 to eliminate the problem permanently. Feel free to take a look.

[hoyFB]

In D82822#2189692, @jdoerfert wrote:

In D82822#2189366, @hoyFB wrote:

In D82822#2187576, @jdoerfert wrote:

@davezarzycki The bots reported this as well, didn't build MLIR locally, this should have been fixed by 4d83aa4771d84940626d86c883193af390812281

I was seeing the same issue with what @davezarzycki has encountered. Not sure why this is related to MLIR. The failure is cause by this line:

74: // CHECK-DEBUG-NEXT:    [[OMP_GLOBAL_THREAD_NUM:%.*]] = call i32 @__kmpc_global_thread_num(%struct.ident_t* **@12**), !dbg !{{[0-9]*}}

where @12 is supposed to be defined as

@11 = private unnamed_addr constant [90 x i8] c";llvm-project/clang/test/OpenMP/irbuilder_nested_parallel_for.c;parallel_for_1;85;1;;\00", align 1
@12 = private unnamed_addr global %struct.ident_t { i32 0, i32 2, i32 0, i32 0, i8* getelementptr inbounds ([90 x i8], [90 x i8]* @11, i32 0, i32 0) }, align 8

The exact match of @12 seems a bit fragile to me since global data names could be changed from time to time. Can you please make it bit more robust? Thank

I just created D85099 to eliminate the problem permanently. Feel free to take a look.

Thanks for the quick turnaround!

clang/lib/CodeGen/CGOpenMPRuntime.cpp

void CGOpenMPRuntime::clearLocThreadIdInsertPt(CodeGenFunction &CGF) {
   auto &Elem = OpenMPLocThreadIDMap.FindAndConstruct(CGF.CurFn);
   if (Elem.second.ServiceInsertPt) {
     llvm::Instruction *Ptr = Elem.second.ServiceInsertPt;
     Elem.second.ServiceInsertPt = nullptr;
     Ptr->eraseFromParent();
   }
 }
 
+static StringRef getIdentStringFromSourceLocation(CodeGenFunction &CGF,
+                                                  SourceLocation Loc,
+                                                  SmallString<128> &Buffer) {
+  llvm::raw_svector_ostream OS(Buffer);
+  // Build debug location
+  PresumedLoc PLoc = CGF.getContext().getSourceManager().getPresumedLoc(Loc);
+  OS << ";" << PLoc.getFilename() << ";";
+  if (const auto *FD = dyn_cast_or_null<FunctionDecl>(CGF.CurFuncDecl))
+    OS << FD->getQualifiedNameAsString();
+  OS << ";" << PLoc.getLine() << ";" << PLoc.getColumn() << ";;";
+  return OS.str();
+}
+
 llvm::Value *CGOpenMPRuntime::emitUpdateLocation(CodeGenFunction &CGF,
                                                  SourceLocation Loc,
                                                  unsigned Flags) {
   Flags |= OMP_IDENT_KMPC;
   // If no debug info is generated - return global default location.
   if (CGM.getCodeGenOpts().getDebugInfo() == codegenoptions::NoDebugInfo ||
       Loc.isInvalid())
     return getOrCreateDefaultLocation(Flags).getPointer();
 
+  // If the OpenMPIRBuilder is used we need to use it for all location handling
+  // as the clang invariants used below might be broken.
+  if (CGM.getLangOpts().OpenMPIRBuilder) {
+    SmallString<128> Buffer;
+    OMPBuilder.updateToLocation(CGF.Builder.saveIP());
+    auto *SrcLocStr = OMPBuilder.getOrCreateSrcLocStr(
+        getIdentStringFromSourceLocation(CGF, Loc, Buffer));
+    return OMPBuilder.getOrCreateIdent(SrcLocStr, IdentFlag(Flags));
+  }
+
   assert(CGF.CurFn && "No function in current CodeGenFunction.");
 
   CharUnits Align = CGM.getContext().getTypeAlignInChars(IdentQTy);
   Address LocValue = Address::invalid();
   auto I = OpenMPLocThreadIDMap.find(CGF.CurFn);
   if (I != OpenMPLocThreadIDMap.end())
     LocValue = Address(I->second.DebugLoc, Align);
 
llvm::Value *CGOpenMPRuntime::emitUpdateLocation(CodeGenFunction &CGF,
   // char **psource = &.kmpc_loc_<flags>.addr.psource;
   LValue Base = CGF.MakeAddrLValue(LocValue, IdentQTy);
   auto Fields = cast<RecordDecl>(IdentQTy->getAsTagDecl())->field_begin();
   LValue PSource =
       CGF.EmitLValueForField(Base, *std::next(Fields, IdentField_PSource));
 
   llvm::Value *OMPDebugLoc = OpenMPDebugLocMap.lookup(Loc.getRawEncoding());
   if (OMPDebugLoc == nullptr) {
-    SmallString<128> Buffer2;
-    llvm::raw_svector_ostream OS2(Buffer2);
-    // Build debug location
-    PresumedLoc PLoc = CGF.getContext().getSourceManager().getPresumedLoc(Loc);
-    OS2 << ";" << PLoc.getFilename() << ";";
-    if (const auto *FD = dyn_cast_or_null<FunctionDecl>(CGF.CurFuncDecl))
-      OS2 << FD->getQualifiedNameAsString();
-    OS2 << ";" << PLoc.getLine() << ";" << PLoc.getColumn() << ";;";
-    OMPDebugLoc = CGF.Builder.CreateGlobalStringPtr(OS2.str());
+    SmallString<128> Buffer;
+    OMPDebugLoc = CGF.Builder.CreateGlobalStringPtr(
+        getIdentStringFromSourceLocation(CGF, Loc, Buffer));
     OpenMPDebugLocMap[Loc.getRawEncoding()] = OMPDebugLoc;
   }
   // *psource = ";<File>;<Function>;<Line>;<Column>;;";
   CGF.EmitStoreOfScalar(OMPDebugLoc, PSource);
 
   // Our callers always pass this to a runtime function, so for
   // convenience, go ahead and return a naked pointer.
   return LocValue.getPointer();
 }
 
 llvm::Value *CGOpenMPRuntime::getThreadID(CodeGenFunction &CGF,
                                           SourceLocation Loc) {
   assert(CGF.CurFn && "No function in current CodeGenFunction.");
+  // If the OpenMPIRBuilder is used we need to use it for all thread id calls as
+  // the clang invariants used below might be broken.
+  if (CGM.getLangOpts().OpenMPIRBuilder) {
+    SmallString<128> Buffer;
+    OMPBuilder.updateToLocation(CGF.Builder.saveIP());
+    auto *SrcLocStr = OMPBuilder.getOrCreateSrcLocStr(
+        getIdentStringFromSourceLocation(CGF, Loc, Buffer));
+    return OMPBuilder.getOrCreateThreadID(
+        OMPBuilder.getOrCreateIdent(SrcLocStr));
+  }
 
   llvm::Value *ThreadID = nullptr;
   // Check whether we've already cached a load of the thread id in this
   // function.
   auto I = OpenMPLocThreadIDMap.find(CGF.CurFn);
   if (I != OpenMPLocThreadIDMap.end()) {
     ThreadID = I->second.ThreadID;
     if (ThreadID != nullptr)

clang/test/OpenMP/cancel_codegen.cpp

 // RUN: %clang_cc1 -fopenmp-simd -fopenmp-version=45 -std=c++11 -include-pch %t -fsyntax-only -verify %s -triple x86_64-apple-darwin13.4.0 -emit-llvm -o - | FileCheck --check-prefix SIMD-ONLY0 %s
 // SIMD-ONLY0-NOT: {{__kmpc|__tgt}}
 // expected-no-diagnostics
 #ifndef HEADER
 #define HEADER
 
 float flag;
 int main (int argc, char **argv) {
-// ALL: [[GTID:%.+]] = call i32 @__kmpc_global_thread_num(
 #pragma omp parallel
 {
 #pragma omp cancel parallel if(flag)
   argv[0][0] = argc;
 #pragma omp barrier
   argv[0][0] += argc;
 }
 // ALL: call void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(

 {
 #pragma omp cancel sections
 #pragma omp section
   {
 #pragma omp cancel sections
   }
 }
 // ALL: call void @__kmpc_for_static_init_4(
-// ALL: [[RES:%.+]] = call i32 @__kmpc_cancel(%struct.ident_t* {{[^,]+}}, i32 [[GTID]], i32 3)
+// ALL: [[RES:%.+]] = call i32 @__kmpc_cancel(%struct.ident_t* {{[^,]+}}, i32 [[GTID:%.*]], i32 3)
 // ALL: [[CMP:%.+]] = icmp ne i32 [[RES]], 0
 // ALL: br i1 [[CMP]], label %[[EXIT:[^,].+]], label %[[CONTINUE:.+]]
 // ALL: [[EXIT]]
 // ALL: br label
 // ALL: [[CONTINUE]]
 // ALL: br label
-// ALL: [[RES:%.+]] = call i32 @__kmpc_cancel(%struct.ident_t* {{[^,]+}}, i32 [[GTID]], i32 3)
+// ALL: [[RES:%.+]] = call i32 @__kmpc_cancel(%struct.ident_t* {{[^,]+}}, i32 [[GTID:%.*]], i32 3)
 // ALL: [[CMP:%.+]] = icmp ne i32 [[RES]], 0
 // ALL: br i1 [[CMP]], label %[[EXIT:[^,].+]], label %[[CONTINUE:.+]]
 // ALL: [[EXIT]]
 // ALL: br label
 // ALL: [[CONTINUE]]
 // ALL: br label
 // ALL: call void @__kmpc_for_static_fini(
 #pragma omp for
 for (int i = 0; i < argc; ++i) {
 #pragma omp cancel for if(cancel: flag)
 }
 // ALL: call void @__kmpc_for_static_init_4(
 // ALL: [[FLAG:%.+]] = load float, float* @{{.+}},
 // ALL: [[BOOL:%.+]] = fcmp une float [[FLAG]], 0.000000e+00
 // ALL: br i1 [[BOOL]], label %[[THEN:[^,]+]], label %[[ELSE:[^,]+]]
 // ALL: [[THEN]]
-// ALL: [[RES:%.+]] = call i32 @__kmpc_cancel(%struct.ident_t* {{[^,]+}}, i32 [[GTID]], i32 2)
+// ALL: [[RES:%.+]] = call i32 @__kmpc_cancel(%struct.ident_t* {{[^,]+}}, i32 [[GTID:%.*]], i32 2)
 // ALL: [[CMP:%.+]] = icmp ne i32 [[RES]], 0
 // ALL: br i1 [[CMP]], label %[[EXIT:[^,].+]], label %[[CONTINUE:.+]]
 // ALL: [[EXIT]]
 // ALL: br label
 // ALL: [[CONTINUE]]
 // ALL: br label
 // ALL: [[ELSE]]
 // ALL: br label

 // CHECK: call void @__kmpc_for_static_init_4(
 // CHECK: [[RES:%.+]] = call i32 @__kmpc_cancel(%struct.ident_t* {{[^,]+}}, i32 [[GTID:%.+]], i32 3)
 // CHECK: [[CMP:%.+]] = icmp ne i32 [[RES]], 0
 // CHECK: br i1 [[CMP]], label %[[EXIT:[^,].+]], label %[[CONTINUE:.+]]
 // CHECK: [[EXIT]]
 // CHECK: br label
 // CHECK: [[CONTINUE]]
 // CHECK: br label
-// CHECK: [[RES:%.+]] = call i32 @__kmpc_cancel(%struct.ident_t* {{[^,]+}}, i32 [[GTID]], i32 3)
+// CHECK: [[RES:%.+]] = call i32 @__kmpc_cancel(%struct.ident_t* {{[^,]+}}, i32 [[GTID:%.*]], i32 3)
 // CHECK: [[CMP:%.+]] = icmp ne i32 [[RES]], 0
 // CHECK: br i1 [[CMP]], label %[[EXIT:[^,].+]], label %[[CONTINUE:.+]]
 // CHECK: [[EXIT]]
 // CHECK: br label
 // CHECK: [[CONTINUE]]
 // CHECK: br label
 // CHECK: call void @__kmpc_for_static_fini(
 // CHECK: ret void

clang/test/OpenMP/irbuilder_nested_parallel_for.c

+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py
+// RUN: %clang_cc1 -verify -fopenmp -fopenmp-enable-irbuilder -x c++ -emit-llvm %s -triple x86_64-unknown-unknown -fexceptions -fcxx-exceptions -o - | FileCheck --check-prefixes=CHECK %s
+// RUN: %clang_cc1 -fopenmp -fopenmp-enable-irbuilder -x c++ -triple x86_64-unknown-unknown -fexceptions -fcxx-exceptions -debug-info-kind=limited -std=c++11 -verify %s -emit-llvm -o - | FileCheck --check-prefixes=CHECK-DEBUG %s
+
+// expected-no-diagnostics
+
+// TODO: Teach the update script to check new functions too.
+
+#ifndef HEADER
+#define HEADER
+
+// CHECK-LABEL: @_Z14parallel_for_0v(
+// CHECK-NEXT:  entry:
+// CHECK-NEXT:    [[OMP_GLOBAL_THREAD_NUM:%.*]] = call i32 @__kmpc_global_thread_num(%struct.ident_t* @1)
+// CHECK-NEXT:    br label [[OMP_PARALLEL:%.*]]
+// CHECK:       omp_parallel:
+// CHECK-NEXT:    call void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%struct.ident_t* @1, i32 0, void (i32*, i32*, ...)* bitcast (void (i32*, i32*)* @_Z14parallel_for_0v..omp_par to void (i32*, i32*, ...)*))
+// CHECK-NEXT:    br label [[OMP_PAR_OUTLINED_EXIT:%.*]]
+// CHECK:       omp.par.outlined.exit:
+// CHECK-NEXT:    br label [[OMP_PAR_EXIT_SPLIT:%.*]]
+// CHECK:       omp.par.exit.split:
+// CHECK-NEXT:    ret void
+//
+// CHECK-DEBUG-LABEL: @_Z14parallel_for_0v(
+// CHECK-DEBUG-NEXT:  entry:
+// CHECK-DEBUG-NEXT:    [[OMP_GLOBAL_THREAD_NUM:%.*]] = call i32 @__kmpc_global_thread_num(%struct.ident_t* @1), !dbg !{{[0-9]*}}
+// CHECK-DEBUG-NEXT:    br label [[OMP_PARALLEL:%.*]]
+// CHECK-DEBUG:       omp_parallel:
+// CHECK-DEBUG-NEXT:    call void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%struct.ident_t* @1, i32 0, void (i32*, i32*, ...)* bitcast (void (i32*, i32*)* @_Z14parallel_for_0v..omp_par to void (i32*, i32*, ...)*)), !dbg !{{[0-9]*}}
+// CHECK-DEBUG-NEXT:    br label [[OMP_PAR_OUTLINED_EXIT:%.*]]
+// CHECK-DEBUG:       omp.par.outlined.exit:
+// CHECK-DEBUG-NEXT:    br label [[OMP_PAR_EXIT_SPLIT:%.*]]
+// CHECK-DEBUG:       omp.par.exit.split:
+// CHECK-DEBUG-NEXT:    ret void, !dbg !{{[0-9]*}}
+//
+void parallel_for_0(void) {
+#pragma omp parallel
+  {
+#pragma omp for
+    for (int i = 0; i < 100; ++i) {
+    }
+  }
+}
+
+// CHECK-LABEL: @_Z14parallel_for_1Pfid(
+// CHECK-NEXT:  entry:
+// CHECK-NEXT:    [[R_ADDR:%.*]] = alloca float*, align 8
+// CHECK-NEXT:    [[A_ADDR:%.*]] = alloca i32, align 4
+// CHECK-NEXT:    [[B_ADDR:%.*]] = alloca double, align 8
+// CHECK-NEXT:    store float* [[R:%.*]], float** [[R_ADDR]], align 8
+// CHECK-NEXT:    store i32 [[A:%.*]], i32* [[A_ADDR]], align 4
+// CHECK-NEXT:    store double [[B:%.*]], double* [[B_ADDR]], align 8
+// CHECK-NEXT:    [[OMP_GLOBAL_THREAD_NUM:%.*]] = call i32 @__kmpc_global_thread_num(%struct.ident_t* @1)
+// CHECK-NEXT:    br label [[OMP_PARALLEL:%.*]]
+// CHECK:       omp_parallel:
+// CHECK-NEXT:    call void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%struct.ident_t* @1, i32 3, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, i32*, double*, float**)* @_Z14parallel_for_1Pfid..omp_par.1 to void (i32*, i32*, ...)*), i32* [[A_ADDR]], double* [[B_ADDR]], float** [[R_ADDR]])
+// CHECK-NEXT:    br label [[OMP_PAR_OUTLINED_EXIT19:%.*]]
+// CHECK:       omp.par.outlined.exit19:
+// CHECK-NEXT:    br label [[OMP_PAR_EXIT_SPLIT:%.*]]
+// CHECK:       omp.par.exit.split:
+// CHECK-NEXT:    ret void
+//
+// CHECK-DEBUG-LABEL: @_Z14parallel_for_1Pfid(
+// CHECK-DEBUG-NEXT:  entry:
+// CHECK-DEBUG-NEXT:    [[R_ADDR:%.*]] = alloca float*, align 8
+// CHECK-DEBUG-NEXT:    [[A_ADDR:%.*]] = alloca i32, align 4
+// CHECK-DEBUG-NEXT:    [[B_ADDR:%.*]] = alloca double, align 8
+// CHECK-DEBUG-NEXT:    store float* [[R:%.*]], float** [[R_ADDR]], align 8
+// CHECK-DEBUG-NEXT:    call void @llvm.dbg.declare(metadata float** [[R_ADDR]], metadata !{{[0-9]*}}, metadata !DIExpression()), !dbg !{{[0-9]*}}
+// CHECK-DEBUG-NEXT:    store i32 [[A:%.*]], i32* [[A_ADDR]], align 4
+// CHECK-DEBUG-NEXT:    call void @llvm.dbg.declare(metadata i32* [[A_ADDR]], metadata !{{[0-9]*}}, metadata !DIExpression()), !dbg !{{[0-9]*}}
+// CHECK-DEBUG-NEXT:    store double [[B:%.*]], double* [[B_ADDR]], align 8
+// CHECK-DEBUG-NEXT:    call void @llvm.dbg.declare(metadata double* [[B_ADDR]], metadata !{{[0-9]*}}, metadata !DIExpression()), !dbg !{{[0-9]*}}
+// CHECK-DEBUG-NEXT:    [[OMP_GLOBAL_THREAD_NUM:%.*]] = call i32 @__kmpc_global_thread_num(%struct.ident_t* @12), !dbg !{{[0-9]*}}
+// CHECK-DEBUG-NEXT:    br label [[OMP_PARALLEL:%.*]]
+// CHECK-DEBUG:       omp_parallel:
+// CHECK-DEBUG-NEXT:    call void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%struct.ident_t* @12, i32 3, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, i32*, double*, float**)* @_Z14parallel_for_1Pfid..omp_par.1 to void (i32*, i32*, ...)*), i32* [[A_ADDR]], double* [[B_ADDR]], float** [[R_ADDR]]), !dbg !{{[0-9]*}}
+// CHECK-DEBUG-NEXT:    br label [[OMP_PAR_OUTLINED_EXIT19:%.*]]
+// CHECK-DEBUG:       omp.par.outlined.exit19:
+// CHECK-DEBUG-NEXT:    br label [[OMP_PAR_EXIT_SPLIT:%.*]]
+// CHECK-DEBUG:       omp.par.exit.split:
+// CHECK-DEBUG-NEXT:    ret void, !dbg !{{[0-9]*}}
+//
+void parallel_for_1(float *r, int a, double b) {
+#pragma omp parallel
+  {
+#pragma omp parallel
+    {
+#pragma omp for
+      for (int i = 0; i < 100; ++i) {
+        *r = a + b;
+      }
+    }
+  }
+}
+
+// CHECK-LABEL: @_Z14parallel_for_2Pfid(
+// CHECK-NEXT:  entry:
+// CHECK-NEXT:    [[R_ADDR:%.*]] = alloca float*, align 8
+// CHECK-NEXT:    [[A_ADDR:%.*]] = alloca i32, align 4
+// CHECK-NEXT:    [[B_ADDR:%.*]] = alloca double, align 8
+// CHECK-NEXT:    [[DOTOMP_IV212:%.*]] = alloca i32, align 4
+// CHECK-NEXT:    [[TMP213:%.*]] = alloca i32, align 4
+// CHECK-NEXT:    [[DOTOMP_LB214:%.*]] = alloca i32, align 4
+// CHECK-NEXT:    [[DOTOMP_UB215:%.*]] = alloca i32, align 4
+// CHECK-NEXT:    [[DOTOMP_STRIDE216:%.*]] = alloca i32, align 4
+// CHECK-NEXT:    [[DOTOMP_IS_LAST217:%.*]] = alloca i32, align 4
+// CHECK-NEXT:    [[I218:%.*]] = alloca i32, align 4
+// CHECK-NEXT:    store float* [[R:%.*]], float** [[R_ADDR]], align 8
+// CHECK-NEXT:    store i32 [[A:%.*]], i32* [[A_ADDR]], align 4
+// CHECK-NEXT:    store double [[B:%.*]], double* [[B_ADDR]], align 8
+// CHECK-NEXT:    [[OMP_GLOBAL_THREAD_NUM:%.*]] = call i32 @__kmpc_global_thread_num(%struct.ident_t* @1)
+// CHECK-NEXT:    br label [[OMP_PARALLEL:%.*]]
+// CHECK:       omp_parallel:
+// CHECK-NEXT:    call void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%struct.ident_t* @1, i32 3, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, i32*, double*, float**)* @_Z14parallel_for_2Pfid..omp_par.4 to void (i32*, i32*, ...)*), i32* [[A_ADDR]], double* [[B_ADDR]], float** [[R_ADDR]])
+// CHECK-NEXT:    br label [[OMP_PAR_OUTLINED_EXIT211:%.*]]
+// CHECK:       omp.par.outlined.exit211:
+// CHECK-NEXT:    br label [[OMP_PAR_EXIT_SPLIT:%.*]]
+// CHECK:       omp.par.exit.split:
+// CHECK-NEXT:    store i32 0, i32* [[DOTOMP_LB214]], align 4
+// CHECK-NEXT:    store i32 99, i32* [[DOTOMP_UB215]], align 4
+// CHECK-NEXT:    store i32 1, i32* [[DOTOMP_STRIDE216]], align 4
+// CHECK-NEXT:    store i32 0, i32* [[DOTOMP_IS_LAST217]], align 4
+// CHECK-NEXT:    [[OMP_GLOBAL_THREAD_NUM219:%.*]] = call i32 @__kmpc_global_thread_num(%struct.ident_t* @41)
+// CHECK-NEXT:    call void @__kmpc_for_static_init_4(%struct.ident_t* @2, i32 [[OMP_GLOBAL_THREAD_NUM219]], i32 34, i32* [[DOTOMP_IS_LAST217]], i32* [[DOTOMP_LB214]], i32* [[DOTOMP_UB215]], i32* [[DOTOMP_STRIDE216]], i32 1, i32 1)
+// CHECK-NEXT:    [[TMP0:%.*]] = load i32, i32* [[DOTOMP_UB215]], align 4
+// CHECK-NEXT:    [[CMP220:%.*]] = icmp sgt i32 [[TMP0]], 99
+// CHECK-NEXT:    br i1 [[CMP220]], label [[COND_TRUE221:%.*]], label [[COND_FALSE222:%.*]]
+// CHECK:       cond.true221:
+// CHECK-NEXT:    br label [[COND_END223:%.*]]
+// CHECK:       cond.false222:
+// CHECK-NEXT:    [[TMP1:%.*]] = load i32, i32* [[DOTOMP_UB215]], align 4
+// CHECK-NEXT:    br label [[COND_END223]]
+// CHECK:       cond.end223:
+// CHECK-NEXT:    [[COND224:%.*]] = phi i32 [ 99, [[COND_TRUE221]] ], [ [[TMP1]], [[COND_FALSE222]] ]
+// CHECK-NEXT:    store i32 [[COND224]], i32* [[DOTOMP_UB215]], align 4
+// CHECK-NEXT:    [[TMP2:%.*]] = load i32, i32* [[DOTOMP_LB214]], align 4
+// CHECK-NEXT:    store i32 [[TMP2]], i32* [[DOTOMP_IV212]], align 4
+// CHECK-NEXT:    br label [[OMP_INNER_FOR_COND225:%.*]]
+// CHECK:       omp.inner.for.cond225:
+// CHECK-NEXT:    [[TMP3:%.*]] = load i32, i32* [[DOTOMP_IV212]], align 4
+// CHECK-NEXT:    [[TMP4:%.*]] = load i32, i32* [[DOTOMP_UB215]], align 4
+// CHECK-NEXT:    [[CMP226:%.*]] = icmp sle i32 [[TMP3]], [[TMP4]]
+// CHECK-NEXT:    br i1 [[CMP226]], label [[OMP_INNER_FOR_BODY227:%.*]], label [[OMP_INNER_FOR_END236:%.*]]
+// CHECK:       omp.inner.for.body227:
+// CHECK-NEXT:    [[TMP5:%.*]] = load i32, i32* [[DOTOMP_IV212]], align 4
+// CHECK-NEXT:    [[MUL228:%.*]] = mul nsw i32 [[TMP5]], 1
+// CHECK-NEXT:    [[ADD229:%.*]] = add nsw i32 0, [[MUL228]]
+// CHECK-NEXT:    store i32 [[ADD229]], i32* [[I218]], align 4
+// CHECK-NEXT:    [[TMP6:%.*]] = load i32, i32* [[A_ADDR]], align 4
+// CHECK-NEXT:    [[CONV230:%.*]] = sitofp i32 [[TMP6]] to double
+// CHECK-NEXT:    [[TMP7:%.*]] = load double, double* [[B_ADDR]], align 8
+// CHECK-NEXT:    [[ADD231:%.*]] = fadd double [[CONV230]], [[TMP7]]
+// CHECK-NEXT:    [[CONV232:%.*]] = fptrunc double [[ADD231]] to float
+// CHECK-NEXT:    [[TMP8:%.*]] = load float*, float** [[R_ADDR]], align 8
+// CHECK-NEXT:    store float [[CONV232]], float* [[TMP8]], align 4
+// CHECK-NEXT:    br label [[OMP_BODY_CONTINUE233:%.*]]
+// CHECK:       omp.body.continue233:
+// CHECK-NEXT:    br label [[OMP_INNER_FOR_INC234:%.*]]
+// CHECK:       omp.inner.for.inc234:
+// CHECK-NEXT:    [[TMP9:%.*]] = load i32, i32* [[DOTOMP_IV212]], align 4
+// CHECK-NEXT:    [[ADD235:%.*]] = add nsw i32 [[TMP9]], 1
+// CHECK-NEXT:    store i32 [[ADD235]], i32* [[DOTOMP_IV212]], align 4
+// CHECK-NEXT:    br label [[OMP_INNER_FOR_COND225]]
+// CHECK:       omp.inner.for.end236:
+// CHECK-NEXT:    br label [[OMP_LOOP_EXIT237:%.*]]
+// CHECK:       omp.loop.exit237:
+// CHECK-NEXT:    [[OMP_GLOBAL_THREAD_NUM238:%.*]] = call i32 @__kmpc_global_thread_num(%struct.ident_t* @43)
+// CHECK-NEXT:    call void @__kmpc_for_static_fini(%struct.ident_t* @2, i32 [[OMP_GLOBAL_THREAD_NUM238]])
+// CHECK-NEXT:    [[OMP_GLOBAL_THREAD_NUM239:%.*]] = call i32 @__kmpc_global_thread_num(%struct.ident_t* @1)
+// CHECK-NEXT:    call void @__kmpc_barrier(%struct.ident_t* @7, i32 [[OMP_GLOBAL_THREAD_NUM239]])
+// CHECK-NEXT:    ret void
+//
+// CHECK-DEBUG-LABEL: @_Z14parallel_for_2Pfid(
+// CHECK-DEBUG-NEXT:  entry:
+// CHECK-DEBUG-NEXT:    [[R_ADDR:%.*]] = alloca float*, align 8
+// CHECK-DEBUG-NEXT:    [[A_ADDR:%.*]] = alloca i32, align 4
+// CHECK-DEBUG-NEXT:    [[B_ADDR:%.*]] = alloca double, align 8
+// CHECK-DEBUG-NEXT:    [[DOTOMP_IV212:%.*]] = alloca i32, align 4
+// CHECK-DEBUG-NEXT:    [[TMP213:%.*]] = alloca i32, align 4
+// CHECK-DEBUG-NEXT:    [[DOTOMP_LB214:%.*]] = alloca i32, align 4
+// CHECK-DEBUG-NEXT:    [[DOTOMP_UB215:%.*]] = alloca i32, align 4
+// CHECK-DEBUG-NEXT:    [[DOTOMP_STRIDE216:%.*]] = alloca i32, align 4
+// CHECK-DEBUG-NEXT:    [[DOTOMP_IS_LAST217:%.*]] = alloca i32, align 4
+// CHECK-DEBUG-NEXT:    [[I218:%.*]] = alloca i32, align 4
+// CHECK-DEBUG-NEXT:    store float* [[R:%.*]], float** [[R_ADDR]], align 8
+// CHECK-DEBUG-NEXT:    call void @llvm.dbg.declare(metadata float** [[R_ADDR]], metadata !{{[0-9]*}}, metadata !DIExpression()), !dbg !{{[0-9]*}}
+// CHECK-DEBUG-NEXT:    store i32 [[A:%.*]], i32* [[A_ADDR]], align 4
+// CHECK-DEBUG-NEXT:    call void @llvm.dbg.declare(metadata i32* [[A_ADDR]], metadata !{{[0-9]*}}, metadata !DIExpression()), !dbg !{{[0-9]*}}
+// CHECK-DEBUG-NEXT:    store double [[B:%.*]], double* [[B_ADDR]], align 8
+// CHECK-DEBUG-NEXT:    call void @llvm.dbg.declare(metadata double* [[B_ADDR]], metadata !{{[0-9]*}}, metadata !DIExpression()), !dbg !{{[0-9]*}}
+// CHECK-DEBUG-NEXT:    [[OMP_GLOBAL_THREAD_NUM:%.*]] = call i32 @__kmpc_global_thread_num(%struct.ident_t* @25), !dbg !{{[0-9]*}}
+// CHECK-DEBUG-NEXT:    br label [[OMP_PARALLEL:%.*]]
+// CHECK-DEBUG:       omp_parallel:
+// CHECK-DEBUG-NEXT:    call void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%struct.ident_t* @25, i32 3, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, i32*, double*, float**)* @_Z14parallel_for_2Pfid..omp_par.4 to void (i32*, i32*, ...)*), i32* [[A_ADDR]], double* [[B_ADDR]], float** [[R_ADDR]]), !dbg !{{[0-9]*}}
+// CHECK-DEBUG-NEXT:    br label [[OMP_PAR_OUTLINED_EXIT211:%.*]]
+// CHECK-DEBUG:       omp.par.outlined.exit211:
+// CHECK-DEBUG-NEXT:    br label [[OMP_PAR_EXIT_SPLIT:%.*]]
+// CHECK-DEBUG:       omp.par.exit.split:
+// CHECK-DEBUG-NEXT:    call void @llvm.dbg.declare(metadata i32* [[DOTOMP_IV212]], metadata !{{[0-9]*}}, metadata !DIExpression()), !dbg !{{[0-9]*}}
+// CHECK-DEBUG-NEXT:    call void @llvm.dbg.declare(metadata i32* [[DOTOMP_LB214]], metadata !{{[0-9]*}}, metadata !DIExpression()), !dbg !{{[0-9]*}}
+// CHECK-DEBUG-NEXT:    store i32 0, i32* [[DOTOMP_LB214]], align 4, !dbg !{{[0-9]*}}
+// CHECK-DEBUG-NEXT:    call void @llvm.dbg.declare(metadata i32* [[DOTOMP_UB215]], metadata !{{[0-9]*}}, metadata !DIExpression()), !dbg !{{[0-9]*}}
+// CHECK-DEBUG-NEXT:    store i32 99, i32* [[DOTOMP_UB215]], align 4, !dbg !{{[0-9]*}}
+// CHECK-DEBUG-NEXT:    call void @llvm.dbg.declare(metadata i32* [[DOTOMP_STRIDE216]], metadata !{{[0-9]*}}, metadata !DIExpression()), !dbg !{{[0-9]*}}
+// CHECK-DEBUG-NEXT:    store i32 1, i32* [[DOTOMP_STRIDE216]], align 4, !dbg !{{[0-9]*}}
+// CHECK-DEBUG-NEXT:    call void @llvm.dbg.declare(metadata i32* [[DOTOMP_IS_LAST217]], metadata !{{[0-9]*}}, metadata !DIExpression()), !dbg !{{[0-9]*}}
+// CHECK-DEBUG-NEXT:    store i32 0, i32* [[DOTOMP_IS_LAST217]], align 4, !dbg !{{[0-9]*}}
+// CHECK-DEBUG-NEXT:    call void @llvm.dbg.declare(metadata i32* [[I218]], metadata !{{[0-9]*}}, metadata !DIExpression()), !dbg !{{[0-9]*}}
+// CHECK-DEBUG-NEXT:    [[OMP_GLOBAL_THREAD_NUM219:%.*]] = call i32 @__kmpc_global_thread_num(%struct.ident_t* @97)
+// CHECK-DEBUG-NEXT:    call void @__kmpc_for_static_init_4(%struct.ident_t* @96, i32 [[OMP_GLOBAL_THREAD_NUM219]], i32 34, i32* [[DOTOMP_IS_LAST217]], i32* [[DOTOMP_LB214]], i32* [[DOTOMP_UB215]], i32* [[DOTOMP_STRIDE216]], i32 1, i32 1), !dbg !{{[0-9]*}}
+// CHECK-DEBUG-NEXT:    [[TMP0:%.*]] = load i32, i32* [[DOTOMP_UB215]], align 4, !dbg !{{[0-9]*}}
+// CHECK-DEBUG-NEXT:    [[CMP220:%.*]] = icmp sgt i32 [[TMP0]], 99, !dbg !{{[0-9]*}}
+// CHECK-DEBUG-NEXT:    br i1 [[CMP220]], label [[COND_TRUE221:%.*]], label [[COND_FALSE222:%.*]], !dbg !{{[0-9]*}}
+// CHECK-DEBUG:       cond.true221:
+// CHECK-DEBUG-NEXT:    br label [[COND_END223:%.*]], !dbg !{{[0-9]*}}
+// CHECK-DEBUG:       cond.false222:
+// CHECK-DEBUG-NEXT:    [[TMP1:%.*]] = load i32, i32* [[DOTOMP_UB215]], align 4, !dbg !{{[0-9]*}}
+// CHECK-DEBUG-NEXT:    br label [[COND_END223]], !dbg !{{[0-9]*}}
+// CHECK-DEBUG:       cond.end223:
+// CHECK-DEBUG-NEXT:    [[COND224:%.*]] = phi i32 [ 99, [[COND_TRUE221]] ], [ [[TMP1]], [[COND_FALSE222]] ], !dbg !{{[0-9]*}}
+// CHECK-DEBUG-NEXT:    store i32 [[COND224]], i32* [[DOTOMP_UB215]], align 4, !dbg !{{[0-9]*}}
+// CHECK-DEBUG-NEXT:    [[TMP2:%.*]] = load i32, i32* [[DOTOMP_LB214]], align 4, !dbg !{{[0-9]*}}
+// CHECK-DEBUG-NEXT:    store i32 [[TMP2]], i32* [[DOTOMP_IV212]], align 4, !dbg !{{[0-9]*}}
+// CHECK-DEBUG-NEXT:    br label [[OMP_INNER_FOR_COND225:%.*]], !dbg !{{[0-9]*}}
+// CHECK-DEBUG:       omp.inner.for.cond225:
+// CHECK-DEBUG-NEXT:    [[TMP3:%.*]] = load i32, i32* [[DOTOMP_IV212]], align 4, !dbg !{{[0-9]*}}
+// CHECK-DEBUG-NEXT:    [[TMP4:%.*]] = load i32, i32* [[DOTOMP_UB215]], align 4, !dbg !{{[0-9]*}}
+// CHECK-DEBUG-NEXT:    [[CMP226:%.*]] = icmp sle i32 [[TMP3]], [[TMP4]], !dbg !{{[0-9]*}}
+// CHECK-DEBUG-NEXT:    br i1 [[CMP226]], label [[OMP_INNER_FOR_BODY227:%.*]], label [[OMP_INNER_FOR_END236:%.*]], !dbg !{{[0-9]*}}
+// CHECK-DEBUG:       omp.inner.for.body227:
+// CHECK-DEBUG-NEXT:    [[TMP5:%.*]] = load i32, i32* [[DOTOMP_IV212]], align 4, !dbg !{{[0-9]*}}
+// CHECK-DEBUG-NEXT:    [[MUL228:%.*]] = mul nsw i32 [[TMP5]], 1, !dbg !{{[0-9]*}}
+// CHECK-DEBUG-NEXT:    [[ADD229:%.*]] = add nsw i32 0, [[MUL228]], !dbg !{{[0-9]*}}
+// CHECK-DEBUG-NEXT:    store i32 [[ADD229]], i32* [[I218]], align 4, !dbg !{{[0-9]*}}
+// CHECK-DEBUG-NEXT:    [[TMP6:%.*]] = load i32, i32* [[A_ADDR]], align 4, !dbg !{{[0-9]*}}
+// CHECK-DEBUG-NEXT:    [[CONV230:%.*]] = sitofp i32 [[TMP6]] to double, !dbg !{{[0-9]*}}
+// CHECK-DEBUG-NEXT:    [[TMP7:%.*]] = load double, double* [[B_ADDR]], align 8, !dbg !{{[0-9]*}}
+// CHECK-DEBUG-NEXT:    [[ADD231:%.*]] = fadd double [[CONV230]], [[TMP7]], !dbg !{{[0-9]*}}
+// CHECK-DEBUG-NEXT:    [[CONV232:%.*]] = fptrunc double [[ADD231]] to float, !dbg !{{[0-9]*}}
+// CHECK-DEBUG-NEXT:    [[TMP8:%.*]] = load float*, float** [[R_ADDR]], align 8, !dbg !{{[0-9]*}}
+// CHECK-DEBUG-NEXT:    store float [[CONV232]], float* [[TMP8]], align 4, !dbg !{{[0-9]*}}
+// CHECK-DEBUG-NEXT:    br label [[OMP_BODY_CONTINUE233:%.*]], !dbg !{{[0-9]*}}
+// CHECK-DEBUG:       omp.body.continue233:
+// CHECK-DEBUG-NEXT:    br label [[OMP_INNER_FOR_INC234:%.*]], !dbg !{{[0-9]*}}
+// CHECK-DEBUG:       omp.inner.for.inc234:
+// CHECK-DEBUG-NEXT:    [[TMP9:%.*]] = load i32, i32* [[DOTOMP_IV212]], align 4, !dbg !{{[0-9]*}}
+// CHECK-DEBUG-NEXT:    [[ADD235:%.*]] = add nsw i32 [[TMP9]], 1, !dbg !{{[0-9]*}}
+// CHECK-DEBUG-NEXT:    store i32 [[ADD235]], i32* [[DOTOMP_IV212]], align 4, !dbg !{{[0-9]*}}
+// CHECK-DEBUG-NEXT:    br label [[OMP_INNER_FOR_COND225]], !dbg !{{[0-9]*}}, !llvm.loop !{{[0-9]*}}
+// CHECK-DEBUG:       omp.inner.for.end236:
+// CHECK-DEBUG-NEXT:    br label [[OMP_LOOP_EXIT237:%.*]], !dbg !{{[0-9]*}}
+// CHECK-DEBUG:       omp.loop.exit237:
+// CHECK-DEBUG-NEXT:    [[OMP_GLOBAL_THREAD_NUM238:%.*]] = call i32 @__kmpc_global_thread_num(%struct.ident_t* @100)
+// CHECK-DEBUG-NEXT:    call void @__kmpc_for_static_fini(%struct.ident_t* @99, i32 [[OMP_GLOBAL_THREAD_NUM238]]), !dbg !{{[0-9]*}}
+// CHECK-DEBUG-NEXT:    [[OMP_GLOBAL_THREAD_NUM239:%.*]] = call i32 @__kmpc_global_thread_num(%struct.ident_t* @103), !dbg !{{[0-9]*}}
+// CHECK-DEBUG-NEXT:    call void @__kmpc_barrier(%struct.ident_t* @102, i32 [[OMP_GLOBAL_THREAD_NUM239]]), !dbg !{{[0-9]*}}
+// CHECK-DEBUG-NEXT:    ret void, !dbg !{{[0-9]*}}
+//
+void parallel_for_2(float *r, int a, double b) {
+#pragma omp parallel
+  {
+#pragma omp for
+    for (int i = 0; i < 100; ++i)
+      *r = a + b;
+#pragma omp parallel
+    {
+#pragma omp for
+      for (int i = 0; i < 100; ++i)
+        *r = a + b;
+#pragma omp parallel
+      {
+#pragma omp for
+        for (int i = 0; i < 100; ++i)
+          *r = a + b;
+      }
+#pragma omp for
+      for (int i = 0; i < 100; ++i)
+        *r = a + b;
+#pragma omp parallel
+      {
+#pragma omp for
+        for (int i = 0; i < 100; ++i)
+          *r = a + b;
+      }
+#pragma omp for
+      for (int i = 0; i < 100; ++i)
+        *r = a + b;
+    }
+#pragma omp for
+    for (int i = 0; i < 100; ++i)
+      *r = a + b;
+  }
+#pragma omp for
+  for (int i = 0; i < 100; ++i)
+    *r = a + b;
+}
+
+#endif

clang/test/OpenMP/task_codegen.cpp

 int a;
 // CHECK-LABEL: @main
 int main() {
 // CHECK: [[B:%.+]] = alloca i8
 // CHECK: [[S:%.+]] = alloca [2 x [[STRUCT_S]]]
   char b;
   S s[2];
   int arr[10][a];
-// CHECK: [[GTID:%.+]] = call i32 @__kmpc_global_thread_num([[IDENT_T]]* @{{.+}})
 // CHECK: [[B_REF:%.+]] = getelementptr inbounds [[STRUCT_SHAREDS]], [[STRUCT_SHAREDS]]* [[CAPTURES:%.+]], i32 0, i32 0
 // CHECK: store i8* [[B]], i8** [[B_REF]]
 // CHECK: [[S_REF:%.+]] = getelementptr inbounds [[STRUCT_SHAREDS]], [[STRUCT_SHAREDS]]* [[CAPTURES]], i32 0, i32 1
 // CHECK: store [2 x [[STRUCT_S]]]* [[S]], [2 x [[STRUCT_S]]]** [[S_REF]]
-// CHECK: [[ORIG_TASK_PTR:%.+]] = call i8* @__kmpc_omp_task_alloc([[IDENT_T]]* @{{.+}}, i32 [[GTID]], i32 33, i64 40, i64 16, i32 (i32, i8*)* bitcast (i32 (i32, [[KMP_TASK_T]]{{.*}}*)* [[TASK_ENTRY1:@.+]] to i32 (i32, i8*)*))
+// CHECK: [[ORIG_TASK_PTR:%.+]] = call i8* @__kmpc_omp_task_alloc([[IDENT_T]]* @{{.+}}, i32 {{%.*}}, i32 33, i64 40, i64 16, i32 (i32, i8*)* bitcast (i32 (i32, [[KMP_TASK_T]]{{.*}}*)* [[TASK_ENTRY1:@.+]] to i32 (i32, i8*)*))
 // CHECK: [[SHAREDS_REF_PTR:%.+]] = getelementptr inbounds [[KMP_TASK_T]], [[KMP_TASK_T]]* [[TASK_PTR:%.+]], i32 0, i32 0
 // CHECK: [[SHAREDS_REF:%.+]] = load i8*, i8** [[SHAREDS_REF_PTR]]
 // CHECK: [[BITCAST:%.+]] = bitcast [[STRUCT_SHAREDS]]* [[CAPTURES]] to i8*
 // CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[SHAREDS_REF]], i8* align 8 [[BITCAST]], i64 16, i1 false)
 // CHECK: [[PRIORITY_REF_PTR:%.+]] = getelementptr inbounds [[KMP_TASK_T]], [[KMP_TASK_T]]* [[TASK_PTR]], i32 0, i32 4
 // CHECK: [[PRIORITY:%.+]] = bitcast %union{{.+}}* [[PRIORITY_REF_PTR]] to i32*
 // CHECK: store i32 {{.+}}, i32* [[PRIORITY]]
-// CHECK: call i32 @__kmpc_omp_task([[IDENT_T]]* @{{.+}}, i32 [[GTID]], i8* [[ORIG_TASK_PTR]])
+// CHECK: call i32 @__kmpc_omp_task([[IDENT_T]]* @{{.+}}, i32 {{%.*}}, i8* [[ORIG_TASK_PTR]])
 #pragma omp task shared(a, b, s) priority(b)
   {
     a = 15;
     b = a;
     s[0].a = 10;
   }
 // CHECK: [[S_REF:%.+]] = getelementptr inbounds [[STRUCT_SHAREDS1]], [[STRUCT_SHAREDS1]]* [[CAPTURES:%.+]], i32 0, i32 0
 // CHECK: store [2 x [[STRUCT_S]]]* [[S]], [2 x [[STRUCT_S]]]** [[S_REF]]
-// CHECK: [[ORIG_TASK_PTR:%.+]] = call i8* @__kmpc_omp_task_alloc([[IDENT_T]]* @{{[^,]+}}, i32 [[GTID]], i32 1, i64 40, i64 8,
+// CHECK: [[ORIG_TASK_PTR:%.+]] = call i8* @__kmpc_omp_task_alloc([[IDENT_T]]* @{{[^,]+}}, i32 {{%.*}}, i32 1, i64 40, i64 8,
 // CHECK: [[SHAREDS_REF_PTR:%.+]] = getelementptr inbounds [[KMP_TASK_T]], [[KMP_TASK_T]]* [[TASK_PTR:%.+]], i32 0, i32 0
 // CHECK: [[SHAREDS_REF:%.+]] = load i8*, i8** [[SHAREDS_REF_PTR]]
 // CHECK: [[BITCAST:%.+]] = bitcast [[STRUCT_SHAREDS1]]* [[CAPTURES]] to i8*
 // CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 8 [[SHAREDS_REF]], i8* align 8 [[BITCAST]], i64 8, i1 false)
 // CHECK: [[DEP_BASE:%.*]] = getelementptr inbounds [4 x [[KMP_DEPEND_INFO]]], [4 x [[KMP_DEPEND_INFO]]]* [[DEPENDENCIES:%.*]], i64 0, i64 0
 // CHECK: [[DEP:%.+]] = getelementptr [[KMP_DEPEND_INFO]], [[KMP_DEPEND_INFO]]* [[DEP_BASE]], i64 0
 // CHECK: [[T0:%.*]] = getelementptr inbounds [[KMP_DEPEND_INFO]], [[KMP_DEPEND_INFO]]* [[DEP]], i32 0, i32 0
 // CHECK: store i64 ptrtoint (i32* @{{.+}} to i64), i64* [[T0]]

 // CHECK: [[T0:%.*]] = getelementptr inbounds [[KMP_DEPEND_INFO]], [[KMP_DEPEND_INFO]]* [[DEP]], i32 0, i32 0
 // CHECK: [[T1:%.*]] = ptrtoint i32* [[START]] to i64
 // CHECK: store i64 [[T1]], i64* [[T0]]
 // CHECK: [[T0:%.*]] = getelementptr inbounds [[KMP_DEPEND_INFO]], [[KMP_DEPEND_INFO]]* %{{[^,]+}}, i32 0, i32 1
 // CHECK: store i64 [[SIZEOF]], i64* [[T0]]
 // CHECK: [[T0:%.*]] = getelementptr inbounds [[KMP_DEPEND_INFO]], [[KMP_DEPEND_INFO]]* %{{[^,]+}}, i32 0, i32 2
 // CHECK: store i8 1, i8* [[T0]]
 // CHECK: bitcast [[KMP_DEPEND_INFO]]* [[DEP_BASE]] to i8*
-// CHECK: call i32 @__kmpc_omp_task_with_deps([[IDENT_T]]* @{{.+}}, i32 [[GTID]], i8* [[ORIG_TASK_PTR]], i32 4, i8* %{{[^,]+}}, i32 0, i8* null)
+// CHECK: call i32 @__kmpc_omp_task_with_deps([[IDENT_T]]* @{{.+}}, i32 {{%.*}}, i8* [[ORIG_TASK_PTR]], i32 4, i8* %{{[^,]+}}, i32 0, i8* null)
 #pragma omp task shared(a, s) depend(in : a, b, s, arr[:])
   {
     a = 15;
     s[1].a = 10;
   }
-// CHECK: [[ORIG_TASK_PTR:%.+]] = call i8* @__kmpc_omp_task_alloc([[IDENT_T]]* @{{.+}}, i32 [[GTID]], i32 0, i64 40, i64 1, i32 (i32, i8*)* bitcast (i32 (i32, [[KMP_TASK_T]]{{.*}}*)* [[TASK_ENTRY2:@.+]] to i32 (i32, i8*)*))
-// CHECK: call i32 @__kmpc_omp_task([[IDENT_T]]* @{{.+}}, i32 [[GTID]], i8* [[ORIG_TASK_PTR]])
+// CHECK: [[ORIG_TASK_PTR:%.+]] = call i8* @__kmpc_omp_task_alloc([[IDENT_T]]* @{{.+}}, i32 {{%.*}}, i32 0, i64 40, i64 1, i32 (i32, i8*)* bitcast (i32 (i32, [[KMP_TASK_T]]{{.*}}*)* [[TASK_ENTRY2:@.+]] to i32 (i32, i8*)*))
+// CHECK: call i32 @__kmpc_omp_task([[IDENT_T]]* @{{.+}}, i32 {{%.*}}, i8* [[ORIG_TASK_PTR]])
 #pragma omp task untied
   {
 #pragma omp critical
     a = 1;
   }
-// CHECK: [[ORIG_TASK_PTR:%.+]] = call i8* @__kmpc_omp_task_alloc([[IDENT_T]]* @{{.+}}, i32 [[GTID]], i32 0, i64 40, i64 1,
+// CHECK: [[ORIG_TASK_PTR:%.+]] = call i8* @__kmpc_omp_task_alloc([[IDENT_T]]* @{{.+}}, i32 {{%.*}}, i32 0, i64 40, i64 1,
 // CHECK: getelementptr inbounds [2 x [[STRUCT_S]]], [2 x [[STRUCT_S]]]* [[S]], i64 0, i64 0
 // CHECK: getelementptr inbounds [[KMP_DEPEND_INFO]], [[KMP_DEPEND_INFO]]* %{{[^,]+}}, i32 0, i32 0
 // CHECK: ptrtoint [[STRUCT_S]]* %{{.+}} to i64
 // CHECK: store i64 %{{[^,]+}}, i64*
 // CHECK: getelementptr inbounds [[KMP_DEPEND_INFO]], [[KMP_DEPEND_INFO]]* %{{[^,]+}}, i32 0, i32 1
 // CHECK: store i64 4, i64*
 // CHECK: getelementptr inbounds [[KMP_DEPEND_INFO]], [[KMP_DEPEND_INFO]]* %{{[^,]+}}, i32 0, i32 2
 // CHECK: store i8 3, i8*

 // CHECK: getelementptr inbounds [[KMP_DEPEND_INFO]], [[KMP_DEPEND_INFO]]* %{{[^,]+}}, i32 0, i32 0
 // CHECK: ptrtoint i32* [[START1]] to i64
 // CHECK: store i64 %{{[^,]+}}, i64*
 // CHECK: getelementptr inbounds [[KMP_DEPEND_INFO]], [[KMP_DEPEND_INFO]]* %{{[^,]+}}, i32 0, i32 1
 // CHECK: store i64 [[SIZEOF]], i64*
 // CHECK: getelementptr inbounds [[KMP_DEPEND_INFO]], [[KMP_DEPEND_INFO]]* %{{[^,]+}}, i32 0, i32 2
 // CHECK: store i8 3, i8*
 // CHECK: bitcast [[KMP_DEPEND_INFO]]* %{{.+}} to i8*
-// CHECK: call i32 @__kmpc_omp_task_with_deps([[IDENT_T]]* @{{.+}}, i32 [[GTID]], i8* [[ORIG_TASK_PTR]], i32 2, i8* %{{[^,]+}}, i32 0, i8* null)
+// CHECK: call i32 @__kmpc_omp_task_with_deps([[IDENT_T]]* @{{.+}}, i32 {{%.*}}, i8* [[ORIG_TASK_PTR]], i32 2, i8* %{{[^,]+}}, i32 0, i8* null)
 #pragma omp task untied depend(out : s[0], arr[4:][b])
   {
     a = 1;
   }
-// CHECK: [[ORIG_TASK_PTR:%.+]] = call i8* @__kmpc_omp_task_alloc([[IDENT_T]]* @{{.+}}, i32 [[GTID]], i32 0, i64 40, i64 1,
+// CHECK: [[ORIG_TASK_PTR:%.+]] = call i8* @__kmpc_omp_task_alloc([[IDENT_T]]* @{{.+}}, i32 {{%.*}}, i32 0, i64 40, i64 1,
 // CHECK: getelementptr inbounds [2 x [[STRUCT_S]]], [2 x [[STRUCT_S]]]* [[S]], i64 0, i64 0
 // CHECK: getelementptr inbounds [[KMP_DEPEND_INFO]], [[KMP_DEPEND_INFO]]* %{{[^,]+}}, i32 0, i32 0
 // CHECK: ptrtoint [[STRUCT_S]]* %{{.+}} to i64
 // CHECK: store i64 %{{[^,]+}}, i64*
 // CHECK: getelementptr inbounds [[KMP_DEPEND_INFO]], [[KMP_DEPEND_INFO]]* %{{[^,]+}}, i32 0, i32 1
 // CHECK: store i64 4, i64*
 // CHECK: getelementptr inbounds [[KMP_DEPEND_INFO]], [[KMP_DEPEND_INFO]]* %{{[^,]+}}, i32 0, i32 2
 // CHECK: store i8 4, i8*

 // CHECK: getelementptr inbounds [[KMP_DEPEND_INFO]], [[KMP_DEPEND_INFO]]* %{{[^,]+}}, i32 0, i32 0
 // CHECK: ptrtoint i32* [[START1]] to i64
 // CHECK: store i64 %{{[^,]+}}, i64*
 // CHECK: getelementptr inbounds [[KMP_DEPEND_INFO]], [[KMP_DEPEND_INFO]]* %{{[^,]+}}, i32 0, i32 1
 // CHECK: store i64 [[SIZEOF]], i64*
 // CHECK: getelementptr inbounds [[KMP_DEPEND_INFO]], [[KMP_DEPEND_INFO]]* %{{[^,]+}}, i32 0, i32 2
 // CHECK: store i8 4, i8*
 // CHECK: bitcast [[KMP_DEPEND_INFO]]* %{{.+}} to i8*
-// CHECK: call i32 @__kmpc_omp_task_with_deps([[IDENT_T]]* @{{.+}}, i32 [[GTID]], i8* [[ORIG_TASK_PTR]], i32 2, i8* %{{[^,]+}}, i32 0, i8* null)
+// CHECK: call i32 @__kmpc_omp_task_with_deps([[IDENT_T]]* @{{.+}}, i32 {{%.*}}, i8* [[ORIG_TASK_PTR]], i32 2, i8* %{{[^,]+}}, i32 0, i8* null)
 #pragma omp task untied depend(mutexinoutset: s[0], arr[4:][b])
   {
     a = 1;
   }
-// CHECK: [[ORIG_TASK_PTR:%.+]] = call i8* @__kmpc_omp_task_alloc([[IDENT_T]]* @{{.+}}, i32 [[GTID]], i32 3, i64 40, i64 1,
+// CHECK: [[ORIG_TASK_PTR:%.+]] = call i8* @__kmpc_omp_task_alloc([[IDENT_T]]* @{{.+}}, i32 {{%.*}}, i32 3, i64 40, i64 1,
 // CHECK: getelementptr inbounds [[KMP_DEPEND_INFO]], [[KMP_DEPEND_INFO]]* %{{[^,]+}}, i32 0, i32 0
 // CHECK: store i64 ptrtoint (i32* @{{.+}} to i64), i64*
 // CHECK: getelementptr inbounds [[KMP_DEPEND_INFO]], [[KMP_DEPEND_INFO]]* %{{[^,]+}}, i32 0, i32 1
 // CHECK: store i64 4, i64*
 // CHECK: getelementptr inbounds [[KMP_DEPEND_INFO]], [[KMP_DEPEND_INFO]]* %{{[^,]+}}, i32 0, i32 2
 // CHECK: store i8 3, i8*
 // CHECK: getelementptr inbounds [2 x [[STRUCT_S]]], [2 x [[STRUCT_S]]]* [[S]], i64 0, i64 1
 // CHECK: getelementptr [[KMP_DEPEND_INFO]], [[KMP_DEPEND_INFO]]* %{{[^,]+}}, i64 1

 // CHECK: getelementptr inbounds [[KMP_DEPEND_INFO]], [[KMP_DEPEND_INFO]]* %{{[^,]+}}, i32 0, i32 0
 // CHECK: ptrtoint i32* [[START1]] to i64
 // CHECK: store i64 %{{[^,]+}}, i64*
 // CHECK: getelementptr inbounds [[KMP_DEPEND_INFO]], [[KMP_DEPEND_INFO]]* %{{[^,]+}}, i32 0, i32 1
 // CHECK: store i64 [[SIZEOF]], i64*
 // CHECK: getelementptr inbounds [[KMP_DEPEND_INFO]], [[KMP_DEPEND_INFO]]* %{{[^,]+}}, i32 0, i32 2
 // CHECK: store i8 3, i8*
 // CHECK: bitcast [[KMP_DEPEND_INFO]]* %{{.+}} to i8*
-// CHECK: call i32 @__kmpc_omp_task_with_deps([[IDENT_T]]* @{{.+}}, i32 [[GTID]], i8* [[ORIG_TASK_PTR]], i32 3, i8* %{{[^,]+}}, i32 0, i8* null)
+// CHECK: call i32 @__kmpc_omp_task_with_deps([[IDENT_T]]* @{{.+}}, i32 {{%.*}}, i8* [[ORIG_TASK_PTR]], i32 3, i8* %{{[^,]+}}, i32 0, i8* null)
 #pragma omp task final(true) depend(inout: a, s[1], arr[:a][3:])
   {
     a = 2;
   }
-// CHECK: [[ORIG_TASK_PTR:%.+]] = call i8* @__kmpc_omp_task_alloc([[IDENT_T]]* @{{.+}}, i32 [[GTID]], i32 3, i64 40, i64 1, i32 (i32, i8*)* bitcast (i32 (i32, [[KMP_TASK_T]]{{.*}}*)* [[TASK_ENTRY3:@.+]] to i32 (i32, i8*)*))
-// CHECK: call i32 @__kmpc_omp_task([[IDENT_T]]* @{{.+}}, i32 [[GTID]], i8* [[ORIG_TASK_PTR]])
+// CHECK: [[ORIG_TASK_PTR:%.+]] = call i8* @__kmpc_omp_task_alloc([[IDENT_T]]* @{{.+}}, i32 {{%.*}}, i32 3, i64 40, i64 1, i32 (i32, i8*)* bitcast (i32 (i32, [[KMP_TASK_T]]{{.*}}*)* [[TASK_ENTRY3:@.+]] to i32 (i32, i8*)*))
+// CHECK: call i32 @__kmpc_omp_task([[IDENT_T]]* @{{.+}}, i32 {{%.*}}, i8* [[ORIG_TASK_PTR]])
 #pragma omp task final(true)
   {
     a = 2;
   }
-// CHECK: [[ORIG_TASK_PTR:%.+]] = call i8* @__kmpc_omp_task_alloc([[IDENT_T]]* @{{.+}}, i32 [[GTID]], i32 1, i64 40, i64 1, i32 (i32, i8*)* bitcast (i32 (i32, [[KMP_TASK_T]]{{.*}}*)* [[TASK_ENTRY4:@.+]] to i32 (i32, i8*)*))
-// CHECK: call i32 @__kmpc_omp_task([[IDENT_T]]* @{{.+}}, i32 [[GTID]], i8* [[ORIG_TASK_PTR]])
+  // CHECK: [[ORIG_TASK_PTR:%.+]] = call i8* @__kmpc_omp_task_alloc([[IDENT_T]]* @{{.+}}, i32 {{%.*}}, i32 1, i64 40, i64 1, i32 (i32, i8*)* bitcast (i32 (i32, [[KMP_TASK_T]]{{.*}}*)* [[TASK_ENTRY4:@.+]] to i32 (i32, i8*)*))
+  // CHECK: call i32 @__kmpc_omp_task([[IDENT_T]]* @{{.+}}, i32 {{%.*}}, i8* [[ORIG_TASK_PTR]])
   const bool flag = false;
 #pragma omp task final(flag)
   {
     a = 3;
   }
   // CHECK: [[B_VAL:%.+]] = load i8, i8* [[B]]
   // CHECK: [[CMP:%.+]] = icmp ne i8 [[B_VAL]], 0
   // CHECK: [[FINAL:%.+]] = select i1 [[CMP]], i32 2, i32 0
   // CHECK: [[FLAGS:%.+]] = or i32 [[FINAL]], 1
-// CHECK: [[ORIG_TASK_PTR:%.+]] = call i8* @__kmpc_omp_task_alloc([[IDENT_T]]* @{{.+}}, i32 [[GTID]], i32 [[FLAGS]], i64 40, i64 8, i32 (i32, i8*)* bitcast (i32 (i32, [[KMP_TASK_T]]{{.*}}*)* [[TASK_ENTRY5:@.+]] to i32 (i32, i8*)*))
-// CHECK: call i32 @__kmpc_omp_task([[IDENT_T]]* @{{.+}}, i32 [[GTID]], i8* [[ORIG_TASK_PTR]])
+  // CHECK: [[ORIG_TASK_PTR:%.+]] = call i8* @__kmpc_omp_task_alloc([[IDENT_T]]* @{{.+}}, i32 {{%.*}}, i32 [[FLAGS]], i64 40, i64 8, i32 (i32, i8*)* bitcast (i32 (i32, [[KMP_TASK_T]]{{.*}}*)* [[TASK_ENTRY5:@.+]] to i32 (i32, i8*)*))
+  // CHECK: call i32 @__kmpc_omp_task([[IDENT_T]]* @{{.+}}, i32 {{%.*}}, i8* [[ORIG_TASK_PTR]])
   int c __attribute__((aligned(128)));
 #pragma omp task final(b) shared(c)
   {
     a = 4;
     c = 5;
   }
-// CHECK: [[ORIG_TASK_PTR:%.+]] = call i8* @__kmpc_omp_task_alloc([[IDENT_T]]* @{{.+}}, i32 [[GTID]], i32 0, i64 40, i64 1, i32 (i32, i8*)* bitcast (i32 (i32, [[KMP_TASK_T]]{{.*}}*)* [[TASK_ENTRY6:@.+]] to i32 (i32, i8*)*))
-// CHECK: call i32 @__kmpc_omp_task([[IDENT_T]]* @{{.+}}, i32 [[GTID]], i8* [[ORIG_TASK_PTR]])
+// CHECK: [[ORIG_TASK_PTR:%.+]] = call i8* @__kmpc_omp_task_alloc([[IDENT_T]]* @{{.+}}, i32 {{%.*}}, i32 0, i64 40, i64 1, i32 (i32, i8*)* bitcast (i32 (i32, [[KMP_TASK_T]]{{.*}}*)* [[TASK_ENTRY6:@.+]] to i32 (i32, i8*)*))
+// CHECK: call i32 @__kmpc_omp_task([[IDENT_T]]* @{{.+}}, i32 {{%.*}}, i8* [[ORIG_TASK_PTR]])
 #pragma omp task untied
   {
     S s1;
 #pragma omp task
     a = 4;
 #pragma omp taskyield
     s1 = S();
 #pragma omp taskwait