[OPENMP] Emit __kmpc_cancel_barrier() and code for 'cancellation point' only if 'cancel' is found.

Patch improves codegen for OpenMP constructs. If the OpenMP region does not have internal 'cancel' construct, a call to 'void __kmpc_barrier()' runtime function is generated for all implicit/explicit barriers. If the region has inner 'cancel' directive, then
```
if (__kmpc_cancel_barrier())
  exit from outer construct;
```
code is generated.
Also, the code for 'canellation point' directive is not generated if parent directive does not have 'cancel' directive.

llvm-svn: 247681

1 files changed, 38 insertions, 22 deletions

diff --git a/clang/lib/CodeGen/CGStmtOpenMP.cpp b/clang/lib/CodeGen/CGStmtOpenMP.cpp
index 03f473631e0..70f86385d57 100644
--- a/clang/lib/CodeGen/CGStmtOpenMP.cpp
+++ b/clang/lib/CodeGen/CGStmtOpenMP.cpp
@@ -624,8 +624,9 @@ void CodeGenFunction::EmitOMPParallelDirective(const OMPParallelDirective &S) {
       // initialization of firstprivate variables or propagation master's thread
       // values of threadprivate variables to local instances of that variables
       // of all other implicit threads.
-      CGF.CGM.getOpenMPRuntime().emitBarrierCall(CGF, S.getLocStart(),
-                                                 OMPD_unknown);
+      CGF.CGM.getOpenMPRuntime().emitBarrierCall(
+          CGF, S.getLocStart(), OMPD_unknown, /*EmitChecks=*/false,
+          /*ForceSimpleCall=*/true);
     }
     CGF.EmitOMPPrivateClause(S, PrivateScope);
     CGF.EmitOMPReductionClauseInit(S, PrivateScope);
@@ -633,8 +634,9 @@ void CodeGenFunction::EmitOMPParallelDirective(const OMPParallelDirective &S) {
     CGF.EmitStmt(cast<CapturedStmt>(S.getAssociatedStmt())->getCapturedStmt());
     CGF.EmitOMPReductionClauseFinal(S);
     // Emit implicit barrier at the end of the 'parallel' directive.
-    CGF.CGM.getOpenMPRuntime().emitBarrierCall(CGF, S.getLocStart(),
-                                               OMPD_unknown);
+    CGF.CGM.getOpenMPRuntime().emitBarrierCall(
+        CGF, S.getLocStart(), OMPD_unknown, /*EmitChecks=*/false,
+        /*ForceSimpleCall=*/true);
   };
   emitCommonOMPParallelDirective(*this, S, OMPD_parallel, CodeGen);
 }
@@ -1238,8 +1240,9 @@ bool CodeGenFunction::EmitOMPWorksharingLoop(const OMPLoopDirective &S) {
       if (EmitOMPFirstprivateClause(S, LoopScope)) {
         // Emit implicit barrier to synchronize threads and avoid data races on
         // initialization of firstprivate variables.
-        CGM.getOpenMPRuntime().emitBarrierCall(*this, S.getLocStart(),
-                                               OMPD_unknown);
+        CGM.getOpenMPRuntime().emitBarrierCall(
+            *this, S.getLocStart(), OMPD_unknown, /*EmitChecks=*/false,
+            /*ForceSimpleCall=*/true);
       }
       EmitOMPPrivateClause(S, LoopScope);
       HasLastprivateClause = EmitOMPLastprivateClauseInit(S, LoopScope);
@@ -1321,7 +1324,8 @@ void CodeGenFunction::EmitOMPForDirective(const OMPForDirective &S) {
   auto &&CodeGen = [&S, &HasLastprivates](CodeGenFunction &CGF) {
     HasLastprivates = CGF.EmitOMPWorksharingLoop(S);
   };
-  CGM.getOpenMPRuntime().emitInlinedDirective(*this, OMPD_for, CodeGen);
+  CGM.getOpenMPRuntime().emitInlinedDirective(*this, OMPD_for, CodeGen,
+                                              S.hasCancel());
 
   // Emit an implicit barrier at the end.
   if (!S.getSingleClause<OMPNowaitClause>() || HasLastprivates) {
@@ -1416,8 +1420,9 @@ CodeGenFunction::EmitSections(const OMPExecutableDirective &S) {
       if (CGF.EmitOMPFirstprivateClause(S, LoopScope)) {
         // Emit implicit barrier to synchronize threads and avoid data races on
         // initialization of firstprivate variables.
-        CGF.CGM.getOpenMPRuntime().emitBarrierCall(CGF, S.getLocStart(),
-                                                   OMPD_unknown);
+        CGF.CGM.getOpenMPRuntime().emitBarrierCall(
+            CGF, S.getLocStart(), OMPD_unknown, /*EmitChecks=*/false,
+            /*ForceSimpleCall=*/true);
       }
       CGF.EmitOMPPrivateClause(S, LoopScope);
       HasLastprivates = CGF.EmitOMPLastprivateClauseInit(S, LoopScope);
@@ -1450,7 +1455,13 @@ CodeGenFunction::EmitSections(const OMPExecutableDirective &S) {
                    CGF.EmitLoadOfScalar(IL, S.getLocStart())));
     };
 
-    CGM.getOpenMPRuntime().emitInlinedDirective(*this, OMPD_sections, CodeGen);
+    bool HasCancel = false;
+    if (auto *OSD = dyn_cast<OMPSectionsDirective>(&S))
+      HasCancel = OSD->hasCancel();
+    else if (auto *OPSD = dyn_cast<OMPParallelSectionsDirective>(&S))
+      HasCancel = OPSD->hasCancel();
+    CGM.getOpenMPRuntime().emitInlinedDirective(*this, OMPD_sections, CodeGen,
+                                                HasCancel);
     // Emit barrier for lastprivates only if 'sections' directive has 'nowait'
     // clause. Otherwise the barrier will be generated by the codegen for the
     // directive.
@@ -1490,7 +1501,9 @@ CodeGenFunction::EmitSections(const OMPExecutableDirective &S) {
       S.getSingleClause<OMPNowaitClause>()) {
     // Emit implicit barrier to synchronize threads and avoid data races on
     // initialization of firstprivate variables.
-    CGM.getOpenMPRuntime().emitBarrierCall(*this, S.getLocStart(), OMPD_unknown);
+    CGM.getOpenMPRuntime().emitBarrierCall(*this, S.getLocStart(), OMPD_unknown,
+                                           /*EmitChecks=*/false,
+                                           /*ForceSimpleCall=*/true);
   }
   return OMPD_single;
 }
@@ -1510,7 +1523,8 @@ void CodeGenFunction::EmitOMPSectionDirective(const OMPSectionDirective &S) {
     CGF.EmitStmt(cast<CapturedStmt>(S.getAssociatedStmt())->getCapturedStmt());
     CGF.EnsureInsertPoint();
   };
-  CGM.getOpenMPRuntime().emitInlinedDirective(*this, OMPD_section, CodeGen);
+  CGM.getOpenMPRuntime().emitInlinedDirective(*this, OMPD_section, CodeGen,
+                                              S.hasCancel());
 }
 
 void CodeGenFunction::EmitOMPSingleDirective(const OMPSingleDirective &S) {
@@ -1586,8 +1600,9 @@ void CodeGenFunction::EmitOMPParallelForDirective(
     // Emit implicit barrier at the end of parallel region, but this barrier
     // is at the end of 'for' directive, so emit it as the implicit barrier for
     // this 'for' directive.
-    CGF.CGM.getOpenMPRuntime().emitBarrierCall(CGF, S.getLocStart(),
-                                               OMPD_parallel);
+    CGF.CGM.getOpenMPRuntime().emitBarrierCall(
+        CGF, S.getLocStart(), OMPD_parallel, /*EmitChecks=*/false,
+        /*ForceSimpleCall=*/true);
   };
   emitCommonOMPParallelDirective(*this, S, OMPD_for, CodeGen);
 }
@@ -1603,8 +1618,9 @@ void CodeGenFunction::EmitOMPParallelForSimdDirective(
     // Emit implicit barrier at the end of parallel region, but this barrier
     // is at the end of 'for' directive, so emit it as the implicit barrier for
     // this 'for' directive.
-    CGF.CGM.getOpenMPRuntime().emitBarrierCall(CGF, S.getLocStart(),
-                                               OMPD_parallel);
+    CGF.CGM.getOpenMPRuntime().emitBarrierCall(
+        CGF, S.getLocStart(), OMPD_parallel, /*EmitChecks=*/false,
+        /*ForceSimpleCall=*/true);
   };
   emitCommonOMPParallelDirective(*this, S, OMPD_simd, CodeGen);
 }
@@ -1617,8 +1633,9 @@ void CodeGenFunction::EmitOMPParallelSectionsDirective(
   auto &&CodeGen = [&S](CodeGenFunction &CGF) {
     (void)CGF.EmitSections(S);
     // Emit implicit barrier at the end of parallel region.
-    CGF.CGM.getOpenMPRuntime().emitBarrierCall(CGF, S.getLocStart(),
-                                               OMPD_parallel);
+    CGF.CGM.getOpenMPRuntime().emitBarrierCall(
+        CGF, S.getLocStart(), OMPD_parallel, /*EmitChecks=*/false,
+        /*ForceSimpleCall=*/true);
   };
   emitCommonOMPParallelDirective(*this, S, OMPD_sections, CodeGen);
 }
@@ -2253,10 +2270,9 @@ CodeGenFunction::JumpDest
 CodeGenFunction::getOMPCancelDestination(OpenMPDirectiveKind Kind) {
   if (Kind == OMPD_parallel || Kind == OMPD_task)
     return ReturnBlock;
-  else if (Kind == OMPD_for || Kind == OMPD_section || Kind == OMPD_sections)
-    return BreakContinueStack.empty() ? JumpDest()
-                                      : BreakContinueStack.back().BreakBlock;
-  return JumpDest();
+  assert(Kind == OMPD_for || Kind == OMPD_section || Kind == OMPD_sections ||
+         Kind == OMPD_parallel_sections || Kind == OMPD_parallel_for);
+  return BreakContinueStack.back().BreakBlock;
 }
 
 // Generate the instructions for '#pragma omp target data' directive.