[OPENMP] Codegen for 'omp barrier' directive.

Adds generation of call to "i32 kmpc_cancel_barrier(ident_t *, i32)" libcall for explicitly specified barriers (OMP_IDENT_BARRIER_EXPL flag is added to "flags" field of "ident_t" structure).
Also this patch replaces all calls to "kmpc_barrier" function by calls of "__kmpc_cancel_barrier" function which provides additional functionality for OpenMP 4.0.
Also, library specific enum OpenMPLocationFlags moved to private section of CGOpenMPRuntime class to make it more independent from library implementation.
Differential Revision: http://reviews.llvm.org/D6447

llvm-svn: 223444

clang/lib/CodeGen/CGOpenMPRuntime.cpp

@@ -73,14 +73,10 @@ void CGOpenMPRegionInfo::EmitBody(CodeGenFunction &CGF, Stmt *S) {
   CodeGenFunction::OMPPrivateScope PrivateScope(CGF);
   CGF.EmitOMPPrivateClause(Directive, PrivateScope);
   CGF.EmitOMPFirstprivateClause(Directive, PrivateScope);
-  if (PrivateScope.Privatize()) {
+  if (PrivateScope.Privatize())
     // Emit implicit barrier to synchronize threads and avoid data races.
-    auto Flags = static_cast<CGOpenMPRuntime::OpenMPLocationFlags>(
-        CGOpenMPRuntime::OMP_IDENT_KMPC |
-        CGOpenMPRuntime::OMP_IDENT_BARRIER_IMPL);
     CGF.CGM.getOpenMPRuntime().EmitOMPBarrierCall(CGF, Directive.getLocStart(),
-                                                  Flags);
+                                                  /*IsExplicit=*/false);
-  }
   CGCapturedStmtInfo::EmitBody(CGF, S);
 }
 
@@ -327,12 +323,13 @@ CGOpenMPRuntime::CreateRuntimeFunction(OpenMPRTLFunction Function) {
     RTLFn = CGM.CreateRuntimeFunction(FnTy, "__kmpc_end_critical");
     break;
   }
-  case OMPRTL__kmpc_barrier: {
+  case OMPRTL__kmpc_cancel_barrier: {
-    // Build void __kmpc_barrier(ident_t *loc, kmp_int32 global_tid);
+    // Build kmp_int32 __kmpc_cancel_barrier(ident_t *loc, kmp_int32
+    // global_tid);
     llvm::Type *TypeParams[] = {getIdentTyPointerTy(), CGM.Int32Ty};
     llvm::FunctionType *FnTy =
-        llvm::FunctionType::get(CGM.VoidTy, TypeParams, /*isVarArg*/ false);
+        llvm::FunctionType::get(CGM.Int32Ty, TypeParams, /*isVarArg*/ false);
-    RTLFn = CGM.CreateRuntimeFunction(FnTy, /*Name*/ "__kmpc_barrier");
+    RTLFn = CGM.CreateRuntimeFunction(FnTy, /*Name*/ "__kmpc_cancel_barrier");
     break;
   }
   case OMPRTL__kmpc_push_num_threads: {
@@ -699,12 +696,20 @@ void CGOpenMPRuntime::EmitOMPMasterRegion(
 }
 
 void CGOpenMPRuntime::EmitOMPBarrierCall(CodeGenFunction &CGF,
-                                         SourceLocation Loc,
+                                         SourceLocation Loc, bool IsExplicit) {
-                                         OpenMPLocationFlags Flags) {
+  // Build call __kmpc_cancel_barrier(loc, thread_id);
-  // Build call __kmpc_barrier(loc, thread_id)
+  auto Flags = static_cast<OpenMPLocationFlags>(
+      OMP_IDENT_KMPC |
+      (IsExplicit ? OMP_IDENT_BARRIER_EXPL : OMP_IDENT_BARRIER_IMPL));
+  // Build call __kmpc_cancel_barrier(loc, thread_id);
+  // Replace __kmpc_barrier() function by __kmpc_cancel_barrier() because this
+  // one provides the same functionality and adds initial support for
+  // cancellation constructs introduced in OpenMP 4.0. __kmpc_cancel_barrier()
+  // is provided default by the runtime library so it safe to make such
+  // replacement.
   llvm::Value *Args[] = {EmitOpenMPUpdateLocation(CGF, Loc, Flags),
                          GetOpenMPThreadID(CGF, Loc)};
-  auto RTLFn = CreateRuntimeFunction(OMPRTL__kmpc_barrier);
+  auto RTLFn = CreateRuntimeFunction(OMPRTL__kmpc_cancel_barrier);
   CGF.EmitRuntimeCall(RTLFn, Args);
 }
 

clang/lib/CodeGen/CGOpenMPRuntime.h

@@ -43,6 +43,46 @@ class CodeGenModule;
 
 class CGOpenMPRuntime {
 public:
+
+private:
+  enum OpenMPRTLFunction {
+    /// \brief Call to void __kmpc_fork_call(ident_t *loc, kmp_int32 argc,
+    /// kmpc_micro microtask, ...);
+    OMPRTL__kmpc_fork_call,
+    /// \brief Call to void *__kmpc_threadprivate_cached(ident_t *loc,
+    /// kmp_int32 global_tid, void *data, size_t size, void ***cache);
+    OMPRTL__kmpc_threadprivate_cached,
+    /// \brief Call to void __kmpc_threadprivate_register( ident_t *,
+    /// void *data, kmpc_ctor ctor, kmpc_cctor cctor, kmpc_dtor dtor);
+    OMPRTL__kmpc_threadprivate_register,
+    // Call to __kmpc_int32 kmpc_global_thread_num(ident_t *loc);
+    OMPRTL__kmpc_global_thread_num,
+    // Call to void __kmpc_critical(ident_t *loc, kmp_int32 global_tid,
+    // kmp_critical_name *crit);
+    OMPRTL__kmpc_critical,
+    // Call to void __kmpc_end_critical(ident_t *loc, kmp_int32 global_tid,
+    // kmp_critical_name *crit);
+    OMPRTL__kmpc_end_critical,
+    // Call to kmp_int32 __kmpc_cancel_barrier(ident_t *loc, kmp_int32
+    // global_tid);
+    OMPRTL__kmpc_cancel_barrier,
+    // Call to void __kmpc_serialized_parallel(ident_t *loc, kmp_int32
+    // global_tid);
+    OMPRTL__kmpc_serialized_parallel,
+    // Call to void __kmpc_end_serialized_parallel(ident_t *loc, kmp_int32
+    // global_tid);
+    OMPRTL__kmpc_end_serialized_parallel,
+    // Call to void __kmpc_push_num_threads(ident_t *loc, kmp_int32 global_tid,
+    // kmp_int32 num_threads);
+    OMPRTL__kmpc_push_num_threads,
+    // Call to void __kmpc_flush(ident_t *loc, ...);
+    OMPRTL__kmpc_flush,
+    // Call to kmp_int32 __kmpc_master(ident_t *, kmp_int32 global_tid);
+    OMPRTL__kmpc_master,
+    // Call to void __kmpc_end_master(ident_t *, kmp_int32 global_tid);
+    OMPRTL__kmpc_end_master,
+  };
+
   /// \brief Values for bit flags used in the ident_t to describe the fields.
   /// All enumeric elements are named and described in accordance with the code
   /// from http://llvm.org/svn/llvm-project/openmp/trunk/runtime/src/kmp.h
@@ -64,45 +104,6 @@ public:
     /// \brief Implicit barrier in 'single' directive.
     OMP_IDENT_BARRIER_IMPL_SINGLE = 0x140
   };
-
-private:
-  enum OpenMPRTLFunction {
-    /// \brief Call to void __kmpc_fork_call(ident_t *loc, kmp_int32 argc,
-    /// kmpc_micro microtask, ...);
-    OMPRTL__kmpc_fork_call,
-    /// \brief Call to void *__kmpc_threadprivate_cached(ident_t *loc,
-    /// kmp_int32 global_tid, void *data, size_t size, void ***cache);
-    OMPRTL__kmpc_threadprivate_cached,
-    /// \brief Call to void __kmpc_threadprivate_register( ident_t *,
-    /// void *data, kmpc_ctor ctor, kmpc_cctor cctor, kmpc_dtor dtor);
-    OMPRTL__kmpc_threadprivate_register,
-    // Call to __kmpc_int32 kmpc_global_thread_num(ident_t *loc);
-    OMPRTL__kmpc_global_thread_num,
-    // Call to void __kmpc_critical(ident_t *loc, kmp_int32 global_tid,
-    // kmp_critical_name *crit);
-    OMPRTL__kmpc_critical,
-    // Call to void __kmpc_end_critical(ident_t *loc, kmp_int32 global_tid,
-    // kmp_critical_name *crit);
-    OMPRTL__kmpc_end_critical,
-    // Call to void __kmpc_barrier(ident_t *loc, kmp_int32 global_tid);
-    OMPRTL__kmpc_barrier,
-    // Call to void __kmpc_serialized_parallel(ident_t *loc, kmp_int32
-    // global_tid);
-    OMPRTL__kmpc_serialized_parallel,
-    // Call to void __kmpc_end_serialized_parallel(ident_t *loc, kmp_int32
-    // global_tid);
-    OMPRTL__kmpc_end_serialized_parallel,
-    // Call to void __kmpc_push_num_threads(ident_t *loc, kmp_int32 global_tid,
-    // kmp_int32 num_threads);
-    OMPRTL__kmpc_push_num_threads,
-    // Call to void __kmpc_flush(ident_t *loc, ...);
-    OMPRTL__kmpc_flush,
-    // Call to kmp_int32 __kmpc_master(ident_t *, kmp_int32 global_tid);
-    OMPRTL__kmpc_master,
-    // Call to void __kmpc_end_master(ident_t *, kmp_int32 global_tid);
-    OMPRTL__kmpc_end_master,
-  };
-
   CodeGenModule &CGM;
   /// \brief Default const ident_t object used for initialization of all other
   /// ident_t objects.
@@ -298,11 +299,11 @@ public:
                                    const std::function<void()> &MasterOpGen,
                                    SourceLocation Loc);
 
-  /// \brief Emits a barrier for OpenMP threads.
+  /// \brief Emits explicit barrier for OpenMP threads.
-  /// \param Flags Flags for the barrier.
+  /// \param IsExplicit true, if it is explicitly specified barrier.
   ///
   virtual void EmitOMPBarrierCall(CodeGenFunction &CGF, SourceLocation Loc,
-                                  OpenMPLocationFlags Flags);
+                                  bool IsExplicit = true);
 
   /// \brief Emits call to void __kmpc_push_num_threads(ident_t *loc, kmp_int32
   /// global_tid, kmp_int32 num_threads) to generate code for 'num_threads'

clang/lib/CodeGen/CGStmtOpenMP.cpp

@@ -531,8 +531,8 @@ void CodeGenFunction::EmitOMPTaskyieldDirective(const OMPTaskyieldDirective &) {
   llvm_unreachable("CodeGen for 'omp taskyield' is not supported yet.");
 }
 
-void CodeGenFunction::EmitOMPBarrierDirective(const OMPBarrierDirective &) {
+void CodeGenFunction::EmitOMPBarrierDirective(const OMPBarrierDirective &S) {
-  llvm_unreachable("CodeGen for 'omp barrier' is not supported yet.");
+  CGM.getOpenMPRuntime().EmitOMPBarrierCall(*this, S.getLocStart());
 }
 
 void CodeGenFunction::EmitOMPTaskwaitDirective(const OMPTaskwaitDirective &) {

clang/test/OpenMP/parallel_firstprivate_codegen.cpp

@@ -112,7 +112,7 @@ int main() {
 // CHECK: call {{.*}} [[ST_TY_DESTR]]([[ST_TY]]* [[ST_TY_TEMP]])
 // CHECK: [[GTID_REF:%.+]] = load i{{[0-9]+}}** [[GTID_ADDR_ADDR]]
 // CHECK: [[GTID:%.+]] = load i{{[0-9]+}}* [[GTID_REF]]
-// CHECK: call void @__kmpc_barrier(%{{.+}}* [[IMPLICIT_BARRIER_LOC]], i{{[0-9]+}} [[GTID]])
+// CHECK: call i32 @__kmpc_cancel_barrier(%{{.+}}* [[IMPLICIT_BARRIER_LOC]], i{{[0-9]+}} [[GTID]])
 // CHECK-DAG: call {{.*}} [[S_FLOAT_TY_DESTR]]([[S_FLOAT_TY]]* [[VAR_PRIV]])
 // CHECK-DAG: call {{.*}} [[S_FLOAT_TY_DESTR]]([[S_FLOAT_TY]]*
 // CHECK: ret void
@@ -167,7 +167,7 @@ int main() {
 // CHECK: call {{.*}} [[ST_TY_DESTR]]([[ST_TY]]* [[ST_TY_TEMP]])
 // CHECK: [[GTID_REF:%.+]] = load i{{[0-9]+}}** [[GTID_ADDR_ADDR]]
 // CHECK: [[GTID:%.+]] = load i{{[0-9]+}}* [[GTID_REF]]
-// CHECK: call void @__kmpc_barrier(%{{.+}}* [[IMPLICIT_BARRIER_LOC]], i{{[0-9]+}} [[GTID]])
+// CHECK: call i32 @__kmpc_cancel_barrier(%{{.+}}* [[IMPLICIT_BARRIER_LOC]], i{{[0-9]+}} [[GTID]])
 // CHECK-DAG: call {{.*}} [[S_INT_TY_DESTR]]([[S_INT_TY]]* [[VAR_PRIV]])
 // CHECK-DAG: call {{.*}} [[S_INT_TY_DESTR]]([[S_INT_TY]]*
 // CHECK: ret void

clang/test/OpenMP/parallel_private_codegen.cpp

@@ -73,7 +73,7 @@ int main() {
 // CHECK: call {{.*}} [[S_FLOAT_TY_DEF_CONSTR]]([[S_FLOAT_TY]]* [[VAR_PRIV]])
 // CHECK: [[GTID_REF:%.+]] = load i{{[0-9]+}}** [[GTID_ADDR_REF]]
 // CHECK: [[GTID:%.+]] = load i{{[0-9]+}}* [[GTID_REF]]
-// CHECK: call void @__kmpc_barrier(%{{.+}}* [[IMPLICIT_BARRIER_LOC]], i{{[0-9]+}} [[GTID]])
+// CHECK: call i32 @__kmpc_cancel_barrier(%{{.+}}* [[IMPLICIT_BARRIER_LOC]], i{{[0-9]+}} [[GTID]])
 // CHECK-DAG: call void [[S_FLOAT_TY_DESTR]]([[S_FLOAT_TY]]* [[VAR_PRIV]])
 // CHECK-DAG: call void [[S_FLOAT_TY_DESTR]]([[S_FLOAT_TY]]*
 // CHECK: ret void
@@ -101,7 +101,7 @@ int main() {
 // CHECK: call {{.*}} [[S_INT_TY_DEF_CONSTR]]([[S_INT_TY]]* [[VAR_PRIV]])
 // CHECK: [[GTID_REF:%.+]] = load i{{[0-9]+}}** [[GTID_ADDR_REF]]
 // CHECK: [[GTID:%.+]] = load i{{[0-9]+}}* [[GTID_REF]]
-// CHECK: call void @__kmpc_barrier(%{{.+}}* [[IMPLICIT_BARRIER_LOC]], i{{[0-9]+}} [[GTID]])
+// CHECK: call i32 @__kmpc_cancel_barrier(%{{.+}}* [[IMPLICIT_BARRIER_LOC]], i{{[0-9]+}} [[GTID]])
 // CHECK-DAG: call void [[S_INT_TY_DESTR]]([[S_INT_TY]]* [[VAR_PRIV]])
 // CHECK-DAG: call void [[S_INT_TY_DESTR]]([[S_INT_TY]]*
 // CHECK: ret void