[analyzer] Add index out of bounds check for CFArrayGetArrayAtIndex.

llvm-svn: 149228

7 files changed, 216 insertions, 7 deletions

diff --git a/clang/include/clang/StaticAnalyzer/Core/PathSensitive/ProgramState.h b/clang/include/clang/StaticAnalyzer/Core/PathSensitive/ProgramState.h
index 8191ed13120..a3e1830be3e 100644
--- a/clang/include/clang/StaticAnalyzer/Core/PathSensitive/ProgramState.h
+++ b/clang/include/clang/StaticAnalyzer/Core/PathSensitive/ProgramState.h
@@ -178,7 +178,8 @@ public:
 
   ProgramStateRef assumeInBound(DefinedOrUnknownSVal idx,
                                DefinedOrUnknownSVal upperBound,
-                               bool assumption) const;
+                               bool assumption,
+                               QualType IndexType = QualType()) const;
 
   /// Utility method for getting regions.
   const VarRegion* getRegion(const VarDecl *D, const LocationContext *LC) const;
diff --git a/clang/include/clang/StaticAnalyzer/Core/PathSensitive/ProgramStateTrait.h b/clang/include/clang/StaticAnalyzer/Core/PathSensitive/ProgramStateTrait.h
index b80d494b88c..e79b5935849 100644
--- a/clang/include/clang/StaticAnalyzer/Core/PathSensitive/ProgramStateTrait.h
+++ b/clang/include/clang/StaticAnalyzer/Core/PathSensitive/ProgramStateTrait.h
@@ -9,8 +9,8 @@
 //
 //  This file defines partial implementations of template specializations of
 //  the class ProgramStateTrait<>.  ProgramStateTrait<> is used by ProgramState 
-//  to implement set/get methods for mapulating a ProgramState's
-// generic data map.
+//  to implement set/get methods for manipulating a ProgramState's
+//  generic data map.
 //
 //===----------------------------------------------------------------------===//
 
diff --git a/clang/lib/StaticAnalyzer/Checkers/CMakeLists.txt b/clang/lib/StaticAnalyzer/Checkers/CMakeLists.txt
index 5c9b6bf1e37..9d628d40bb7 100644
--- a/clang/lib/StaticAnalyzer/Checkers/CMakeLists.txt
+++ b/clang/lib/StaticAnalyzer/Checkers/CMakeLists.txt
@@ -44,6 +44,7 @@ add_clang_library(clangStaticAnalyzerCheckers
   OSAtomicChecker.cpp
   ObjCAtSyncChecker.cpp
   ObjCContainersASTChecker.cpp
+  ObjCContainersChecker.cpp
   ObjCSelfInitChecker.cpp
   ObjCUnusedIVarsChecker.cpp
   PointerArithChecker.cpp
diff --git a/clang/lib/StaticAnalyzer/Checkers/Checkers.td b/clang/lib/StaticAnalyzer/Checkers/Checkers.td
index 16fc53a2416..f7c2a5735f0 100644
--- a/clang/lib/StaticAnalyzer/Checkers/Checkers.td
+++ b/clang/lib/StaticAnalyzer/Checkers/Checkers.td
@@ -372,6 +372,10 @@ def ObjCContainersASTChecker : Checker<"ContainerAPI">,
   HelpText<"Check for common pitfalls when using 'CFArray', 'CFDictionary', 'CFSet' APIs">,
   DescFile<"ObjCContainersASTChecker.cpp">;
 
+def ObjCContainersChecker : Checker<"Containers">,
+  HelpText<"Deep checks for common pitfalls when using 'CFArray' APIs">,
+  DescFile<"ObjCContainersChecker.cpp">;
+
 def ObjCSelfInitChecker : Checker<"SelfInit">,
   HelpText<"Check that 'self' is properly initialized inside an initializer method">,
   DescFile<"ObjCSelfInitChecker.cpp">;
diff --git a/clang/lib/StaticAnalyzer/Checkers/ObjCContainersChecker.cpp b/clang/lib/StaticAnalyzer/Checkers/ObjCContainersChecker.cpp
new file mode 100644
index 00000000000..8c8020be5cd
--- /dev/null
+++ b/clang/lib/StaticAnalyzer/Checkers/ObjCContainersChecker.cpp
@@ -0,0 +1,161 @@
+//== ObjCContainersChecker.cpp - Path sensitive checker for CFArray *- C++ -*=//
+//
+//                     The LLVM Compiler Infrastructure
+//
+// This file is distributed under the University of Illinois Open Source
+// License. See LICENSE.TXT for details.
+//
+//===----------------------------------------------------------------------===//
+//
+// Performs path sensitive checks of Core Foundation static containers like
+// CFArray.
+// 1) Check for buffer overflows:
+//      In CFArrayGetArrayAtIndex( myArray, index), if the index is outside the
+//      index space of theArray (0 to N-1 inclusive (where N is the count of
+//      theArray), the behavior is undefined.
+//
+//===----------------------------------------------------------------------===//
+
+#include "ClangSACheckers.h"
+#include "clang/StaticAnalyzer/Core/Checker.h"
+#include "clang/StaticAnalyzer/Core/CheckerManager.h"
+#include "clang/StaticAnalyzer/Core/PathSensitive/CheckerContext.h"
+#include "clang/StaticAnalyzer/Core/PathSensitive/ProgramStateTrait.h"
+#include "clang/StaticAnalyzer/Core/PathSensitive/ObjCMessage.h"
+#include "clang/StaticAnalyzer/Core/BugReporter/BugType.h"
+#include "clang/AST/ParentMap.h"
+
+using namespace clang;
+using namespace ento;
+
+namespace {
+class ObjCContainersChecker : public Checker< check::PreStmt<CallExpr>,
+                                             check::PostStmt<CallExpr> > {
+  mutable llvm::OwningPtr<BugType> BT;
+  inline void initBugType() const {
+    if (!BT)
+      BT.reset(new BugType("CFArray API", "Core Foundation/Objective-C API"));
+  }
+
+  inline SymbolRef getArraySym(const Expr *E, CheckerContext &C) const {
+    SVal ArrayRef = C.getState()->getSVal(E, C.getLocationContext());
+    SymbolRef ArraySym = ArrayRef.getAsSymbol();
+    assert(ArraySym);
+    return ArraySym;
+  }
+
+  void addSizeInfo(const Expr *Array, const Expr *Size,
+                   CheckerContext &C) const;
+
+public:
+  /// A tag to id this checker.
+  static void *getTag() { static int Tag; return &Tag; }
+
+  void checkPostStmt(const CallExpr *CE, CheckerContext &C) const;
+  void checkPreStmt(const CallExpr *CE, CheckerContext &C) const;
+};
+} // end anonymous namespace
+
+// Array state stores the array size on creation (allocation).
+// Size might be an unsigned or a symbol.
+struct ArraySize {
+  SVal NLSize;
+  ArraySize(SVal s) : NLSize(s) {}
+  void Profile(llvm::FoldingSetNodeID &ID) const {
+    NLSize.Profile(ID);
+  }
+};
+
+// ProgramState trait - a map from array symbol to it's state.
+typedef llvm::ImmutableMap<SymbolRef, SVal> ArraySizeM;
+
+namespace { struct ArraySizeMap {}; }
+namespace clang { namespace ento {
+template<> struct ProgramStateTrait<ArraySizeMap>
+    :  public ProgramStatePartialTrait<ArraySizeM > {
+  static void *GDMIndex() { return ObjCContainersChecker::getTag(); }
+};
+}}
+
+void ObjCContainersChecker::addSizeInfo(const Expr *Array, const Expr *Size,
+                                        CheckerContext &C) const {
+  const ProgramState *State = C.getState();
+  SVal SizeV = State->getSVal(Size, C.getLocationContext());
+  // Undefined is reported by another checker.
+  if (SizeV.isUnknownOrUndef())
+    return;
+
+  // Get the ArrayRef symbol.
+  SVal ArrayRef = State->getSVal(Array, C.getLocationContext());
+  SymbolRef ArraySym = ArrayRef.getAsSymbol();
+  if (!ArraySym)
+    return;
+
+  C.addTransition(State->set<ArraySizeMap>(ArraySym, SizeV));
+  return;
+}
+
+void ObjCContainersChecker::checkPostStmt(const CallExpr *CE,
+                                          CheckerContext &C) const {
+  StringRef Name = C.getCalleeName(CE);
+  if (Name.empty())
+    return;
+
+  // Add array size information to the state.
+  if (Name.equals("CFArrayCreate")) {
+    // Note, we can visit the Create method in the post-visit because
+    // the CFIndex parameter is passed in by value and will not be invalidated
+    // by the call.
+    addSizeInfo(CE, CE->getArg(2), C);
+    return;
+  }
+
+  if (Name.equals("CFArrayGetCount")) {
+    addSizeInfo(CE->getArg(0), CE, C);
+    return;
+  }
+}
+
+void ObjCContainersChecker::checkPreStmt(const CallExpr *CE,
+                                         CheckerContext &C) const {
+  StringRef Name = C.getCalleeName(CE);
+  if (Name.empty())
+    return;
+
+  // Check the array access.
+  if (Name.equals("CFArrayGetValueAtIndex")) {
+    const ProgramState *State = C.getState();
+    // Retrieve the size.
+    // Find out if we saw this array symbol before and have information about it.
+    const Expr *ArrayExpr = CE->getArg(0);
+    const SVal *SizeVal = State->get<ArraySizeMap>(getArraySym(ArrayExpr, C));
+    if (!SizeVal)
+      return;
+    DefinedOrUnknownSVal Size = cast<DefinedOrUnknownSVal>(*SizeVal);
+
+    // Get the index.
+    const Expr *IdxExpr = CE->getArg(1);
+    SVal IdxVal = State->getSVal(IdxExpr, C.getLocationContext());
+    DefinedOrUnknownSVal Idx = cast<DefinedOrUnknownSVal>(IdxVal);
+    
+    // Now, check if 'Idx in [0, Size-1]'.
+    const QualType T = IdxExpr->getType();
+    const ProgramState *StInBound = State->assumeInBound(Idx, Size, true, T);
+    const ProgramState *StOutBound = State->assumeInBound(Idx, Size, false, T);
+    if (StOutBound && !StInBound) {
+      ExplodedNode *N = C.generateSink(StOutBound);
+      if (!N)
+        return;
+      initBugType();
+      BugReport *R = new BugReport(*BT, "Index is out of bounds", N);
+      R->addRange(IdxExpr->getSourceRange());
+      C.EmitReport(R);
+      return;
+    }
+  }
+}
+
+/// Register checker.
+void ento::registerObjCContainersChecker(CheckerManager &mgr) {
+  mgr.registerChecker<ObjCContainersChecker>();
+}
diff --git a/clang/lib/StaticAnalyzer/Core/ProgramState.cpp b/clang/lib/StaticAnalyzer/Core/ProgramState.cpp
index 8c48a440738..b8b9ad94df3 100644
--- a/clang/lib/StaticAnalyzer/Core/ProgramState.cpp
+++ b/clang/lib/StaticAnalyzer/Core/ProgramState.cpp
@@ -274,7 +274,8 @@ ProgramState::bindExprAndLocation(const Stmt *S, const LocationContext *LCtx,
 
 ProgramStateRef ProgramState::assumeInBound(DefinedOrUnknownSVal Idx,
                                       DefinedOrUnknownSVal UpperBound,
-                                      bool Assumption) const {
+                                      bool Assumption,
+                                      QualType indexTy) const {
   if (Idx.isUnknown() || UpperBound.isUnknown())
     return this;
 
@@ -288,7 +289,8 @@ ProgramStateRef ProgramState::assumeInBound(DefinedOrUnknownSVal Idx,
   // Get the offset: the minimum value of the array index type.
   BasicValueFactory &BVF = svalBuilder.getBasicValueFactory();
   // FIXME: This should be using ValueManager::ArrayindexTy...somehow.
-  QualType indexTy = Ctx.IntTy;
+  if (indexTy.isNull())
+    indexTy = Ctx.IntTy;
   nonloc::ConcreteInt Min(BVF.getMinValue(indexTy));
 
   // Adjust the index.
diff --git a/clang/test/Analysis/CFContainers.mm b/clang/test/Analysis/CFContainers.mm
index ba2f386cf76..cead1d3ca98 100644
--- a/clang/test/Analysis/CFContainers.mm
+++ b/clang/test/Analysis/CFContainers.mm
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -analyze -analyzer-checker=experimental.osx.cocoa.ContainerAPI -analyzer-store=region -triple x86_64-apple-darwin -verify %s
+// RUN: %clang_cc1 -analyze -analyzer-checker=experimental.osx.cocoa.ContainerAPI,experimental.osx.cocoa.Containers -analyzer-store=region -triple x86_64-apple-darwin -verify %s
 
 typedef const struct __CFAllocator * CFAllocatorRef;
 typedef const struct __CFString * CFStringRef;
@@ -81,6 +81,10 @@ extern
 const CFSetCallBacks kCFTypeSetCallBacks;
 extern
 const CFDictionaryKeyCallBacks kCFCopyStringDictionaryKeyCallBacks;
+extern
+const void *CFArrayGetValueAtIndex(CFArrayRef theArray, CFIndex idx);
+extern
+CFIndex CFArrayGetCount(CFArrayRef theArray);
 CFDictionaryRef CFDictionaryCreate(CFAllocatorRef allocator, const void **keys, const void **values, CFIndex numValues, const 
 CFDictionaryKeyCallBacks *keyCallBacks, const CFDictionaryValueCallBacks *valueCallBacks);
 CFNumberRef CFNumberCreate(CFAllocatorRef allocator, CFNumberType theType, const void *valuePtr);
@@ -123,4 +127,40 @@ void CreateDict(int *elems) {
   CFDictionaryRef dict1 = CFDictionaryCreate(kCFAllocatorDefault, (const void**)keys, (const void**)values, numValues, &keyCB, &valCB); // no warning
   CFDictionaryRef dict2 = CFDictionaryCreate(kCFAllocatorDefault, (const void**)elems[0], (const void**)values, numValues, &keyCB, &valCB); //expected-warning {{The first argument to 'CFDictionaryCreate' must be a C array of}}
   CFDictionaryRef dict3 = CFDictionaryCreate(kCFAllocatorDefault, (const void**)keys, (const void**)elems, numValues, &keyCB, &valCB); // expected-warning {{The second argument to 'CFDictionaryCreate' must be a C array of pointer-sized values}}
-}
\ No newline at end of file
+}
+
+void OutOfBoundsSymbolicOffByOne(const void ** input, CFIndex S) {
+  CFArrayRef array;
+  array = CFArrayCreate(kCFAllocatorDefault, input, S, 0);
+  const void *s1 = CFArrayGetValueAtIndex(array, 0);   // no warning
+  const void *s2 = CFArrayGetValueAtIndex(array, S-1); // no warning
+  const void *s3 = CFArrayGetValueAtIndex(array, S);   // expected-warning {{Index is out of bounds}}
+}
+
+void OutOfBoundsConst(const void ** input, CFIndex S) {
+  CFArrayRef array;
+  array = CFArrayCreate(kCFAllocatorDefault, input, 3, 0);
+  const void *s1 = CFArrayGetValueAtIndex(array, 0); // no warning
+  const void *s2 = CFArrayGetValueAtIndex(array, 2); // no warning
+  const void *s3 = CFArrayGetValueAtIndex(array, 5); // expected-warning {{Index is out of bounds}}
+  
+  // TODO: The solver is probably not strong enough here.
+  CFIndex sIndex;
+  for (sIndex = 0 ; sIndex <= 5 ; sIndex += 3 ) {
+    const void *s = CFArrayGetValueAtIndex(array, sIndex); 
+  }  
+}
+
+void OutOfBoundsZiro(const void ** input, CFIndex S) {
+  CFArrayRef array;
+  // The API allows to set the size to 0. Check that we don't undeflow when the size is 0.
+  array = CFArrayCreate(kCFAllocatorDefault, 0, 0, 0);
+  const void *s1 = CFArrayGetValueAtIndex(array, 0); // expected-warning {{Index is out of bounds}}
+}
+
+void TestGetCount(CFArrayRef A, CFIndex sIndex, CFIndex badIndex) {
+  CFIndex sCount = CFArrayGetCount(A);
+  if (sCount > sIndex)
+    const void *s1 = CFArrayGetValueAtIndex(A, sIndex);
+  const void *s2 = CFArrayGetValueAtIndex(A, sCount);// expected-warning {{Index is out of bounds}}
+}