[analyzer] Add checks for common anti-patterns in strncat.

(Since this is syntax only, might be a good candidate for turning into a
compiler warning.)

llvm-svn: 149407

clang/include/clang/StaticAnalyzer/Core/PathSensitive/CheckerContext.h

@@ -164,6 +164,8 @@ public:
   /// Given a function declaration and a name checks if this is a C lib
   /// function with the given name.
   bool isCLibraryFunction(const FunctionDecl *FD, StringRef Name);
+  static bool isCLibraryFunction(const FunctionDecl *FD, StringRef Name,
+                                 ASTContext &Context);
 
   /// \brief Depending on wither the location corresponds to a macro, return 
   /// either the macro name or the token spelling.

clang/lib/StaticAnalyzer/Checkers/CMakeLists.txt

@@ -14,6 +14,7 @@ add_clang_library(clangStaticAnalyzerCheckers
   BasicObjCFoundationChecks.cpp
   BuiltinFunctionChecker.cpp
   CStringChecker.cpp
+  CStringSyntaxChecker.cpp
   CallAndMessageChecker.cpp
   CastSizeChecker.cpp
   CastToStructChecker.cpp

clang/lib/StaticAnalyzer/Checkers/CStringSyntaxChecker.cpp

@@ -0,0 +1,190 @@
+//== CStringSyntaxChecker.cpp - CoreFoundation containers API *- C++ -*-==//
+//
+//                     The LLVM Compiler Infrastructure
+//
+// This file is distributed under the University of Illinois Open Source
+// License. See LICENSE.TXT for details.
+//
+//===----------------------------------------------------------------------===//
+//
+// An AST checker that looks for common pitfalls when using C string APIs.
+//  - Identifies erroneous patterns in the last argument to strncat - the number
+//    of bytes to copy.
+//
+//===----------------------------------------------------------------------===//
+#include "ClangSACheckers.h"
+#include "clang/Analysis/AnalysisContext.h"
+#include "clang/AST/Expr.h"
+#include "clang/AST/OperationKinds.h"
+#include "clang/AST/StmtVisitor.h"
+#include "clang/Basic/TargetInfo.h"
+#include "clang/Basic/TypeTraits.h"
+#include "clang/StaticAnalyzer/Core/Checker.h"
+#include "clang/StaticAnalyzer/Core/BugReporter/BugReporter.h"
+#include "clang/StaticAnalyzer/Core/PathSensitive/AnalysisManager.h"
+#include "clang/StaticAnalyzer/Core/PathSensitive/CheckerContext.h"
+#include "llvm/Support/raw_ostream.h"
+
+using namespace clang;
+using namespace ento;
+
+namespace {
+class WalkAST: public StmtVisitor<WalkAST> {
+  BugReporter &BR;
+  AnalysisDeclContext* AC;
+  ASTContext &ASTC;
+
+  /// Check if two expressions refer to the same declaration.
+  inline bool sameDecl(const Expr *A1, const Expr *A2) {
+    if (const DeclRefExpr *D1 = dyn_cast<DeclRefExpr>(A1->IgnoreParenCasts()))
+      if (const DeclRefExpr *D2 = dyn_cast<DeclRefExpr>(A2->IgnoreParenCasts()))
+        return D1->getDecl() == D2->getDecl();
+    return false;
+  }
+
+  /// Check if the expression E is a sizeof(WithArg).
+  inline bool isSizeof(const Expr *E, const Expr *WithArg) {
+    if (const UnaryExprOrTypeTraitExpr *UE =
+    dyn_cast<UnaryExprOrTypeTraitExpr>(E))
+      if (UE->getKind() == UETT_SizeOf)
+        return sameDecl(UE->getArgumentExpr(), WithArg);
+    return false;
+  }
+
+  /// Check if the expression E is a strlen(WithArg).
+  inline bool isStrlen(const Expr *E, const Expr *WithArg) {
+    if (const CallExpr *CE = dyn_cast<CallExpr>(E)) {
+      const FunctionDecl *FD = CE->getDirectCallee();
+      if (!FD)
+        return false;
+      return (CheckerContext::isCLibraryFunction(FD, "strlen", ASTC)
+          && sameDecl(CE->getArg(0), WithArg));
+    }
+    return false;
+  }
+
+  /// Check if the expression is an integer literal with value 1.
+  inline bool isOne(const Expr *E) {
+    if (const IntegerLiteral *IL = dyn_cast<IntegerLiteral>(E))
+      return (IL->getValue().isIntN(1));
+    return false;
+  }
+
+  inline StringRef getPrintableName(const Expr *E) {
+    if (const DeclRefExpr *D = dyn_cast<DeclRefExpr>(E->IgnoreParenCasts()))
+      return D->getDecl()->getName();
+    return StringRef();
+  }
+
+  /// Identify erroneous patterns in the last argument to strncat - the number
+  /// of bytes to copy.
+  bool containsBadStrncatPattern(const CallExpr *CE);
+
+public:
+  WalkAST(BugReporter &br, AnalysisDeclContext* ac) :
+      BR(br), AC(ac), ASTC(AC->getASTContext()) {
+  }
+
+  // Statement visitor methods.
+  void VisitChildren(Stmt *S);
+  void VisitStmt(Stmt *S) {
+    VisitChildren(S);
+  }
+  void VisitCallExpr(CallExpr *CE);
+};
+} // end anonymous namespace
+
+// The correct size argument should look like following:
+//   strncat(dst, src, sizeof(dst) - strlen(dest) - 1);
+// We look for the following anti-patterns:
+//   - strncat(dst, src, sizeof(dst) - strlen(dst));
+//   - strncat(dst, src, sizeof(dst) - 1);
+//   - strncat(dst, src, sizeof(dst));
+bool WalkAST::containsBadStrncatPattern(const CallExpr *CE) {
+  const Expr *DstArg = CE->getArg(0);
+  const Expr *SrcArg = CE->getArg(1);
+  const Expr *LenArg = CE->getArg(2);
+
+  // Identify wrong size expressions, which are commonly used instead.
+  if (const BinaryOperator *BE =
+              dyn_cast<BinaryOperator>(LenArg->IgnoreParenCasts())) {
+    // - sizeof(dst) - strlen(dst)
+    if (BE->getOpcode() == BO_Sub) {
+      const Expr *L = BE->getLHS();
+      const Expr *R = BE->getRHS();
+      if (isSizeof(L, DstArg) && isStrlen(R, DstArg))
+        return true;
+
+      // - sizeof(dst) - 1
+      if (isSizeof(L, DstArg) && isOne(R->IgnoreParenCasts()))
+        return true;
+    }
+  }
+  // - sizeof(dst)
+  if (isSizeof(LenArg, DstArg))
+    return true;
+
+  // - sizeof(src)
+  if (isSizeof(LenArg, SrcArg))
+    return true;
+  return false;
+}
+
+void WalkAST::VisitCallExpr(CallExpr *CE) {
+  const FunctionDecl *FD = CE->getDirectCallee();
+  if (!FD)
+    return;
+
+  if (CheckerContext::isCLibraryFunction(FD, "strncat", ASTC)) {
+    if (containsBadStrncatPattern(CE)) {
+      const Expr *DstArg = CE->getArg(0);
+      const Expr *LenArg = CE->getArg(2);
+      SourceRange R = LenArg->getSourceRange();
+      PathDiagnosticLocation Loc =
+        PathDiagnosticLocation::createBegin(LenArg, BR.getSourceManager(), AC);
+
+      StringRef DstName = getPrintableName(DstArg);
+
+      llvm::SmallString<256> S;
+      llvm::raw_svector_ostream os(S);
+      os << "Potential buffer overflow. ";
+      if (!DstName.empty()) {
+        os << "Replace with 'sizeof(" << DstName << ") "
+              "- strlen(" << DstName <<") - 1'";
+        os << " or u";
+      } else
+        os << "U";
+      os << "se a safer 'strlcat' API";
+
+      BR.EmitBasicReport("Anti-pattern in the argument", "C String API",
+                         os.str(), Loc, &R, 1);
+    }
+  }
+
+  // Recurse and check children.
+  VisitChildren(CE);
+}
+
+void WalkAST::VisitChildren(Stmt *S) {
+  for (Stmt::child_iterator I = S->child_begin(), E = S->child_end(); I != E;
+      ++I)
+    if (Stmt *child = *I)
+      Visit(child);
+}
+
+namespace {
+class CStringSyntaxChecker: public Checker<check::ASTCodeBody> {
+public:
+
+  void checkASTCodeBody(const Decl *D, AnalysisManager& Mgr,
+      BugReporter &BR) const {
+    WalkAST walker(BR, Mgr.getAnalysisDeclContext(D));
+    walker.Visit(D->getBody());
+  }
+};
+}
+
+void ento::registerCStringSyntaxChecker(CheckerManager &mgr) {
+  mgr.registerChecker<CStringSyntaxChecker>();
+}
+

clang/lib/StaticAnalyzer/Checkers/Checkers.td

@@ -33,6 +33,7 @@ def Taint : Package<"taint">, InPackage<SecurityExperimental>, Hidden;
 
 def Unix : Package<"unix">;
 def UnixExperimental : Package<"unix">, InPackage<Experimental>, Hidden;
+def CString : Package<"cstring">, InPackage<UnixExperimental>, Hidden;
 
 def OSX : Package<"osx">;
 def OSXExperimental : Package<"osx">, InPackage<Experimental>, Hidden;
@@ -281,10 +282,6 @@ def ChrootChecker : Checker<"Chroot">,
   HelpText<"Check improper use of chroot">,
   DescFile<"ChrootChecker.cpp">;
 
-def CStringChecker : Checker<"CString">,
-  HelpText<"Check calls to functions in <string.h>">,
-  DescFile<"CStringChecker.cpp">;
-
 def MallocChecker : Checker<"Malloc">,
   HelpText<"Check for potential memory leaks, double free, and use-after-free problems">,
   DescFile<"MallocChecker.cpp">;
@@ -303,6 +300,17 @@ def StreamChecker : Checker<"Stream">,
 
 } // end "unix.experimental"
 
+let ParentPackage = CString in {
+
+def CStringChecker : Checker<"Generic">,
+  HelpText<"Check calls to functions in <string.h>">,
+  DescFile<"CStringChecker.cpp">;
+
+def CStringSyntaxChecker : Checker<"BadSizeArg">,
+  HelpText<"Check the size argument passed into C string functions for common erroneous patterns">,
+  DescFile<"CStringSyntaxChecker.cpp">;  
+}
+
 //===----------------------------------------------------------------------===//
 // Mac OS X, Cocoa, and Core Foundation checkers.
 //===----------------------------------------------------------------------===//

clang/lib/StaticAnalyzer/Core/CheckerContext.cpp

@@ -37,19 +37,27 @@ StringRef CheckerContext::getCalleeName(const FunctionDecl *FunDecl) const {
 
 
 bool CheckerContext::isCLibraryFunction(const FunctionDecl *FD,
-                                        StringRef Name){
+                                        StringRef Name) {
+  return isCLibraryFunction(FD, Name, getASTContext());
+}
+
+bool CheckerContext::isCLibraryFunction(const FunctionDecl *FD,
+                                        StringRef Name, ASTContext &Context) {
   // To avoid false positives (Ex: finding user defined functions with
   // similar names), only perform fuzzy name matching when it's a builtin.
   // Using a string compare is slow, we might want to switch on BuiltinID here.
   unsigned BId = FD->getBuiltinID();
   if (BId != 0) {
-    ASTContext &Context = getASTContext();
     StringRef BName = Context.BuiltinInfo.GetName(BId);
-    if (StringRef(BName).find(Name) != StringRef::npos)
+    if (BName.find(Name) != StringRef::npos)
       return true;
   }
 
-  if (FD->isExternC() && FD->getIdentifier()->getName().equals(Name))
+  StringRef FName = FD->getIdentifier()->getName();
+  if (FName.startswith("__inline"))
+    return (FName.find(Name) != StringRef::npos);
+
+  if (FD->isExternC() && FName.equals(Name))
     return true;
 
   return false;

clang/test/Analysis/bstring.c

@@ -1,7 +1,7 @@
-// RUN: %clang_cc1 -analyze -analyzer-checker=core,experimental.unix.CString -analyzer-store=region -Wno-null-dereference -verify %s
-// RUN: %clang_cc1 -analyze -DUSE_BUILTINS -analyzer-checker=core,experimental.unix.CString -analyzer-store=region -Wno-null-dereference -verify %s
-// RUN: %clang_cc1 -analyze -DVARIANT -analyzer-checker=core,experimental.unix.CString -analyzer-store=region -Wno-null-dereference -verify %s
-// RUN: %clang_cc1 -analyze -DUSE_BUILTINS -DVARIANT -analyzer-checker=core,experimental.unix.CString -analyzer-store=region -Wno-null-dereference -verify %s
+// RUN: %clang_cc1 -analyze -analyzer-checker=core,experimental.unix.cstring.Generic -analyzer-store=region -Wno-null-dereference -verify %s
+// RUN: %clang_cc1 -analyze -DUSE_BUILTINS -analyzer-checker=core,experimental.unix.cstring.Generic -analyzer-store=region -Wno-null-dereference -verify %s
+// RUN: %clang_cc1 -analyze -DVARIANT -analyzer-checker=core,experimental.unix.cstring.Generic -analyzer-store=region -Wno-null-dereference -verify %s
+// RUN: %clang_cc1 -analyze -DUSE_BUILTINS -DVARIANT -analyzer-checker=core,experimental.unix.cstring.Generic -analyzer-store=region -Wno-null-dereference -verify %s
 
 //===----------------------------------------------------------------------===
 // Declarations

clang/test/Analysis/cstring-syntax.c

@@ -0,0 +1,13 @@
+// RUN: %clang_cc1 -analyze -analyzer-checker=experimental.unix.cstring.BadSizeArg -analyzer-store=region -Wno-strlcpy-strlcat-size -Wno-sizeof-array-argument -Wno-sizeof-pointer-memaccess -verify %s
+
+typedef __SIZE_TYPE__ size_t;
+char  *strncat(char *, const char *, size_t);
+size_t strlen (const char *s);
+
+void testStrncat(const char *src) {
+  char dest[10];
+  strncat(dest, "AAAAAAAAAAAAAAAAAAAAAAAAAAAAA", sizeof(dest) - 1); // expected-warning {{Potential buffer overflow. Replace with 'sizeof(dest) - strlen(dest) - 1' or use a safer 'strlcat' API}}
+  strncat(dest, "AAAAAAAAAAAAAAAAAAAAAAAAAAA", sizeof(dest)); // expected-warning {{Potential buffer overflow. Replace with}}
+  strncat(dest, "AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA", sizeof(dest) - strlen(dest)); // expected-warning {{Potential buffer overflow. Replace with}}
+  strncat(dest, src, sizeof(src)); // expected-warning {{Potential buffer overflow. Replace with}}
+}

clang/test/Analysis/string.c

@@ -1,7 +1,7 @@
-// RUN: %clang_cc1 -analyze -analyzer-checker=core,experimental.unix.CString,experimental.deadcode.UnreachableCode -analyzer-store=region -Wno-null-dereference -verify %s
-// RUN: %clang_cc1 -analyze -DUSE_BUILTINS -analyzer-checker=core,experimental.unix.CString,experimental.deadcode.UnreachableCode -analyzer-store=region -Wno-null-dereference -verify %s
-// RUN: %clang_cc1 -analyze -DVARIANT -analyzer-checker=core,experimental.unix.CString,experimental.deadcode.UnreachableCode -analyzer-store=region -Wno-null-dereference -verify %s
-// RUN: %clang_cc1 -analyze -DUSE_BUILTINS -DVARIANT -analyzer-checker=experimental.security.taint,core,experimental.unix.CString,experimental.deadcode.UnreachableCode -analyzer-store=region -Wno-null-dereference -verify %s
+// RUN: %clang_cc1 -analyze -analyzer-checker=core,experimental.unix.cstring.Generic,experimental.deadcode.UnreachableCode -analyzer-store=region -Wno-null-dereference -verify %s
+// RUN: %clang_cc1 -analyze -DUSE_BUILTINS -analyzer-checker=core,experimental.unix.cstring.Generic,experimental.deadcode.UnreachableCode -analyzer-store=region -Wno-null-dereference -verify %s
+// RUN: %clang_cc1 -analyze -DVARIANT -analyzer-checker=core,experimental.unix.cstring.Generic,experimental.deadcode.UnreachableCode -analyzer-store=region -Wno-null-dereference -verify %s
+// RUN: %clang_cc1 -analyze -DUSE_BUILTINS -DVARIANT -analyzer-checker=experimental.security.taint,core,experimental.unix.cstring.Generic,experimental.deadcode.UnreachableCode -analyzer-store=region -Wno-null-dereference -verify %s
 
 //===----------------------------------------------------------------------===
 // Declarations