1 files changed, 42 insertions, 11 deletions

diff --git a/clang/lib/CodeGen/CGOpenMPRuntime.cpp b/clang/lib/CodeGen/CGOpenMPRuntime.cpp
index 8606e80418d7..75709b3c7e78 100644
--- a/clang/lib/CodeGen/CGOpenMPRuntime.cpp
+++ b/clang/lib/CodeGen/CGOpenMPRuntime.cpp
@@ -6264,21 +6264,51 @@ Address CGOpenMPRuntime::getTaskReductionItem(CodeGenFunction &CGF,
       SharedLVal.getAlignment());
 }
 
-void CGOpenMPRuntime::emitTaskwaitCall(CodeGenFunction &CGF,
-                                       SourceLocation Loc) {
+void CGOpenMPRuntime::emitTaskwaitCall(CodeGenFunction &CGF, SourceLocation Loc,
+                                       const OMPTaskDataTy &Data) {
   if (!CGF.HaveInsertPoint())
     return;
 
-  if (CGF.CGM.getLangOpts().OpenMPIRBuilder) {
+  if (CGF.CGM.getLangOpts().OpenMPIRBuilder && Data.Dependences.empty()) {
+    // TODO: Need to support taskwait with dependences in the OpenMPIRBuilder.
     OMPBuilder.createTaskwait(CGF.Builder);
   } else {
-    // Build call kmp_int32 __kmpc_omp_taskwait(ident_t *loc, kmp_int32
-    // global_tid);
-    llvm::Value *Args[] = {emitUpdateLocation(CGF, Loc), getThreadID(CGF, Loc)};
-    // Ignore return result until untied tasks are supported.
-    CGF.EmitRuntimeCall(OMPBuilder.getOrCreateRuntimeFunction(
-                            CGM.getModule(), OMPRTL___kmpc_omp_taskwait),
-                        Args);
+    llvm::Value *ThreadID = getThreadID(CGF, Loc);
+    llvm::Value *UpLoc = emitUpdateLocation(CGF, Loc);
+    auto &M = CGM.getModule();
+    Address DependenciesArray = Address::invalid();
+    llvm::Value *NumOfElements;
+    std::tie(NumOfElements, DependenciesArray) =
+        emitDependClause(CGF, Data.Dependences, Loc);
+    llvm::Value *DepWaitTaskArgs[6];
+    if (!Data.Dependences.empty()) {
+      DepWaitTaskArgs[0] = UpLoc;
+      DepWaitTaskArgs[1] = ThreadID;
+      DepWaitTaskArgs[2] = NumOfElements;
+      DepWaitTaskArgs[3] = DependenciesArray.getPointer();
+      DepWaitTaskArgs[4] = CGF.Builder.getInt32(0);
+      DepWaitTaskArgs[5] = llvm::ConstantPointerNull::get(CGF.VoidPtrTy);
+
+      CodeGenFunction::RunCleanupsScope LocalScope(CGF);
+
+      // Build void __kmpc_omp_wait_deps(ident_t *, kmp_int32 gtid,
+      // kmp_int32 ndeps, kmp_depend_info_t *dep_list, kmp_int32
+      // ndeps_noalias, kmp_depend_info_t *noalias_dep_list); if dependence info
+      // is specified.
+      CGF.EmitRuntimeCall(
+          OMPBuilder.getOrCreateRuntimeFunction(M, OMPRTL___kmpc_omp_wait_deps),
+          DepWaitTaskArgs);
+
+    } else {
+
+      // Build call kmp_int32 __kmpc_omp_taskwait(ident_t *loc, kmp_int32
+      // global_tid);
+      llvm::Value *Args[] = {UpLoc, ThreadID};
+      // Ignore return result until untied tasks are supported.
+      CGF.EmitRuntimeCall(
+          OMPBuilder.getOrCreateRuntimeFunction(M, OMPRTL___kmpc_omp_taskwait),
+          Args);
+    }
   }
 
   if (auto *Region = dyn_cast_or_null<CGOpenMPRegionInfo>(CGF.CapturedStmtInfo))
@@ -13044,7 +13074,8 @@ Address CGOpenMPSIMDRuntime::getTaskReductionItem(CodeGenFunction &CGF,
 }
 
 void CGOpenMPSIMDRuntime::emitTaskwaitCall(CodeGenFunction &CGF,
-                                           SourceLocation Loc) {
+                                           SourceLocation Loc,
+                                           const OMPTaskDataTy &Data) {
   llvm_unreachable("Not supported in SIMD-only mode");
 }