llvm-6502/lib/Bitcode/Reader/BitstreamReader.cpp
Filipe Cabecinhas 5d8bb5c7c5 [Bitcode] Diagnose errors instead of asserting from bad input
Eventually we can make some of these pass the error along to the caller.

Reports a fatal error if:
We find an invalid abbrev record
We try to get an invalid abbrev number
We can't fill the current word due to an EOF

Fixed an invalid bitcode test to check for output with FileCheck

Bugs found with afl-fuzz

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@226986 91177308-0d34-0410-b5e6-96231b3b80d8
2015-01-24 04:15:05 +00:00

342 lines
11 KiB
C++

//===- BitstreamReader.cpp - BitstreamReader implementation ---------------===//
//
// The LLVM Compiler Infrastructure
//
// This file is distributed under the University of Illinois Open Source
// License. See LICENSE.TXT for details.
//
//===----------------------------------------------------------------------===//
#include "llvm/Bitcode/BitstreamReader.h"
using namespace llvm;
//===----------------------------------------------------------------------===//
// BitstreamCursor implementation
//===----------------------------------------------------------------------===//
void BitstreamCursor::freeState() {
// Free all the Abbrevs.
CurAbbrevs.clear();
// Free all the Abbrevs in the block scope.
BlockScope.clear();
}
/// EnterSubBlock - Having read the ENTER_SUBBLOCK abbrevid, enter
/// the block, and return true if the block has an error.
bool BitstreamCursor::EnterSubBlock(unsigned BlockID, unsigned *NumWordsP) {
// Save the current block's state on BlockScope.
BlockScope.push_back(Block(CurCodeSize));
BlockScope.back().PrevAbbrevs.swap(CurAbbrevs);
// Add the abbrevs specific to this block to the CurAbbrevs list.
if (const BitstreamReader::BlockInfo *Info =
BitStream->getBlockInfo(BlockID)) {
CurAbbrevs.insert(CurAbbrevs.end(), Info->Abbrevs.begin(),
Info->Abbrevs.end());
}
// Get the codesize of this block.
CurCodeSize = ReadVBR(bitc::CodeLenWidth);
SkipToFourByteBoundary();
unsigned NumWords = Read(bitc::BlockSizeWidth);
if (NumWordsP) *NumWordsP = NumWords;
// Validate that this block is sane.
if (CurCodeSize == 0 || AtEndOfStream())
return true;
return false;
}
static uint64_t readAbbreviatedField(BitstreamCursor &Cursor,
const BitCodeAbbrevOp &Op) {
assert(!Op.isLiteral() && "Not to be used with literals!");
// Decode the value as we are commanded.
switch (Op.getEncoding()) {
case BitCodeAbbrevOp::Array:
case BitCodeAbbrevOp::Blob:
llvm_unreachable("Should not reach here");
case BitCodeAbbrevOp::Fixed:
return Cursor.Read((unsigned)Op.getEncodingData());
case BitCodeAbbrevOp::VBR:
return Cursor.ReadVBR64((unsigned)Op.getEncodingData());
case BitCodeAbbrevOp::Char6:
return BitCodeAbbrevOp::DecodeChar6(Cursor.Read(6));
}
llvm_unreachable("invalid abbreviation encoding");
}
static void skipAbbreviatedField(BitstreamCursor &Cursor,
const BitCodeAbbrevOp &Op) {
assert(!Op.isLiteral() && "Not to be used with literals!");
// Decode the value as we are commanded.
switch (Op.getEncoding()) {
case BitCodeAbbrevOp::Array:
case BitCodeAbbrevOp::Blob:
llvm_unreachable("Should not reach here");
case BitCodeAbbrevOp::Fixed:
Cursor.Read((unsigned)Op.getEncodingData());
break;
case BitCodeAbbrevOp::VBR:
Cursor.ReadVBR64((unsigned)Op.getEncodingData());
break;
case BitCodeAbbrevOp::Char6:
Cursor.Read(6);
break;
}
}
/// skipRecord - Read the current record and discard it.
void BitstreamCursor::skipRecord(unsigned AbbrevID) {
// Skip unabbreviated records by reading past their entries.
if (AbbrevID == bitc::UNABBREV_RECORD) {
unsigned Code = ReadVBR(6);
(void)Code;
unsigned NumElts = ReadVBR(6);
for (unsigned i = 0; i != NumElts; ++i)
(void)ReadVBR64(6);
return;
}
const BitCodeAbbrev *Abbv = getAbbrev(AbbrevID);
for (unsigned i = 0, e = Abbv->getNumOperandInfos(); i != e; ++i) {
const BitCodeAbbrevOp &Op = Abbv->getOperandInfo(i);
if (Op.isLiteral())
continue;
if (Op.getEncoding() != BitCodeAbbrevOp::Array &&
Op.getEncoding() != BitCodeAbbrevOp::Blob) {
skipAbbreviatedField(*this, Op);
continue;
}
if (Op.getEncoding() == BitCodeAbbrevOp::Array) {
// Array case. Read the number of elements as a vbr6.
unsigned NumElts = ReadVBR(6);
// Get the element encoding.
assert(i+2 == e && "array op not second to last?");
const BitCodeAbbrevOp &EltEnc = Abbv->getOperandInfo(++i);
// Read all the elements.
for (; NumElts; --NumElts)
skipAbbreviatedField(*this, EltEnc);
continue;
}
assert(Op.getEncoding() == BitCodeAbbrevOp::Blob);
// Blob case. Read the number of bytes as a vbr6.
unsigned NumElts = ReadVBR(6);
SkipToFourByteBoundary(); // 32-bit alignment
// Figure out where the end of this blob will be including tail padding.
size_t NewEnd = GetCurrentBitNo()+((NumElts+3)&~3)*8;
// If this would read off the end of the bitcode file, just set the
// record to empty and return.
if (!canSkipToPos(NewEnd/8)) {
NextChar = BitStream->getBitcodeBytes().getExtent();
break;
}
// Skip over the blob.
JumpToBit(NewEnd);
}
}
unsigned BitstreamCursor::readRecord(unsigned AbbrevID,
SmallVectorImpl<uint64_t> &Vals,
StringRef *Blob) {
if (AbbrevID == bitc::UNABBREV_RECORD) {
unsigned Code = ReadVBR(6);
unsigned NumElts = ReadVBR(6);
for (unsigned i = 0; i != NumElts; ++i)
Vals.push_back(ReadVBR64(6));
return Code;
}
const BitCodeAbbrev *Abbv = getAbbrev(AbbrevID);
// Read the record code first.
assert(Abbv->getNumOperandInfos() != 0 && "no record code in abbreviation?");
const BitCodeAbbrevOp &CodeOp = Abbv->getOperandInfo(0);
unsigned Code;
if (CodeOp.isLiteral())
Code = CodeOp.getLiteralValue();
else {
if (CodeOp.getEncoding() == BitCodeAbbrevOp::Array ||
CodeOp.getEncoding() == BitCodeAbbrevOp::Blob)
report_fatal_error("Abbreviation starts with an Array or a Blob");
Code = readAbbreviatedField(*this, CodeOp);
}
for (unsigned i = 1, e = Abbv->getNumOperandInfos(); i != e; ++i) {
const BitCodeAbbrevOp &Op = Abbv->getOperandInfo(i);
if (Op.isLiteral()) {
Vals.push_back(Op.getLiteralValue());
continue;
}
if (Op.getEncoding() != BitCodeAbbrevOp::Array &&
Op.getEncoding() != BitCodeAbbrevOp::Blob) {
Vals.push_back(readAbbreviatedField(*this, Op));
continue;
}
if (Op.getEncoding() == BitCodeAbbrevOp::Array) {
// Array case. Read the number of elements as a vbr6.
unsigned NumElts = ReadVBR(6);
// Get the element encoding.
assert(i+2 == e && "array op not second to last?");
const BitCodeAbbrevOp &EltEnc = Abbv->getOperandInfo(++i);
// Read all the elements.
for (; NumElts; --NumElts)
Vals.push_back(readAbbreviatedField(*this, EltEnc));
continue;
}
assert(Op.getEncoding() == BitCodeAbbrevOp::Blob);
// Blob case. Read the number of bytes as a vbr6.
unsigned NumElts = ReadVBR(6);
SkipToFourByteBoundary(); // 32-bit alignment
// Figure out where the end of this blob will be including tail padding.
size_t CurBitPos = GetCurrentBitNo();
size_t NewEnd = CurBitPos+((NumElts+3)&~3)*8;
// If this would read off the end of the bitcode file, just set the
// record to empty and return.
if (!canSkipToPos(NewEnd/8)) {
Vals.append(NumElts, 0);
NextChar = BitStream->getBitcodeBytes().getExtent();
break;
}
// Otherwise, inform the streamer that we need these bytes in memory.
const char *Ptr = (const char*)
BitStream->getBitcodeBytes().getPointer(CurBitPos/8, NumElts);
// If we can return a reference to the data, do so to avoid copying it.
if (Blob) {
*Blob = StringRef(Ptr, NumElts);
} else {
// Otherwise, unpack into Vals with zero extension.
for (; NumElts; --NumElts)
Vals.push_back((unsigned char)*Ptr++);
}
// Skip over tail padding.
JumpToBit(NewEnd);
}
return Code;
}
void BitstreamCursor::ReadAbbrevRecord() {
BitCodeAbbrev *Abbv = new BitCodeAbbrev();
unsigned NumOpInfo = ReadVBR(5);
for (unsigned i = 0; i != NumOpInfo; ++i) {
bool IsLiteral = Read(1) ? true : false;
if (IsLiteral) {
Abbv->Add(BitCodeAbbrevOp(ReadVBR64(8)));
continue;
}
BitCodeAbbrevOp::Encoding E = (BitCodeAbbrevOp::Encoding)Read(3);
if (BitCodeAbbrevOp::hasEncodingData(E)) {
unsigned Data = ReadVBR64(5);
// As a special case, handle fixed(0) (i.e., a fixed field with zero bits)
// and vbr(0) as a literal zero. This is decoded the same way, and avoids
// a slow path in Read() to have to handle reading zero bits.
if ((E == BitCodeAbbrevOp::Fixed || E == BitCodeAbbrevOp::VBR) &&
Data == 0) {
Abbv->Add(BitCodeAbbrevOp(0));
continue;
}
Abbv->Add(BitCodeAbbrevOp(E, Data));
} else
Abbv->Add(BitCodeAbbrevOp(E));
}
CurAbbrevs.push_back(Abbv);
}
bool BitstreamCursor::ReadBlockInfoBlock() {
// If this is the second stream to get to the block info block, skip it.
if (BitStream->hasBlockInfoRecords())
return SkipBlock();
if (EnterSubBlock(bitc::BLOCKINFO_BLOCK_ID)) return true;
SmallVector<uint64_t, 64> Record;
BitstreamReader::BlockInfo *CurBlockInfo = nullptr;
// Read all the records for this module.
while (1) {
BitstreamEntry Entry = advanceSkippingSubblocks(AF_DontAutoprocessAbbrevs);
switch (Entry.Kind) {
case llvm::BitstreamEntry::SubBlock: // Handled for us already.
case llvm::BitstreamEntry::Error:
return true;
case llvm::BitstreamEntry::EndBlock:
return false;
case llvm::BitstreamEntry::Record:
// The interesting case.
break;
}
// Read abbrev records, associate them with CurBID.
if (Entry.ID == bitc::DEFINE_ABBREV) {
if (!CurBlockInfo) return true;
ReadAbbrevRecord();
// ReadAbbrevRecord installs the abbrev in CurAbbrevs. Move it to the
// appropriate BlockInfo.
CurBlockInfo->Abbrevs.push_back(std::move(CurAbbrevs.back()));
CurAbbrevs.pop_back();
continue;
}
// Read a record.
Record.clear();
switch (readRecord(Entry.ID, Record)) {
default: break; // Default behavior, ignore unknown content.
case bitc::BLOCKINFO_CODE_SETBID:
if (Record.size() < 1) return true;
CurBlockInfo = &BitStream->getOrCreateBlockInfo((unsigned)Record[0]);
break;
case bitc::BLOCKINFO_CODE_BLOCKNAME: {
if (!CurBlockInfo) return true;
if (BitStream->isIgnoringBlockInfoNames()) break; // Ignore name.
std::string Name;
for (unsigned i = 0, e = Record.size(); i != e; ++i)
Name += (char)Record[i];
CurBlockInfo->Name = Name;
break;
}
case bitc::BLOCKINFO_CODE_SETRECORDNAME: {
if (!CurBlockInfo) return true;
if (BitStream->isIgnoringBlockInfoNames()) break; // Ignore name.
std::string Name;
for (unsigned i = 1, e = Record.size(); i != e; ++i)
Name += (char)Record[i];
CurBlockInfo->RecordNames.push_back(std::make_pair((unsigned)Record[0],
Name));
break;
}
}
}
}