[OPENMP][NVPTX]Fix critical region codegen.

Summary:
Previously critical regions were emitted with the barrier making it a
worksharing construct though it is not. Also, it leads to incorrect
behavior in Cuda9+. Patch fixes this problem.

Reviewers: ABataev, jdoerfert

Subscribers: jholewinski, guansong, cfe-commits, grokos

Tags: #clang

Differential Revision: https://reviews.llvm.org/D66673

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@369946 91177308-0d34-0410-b5e6-96231b3b80d8

1 files changed, 24 insertions, 2 deletions

diff --git a/lib/CodeGen/CGOpenMPRuntimeNVPTX.cpp b/lib/CodeGen/CGOpenMPRuntimeNVPTX.cpp
index 284d14b995..6c908412e0 100644
--- a/lib/CodeGen/CGOpenMPRuntimeNVPTX.cpp
+++ b/lib/CodeGen/CGOpenMPRuntimeNVPTX.cpp
@@ -107,6 +107,10 @@ enum OpenMPRTLFunctionNVPTX {
   /// Call to void __kmpc_barrier_simple_spmd(ident_t *loc, kmp_int32
   /// global_tid);
   OMPRTL__kmpc_barrier_simple_spmd,
+  /// Call to int32_t __kmpc_warp_active_thread_mask(void);
+  OMPRTL_NVPTX__kmpc_warp_active_thread_mask,
+  /// Call to void __kmpc_syncwarp(int32_t Mask);
+  OMPRTL_NVPTX__kmpc_syncwarp,
 };
 
 /// Pre(post)-action for different OpenMP constructs specialized for NVPTX.
@@ -1794,6 +1798,20 @@ CGOpenMPRuntimeNVPTX::createNVPTXRuntimeFunction(unsigned Function) {
         ->addFnAttr(llvm::Attribute::Convergent);
     break;
   }
+  case OMPRTL_NVPTX__kmpc_warp_active_thread_mask: {
+    // Build int32_t __kmpc_warp_active_thread_mask(void);
+    auto *FnTy =
+        llvm::FunctionType::get(CGM.Int32Ty, llvm::None, /*isVarArg=*/false);
+    RTLFn = CGM.CreateRuntimeFunction(FnTy, "__kmpc_warp_active_thread_mask");
+    break;
+  }
+  case OMPRTL_NVPTX__kmpc_syncwarp: {
+    // Build void __kmpc_syncwarp(kmp_int32 Mask);
+    auto *FnTy =
+        llvm::FunctionType::get(CGM.VoidTy, CGM.Int32Ty, /*isVarArg=*/false);
+    RTLFn = CGM.CreateRuntimeFunction(FnTy, "__kmpc_syncwarp");
+    break;
+  }
   }
   return RTLFn;
 }
@@ -2700,6 +2718,9 @@ void CGOpenMPRuntimeNVPTX::emitCriticalRegion(
   llvm::BasicBlock *BodyBB = CGF.createBasicBlock("omp.critical.body");
   llvm::BasicBlock *ExitBB = CGF.createBasicBlock("omp.critical.exit");
 
+  // Get the mask of active threads in the warp.
+  llvm::Value *Mask = CGF.EmitRuntimeCall(
+      createNVPTXRuntimeFunction(OMPRTL_NVPTX__kmpc_warp_active_thread_mask));
   // Fetch team-local id of the thread.
   llvm::Value *ThreadID = getNVPTXThreadID(CGF);
 
@@ -2740,8 +2761,9 @@ void CGOpenMPRuntimeNVPTX::emitCriticalRegion(
   // Block waits for all threads in current team to finish then increments the
   // counter variable and returns to the loop.
   CGF.EmitBlock(SyncBB);
-  emitBarrierCall(CGF, Loc, OMPD_unknown, /*EmitChecks=*/false,
-                  /*ForceSimpleCall=*/true);
+  // Reconverge active threads in the warp.
+  (void)CGF.EmitRuntimeCall(
+      createNVPTXRuntimeFunction(OMPRTL_NVPTX__kmpc_syncwarp), Mask);
 
   llvm::Value *IncCounterVal =
       CGF.Builder.CreateNSWAdd(CounterVal, CGF.Builder.getInt32(1));