Handle ARM "ret void" and function alignment with proper padding.

src/IceTargetLoweringARM32.h

 //===- subzero/src/IceTargetLoweringARM32.h - ARM32 lowering ----*- C++ -*-===//
 //
 //                        The Subzero Code Generator
 //
 // This file is distributed under the University of Illinois Open Source
 // License. See LICENSE.TXT for details.
 //
 //===----------------------------------------------------------------------===//
 //
 // This file declares the TargetLoweringARM32 class, which implements the
 // TargetLowering interface for the ARM 32-bit architecture.
 //
 //===----------------------------------------------------------------------===//
 
 #ifndef SUBZERO_SRC_ICETARGETLOWERINGARM32_H
 #define SUBZERO_SRC_ICETARGETLOWERINGARM32_H
 
 #include "IceDefs.h"
+#include "IceInstARM32.h"
 #include "IceRegistersARM32.h"
 #include "IceTargetLowering.h"
 
 namespace Ice {
 
 class TargetARM32 : public TargetLowering {
   TargetARM32() = delete;
   TargetARM32(const TargetARM32 &) = delete;
   TargetARM32 &operator=(const TargetARM32 &) = delete;
 
@@ skipping 55 matching lines @@
                            const AssignList &Assignments) override;
   void doAddressOptLoad() override;
   void doAddressOptStore() override;
   void randomlyInsertNop(float Probability) override;
   void makeRandomRegisterPermutation(
       llvm::SmallVectorImpl<int32_t> &Permutation,
       const llvm::SmallBitVector &ExcludeRegisters) const override;
 
   static Type stackSlotType();
 
+  // The following are helpers that insert lowered ARM32 instructions
+  // with minimal syntactic overhead, so that the lowering code can
+  // look as close to assembly as practical.
+
+  void _ret(Variable *LR, Variable *Src0 = nullptr) {
+    Context.insert(InstARM32Ret::create(Func, LR, Src0));
+  }
+
   bool UsesFramePointer;
   bool NeedsStackAlignment;
   llvm::SmallBitVector TypeToRegisterSet[IceType_NUM];
   llvm::SmallBitVector ScratchRegs;
   llvm::SmallBitVector RegsUsed;
   VarList PhysicalRegisters[IceType_NUM];
   static IceString RegNames[];
 
 private:
   ~TargetARM32() override {}
@@ skipping 17 matching lines @@
 
 private:
   void lowerGlobal(const VariableDeclaration &Var) const;
   ~TargetDataARM32() override {}
   template <typename T> static void emitConstantPool(GlobalContext *Ctx);
 };
 
 } // end of namespace Ice
 
 #endif // SUBZERO_SRC_ICETARGETLOWERINGARM32_H