//===-- ExprPPrinter.cpp - ----------------------------------------------===// // // The KLEE Symbolic Virtual Machine // // This file is distributed under the University of Illinois Open Source // License. See LICENSE.TXT for details. // //===----------------------------------------------------------------------===// #include "klee/Expr/ExprPPrinter.h" #include "klee/Expr/Constraints.h" #include "klee/OptionCategories.h" #include "klee/util/PrintContext.h" #include "llvm/Support/CommandLine.h" #include "llvm/Support/raw_ostream.h" #include #include using namespace klee; namespace { llvm::cl::opt PCWidthAsArg( "pc-width-as-arg", llvm::cl::init(true), llvm::cl::desc( "Print the width as a separate argument, as opposed to a prefix " "to the operation (default=true)"), llvm::cl::cat(klee::ExprCat)); llvm::cl::opt PCAllWidths("pc-all-widths", llvm::cl::init(false), llvm::cl::desc("Print the width of all operations, including " "booleans (default=false)"), llvm::cl::cat(klee::ExprCat)); llvm::cl::opt PCPrefixWidth("pc-prefix-width", llvm::cl::init(true), llvm::cl::desc("Print width with 'w' prefix (default=true)"), llvm::cl::cat(klee::ExprCat)); llvm::cl::opt PCMultibyteReads("pc-multibyte-reads", llvm::cl::init(true), llvm::cl::desc("Print ReadLSB and ReadMSB expressions " "when possible (default=true)"), llvm::cl::cat(klee::ExprCat)); llvm::cl::opt PCAllConstWidths( "pc-all-const-widths", llvm::cl::init(false), llvm::cl::desc( "Always print the width of constant expressions (default=false)"), llvm::cl::cat(klee::ExprCat)); } // namespace class PPrinter : public ExprPPrinter { public: std::set usedArrays; private: std::map, unsigned> bindings; std::map updateBindings; std::set< ref > couldPrint, shouldPrint; std::set couldPrintUpdates, shouldPrintUpdates; llvm::raw_ostream &os; unsigned counter; unsigned updateCounter; bool hasScan; bool forceNoLineBreaks; std::string newline; /// shouldPrintWidth - Predicate for whether this expression should /// be printed with its width. bool shouldPrintWidth(ref e) { if (PCAllWidths) return true; return e->getWidth() != Expr::Bool; } bool isVerySimple(const ref &e) { return isa(e) || bindings.find(e)!=bindings.end(); } bool isVerySimpleUpdate(const UpdateNode *un) { return !un || updateBindings.find(un)!=updateBindings.end(); } // document me! bool isSimple(const ref &e) { if (isVerySimple(e)) { return true; } else if (const ReadExpr *re = dyn_cast(e)) { return isVerySimple(re->index) && isVerySimpleUpdate(re->updates.head); } else { Expr *ep = e.get(); for (unsigned i=0; igetNumKids(); i++) if (!isVerySimple(ep->getKid(i))) return false; return true; } } bool hasSimpleKids(const Expr *ep) { for (unsigned i=0; igetNumKids(); i++) if (!isSimple(ep->getKid(i))) return false; return true; } void scanUpdate(const UpdateNode *un) { // FIXME: This needs to be non-recursive. if (un) { if (couldPrintUpdates.insert(un).second) { scanUpdate(un->next); scan1(un->index); scan1(un->value); } else { shouldPrintUpdates.insert(un); } } } void scan1(const ref &e) { if (!isa(e)) { if (couldPrint.insert(e).second) { Expr *ep = e.get(); for (unsigned i=0; igetNumKids(); i++) scan1(ep->getKid(i)); if (const ReadExpr *re = dyn_cast(e)) { usedArrays.insert(re->updates.root); scanUpdate(re->updates.head); } } else { shouldPrint.insert(e); } } } void printUpdateList(const UpdateList &updates, PrintContext &PC) { const UpdateNode *head = updates.head; // Special case empty list. if (!head) { // FIXME: We need to do something (assert, mangle, etc.) so that printing // distinct arrays with the same name doesn't fail. PC << updates.root->name; return; } // FIXME: Explain this breaking policy. bool openedList = false, nextShouldBreak = false; unsigned outerIndent = PC.pos; unsigned middleIndent = 0; for (const UpdateNode *un = head; un; un = un->next) { // We are done if we hit the cache. std::map::iterator it = updateBindings.find(un); if (it!=updateBindings.end()) { if (openedList) PC << "] @ "; PC << "U" << it->second; return; } else if (!hasScan || shouldPrintUpdates.count(un)) { if (openedList) PC << "] @"; if (un != head) PC.breakLine(outerIndent); PC << "U" << updateCounter << ":"; updateBindings.insert(std::make_pair(un, updateCounter++)); openedList = nextShouldBreak = false; } if (!openedList) { openedList = 1; PC << '['; middleIndent = PC.pos; } else { PC << ','; printSeparator(PC, !nextShouldBreak, middleIndent); } //PC << "(="; //unsigned innerIndent = PC.pos; print(un->index, PC); //printSeparator(PC, isSimple(un->index), innerIndent); PC << "="; print(un->value, PC); //PC << ')'; nextShouldBreak = !(isa(un->index) && isa(un->value)); } if (openedList) PC << ']'; PC << " @ " << updates.root->name; } void printWidth(PrintContext &PC, ref e) { if (!shouldPrintWidth(e)) return; if (PCWidthAsArg) { PC << ' '; if (PCPrefixWidth) PC << 'w'; } PC << e->getWidth(); } bool isReadExprAtOffset(ref e, const ReadExpr *base, ref offset) { const ReadExpr *re = dyn_cast(e.get()); // right now, all Reads are byte reads but some // transformations might change this if (!re || (re->getWidth() != Expr::Int8)) return false; // Check if the index follows the stride. // FIXME: How aggressive should this be simplified. The // canonicalizing builder is probably the right choice, but this // is yet another area where we would really prefer it to be // global or else use static methods. return SubExpr::create(re->index, base->index) == offset; } /// hasOrderedReads: \arg e must be a ConcatExpr, \arg stride must /// be 1 or -1. /// /// If all children of this Concat are reads or concats of reads /// with consecutive offsets according to the given \arg stride, it /// returns the base ReadExpr according to \arg stride: first Read /// for 1 (MSB), last Read for -1 (LSB). Otherwise, it returns /// null. const ReadExpr* hasOrderedReads(ref e, int stride) { assert(e->getKind() == Expr::Concat); assert(stride == 1 || stride == -1); const ReadExpr *base = dyn_cast(e->getKid(0)); // right now, all Reads are byte reads but some // transformations might change this if (!base || base->getWidth() != Expr::Int8) return NULL; // Get stride expr in proper index width. Expr::Width idxWidth = base->index->getWidth(); ref strideExpr = ConstantExpr::alloc(stride, idxWidth); ref offset = ConstantExpr::create(0, idxWidth); e = e->getKid(1); // concat chains are unbalanced to the right while (e->getKind() == Expr::Concat) { offset = AddExpr::create(offset, strideExpr); if (!isReadExprAtOffset(e->getKid(0), base, offset)) return NULL; e = e->getKid(1); } offset = AddExpr::create(offset, strideExpr); if (!isReadExprAtOffset(e, base, offset)) return NULL; if (stride == -1) return cast(e.get()); else return base; } #if 0 /// hasAllByteReads - True iff all children are byte level reads or /// concats of byte level reads. bool hasAllByteReads(const Expr *ep) { switch (ep->kind) { Expr::Read: { // right now, all Reads are byte reads but some // transformations might change this return ep->getWidth() == Expr::Int8; } Expr::Concat: { for (unsigned i=0; igetNumKids(); ++i) { if (!hashAllByteReads(ep->getKid(i))) return false; } } default: return false; } } #endif void printRead(const ReadExpr *re, PrintContext &PC, unsigned indent) { print(re->index, PC); printSeparator(PC, isVerySimple(re->index), indent); printUpdateList(re->updates, PC); } void printExtract(const ExtractExpr *ee, PrintContext &PC, unsigned indent) { PC << ee->offset << ' '; print(ee->expr, PC); } void printExpr(const Expr *ep, PrintContext &PC, unsigned indent, bool printConstWidth=false) { bool simple = hasSimpleKids(ep); print(ep->getKid(0), PC); for (unsigned i=1; igetNumKids(); i++) { printSeparator(PC, simple, indent); print(ep->getKid(i), PC, printConstWidth); } } public: PPrinter(llvm::raw_ostream &_os) : os(_os), newline("\n") { reset(); } void setNewline(const std::string &_newline) { newline = _newline; } void setForceNoLineBreaks(bool _forceNoLineBreaks) { forceNoLineBreaks = _forceNoLineBreaks; } void reset() { counter = 0; updateCounter = 0; hasScan = false; forceNoLineBreaks = false; bindings.clear(); updateBindings.clear(); couldPrint.clear(); shouldPrint.clear(); couldPrintUpdates.clear(); shouldPrintUpdates.clear(); } void scan(const ref &e) { hasScan = true; scan1(e); } void print(const ref &e, unsigned level=0) { PrintContext PC(os); PC.pos = level; print(e, PC); } void printConst(const ref &e, PrintContext &PC, bool printWidth) { if (e->getWidth() == Expr::Bool) PC << (e->isTrue() ? "true" : "false"); else { if (PCAllConstWidths) printWidth = true; if (printWidth) PC << "(w" << e->getWidth() << " "; if (e->getWidth() <= 64) { PC << e->getZExtValue(); } else { std::string S; e->toString(S); PC << S; } if (printWidth) PC << ")"; } } void print(const ref &e, PrintContext &PC, bool printConstWidth=false) { if (ConstantExpr *CE = dyn_cast(e)) printConst(CE, PC, printConstWidth); else { std::map, unsigned>::iterator it = bindings.find(e); if (it!=bindings.end()) { PC << 'N' << it->second; } else { if (!hasScan || shouldPrint.count(e)) { PC << 'N' << counter << ':'; bindings.insert(std::make_pair(e, counter++)); } // Detect multibyte reads. // FIXME: Hrm. One problem with doing this is that we are // masking the sharing of the indices which aren't // visible. Need to think if this matters... probably not // because if they are offset reads then its either constant, // or they are (base + offset) and base will get printed with // a declaration. if (PCMultibyteReads && e->getKind() == Expr::Concat) { const ReadExpr *base = hasOrderedReads(e, -1); const bool isLSB = (base != nullptr); if (!isLSB) base = hasOrderedReads(e, 1); if (base) { PC << "(Read" << (isLSB ? "LSB" : "MSB"); printWidth(PC, e); PC << ' '; printRead(base, PC, PC.pos); PC << ')'; return; } } PC << '(' << e->getKind(); printWidth(PC, e); PC << ' '; // Indent at first argument and dispatch to appropriate print // routine for exprs which require special handling. unsigned indent = PC.pos; if (const ReadExpr *re = dyn_cast(e)) { printRead(re, PC, indent); } else if (const ExtractExpr *ee = dyn_cast(e)) { printExtract(ee, PC, indent); } else if (e->getKind() == Expr::Concat || e->getKind() == Expr::SExt) printExpr(e.get(), PC, indent, true); else printExpr(e.get(), PC, indent); PC << ")"; } } } /* Public utility functions */ void printSeparator(PrintContext &PC, bool simple, unsigned indent) { if (simple || forceNoLineBreaks) { PC << ' '; } else { PC.breakLine(indent); } } }; ExprPPrinter *klee::ExprPPrinter::create(llvm::raw_ostream &os) { return new PPrinter(os); } void ExprPPrinter::printOne(llvm::raw_ostream &os, const char *message, const ref &e) { PPrinter p(os); p.scan(e); // FIXME: Need to figure out what to do here. Probably print as a // "forward declaration" with whatever syntax we pick for that. PrintContext PC(os); PC << message << ": "; p.print(e, PC); PC.breakLine(); } void ExprPPrinter::printSingleExpr(llvm::raw_ostream &os, const ref &e) { PPrinter p(os); p.scan(e); // FIXME: Need to figure out what to do here. Probably print as a // "forward declaration" with whatever syntax we pick for that. PrintContext PC(os); p.print(e, PC); } void ExprPPrinter::printConstraints(llvm::raw_ostream &os, const ConstraintManager &constraints) { printQuery(os, constraints, ConstantExpr::alloc(false, Expr::Bool)); } namespace { struct ArrayPtrsByName { bool operator()(const Array *a1, const Array *a2) const { return a1->name < a2->name; } }; } void ExprPPrinter::printQuery(llvm::raw_ostream &os, const ConstraintManager &constraints, const ref &q, const ref *evalExprsBegin, const ref *evalExprsEnd, const Array * const *evalArraysBegin, const Array * const *evalArraysEnd, bool printArrayDecls) { PPrinter p(os); for (ConstraintManager::const_iterator it = constraints.begin(), ie = constraints.end(); it != ie; ++it) p.scan(*it); p.scan(q); for (const ref *it = evalExprsBegin; it != evalExprsEnd; ++it) p.scan(*it); PrintContext PC(os); // Print array declarations. if (printArrayDecls) { for (const Array * const* it = evalArraysBegin; it != evalArraysEnd; ++it) p.usedArrays.insert(*it); std::vector sortedArray(p.usedArrays.begin(), p.usedArrays.end()); std::sort(sortedArray.begin(), sortedArray.end(), ArrayPtrsByName()); for (std::vector::iterator it = sortedArray.begin(), ie = sortedArray.end(); it != ie; ++it) { const Array *A = *it; PC << "array " << A->name << "[" << A->size << "]" << " : w" << A->domain << " -> w" << A->range << " = "; if (A->isSymbolicArray()) { PC << "symbolic"; } else { PC << "["; for (unsigned i = 0, e = A->size; i != e; ++i) { if (i) PC << " "; PC << A->constantValues[i]; } PC << "]"; } PC.breakLine(); } } PC << "(query ["; // Ident at constraint list; unsigned indent = PC.pos; for (ConstraintManager::const_iterator it = constraints.begin(), ie = constraints.end(); it != ie;) { p.print(*it, PC); ++it; if (it != ie) PC.breakLine(indent); } PC << ']'; p.printSeparator(PC, constraints.empty(), indent-1); p.print(q, PC); // Print expressions to obtain values for, if any. if (evalExprsBegin != evalExprsEnd) { p.printSeparator(PC, q->isFalse(), indent-1); PC << '['; for (const ref *it = evalExprsBegin; it != evalExprsEnd; ++it) { p.print(*it, PC, /*printConstWidth*/true); if (it + 1 != evalExprsEnd) PC.breakLine(indent); } PC << ']'; } // Print arrays to obtain values for, if any. if (evalArraysBegin != evalArraysEnd) { if (evalExprsBegin == evalExprsEnd) PC << " []"; PC.breakLine(indent - 1); PC << '['; for (const Array * const* it = evalArraysBegin; it != evalArraysEnd; ++it) { PC << (*it)->name; if (it + 1 != evalArraysEnd) PC.breakLine(indent); } PC << ']'; } PC << ')'; PC.breakLine(); }