[OpenMP] Migrate dispatch related utility functions from Clang codegen to OMPIRBuilder

[TIFitis]

Summary

Migrate createForStaticInitFunction, createDispatchInitFunction, createDispatchNextFunction and createDispatchFiniFunction from Clang CodeGen to OMPIRBuilder.

[TIFitis]

llvm/include/llvm/Frontend/OpenMP/OMPKinds.def

532 ↗

(On Diff #550365)

Clang test llvm/clang/test/OpenMP/ordered_codegen.cpp complains about the presence of NoUnwind attribute for the __kmpc_dispatch_* functions.

This change doesn't seem to break any other test, although I am not sure if we should be creating a new Attr group for the __kmpc_dispatch functions and remove the NoUnwind Attr only there.

[jdoerfert]

LG, but do not remove the Attribute from our set, add it to the tests. The functions are nounwind.

llvm/include/llvm/Frontend/OpenMP/OMPKinds.def

532 ↗

(On Diff #550365)

Add nounwind to the tests?

[TIFitis]

Updated test with correct function attributes.

clang/lib/CodeGen/CGOpenMPRuntime.h

protected:
   void scanForTargetRegionsFunctions(const Stmt *S, StringRef ParentName);
 
   /// Build type kmp_routine_entry_t (if not built yet).
   void emitKmpRoutineEntryT(QualType KmpInt32Ty);
 
   /// Returns pointer to kmpc_micro type.
   llvm::Type *getKmpc_MicroPointerTy();
 
-  /// Returns __kmpc_for_static_init_* runtime function for the specified
-  /// size \a IVSize and sign \a IVSigned. Will create a distribute call
-  /// __kmpc_distribute_static_init* if \a IsGPUDistribute is set.
-  llvm::FunctionCallee createForStaticInitFunction(unsigned IVSize,
-                                                   bool IVSigned,
-                                                   bool IsGPUDistribute);
-
-  /// Returns __kmpc_dispatch_init_* runtime function for the specified
-  /// size \a IVSize and sign \a IVSigned.
-  llvm::FunctionCallee createDispatchInitFunction(unsigned IVSize,
-                                                  bool IVSigned);
-
-  /// Returns __kmpc_dispatch_next_* runtime function for the specified
-  /// size \a IVSize and sign \a IVSigned.
-  llvm::FunctionCallee createDispatchNextFunction(unsigned IVSize,
-                                                  bool IVSigned);
-
-  /// Returns __kmpc_dispatch_fini_* runtime function for the specified
-  /// size \a IVSize and sign \a IVSigned.
-  llvm::FunctionCallee createDispatchFiniFunction(unsigned IVSize,
-                                                  bool IVSigned);
-
   /// If the specified mangled name is not in the module, create and
   /// return threadprivate cache object. This object is a pointer's worth of
   /// storage that's reserved for use by the OpenMP runtime.
   /// \param VD Threadprivate variable.
   /// \return Cache variable for the specified threadprivate.
   llvm::Constant *getOrCreateThreadPrivateCache(const VarDecl *VD);
 
   /// Set of threadprivate variables with the generated initializer.

clang/lib/CodeGen/CGOpenMPRuntime.cpp

if (!Kmpc_MicroTy) {
     // Build void (*kmpc_micro)(kmp_int32 *global_tid, kmp_int32 *bound_tid,...)
     llvm::Type *MicroParams[] = {llvm::PointerType::getUnqual(CGM.Int32Ty),
                                  llvm::PointerType::getUnqual(CGM.Int32Ty)};
     Kmpc_MicroTy = llvm::FunctionType::get(CGM.VoidTy, MicroParams, true);
   }
   return llvm::PointerType::getUnqual(Kmpc_MicroTy);
 }
 
-llvm::FunctionCallee
-CGOpenMPRuntime::createForStaticInitFunction(unsigned IVSize, bool IVSigned,
-                                             bool IsGPUDistribute) {
-  assert((IVSize == 32 || IVSize == 64) &&
-         "IV size is not compatible with the omp runtime");
-  StringRef Name;
-  if (IsGPUDistribute)
-    Name = IVSize == 32 ? (IVSigned ? "__kmpc_distribute_static_init_4"
-                                    : "__kmpc_distribute_static_init_4u")
-                        : (IVSigned ? "__kmpc_distribute_static_init_8"
-                                    : "__kmpc_distribute_static_init_8u");
-  else
-    Name = IVSize == 32 ? (IVSigned ? "__kmpc_for_static_init_4"
-                                    : "__kmpc_for_static_init_4u")
-                        : (IVSigned ? "__kmpc_for_static_init_8"
-                                    : "__kmpc_for_static_init_8u");
-
-  llvm::Type *ITy = IVSize == 32 ? CGM.Int32Ty : CGM.Int64Ty;
-  auto *PtrTy = llvm::PointerType::getUnqual(ITy);
-  llvm::Type *TypeParams[] = {
-    getIdentTyPointerTy(),                     // loc
-    CGM.Int32Ty,                               // tid
-    CGM.Int32Ty,                               // schedtype
-    llvm::PointerType::getUnqual(CGM.Int32Ty), // p_lastiter
-    PtrTy,                                     // p_lower
-    PtrTy,                                     // p_upper
-    PtrTy,                                     // p_stride
-    ITy,                                       // incr
-    ITy                                        // chunk
-  };
-  auto *FnTy =
-      llvm::FunctionType::get(CGM.VoidTy, TypeParams, /*isVarArg*/ false);
-  return CGM.CreateRuntimeFunction(FnTy, Name);
-}
-
-llvm::FunctionCallee
-CGOpenMPRuntime::createDispatchInitFunction(unsigned IVSize, bool IVSigned) {
-  assert((IVSize == 32 || IVSize == 64) &&
-         "IV size is not compatible with the omp runtime");
-  StringRef Name =
-      IVSize == 32
-          ? (IVSigned ? "__kmpc_dispatch_init_4" : "__kmpc_dispatch_init_4u")
-          : (IVSigned ? "__kmpc_dispatch_init_8" : "__kmpc_dispatch_init_8u");
-  llvm::Type *ITy = IVSize == 32 ? CGM.Int32Ty : CGM.Int64Ty;
-  llvm::Type *TypeParams[] = { getIdentTyPointerTy(), // loc
-                               CGM.Int32Ty,           // tid
-                               CGM.Int32Ty,           // schedtype
-                               ITy,                   // lower
-                               ITy,                   // upper
-                               ITy,                   // stride
-                               ITy                    // chunk
-  };
-  auto *FnTy =
-      llvm::FunctionType::get(CGM.VoidTy, TypeParams, /*isVarArg*/ false);
-  return CGM.CreateRuntimeFunction(FnTy, Name);
-}
-
-llvm::FunctionCallee
-CGOpenMPRuntime::createDispatchFiniFunction(unsigned IVSize, bool IVSigned) {
-  assert((IVSize == 32 || IVSize == 64) &&
-         "IV size is not compatible with the omp runtime");
-  StringRef Name =
-      IVSize == 32
-          ? (IVSigned ? "__kmpc_dispatch_fini_4" : "__kmpc_dispatch_fini_4u")
-          : (IVSigned ? "__kmpc_dispatch_fini_8" : "__kmpc_dispatch_fini_8u");
-  llvm::Type *TypeParams[] = {
-      getIdentTyPointerTy(), // loc
-      CGM.Int32Ty,           // tid
-  };
-  auto *FnTy =
-      llvm::FunctionType::get(CGM.VoidTy, TypeParams, /*isVarArg=*/false);
-  return CGM.CreateRuntimeFunction(FnTy, Name);
-}
-
-llvm::FunctionCallee
-CGOpenMPRuntime::createDispatchNextFunction(unsigned IVSize, bool IVSigned) {
-  assert((IVSize == 32 || IVSize == 64) &&
-         "IV size is not compatible with the omp runtime");
-  StringRef Name =
-      IVSize == 32
-          ? (IVSigned ? "__kmpc_dispatch_next_4" : "__kmpc_dispatch_next_4u")
-          : (IVSigned ? "__kmpc_dispatch_next_8" : "__kmpc_dispatch_next_8u");
-  llvm::Type *ITy = IVSize == 32 ? CGM.Int32Ty : CGM.Int64Ty;
-  auto *PtrTy = llvm::PointerType::getUnqual(ITy);
-  llvm::Type *TypeParams[] = {
-    getIdentTyPointerTy(),                     // loc
-    CGM.Int32Ty,                               // tid
-    llvm::PointerType::getUnqual(CGM.Int32Ty), // p_lastiter
-    PtrTy,                                     // p_lower
-    PtrTy,                                     // p_upper
-    PtrTy                                      // p_stride
-  };
-  auto *FnTy =
-      llvm::FunctionType::get(CGM.Int32Ty, TypeParams, /*isVarArg*/ false);
-  return CGM.CreateRuntimeFunction(FnTy, Name);
-}
-
 llvm::OffloadEntriesInfoManager::OMPTargetDeviceClauseKind
 convertDeviceClause(const VarDecl *VD) {
   std::optional<OMPDeclareTargetDeclAttr::DevTypeTy> DevTy =
       OMPDeclareTargetDeclAttr::getDeviceType(VD);
   if (!DevTy)
     return llvm::OffloadEntriesInfoManager::OMPTargetDeviceClauseNone;
 
   switch ((int)*DevTy) { // Avoid -Wcovered-switch-default
llvm::Value *Args[] = {
       getThreadID(CGF, Loc),
       CGF.Builder.getInt32(addMonoNonMonoModifier(
           CGM, Schedule, ScheduleKind.M1, ScheduleKind.M2)), // Schedule type
       DispatchValues.LB,                                     // Lower
       DispatchValues.UB,                                     // Upper
       CGF.Builder.getIntN(IVSize, 1),                        // Stride
       Chunk                                                  // Chunk
   };
-  CGF.EmitRuntimeCall(createDispatchInitFunction(IVSize, IVSigned), Args);
+  CGF.EmitRuntimeCall(OMPBuilder.createDispatchInitFunction(IVSize, IVSigned),
+                      Args);
 }
 
 static void emitForStaticInitCall(
     CodeGenFunction &CGF, llvm::Value *UpdateLocation, llvm::Value *ThreadId,
     llvm::FunctionCallee ForStaticInitFunction, OpenMPSchedType Schedule,
     OpenMPScheduleClauseModifier M1, OpenMPScheduleClauseModifier M2,
     const CGOpenMPRuntime::StaticRTInput &Values) {
   if (!CGF.HaveInsertPoint())
void CGOpenMPRuntime::emitForStaticInit(CodeGenFunction &CGF,
   assert((isOpenMPWorksharingDirective(DKind) || (DKind == OMPD_loop)) &&
          "Expected loop-based or sections-based directive.");
   llvm::Value *UpdatedLocation = emitUpdateLocation(CGF, Loc,
                                              isOpenMPLoopDirective(DKind)
                                                  ? OMP_IDENT_WORK_LOOP
                                                  : OMP_IDENT_WORK_SECTIONS);
   llvm::Value *ThreadId = getThreadID(CGF, Loc);
   llvm::FunctionCallee StaticInitFunction =
-      createForStaticInitFunction(Values.IVSize, Values.IVSigned, false);
+      OMPBuilder.createForStaticInitFunction(Values.IVSize, Values.IVSigned,
+                                             false);
   auto DL = ApplyDebugLocation::CreateDefaultArtificial(CGF, Loc);
   emitForStaticInitCall(CGF, UpdatedLocation, ThreadId, StaticInitFunction,
                         ScheduleNum, ScheduleKind.M1, ScheduleKind.M2, Values);
 }
 
 void CGOpenMPRuntime::emitDistributeStaticInit(
     CodeGenFunction &CGF, SourceLocation Loc,
     OpenMPDistScheduleClauseKind SchedKind,
     const CGOpenMPRuntime::StaticRTInput &Values) {
   OpenMPSchedType ScheduleNum =
       getRuntimeSchedule(SchedKind, Values.Chunk != nullptr);
   llvm::Value *UpdatedLocation =
       emitUpdateLocation(CGF, Loc, OMP_IDENT_WORK_DISTRIBUTE);
   llvm::Value *ThreadId = getThreadID(CGF, Loc);
   llvm::FunctionCallee StaticInitFunction;
   bool isGPUDistribute =
       CGM.getLangOpts().OpenMPIsTargetDevice &&
       (CGM.getTriple().isAMDGCN() || CGM.getTriple().isNVPTX());
-  StaticInitFunction = createForStaticInitFunction(
+  StaticInitFunction = OMPBuilder.createForStaticInitFunction(
       Values.IVSize, Values.IVSigned, isGPUDistribute);
 
   emitForStaticInitCall(CGF, UpdatedLocation, ThreadId, StaticInitFunction,
                         ScheduleNum, OMPC_SCHEDULE_MODIFIER_unknown,
                         OMPC_SCHEDULE_MODIFIER_unknown, Values);
 }
 
 void CGOpenMPRuntime::emitForStaticFinish(CodeGenFunction &CGF,

 void CGOpenMPRuntime::emitForOrderedIterationEnd(CodeGenFunction &CGF,
                                                  SourceLocation Loc,
                                                  unsigned IVSize,
                                                  bool IVSigned) {
   if (!CGF.HaveInsertPoint())
     return;
   // Call __kmpc_for_dynamic_fini_(4|8)[u](ident_t *loc, kmp_int32 tid);
   llvm::Value *Args[] = {emitUpdateLocation(CGF, Loc), getThreadID(CGF, Loc)};
-  CGF.EmitRuntimeCall(createDispatchFiniFunction(IVSize, IVSigned), Args);
+  CGF.EmitRuntimeCall(OMPBuilder.createDispatchFiniFunction(IVSize, IVSigned),
+                      Args);
 }
 
 llvm::Value *CGOpenMPRuntime::emitForNext(CodeGenFunction &CGF,
                                           SourceLocation Loc, unsigned IVSize,
                                           bool IVSigned, Address IL,
                                           Address LB, Address UB,
                                           Address ST) {
   // Call __kmpc_dispatch_next(
   //          ident_t *loc, kmp_int32 tid, kmp_int32 *p_lastiter,
   //          kmp_int[32|64] *p_lower, kmp_int[32|64] *p_upper,
   //          kmp_int[32|64] *p_stride);
   llvm::Value *Args[] = {
       emitUpdateLocation(CGF, Loc),
       getThreadID(CGF, Loc),
       IL.getPointer(), // &isLastIter
       LB.getPointer(), // &Lower
       UB.getPointer(), // &Upper
       ST.getPointer()  // &Stride
   };
-  llvm::Value *Call =
-      CGF.EmitRuntimeCall(createDispatchNextFunction(IVSize, IVSigned), Args);
+  llvm::Value *Call = CGF.EmitRuntimeCall(
+      OMPBuilder.createDispatchNextFunction(IVSize, IVSigned), Args);
   return CGF.EmitScalarConversion(
       Call, CGF.getContext().getIntTypeForBitwidth(32, /*Signed=*/1),
       CGF.getContext().BoolTy, Loc);
 }
 
 void CGOpenMPRuntime::emitNumThreadsClause(CodeGenFunction &CGF,
                                            llvm::Value *NumThreads,
                                            SourceLocation Loc) {

clang/test/OpenMP/ordered_codegen.cpp

 // CHECK1:       omp.dispatch.inc:
 // CHECK1-NEXT:    br label [[OMP_DISPATCH_COND]]
 // CHECK1:       omp.dispatch.end:
 // CHECK1-NEXT:    call void @__kmpc_barrier(ptr @[[GLOB2]], i32 [[TMP0]])
 // CHECK1-NEXT:    ret void
 //
 //
 // CHECK1-LABEL: define {{[^@]+}}@_Z8foo_simdii
-// CHECK1-SAME: (i32 noundef [[LOW:%.*]], i32 noundef [[UP:%.*]]) #[[ATTR0]] {
+// CHECK1-SAME: (i32 noundef [[LOW:%.*]], i32 noundef [[UP:%.*]]) #[[ATTR0:[0-9]+]] {
 // CHECK1-NEXT:  entry:
 // CHECK1-NEXT:    [[LOW_ADDR:%.*]] = alloca i32, align 4
 // CHECK1-NEXT:    [[UP_ADDR:%.*]] = alloca i32, align 4
 // CHECK1-NEXT:    [[TMP:%.*]] = alloca i32, align 4
 // CHECK1-NEXT:    [[DOTCAPTURE_EXPR_:%.*]] = alloca i32, align 4
 // CHECK1-NEXT:    [[DOTCAPTURE_EXPR_1:%.*]] = alloca i32, align 4
 // CHECK1-NEXT:    [[DOTCAPTURE_EXPR_2:%.*]] = alloca i32, align 4
 // CHECK1-NEXT:    [[I:%.*]] = alloca i32, align 4

 // CHECK1-IRBUILDER-NEXT:    br label [[OMP_DISPATCH_COND]]
 // CHECK1-IRBUILDER:       omp.dispatch.end:
 // CHECK1-IRBUILDER-NEXT:    [[OMP_GLOBAL_THREAD_NUM19:%.*]] = call i32 @__kmpc_global_thread_num(ptr @[[GLOB1]])
 // CHECK1-IRBUILDER-NEXT:    call void @__kmpc_barrier(ptr @[[GLOB4]], i32 [[OMP_GLOBAL_THREAD_NUM19]])
 // CHECK1-IRBUILDER-NEXT:    ret void
 //
 //
 // CHECK1-IRBUILDER-LABEL: define {{[^@]+}}@_Z8foo_simdii
-// CHECK1-IRBUILDER-SAME: (i32 noundef [[LOW:%.*]], i32 noundef [[UP:%.*]]) #[[ATTR0]] {
+// CHECK1-IRBUILDER-SAME: (i32 noundef [[LOW:%.*]], i32 noundef [[UP:%.*]]) #[[ATTR0:[0-9]+]] {
 // CHECK1-IRBUILDER-NEXT:  entry:
 // CHECK1-IRBUILDER-NEXT:    [[LOW_ADDR:%.*]] = alloca i32, align 4
 // CHECK1-IRBUILDER-NEXT:    [[UP_ADDR:%.*]] = alloca i32, align 4
 // CHECK1-IRBUILDER-NEXT:    [[TMP:%.*]] = alloca i32, align 4
 // CHECK1-IRBUILDER-NEXT:    [[DOTCAPTURE_EXPR_:%.*]] = alloca i32, align 4
 // CHECK1-IRBUILDER-NEXT:    [[DOTCAPTURE_EXPR_1:%.*]] = alloca i32, align 4
 // CHECK1-IRBUILDER-NEXT:    [[DOTCAPTURE_EXPR_2:%.*]] = alloca i32, align 4
 // CHECK1-IRBUILDER-NEXT:    [[I:%.*]] = alloca i32, align 4

 // CHECK3:       omp.dispatch.inc:
 // CHECK3-NEXT:    br label [[OMP_DISPATCH_COND]]
 // CHECK3:       omp.dispatch.end:
 // CHECK3-NEXT:    call void @__kmpc_barrier(ptr @[[GLOB2]], i32 [[TMP0]])
 // CHECK3-NEXT:    ret void
 //
 //
 // CHECK3-LABEL: define {{[^@]+}}@_Z8foo_simdii
-// CHECK3-SAME: (i32 noundef [[LOW:%.*]], i32 noundef [[UP:%.*]]) #[[ATTR0]] {
+// CHECK3-SAME: (i32 noundef [[LOW:%.*]], i32 noundef [[UP:%.*]]) #[[ATTR0:[0-9]+]] {
 // CHECK3-NEXT:  entry:
 // CHECK3-NEXT:    [[LOW_ADDR:%.*]] = alloca i32, align 4
 // CHECK3-NEXT:    [[UP_ADDR:%.*]] = alloca i32, align 4
 // CHECK3-NEXT:    [[TMP:%.*]] = alloca i32, align 4
 // CHECK3-NEXT:    [[DOTCAPTURE_EXPR_:%.*]] = alloca i32, align 4
 // CHECK3-NEXT:    [[DOTCAPTURE_EXPR_1:%.*]] = alloca i32, align 4
 // CHECK3-NEXT:    [[DOTCAPTURE_EXPR_2:%.*]] = alloca i32, align 4
 // CHECK3-NEXT:    [[I:%.*]] = alloca i32, align 4

 // CHECK3-IRBUILDER-NEXT:    br label [[OMP_DISPATCH_COND]]
 // CHECK3-IRBUILDER:       omp.dispatch.end:
 // CHECK3-IRBUILDER-NEXT:    [[OMP_GLOBAL_THREAD_NUM19:%.*]] = call i32 @__kmpc_global_thread_num(ptr @[[GLOB1]])
 // CHECK3-IRBUILDER-NEXT:    call void @__kmpc_barrier(ptr @[[GLOB4]], i32 [[OMP_GLOBAL_THREAD_NUM19]])
 // CHECK3-IRBUILDER-NEXT:    ret void
 //
 //
 // CHECK3-IRBUILDER-LABEL: define {{[^@]+}}@_Z8foo_simdii
-// CHECK3-IRBUILDER-SAME: (i32 noundef [[LOW:%.*]], i32 noundef [[UP:%.*]]) #[[ATTR0]] {
+// CHECK3-IRBUILDER-SAME: (i32 noundef [[LOW:%.*]], i32 noundef [[UP:%.*]]) #[[ATTR0:[0-9]+]] {
 // CHECK3-IRBUILDER-NEXT:  entry:
 // CHECK3-IRBUILDER-NEXT:    [[LOW_ADDR:%.*]] = alloca i32, align 4
 // CHECK3-IRBUILDER-NEXT:    [[UP_ADDR:%.*]] = alloca i32, align 4
 // CHECK3-IRBUILDER-NEXT:    [[TMP:%.*]] = alloca i32, align 4
 // CHECK3-IRBUILDER-NEXT:    [[DOTCAPTURE_EXPR_:%.*]] = alloca i32, align 4
 // CHECK3-IRBUILDER-NEXT:    [[DOTCAPTURE_EXPR_1:%.*]] = alloca i32, align 4
 // CHECK3-IRBUILDER-NEXT:    [[DOTCAPTURE_EXPR_2:%.*]] = alloca i32, align 4
 // CHECK3-IRBUILDER-NEXT:    [[I:%.*]] = alloca i32, align 4

llvm/include/llvm/Frontend/OpenMP/OMPIRBuilder.h

InsertPointTy createTarget(const LocationDescription &Loc,
                              OpenMPIRBuilder::InsertPointTy AllocaIP,
                              OpenMPIRBuilder::InsertPointTy CodeGenIP,
                              TargetRegionEntryInfo &EntryInfo, int32_t NumTeams,
                              int32_t NumThreads,
                              SmallVectorImpl<Value *> &Inputs,
                              GenMapInfoCallbackTy GenMapInfoCB,
                              TargetBodyGenCallbackTy BodyGenCB);
 
+  /// Returns __kmpc_for_static_init_* runtime function for the specified
+  /// size \a IVSize and sign \a IVSigned. Will create a distribute call
+  /// __kmpc_distribute_static_init* if \a IsGPUDistribute is set.
+  FunctionCallee createForStaticInitFunction(unsigned IVSize, bool IVSigned,
+                                             bool IsGPUDistribute);
+
+  /// Returns __kmpc_dispatch_init_* runtime function for the specified
+  /// size \a IVSize and sign \a IVSigned.
+  FunctionCallee createDispatchInitFunction(unsigned IVSize, bool IVSigned);
+
+  /// Returns __kmpc_dispatch_next_* runtime function for the specified
+  /// size \a IVSize and sign \a IVSigned.
+  FunctionCallee createDispatchNextFunction(unsigned IVSize, bool IVSigned);
+
+  /// Returns __kmpc_dispatch_fini_* runtime function for the specified
+  /// size \a IVSize and sign \a IVSigned.
+  FunctionCallee createDispatchFiniFunction(unsigned IVSize, bool IVSigned);
+
   /// Declarations for LLVM-IR types (simple, array, function and structure) are
   /// generated below. Their names are defined and used in OpenMPKinds.def. Here
   /// we provide the declarations, the initializeTypes function will provide the
   /// values.
   ///
   ///{
 #define OMP_TYPE(VarName, InitValue) Type *VarName = nullptr;
 #define OMP_ARRAY_TYPE(VarName, ElemTy, ArraySize)                             \

llvm/lib/Frontend/OpenMP/OMPIRBuilder.cpp

if (BodyGenCB) {
     } else {
       BeginThenGen(AllocaIP, Builder.saveIP());
     }
   }
 
   return Builder.saveIP();
 }
 
+FunctionCallee
+OpenMPIRBuilder::createForStaticInitFunction(unsigned IVSize, bool IVSigned,
+                                             bool IsGPUDistribute) {
+  assert((IVSize == 32 || IVSize == 64) &&
+         "IV size is not compatible with the omp runtime");
+  RuntimeFunction Name;
+  if (IsGPUDistribute)
+    Name = IVSize == 32
+               ? (IVSigned ? omp::OMPRTL___kmpc_distribute_static_init_4
+                           : omp::OMPRTL___kmpc_distribute_static_init_4u)
+               : (IVSigned ? omp::OMPRTL___kmpc_distribute_static_init_8
+                           : omp::OMPRTL___kmpc_distribute_static_init_8u);
+  else
+    Name = IVSize == 32 ? (IVSigned ? omp::OMPRTL___kmpc_for_static_init_4
+                                    : omp::OMPRTL___kmpc_for_static_init_4u)
+                        : (IVSigned ? omp::OMPRTL___kmpc_for_static_init_8
+                                    : omp::OMPRTL___kmpc_for_static_init_8u);
+
+  return getOrCreateRuntimeFunction(M, Name);
+}
+
+FunctionCallee OpenMPIRBuilder::createDispatchInitFunction(unsigned IVSize,
+                                                           bool IVSigned) {
+  assert((IVSize == 32 || IVSize == 64) &&
+         "IV size is not compatible with the omp runtime");
+  RuntimeFunction Name = IVSize == 32
+                             ? (IVSigned ? omp::OMPRTL___kmpc_dispatch_init_4
+                                         : omp::OMPRTL___kmpc_dispatch_init_4u)
+                             : (IVSigned ? omp::OMPRTL___kmpc_dispatch_init_8
+                                         : omp::OMPRTL___kmpc_dispatch_init_8u);
+
+  return getOrCreateRuntimeFunction(M, Name);
+}
+
+FunctionCallee OpenMPIRBuilder::createDispatchNextFunction(unsigned IVSize,
+                                                           bool IVSigned) {
+  assert((IVSize == 32 || IVSize == 64) &&
+         "IV size is not compatible with the omp runtime");
+  RuntimeFunction Name = IVSize == 32
+                             ? (IVSigned ? omp::OMPRTL___kmpc_dispatch_next_4
+                                         : omp::OMPRTL___kmpc_dispatch_next_4u)
+                             : (IVSigned ? omp::OMPRTL___kmpc_dispatch_next_8
+                                         : omp::OMPRTL___kmpc_dispatch_next_8u);
+
+  return getOrCreateRuntimeFunction(M, Name);
+}
+
+FunctionCallee OpenMPIRBuilder::createDispatchFiniFunction(unsigned IVSize,
+                                                           bool IVSigned) {
+  assert((IVSize == 32 || IVSize == 64) &&
+         "IV size is not compatible with the omp runtime");
+  RuntimeFunction Name = IVSize == 32
+                             ? (IVSigned ? omp::OMPRTL___kmpc_dispatch_fini_4
+                                         : omp::OMPRTL___kmpc_dispatch_fini_4u)
+                             : (IVSigned ? omp::OMPRTL___kmpc_dispatch_fini_8
+                                         : omp::OMPRTL___kmpc_dispatch_fini_8u);
+
+  return getOrCreateRuntimeFunction(M, Name);
+}
+
 // Copy input from pointer or i64 to the expected argument type.
 static Value *copyInput(IRBuilderBase &Builder, unsigned AddrSpace,
                         Value *Input, Argument &Arg) {
   auto Addr = Builder.CreateAlloca(Arg.getType()->isPointerTy()
                                        ? Arg.getType()
                                        : Type::getInt64Ty(Builder.getContext()),
                                    AddrSpace);
   auto AddrAscast =