2009-12-22 22:50:29 +00:00
|
|
|
//===- Disassembler.cpp - Disassembler for hex strings --------------------===//
|
2009-12-17 01:49:59 +00:00
|
|
|
//
|
|
|
|
// The LLVM Compiler Infrastructure
|
|
|
|
//
|
|
|
|
// This file is distributed under the University of Illinois Open Source
|
|
|
|
// License. See LICENSE.TXT for details.
|
|
|
|
//
|
|
|
|
//===----------------------------------------------------------------------===//
|
|
|
|
//
|
|
|
|
// This class implements the disassembler of strings of bytes written in
|
|
|
|
// hexadecimal, from standard input or from a file.
|
|
|
|
//
|
|
|
|
//===----------------------------------------------------------------------===//
|
|
|
|
|
2009-12-22 22:50:29 +00:00
|
|
|
#include "Disassembler.h"
|
2009-12-17 01:49:59 +00:00
|
|
|
|
|
|
|
#include "llvm/ADT/OwningPtr.h"
|
2010-04-12 19:43:00 +00:00
|
|
|
#include "llvm/ADT/Triple.h"
|
2009-12-17 01:49:59 +00:00
|
|
|
#include "llvm/MC/MCAsmInfo.h"
|
|
|
|
#include "llvm/MC/MCDisassembler.h"
|
|
|
|
#include "llvm/MC/MCInst.h"
|
|
|
|
#include "llvm/MC/MCInstPrinter.h"
|
|
|
|
#include "llvm/Target/TargetRegistry.h"
|
|
|
|
#include "llvm/Support/MemoryBuffer.h"
|
|
|
|
#include "llvm/Support/MemoryObject.h"
|
|
|
|
#include "llvm/Support/raw_ostream.h"
|
2009-12-22 06:45:48 +00:00
|
|
|
#include "llvm/Support/SourceMgr.h"
|
2010-04-12 19:43:00 +00:00
|
|
|
|
|
|
|
#include "llvm-c/EnhancedDisassembly.h"
|
|
|
|
|
2009-12-17 01:49:59 +00:00
|
|
|
using namespace llvm;
|
|
|
|
|
2009-12-22 06:56:51 +00:00
|
|
|
typedef std::vector<std::pair<unsigned char, const char*> > ByteArrayTy;
|
|
|
|
|
|
|
|
namespace {
|
2009-12-17 01:49:59 +00:00
|
|
|
class VectorMemoryObject : public MemoryObject {
|
|
|
|
private:
|
2009-12-22 06:56:51 +00:00
|
|
|
const ByteArrayTy &Bytes;
|
2009-12-17 01:49:59 +00:00
|
|
|
public:
|
2009-12-22 06:56:51 +00:00
|
|
|
VectorMemoryObject(const ByteArrayTy &bytes) : Bytes(bytes) {}
|
2009-12-17 01:49:59 +00:00
|
|
|
|
2009-12-22 06:56:51 +00:00
|
|
|
uint64_t getBase() const { return 0; }
|
|
|
|
uint64_t getExtent() const { return Bytes.size(); }
|
2009-12-17 01:49:59 +00:00
|
|
|
|
2009-12-22 06:56:51 +00:00
|
|
|
int readByte(uint64_t Addr, uint8_t *Byte) const {
|
|
|
|
if (Addr > getExtent())
|
2009-12-17 01:49:59 +00:00
|
|
|
return -1;
|
2009-12-22 06:56:51 +00:00
|
|
|
*Byte = Bytes[Addr].first;
|
2009-12-17 01:49:59 +00:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
};
|
2009-12-22 06:56:51 +00:00
|
|
|
}
|
2009-12-17 01:49:59 +00:00
|
|
|
|
2010-03-20 22:36:35 +00:00
|
|
|
static bool PrintInsts(const MCDisassembler &DisAsm,
|
2010-04-04 05:04:31 +00:00
|
|
|
MCInstPrinter &Printer, const ByteArrayTy &Bytes,
|
|
|
|
SourceMgr &SM) {
|
2009-12-17 01:49:59 +00:00
|
|
|
// Wrap the vector in a MemoryObject.
|
2009-12-22 06:56:51 +00:00
|
|
|
VectorMemoryObject memoryObject(Bytes);
|
2009-12-17 01:49:59 +00:00
|
|
|
|
2010-02-03 03:46:41 +00:00
|
|
|
// Disassemble it to strings.
|
2009-12-22 06:56:51 +00:00
|
|
|
uint64_t Size;
|
2010-02-03 03:46:41 +00:00
|
|
|
uint64_t Index;
|
2009-12-17 01:49:59 +00:00
|
|
|
|
2010-02-03 03:46:41 +00:00
|
|
|
for (Index = 0; Index < Bytes.size(); Index += Size) {
|
|
|
|
MCInst Inst;
|
|
|
|
|
|
|
|
if (DisAsm.getInstruction(Inst, Size, memoryObject, Index,
|
|
|
|
/*REMOVE*/ nulls())) {
|
2010-04-04 05:04:31 +00:00
|
|
|
Printer.printInst(&Inst, outs());
|
2010-02-03 03:46:41 +00:00
|
|
|
outs() << "\n";
|
2010-04-12 19:43:00 +00:00
|
|
|
} else {
|
2010-02-03 03:46:41 +00:00
|
|
|
SM.PrintMessage(SMLoc::getFromPointer(Bytes[Index].second),
|
|
|
|
"invalid instruction encoding", "warning");
|
|
|
|
if (Size == 0)
|
|
|
|
Size = 1; // skip illegible bytes
|
|
|
|
}
|
2009-12-17 01:49:59 +00:00
|
|
|
}
|
2009-12-22 06:56:51 +00:00
|
|
|
|
|
|
|
return false;
|
2009-12-17 01:49:59 +00:00
|
|
|
}
|
|
|
|
|
2010-04-12 19:43:00 +00:00
|
|
|
static bool ByteArrayFromString(ByteArrayTy &ByteArray,
|
|
|
|
StringRef &Str,
|
|
|
|
SourceMgr &SM) {
|
2009-12-22 06:24:00 +00:00
|
|
|
while (!Str.empty()) {
|
2009-12-22 06:37:58 +00:00
|
|
|
// Strip horizontal whitespace.
|
|
|
|
if (size_t Pos = Str.find_first_not_of(" \t\r")) {
|
|
|
|
Str = Str.substr(Pos);
|
|
|
|
continue;
|
|
|
|
}
|
|
|
|
|
2010-02-03 03:46:41 +00:00
|
|
|
// If this is the end of a line or start of a comment, remove the rest of
|
|
|
|
// the line.
|
2009-12-22 06:37:58 +00:00
|
|
|
if (Str[0] == '\n' || Str[0] == '#') {
|
|
|
|
// Strip to the end of line if we already processed any bytes on this
|
|
|
|
// line. This strips the comment and/or the \n.
|
2010-04-12 19:43:00 +00:00
|
|
|
if (Str[0] == '\n') {
|
2009-12-22 06:37:58 +00:00
|
|
|
Str = Str.substr(1);
|
2010-04-12 19:43:00 +00:00
|
|
|
} else {
|
2009-12-22 06:37:58 +00:00
|
|
|
Str = Str.substr(Str.find_first_of('\n'));
|
|
|
|
if (!Str.empty())
|
|
|
|
Str = Str.substr(1);
|
|
|
|
}
|
|
|
|
continue;
|
2009-12-17 01:49:59 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
// Get the current token.
|
2009-12-22 06:37:58 +00:00
|
|
|
size_t Next = Str.find_first_of(" \t\n\r#");
|
|
|
|
StringRef Value = Str.substr(0, Next);
|
2009-12-17 01:49:59 +00:00
|
|
|
|
|
|
|
// Convert to a byte and add to the byte vector.
|
2009-12-22 06:24:00 +00:00
|
|
|
unsigned ByteVal;
|
|
|
|
if (Value.getAsInteger(0, ByteVal) || ByteVal > 255) {
|
2009-12-22 06:45:48 +00:00
|
|
|
// If we have an error, print it and skip to the end of line.
|
2009-12-22 06:56:51 +00:00
|
|
|
SM.PrintMessage(SMLoc::getFromPointer(Value.data()),
|
2010-04-12 19:43:00 +00:00
|
|
|
"invalid input token", "error");
|
2009-12-22 06:45:48 +00:00
|
|
|
Str = Str.substr(Str.find('\n'));
|
|
|
|
ByteArray.clear();
|
|
|
|
continue;
|
2009-12-17 01:49:59 +00:00
|
|
|
}
|
2009-12-22 06:45:48 +00:00
|
|
|
|
2009-12-22 06:56:51 +00:00
|
|
|
ByteArray.push_back(std::make_pair((unsigned char)ByteVal, Value.data()));
|
2009-12-22 06:24:00 +00:00
|
|
|
Str = Str.substr(Next);
|
2009-12-17 01:49:59 +00:00
|
|
|
}
|
|
|
|
|
2010-04-12 19:43:00 +00:00
|
|
|
return false;
|
|
|
|
}
|
|
|
|
|
|
|
|
int Disassembler::disassemble(const Target &T, const std::string &Triple,
|
|
|
|
MemoryBuffer &Buffer) {
|
|
|
|
// Set up disassembler.
|
|
|
|
OwningPtr<const MCAsmInfo> AsmInfo(T.createAsmInfo(Triple));
|
|
|
|
|
|
|
|
if (!AsmInfo) {
|
|
|
|
errs() << "error: no assembly info for target " << Triple << "\n";
|
|
|
|
return -1;
|
|
|
|
}
|
|
|
|
|
|
|
|
OwningPtr<const MCDisassembler> DisAsm(T.createMCDisassembler());
|
|
|
|
if (!DisAsm) {
|
|
|
|
errs() << "error: no disassembler for target " << Triple << "\n";
|
|
|
|
return -1;
|
|
|
|
}
|
|
|
|
|
2010-04-13 18:41:17 +00:00
|
|
|
int AsmPrinterVariant = AsmInfo->getAssemblerDialect();
|
|
|
|
OwningPtr<MCInstPrinter> IP(T.createMCInstPrinter(AsmPrinterVariant,
|
|
|
|
*AsmInfo));
|
2010-04-12 19:43:00 +00:00
|
|
|
if (!IP) {
|
|
|
|
errs() << "error: no instruction printer for target " << Triple << '\n';
|
|
|
|
return -1;
|
|
|
|
}
|
|
|
|
|
|
|
|
bool ErrorOccurred = false;
|
|
|
|
|
|
|
|
SourceMgr SM;
|
|
|
|
SM.AddNewSourceBuffer(&Buffer, SMLoc());
|
|
|
|
|
|
|
|
// Convert the input to a vector for disassembly.
|
|
|
|
ByteArrayTy ByteArray;
|
|
|
|
StringRef Str = Buffer.getBuffer();
|
|
|
|
|
|
|
|
ErrorOccurred |= ByteArrayFromString(ByteArray, Str, SM);
|
|
|
|
|
2009-12-22 06:24:00 +00:00
|
|
|
if (!ByteArray.empty())
|
2010-03-20 22:36:35 +00:00
|
|
|
ErrorOccurred |= PrintInsts(*DisAsm, *IP, ByteArray, SM);
|
2009-12-17 01:49:59 +00:00
|
|
|
|
2009-12-22 06:56:51 +00:00
|
|
|
return ErrorOccurred;
|
2009-12-17 01:49:59 +00:00
|
|
|
}
|
2010-04-12 19:43:00 +00:00
|
|
|
|
|
|
|
static int byteArrayReader(uint8_t *B, uint64_t A, void *Arg) {
|
|
|
|
ByteArrayTy &ByteArray = *((ByteArrayTy*)Arg);
|
|
|
|
|
|
|
|
if (A >= ByteArray.size())
|
|
|
|
return -1;
|
|
|
|
|
|
|
|
*B = ByteArray[A].first;
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int verboseEvaluator(uint64_t *V, unsigned R, void *Arg) {
|
|
|
|
EDDisassemblerRef &disassembler = *((EDDisassemblerRef*)Arg);
|
|
|
|
|
|
|
|
const char *regName;
|
|
|
|
|
|
|
|
if (!EDGetRegisterName(®Name,
|
|
|
|
disassembler,
|
|
|
|
R))
|
|
|
|
outs() << "[" << regName << "/" << R << "]";
|
|
|
|
if (EDRegisterIsStackPointer(disassembler, R))
|
|
|
|
outs() << "(sp)";
|
|
|
|
if (EDRegisterIsProgramCounter(disassembler, R))
|
|
|
|
outs() << "(pc)";
|
|
|
|
|
|
|
|
*V = 0;
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
int Disassembler::disassembleEnhanced(const std::string &TS,
|
|
|
|
MemoryBuffer &Buffer) {
|
|
|
|
ByteArrayTy ByteArray;
|
|
|
|
StringRef Str = Buffer.getBuffer();
|
|
|
|
SourceMgr SM;
|
|
|
|
|
|
|
|
SM.AddNewSourceBuffer(&Buffer, SMLoc());
|
|
|
|
|
|
|
|
if (ByteArrayFromString(ByteArray, Str, SM)) {
|
|
|
|
return -1;
|
|
|
|
}
|
|
|
|
|
|
|
|
EDDisassemblerRef disassembler;
|
|
|
|
|
|
|
|
Triple T(TS);
|
|
|
|
EDAssemblySyntax_t AS;
|
|
|
|
|
|
|
|
switch (T.getArch()) {
|
|
|
|
default:
|
|
|
|
errs() << "error: no default assembly syntax for " << TS.c_str() << "\n";
|
|
|
|
return -1;
|
|
|
|
case Triple::arm:
|
|
|
|
case Triple::thumb:
|
|
|
|
AS = kEDAssemblySyntaxARMUAL;
|
|
|
|
break;
|
|
|
|
case Triple::x86:
|
|
|
|
case Triple::x86_64:
|
|
|
|
AS = kEDAssemblySyntaxX86ATT;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (EDGetDisassembler(&disassembler,
|
|
|
|
TS.c_str(),
|
|
|
|
AS)) {
|
|
|
|
errs() << "error: couldn't get disassembler for " << TS.c_str() << "\n";
|
|
|
|
return -1;
|
|
|
|
}
|
|
|
|
|
|
|
|
EDInstRef inst;
|
|
|
|
|
|
|
|
if (EDCreateInsts(&inst, 1, disassembler, byteArrayReader, 0,&ByteArray)
|
|
|
|
!= 1) {
|
|
|
|
errs() << "error: Didn't get an instruction\n";
|
|
|
|
return -1;
|
|
|
|
}
|
|
|
|
|
|
|
|
int numTokens = EDNumTokens(inst);
|
|
|
|
|
|
|
|
if (numTokens < 0) {
|
|
|
|
errs() << "error: Couldn't count the instruction's tokens\n";
|
|
|
|
return -1;
|
|
|
|
}
|
|
|
|
|
|
|
|
int tokenIndex;
|
|
|
|
|
|
|
|
for (tokenIndex = 0; tokenIndex < numTokens; ++tokenIndex) {
|
|
|
|
EDTokenRef token;
|
|
|
|
|
|
|
|
if (EDGetToken(&token, inst, tokenIndex)) {
|
|
|
|
errs() << "error: Couldn't get token\n";
|
|
|
|
return -1;
|
|
|
|
}
|
|
|
|
|
|
|
|
const char *buf;
|
|
|
|
|
|
|
|
if (EDGetTokenString(&buf, token)) {
|
|
|
|
errs() << "error: Couldn't get string for token\n";
|
|
|
|
return -1;
|
|
|
|
}
|
|
|
|
|
|
|
|
outs() << "[";
|
|
|
|
|
|
|
|
int operandIndex = EDOperandIndexForToken(token);
|
|
|
|
|
|
|
|
if (operandIndex >= 0)
|
|
|
|
outs() << operandIndex << "-";
|
|
|
|
|
|
|
|
if (EDTokenIsWhitespace(token)) {
|
|
|
|
outs() << "w";
|
|
|
|
} else if (EDTokenIsPunctuation(token)) {
|
|
|
|
outs() << "p";
|
|
|
|
} else if (EDTokenIsOpcode(token)) {
|
|
|
|
outs() << "o";
|
|
|
|
} else if (EDTokenIsLiteral(token)) {
|
|
|
|
outs() << "l";
|
|
|
|
} else if (EDTokenIsRegister(token)) {
|
|
|
|
outs() << "r";
|
|
|
|
} else {
|
|
|
|
outs() << "?";
|
|
|
|
}
|
|
|
|
|
|
|
|
outs() << ":" << buf;
|
|
|
|
|
|
|
|
if (EDTokenIsLiteral(token)) {
|
|
|
|
outs() << "=";
|
|
|
|
if (EDTokenIsNegativeLiteral(token))
|
|
|
|
outs() << "-";
|
|
|
|
uint64_t absoluteValue;
|
|
|
|
if (EDLiteralTokenAbsoluteValue(&absoluteValue, token)) {
|
|
|
|
errs() << "error: Couldn't get the value of a literal token\n";
|
|
|
|
return -1;
|
|
|
|
}
|
|
|
|
outs() << absoluteValue;
|
|
|
|
} else if (EDTokenIsRegister(token)) {
|
|
|
|
outs() << "=";
|
|
|
|
unsigned regID;
|
|
|
|
if (EDRegisterTokenValue(®ID, token)) {
|
|
|
|
errs() << "error: Couldn't get the ID of a register token\n";
|
|
|
|
return -1;
|
|
|
|
}
|
|
|
|
outs() << "r" << regID;
|
|
|
|
}
|
|
|
|
|
|
|
|
outs() << "]";
|
|
|
|
}
|
|
|
|
|
|
|
|
outs() << " ";
|
|
|
|
|
|
|
|
int numOperands = EDNumOperands(inst);
|
|
|
|
|
|
|
|
if (numOperands < 0) {
|
|
|
|
errs() << "error: Couldn't count operands\n";
|
|
|
|
return -1;
|
|
|
|
}
|
|
|
|
|
|
|
|
int operandIndex;
|
|
|
|
|
|
|
|
for (operandIndex = 0; operandIndex < numOperands; ++operandIndex) {
|
|
|
|
outs() << operandIndex << ":";
|
|
|
|
|
|
|
|
EDOperandRef operand;
|
|
|
|
|
|
|
|
if (EDGetOperand(&operand,
|
|
|
|
inst,
|
|
|
|
operandIndex)) {
|
|
|
|
errs() << "error: Couldn't get operand\n";
|
|
|
|
return -1;
|
|
|
|
}
|
|
|
|
|
|
|
|
uint64_t evaluatedResult;
|
|
|
|
|
|
|
|
EDEvaluateOperand(&evaluatedResult,
|
|
|
|
operand,
|
|
|
|
verboseEvaluator,
|
|
|
|
&disassembler);
|
|
|
|
|
|
|
|
outs() << " ";
|
|
|
|
}
|
|
|
|
|
|
|
|
outs() << "\n";
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|