Subzero: Loop Invariant Code Motion

src/IceCfg.h

 //===- subzero/src/IceCfg.h - Control flow graph ----------------*- C++ -*-===//
 //
 //                        The Subzero Code Generator
 //
 // This file is distributed under the University of Illinois Open Source
 // License. See LICENSE.TXT for details.
 //
 //===----------------------------------------------------------------------===//
 ///
 /// \file
@@ skipping 184 matching lines @@
   /// instructions from the Cfg, to speed up later passes over the instructions.
   void renumberInstructions();
   void placePhiLoads();
   void placePhiStores();
   void deletePhis();
   void advancedPhiLowering();
   void reorderNodes();
   void shuffleNodes();
   void localCSE();
   void shortCircuitJumps();
+  void loopInvariantCodeMotion();
 
   /// Scan allocas to determine whether we need to use a frame pointer.
   /// If SortAndCombine == true, merge all the fixed-size allocas in the
   /// entry block and emit stack or frame pointer-relative addressing.
   void processAllocas(bool SortAndCombine);
   void doAddressOpt();
   /// Find clusters of insertelement/extractelement instructions that can be
   /// replaced by a shufflevector instruction.
   void materializeVectorShuffles();
   void doArgLowering();
   void doNopInsertion();
   void genCode();
   void genFrame();
-  void computeLoopNestDepth();
+  void generateLoopInfo();
   void livenessLightweight();
   void liveness(LivenessMode Mode);
   bool validateLiveness() const;
   void contractEmptyNodes();
   void doBranchOpt();
   void markNodesForSandboxing();
 
   /// \name  Manage the CurrentNode field.
   /// CurrentNode is used for validating the Variable::DefNode field during
   /// dumping/emitting.
@@ skipping 64 matching lines @@
 
   enum AllocaBaseVariableType {
     BVT_StackPointer,
     BVT_FramePointer,
     BVT_UserPointer
   };
   void sortAndCombineAllocas(CfgVector<InstAlloca *> &Allocas,
                              uint32_t CombinedAlignment, InstList &Insts,
                              AllocaBaseVariableType BaseVariableType);
   void findRematerializable();
+  CfgVector<Inst *> findLoopInvariantInstructions(SizeT LoopHeaderIndex);
 
   GlobalContext *Ctx;
   uint32_t SequenceNumber; /// output order for emission
   OptLevel OptimizationLevel = Opt_m1;
   uint32_t ConstantBlindingCookie = 0; /// cookie for constant blinding
   VerboseMask VMask;
   GlobalString FunctionName;
   Type ReturnType = IceType_void;
   bool IsInternalLinkage = false;
   bool HasError = false;
@@ skipping 10 matching lines @@
   // of the uniqueness requirement, and assumptions in lit tests.
   std::unique_ptr<StringPool> NodeStrings;
   std::unique_ptr<StringPool> VarStrings;
   std::unique_ptr<Liveness> Live;
   std::unique_ptr<TargetLowering> Target;
   std::unique_ptr<VariablesMetadata> VMetadata;
   std::unique_ptr<Assembler> TargetAssembler;
   /// Globals required by this CFG. Mostly used for the profiler's globals.
   std::unique_ptr<VariableDeclarationList> GlobalInits;
   CfgVector<InstJumpTable *> JumpTables;
-
+  CfgUnorderedMap<SizeT, CfgVector<SizeT>> LoopInfo;
   /// CurrentNode is maintained during dumping/emitting just for validating
   /// Variable::DefNode. Normally, a traversal over CfgNodes maintains this, but
   /// before global operations like register allocation, resetCurrentNode()
   /// should be called to avoid spurious validation failures.
   const CfgNode *CurrentNode = nullptr;
 
 public:
   static void TlsInit() { CfgAllocatorTraits::init(); }
 };
 
@@ skipping 11 matching lines @@
   using OwnerType = Cfg;
   static StringPool *getStrings(const OwnerType *PoolOwner) {
     return PoolOwner->getVarStrings();
   }
 };
 using VariableString = StringID<VariableStringPoolTraits>;
 
 } // end of namespace Ice
 
 #endif // SUBZERO_SRC_ICECFG_H