[OPENMP] Codegen for 'taskyield' directive

[ABataev]

Summary

For 'taskyield' directive emit call to kmp_int32 __kmpc_omp_taskyield(ident_t *, kmp_int32 global_tid, int end_part); runtime function call with end_part arg set to 0 (it is ignored).

[ABataev]

Ping!

[rjmccall]

LGTM.

cfe/trunk/lib/CodeGen/CGOpenMPRuntime.h

enum OpenMPRTLFunction {
     // 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,
+    // Call to kmp_int32 __kmpc_omp_taskyield(ident_t *, kmp_int32 global_tid,
+    // int end_part);
+    OMPRTL__kmpc_omp_taskyield,
   };
 
   /// \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
   enum OpenMPLocationFlags {
     /// \brief Use trampoline for internal microtask.
     OMP_IDENT_IMD = 0x01,
public:
 
   /// \brief Emits a master region.
   /// \param MasterOpGen Generator for the statement associated with the given
   /// master region.
   virtual void EmitOMPMasterRegion(CodeGenFunction &CGF,
                                    const std::function<void()> &MasterOpGen,
                                    SourceLocation Loc);
 
+  /// \brief Emits code for a taskyield directive.
+  virtual void EmitOMPTaskyieldCall(CodeGenFunction &CGF, SourceLocation Loc);
+
   /// \brief Emits explicit barrier for OpenMP threads.
   /// \param IsExplicit true, if it is explicitly specified barrier.
   ///
   virtual void EmitOMPBarrierCall(CodeGenFunction &CGF, SourceLocation Loc,
                                   bool IsExplicit = true);
 
   /// \brief Check if the specified \a ScheduleKind is static non-chunked.
   /// This kind of worksharing directive is emitted without outer loop.

cfe/trunk/lib/CodeGen/CGOpenMPRuntime.cpp

CGOpenMPRuntime::CreateRuntimeFunction(OpenMPRTLFunction Function) {
   case OMPRTL__kmpc_end_master: {
     // Build void __kmpc_end_master(ident_t *loc, kmp_int32 global_tid);
     llvm::Type *TypeParams[] = {getIdentTyPointerTy(), CGM.Int32Ty};
     llvm::FunctionType *FnTy =
         llvm::FunctionType::get(CGM.VoidTy, TypeParams, /*isVarArg=*/false);
     RTLFn = CGM.CreateRuntimeFunction(FnTy, /*Name=*/"__kmpc_end_master");
     break;
   }
+  case OMPRTL__kmpc_omp_taskyield: {
+    // Build kmp_int32 __kmpc_omp_taskyield(ident_t *, kmp_int32 global_tid,
+    // int end_part);
+    llvm::Type *TypeParams[] = {getIdentTyPointerTy(), CGM.Int32Ty, CGM.IntTy};
+    llvm::FunctionType *FnTy =
+        llvm::FunctionType::get(CGM.Int32Ty, TypeParams, /*isVarArg=*/false);
+    RTLFn = CGM.CreateRuntimeFunction(FnTy, /*Name=*/"__kmpc_omp_taskyield");
+    break;
+  }
   }
   return RTLFn;
 }
 
 llvm::Constant *
 CGOpenMPRuntime::getOrCreateThreadPrivateCache(const VarDecl *VD) {
   // Lookup the entry, lazily creating it if necessary.
   return GetOrCreateInternalVariable(CGM.Int8PtrPtrTy,
EmitOMPIfStmt(CGF, IsMaster, [&]() -> void {
     // structured block.
     // It is analyzed in Sema, so we can just call __kmpc_end_master() on
     // fallthrough rather than pushing a normal cleanup for it.
     RTLFn = CreateRuntimeFunction(OMPRTL__kmpc_end_master);
     CGF.EmitRuntimeCall(RTLFn, Args);
   });
 }
 
+void CGOpenMPRuntime::EmitOMPTaskyieldCall(CodeGenFunction &CGF,
+                                           SourceLocation Loc) {
+  // Build call __kmpc_omp_taskyield(loc, thread_id, 0);
+  llvm::Value *Args[] = {
+      EmitOpenMPUpdateLocation(CGF, Loc), GetOpenMPThreadID(CGF, Loc),
+      llvm::ConstantInt::get(CGM.IntTy, /*V=*/0, /*isSigned=*/true)};
+  auto RTLFn = CreateRuntimeFunction(OMPRTL__kmpc_omp_taskyield);
+  CGF.EmitRuntimeCall(RTLFn, Args);
+}
+
 void CGOpenMPRuntime::EmitOMPBarrierCall(CodeGenFunction &CGF,
                                          SourceLocation Loc, bool IsExplicit) {
   // Build call __kmpc_cancel_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

cfe/trunk/lib/CodeGen/CGStmtOpenMP.cpp

void CodeGenFunction::EmitOMPParallelSectionsDirective(
     const OMPParallelSectionsDirective &) {
   llvm_unreachable("CodeGen for 'omp parallel sections' is not supported yet.");
 }
 
 void CodeGenFunction::EmitOMPTaskDirective(const OMPTaskDirective &) {
   llvm_unreachable("CodeGen for 'omp task' is not supported yet.");
 }
 
-void CodeGenFunction::EmitOMPTaskyieldDirective(const OMPTaskyieldDirective &) {
-  llvm_unreachable("CodeGen for 'omp taskyield' is not supported yet.");
+void CodeGenFunction::EmitOMPTaskyieldDirective(
+    const OMPTaskyieldDirective &S) {
+  CGM.getOpenMPRuntime().EmitOMPTaskyieldCall(*this, S.getLocStart());
 }
 
 void CodeGenFunction::EmitOMPBarrierDirective(const OMPBarrierDirective &S) {
   CGM.getOpenMPRuntime().EmitOMPBarrierCall(*this, S.getLocStart());
 }
 
 void CodeGenFunction::EmitOMPTaskwaitDirective(const OMPTaskwaitDirective &) {
   llvm_unreachable("CodeGen for 'omp taskwait' is not supported yet.");

cfe/trunk/test/OpenMP/taskyield_codegen.cpp

+// RUN: %clang_cc1 -verify -fopenmp=libiomp5 -x c++ -triple x86_64-unknown-unknown -emit-llvm %s -o - | FileCheck %s
+// RUN: %clang_cc1 -fopenmp=libiomp5 -x c++ -std=c++11 -triple x86_64-unknown-unknown -emit-pch -o %t %s
+// RUN: %clang_cc1 -fopenmp=libiomp5 -x c++ -triple x86_64-unknown-unknown -std=c++11 -include-pch %t -verify %s -emit-llvm -o - | FileCheck %s
+// expected-no-diagnostics
+
+#ifndef HEADER
+#define HEADER
+
+// CHECK: [[IDENT_T:%.+]] = type { i32, i32, i32, i32, i8* }
+
+void foo() {}
+
+template <class T>
+T tmain(T argc) {
+  static T a;
+#pragma omp taskyield
+  return a + argc;
+}
+
+// CHECK-LABEL: @main
+int main(int argc, char **argv) {
+  static int a;
+#pragma omp taskyield
+  // CHECK: [[GTID:%.+]] = call i32 @__kmpc_global_thread_num([[IDENT_T]]* @{{.+}})
+  // CHECK: call i32 @__kmpc_omp_taskyield([[IDENT_T]]* @{{.+}}, i32 [[GTID]], i32 0)
+  // CHECK: call {{.+}} [[TMAIN_INT:@.+]](i{{[0-9][0-9]}}
+  // CHECK: call {{.+}} [[TMAIN_CHAR:@.+]](i{{[0-9]}}
+  return tmain(argc) + tmain(argv[0][0]) + a;
+}
+
+// CHECK: define {{.+}} [[TMAIN_INT]](
+// CHECK: [[GTID:%.+]] = call i32 @__kmpc_global_thread_num([[IDENT_T]]* @{{.+}})
+// CHECK: call i32 @__kmpc_omp_taskyield([[IDENT_T]]* @{{.+}}, i32 [[GTID]], i32 0)
+
+// CHECK: define {{.+}} [[TMAIN_CHAR]](
+// CHECK: [[GTID:%.+]] = call i32 @__kmpc_global_thread_num([[IDENT_T]]* @{{.+}})
+// CHECK: call i32 @__kmpc_omp_taskyield([[IDENT_T]]* @{{.+}}, i32 [[GTID]], i32 0)
+
+#endif