llvm-project/clang/lib/Checker/BugReporter.cpp

1897 lines
57 KiB
C++
Raw Normal View History

// BugReporter.cpp - Generate PathDiagnostics for Bugs ------------*- C++ -*--//
//
// The LLVM Compiler Infrastructure
//
// This file is distributed under the University of Illinois Open Source
// License. See LICENSE.TXT for details.
//
//===----------------------------------------------------------------------===//
//
// This file defines BugReporter, a utility class for generating
Introduce a new concept to the static analyzer: SValuator. GRTransferFuncs had the conflated role of both constructing SVals (symbolic expressions) as well as handling checker-specific logic. Now SValuator has the role of constructing SVals from expressions and GRTransferFuncs just handles checker-specific logic. The motivation is by separating these two concepts we will be able to much more easily create richer constraint-generating logic without coupling it to the main checker transfer function logic. We now have one implementation of SValuator: SimpleSValuator. SimpleSValuator is essentially the SVal-related logic that was in GRSimpleVals (which is removed in this patch). This includes the logic for EvalBinOp, EvalCast, etc. Because SValuator has a narrower role than the old GRTransferFuncs, the interfaces are much simpler, and so is the implementation of SimpleSValuator compared to GRSimpleVals. I also did a line-by-line review of SVal-related logic in GRSimpleVals and cleaned it up while moving it over to SimpleSValuator. As a consequence of removing GRSimpleVals, there is no longer a '-checker-simple' option. The '-checker-cfref' did everything that option did but also ran the retain/release checker. Of course a user may not always wish to run the retain/release checker, nor do we wish core analysis logic buried in the checker-specific logic. The next step is to refactor the logic in CFRefCount.cpp to separate out these pieces into the core analysis engine. llvm-svn: 74229
2009-06-26 08:05:51 +08:00
// PathDiagnostics.
//
//===----------------------------------------------------------------------===//
#include "clang/Checker/BugReporter/BugReporter.h"
#include "clang/Checker/BugReporter/BugType.h"
#include "clang/Checker/PathSensitive/GRExprEngine.h"
#include "clang/AST/ASTContext.h"
#include "clang/Analysis/CFG.h"
#include "clang/AST/Expr.h"
#include "clang/AST/ParentMap.h"
#include "clang/AST/StmtObjC.h"
#include "clang/Basic/SourceManager.h"
#include "clang/Analysis/ProgramPoint.h"
#include "clang/Checker/BugReporter/PathDiagnostic.h"
#include "llvm/Support/raw_ostream.h"
#include "llvm/ADT/DenseMap.h"
#include "llvm/ADT/STLExtras.h"
#include "llvm/ADT/OwningPtr.h"
#include <queue>
using namespace clang;
BugReporterVisitor::~BugReporterVisitor() {}
BugReporterContext::~BugReporterContext() {
for (visitor_iterator I = visitor_begin(), E = visitor_end(); I != E; ++I)
if ((*I)->isOwnedByReporterContext()) delete *I;
}
void BugReporterContext::addVisitor(BugReporterVisitor* visitor) {
if (!visitor)
return;
llvm::FoldingSetNodeID ID;
visitor->Profile(ID);
void *InsertPos;
if (CallbacksSet.FindNodeOrInsertPos(ID, InsertPos)) {
delete visitor;
return;
}
CallbacksSet.InsertNode(visitor, InsertPos);
Callbacks = F.Add(visitor, Callbacks);
}
//===----------------------------------------------------------------------===//
// Helper routines for walking the ExplodedGraph and fetching statements.
//===----------------------------------------------------------------------===//
static inline const Stmt* GetStmt(ProgramPoint P) {
if (const StmtPoint* SP = dyn_cast<StmtPoint>(&P))
return SP->getStmt();
else if (const BlockEdge* BE = dyn_cast<BlockEdge>(&P))
return BE->getSrc()->getTerminator();
return 0;
}
static inline const ExplodedNode*
GetPredecessorNode(const ExplodedNode* N) {
return N->pred_empty() ? NULL : *(N->pred_begin());
}
static inline const ExplodedNode*
GetSuccessorNode(const ExplodedNode* N) {
return N->succ_empty() ? NULL : *(N->succ_begin());
}
static const Stmt* GetPreviousStmt(const ExplodedNode* N) {
for (N = GetPredecessorNode(N); N; N = GetPredecessorNode(N))
if (const Stmt *S = GetStmt(N->getLocation()))
return S;
return 0;
}
static const Stmt* GetNextStmt(const ExplodedNode* N) {
for (N = GetSuccessorNode(N); N; N = GetSuccessorNode(N))
if (const Stmt *S = GetStmt(N->getLocation())) {
// Check if the statement is '?' or '&&'/'||'. These are "merges",
// not actual statement points.
switch (S->getStmtClass()) {
case Stmt::ChooseExprClass:
case Stmt::ConditionalOperatorClass: continue;
case Stmt::BinaryOperatorClass: {
BinaryOperator::Opcode Op = cast<BinaryOperator>(S)->getOpcode();
if (Op == BinaryOperator::LAnd || Op == BinaryOperator::LOr)
continue;
break;
}
default:
break;
}
// Some expressions don't have locations.
if (S->getLocStart().isInvalid())
continue;
return S;
}
return 0;
}
static inline const Stmt*
GetCurrentOrPreviousStmt(const ExplodedNode* N) {
if (const Stmt *S = GetStmt(N->getLocation()))
return S;
return GetPreviousStmt(N);
}
static inline const Stmt*
GetCurrentOrNextStmt(const ExplodedNode* N) {
if (const Stmt *S = GetStmt(N->getLocation()))
return S;
return GetNextStmt(N);
}
//===----------------------------------------------------------------------===//
// PathDiagnosticBuilder and its associated routines and helper objects.
//===----------------------------------------------------------------------===//
typedef llvm::DenseMap<const ExplodedNode*,
const ExplodedNode*> NodeBackMap;
namespace {
class NodeMapClosure : public BugReport::NodeResolver {
NodeBackMap& M;
public:
NodeMapClosure(NodeBackMap *m) : M(*m) {}
~NodeMapClosure() {}
const ExplodedNode* getOriginalNode(const ExplodedNode* N) {
NodeBackMap::iterator I = M.find(N);
return I == M.end() ? 0 : I->second;
}
};
class PathDiagnosticBuilder : public BugReporterContext {
BugReport *R;
PathDiagnosticClient *PDC;
llvm::OwningPtr<ParentMap> PM;
NodeMapClosure NMC;
public:
PathDiagnosticBuilder(GRBugReporter &br,
BugReport *r, NodeBackMap *Backmap,
PathDiagnosticClient *pdc)
: BugReporterContext(br),
R(r), PDC(pdc), NMC(Backmap) {
addVisitor(R);
}
PathDiagnosticLocation ExecutionContinues(const ExplodedNode* N);
PathDiagnosticLocation ExecutionContinues(llvm::raw_string_ostream& os,
const ExplodedNode* N);
Decl const &getCodeDecl() { return R->getEndNode()->getCodeDecl(); }
ParentMap& getParentMap() { return R->getEndNode()->getParentMap(); }
const Stmt *getParent(const Stmt *S) {
return getParentMap().getParent(S);
}
virtual NodeMapClosure& getNodeResolver() { return NMC; }
BugReport& getReport() { return *R; }
PathDiagnosticLocation getEnclosingStmtLocation(const Stmt *S);
PathDiagnosticLocation
getEnclosingStmtLocation(const PathDiagnosticLocation &L) {
if (const Stmt *S = L.asStmt())
return getEnclosingStmtLocation(S);
return L;
}
PathDiagnosticClient::PathGenerationScheme getGenerationScheme() const {
return PDC ? PDC->getGenerationScheme() : PathDiagnosticClient::Extensive;
}
bool supportsLogicalOpControlFlow() const {
return PDC ? PDC->supportsLogicalOpControlFlow() : true;
}
};
} // end anonymous namespace
PathDiagnosticLocation
PathDiagnosticBuilder::ExecutionContinues(const ExplodedNode* N) {
if (const Stmt *S = GetNextStmt(N))
return PathDiagnosticLocation(S, getSourceManager());
return FullSourceLoc(N->getLocationContext()->getDecl()->getBodyRBrace(),
getSourceManager());
}
PathDiagnosticLocation
PathDiagnosticBuilder::ExecutionContinues(llvm::raw_string_ostream& os,
const ExplodedNode* N) {
// Slow, but probably doesn't matter.
if (os.str().empty())
os << ' ';
const PathDiagnosticLocation &Loc = ExecutionContinues(N);
if (Loc.asStmt())
os << "Execution continues on line "
<< getSourceManager().getInstantiationLineNumber(Loc.asLocation())
<< '.';
else {
os << "Execution jumps to the end of the ";
const Decl *D = N->getLocationContext()->getDecl();
if (isa<ObjCMethodDecl>(D))
os << "method";
else if (isa<FunctionDecl>(D))
os << "function";
else {
assert(isa<BlockDecl>(D));
os << "anonymous block";
}
os << '.';
}
return Loc;
}
static bool IsNested(const Stmt *S, ParentMap &PM) {
if (isa<Expr>(S) && PM.isConsumedExpr(cast<Expr>(S)))
return true;
const Stmt *Parent = PM.getParentIgnoreParens(S);
if (Parent)
switch (Parent->getStmtClass()) {
case Stmt::ForStmtClass:
case Stmt::DoStmtClass:
case Stmt::WhileStmtClass:
return true;
default:
break;
}
return false;
}
PathDiagnosticLocation
PathDiagnosticBuilder::getEnclosingStmtLocation(const Stmt *S) {
assert(S && "Null Stmt* passed to getEnclosingStmtLocation");
ParentMap &P = getParentMap();
SourceManager &SMgr = getSourceManager();
while (IsNested(S, P)) {
const Stmt *Parent = P.getParentIgnoreParens(S);
if (!Parent)
break;
switch (Parent->getStmtClass()) {
case Stmt::BinaryOperatorClass: {
const BinaryOperator *B = cast<BinaryOperator>(Parent);
if (B->isLogicalOp())
return PathDiagnosticLocation(S, SMgr);
break;
}
case Stmt::CompoundStmtClass:
case Stmt::StmtExprClass:
return PathDiagnosticLocation(S, SMgr);
case Stmt::ChooseExprClass:
// Similar to '?' if we are referring to condition, just have the edge
// point to the entire choose expression.
if (cast<ChooseExpr>(Parent)->getCond() == S)
return PathDiagnosticLocation(Parent, SMgr);
else
return PathDiagnosticLocation(S, SMgr);
case Stmt::ConditionalOperatorClass:
// For '?', if we are referring to condition, just have the edge point
// to the entire '?' expression.
if (cast<ConditionalOperator>(Parent)->getCond() == S)
return PathDiagnosticLocation(Parent, SMgr);
else
return PathDiagnosticLocation(S, SMgr);
case Stmt::DoStmtClass:
return PathDiagnosticLocation(S, SMgr);
case Stmt::ForStmtClass:
if (cast<ForStmt>(Parent)->getBody() == S)
return PathDiagnosticLocation(S, SMgr);
break;
case Stmt::IfStmtClass:
if (cast<IfStmt>(Parent)->getCond() != S)
return PathDiagnosticLocation(S, SMgr);
break;
case Stmt::ObjCForCollectionStmtClass:
if (cast<ObjCForCollectionStmt>(Parent)->getBody() == S)
return PathDiagnosticLocation(S, SMgr);
break;
case Stmt::WhileStmtClass:
if (cast<WhileStmt>(Parent)->getCond() != S)
return PathDiagnosticLocation(S, SMgr);
break;
default:
break;
}
S = Parent;
}
assert(S && "Cannot have null Stmt for PathDiagnosticLocation");
// Special case: DeclStmts can appear in for statement declarations, in which
// case the ForStmt is the context.
if (isa<DeclStmt>(S)) {
if (const Stmt *Parent = P.getParent(S)) {
switch (Parent->getStmtClass()) {
case Stmt::ForStmtClass:
case Stmt::ObjCForCollectionStmtClass:
return PathDiagnosticLocation(Parent, SMgr);
default:
break;
}
}
}
else if (isa<BinaryOperator>(S)) {
// Special case: the binary operator represents the initialization
// code in a for statement (this can happen when the variable being
// initialized is an old variable.
if (const ForStmt *FS =
dyn_cast_or_null<ForStmt>(P.getParentIgnoreParens(S))) {
if (FS->getInit() == S)
return PathDiagnosticLocation(FS, SMgr);
}
}
return PathDiagnosticLocation(S, SMgr);
}
//===----------------------------------------------------------------------===//
// ScanNotableSymbols: closure-like callback for scanning Store bindings.
//===----------------------------------------------------------------------===//
static const VarDecl*
GetMostRecentVarDeclBinding(const ExplodedNode* N,
GRStateManager& VMgr, SVal X) {
for ( ; N ; N = N->pred_empty() ? 0 : *N->pred_begin()) {
ProgramPoint P = N->getLocation();
if (!isa<PostStmt>(P))
continue;
const DeclRefExpr* DR = dyn_cast<DeclRefExpr>(cast<PostStmt>(P).getStmt());
if (!DR)
continue;
2010-02-09 00:18:51 +08:00
SVal Y = N->getState()->getSVal(DR);
if (X != Y)
continue;
const VarDecl* VD = dyn_cast<VarDecl>(DR->getDecl());
if (!VD)
continue;
return VD;
}
return 0;
}
namespace {
class NotableSymbolHandler
: public StoreManager::BindingsHandler {
SymbolRef Sym;
const GRState* PrevSt;
const Stmt* S;
GRStateManager& VMgr;
const ExplodedNode* Pred;
PathDiagnostic& PD;
BugReporter& BR;
public:
NotableSymbolHandler(SymbolRef sym, const GRState* prevst, const Stmt* s,
GRStateManager& vmgr, const ExplodedNode* pred,
PathDiagnostic& pd, BugReporter& br)
: Sym(sym), PrevSt(prevst), S(s), VMgr(vmgr), Pred(pred), PD(pd), BR(br) {}
bool HandleBinding(StoreManager& SMgr, Store store, const MemRegion* R,
SVal V) {
SymbolRef ScanSym = V.getAsSymbol();
if (ScanSym != Sym)
return true;
// Check if the previous state has this binding.
2010-02-09 00:18:51 +08:00
SVal X = PrevSt->getSVal(loc::MemRegionVal(R));
if (X == V) // Same binding?
return true;
// Different binding. Only handle assignments for now. We don't pull
// this check out of the loop because we will eventually handle other
// cases.
VarDecl *VD = 0;
if (const BinaryOperator* B = dyn_cast<BinaryOperator>(S)) {
if (!B->isAssignmentOp())
return true;
// What variable did we assign to?
DeclRefExpr* DR = dyn_cast<DeclRefExpr>(B->getLHS()->IgnoreParenCasts());
if (!DR)
return true;
VD = dyn_cast<VarDecl>(DR->getDecl());
}
else if (const DeclStmt* DS = dyn_cast<DeclStmt>(S)) {
// FIXME: Eventually CFGs won't have DeclStmts. Right now we
// assume that each DeclStmt has a single Decl. This invariant
// holds by contruction in the CFG.
VD = dyn_cast<VarDecl>(*DS->decl_begin());
}
if (!VD)
return true;
// What is the most recently referenced variable with this binding?
const VarDecl* MostRecent = GetMostRecentVarDeclBinding(Pred, VMgr, V);
if (!MostRecent)
return true;
// Create the diagnostic.
FullSourceLoc L(S->getLocStart(), BR.getSourceManager());
if (Loc::IsLocType(VD->getType())) {
std::string msg = "'" + std::string(VD->getNameAsString()) +
"' now aliases '" + MostRecent->getNameAsString() + "'";
PD.push_front(new PathDiagnosticEventPiece(L, msg));
}
return true;
}
};
}
static void HandleNotableSymbol(const ExplodedNode* N,
const Stmt* S,
SymbolRef Sym, BugReporter& BR,
PathDiagnostic& PD) {
const ExplodedNode* Pred = N->pred_empty() ? 0 : *N->pred_begin();
const GRState* PrevSt = Pred ? Pred->getState() : 0;
if (!PrevSt)
return;
// Look at the region bindings of the current state that map to the
// specified symbol. Are any of them not in the previous state?
GRStateManager& VMgr = cast<GRBugReporter>(BR).getStateManager();
NotableSymbolHandler H(Sym, PrevSt, S, VMgr, Pred, PD, BR);
cast<GRBugReporter>(BR).getStateManager().iterBindings(N->getState(), H);
}
namespace {
class ScanNotableSymbols
: public StoreManager::BindingsHandler {
llvm::SmallSet<SymbolRef, 10> AlreadyProcessed;
const ExplodedNode* N;
const Stmt* S;
GRBugReporter& BR;
PathDiagnostic& PD;
public:
ScanNotableSymbols(const ExplodedNode* n, const Stmt* s,
GRBugReporter& br, PathDiagnostic& pd)
: N(n), S(s), BR(br), PD(pd) {}
bool HandleBinding(StoreManager& SMgr, Store store,
const MemRegion* R, SVal V) {
SymbolRef ScanSym = V.getAsSymbol();
if (!ScanSym)
return true;
if (!BR.isNotable(ScanSym))
return true;
if (AlreadyProcessed.count(ScanSym))
return true;
AlreadyProcessed.insert(ScanSym);
HandleNotableSymbol(N, S, ScanSym, BR, PD);
return true;
}
};
} // end anonymous namespace
//===----------------------------------------------------------------------===//
// "Minimal" path diagnostic generation algorithm.
//===----------------------------------------------------------------------===//
static void CompactPathDiagnostic(PathDiagnostic &PD, const SourceManager& SM);
static void GenerateMinimalPathDiagnostic(PathDiagnostic& PD,
PathDiagnosticBuilder &PDB,
const ExplodedNode *N) {
SourceManager& SMgr = PDB.getSourceManager();
const ExplodedNode* NextNode = N->pred_empty()
? NULL : *(N->pred_begin());
while (NextNode) {
N = NextNode;
NextNode = GetPredecessorNode(N);
ProgramPoint P = N->getLocation();
if (const BlockEdge* BE = dyn_cast<BlockEdge>(&P)) {
CFGBlock* Src = BE->getSrc();
CFGBlock* Dst = BE->getDst();
Stmt* T = Src->getTerminator();
if (!T)
continue;
FullSourceLoc Start(T->getLocStart(), SMgr);
switch (T->getStmtClass()) {
default:
break;
case Stmt::GotoStmtClass:
case Stmt::IndirectGotoStmtClass: {
const Stmt* S = GetNextStmt(N);
if (!S)
continue;
std::string sbuf;
llvm::raw_string_ostream os(sbuf);
const PathDiagnosticLocation &End = PDB.getEnclosingStmtLocation(S);
os << "Control jumps to line "
<< End.asLocation().getInstantiationLineNumber();
PD.push_front(new PathDiagnosticControlFlowPiece(Start, End,
os.str()));
break;
}
case Stmt::SwitchStmtClass: {
// Figure out what case arm we took.
std::string sbuf;
llvm::raw_string_ostream os(sbuf);
if (Stmt* S = Dst->getLabel()) {
PathDiagnosticLocation End(S, SMgr);
switch (S->getStmtClass()) {
default:
os << "No cases match in the switch statement. "
"Control jumps to line "
<< End.asLocation().getInstantiationLineNumber();
break;
case Stmt::DefaultStmtClass:
os << "Control jumps to the 'default' case at line "
<< End.asLocation().getInstantiationLineNumber();
break;
case Stmt::CaseStmtClass: {
os << "Control jumps to 'case ";
CaseStmt* Case = cast<CaseStmt>(S);
Expr* LHS = Case->getLHS()->IgnoreParenCasts();
// Determine if it is an enum.
bool GetRawInt = true;
if (DeclRefExpr* DR = dyn_cast<DeclRefExpr>(LHS)) {
// FIXME: Maybe this should be an assertion. Are there cases
// were it is not an EnumConstantDecl?
EnumConstantDecl* D =
dyn_cast<EnumConstantDecl>(DR->getDecl());
if (D) {
GetRawInt = false;
os << D->getNameAsString();
}
}
2009-04-27 03:04:51 +08:00
if (GetRawInt)
os << LHS->EvaluateAsInt(PDB.getASTContext());
2009-04-27 03:04:51 +08:00
os << ":' at line "
<< End.asLocation().getInstantiationLineNumber();
break;
}
}
PD.push_front(new PathDiagnosticControlFlowPiece(Start, End,
os.str()));
}
else {
os << "'Default' branch taken. ";
const PathDiagnosticLocation &End = PDB.ExecutionContinues(os, N);
PD.push_front(new PathDiagnosticControlFlowPiece(Start, End,
os.str()));
}
break;
}
case Stmt::BreakStmtClass:
case Stmt::ContinueStmtClass: {
std::string sbuf;
llvm::raw_string_ostream os(sbuf);
PathDiagnosticLocation End = PDB.ExecutionContinues(os, N);
PD.push_front(new PathDiagnosticControlFlowPiece(Start, End,
os.str()));
break;
}
// Determine control-flow for ternary '?'.
case Stmt::ConditionalOperatorClass: {
std::string sbuf;
llvm::raw_string_ostream os(sbuf);
os << "'?' condition is ";
if (*(Src->succ_begin()+1) == Dst)
os << "false";
else
os << "true";
PathDiagnosticLocation End = PDB.ExecutionContinues(N);
if (const Stmt *S = End.asStmt())
End = PDB.getEnclosingStmtLocation(S);
PD.push_front(new PathDiagnosticControlFlowPiece(Start, End,
os.str()));
break;
}
// Determine control-flow for short-circuited '&&' and '||'.
case Stmt::BinaryOperatorClass: {
if (!PDB.supportsLogicalOpControlFlow())
break;
BinaryOperator *B = cast<BinaryOperator>(T);
std::string sbuf;
llvm::raw_string_ostream os(sbuf);
os << "Left side of '";
if (B->getOpcode() == BinaryOperator::LAnd) {
os << "&&" << "' is ";
if (*(Src->succ_begin()+1) == Dst) {
os << "false";
PathDiagnosticLocation End(B->getLHS(), SMgr);
PathDiagnosticLocation Start(B->getOperatorLoc(), SMgr);
PD.push_front(new PathDiagnosticControlFlowPiece(Start, End,
os.str()));
}
else {
os << "true";
PathDiagnosticLocation Start(B->getLHS(), SMgr);
PathDiagnosticLocation End = PDB.ExecutionContinues(N);
PD.push_front(new PathDiagnosticControlFlowPiece(Start, End,
os.str()));
}
}
else {
assert(B->getOpcode() == BinaryOperator::LOr);
os << "||" << "' is ";
if (*(Src->succ_begin()+1) == Dst) {
os << "false";
PathDiagnosticLocation Start(B->getLHS(), SMgr);
PathDiagnosticLocation End = PDB.ExecutionContinues(N);
PD.push_front(new PathDiagnosticControlFlowPiece(Start, End,
os.str()));
}
else {
os << "true";
PathDiagnosticLocation End(B->getLHS(), SMgr);
PathDiagnosticLocation Start(B->getOperatorLoc(), SMgr);
PD.push_front(new PathDiagnosticControlFlowPiece(Start, End,
os.str()));
}
}
break;
}
case Stmt::DoStmtClass: {
if (*(Src->succ_begin()) == Dst) {
std::string sbuf;
llvm::raw_string_ostream os(sbuf);
os << "Loop condition is true. ";
PathDiagnosticLocation End = PDB.ExecutionContinues(os, N);
if (const Stmt *S = End.asStmt())
End = PDB.getEnclosingStmtLocation(S);
PD.push_front(new PathDiagnosticControlFlowPiece(Start, End,
os.str()));
}
else {
PathDiagnosticLocation End = PDB.ExecutionContinues(N);
if (const Stmt *S = End.asStmt())
End = PDB.getEnclosingStmtLocation(S);
PD.push_front(new PathDiagnosticControlFlowPiece(Start, End,
"Loop condition is false. Exiting loop"));
}
break;
}
case Stmt::WhileStmtClass:
case Stmt::ForStmtClass: {
if (*(Src->succ_begin()+1) == Dst) {
std::string sbuf;
llvm::raw_string_ostream os(sbuf);
os << "Loop condition is false. ";
PathDiagnosticLocation End = PDB.ExecutionContinues(os, N);
if (const Stmt *S = End.asStmt())
End = PDB.getEnclosingStmtLocation(S);
PD.push_front(new PathDiagnosticControlFlowPiece(Start, End,
os.str()));
}
else {
PathDiagnosticLocation End = PDB.ExecutionContinues(N);
if (const Stmt *S = End.asStmt())
End = PDB.getEnclosingStmtLocation(S);
PD.push_front(new PathDiagnosticControlFlowPiece(Start, End,
"Loop condition is true. Entering loop body"));
}
break;
}
case Stmt::IfStmtClass: {
PathDiagnosticLocation End = PDB.ExecutionContinues(N);
if (const Stmt *S = End.asStmt())
End = PDB.getEnclosingStmtLocation(S);
if (*(Src->succ_begin()+1) == Dst)
PD.push_front(new PathDiagnosticControlFlowPiece(Start, End,
"Taking false branch"));
else
PD.push_front(new PathDiagnosticControlFlowPiece(Start, End,
"Taking true branch"));
break;
}
}
}
if (NextNode) {
for (BugReporterContext::visitor_iterator I = PDB.visitor_begin(),
E = PDB.visitor_end(); I!=E; ++I) {
if (PathDiagnosticPiece* p = (*I)->VisitNode(N, NextNode, PDB))
PD.push_front(p);
}
}
if (const PostStmt* PS = dyn_cast<PostStmt>(&P)) {
// Scan the region bindings, and see if a "notable" symbol has a new
// lval binding.
ScanNotableSymbols SNS(N, PS->getStmt(), PDB.getBugReporter(), PD);
PDB.getStateManager().iterBindings(N->getState(), SNS);
}
}
// After constructing the full PathDiagnostic, do a pass over it to compact
// PathDiagnosticPieces that occur within a macro.
CompactPathDiagnostic(PD, PDB.getSourceManager());
}
//===----------------------------------------------------------------------===//
// "Extensive" PathDiagnostic generation.
//===----------------------------------------------------------------------===//
static bool IsControlFlowExpr(const Stmt *S) {
const Expr *E = dyn_cast<Expr>(S);
if (!E)
return false;
E = E->IgnoreParenCasts();
if (isa<ConditionalOperator>(E))
return true;
if (const BinaryOperator *B = dyn_cast<BinaryOperator>(E))
if (B->isLogicalOp())
return true;
return false;
}
namespace {
class ContextLocation : public PathDiagnosticLocation {
bool IsDead;
public:
ContextLocation(const PathDiagnosticLocation &L, bool isdead = false)
: PathDiagnosticLocation(L), IsDead(isdead) {}
void markDead() { IsDead = true; }
bool isDead() const { return IsDead; }
};
class EdgeBuilder {
std::vector<ContextLocation> CLocs;
typedef std::vector<ContextLocation>::iterator iterator;
PathDiagnostic &PD;
PathDiagnosticBuilder &PDB;
PathDiagnosticLocation PrevLoc;
bool IsConsumedExpr(const PathDiagnosticLocation &L);
bool containsLocation(const PathDiagnosticLocation &Container,
const PathDiagnosticLocation &Containee);
PathDiagnosticLocation getContextLocation(const PathDiagnosticLocation &L);
PathDiagnosticLocation cleanUpLocation(PathDiagnosticLocation L,
bool firstCharOnly = false) {
if (const Stmt *S = L.asStmt()) {
const Stmt *Original = S;
while (1) {
// Adjust the location for some expressions that are best referenced
// by one of their subexpressions.
switch (S->getStmtClass()) {
default:
break;
case Stmt::ParenExprClass:
S = cast<ParenExpr>(S)->IgnoreParens();
firstCharOnly = true;
continue;
case Stmt::ConditionalOperatorClass:
S = cast<ConditionalOperator>(S)->getCond();
firstCharOnly = true;
continue;
case Stmt::ChooseExprClass:
S = cast<ChooseExpr>(S)->getCond();
firstCharOnly = true;
continue;
case Stmt::BinaryOperatorClass:
S = cast<BinaryOperator>(S)->getLHS();
firstCharOnly = true;
continue;
}
break;
}
if (S != Original)
L = PathDiagnosticLocation(S, L.getManager());
}
if (firstCharOnly)
L = PathDiagnosticLocation(L.asLocation());
return L;
}
void popLocation() {
if (!CLocs.back().isDead() && CLocs.back().asLocation().isFileID()) {
// For contexts, we only one the first character as the range.
rawAddEdge(cleanUpLocation(CLocs.back(), true));
}
CLocs.pop_back();
}
PathDiagnosticLocation IgnoreParens(const PathDiagnosticLocation &L);
public:
EdgeBuilder(PathDiagnostic &pd, PathDiagnosticBuilder &pdb)
: PD(pd), PDB(pdb) {
// If the PathDiagnostic already has pieces, add the enclosing statement
// of the first piece as a context as well.
if (!PD.empty()) {
PrevLoc = PD.begin()->getLocation();
if (const Stmt *S = PrevLoc.asStmt())
addExtendedContext(PDB.getEnclosingStmtLocation(S).asStmt());
}
}
~EdgeBuilder() {
while (!CLocs.empty()) popLocation();
// Finally, add an initial edge from the start location of the first
// statement (if it doesn't already exist).
// FIXME: Should handle CXXTryStmt if analyser starts supporting C++.
if (const CompoundStmt *CS =
PDB.getCodeDecl().getCompoundBody())
if (!CS->body_empty()) {
SourceLocation Loc = (*CS->body_begin())->getLocStart();
rawAddEdge(PathDiagnosticLocation(Loc, PDB.getSourceManager()));
}
}
void addEdge(PathDiagnosticLocation NewLoc, bool alwaysAdd = false);
void addEdge(const Stmt *S, bool alwaysAdd = false) {
addEdge(PathDiagnosticLocation(S, PDB.getSourceManager()), alwaysAdd);
}
void rawAddEdge(PathDiagnosticLocation NewLoc);
void addContext(const Stmt *S);
void addExtendedContext(const Stmt *S);
};
} // end anonymous namespace
PathDiagnosticLocation
EdgeBuilder::getContextLocation(const PathDiagnosticLocation &L) {
if (const Stmt *S = L.asStmt()) {
if (IsControlFlowExpr(S))
return L;
return PDB.getEnclosingStmtLocation(S);
}
return L;
}
bool EdgeBuilder::containsLocation(const PathDiagnosticLocation &Container,
const PathDiagnosticLocation &Containee) {
if (Container == Containee)
return true;
if (Container.asDecl())
return true;
if (const Stmt *S = Containee.asStmt())
if (const Stmt *ContainerS = Container.asStmt()) {
while (S) {
if (S == ContainerS)
return true;
S = PDB.getParent(S);
}
return false;
}
// Less accurate: compare using source ranges.
SourceRange ContainerR = Container.asRange();
SourceRange ContaineeR = Containee.asRange();
SourceManager &SM = PDB.getSourceManager();
SourceLocation ContainerRBeg = SM.getInstantiationLoc(ContainerR.getBegin());
SourceLocation ContainerREnd = SM.getInstantiationLoc(ContainerR.getEnd());
SourceLocation ContaineeRBeg = SM.getInstantiationLoc(ContaineeR.getBegin());
SourceLocation ContaineeREnd = SM.getInstantiationLoc(ContaineeR.getEnd());
unsigned ContainerBegLine = SM.getInstantiationLineNumber(ContainerRBeg);
unsigned ContainerEndLine = SM.getInstantiationLineNumber(ContainerREnd);
unsigned ContaineeBegLine = SM.getInstantiationLineNumber(ContaineeRBeg);
unsigned ContaineeEndLine = SM.getInstantiationLineNumber(ContaineeREnd);
assert(ContainerBegLine <= ContainerEndLine);
assert(ContaineeBegLine <= ContaineeEndLine);
return (ContainerBegLine <= ContaineeBegLine &&
ContainerEndLine >= ContaineeEndLine &&
(ContainerBegLine != ContaineeBegLine ||
SM.getInstantiationColumnNumber(ContainerRBeg) <=
SM.getInstantiationColumnNumber(ContaineeRBeg)) &&
(ContainerEndLine != ContaineeEndLine ||
SM.getInstantiationColumnNumber(ContainerREnd) >=
SM.getInstantiationColumnNumber(ContainerREnd)));
}
PathDiagnosticLocation
EdgeBuilder::IgnoreParens(const PathDiagnosticLocation &L) {
if (const Expr* E = dyn_cast_or_null<Expr>(L.asStmt()))
return PathDiagnosticLocation(E->IgnoreParenCasts(),
PDB.getSourceManager());
return L;
}
void EdgeBuilder::rawAddEdge(PathDiagnosticLocation NewLoc) {
if (!PrevLoc.isValid()) {
PrevLoc = NewLoc;
return;
}
const PathDiagnosticLocation &NewLocClean = cleanUpLocation(NewLoc);
const PathDiagnosticLocation &PrevLocClean = cleanUpLocation(PrevLoc);
if (NewLocClean.asLocation() == PrevLocClean.asLocation())
return;
// FIXME: Ignore intra-macro edges for now.
if (NewLocClean.asLocation().getInstantiationLoc() ==
PrevLocClean.asLocation().getInstantiationLoc())
return;
PD.push_front(new PathDiagnosticControlFlowPiece(NewLocClean, PrevLocClean));
PrevLoc = NewLoc;
}
void EdgeBuilder::addEdge(PathDiagnosticLocation NewLoc, bool alwaysAdd) {
if (!alwaysAdd && NewLoc.asLocation().isMacroID())
return;
const PathDiagnosticLocation &CLoc = getContextLocation(NewLoc);
while (!CLocs.empty()) {
ContextLocation &TopContextLoc = CLocs.back();
// Is the top location context the same as the one for the new location?
if (TopContextLoc == CLoc) {
if (alwaysAdd) {
if (IsConsumedExpr(TopContextLoc) &&
!IsControlFlowExpr(TopContextLoc.asStmt()))
TopContextLoc.markDead();
rawAddEdge(NewLoc);
}
return;
}
if (containsLocation(TopContextLoc, CLoc)) {
if (alwaysAdd) {
rawAddEdge(NewLoc);
if (IsConsumedExpr(CLoc) && !IsControlFlowExpr(CLoc.asStmt())) {
CLocs.push_back(ContextLocation(CLoc, true));
return;
}
}
CLocs.push_back(CLoc);
return;
}
// Context does not contain the location. Flush it.
popLocation();
}
// If we reach here, there is no enclosing context. Just add the edge.
rawAddEdge(NewLoc);
}
bool EdgeBuilder::IsConsumedExpr(const PathDiagnosticLocation &L) {
if (const Expr *X = dyn_cast_or_null<Expr>(L.asStmt()))
return PDB.getParentMap().isConsumedExpr(X) && !IsControlFlowExpr(X);
return false;
}
void EdgeBuilder::addExtendedContext(const Stmt *S) {
if (!S)
return;
const Stmt *Parent = PDB.getParent(S);
while (Parent) {
if (isa<CompoundStmt>(Parent))
Parent = PDB.getParent(Parent);
else
break;
}
if (Parent) {
switch (Parent->getStmtClass()) {
case Stmt::DoStmtClass:
case Stmt::ObjCAtSynchronizedStmtClass:
addContext(Parent);
default:
break;
}
}
addContext(S);
}
void EdgeBuilder::addContext(const Stmt *S) {
if (!S)
return;
PathDiagnosticLocation L(S, PDB.getSourceManager());
while (!CLocs.empty()) {
const PathDiagnosticLocation &TopContextLoc = CLocs.back();
// Is the top location context the same as the one for the new location?
if (TopContextLoc == L)
return;
if (containsLocation(TopContextLoc, L)) {
CLocs.push_back(L);
return;
}
// Context does not contain the location. Flush it.
popLocation();
}
CLocs.push_back(L);
}
static void GenerateExtensivePathDiagnostic(PathDiagnostic& PD,
PathDiagnosticBuilder &PDB,
const ExplodedNode *N) {
EdgeBuilder EB(PD, PDB);
2010-03-23 13:13:26 +08:00
const ExplodedNode* NextNode = N->pred_empty() ? NULL : *(N->pred_begin());
while (NextNode) {
N = NextNode;
NextNode = GetPredecessorNode(N);
ProgramPoint P = N->getLocation();
do {
// Block edges.
if (const BlockEdge *BE = dyn_cast<BlockEdge>(&P)) {
const CFGBlock &Blk = *BE->getSrc();
const Stmt *Term = Blk.getTerminator();
// Are we jumping to the head of a loop? Add a special diagnostic.
if (const Stmt *Loop = BE->getDst()->getLoopTarget()) {
PathDiagnosticLocation L(Loop, PDB.getSourceManager());
const CompoundStmt *CS = NULL;
if (!Term) {
if (const ForStmt *FS = dyn_cast<ForStmt>(Loop))
CS = dyn_cast<CompoundStmt>(FS->getBody());
else if (const WhileStmt *WS = dyn_cast<WhileStmt>(Loop))
CS = dyn_cast<CompoundStmt>(WS->getBody());
}
PathDiagnosticEventPiece *p =
new PathDiagnosticEventPiece(L,
"Looping back to the head of the loop");
EB.addEdge(p->getLocation(), true);
PD.push_front(p);
if (CS) {
PathDiagnosticLocation BL(CS->getRBracLoc(),
PDB.getSourceManager());
BL = PathDiagnosticLocation(BL.asLocation());
EB.addEdge(BL);
}
}
if (Term)
EB.addContext(Term);
break;
}
if (const BlockEntrance *BE = dyn_cast<BlockEntrance>(&P)) {
if (const Stmt* S = BE->getFirstStmt()) {
if (IsControlFlowExpr(S)) {
// Add the proper context for '&&', '||', and '?'.
EB.addContext(S);
}
else
EB.addExtendedContext(PDB.getEnclosingStmtLocation(S).asStmt());
}
break;
}
} while (0);
if (!NextNode)
continue;
for (BugReporterContext::visitor_iterator I = PDB.visitor_begin(),
E = PDB.visitor_end(); I!=E; ++I) {
if (PathDiagnosticPiece* p = (*I)->VisitNode(N, NextNode, PDB)) {
const PathDiagnosticLocation &Loc = p->getLocation();
EB.addEdge(Loc, true);
PD.push_front(p);
if (const Stmt *S = Loc.asStmt())
EB.addExtendedContext(PDB.getEnclosingStmtLocation(S).asStmt());
}
}
}
}
//===----------------------------------------------------------------------===//
// Methods for BugType and subclasses.
//===----------------------------------------------------------------------===//
BugType::~BugType() {
// Free up the equivalence class objects. Observe that we get a pointer to
// the object first before incrementing the iterator, as destroying the
// node before doing so means we will read from freed memory.
for (iterator I = begin(), E = end(); I !=E; ) {
BugReportEquivClass *EQ = &*I;
++I;
delete EQ;
}
}
void BugType::FlushReports(BugReporter &BR) {}
//===----------------------------------------------------------------------===//
// Methods for BugReport and subclasses.
//===----------------------------------------------------------------------===//
BugReport::~BugReport() {}
RangedBugReport::~RangedBugReport() {}
const Stmt* BugReport::getStmt() const {
ProgramPoint ProgP = EndNode->getLocation();
const Stmt *S = NULL;
if (BlockEntrance* BE = dyn_cast<BlockEntrance>(&ProgP)) {
CFGBlock &Exit = ProgP.getLocationContext()->getCFG()->getExit();
if (BE->getBlock() == &Exit)
S = GetPreviousStmt(EndNode);
}
if (!S)
S = GetStmt(ProgP);
return S;
}
PathDiagnosticPiece*
BugReport::getEndPath(BugReporterContext& BRC,
const ExplodedNode* EndPathNode) {
const Stmt* S = getStmt();
if (!S)
return NULL;
const SourceRange *Beg, *End;
getRanges(Beg, End);
PathDiagnosticLocation L(S, BRC.getSourceManager());
// Only add the statement itself as a range if we didn't specify any
// special ranges for this report.
PathDiagnosticPiece* P = new PathDiagnosticEventPiece(L, getDescription(),
Beg == End);
for (; Beg != End; ++Beg)
P->addRange(*Beg);
return P;
}
void BugReport::getRanges(const SourceRange*& beg, const SourceRange*& end) {
if (const Expr* E = dyn_cast_or_null<Expr>(getStmt())) {
R = E->getSourceRange();
assert(R.isValid());
beg = &R;
end = beg+1;
}
else
beg = end = 0;
}
SourceLocation BugReport::getLocation() const {
if (EndNode)
if (const Stmt* S = GetCurrentOrPreviousStmt(EndNode)) {
// For member expressions, return the location of the '.' or '->'.
if (const MemberExpr *ME = dyn_cast<MemberExpr>(S))
return ME->getMemberLoc();
// For binary operators, return the location of the operator.
if (const BinaryOperator *B = dyn_cast<BinaryOperator>(S))
return B->getOperatorLoc();
return S->getLocStart();
}
return FullSourceLoc();
}
PathDiagnosticPiece* BugReport::VisitNode(const ExplodedNode* N,
const ExplodedNode* PrevN,
BugReporterContext &BRC) {
return NULL;
}
//===----------------------------------------------------------------------===//
// Methods for BugReporter and subclasses.
//===----------------------------------------------------------------------===//
BugReportEquivClass::~BugReportEquivClass() {
for (iterator I=begin(), E=end(); I!=E; ++I) delete *I;
}
GRBugReporter::~GRBugReporter() { }
BugReporterData::~BugReporterData() {}
ExplodedGraph &GRBugReporter::getGraph() { return Eng.getGraph(); }
GRStateManager&
GRBugReporter::getStateManager() { return Eng.getStateManager(); }
BugReporter::~BugReporter() { FlushReports(); }
void BugReporter::FlushReports() {
if (BugTypes.isEmpty())
return;
// First flush the warnings for each BugType. This may end up creating new
// warnings and new BugTypes. Because ImmutableSet is a functional data
// structure, we do not need to worry about the iterators being invalidated.
for (BugTypesTy::iterator I=BugTypes.begin(), E=BugTypes.end(); I!=E; ++I)
const_cast<BugType*>(*I)->FlushReports(*this);
// Iterate through BugTypes a second time. BugTypes may have been updated
// with new BugType objects and new warnings.
for (BugTypesTy::iterator I=BugTypes.begin(), E=BugTypes.end(); I!=E; ++I) {
BugType *BT = const_cast<BugType*>(*I);
typedef llvm::FoldingSet<BugReportEquivClass> SetTy;
SetTy& EQClasses = BT->EQClasses;
for (SetTy::iterator EI=EQClasses.begin(), EE=EQClasses.end(); EI!=EE;++EI){
BugReportEquivClass& EQ = *EI;
FlushReport(EQ);
}
// Delete the BugType object.
delete BT;
}
// Remove all references to the BugType objects.
BugTypes = F.GetEmptySet();
}
//===----------------------------------------------------------------------===//
// PathDiagnostics generation.
//===----------------------------------------------------------------------===//
static std::pair<std::pair<ExplodedGraph*, NodeBackMap*>,
std::pair<ExplodedNode*, unsigned> >
MakeReportGraph(const ExplodedGraph* G,
const ExplodedNode** NStart,
const ExplodedNode** NEnd) {
// Create the trimmed graph. It will contain the shortest paths from the
// error nodes to the root. In the new graph we should only have one
// error node unless there are two or more error nodes with the same minimum
// path length.
ExplodedGraph* GTrim;
InterExplodedGraphMap* NMap;
llvm::DenseMap<const void*, const void*> InverseMap;
llvm::tie(GTrim, NMap) = G->Trim(NStart, NEnd, &InverseMap);
// Create owning pointers for GTrim and NMap just to ensure that they are
// released when this function exists.
llvm::OwningPtr<ExplodedGraph> AutoReleaseGTrim(GTrim);
llvm::OwningPtr<InterExplodedGraphMap> AutoReleaseNMap(NMap);
// Find the (first) error node in the trimmed graph. We just need to consult
// the node map (NMap) which maps from nodes in the original graph to nodes
// in the new graph.
std::queue<const ExplodedNode*> WS;
typedef llvm::DenseMap<const ExplodedNode*, unsigned> IndexMapTy;
IndexMapTy IndexMap;
for (const ExplodedNode** I = NStart; I != NEnd; ++I)
if (const ExplodedNode *N = NMap->getMappedNode(*I)) {
unsigned NodeIndex = (I - NStart) / sizeof(*I);
WS.push(N);
IndexMap[*I] = NodeIndex;
}
assert(!WS.empty() && "No error node found in the trimmed graph.");
// Create a new (third!) graph with a single path. This is the graph
// that will be returned to the caller.
ExplodedGraph *GNew = new ExplodedGraph(GTrim->getContext());
// Sometimes the trimmed graph can contain a cycle. Perform a reverse BFS
// to the root node, and then construct a new graph that contains only
// a single path.
llvm::DenseMap<const void*,unsigned> Visited;
unsigned cnt = 0;
const ExplodedNode* Root = 0;
while (!WS.empty()) {
const ExplodedNode* Node = WS.front();
WS.pop();
if (Visited.find(Node) != Visited.end())
continue;
Visited[Node] = cnt++;
if (Node->pred_empty()) {
Root = Node;
break;
}
for (ExplodedNode::const_pred_iterator I=Node->pred_begin(),
E=Node->pred_end(); I!=E; ++I)
WS.push(*I);
}
assert(Root);
// Now walk from the root down the BFS path, always taking the successor
// with the lowest number.
ExplodedNode *Last = 0, *First = 0;
NodeBackMap *BM = new NodeBackMap();
unsigned NodeIndex = 0;
for ( const ExplodedNode *N = Root ;;) {
// Lookup the number associated with the current node.
llvm::DenseMap<const void*,unsigned>::iterator I = Visited.find(N);
assert(I != Visited.end());
// Create the equivalent node in the new graph with the same state
// and location.
ExplodedNode* NewN = GNew->getNode(N->getLocation(), N->getState());
// Store the mapping to the original node.
llvm::DenseMap<const void*, const void*>::iterator IMitr=InverseMap.find(N);
assert(IMitr != InverseMap.end() && "No mapping to original node.");
(*BM)[NewN] = (const ExplodedNode*) IMitr->second;
// Link up the new node with the previous node.
if (Last)
NewN->addPredecessor(Last, *GNew);
Last = NewN;
// Are we at the final node?
IndexMapTy::iterator IMI =
IndexMap.find((const ExplodedNode*)(IMitr->second));
if (IMI != IndexMap.end()) {
First = NewN;
NodeIndex = IMI->second;
break;
}
// Find the next successor node. We choose the node that is marked
// with the lowest DFS number.
ExplodedNode::const_succ_iterator SI = N->succ_begin();
ExplodedNode::const_succ_iterator SE = N->succ_end();
N = 0;
for (unsigned MinVal = 0; SI != SE; ++SI) {
I = Visited.find(*SI);
if (I == Visited.end())
continue;
if (!N || I->second < MinVal) {
N = *SI;
MinVal = I->second;
}
}
assert(N);
}
assert(First);
return std::make_pair(std::make_pair(GNew, BM),
std::make_pair(First, NodeIndex));
}
/// CompactPathDiagnostic - This function postprocesses a PathDiagnostic object
/// and collapses PathDiagosticPieces that are expanded by macros.
static void CompactPathDiagnostic(PathDiagnostic &PD, const SourceManager& SM) {
typedef std::vector<std::pair<PathDiagnosticMacroPiece*, SourceLocation> >
MacroStackTy;
typedef std::vector<PathDiagnosticPiece*>
PiecesTy;
MacroStackTy MacroStack;
PiecesTy Pieces;
for (PathDiagnostic::iterator I = PD.begin(), E = PD.end(); I!=E; ++I) {
// Get the location of the PathDiagnosticPiece.
const FullSourceLoc Loc = I->getLocation().asLocation();
// Determine the instantiation location, which is the location we group
// related PathDiagnosticPieces.
SourceLocation InstantiationLoc = Loc.isMacroID() ?
SM.getInstantiationLoc(Loc) :
SourceLocation();
if (Loc.isFileID()) {
MacroStack.clear();
Pieces.push_back(&*I);
continue;
}
assert(Loc.isMacroID());
// Is the PathDiagnosticPiece within the same macro group?
if (!MacroStack.empty() && InstantiationLoc == MacroStack.back().second) {
MacroStack.back().first->push_back(&*I);
continue;
}
// We aren't in the same group. Are we descending into a new macro
// or are part of an old one?
PathDiagnosticMacroPiece *MacroGroup = 0;
SourceLocation ParentInstantiationLoc = InstantiationLoc.isMacroID() ?
SM.getInstantiationLoc(Loc) :
SourceLocation();
// Walk the entire macro stack.
while (!MacroStack.empty()) {
if (InstantiationLoc == MacroStack.back().second) {
MacroGroup = MacroStack.back().first;
break;
}
if (ParentInstantiationLoc == MacroStack.back().second) {
MacroGroup = MacroStack.back().first;
break;
}
MacroStack.pop_back();
}
if (!MacroGroup || ParentInstantiationLoc == MacroStack.back().second) {
// Create a new macro group and add it to the stack.
PathDiagnosticMacroPiece *NewGroup = new PathDiagnosticMacroPiece(Loc);
if (MacroGroup)
MacroGroup->push_back(NewGroup);
else {
assert(InstantiationLoc.isFileID());
Pieces.push_back(NewGroup);
}
MacroGroup = NewGroup;
MacroStack.push_back(std::make_pair(MacroGroup, InstantiationLoc));
}
// Finally, add the PathDiagnosticPiece to the group.
MacroGroup->push_back(&*I);
}
// Now take the pieces and construct a new PathDiagnostic.
PD.resetPath(false);
for (PiecesTy::iterator I=Pieces.begin(), E=Pieces.end(); I!=E; ++I) {
if (PathDiagnosticMacroPiece *MP=dyn_cast<PathDiagnosticMacroPiece>(*I))
if (!MP->containsEvent()) {
delete MP;
continue;
}
PD.push_back(*I);
}
}
void GRBugReporter::GeneratePathDiagnostic(PathDiagnostic& PD,
BugReportEquivClass& EQ) {
std::vector<const ExplodedNode*> Nodes;
for (BugReportEquivClass::iterator I=EQ.begin(), E=EQ.end(); I!=E; ++I) {
const ExplodedNode* N = I->getEndNode();
if (N) Nodes.push_back(N);
}
if (Nodes.empty())
return;
// Construct a new graph that contains only a single path from the error
// node to a root.
const std::pair<std::pair<ExplodedGraph*, NodeBackMap*>,
std::pair<ExplodedNode*, unsigned> >&
GPair = MakeReportGraph(&getGraph(), &Nodes[0], &Nodes[0] + Nodes.size());
// Find the BugReport with the original location.
BugReport *R = 0;
unsigned i = 0;
for (BugReportEquivClass::iterator I=EQ.begin(), E=EQ.end(); I!=E; ++I, ++i)
if (i == GPair.second.second) { R = *I; break; }
assert(R && "No original report found for sliced graph.");
llvm::OwningPtr<ExplodedGraph> ReportGraph(GPair.first.first);
llvm::OwningPtr<NodeBackMap> BackMap(GPair.first.second);
const ExplodedNode *N = GPair.second.first;
// Start building the path diagnostic...
PathDiagnosticBuilder PDB(*this, R, BackMap.get(), getPathDiagnosticClient());
if (PathDiagnosticPiece* Piece = R->getEndPath(PDB, N))
PD.push_back(Piece);
else
return;
// Register node visitors.
R->registerInitialVisitors(PDB, N);
bugreporter::registerNilReceiverVisitor(PDB);
switch (PDB.getGenerationScheme()) {
case PathDiagnosticClient::Extensive:
GenerateExtensivePathDiagnostic(PD, PDB, N);
break;
case PathDiagnosticClient::Minimal:
GenerateMinimalPathDiagnostic(PD, PDB, N);
break;
}
}
void BugReporter::Register(BugType *BT) {
BugTypes = F.Add(BugTypes, BT);
}
void BugReporter::EmitReport(BugReport* R) {
// Compute the bug report's hash to determine its equivalence class.
llvm::FoldingSetNodeID ID;
R->Profile(ID);
// Lookup the equivance class. If there isn't one, create it.
BugType& BT = R->getBugType();
Register(&BT);
void *InsertPos;
BugReportEquivClass* EQ = BT.EQClasses.FindNodeOrInsertPos(ID, InsertPos);
if (!EQ) {
EQ = new BugReportEquivClass(R);
BT.EQClasses.InsertNode(EQ, InsertPos);
}
else
EQ->AddReport(R);
}
//===----------------------------------------------------------------------===//
// Emitting reports in equivalence classes.
//===----------------------------------------------------------------------===//
namespace {
struct FRIEC_WLItem {
const ExplodedNode *N;
ExplodedNode::const_succ_iterator I, E;
FRIEC_WLItem(const ExplodedNode *n)
: N(n), I(N->succ_begin()), E(N->succ_end()) {}
};
}
static BugReport *FindReportInEquivalenceClass(BugReportEquivClass& EQ) {
BugReportEquivClass::iterator I = EQ.begin(), E = EQ.end();
assert(I != E);
BugReport *R = *I;
BugType& BT = R->getBugType();
if (!BT.isSuppressOnSink())
return R;
// For bug reports that should be suppressed when all paths are post-dominated
// by a sink node, iterate through the reports in the equivalence class
// until we find one that isn't post-dominated (if one exists). We use a
// DFS traversal of the ExplodedGraph to find a non-sink node. We could write
// this as a recursive function, but we don't want to risk blowing out the
// stack for very long paths.
for (; I != E; ++I) {
R = *I;
const ExplodedNode *N = R->getEndNode();
if (!N)
continue;
if (N->isSink()) {
assert(false &&
"BugType::isSuppressSink() should not be 'true' for sink end nodes");
return R;
}
if (N->succ_empty())
return R;
// At this point we know that 'N' is not a sink and it has at least one
// successor. Use a DFS worklist to find a non-sink end-of-path node.
typedef FRIEC_WLItem WLItem;
typedef llvm::SmallVector<WLItem, 10> DFSWorkList;
llvm::DenseMap<const ExplodedNode *, unsigned> Visited;
DFSWorkList WL;
WL.push_back(N);
Visited[N] = 1;
while (!WL.empty()) {
WLItem &WI = WL.back();
assert(!WI.N->succ_empty());
for (; WI.I != WI.E; ++WI.I) {
const ExplodedNode *Succ = *WI.I;
// End-of-path node?
if (Succ->succ_empty()) {
// If we found an end-of-path node that is not a sink, then return
// this report.
if (!Succ->isSink())
return R;
// Found a sink? Continue on to the next successor.
continue;
}
// Mark the successor as visited. If it hasn't been explored,
// enqueue it to the DFS worklist.
unsigned &mark = Visited[Succ];
if (!mark) {
mark = 1;
WL.push_back(Succ);
break;
}
}
if (&WL.back() == &WI)
WL.pop_back();
}
}
2009-09-15 11:28:00 +08:00
// If we reach here, the end nodes for all reports in the equivalence
// class are post-dominated by a sink node.
return NULL;
}
//===----------------------------------------------------------------------===//
// DiagnosticCache. This is a hack to cache analyzer diagnostics. It
// uses global state, which eventually should go elsewhere.
//===----------------------------------------------------------------------===//
namespace {
class DiagCacheItem : public llvm::FoldingSetNode {
llvm::FoldingSetNodeID ID;
public:
DiagCacheItem(BugReport *R, PathDiagnostic *PD) {
ID.AddString(R->getBugType().getName());
ID.AddString(R->getBugType().getCategory());
ID.AddString(R->getDescription());
ID.AddInteger(R->getLocation().getRawEncoding());
PD->Profile(ID);
}
void Profile(llvm::FoldingSetNodeID &id) {
id = ID;
}
llvm::FoldingSetNodeID &getID() { return ID; }
};
}
static bool IsCachedDiagnostic(BugReport *R, PathDiagnostic *PD) {
// FIXME: Eventually this diagnostic cache should reside in something
// like AnalysisManager instead of being a static variable. This is
// really unsafe in the long term.
typedef llvm::FoldingSet<DiagCacheItem> DiagnosticCache;
static DiagnosticCache DC;
void *InsertPos;
DiagCacheItem *Item = new DiagCacheItem(R, PD);
if (DC.FindNodeOrInsertPos(Item->getID(), InsertPos)) {
delete Item;
return true;
}
DC.InsertNode(Item, InsertPos);
return false;
}
void BugReporter::FlushReport(BugReportEquivClass& EQ) {
BugReport *R = FindReportInEquivalenceClass(EQ);
if (!R)
return;
PathDiagnosticClient* PD = getPathDiagnosticClient();
// FIXME: Make sure we use the 'R' for the path that was actually used.
// Probably doesn't make a difference in practice.
BugType& BT = R->getBugType();
llvm::OwningPtr<PathDiagnostic>
D(new PathDiagnostic(R->getBugType().getName(),
2009-04-30 06:05:03 +08:00
!PD || PD->useVerboseDescription()
? R->getDescription() : R->getShortDescription(),
BT.getCategory()));
GeneratePathDiagnostic(*D.get(), EQ);
if (IsCachedDiagnostic(R, D.get()))
return;
// Get the meta data.
std::pair<const char**, const char**> Meta = R->getExtraDescriptiveText();
for (const char** s = Meta.first; s != Meta.second; ++s)
D->addMeta(*s);
// Emit a summary diagnostic to the regular Diagnostics engine.
const SourceRange *Beg = 0, *End = 0;
R->getRanges(Beg, End);
Diagnostic& Diag = getDiagnostic();
FullSourceLoc L(R->getLocation(), getSourceManager());
// Search the description for '%', as that will be interpretted as a
// format character by FormatDiagnostics.
llvm::StringRef desc = R->getShortDescription();
unsigned ErrorDiag;
{
llvm::SmallString<512> TmpStr;
llvm::raw_svector_ostream Out(TmpStr);
for (llvm::StringRef::iterator I=desc.begin(), E=desc.end(); I!=E; ++I)
if (*I == '%')
Out << "%%";
else
Out << *I;
Out.flush();
ErrorDiag = Diag.getCustomDiagID(Diagnostic::Warning, TmpStr);
}
This reworks some of the Diagnostic interfaces a bit to change how diagnostics are formed. In particular, a diagnostic with all its strings and ranges is now packaged up and sent to DiagnosticClients as a DiagnosticInfo instead of as a ton of random stuff. This has the benefit of simplifying the interface, making it more extensible, and allowing us to do more checking for things like access past the end of the various arrays passed in. In addition to introducing DiagnosticInfo, this also substantially changes how Diagnostic::Report works. Instead of being passed in all of the info required to issue a diagnostic, Report now takes only the required info (a location and ID) and returns a fresh DiagnosticInfo *by value*. The caller is then free to stuff strings and ranges into the DiagnosticInfo with the << operator. When the dtor runs on the DiagnosticInfo object (which should happen at the end of the statement), the diagnostic is actually emitted with all of the accumulated information. This is a somewhat tricky dance, but it means that the accumulated DiagnosticInfo is allowed to keep pointers to other expression temporaries without those pointers getting invalidated. This is just the minimal change to get this stuff working, but this will allow us to eliminate the zillions of variant "Diag" methods scattered throughout (e.g.) sema. For example, instead of calling: Diag(BuiltinLoc, diag::err_overload_no_match, typeNames, SourceRange(BuiltinLoc, RParenLoc)); We will soon be able to just do: Diag(BuiltinLoc, diag::err_overload_no_match) << typeNames << SourceRange(BuiltinLoc, RParenLoc)); This scales better to support arbitrary types being passed in (not just strings) in a type-safe way. Go operator overloading?! llvm-svn: 59502
2008-11-18 15:04:44 +08:00
switch (End-Beg) {
This reworks some of the Diagnostic interfaces a bit to change how diagnostics are formed. In particular, a diagnostic with all its strings and ranges is now packaged up and sent to DiagnosticClients as a DiagnosticInfo instead of as a ton of random stuff. This has the benefit of simplifying the interface, making it more extensible, and allowing us to do more checking for things like access past the end of the various arrays passed in. In addition to introducing DiagnosticInfo, this also substantially changes how Diagnostic::Report works. Instead of being passed in all of the info required to issue a diagnostic, Report now takes only the required info (a location and ID) and returns a fresh DiagnosticInfo *by value*. The caller is then free to stuff strings and ranges into the DiagnosticInfo with the << operator. When the dtor runs on the DiagnosticInfo object (which should happen at the end of the statement), the diagnostic is actually emitted with all of the accumulated information. This is a somewhat tricky dance, but it means that the accumulated DiagnosticInfo is allowed to keep pointers to other expression temporaries without those pointers getting invalidated. This is just the minimal change to get this stuff working, but this will allow us to eliminate the zillions of variant "Diag" methods scattered throughout (e.g.) sema. For example, instead of calling: Diag(BuiltinLoc, diag::err_overload_no_match, typeNames, SourceRange(BuiltinLoc, RParenLoc)); We will soon be able to just do: Diag(BuiltinLoc, diag::err_overload_no_match) << typeNames << SourceRange(BuiltinLoc, RParenLoc)); This scales better to support arbitrary types being passed in (not just strings) in a type-safe way. Go operator overloading?! llvm-svn: 59502
2008-11-18 15:04:44 +08:00
default: assert(0 && "Don't handle this many ranges yet!");
case 0: Diag.Report(L, ErrorDiag); break;
case 1: Diag.Report(L, ErrorDiag) << Beg[0]; break;
case 2: Diag.Report(L, ErrorDiag) << Beg[0] << Beg[1]; break;
case 3: Diag.Report(L, ErrorDiag) << Beg[0] << Beg[1] << Beg[2]; break;
}
// Emit a full diagnostic for the path if we have a PathDiagnosticClient.
if (!PD)
return;
if (D->empty()) {
PathDiagnosticPiece* piece =
new PathDiagnosticEventPiece(L, R->getDescription());
for ( ; Beg != End; ++Beg) piece->addRange(*Beg);
D->push_back(piece);
}
PD->HandlePathDiagnostic(D.take());
}
void BugReporter::EmitBasicReport(llvm::StringRef name, llvm::StringRef str,
SourceLocation Loc,
SourceRange* RBeg, unsigned NumRanges) {
EmitBasicReport(name, "", str, Loc, RBeg, NumRanges);
}
void BugReporter::EmitBasicReport(llvm::StringRef name,
llvm::StringRef category,
llvm::StringRef str, SourceLocation Loc,
SourceRange* RBeg, unsigned NumRanges) {
// 'BT' will be owned by BugReporter as soon as we call 'EmitReport'.
BugType *BT = new BugType(name, category);
This reworks some of the Diagnostic interfaces a bit to change how diagnostics are formed. In particular, a diagnostic with all its strings and ranges is now packaged up and sent to DiagnosticClients as a DiagnosticInfo instead of as a ton of random stuff. This has the benefit of simplifying the interface, making it more extensible, and allowing us to do more checking for things like access past the end of the various arrays passed in. In addition to introducing DiagnosticInfo, this also substantially changes how Diagnostic::Report works. Instead of being passed in all of the info required to issue a diagnostic, Report now takes only the required info (a location and ID) and returns a fresh DiagnosticInfo *by value*. The caller is then free to stuff strings and ranges into the DiagnosticInfo with the << operator. When the dtor runs on the DiagnosticInfo object (which should happen at the end of the statement), the diagnostic is actually emitted with all of the accumulated information. This is a somewhat tricky dance, but it means that the accumulated DiagnosticInfo is allowed to keep pointers to other expression temporaries without those pointers getting invalidated. This is just the minimal change to get this stuff working, but this will allow us to eliminate the zillions of variant "Diag" methods scattered throughout (e.g.) sema. For example, instead of calling: Diag(BuiltinLoc, diag::err_overload_no_match, typeNames, SourceRange(BuiltinLoc, RParenLoc)); We will soon be able to just do: Diag(BuiltinLoc, diag::err_overload_no_match) << typeNames << SourceRange(BuiltinLoc, RParenLoc)); This scales better to support arbitrary types being passed in (not just strings) in a type-safe way. Go operator overloading?! llvm-svn: 59502
2008-11-18 15:04:44 +08:00
FullSourceLoc L = getContext().getFullLoc(Loc);
RangedBugReport *R = new DiagBugReport(*BT, str, L);
for ( ; NumRanges > 0 ; --NumRanges, ++RBeg) R->addRange(*RBeg);
EmitReport(R);
}