[OPENMP][NVPTX]Emit default locations as constant with undefined mode.

For the NVPTX target default locations should be emitted as constants +
additional info must be emitted in the reserved_2 field of the ident_t
structure. The 1st bit controls the execution mode and the 2nd bit
controls use of the lightweight runtime. The combination of the bits for
Non-SPMD mode + lightweight runtime represents special undefined mode,
used outside of the target regions for orphaned directives or functions.
Should allow and additional optimization inside of the target regions.

llvm-svn: 347425

4 files changed, 50 insertions, 9 deletions

diff --git a/clang/lib/CodeGen/CGOpenMPRuntime.cpp b/clang/lib/CodeGen/CGOpenMPRuntime.cpp
index cf2c6cf6f22..cade093941a 100644
--- a/clang/lib/CodeGen/CGOpenMPRuntime.cpp
+++ b/clang/lib/CodeGen/CGOpenMPRuntime.cpp
@@ -1467,7 +1467,9 @@ createConstantGlobalStructAndAddToParent(CodeGenModule &CGM, QualType Ty,
 
 Address CGOpenMPRuntime::getOrCreateDefaultLocation(unsigned Flags) {
   CharUnits Align = CGM.getContext().getTypeAlignInChars(IdentQTy);
-  llvm::Value *Entry = OpenMPDefaultLocMap.lookup(Flags);
+  unsigned Reserved2Flags = getDefaultLocationReserved2Flags();
+  FlagsTy FlagsKey(Flags, Reserved2Flags);
+  llvm::Value *Entry = OpenMPDefaultLocMap.lookup(FlagsKey);
   if (!Entry) {
     if (!DefaultOpenMPPSource) {
       // Initialize default location for psource field of ident_t structure of
@@ -1480,18 +1482,18 @@ Address CGOpenMPRuntime::getOrCreateDefaultLocation(unsigned Flags) {
           llvm::ConstantExpr::getBitCast(DefaultOpenMPPSource, CGM.Int8PtrTy);
     }
 
-    llvm::Constant *Data[] = {llvm::ConstantInt::getNullValue(CGM.Int32Ty),
-                              llvm::ConstantInt::get(CGM.Int32Ty, Flags),
-                              llvm::ConstantInt::getNullValue(CGM.Int32Ty),
-                              llvm::ConstantInt::getNullValue(CGM.Int32Ty),
-                              DefaultOpenMPPSource};
+    llvm::Constant *Data[] = {
+        llvm::ConstantInt::getNullValue(CGM.Int32Ty),
+        llvm::ConstantInt::get(CGM.Int32Ty, Flags),
+        llvm::ConstantInt::get(CGM.Int32Ty, Reserved2Flags),
+        llvm::ConstantInt::getNullValue(CGM.Int32Ty), DefaultOpenMPPSource};
     llvm::GlobalValue *DefaultOpenMPLocation =
-        createGlobalStruct(CGM, IdentQTy, /*IsConstant=*/false, Data, "",
+        createGlobalStruct(CGM, IdentQTy, isDefaultLocationConstant(), Data, "",
                            llvm::GlobalValue::PrivateLinkage);
     DefaultOpenMPLocation->setUnnamedAddr(
         llvm::GlobalValue::UnnamedAddr::Global);
 
-    OpenMPDefaultLocMap[Flags] = Entry = DefaultOpenMPLocation;
+    OpenMPDefaultLocMap[FlagsKey] = Entry = DefaultOpenMPLocation;
   }
   return Address(Entry, Align);
 }
diff --git a/clang/lib/CodeGen/CGOpenMPRuntime.h b/clang/lib/CodeGen/CGOpenMPRuntime.h
index 1a27db15746..489ecdf15e1 100644
--- a/clang/lib/CodeGen/CGOpenMPRuntime.h
+++ b/clang/lib/CodeGen/CGOpenMPRuntime.h
@@ -282,12 +282,21 @@ protected:
                               bool AtCurrentPoint = false);
   void clearLocThreadIdInsertPt(CodeGenFunction &CGF);
 
+  /// Check if the default location must be constant.
+  /// Default is false to support OMPT/OMPD.
+  virtual bool isDefaultLocationConstant() const { return false; }
+
+  /// Returns additional flags that can be stored in reserved_2 field of the
+  /// default location.
+  virtual unsigned getDefaultLocationReserved2Flags() const { return 0; }
+
 private:
   /// Default const ident_t object used for initialization of all other
   /// ident_t objects.
   llvm::Constant *DefaultOpenMPPSource = nullptr;
+  using FlagsTy = std::pair<unsigned, unsigned>;
   /// Map of flags and corresponding default locations.
-  typedef llvm::DenseMap<unsigned, llvm::Value *> OpenMPDefaultLocMapTy;
+  using OpenMPDefaultLocMapTy = llvm::DenseMap<FlagsTy, llvm::Value *>;
   OpenMPDefaultLocMapTy OpenMPDefaultLocMap;
   Address getOrCreateDefaultLocation(unsigned Flags);
 
diff --git a/clang/lib/CodeGen/CGOpenMPRuntimeNVPTX.cpp b/clang/lib/CodeGen/CGOpenMPRuntimeNVPTX.cpp
index 74b24eacce4..60fbf668114 100644
--- a/clang/lib/CodeGen/CGOpenMPRuntimeNVPTX.cpp
+++ b/clang/lib/CodeGen/CGOpenMPRuntimeNVPTX.cpp
@@ -1902,6 +1902,26 @@ void CGOpenMPRuntimeNVPTX::emitTargetOutlinedFunction(
   setPropertyExecutionMode(CGM, OutlinedFn->getName(), Mode);
 }
 
+namespace {
+LLVM_ENABLE_BITMASK_ENUMS_IN_NAMESPACE();
+/// Enum for accesseing the reserved_2 field of the ident_t struct.
+enum ModeFlagsTy : unsigned {
+  /// Bit set to 1 when in SPMD mode.
+  KMP_IDENT_SPMD_MODE = 0x01,
+  /// Bit set to 1 when a simplified runtime is used.
+  KMP_IDENT_SIMPLE_RT_MODE = 0x02,
+  LLVM_MARK_AS_BITMASK_ENUM(/*LargestValue=*/KMP_IDENT_SIMPLE_RT_MODE)
+};
+
+/// Special mode Undefined. Is the combination of Non-SPMD mode + SimpleRuntime.
+static const ModeFlagsTy UndefinedMode =
+    (~KMP_IDENT_SPMD_MODE) & KMP_IDENT_SIMPLE_RT_MODE;
+} // anonymous namespace
+
+unsigned CGOpenMPRuntimeNVPTX::getDefaultLocationReserved2Flags() const {
+  return UndefinedMode;
+}
+
 CGOpenMPRuntimeNVPTX::CGOpenMPRuntimeNVPTX(CodeGenModule &CGM)
     : CGOpenMPRuntime(CGM, "_", "$") {
   if (!CGM.getLangOpts().OpenMPIsDevice)
diff --git a/clang/lib/CodeGen/CGOpenMPRuntimeNVPTX.h b/clang/lib/CodeGen/CGOpenMPRuntimeNVPTX.h
index 2acab073580..b03ff782d75 100644
--- a/clang/lib/CodeGen/CGOpenMPRuntimeNVPTX.h
+++ b/clang/lib/CodeGen/CGOpenMPRuntimeNVPTX.h
@@ -180,6 +180,16 @@ protected:
     return "__omp_outlined__";
   }
 
+  /// Check if the default location must be constant.
+  /// Constant for NVPTX for better optimization.
+  bool isDefaultLocationConstant() const override { return true; }
+
+  /// Returns additional flags that can be stored in reserved_2 field of the
+  /// default location.
+  /// For NVPTX target contains data about SPMD/Non-SPMD execution mode +
+  /// Full/Lightweight runtime mode. Used for better optimization.
+  unsigned getDefaultLocationReserved2Flags() const override;
+
 public:
   explicit CGOpenMPRuntimeNVPTX(CodeGenModule &CGM);
   void clear() override;