add a pass that can extract all kinds of global values, not just functions. Update llvm-extract to use it and optionally extract a global variable if you want it too

llvm-svn: 48015

include/llvm/Transforms/IPO.h

@@ -24,6 +24,7 @@ class ModulePass;
 class Pass;
 class Function;
 class BasicBlock;
+class GlobalValue;
 
 //===----------------------------------------------------------------------===//
 //
@@ -85,6 +86,14 @@ ModulePass *createGlobalDCEPass();
 ModulePass *createFunctionExtractionPass(Function *F, bool deleteFn = false,
                                          bool relinkCallees = false);
 
+//===----------------------------------------------------------------------===//
+/// createGVExtractionPass - If deleteFn is true, this pass deletes as
+/// the specified global values. Otherwise, it deletes as much of the module as
+/// possible, except for the global values specified.
+///
+ModulePass *createGVExtractionPass(std::vector<GlobalValue*>& GVs, bool 
+                                   deleteFn = false, 
+                                   bool relinkCallees = false);
 
 //===----------------------------------------------------------------------===//
 /// createFunctionInliningPass - Return a new pass object that uses a heuristic

lib/Transforms/IPO/ExtractGV.cpp

@@ -0,0 +1,174 @@
+//===-- ExtractGV.cpp - Global Value extraction pass ----------------------===//
+//
+//                     The LLVM Compiler Infrastructure
+//
+// This file is distributed under the University of Illinois Open Source
+// License. See LICENSE.TXT for details.
+//
+//===----------------------------------------------------------------------===//
+//
+// This pass extracts global values
+//
+//===----------------------------------------------------------------------===//
+
+#include "llvm/Instructions.h"
+#include "llvm/Module.h"
+#include "llvm/Pass.h"
+#include "llvm/Constants.h"
+#include "llvm/Transforms/IPO.h"
+#include "llvm/Support/Compiler.h"
+using namespace llvm;
+
+namespace {
+  /// @brief A pass to extract specific functions and their dependencies.
+  class VISIBILITY_HIDDEN GVExtractorPass : public ModulePass {
+    std::vector<GlobalValue*> Named;
+    bool deleteStuff;
+    bool reLink;
+  public:
+    static char ID; // Pass identification, replacement for typeid
+
+    /// FunctionExtractorPass - If deleteFn is true, this pass deletes as the
+    /// specified function. Otherwise, it deletes as much of the module as
+    /// possible, except for the function specified.
+    ///
+    explicit GVExtractorPass(std::vector<GlobalValue*>& GVs, bool deleteS = true,
+                             bool relinkCallees = false)
+      : ModulePass((intptr_t)&ID), Named(GVs), deleteStuff(deleteS),
+        reLink(relinkCallees) {}
+
+    bool runOnModule(Module &M) {
+      if (Named.size() == 0) {
+        return false;  // Nothing to extract
+      }
+      
+      if (deleteStuff)
+        return deleteGV();
+      M.setModuleInlineAsm("");
+      return isolateGV(M);
+    }
+
+    bool deleteGV() {
+      for (std::vector<GlobalValue*>::iterator GI = Named.begin(), 
+             GE = Named.end(); GI != GE; ++GI) {
+        if (Function* NamedFunc = dyn_cast<Function>(&*GI)) {
+         // If we're in relinking mode, set linkage of all internal callees to
+         // external. This will allow us extract function, and then - link
+         // everything together
+         if (reLink) {
+           for (Function::iterator B = NamedFunc->begin(), BE = NamedFunc->end();
+                B != BE; ++B) {
+             for (BasicBlock::iterator I = B->begin(), E = B->end();
+                  I != E; ++I) {
+               if (CallInst* callInst = dyn_cast<CallInst>(&*I)) {
+                 Function* Callee = callInst->getCalledFunction();
+                 if (Callee && Callee->hasInternalLinkage())
+                   Callee->setLinkage(GlobalValue::ExternalLinkage);
+               }
+             }
+           }
+         }
+         
+         NamedFunc->setLinkage(GlobalValue::ExternalLinkage);
+         NamedFunc->deleteBody();
+         assert(NamedFunc->isDeclaration() && "This didn't make the function external!");
+       } else {
+          if (!(*GI)->isDeclaration()) {
+            cast<GlobalVariable>(*GI)->setInitializer(0);  //clear the initializer
+            (*GI)->setLinkage(GlobalValue::ExternalLinkage);
+          }
+        }
+      }
+      return true;
+    }
+
+    bool isolateGV(Module &M) {
+      // Mark all globals internal
+      for (Module::global_iterator I = M.global_begin(), E = M.global_end(); I != E; ++I)
+        if (!I->isDeclaration()) {
+          I->setLinkage(GlobalValue::InternalLinkage);
+        }
+      for (Module::iterator I = M.begin(), E = M.end(); I != E; ++I)
+        if (!I->isDeclaration()) {
+          I->setLinkage(GlobalValue::InternalLinkage);
+        }
+
+      // Make sure our result is globally accessible...
+      // by putting them in the used array
+      {
+        std::vector<Constant *> AUGs;
+        const Type *SBP= PointerType::getUnqual(Type::Int8Ty);
+        for (std::vector<GlobalValue*>::iterator GI = Named.begin(), 
+               GE = Named.end(); GI != GE; ++GI) {
+          (*GI)->setLinkage(GlobalValue::ExternalLinkage);
+          AUGs.push_back(ConstantExpr::getBitCast(*GI, SBP));
+        }
+        ArrayType *AT = ArrayType::get(SBP, AUGs.size());
+        Constant *Init = ConstantArray::get(AT, AUGs);
+        GlobalValue *gv = new GlobalVariable(AT, false, 
+                                             GlobalValue::AppendingLinkage, 
+                                             Init, "llvm.used", &M);
+        gv->setSection("llvm.metadata");
+      }
+
+      // All of the functions may be used by global variables or the named
+      // globals.  Loop through them and create a new, external functions that
+      // can be "used", instead of ones with bodies.
+      std::vector<Function*> NewFunctions;
+
+      Function *Last = --M.end();  // Figure out where the last real fn is.
+
+      for (Module::iterator I = M.begin(); ; ++I) {
+        if (std::find(Named.begin(), Named.end(), &*I) == Named.end()) {
+          Function *New = new Function(I->getFunctionType(),
+                                       GlobalValue::ExternalLinkage);
+          New->setCallingConv(I->getCallingConv());
+          New->setParamAttrs(I->getParamAttrs());
+          if (I->hasCollector())
+            New->setCollector(I->getCollector());
+
+          // If it's not the named function, delete the body of the function
+          I->dropAllReferences();
+
+          M.getFunctionList().push_back(New);
+          NewFunctions.push_back(New);
+          New->takeName(I);
+        }
+
+        if (&*I == Last) break;  // Stop after processing the last function
+      }
+
+      // Now that we have replacements all set up, loop through the module,
+      // deleting the old functions, replacing them with the newly created
+      // functions.
+      if (!NewFunctions.empty()) {
+        unsigned FuncNum = 0;
+        Module::iterator I = M.begin();
+        do {
+          if (std::find(Named.begin(), Named.end(), &*I) == Named.end()) {
+            // Make everything that uses the old function use the new dummy fn
+            I->replaceAllUsesWith(NewFunctions[FuncNum++]);
+
+            Function *Old = I;
+            ++I;  // Move the iterator to the new function
+
+            // Delete the old function!
+            M.getFunctionList().erase(Old);
+
+          } else {
+            ++I;  // Skip the function we are extracting
+          }
+        } while (&*I != NewFunctions[0]);
+      }
+
+      return true;
+    }
+  };
+
+  char GVExtractorPass::ID = 0;
+}
+
+ModulePass *llvm::createGVExtractionPass(std::vector<GlobalValue*>& GVs, 
+                                         bool deleteFn, bool relinkCallees) {
+  return new GVExtractorPass(GVs, deleteFn, relinkCallees);
+}

tools/llvm-extract/llvm-extract.cpp

@@ -39,17 +39,22 @@ static cl::opt<bool>
 Force("f", cl::desc("Overwrite output files"));
 
 static cl::opt<bool>
-DeleteFn("delete", cl::desc("Delete specified function from Module"));
+DeleteFn("delete", cl::desc("Delete specified Globals from Module"));
 
 static cl::opt<bool>
 Relink("relink",
        cl::desc("Turn external linkage for callees of function to delete"));
 
-// ExtractFunc - The function to extract from the module... defaults to main.
+// ExtractFunc - The function to extract from the module... 
 static cl::opt<std::string>
-ExtractFunc("func", cl::desc("Specify function to extract"), cl::init("main"),
+ExtractFunc("func", cl::desc("Specify function to extract"), cl::init(""),
             cl::value_desc("function"));
 
+// ExtractGlobal - The global to extract from the module...
+static cl::opt<std::string>
+ExtractGlobal("glob", cl::desc("Specify global to extract"), cl::init(""),
+              cl::value_desc("global"));
+
 int main(int argc, char **argv) {
   llvm_shutdown_obj X;  // Call llvm_shutdown() on exit.
   cl::ParseCommandLineOptions(argc, argv, "llvm extractor\n");
@@ -72,10 +77,17 @@ int main(int argc, char **argv) {
   }
 
   // Figure out which function we should extract
-  Function *F = M.get()->getFunction(ExtractFunc);
-  if (F == 0) {
+  GlobalVariable *G = ExtractGlobal.size() ? 
+    M.get()->getNamedGlobal(ExtractGlobal) : 0;
+
+  // Figure out which function we should extract
+  Function *F = ExtractFunc.size() ? 
+    M.get()->getFunction(ExtractFunc) :
+    M.get()->getFunction("main");
+
+  if (F == 0 && G == 0) {
     cerr << argv[0] << ": program doesn't contain function named '"
-         << ExtractFunc << "'!\n";
+         << ExtractFunc << "' or a global named '" << ExtractGlobal << "'!\n";
     return 1;
   }
 
@@ -84,7 +96,11 @@ int main(int argc, char **argv) {
   PassManager Passes;
   Passes.add(new TargetData(M.get())); // Use correct TargetData
   // Either isolate the function or delete it from the Module
-  Passes.add(createFunctionExtractionPass(F, DeleteFn, Relink));
+  std::vector<GlobalValue*> GVs;
+  if (F) GVs.push_back(F);
+  if (G) GVs.push_back(G);
+
+  Passes.add(createGVExtractionPass(GVs, DeleteFn, Relink));
   if (!DeleteFn)
     Passes.add(createGlobalDCEPass());           // Delete unreachable globals
   Passes.add(createDeadTypeEliminationPass());   // Remove dead types...