Introduce model of global initializers in Subzero.

src/IceConverter.cpp

Index: src/IceConverter.cpp
diff --git a/src/IceConverter.cpp b/src/IceConverter.cpp
index 9d59c56980f072b8215de1b0acda4dfcfafce45e..ed9f13aabd87e7ceea8c2ebcf05bc99bd42117e1 100644
--- a/src/IceConverter.cpp
+++ b/src/IceConverter.cpp
@@ -27,6 +27,7 @@
 #include "IceConverter.h"
 #include "IceDefs.h"
 #include "IceGlobalContext.h"
+#include "IceGlobalInits.h"
 #include "IceInst.h"
 #include "IceOperand.h"
 #include "IceTargetLowering.h"
@@ -45,17 +46,29 @@ template <typename T> static std::string LLVMObjectAsString(const T *O) {
   return Stream.str();
 }
 
-// Converter from LLVM to ICE. The entry point is the convertFunction method.
+// Base class for converting LLVM to ICE.
+class LLVM2ICEConverter {
+public:
+  LLVM2ICEConverter(Ice::GlobalContext *Ctx, LLVMContext &LLVMContext)
+      : Ctx(Ctx), TypeConverter(LLVMContext) {}
+
+protected:
+  // Data
+  Ice::GlobalContext *Ctx;
+  Ice::TypeConverter TypeConverter;

[jvoung (off chromium), 2014/10/03 16:15:29]

Could disallow copy / assign.

[Karl, 2014/10/04 16:28:33]

Done.

+};
+
+// Converter from LLVM functions to ICE. The entry point is the
+// convertFunction method.
 //
-// Note: this currently assumes that the given IR was verified to be valid PNaCl
-// bitcode:
+// Note: this currently assumes that the given IR was verified to be
+// valid PNaCl bitcode:
 // https://developers.google.com/native-client/dev/reference/pnacl-bitcode-abi
 // If not, all kinds of assertions may fire.
-//
-class LLVM2ICEConverter {
+class LLVM2ICEFunctionConverter : LLVM2ICEConverter {
 public:
-  LLVM2ICEConverter(Ice::GlobalContext *Ctx, LLVMContext &LLVMContext)
-      : Ctx(Ctx), Func(NULL), TypeConverter(LLVMContext) {}
+  LLVM2ICEFunctionConverter(Ice::GlobalContext *Ctx, LLVMContext &LLVMContext)
+      : LLVM2ICEConverter(Ctx, LLVMContext), Func(NULL) {}
 
   // Caller is expected to delete the returned Ice::Cfg object.
   Ice::Cfg *convertFunction(const Function *F) {
@@ -611,13 +624,185 @@ private:
 
 private:
   // Data
-  Ice::GlobalContext *Ctx;
   Ice::Cfg *Func;
   std::map<const Value *, Ice::Variable *> VarMap;
   std::map<const BasicBlock *, Ice::CfgNode *> NodeMap;
-  Ice::TypeConverter TypeConverter;
 };
 
+// Converter from LLVM global variables to ICE. The entry point is the
+// convertGlobalsToIce method.
+//
+// Note: this currently assumes that the given IR was verified to be
+// valid PNaCl bitcode:
+// https://developers.google.com/native-client/dev/reference/pnacl-bitcode-abi
+// If not, all kinds of assertions may fire.
+class LLVM2ICEGlobalsConverter : public LLVM2ICEConverter {
+public:
+  LLVM2ICEGlobalsConverter(Ice::GlobalContext *Ctx, LLVMContext &LLVMContext)
+      : LLVM2ICEConverter(Ctx, LLVMContext) {}
+
+  ~LLVM2ICEGlobalsConverter() {
+    llvm::DeleteContainerSeconds(GlobalVarAddressMap);
+  }
+
+  /// Converts global variables, and their initializers into ICE global
+  /// addresses, for module Mod. Puts corresponding converted global
+  /// addresses into GlobalAddresses.
+  void convertGlobalsToIce(llvm::Module *Mod,
+                           std::vector<Ice::GlobalAddress *> &GlobalAddresses);

[Jim Stichnoth, 2014/10/03 01:28:08]

Can you make a typedef for std::vector<Ice::GlobalAddress *> ?
Maybe std::map<const llvm::GlobalVariable *, Ice::GlobalAddress *> as well.

[Karl, 2014/10/04 16:28:33]

Added typedef Translator::GlobalAddressList.

Added typedef GlobalVarAddressMapType for GlobalVarAddressMap.

+
+private:
+  // Map from global variables to thier corresponding global address.

[Jim Stichnoth, 2014/10/03 01:28:08]

their

[Karl, 2014/10/04 16:28:33]

Done.

+  std::map<const llvm::GlobalVariable *, Ice::GlobalAddress *>
+      GlobalVarAddressMap;
+
+  // Adds the Initializer to the list of initializers for Global address.
+  void AddGlobalInitializer(Ice::GlobalAddress &Global,

[jvoung (off chromium), 2014/10/03 16:15:29]

method name consistency: "addGlobalInitializer" instead?

[Karl, 2014/10/04 16:28:33]

Done.

+                            const llvm::Constant *Initializer) {
+    AddGlobalInitializer(Global, Initializer, false, 0);

[jvoung (off chromium), 2014/10/03 16:15:29]

nit: For some of the other code in Subzero, there is a style of documenting
constants w/ const local vars:

const bool HasOffset = false;
const ... Offset = 0;
foo(..., ..., HasOffset, Offset);

[Karl, 2014/10/04 16:28:33]

Done.

+  }
+
+  // Adds Initializer to the list of initializers for Global
+  // address. HasOffset is true only if Initializer is a relocation
+  // initializer and Offset should be added to the relocation.
+  void AddGlobalInitializer(Ice::GlobalAddress &Global,
+                            const llvm::Constant *Initializer, bool HasOffset,
+                            Ice::GlobalAddress::RelocOffsetType Offset);
+
+  // Returns the global address associated with global variable GV.
+  Ice::GlobalAddress *getGlobalVarAddress(const llvm::GlobalVariable *GV) {
+    Ice::GlobalAddress *Addr = GlobalVarAddressMap[GV];

[Jim Stichnoth, 2014/10/03 01:28:08]

Maybe better form to do something like this?

  if (GlobalVarAddressMap.find(GV) == GlobalVarAddressMap.end())
    GlobalVarAddressMap[GV] = new Ice::GlobalAddress();
  return GlobalVarAddressMap[GV];

The point being to only use the [] operator on the RHS if you know it has a
valid entry.

[Karl, 2014/10/04 16:28:33]

Done.

+    if (Addr)
+      return Addr;
+    Addr = new Ice::GlobalAddress();
+    GlobalVarAddressMap[GV] = Addr;
+    return Addr;
+  }
+
+  // Converts the given constant C to the corresponding integer
+  // literal it contains.
+  Ice::GlobalAddress::RelocOffsetType
+  getIntegerLiteralConstant(const llvm::Value *C) {
+    const llvm::ConstantInt *CI = llvm::dyn_cast<llvm::ConstantInt>(C);

[Jim Stichnoth, 2014/10/03 01:28:08]

if (auto CI = llvm::dyn_cast<const llvm::ConstantInt>(C))
  if (CI->getType()->isIntegerTy(32))
    return CI->getSExtValue();

And BTW, it seems like cast<> and dyn_cast<> are good opportunities for auto.

[Karl, 2014/10/04 16:28:33]

Fixed all occurrences in file.

+    if (CI && CI->getType()->isIntegerTy(32))
+      return CI->getSExtValue();
+
+    std::string Buffer;
+    raw_string_ostream StrBuf(Buffer);
+    StrBuf << "Constant not i32 literal: " << *C;
+    report_fatal_error(StrBuf.str());
+    return 0;
+  }
+};
+
+void LLVM2ICEGlobalsConverter::convertGlobalsToIce(
+    llvm::Module *Mod, std::vector<Ice::GlobalAddress *> &GlobalAddresses) {
+  for (llvm::Module::const_global_iterator I = Mod->global_begin(),
+                                           E = Mod->global_end();
+       I != E; ++I) {
+    if (!I->hasInitializer())
+      continue;

[jvoung (off chromium), 2014/10/03 16:15:29]

For the PNaCl ABI, should we be stricter? I thought "Every global variable must
have an initializer. Each initializer must be either a SimpleElement or a
CompoundElement, defined as follows..."

[Karl, 2014/10/04 16:28:33]

I agree. I was being lazy because there is a test for an external global. Fixing
test instead.

Note: For the cross test tool, we need to allow uninitialized globals. Fixing by
adding command line flag for cross tests.

+
+    const llvm::GlobalVariable *GV = llvm::dyn_cast<GlobalVariable>(I);
+    assert(GV);

[Jim Stichnoth, 2014/10/03 01:28:08]

If you're going to assert(GV), might as well use llvm::cast<>() above.

[Karl, 2014/10/04 16:28:33]

Not sure what you are asking. Are you asking to put the cast in the assert? I
moved it out because the cased value is used below as well (and did not want to
repeat usage).

+    Ice::IceString Name = GV->getName();
+    if (!GV->hasInternalLinkage()) {
+      std::string Buffer;
+      raw_string_ostream StrBuf(Buffer);
+      StrBuf << "Can't define external global address: " << Name;
+      report_fatal_error(StrBuf.str());
+    }
+
+    Ice::GlobalAddress *Addr = getGlobalVarAddress(GV);
+    GlobalAddresses.push_back(Addr);
+    Addr->setAlignment(GV->getAlignment());
+    Addr->setIsConstant(GV->isConstant());
+    Addr->setName(Name);
+
+    const llvm::Constant *Initializer = GV->getInitializer();
+    if (const llvm::ConstantStruct *CompoundInit =
+            dyn_cast<llvm::ConstantStruct>(Initializer)) {
+      for (llvm::ConstantStruct::const_op_iterator i = CompoundInit->op_begin(),

[Jim Stichnoth, 2014/10/03 01:28:08]

Use I and E instead of i and e for consistency?

[Karl, 2014/10/04 16:28:33]

Done.

+                                                   e = CompoundInit->op_end();
+           i != e; ++i) {
+        if (const llvm::Constant *Init = dyn_cast<llvm::Constant>(i)) {
+          AddGlobalInitializer(*Addr, Init);
+        }

[jvoung (off chromium), 2014/10/03 16:15:29]

What are the possible "else" here, and should it be an error? Similar for the
ConstantExpr checks below.

[Karl, 2014/10/04 16:28:33]

The format in PNaCl ABI is that the struct is part of the ConstantExpr only if
it has multiple simple entries. If it has a single entry, it is just a simple
entry.

+      }
+    } else {
+      AddGlobalInitializer(*Addr, Initializer);
+    }
+  }
+}
+
+void LLVM2ICEGlobalsConverter::AddGlobalInitializer(
+    Ice::GlobalAddress &Global, const llvm::Constant *Initializer,
+    bool HasOffset, Ice::GlobalAddress::RelocOffsetType Offset) {
+  assert(HasOffset || Offset == 0);
+
+  if (const llvm::ConstantDataArray *CDA =
+          llvm::dyn_cast<llvm::ConstantDataArray>(Initializer)) {
+    assert(!HasOffset && llvm::isa<llvm::IntegerType>(CDA->getElementType()) &&
+           (llvm::cast<llvm::IntegerType>(CDA->getElementType())
+                ->getBitWidth() == 8));
+    Global.AddInitializer(new Ice::GlobalAddress::DataInitializer(
+        CDA->getRawDataValues().data(), CDA->getNumElements()));
+    return;
+  }
+
+  if (llvm::isa<llvm::ConstantAggregateZero>(Initializer)) {
+    if (const llvm::ArrayType *AT =
+            llvm::dyn_cast<llvm::ArrayType>(Initializer->getType())) {
+      assert(!HasOffset && llvm::isa<llvm::IntegerType>(AT->getElementType()) &&
+             (llvm::cast<llvm::IntegerType>(AT->getElementType())
+                  ->getBitWidth() == 8));
+      Global.AddInitializer(
+          new Ice::GlobalAddress::ZeroInitializer(AT->getNumElements()));
+    } else {
+      llvm_unreachable("Unhandled constant aggregate zero type");
+    }
+    return;
+  }
+
+  if (const ConstantExpr *Exp =
+          llvm::dyn_cast<llvm::ConstantExpr>(Initializer)) {
+    switch (Exp->getOpcode()) {
+    case Instruction::Add:
+      assert(!HasOffset);
+      AddGlobalInitializer(Global, Exp->getOperand(0), true,
+                           getIntegerLiteralConstant(Exp->getOperand(1)));
+      return;
+    case Instruction::PtrToInt: {
+      assert(TypeConverter.convertToIceType(Exp->getType()) ==
+             TypeConverter.getIcePointerType());
+      llvm::GlobalValue *GV =
+          llvm::dyn_cast<llvm::GlobalValue>(Exp->getOperand(0));
+      assert(GV);
+      if (llvm::Function *Fcn = llvm::dyn_cast<Function>(GV)) {
+        Ice::GlobalAddress::RelocationAddress Addr(Fcn);
+        Global.AddInitializer(
+            new Ice::GlobalAddress::RelocInitializer(Addr, Offset));
+        return;
+      } else if (llvm::GlobalVariable *Var =
+                     llvm::dyn_cast<llvm::GlobalVariable>(GV)) {
+        Ice::GlobalAddress::RelocationAddress Addr(getGlobalVarAddress(Var));
+        Global.AddInitializer(
+            new Ice::GlobalAddress::RelocInitializer(Addr, Offset));
+        return;
+      }
+      break;
+    }
+    default:
+      break;
+    }
+  }
+
+  std::string Buffer;
+  raw_string_ostream StrBuf(Buffer);
+  StrBuf << "Unhandled global initializer: " << Initializer;
+  report_fatal_error(StrBuf.str());
+}
+
 } // end of anonymous namespace
 
 namespace Ice {
@@ -626,16 +811,24 @@ void Converter::convertToIce() {
   static TimerIdT IDconvertToIce = GlobalContext::getTimerID("convertToIce");
   TimerMarker T(IDconvertToIce, Ctx);
   nameUnnamedGlobalAddresses(Mod);
+  nameUnnamedFunctions(Mod);
   if (!Ctx->getFlags().DisableGlobals)
     convertGlobals(Mod);
   convertFunctions();
 }
 
+void Converter::convertGlobals(llvm::Module *Mod) {
+  LLVM2ICEGlobalsConverter GlobalsConverter(Ctx, Mod->getContext());
+  std::vector<Ice::GlobalAddress *> GlobalAddresses;
+  GlobalsConverter.convertGlobalsToIce(Mod, GlobalAddresses);
+  lowerIceGlobals(GlobalAddresses);
+}
+
 void Converter::convertFunctions() {
   for (Module::const_iterator I = Mod->begin(), E = Mod->end(); I != E; ++I) {
     if (I->empty())
       continue;
-    LLVM2ICEConverter FunctionConverter(Ctx, Mod->getContext());
+    LLVM2ICEFunctionConverter FunctionConverter(Ctx, Mod->getContext());
 
     Timer TConvert;
     Cfg *Fcn = FunctionConverter.convertFunction(I);