blob: 0a487e3f3f80a464ff1a82e7dcdca41bd725e128 [file] [log] [blame]
/*
* Copyright (C) 2011 The Android Open Source Project
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
#include "method_verifier.h"
#include <iostream>
#include "class_linker.h"
#include "compiler.h"
#include "dex_cache.h"
#include "dex_file.h"
#include "dex_instruction.h"
#include "dex_instruction_visitor.h"
#include "gc_map.h"
#include "intern_table.h"
#include "leb128.h"
#include "logging.h"
#include "object_utils.h"
#include "runtime.h"
#include "stringpiece.h"
#if defined(ART_USE_LLVM_COMPILER)
#include "compiler_llvm/backend_types.h"
#include "compiler_llvm/inferred_reg_category_map.h"
using namespace art::compiler_llvm;
#endif
#if defined(ART_USE_GREENLAND_COMPILER)
#include "greenland/backend_types.h"
#include "greenland/inferred_reg_category_map.h"
using namespace art::greenland;
#endif
namespace art {
namespace verifier {
static const bool gDebugVerify = false;
class InsnFlags {
public:
InsnFlags() : length_(0), flags_(0) {}
void SetLengthInCodeUnits(size_t length) {
CHECK_LT(length, 65536u);
length_ = length;
}
size_t GetLengthInCodeUnits() {
return length_;
}
bool IsOpcode() const {
return length_ != 0;
}
void SetInTry() {
flags_ |= 1 << kInTry;
}
void ClearInTry() {
flags_ &= ~(1 << kInTry);
}
bool IsInTry() const {
return (flags_ & (1 << kInTry)) != 0;
}
void SetBranchTarget() {
flags_ |= 1 << kBranchTarget;
}
void ClearBranchTarget() {
flags_ &= ~(1 << kBranchTarget);
}
bool IsBranchTarget() const {
return (flags_ & (1 << kBranchTarget)) != 0;
}
void SetGcPoint() {
flags_ |= 1 << kGcPoint;
}
void ClearGcPoint() {
flags_ &= ~(1 << kGcPoint);
}
bool IsGcPoint() const {
return (flags_ & (1 << kGcPoint)) != 0;
}
void SetVisited() {
flags_ |= 1 << kVisited;
}
void ClearVisited() {
flags_ &= ~(1 << kVisited);
}
bool IsVisited() const {
return (flags_ & (1 << kVisited)) != 0;
}
void SetChanged() {
flags_ |= 1 << kChanged;
}
void ClearChanged() {
flags_ &= ~(1 << kChanged);
}
bool IsChanged() const {
return (flags_ & (1 << kChanged)) != 0;
}
bool IsVisitedOrChanged() const {
return IsVisited() || IsChanged();
}
std::string Dump() {
char encoding[6];
if (!IsOpcode()) {
strncpy(encoding, "XXXXX", sizeof(encoding));
} else {
strncpy(encoding, "-----", sizeof(encoding));
if (IsInTry()) encoding[kInTry] = 'T';
if (IsBranchTarget()) encoding[kBranchTarget] = 'B';
if (IsGcPoint()) encoding[kGcPoint] = 'G';
if (IsVisited()) encoding[kVisited] = 'V';
if (IsChanged()) encoding[kChanged] = 'C';
}
return std::string(encoding);
}
private:
enum {
kInTry,
kBranchTarget,
kGcPoint,
kVisited,
kChanged,
};
// Size of instruction in code units
uint16_t length_;
uint8_t flags_;
};
void PcToRegisterLineTable::Init(RegisterTrackingMode mode, InsnFlags* flags,
uint32_t insns_size, uint16_t registers_size,
MethodVerifier* verifier) {
DCHECK_GT(insns_size, 0U);
for (uint32_t i = 0; i < insns_size; i++) {
bool interesting = false;
switch (mode) {
case kTrackRegsAll:
interesting = flags[i].IsOpcode();
break;
case kTrackRegsGcPoints:
interesting = flags[i].IsGcPoint() || flags[i].IsBranchTarget();
break;
case kTrackRegsBranches:
interesting = flags[i].IsBranchTarget();
break;
default:
break;
}
if (interesting) {
pc_to_register_line_.Put(i, new RegisterLine(registers_size, verifier));
}
}
}
MethodVerifier::FailureKind MethodVerifier::VerifyClass(const Class* klass, std::string& error) {
if (klass->IsVerified()) {
return kNoFailure;
}
Class* super = klass->GetSuperClass();
if (super == NULL && StringPiece(ClassHelper(klass).GetDescriptor()) != "Ljava/lang/Object;") {
error = "Verifier rejected class ";
error += PrettyDescriptor(klass);
error += " that has no super class";
return kHardFailure;
}
if (super != NULL && super->IsFinal()) {
error = "Verifier rejected class ";
error += PrettyDescriptor(klass);
error += " that attempts to sub-class final class ";
error += PrettyDescriptor(super);
return kHardFailure;
}
ClassHelper kh(klass);
const DexFile& dex_file = kh.GetDexFile();
uint32_t class_def_idx;
if (!dex_file.FindClassDefIndex(kh.GetDescriptor(), class_def_idx)) {
error = "Verifier rejected class ";
error += PrettyDescriptor(klass);
error += " that isn't present in dex file ";
error += dex_file.GetLocation();
return kHardFailure;
}
return VerifyClass(&dex_file, kh.GetDexCache(), klass->GetClassLoader(), class_def_idx, error);
}
MethodVerifier::FailureKind MethodVerifier::VerifyClass(const DexFile* dex_file,
DexCache* dex_cache, ClassLoader* class_loader, uint32_t class_def_idx, std::string& error) {
const DexFile::ClassDef& class_def = dex_file->GetClassDef(class_def_idx);
const byte* class_data = dex_file->GetClassData(class_def);
if (class_data == NULL) {
// empty class, probably a marker interface
return kNoFailure;
}
ClassDataItemIterator it(*dex_file, class_data);
while (it.HasNextStaticField() || it.HasNextInstanceField()) {
it.Next();
}
size_t error_count = 0;
bool hard_fail = false;
ClassLinker* linker = Runtime::Current()->GetClassLinker();
while (it.HasNextDirectMethod()) {
uint32_t method_idx = it.GetMemberIndex();
Method* method = linker->ResolveMethod(*dex_file, method_idx, dex_cache, class_loader, true);
if (method == NULL) {
DCHECK(Thread::Current()->IsExceptionPending());
// We couldn't resolve the method, but continue regardless.
Thread::Current()->ClearException();
}
MethodVerifier::FailureKind result = VerifyMethod(method_idx, dex_file, dex_cache, class_loader,
class_def_idx, it.GetMethodCodeItem(), method, it.GetMemberAccessFlags());
if (result != kNoFailure) {
if (result == kHardFailure) {
hard_fail = true;
if (error_count > 0) {
error += "\n";
}
error = "Verifier rejected class ";
error += PrettyDescriptor(dex_file->GetClassDescriptor(class_def));
error += " due to bad method ";
error += PrettyMethod(method_idx, *dex_file);
}
++error_count;
}
it.Next();
}
while (it.HasNextVirtualMethod()) {
uint32_t method_idx = it.GetMemberIndex();
Method* method = linker->ResolveMethod(*dex_file, method_idx, dex_cache, class_loader, false);
if (method == NULL) {
DCHECK(Thread::Current()->IsExceptionPending());
// We couldn't resolve the method, but continue regardless.
Thread::Current()->ClearException();
}
MethodVerifier::FailureKind result = VerifyMethod(method_idx, dex_file, dex_cache, class_loader,
class_def_idx, it.GetMethodCodeItem(), method, it.GetMemberAccessFlags());
if (result != kNoFailure) {
if (result == kHardFailure) {
hard_fail = true;
if (error_count > 0) {
error += "\n";
}
error = "Verifier rejected class ";
error += PrettyDescriptor(dex_file->GetClassDescriptor(class_def));
error += " due to bad method ";
error += PrettyMethod(method_idx, *dex_file);
}
++error_count;
}
it.Next();
}
if (error_count == 0) {
return kNoFailure;
} else {
return hard_fail ? kHardFailure : kSoftFailure;
}
}
MethodVerifier::FailureKind MethodVerifier::VerifyMethod(uint32_t method_idx, const DexFile* dex_file,
DexCache* dex_cache, ClassLoader* class_loader, uint32_t class_def_idx,
const DexFile::CodeItem* code_item, Method* method, uint32_t method_access_flags) {
MethodVerifier verifier(dex_file, dex_cache, class_loader, class_def_idx, code_item, method_idx,
method, method_access_flags);
if (verifier.Verify()) {
// Verification completed, however failures may be pending that didn't cause the verification
// to hard fail.
CHECK(!verifier.have_pending_hard_failure_);
if (verifier.failures_.size() != 0) {
verifier.DumpFailures(LOG(INFO) << "Soft verification failures in "
<< PrettyMethod(method_idx, *dex_file) << "\n");
return kSoftFailure;
}
} else {
// Bad method data.
CHECK_NE(verifier.failures_.size(), 0U);
CHECK(verifier.have_pending_hard_failure_);
verifier.DumpFailures(LOG(INFO) << "Verification error in "
<< PrettyMethod(method_idx, *dex_file) << "\n");
if (gDebugVerify) {
std::cout << "\n" << verifier.info_messages_.str();
verifier.Dump(std::cout);
}
return kHardFailure;
}
return kNoFailure;
}
void MethodVerifier::VerifyMethodAndDump(Method* method) {
CHECK(method != NULL);
MethodHelper mh(method);
MethodVerifier verifier(&mh.GetDexFile(), mh.GetDexCache(), mh.GetClassLoader(),
mh.GetClassDefIndex(), mh.GetCodeItem(), method->GetDexMethodIndex(),
method, method->GetAccessFlags());
verifier.Verify();
verifier.DumpFailures(LOG(INFO) << "Dump of method " << PrettyMethod(method) << "\n")
<< verifier.info_messages_.str() << Dumpable<MethodVerifier>(verifier);
}
MethodVerifier::MethodVerifier(const DexFile* dex_file, DexCache* dex_cache,
ClassLoader* class_loader, uint32_t class_def_idx, const DexFile::CodeItem* code_item,
uint32_t method_idx, Method* method, uint32_t method_access_flags)
: work_insn_idx_(-1),
method_idx_(method_idx),
foo_method_(method),
method_access_flags_(method_access_flags),
dex_file_(dex_file),
dex_cache_(dex_cache),
class_loader_(class_loader),
class_def_idx_(class_def_idx),
code_item_(code_item),
interesting_dex_pc_(-1),
monitor_enter_dex_pcs_(NULL),
have_pending_hard_failure_(false),
have_pending_rewrite_failure_(false),
new_instance_count_(0),
monitor_enter_count_(0) {
}
void MethodVerifier::FindLocksAtDexPc(Method* m, uint32_t dex_pc, std::vector<uint32_t>& monitor_enter_dex_pcs) {
MethodHelper mh(m);
MethodVerifier verifier(&mh.GetDexFile(), mh.GetDexCache(), mh.GetClassLoader(),
mh.GetClassDefIndex(), mh.GetCodeItem(), m->GetDexMethodIndex(),
m, m->GetAccessFlags());
verifier.interesting_dex_pc_ = dex_pc;
verifier.monitor_enter_dex_pcs_ = &monitor_enter_dex_pcs;
verifier.FindLocksAtDexPc();
}
void MethodVerifier::FindLocksAtDexPc() {
CHECK(monitor_enter_dex_pcs_ != NULL);
CHECK(code_item_ != NULL); // This only makes sense for methods with code.
// Strictly speaking, we ought to be able to get away with doing a subset of the full method
// verification. In practice, the phase we want relies on data structures set up by all the
// earlier passes, so we just run the full method verification and bail out early when we've
// got what we wanted.
Verify();
}
bool MethodVerifier::Verify() {
// If there aren't any instructions, make sure that's expected, then exit successfully.
if (code_item_ == NULL) {
if ((method_access_flags_ & (kAccNative | kAccAbstract)) == 0) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "zero-length code in concrete non-native method";
return false;
} else {
return true;
}
}
// Sanity-check the register counts. ins + locals = registers, so make sure that ins <= registers.
if (code_item_->ins_size_ > code_item_->registers_size_) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "bad register counts (ins=" << code_item_->ins_size_
<< " regs=" << code_item_->registers_size_;
return false;
}
// Allocate and initialize an array to hold instruction data.
insn_flags_.reset(new InsnFlags[code_item_->insns_size_in_code_units_]());
// Run through the instructions and see if the width checks out.
bool result = ComputeWidthsAndCountOps();
// Flag instructions guarded by a "try" block and check exception handlers.
result = result && ScanTryCatchBlocks();
// Perform static instruction verification.
result = result && VerifyInstructions();
// Perform code-flow analysis and return.
return result && VerifyCodeFlow();
}
std::ostream& MethodVerifier::Fail(VerifyError error) {
switch (error) {
case VERIFY_ERROR_NO_CLASS:
case VERIFY_ERROR_NO_FIELD:
case VERIFY_ERROR_NO_METHOD:
case VERIFY_ERROR_ACCESS_CLASS:
case VERIFY_ERROR_ACCESS_FIELD:
case VERIFY_ERROR_ACCESS_METHOD:
if (Runtime::Current()->IsCompiler()) {
// If we're optimistically running verification at compile time, turn NO_xxx and ACCESS_xxx
// errors into soft verification errors so that we re-verify at runtime. We may fail to find
// or to agree on access because of not yet available class loaders, or class loaders that
// will differ at runtime.
error = VERIFY_ERROR_BAD_CLASS_SOFT;
} else {
have_pending_rewrite_failure_ = true;
}
break;
// Errors that are bad at both compile and runtime, but don't cause rejection of the class.
case VERIFY_ERROR_CLASS_CHANGE:
case VERIFY_ERROR_INSTANTIATION:
have_pending_rewrite_failure_ = true;
break;
// Indication that verification should be retried at runtime.
case VERIFY_ERROR_BAD_CLASS_SOFT:
if (!Runtime::Current()->IsCompiler()) {
// It is runtime so hard fail.
have_pending_hard_failure_ = true;
}
break;
// Hard verification failures at compile time will still fail at runtime, so the class is
// marked as rejected to prevent it from being compiled.
case VERIFY_ERROR_BAD_CLASS_HARD: {
if (Runtime::Current()->IsCompiler()) {
Compiler::ClassReference ref(dex_file_, class_def_idx_);
AddRejectedClass(ref);
}
have_pending_hard_failure_ = true;
break;
}
}
failures_.push_back(error);
std::string location(StringPrintf("%s: [0x%X]", PrettyMethod(method_idx_, *dex_file_).c_str(),
work_insn_idx_));
std::ostringstream* failure_message = new std::ostringstream(location);
failure_messages_.push_back(failure_message);
return *failure_message;
}
void MethodVerifier::PrependToLastFailMessage(std::string prepend) {
size_t failure_num = failure_messages_.size();
DCHECK_NE(failure_num, 0U);
std::ostringstream* last_fail_message = failure_messages_[failure_num - 1];
prepend += last_fail_message->str();
failure_messages_[failure_num - 1] = new std::ostringstream(prepend);
delete last_fail_message;
}
void MethodVerifier::AppendToLastFailMessage(std::string append) {
size_t failure_num = failure_messages_.size();
DCHECK_NE(failure_num, 0U);
std::ostringstream* last_fail_message = failure_messages_[failure_num - 1];
(*last_fail_message) << append;
}
bool MethodVerifier::ComputeWidthsAndCountOps() {
const uint16_t* insns = code_item_->insns_;
size_t insns_size = code_item_->insns_size_in_code_units_;
const Instruction* inst = Instruction::At(insns);
size_t new_instance_count = 0;
size_t monitor_enter_count = 0;
size_t dex_pc = 0;
while (dex_pc < insns_size) {
Instruction::Code opcode = inst->Opcode();
if (opcode == Instruction::NEW_INSTANCE) {
new_instance_count++;
} else if (opcode == Instruction::MONITOR_ENTER) {
monitor_enter_count++;
}
size_t inst_size = inst->SizeInCodeUnits();
insn_flags_[dex_pc].SetLengthInCodeUnits(inst_size);
dex_pc += inst_size;
inst = inst->Next();
}
if (dex_pc != insns_size) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "code did not end where expected ("
<< dex_pc << " vs. " << insns_size << ")";
return false;
}
new_instance_count_ = new_instance_count;
monitor_enter_count_ = monitor_enter_count;
return true;
}
bool MethodVerifier::ScanTryCatchBlocks() {
uint32_t tries_size = code_item_->tries_size_;
if (tries_size == 0) {
return true;
}
uint32_t insns_size = code_item_->insns_size_in_code_units_;
const DexFile::TryItem* tries = DexFile::GetTryItems(*code_item_, 0);
for (uint32_t idx = 0; idx < tries_size; idx++) {
const DexFile::TryItem* try_item = &tries[idx];
uint32_t start = try_item->start_addr_;
uint32_t end = start + try_item->insn_count_;
if ((start >= end) || (start >= insns_size) || (end > insns_size)) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "bad exception entry: startAddr=" << start
<< " endAddr=" << end << " (size=" << insns_size << ")";
return false;
}
if (!insn_flags_[start].IsOpcode()) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "'try' block starts inside an instruction (" << start << ")";
return false;
}
for (uint32_t dex_pc = start; dex_pc < end;
dex_pc += insn_flags_[dex_pc].GetLengthInCodeUnits()) {
insn_flags_[dex_pc].SetInTry();
}
}
// Iterate over each of the handlers to verify target addresses.
const byte* handlers_ptr = DexFile::GetCatchHandlerData(*code_item_, 0);
uint32_t handlers_size = DecodeUnsignedLeb128(&handlers_ptr);
ClassLinker* linker = Runtime::Current()->GetClassLinker();
for (uint32_t idx = 0; idx < handlers_size; idx++) {
CatchHandlerIterator iterator(handlers_ptr);
for (; iterator.HasNext(); iterator.Next()) {
uint32_t dex_pc= iterator.GetHandlerAddress();
if (!insn_flags_[dex_pc].IsOpcode()) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "exception handler starts at bad address (" << dex_pc << ")";
return false;
}
const Instruction* inst = Instruction::At(code_item_->insns_ + dex_pc);
if (inst->Opcode() != Instruction::MOVE_EXCEPTION) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "exception handler doesn't start with move-exception ("
<< dex_pc << ")";
return false;
}
insn_flags_[dex_pc].SetBranchTarget();
// Ensure exception types are resolved so that they don't need resolution to be delivered,
// unresolved exception types will be ignored by exception delivery
if (iterator.GetHandlerTypeIndex() != DexFile::kDexNoIndex16) {
Class* exception_type = linker->ResolveType(*dex_file_, iterator.GetHandlerTypeIndex(),
dex_cache_, class_loader_);
if (exception_type == NULL) {
DCHECK(Thread::Current()->IsExceptionPending());
Thread::Current()->ClearException();
}
}
}
handlers_ptr = iterator.EndDataPointer();
}
return true;
}
bool MethodVerifier::VerifyInstructions() {
const Instruction* inst = Instruction::At(code_item_->insns_);
/* Flag the start of the method as a branch target. */
insn_flags_[0].SetBranchTarget();
uint32_t insns_size = code_item_->insns_size_in_code_units_;
for (uint32_t dex_pc = 0; dex_pc < insns_size;) {
if (!VerifyInstruction(inst, dex_pc)) {
DCHECK_NE(failures_.size(), 0U);
return false;
}
/* Flag instructions that are garbage collection points */
if (inst->IsBranch() || inst->IsSwitch() || inst->IsThrow() || inst->IsReturn()) {
insn_flags_[dex_pc].SetGcPoint();
}
dex_pc += inst->SizeInCodeUnits();
inst = inst->Next();
}
return true;
}
bool MethodVerifier::VerifyInstruction(const Instruction* inst, uint32_t code_offset) {
DecodedInstruction dec_insn(inst);
bool result = true;
switch (inst->GetVerifyTypeArgumentA()) {
case Instruction::kVerifyRegA:
result = result && CheckRegisterIndex(dec_insn.vA);
break;
case Instruction::kVerifyRegAWide:
result = result && CheckWideRegisterIndex(dec_insn.vA);
break;
}
switch (inst->GetVerifyTypeArgumentB()) {
case Instruction::kVerifyRegB:
result = result && CheckRegisterIndex(dec_insn.vB);
break;
case Instruction::kVerifyRegBField:
result = result && CheckFieldIndex(dec_insn.vB);
break;
case Instruction::kVerifyRegBMethod:
result = result && CheckMethodIndex(dec_insn.vB);
break;
case Instruction::kVerifyRegBNewInstance:
result = result && CheckNewInstance(dec_insn.vB);
break;
case Instruction::kVerifyRegBString:
result = result && CheckStringIndex(dec_insn.vB);
break;
case Instruction::kVerifyRegBType:
result = result && CheckTypeIndex(dec_insn.vB);
break;
case Instruction::kVerifyRegBWide:
result = result && CheckWideRegisterIndex(dec_insn.vB);
break;
}
switch (inst->GetVerifyTypeArgumentC()) {
case Instruction::kVerifyRegC:
result = result && CheckRegisterIndex(dec_insn.vC);
break;
case Instruction::kVerifyRegCField:
result = result && CheckFieldIndex(dec_insn.vC);
break;
case Instruction::kVerifyRegCNewArray:
result = result && CheckNewArray(dec_insn.vC);
break;
case Instruction::kVerifyRegCType:
result = result && CheckTypeIndex(dec_insn.vC);
break;
case Instruction::kVerifyRegCWide:
result = result && CheckWideRegisterIndex(dec_insn.vC);
break;
}
switch (inst->GetVerifyExtraFlags()) {
case Instruction::kVerifyArrayData:
result = result && CheckArrayData(code_offset);
break;
case Instruction::kVerifyBranchTarget:
result = result && CheckBranchTarget(code_offset);
break;
case Instruction::kVerifySwitchTargets:
result = result && CheckSwitchTargets(code_offset);
break;
case Instruction::kVerifyVarArg:
result = result && CheckVarArgRegs(dec_insn.vA, dec_insn.arg);
break;
case Instruction::kVerifyVarArgRange:
result = result && CheckVarArgRangeRegs(dec_insn.vA, dec_insn.vC);
break;
case Instruction::kVerifyError:
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "unexpected opcode " << inst->Name();
result = false;
break;
}
return result;
}
bool MethodVerifier::CheckRegisterIndex(uint32_t idx) {
if (idx >= code_item_->registers_size_) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "register index out of range (" << idx << " >= "
<< code_item_->registers_size_ << ")";
return false;
}
return true;
}
bool MethodVerifier::CheckWideRegisterIndex(uint32_t idx) {
if (idx + 1 >= code_item_->registers_size_) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "wide register index out of range (" << idx
<< "+1 >= " << code_item_->registers_size_ << ")";
return false;
}
return true;
}
bool MethodVerifier::CheckFieldIndex(uint32_t idx) {
if (idx >= dex_file_->GetHeader().field_ids_size_) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "bad field index " << idx << " (max "
<< dex_file_->GetHeader().field_ids_size_ << ")";
return false;
}
return true;
}
bool MethodVerifier::CheckMethodIndex(uint32_t idx) {
if (idx >= dex_file_->GetHeader().method_ids_size_) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "bad method index " << idx << " (max "
<< dex_file_->GetHeader().method_ids_size_ << ")";
return false;
}
return true;
}
bool MethodVerifier::CheckNewInstance(uint32_t idx) {
if (idx >= dex_file_->GetHeader().type_ids_size_) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "bad type index " << idx << " (max "
<< dex_file_->GetHeader().type_ids_size_ << ")";
return false;
}
// We don't need the actual class, just a pointer to the class name.
const char* descriptor = dex_file_->StringByTypeIdx(idx);
if (descriptor[0] != 'L') {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "can't call new-instance on type '" << descriptor << "'";
return false;
}
return true;
}
bool MethodVerifier::CheckStringIndex(uint32_t idx) {
if (idx >= dex_file_->GetHeader().string_ids_size_) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "bad string index " << idx << " (max "
<< dex_file_->GetHeader().string_ids_size_ << ")";
return false;
}
return true;
}
bool MethodVerifier::CheckTypeIndex(uint32_t idx) {
if (idx >= dex_file_->GetHeader().type_ids_size_) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "bad type index " << idx << " (max "
<< dex_file_->GetHeader().type_ids_size_ << ")";
return false;
}
return true;
}
bool MethodVerifier::CheckNewArray(uint32_t idx) {
if (idx >= dex_file_->GetHeader().type_ids_size_) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "bad type index " << idx << " (max "
<< dex_file_->GetHeader().type_ids_size_ << ")";
return false;
}
int bracket_count = 0;
const char* descriptor = dex_file_->StringByTypeIdx(idx);
const char* cp = descriptor;
while (*cp++ == '[') {
bracket_count++;
}
if (bracket_count == 0) {
/* The given class must be an array type. */
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "can't new-array class '" << descriptor << "' (not an array)";
return false;
} else if (bracket_count > 255) {
/* It is illegal to create an array of more than 255 dimensions. */
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "can't new-array class '" << descriptor << "' (exceeds limit)";
return false;
}
return true;
}
bool MethodVerifier::CheckArrayData(uint32_t cur_offset) {
const uint32_t insn_count = code_item_->insns_size_in_code_units_;
const uint16_t* insns = code_item_->insns_ + cur_offset;
const uint16_t* array_data;
int32_t array_data_offset;
DCHECK_LT(cur_offset, insn_count);
/* make sure the start of the array data table is in range */
array_data_offset = insns[1] | (((int32_t) insns[2]) << 16);
if ((int32_t) cur_offset + array_data_offset < 0 ||
cur_offset + array_data_offset + 2 >= insn_count) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "invalid array data start: at " << cur_offset
<< ", data offset " << array_data_offset << ", count " << insn_count;
return false;
}
/* offset to array data table is a relative branch-style offset */
array_data = insns + array_data_offset;
/* make sure the table is 32-bit aligned */
if ((((uint32_t) array_data) & 0x03) != 0) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "unaligned array data table: at " << cur_offset
<< ", data offset " << array_data_offset;
return false;
}
uint32_t value_width = array_data[1];
uint32_t value_count = *reinterpret_cast<const uint32_t*>(&array_data[2]);
uint32_t table_size = 4 + (value_width * value_count + 1) / 2;
/* make sure the end of the switch is in range */
if (cur_offset + array_data_offset + table_size > insn_count) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "invalid array data end: at " << cur_offset
<< ", data offset " << array_data_offset << ", end "
<< cur_offset + array_data_offset + table_size
<< ", count " << insn_count;
return false;
}
return true;
}
bool MethodVerifier::CheckBranchTarget(uint32_t cur_offset) {
int32_t offset;
bool isConditional, selfOkay;
if (!GetBranchOffset(cur_offset, &offset, &isConditional, &selfOkay)) {
return false;
}
if (!selfOkay && offset == 0) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "branch offset of zero not allowed at" << reinterpret_cast<void*>(cur_offset);
return false;
}
// Check for 32-bit overflow. This isn't strictly necessary if we can depend on the runtime
// to have identical "wrap-around" behavior, but it's unwise to depend on that.
if (((int64_t) cur_offset + (int64_t) offset) != (int64_t) (cur_offset + offset)) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "branch target overflow " << reinterpret_cast<void*>(cur_offset) << " +" << offset;
return false;
}
const uint32_t insn_count = code_item_->insns_size_in_code_units_;
int32_t abs_offset = cur_offset + offset;
if (abs_offset < 0 || (uint32_t) abs_offset >= insn_count || !insn_flags_[abs_offset].IsOpcode()) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "invalid branch target " << offset << " (-> "
<< reinterpret_cast<void*>(abs_offset) << ") at "
<< reinterpret_cast<void*>(cur_offset);
return false;
}
insn_flags_[abs_offset].SetBranchTarget();
return true;
}
bool MethodVerifier::GetBranchOffset(uint32_t cur_offset, int32_t* pOffset, bool* pConditional,
bool* selfOkay) {
const uint16_t* insns = code_item_->insns_ + cur_offset;
*pConditional = false;
*selfOkay = false;
switch (*insns & 0xff) {
case Instruction::GOTO:
*pOffset = ((int16_t) *insns) >> 8;
break;
case Instruction::GOTO_32:
*pOffset = insns[1] | (((uint32_t) insns[2]) << 16);
*selfOkay = true;
break;
case Instruction::GOTO_16:
*pOffset = (int16_t) insns[1];
break;
case Instruction::IF_EQ:
case Instruction::IF_NE:
case Instruction::IF_LT:
case Instruction::IF_GE:
case Instruction::IF_GT:
case Instruction::IF_LE:
case Instruction::IF_EQZ:
case Instruction::IF_NEZ:
case Instruction::IF_LTZ:
case Instruction::IF_GEZ:
case Instruction::IF_GTZ:
case Instruction::IF_LEZ:
*pOffset = (int16_t) insns[1];
*pConditional = true;
break;
default:
return false;
break;
}
return true;
}
bool MethodVerifier::CheckSwitchTargets(uint32_t cur_offset) {
const uint32_t insn_count = code_item_->insns_size_in_code_units_;
DCHECK_LT(cur_offset, insn_count);
const uint16_t* insns = code_item_->insns_ + cur_offset;
/* make sure the start of the switch is in range */
int32_t switch_offset = insns[1] | ((int32_t) insns[2]) << 16;
if ((int32_t) cur_offset + switch_offset < 0 || cur_offset + switch_offset + 2 >= insn_count) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "invalid switch start: at " << cur_offset
<< ", switch offset " << switch_offset << ", count " << insn_count;
return false;
}
/* offset to switch table is a relative branch-style offset */
const uint16_t* switch_insns = insns + switch_offset;
/* make sure the table is 32-bit aligned */
if ((((uint32_t) switch_insns) & 0x03) != 0) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "unaligned switch table: at " << cur_offset
<< ", switch offset " << switch_offset;
return false;
}
uint32_t switch_count = switch_insns[1];
int32_t keys_offset, targets_offset;
uint16_t expected_signature;
if ((*insns & 0xff) == Instruction::PACKED_SWITCH) {
/* 0=sig, 1=count, 2/3=firstKey */
targets_offset = 4;
keys_offset = -1;
expected_signature = Instruction::kPackedSwitchSignature;
} else {
/* 0=sig, 1=count, 2..count*2 = keys */
keys_offset = 2;
targets_offset = 2 + 2 * switch_count;
expected_signature = Instruction::kSparseSwitchSignature;
}
uint32_t table_size = targets_offset + switch_count * 2;
if (switch_insns[0] != expected_signature) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << StringPrintf("wrong signature for switch table (%x, wanted %x)",
switch_insns[0], expected_signature);
return false;
}
/* make sure the end of the switch is in range */
if (cur_offset + switch_offset + table_size > (uint32_t) insn_count) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "invalid switch end: at " << cur_offset << ", switch offset "
<< switch_offset << ", end "
<< (cur_offset + switch_offset + table_size)
<< ", count " << insn_count;
return false;
}
/* for a sparse switch, verify the keys are in ascending order */
if (keys_offset > 0 && switch_count > 1) {
int32_t last_key = switch_insns[keys_offset] | (switch_insns[keys_offset + 1] << 16);
for (uint32_t targ = 1; targ < switch_count; targ++) {
int32_t key = (int32_t) switch_insns[keys_offset + targ * 2] |
(int32_t) (switch_insns[keys_offset + targ * 2 + 1] << 16);
if (key <= last_key) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "invalid packed switch: last key=" << last_key
<< ", this=" << key;
return false;
}
last_key = key;
}
}
/* verify each switch target */
for (uint32_t targ = 0; targ < switch_count; targ++) {
int32_t offset = (int32_t) switch_insns[targets_offset + targ * 2] |
(int32_t) (switch_insns[targets_offset + targ * 2 + 1] << 16);
int32_t abs_offset = cur_offset + offset;
if (abs_offset < 0 || abs_offset >= (int32_t) insn_count || !insn_flags_[abs_offset].IsOpcode()) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "invalid switch target " << offset << " (-> "
<< reinterpret_cast<void*>(abs_offset) << ") at "
<< reinterpret_cast<void*>(cur_offset) << "[" << targ << "]";
return false;
}
insn_flags_[abs_offset].SetBranchTarget();
}
return true;
}
bool MethodVerifier::CheckVarArgRegs(uint32_t vA, uint32_t arg[]) {
if (vA > 5) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "invalid arg count (" << vA << ") in non-range invoke)";
return false;
}
uint16_t registers_size = code_item_->registers_size_;
for (uint32_t idx = 0; idx < vA; idx++) {
if (arg[idx] >= registers_size) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "invalid reg index (" << arg[idx]
<< ") in non-range invoke (>= " << registers_size << ")";
return false;
}
}
return true;
}
bool MethodVerifier::CheckVarArgRangeRegs(uint32_t vA, uint32_t vC) {
uint16_t registers_size = code_item_->registers_size_;
// vA/vC are unsigned 8-bit/16-bit quantities for /range instructions, so there's no risk of
// integer overflow when adding them here.
if (vA + vC > registers_size) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "invalid reg index " << vA << "+" << vC << " in range invoke (> "
<< registers_size << ")";
return false;
}
return true;
}
const std::vector<uint8_t>* CreateLengthPrefixedGcMap(const std::vector<uint8_t>& gc_map) {
std::vector<uint8_t>* length_prefixed_gc_map = new std::vector<uint8_t>;
length_prefixed_gc_map->push_back((gc_map.size() & 0xff000000) >> 24);
length_prefixed_gc_map->push_back((gc_map.size() & 0x00ff0000) >> 16);
length_prefixed_gc_map->push_back((gc_map.size() & 0x0000ff00) >> 8);
length_prefixed_gc_map->push_back((gc_map.size() & 0x000000ff) >> 0);
length_prefixed_gc_map->insert(length_prefixed_gc_map->end(),
gc_map.begin(),
gc_map.end());
DCHECK_EQ(gc_map.size() + 4, length_prefixed_gc_map->size());
DCHECK_EQ(gc_map.size(),
static_cast<size_t>((length_prefixed_gc_map->at(0) << 24) |
(length_prefixed_gc_map->at(1) << 16) |
(length_prefixed_gc_map->at(2) << 8) |
(length_prefixed_gc_map->at(3) << 0)));
return length_prefixed_gc_map;
}
bool MethodVerifier::VerifyCodeFlow() {
uint16_t registers_size = code_item_->registers_size_;
uint32_t insns_size = code_item_->insns_size_in_code_units_;
if (registers_size * insns_size > 4*1024*1024) {
LOG(WARNING) << "warning: method is huge (regs=" << registers_size
<< " insns_size=" << insns_size << ")";
}
/* Create and initialize table holding register status */
reg_table_.Init(kTrackRegsGcPoints, insn_flags_.get(), insns_size, registers_size, this);
work_line_.reset(new RegisterLine(registers_size, this));
saved_line_.reset(new RegisterLine(registers_size, this));
/* Initialize register types of method arguments. */
if (!SetTypesFromSignature()) {
DCHECK_NE(failures_.size(), 0U);
std::string prepend("Bad signature in ");
prepend += PrettyMethod(method_idx_, *dex_file_);
PrependToLastFailMessage(prepend);
return false;
}
/* Perform code flow verification. */
if (!CodeFlowVerifyMethod()) {
DCHECK_NE(failures_.size(), 0U);
return false;
}
Compiler::MethodReference ref(dex_file_, method_idx_);
#if !defined(ART_USE_LLVM_COMPILER) && !defined(ART_USE_GREENLAND_COMPILER)
/* Generate a register map and add it to the method. */
UniquePtr<const std::vector<uint8_t> > map(GenerateGcMap());
if (map.get() == NULL) {
DCHECK_NE(failures_.size(), 0U);
return false; // Not a real failure, but a failure to encode
}
#ifndef NDEBUG
VerifyGcMap(*map);
#endif
const std::vector<uint8_t>* gc_map = CreateLengthPrefixedGcMap(*(map.get()));
verifier::MethodVerifier::SetGcMap(ref, *gc_map);
if (foo_method_ != NULL) {
foo_method_->SetGcMap(&gc_map->at(0));
}
#else // defined(ART_USE_LLVM_COMPILER) || defined(ART_USE_GREENLAND_COMPILER)
/* Generate Inferred Register Category for LLVM-based Code Generator */
const InferredRegCategoryMap* table = GenerateInferredRegCategoryMap();
verifier::MethodVerifier::SetInferredRegCategoryMap(ref, *table);
#endif
return true;
}
std::ostream& MethodVerifier::DumpFailures(std::ostream& os) {
DCHECK_EQ(failures_.size(), failure_messages_.size());
for (size_t i = 0; i < failures_.size(); ++i) {
os << failure_messages_[i]->str() << "\n";
}
return os;
}
extern "C" void MethodVerifierGdbDump(MethodVerifier* v) {
v->Dump(std::cerr);
}
void MethodVerifier::Dump(std::ostream& os) {
if (code_item_ == NULL) {
os << "Native method\n";
return;
}
DCHECK(code_item_ != NULL);
const Instruction* inst = Instruction::At(code_item_->insns_);
for (size_t dex_pc = 0; dex_pc < code_item_->insns_size_in_code_units_;
dex_pc += insn_flags_[dex_pc].GetLengthInCodeUnits()) {
os << StringPrintf("0x%04zx", dex_pc) << ": " << insn_flags_[dex_pc].Dump()
<< " " << inst->DumpHex(5) << " " << inst->DumpString(dex_file_) << "\n";
RegisterLine* reg_line = reg_table_.GetLine(dex_pc);
if (reg_line != NULL) {
os << reg_line->Dump() << "\n";
}
inst = inst->Next();
}
}
static bool IsPrimitiveDescriptor(char descriptor) {
switch (descriptor) {
case 'I':
case 'C':
case 'S':
case 'B':
case 'Z':
case 'F':
case 'D':
case 'J':
return true;
default:
return false;
}
}
bool MethodVerifier::SetTypesFromSignature() {
RegisterLine* reg_line = reg_table_.GetLine(0);
int arg_start = code_item_->registers_size_ - code_item_->ins_size_;
size_t expected_args = code_item_->ins_size_; /* long/double count as two */
DCHECK_GE(arg_start, 0); /* should have been verified earlier */
//Include the "this" pointer.
size_t cur_arg = 0;
if (!IsStatic()) {
// If this is a constructor for a class other than java.lang.Object, mark the first ("this")
// argument as uninitialized. This restricts field access until the superclass constructor is
// called.
const RegType& declaring_class = GetDeclaringClass();
if (IsConstructor() && !declaring_class.IsJavaLangObject()) {
reg_line->SetRegisterType(arg_start + cur_arg,
reg_types_.UninitializedThisArgument(declaring_class));
} else {
reg_line->SetRegisterType(arg_start + cur_arg, declaring_class);
}
cur_arg++;
}
const DexFile::ProtoId& proto_id =
dex_file_->GetMethodPrototype(dex_file_->GetMethodId(method_idx_));
DexFileParameterIterator iterator(*dex_file_, proto_id);
for (; iterator.HasNext(); iterator.Next()) {
const char* descriptor = iterator.GetDescriptor();
if (descriptor == NULL) {
LOG(FATAL) << "Null descriptor";
}
if (cur_arg >= expected_args) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "expected " << expected_args
<< " args, found more (" << descriptor << ")";
return false;
}
switch (descriptor[0]) {
case 'L':
case '[':
// We assume that reference arguments are initialized. The only way it could be otherwise
// (assuming the caller was verified) is if the current method is <init>, but in that case
// it's effectively considered initialized the instant we reach here (in the sense that we
// can return without doing anything or call virtual methods).
{
const RegType& reg_type = reg_types_.FromDescriptor(class_loader_, descriptor);
reg_line->SetRegisterType(arg_start + cur_arg, reg_type);
}
break;
case 'Z':
reg_line->SetRegisterType(arg_start + cur_arg, reg_types_.Boolean());
break;
case 'C':
reg_line->SetRegisterType(arg_start + cur_arg, reg_types_.Char());
break;
case 'B':
reg_line->SetRegisterType(arg_start + cur_arg, reg_types_.Byte());
break;
case 'I':
reg_line->SetRegisterType(arg_start + cur_arg, reg_types_.Integer());
break;
case 'S':
reg_line->SetRegisterType(arg_start + cur_arg, reg_types_.Short());
break;
case 'F':
reg_line->SetRegisterType(arg_start + cur_arg, reg_types_.Float());
break;
case 'J':
case 'D': {
const RegType& low_half = descriptor[0] == 'J' ? reg_types_.Long() : reg_types_.Double();
reg_line->SetRegisterType(arg_start + cur_arg, low_half); // implicitly sets high-register
cur_arg++;
break;
}
default:
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "unexpected signature type char '" << descriptor << "'";
return false;
}
cur_arg++;
}
if (cur_arg != expected_args) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "expected " << expected_args << " arguments, found " << cur_arg;
return false;
}
const char* descriptor = dex_file_->GetReturnTypeDescriptor(proto_id);
// Validate return type. We don't do the type lookup; just want to make sure that it has the right
// format. Only major difference from the method argument format is that 'V' is supported.
bool result;
if (IsPrimitiveDescriptor(descriptor[0]) || descriptor[0] == 'V') {
result = descriptor[1] == '\0';
} else if (descriptor[0] == '[') { // single/multi-dimensional array of object/primitive
size_t i = 0;
do {
i++;
} while (descriptor[i] == '['); // process leading [
if (descriptor[i] == 'L') { // object array
do {
i++; // find closing ;
} while (descriptor[i] != ';' && descriptor[i] != '\0');
result = descriptor[i] == ';';
} else { // primitive array
result = IsPrimitiveDescriptor(descriptor[i]) && descriptor[i + 1] == '\0';
}
} else if (descriptor[0] == 'L') {
// could be more thorough here, but shouldn't be required
size_t i = 0;
do {
i++;
} while (descriptor[i] != ';' && descriptor[i] != '\0');
result = descriptor[i] == ';';
} else {
result = false;
}
if (!result) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "unexpected char in return type descriptor '"
<< descriptor << "'";
}
return result;
}
bool MethodVerifier::CodeFlowVerifyMethod() {
const uint16_t* insns = code_item_->insns_;
const uint32_t insns_size = code_item_->insns_size_in_code_units_;
/* Begin by marking the first instruction as "changed". */
insn_flags_[0].SetChanged();
uint32_t start_guess = 0;
/* Continue until no instructions are marked "changed". */
while (true) {
// Find the first marked one. Use "start_guess" as a way to find one quickly.
uint32_t insn_idx = start_guess;
for (; insn_idx < insns_size; insn_idx++) {
if (insn_flags_[insn_idx].IsChanged())
break;
}
if (insn_idx == insns_size) {
if (start_guess != 0) {
/* try again, starting from the top */
start_guess = 0;
continue;
} else {
/* all flags are clear */
break;
}
}
// We carry the working set of registers from instruction to instruction. If this address can
// be the target of a branch (or throw) instruction, or if we're skipping around chasing
// "changed" flags, we need to load the set of registers from the table.
// Because we always prefer to continue on to the next instruction, we should never have a
// situation where we have a stray "changed" flag set on an instruction that isn't a branch
// target.
work_insn_idx_ = insn_idx;
if (insn_flags_[insn_idx].IsBranchTarget()) {
work_line_->CopyFromLine(reg_table_.GetLine(insn_idx));
} else {
#ifndef NDEBUG
/*
* Sanity check: retrieve the stored register line (assuming
* a full table) and make sure it actually matches.
*/
RegisterLine* register_line = reg_table_.GetLine(insn_idx);
if (register_line != NULL) {
if (work_line_->CompareLine(register_line) != 0) {
Dump(std::cout);
std::cout << info_messages_.str();
LOG(FATAL) << "work_line diverged in " << PrettyMethod(method_idx_, *dex_file_)
<< "@" << reinterpret_cast<void*>(work_insn_idx_) << "\n"
<< " work_line=" << *work_line_ << "\n"
<< " expected=" << *register_line;
}
}
#endif
}
if (!CodeFlowVerifyInstruction(&start_guess)) {
std::string prepend(PrettyMethod(method_idx_, *dex_file_));
prepend += " failed to verify: ";
PrependToLastFailMessage(prepend);
return false;
}
/* Clear "changed" and mark as visited. */
insn_flags_[insn_idx].SetVisited();
insn_flags_[insn_idx].ClearChanged();
}
if (gDebugVerify) {
/*
* Scan for dead code. There's nothing "evil" about dead code
* (besides the wasted space), but it indicates a flaw somewhere
* down the line, possibly in the verifier.
*
* If we've substituted "always throw" instructions into the stream,
* we are almost certainly going to have some dead code.
*/
int dead_start = -1;
uint32_t insn_idx = 0;
for (; insn_idx < insns_size; insn_idx += insn_flags_[insn_idx].GetLengthInCodeUnits()) {
/*
* Switch-statement data doesn't get "visited" by scanner. It
* may or may not be preceded by a padding NOP (for alignment).
*/
if (insns[insn_idx] == Instruction::kPackedSwitchSignature ||
insns[insn_idx] == Instruction::kSparseSwitchSignature ||
insns[insn_idx] == Instruction::kArrayDataSignature ||
(insns[insn_idx] == Instruction::NOP &&
(insns[insn_idx + 1] == Instruction::kPackedSwitchSignature ||
insns[insn_idx + 1] == Instruction::kSparseSwitchSignature ||
insns[insn_idx + 1] == Instruction::kArrayDataSignature))) {
insn_flags_[insn_idx].SetVisited();
}
if (!insn_flags_[insn_idx].IsVisited()) {
if (dead_start < 0)
dead_start = insn_idx;
} else if (dead_start >= 0) {
LogVerifyInfo() << "dead code " << reinterpret_cast<void*>(dead_start) << "-" << reinterpret_cast<void*>(insn_idx - 1);
dead_start = -1;
}
}
if (dead_start >= 0) {
LogVerifyInfo() << "dead code " << reinterpret_cast<void*>(dead_start) << "-" << reinterpret_cast<void*>(insn_idx - 1);
}
}
return true;
}
bool MethodVerifier::CodeFlowVerifyInstruction(uint32_t* start_guess) {
#ifdef VERIFIER_STATS
if (CurrentInsnFlags().IsVisited()) {
gDvm.verifierStats.instrsReexamined++;
} else {
gDvm.verifierStats.instrsExamined++;
}
#endif
// If we're doing FindLocksAtDexPc, check whether we're at the dex pc we care about.
// We want the state _before_ the instruction, for the case where the dex pc we're
// interested in is itself a monitor-enter instruction (which is a likely place
// for a thread to be suspended).
if (monitor_enter_dex_pcs_ != NULL && work_insn_idx_ == interesting_dex_pc_) {
for (size_t i = 0; i < work_line_->GetMonitorEnterCount(); ++i) {
monitor_enter_dex_pcs_->push_back(work_line_->GetMonitorEnterDexPc(i));
}
}
/*
* Once we finish decoding the instruction, we need to figure out where
* we can go from here. There are three possible ways to transfer
* control to another statement:
*
* (1) Continue to the next instruction. Applies to all but
* unconditional branches, method returns, and exception throws.
* (2) Branch to one or more possible locations. Applies to branches
* and switch statements.
* (3) Exception handlers. Applies to any instruction that can
* throw an exception that is handled by an encompassing "try"
* block.
*
* We can also return, in which case there is no successor instruction
* from this point.
*
* The behavior can be determined from the opcode flags.
*/
const uint16_t* insns = code_item_->insns_ + work_insn_idx_;
const Instruction* inst = Instruction::At(insns);
DecodedInstruction dec_insn(inst);
int opcode_flags = Instruction::Flags(inst->Opcode());
int32_t branch_target = 0;
bool just_set_result = false;
if (gDebugVerify) {
// Generate processing back trace to debug verifier
LogVerifyInfo() << "Processing " << inst->DumpString(dex_file_) << "\n"
<< *work_line_.get() << "\n";
}
/*
* Make a copy of the previous register state. If the instruction
* can throw an exception, we will copy/merge this into the "catch"
* address rather than work_line, because we don't want the result
* from the "successful" code path (e.g. a check-cast that "improves"
* a type) to be visible to the exception handler.
*/
if ((opcode_flags & Instruction::kThrow) != 0 && CurrentInsnFlags()->IsInTry()) {
saved_line_->CopyFromLine(work_line_.get());
} else {
#ifndef NDEBUG
saved_line_->FillWithGarbage();
#endif
}
switch (dec_insn.opcode) {
case Instruction::NOP:
/*
* A "pure" NOP has no effect on anything. Data tables start with
* a signature that looks like a NOP; if we see one of these in
* the course of executing code then we have a problem.
*/
if (dec_insn.vA != 0) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "encountered data table in instruction stream";
}
break;
case Instruction::MOVE:
case Instruction::MOVE_FROM16:
case Instruction::MOVE_16:
work_line_->CopyRegister1(dec_insn.vA, dec_insn.vB, kTypeCategory1nr);
break;
case Instruction::MOVE_WIDE:
case Instruction::MOVE_WIDE_FROM16:
case Instruction::MOVE_WIDE_16:
work_line_->CopyRegister2(dec_insn.vA, dec_insn.vB);
break;
case Instruction::MOVE_OBJECT:
case Instruction::MOVE_OBJECT_FROM16:
case Instruction::MOVE_OBJECT_16:
work_line_->CopyRegister1(dec_insn.vA, dec_insn.vB, kTypeCategoryRef);
break;
/*
* The move-result instructions copy data out of a "pseudo-register"
* with the results from the last method invocation. In practice we
* might want to hold the result in an actual CPU register, so the
* Dalvik spec requires that these only appear immediately after an
* invoke or filled-new-array.
*
* These calls invalidate the "result" register. (This is now
* redundant with the reset done below, but it can make the debug info
* easier to read in some cases.)
*/
case Instruction::MOVE_RESULT:
work_line_->CopyResultRegister1(dec_insn.vA, false);
break;
case Instruction::MOVE_RESULT_WIDE:
work_line_->CopyResultRegister2(dec_insn.vA);
break;
case Instruction::MOVE_RESULT_OBJECT:
work_line_->CopyResultRegister1(dec_insn.vA, true);
break;
case Instruction::MOVE_EXCEPTION: {
/*
* This statement can only appear as the first instruction in an exception handler. We verify
* that as part of extracting the exception type from the catch block list.
*/
const RegType& res_type = GetCaughtExceptionType();
work_line_->SetRegisterType(dec_insn.vA, res_type);
break;
}
case Instruction::RETURN_VOID:
if (!IsConstructor() || work_line_->CheckConstructorReturn()) {
if (!GetMethodReturnType().IsConflict()) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "return-void not expected";
}
}
break;
case Instruction::RETURN:
if (!IsConstructor() || work_line_->CheckConstructorReturn()) {
/* check the method signature */
const RegType& return_type = GetMethodReturnType();
if (!return_type.IsCategory1Types()) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "unexpected non-category 1 return type " << return_type;
} else {
// Compilers may generate synthetic functions that write byte values into boolean fields.
// Also, it may use integer values for boolean, byte, short, and character return types.
const RegType& src_type = work_line_->GetRegisterType(dec_insn.vA);
bool use_src = ((return_type.IsBoolean() && src_type.IsByte()) ||
((return_type.IsBoolean() || return_type.IsByte() ||
return_type.IsShort() || return_type.IsChar()) &&
src_type.IsInteger()));
/* check the register contents */
bool success =
work_line_->VerifyRegisterType(dec_insn.vA, use_src ? src_type : return_type);
if (!success) {
AppendToLastFailMessage(StringPrintf(" return-1nr on invalid register v%d", dec_insn.vA));
}
}
}
break;
case Instruction::RETURN_WIDE:
if (!IsConstructor() || work_line_->CheckConstructorReturn()) {
/* check the method signature */
const RegType& return_type = GetMethodReturnType();
if (!return_type.IsCategory2Types()) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "return-wide not expected";
} else {
/* check the register contents */
bool success = work_line_->VerifyRegisterType(dec_insn.vA, return_type);
if (!success) {
AppendToLastFailMessage(StringPrintf(" return-wide on invalid register v%d", dec_insn.vA));
}
}
}
break;
case Instruction::RETURN_OBJECT:
if (!IsConstructor() || work_line_->CheckConstructorReturn()) {
const RegType& return_type = GetMethodReturnType();
if (!return_type.IsReferenceTypes()) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "return-object not expected";
} else {
/* return_type is the *expected* return type, not register value */
DCHECK(!return_type.IsZero());
DCHECK(!return_type.IsUninitializedReference());
const RegType& reg_type = work_line_->GetRegisterType(dec_insn.vA);
// Disallow returning uninitialized values and verify that the reference in vAA is an
// instance of the "return_type"
if (reg_type.IsUninitializedTypes()) {
Fail(VERIFY_ERROR_BAD_CLASS_SOFT) << "returning uninitialized object '" << reg_type << "'";
} else if (!return_type.IsAssignableFrom(reg_type)) {
Fail(reg_type.IsUnresolvedTypes() ? VERIFY_ERROR_BAD_CLASS_SOFT : VERIFY_ERROR_BAD_CLASS_HARD)
<< "returning '" << reg_type << "', but expected from declaration '" << return_type << "'";
}
}
}
break;
case Instruction::CONST_4:
case Instruction::CONST_16:
case Instruction::CONST:
/* could be boolean, int, float, or a null reference */
work_line_->SetRegisterType(dec_insn.vA, reg_types_.FromCat1Const((int32_t) dec_insn.vB));
break;
case Instruction::CONST_HIGH16:
/* could be boolean, int, float, or a null reference */
work_line_->SetRegisterType(dec_insn.vA,
reg_types_.FromCat1Const((int32_t) dec_insn.vB << 16));
break;
case Instruction::CONST_WIDE_16:
case Instruction::CONST_WIDE_32:
case Instruction::CONST_WIDE:
case Instruction::CONST_WIDE_HIGH16:
/* could be long or double; resolved upon use */
work_line_->SetRegisterType(dec_insn.vA, reg_types_.ConstLo());
break;
case Instruction::CONST_STRING:
case Instruction::CONST_STRING_JUMBO:
work_line_->SetRegisterType(dec_insn.vA, reg_types_.JavaLangString());
break;
case Instruction::CONST_CLASS: {
// Get type from instruction if unresolved then we need an access check
// TODO: check Compiler::CanAccessTypeWithoutChecks returns false when res_type is unresolved
const RegType& res_type = ResolveClassAndCheckAccess(dec_insn.vB);
// Register holds class, ie its type is class, on error it will hold Conflict.
work_line_->SetRegisterType(dec_insn.vA,
res_type.IsConflict() ? res_type : reg_types_.JavaLangClass());
break;
}
case Instruction::MONITOR_ENTER:
work_line_->PushMonitor(dec_insn.vA, work_insn_idx_);
break;
case Instruction::MONITOR_EXIT:
/*
* monitor-exit instructions are odd. They can throw exceptions,
* but when they do they act as if they succeeded and the PC is
* pointing to the following instruction. (This behavior goes back
* to the need to handle asynchronous exceptions, a now-deprecated
* feature that Dalvik doesn't support.)
*
* In practice we don't need to worry about this. The only
* exceptions that can be thrown from monitor-exit are for a
* null reference and -exit without a matching -enter. If the
* structured locking checks are working, the former would have
* failed on the -enter instruction, and the latter is impossible.
*
* This is fortunate, because issue 3221411 prevents us from
* chasing the "can throw" path when monitor verification is
* enabled. If we can fully verify the locking we can ignore
* some catch blocks (which will show up as "dead" code when
* we skip them here); if we can't, then the code path could be
* "live" so we still need to check it.
*/
opcode_flags &= ~Instruction::kThrow;
work_line_->PopMonitor(dec_insn.vA);
break;
case Instruction::CHECK_CAST:
case Instruction::INSTANCE_OF: {
/*
* If this instruction succeeds, we will "downcast" register vA to the type in vB. (This
* could be a "upcast" -- not expected, so we don't try to address it.)
*
* If it fails, an exception is thrown, which we deal with later by ignoring the update to
* dec_insn.vA when branching to a handler.
*/
bool is_checkcast = dec_insn.opcode == Instruction::CHECK_CAST;
const RegType& res_type =
ResolveClassAndCheckAccess(is_checkcast ? dec_insn.vB : dec_insn.vC);
if (res_type.IsConflict()) {
DCHECK_NE(failures_.size(), 0U);
if (!is_checkcast) {
work_line_->SetRegisterType(dec_insn.vA, reg_types_.Boolean());
}
break; // bad class
}
// TODO: check Compiler::CanAccessTypeWithoutChecks returns false when res_type is unresolved
const RegType& orig_type =
work_line_->GetRegisterType(is_checkcast ? dec_insn.vA : dec_insn.vB);
if (!res_type.IsNonZeroReferenceTypes()) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "check-cast on unexpected class " << res_type;
} else if (!orig_type.IsReferenceTypes()) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "check-cast on non-reference in v" << dec_insn.vA;
} else {
if (is_checkcast) {
work_line_->SetRegisterType(dec_insn.vA, res_type);
} else {
work_line_->SetRegisterType(dec_insn.vA, reg_types_.Boolean());
}
}
break;
}
case Instruction::ARRAY_LENGTH: {
const RegType& res_type = work_line_->GetRegisterType(dec_insn.vB);
if (res_type.IsReferenceTypes()) {
if (!res_type.IsArrayTypes() && !res_type.IsZero()) { // ie not an array or null
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "array-length on non-array " << res_type;
} else {
work_line_->SetRegisterType(dec_insn.vA, reg_types_.Integer());
}
}
break;
}
case Instruction::NEW_INSTANCE: {
const RegType& res_type = ResolveClassAndCheckAccess(dec_insn.vB);
if (res_type.IsConflict()) {
DCHECK_NE(failures_.size(), 0U);
break; // bad class
}
// TODO: check Compiler::CanAccessTypeWithoutChecks returns false when res_type is unresolved
// can't create an instance of an interface or abstract class */
if (!res_type.IsInstantiableTypes()) {
Fail(VERIFY_ERROR_INSTANTIATION)
<< "new-instance on primitive, interface or abstract class" << res_type;
} else {
const RegType& uninit_type = reg_types_.Uninitialized(res_type, work_insn_idx_);
// Any registers holding previous allocations from this address that have not yet been
// initialized must be marked invalid.
work_line_->MarkUninitRefsAsInvalid(uninit_type);
// add the new uninitialized reference to the register state
work_line_->SetRegisterType(dec_insn.vA, uninit_type);
}
break;
}
case Instruction::NEW_ARRAY:
VerifyNewArray(dec_insn, false, false);
break;
case Instruction::FILLED_NEW_ARRAY:
VerifyNewArray(dec_insn, true, false);
just_set_result = true; // Filled new array sets result register
break;
case Instruction::FILLED_NEW_ARRAY_RANGE:
VerifyNewArray(dec_insn, true, true);
just_set_result = true; // Filled new array range sets result register
break;
case Instruction::CMPL_FLOAT:
case Instruction::CMPG_FLOAT:
if (!work_line_->VerifyRegisterType(dec_insn.vB, reg_types_.Float())) {
break;
}
if (!work_line_->VerifyRegisterType(dec_insn.vC, reg_types_.Float())) {
break;
}
work_line_->SetRegisterType(dec_insn.vA, reg_types_.Integer());
break;
case Instruction::CMPL_DOUBLE:
case Instruction::CMPG_DOUBLE:
if (!work_line_->VerifyRegisterType(dec_insn.vB, reg_types_.Double())) {
break;
}
if (!work_line_->VerifyRegisterType(dec_insn.vC, reg_types_.Double())) {
break;
}
work_line_->SetRegisterType(dec_insn.vA, reg_types_.Integer());
break;
case Instruction::CMP_LONG:
if (!work_line_->VerifyRegisterType(dec_insn.vB, reg_types_.Long())) {
break;
}
if (!work_line_->VerifyRegisterType(dec_insn.vC, reg_types_.Long())) {
break;
}
work_line_->SetRegisterType(dec_insn.vA, reg_types_.Integer());
break;
case Instruction::THROW: {
const RegType& res_type = work_line_->GetRegisterType(dec_insn.vA);
if (!reg_types_.JavaLangThrowable().IsAssignableFrom(res_type)) {
Fail(VERIFY_ERROR_BAD_CLASS_SOFT) << "thrown class " << res_type << " not instanceof Throwable";
}
break;
}
case Instruction::GOTO:
case Instruction::GOTO_16:
case Instruction::GOTO_32:
/* no effect on or use of registers */
break;
case Instruction::PACKED_SWITCH:
case Instruction::SPARSE_SWITCH:
/* verify that vAA is an integer, or can be converted to one */
work_line_->VerifyRegisterType(dec_insn.vA, reg_types_.Integer());
break;
case Instruction::FILL_ARRAY_DATA: {
/* Similar to the verification done for APUT */
const RegType& array_type = work_line_->GetRegisterType(dec_insn.vA);
/* array_type can be null if the reg type is Zero */
if (!array_type.IsZero()) {
if (!array_type.IsArrayTypes()) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "invalid fill-array-data with array type " << array_type;
} else {
const RegType& component_type = reg_types_.GetComponentType(array_type, class_loader_);
DCHECK(!component_type.IsConflict());
if (component_type.IsNonZeroReferenceTypes()) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "invalid fill-array-data with component type "
<< component_type;
} else {
// Now verify if the element width in the table matches the element width declared in
// the array
const uint16_t* array_data = insns + (insns[1] | (((int32_t) insns[2]) << 16));
if (array_data[0] != Instruction::kArrayDataSignature) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "invalid magic for array-data";
} else {
size_t elem_width = Primitive::ComponentSize(component_type.GetPrimitiveType());
// Since we don't compress the data in Dex, expect to see equal width of data stored
// in the table and expected from the array class.
if (array_data[1] != elem_width) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "array-data size mismatch (" << array_data[1]
<< " vs " << elem_width << ")";
}
}
}
}
}
break;
}
case Instruction::IF_EQ:
case Instruction::IF_NE: {
const RegType& reg_type1 = work_line_->GetRegisterType(dec_insn.vA);
const RegType& reg_type2 = work_line_->GetRegisterType(dec_insn.vB);
bool mismatch = false;
if (reg_type1.IsZero()) { // zero then integral or reference expected
mismatch = !reg_type2.IsReferenceTypes() && !reg_type2.IsIntegralTypes();
} else if (reg_type1.IsReferenceTypes()) { // both references?
mismatch = !reg_type2.IsReferenceTypes();
} else { // both integral?
mismatch = !reg_type1.IsIntegralTypes() || !reg_type2.IsIntegralTypes();
}
if (mismatch) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "args to if-eq/if-ne (" << reg_type1 << "," << reg_type2
<< ") must both be references or integral";
}
break;
}
case Instruction::IF_LT:
case Instruction::IF_GE:
case Instruction::IF_GT:
case Instruction::IF_LE: {
const RegType& reg_type1 = work_line_->GetRegisterType(dec_insn.vA);
const RegType& reg_type2 = work_line_->GetRegisterType(dec_insn.vB);
if (!reg_type1.IsIntegralTypes() || !reg_type2.IsIntegralTypes()) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "args to 'if' (" << reg_type1 << ","
<< reg_type2 << ") must be integral";
}
break;
}
case Instruction::IF_EQZ:
case Instruction::IF_NEZ: {
const RegType& reg_type = work_line_->GetRegisterType(dec_insn.vA);
if (!reg_type.IsReferenceTypes() && !reg_type.IsIntegralTypes()) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "type " << reg_type << " unexpected as arg to if-eqz/if-nez";
}
break;
}
case Instruction::IF_LTZ:
case Instruction::IF_GEZ:
case Instruction::IF_GTZ:
case Instruction::IF_LEZ: {
const RegType& reg_type = work_line_->GetRegisterType(dec_insn.vA);
if (!reg_type.IsIntegralTypes()) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "type " << reg_type
<< " unexpected as arg to if-ltz/if-gez/if-gtz/if-lez";
}
break;
}
case Instruction::AGET_BOOLEAN:
VerifyAGet(dec_insn, reg_types_.Boolean(), true);
break;
case Instruction::AGET_BYTE:
VerifyAGet(dec_insn, reg_types_.Byte(), true);
break;
case Instruction::AGET_CHAR:
VerifyAGet(dec_insn, reg_types_.Char(), true);
break;
case Instruction::AGET_SHORT:
VerifyAGet(dec_insn, reg_types_.Short(), true);
break;
case Instruction::AGET:
VerifyAGet(dec_insn, reg_types_.Integer(), true);
break;
case Instruction::AGET_WIDE:
VerifyAGet(dec_insn, reg_types_.Long(), true);
break;
case Instruction::AGET_OBJECT:
VerifyAGet(dec_insn, reg_types_.JavaLangObject(), false);
break;
case Instruction::APUT_BOOLEAN:
VerifyAPut(dec_insn, reg_types_.Boolean(), true);
break;
case Instruction::APUT_BYTE:
VerifyAPut(dec_insn, reg_types_.Byte(), true);
break;
case Instruction::APUT_CHAR:
VerifyAPut(dec_insn, reg_types_.Char(), true);
break;
case Instruction::APUT_SHORT:
VerifyAPut(dec_insn, reg_types_.Short(), true);
break;
case Instruction::APUT:
VerifyAPut(dec_insn, reg_types_.Integer(), true);
break;
case Instruction::APUT_WIDE:
VerifyAPut(dec_insn, reg_types_.Long(), true);
break;
case Instruction::APUT_OBJECT:
VerifyAPut(dec_insn, reg_types_.JavaLangObject(), false);
break;
case Instruction::IGET_BOOLEAN:
VerifyISGet(dec_insn, reg_types_.Boolean(), true, false);
break;
case Instruction::IGET_BYTE:
VerifyISGet(dec_insn, reg_types_.Byte(), true, false);
break;
case Instruction::IGET_CHAR:
VerifyISGet(dec_insn, reg_types_.Char(), true, false);
break;
case Instruction::IGET_SHORT:
VerifyISGet(dec_insn, reg_types_.Short(), true, false);
break;
case Instruction::IGET:
VerifyISGet(dec_insn, reg_types_.Integer(), true, false);
break;
case Instruction::IGET_WIDE:
VerifyISGet(dec_insn, reg_types_.Long(), true, false);
break;
case Instruction::IGET_OBJECT:
VerifyISGet(dec_insn, reg_types_.JavaLangObject(), false, false);
break;
case Instruction::IPUT_BOOLEAN:
VerifyISPut(dec_insn, reg_types_.Boolean(), true, false);
break;
case Instruction::IPUT_BYTE:
VerifyISPut(dec_insn, reg_types_.Byte(), true, false);
break;
case Instruction::IPUT_CHAR:
VerifyISPut(dec_insn, reg_types_.Char(), true, false);
break;
case Instruction::IPUT_SHORT:
VerifyISPut(dec_insn, reg_types_.Short(), true, false);
break;
case Instruction::IPUT:
VerifyISPut(dec_insn, reg_types_.Integer(), true, false);
break;
case Instruction::IPUT_WIDE:
VerifyISPut(dec_insn, reg_types_.Long(), true, false);
break;
case Instruction::IPUT_OBJECT:
VerifyISPut(dec_insn, reg_types_.JavaLangObject(), false, false);
break;
case Instruction::SGET_BOOLEAN:
VerifyISGet(dec_insn, reg_types_.Boolean(), true, true);
break;
case Instruction::SGET_BYTE:
VerifyISGet(dec_insn, reg_types_.Byte(), true, true);
break;
case Instruction::SGET_CHAR:
VerifyISGet(dec_insn, reg_types_.Char(), true, true);
break;
case Instruction::SGET_SHORT:
VerifyISGet(dec_insn, reg_types_.Short(), true, true);
break;
case Instruction::SGET:
VerifyISGet(dec_insn, reg_types_.Integer(), true, true);
break;
case Instruction::SGET_WIDE:
VerifyISGet(dec_insn, reg_types_.Long(), true, true);
break;
case Instruction::SGET_OBJECT:
VerifyISGet(dec_insn, reg_types_.JavaLangObject(), false, true);
break;
case Instruction::SPUT_BOOLEAN:
VerifyISPut(dec_insn, reg_types_.Boolean(), true, true);
break;
case Instruction::SPUT_BYTE:
VerifyISPut(dec_insn, reg_types_.Byte(), true, true);
break;
case Instruction::SPUT_CHAR:
VerifyISPut(dec_insn, reg_types_.Char(), true, true);
break;
case Instruction::SPUT_SHORT:
VerifyISPut(dec_insn, reg_types_.Short(), true, true);
break;
case Instruction::SPUT:
VerifyISPut(dec_insn, reg_types_.Integer(), true, true);
break;
case Instruction::SPUT_WIDE:
VerifyISPut(dec_insn, reg_types_.Long(), true, true);
break;
case Instruction::SPUT_OBJECT:
VerifyISPut(dec_insn, reg_types_.JavaLangObject(), false, true);
break;
case Instruction::INVOKE_VIRTUAL:
case Instruction::INVOKE_VIRTUAL_RANGE:
case Instruction::INVOKE_SUPER:
case Instruction::INVOKE_SUPER_RANGE: {
bool is_range = (dec_insn.opcode == Instruction::INVOKE_VIRTUAL_RANGE ||
dec_insn.opcode == Instruction::INVOKE_SUPER_RANGE);
bool is_super = (dec_insn.opcode == Instruction::INVOKE_SUPER ||
dec_insn.opcode == Instruction::INVOKE_SUPER_RANGE);
Method* called_method = VerifyInvocationArgs(dec_insn, METHOD_VIRTUAL, is_range, is_super);
const char* descriptor;
if (called_method == NULL) {
uint32_t method_idx = dec_insn.vB;
const DexFile::MethodId& method_id = dex_file_->GetMethodId(method_idx);
uint32_t return_type_idx = dex_file_->GetProtoId(method_id.proto_idx_).return_type_idx_;
descriptor = dex_file_->StringByTypeIdx(return_type_idx);
} else {
descriptor = MethodHelper(called_method).GetReturnTypeDescriptor();
}
const RegType& return_type = reg_types_.FromDescriptor(class_loader_, descriptor);
work_line_->SetResultRegisterType(return_type);
just_set_result = true;
break;
}
case Instruction::INVOKE_DIRECT:
case Instruction::INVOKE_DIRECT_RANGE: {
bool is_range = (dec_insn.opcode == Instruction::INVOKE_DIRECT_RANGE);
Method* called_method = VerifyInvocationArgs(dec_insn, METHOD_DIRECT, is_range, false);
const char* return_type_descriptor;
bool is_constructor;
if (called_method == NULL) {
uint32_t method_idx = dec_insn.vB;
const DexFile::MethodId& method_id = dex_file_->GetMethodId(method_idx);
is_constructor = StringPiece(dex_file_->GetMethodName(method_id)) == "<init>";
uint32_t return_type_idx = dex_file_->GetProtoId(method_id.proto_idx_).return_type_idx_;
return_type_descriptor = dex_file_->StringByTypeIdx(return_type_idx);
} else {
is_constructor = called_method->IsConstructor();
return_type_descriptor = MethodHelper(called_method).GetReturnTypeDescriptor();
}
if (is_constructor) {
/*
* Some additional checks when calling a constructor. We know from the invocation arg check
* that the "this" argument is an instance of called_method->klass. Now we further restrict
* that to require that called_method->klass is the same as this->klass or this->super,
* allowing the latter only if the "this" argument is the same as the "this" argument to
* this method (which implies that we're in a constructor ourselves).
*/
const RegType& this_type = work_line_->GetInvocationThis(dec_insn);
if (this_type.IsConflict()) // failure.
break;
/* no null refs allowed (?) */
if (this_type.IsZero()) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "unable to initialize null ref";
break;
}
/* must be in same class or in superclass */
// const RegType& this_super_klass = this_type.GetSuperClass(&reg_types_);
// TODO: re-enable constructor type verification
// if (this_super_klass.IsConflict()) {
// Unknown super class, fail so we re-check at runtime.
// Fail(VERIFY_ERROR_BAD_CLASS_SOFT) << "super class unknown for '" << this_type << "'";
// break;
// }
/* arg must be an uninitialized reference */
if (!this_type.IsUninitializedTypes()) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "Expected initialization on uninitialized reference "
<< this_type;
break;
}
/*
* Replace the uninitialized reference with an initialized one. We need to do this for all
* registers that have the same object instance in them, not just the "this" register.
*/
work_line_->MarkRefsAsInitialized(this_type);
}
const RegType& return_type = reg_types_.FromDescriptor(class_loader_, return_type_descriptor);
work_line_->SetResultRegisterType(return_type);
just_set_result = true;
break;
}
case Instruction::INVOKE_STATIC:
case Instruction::INVOKE_STATIC_RANGE: {
bool is_range = (dec_insn.opcode == Instruction::INVOKE_STATIC_RANGE);
Method* called_method = VerifyInvocationArgs(dec_insn, METHOD_STATIC, is_range, false);
const char* descriptor;
if (called_method == NULL) {
uint32_t method_idx = dec_insn.vB;
const DexFile::MethodId& method_id = dex_file_->GetMethodId(method_idx);
uint32_t return_type_idx = dex_file_->GetProtoId(method_id.proto_idx_).return_type_idx_;
descriptor = dex_file_->StringByTypeIdx(return_type_idx);
} else {
descriptor = MethodHelper(called_method).GetReturnTypeDescriptor();
}
const RegType& return_type = reg_types_.FromDescriptor(class_loader_, descriptor);
work_line_->SetResultRegisterType(return_type);
just_set_result = true;
}
break;
case Instruction::INVOKE_INTERFACE:
case Instruction::INVOKE_INTERFACE_RANGE: {
bool is_range = (dec_insn.opcode == Instruction::INVOKE_INTERFACE_RANGE);
Method* abs_method = VerifyInvocationArgs(dec_insn, METHOD_INTERFACE, is_range, false);
if (abs_method != NULL) {
Class* called_interface = abs_method->GetDeclaringClass();
if (!called_interface->IsInterface() && !called_interface->IsObjectClass()) {
Fail(VERIFY_ERROR_CLASS_CHANGE) << "expected interface class in invoke-interface '"
<< PrettyMethod(abs_method) << "'";
break;
}
}
/* Get the type of the "this" arg, which should either be a sub-interface of called
* interface or Object (see comments in RegType::JoinClass).
*/
const RegType& this_type = work_line_->GetInvocationThis(dec_insn);
if (this_type.IsZero()) {
/* null pointer always passes (and always fails at runtime) */
} else {
if (this_type.IsUninitializedTypes()) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "interface call on uninitialized object "
<< this_type;
break;
}
// In the past we have tried to assert that "called_interface" is assignable
// from "this_type.GetClass()", however, as we do an imprecise Join
// (RegType::JoinClass) we don't have full information on what interfaces are
// implemented by "this_type". For example, two classes may implement the same
// interfaces and have a common parent that doesn't implement the interface. The
// join will set "this_type" to the parent class and a test that this implements
// the interface will incorrectly fail.
}
/*
* We don't have an object instance, so we can't find the concrete method. However, all of
* the type information is in the abstract method, so we're good.
*/
const char* descriptor;
if (abs_method == NULL) {
uint32_t method_idx = dec_insn.vB;
const DexFile::MethodId& method_id = dex_file_->GetMethodId(method_idx);
uint32_t return_type_idx = dex_file_->GetProtoId(method_id.proto_idx_).return_type_idx_;
descriptor = dex_file_->StringByTypeIdx(return_type_idx);
} else {
descriptor = MethodHelper(abs_method).GetReturnTypeDescriptor();
}
const RegType& return_type = reg_types_.FromDescriptor(class_loader_, descriptor);
work_line_->SetResultRegisterType(return_type);
work_line_->SetResultRegisterType(return_type);
just_set_result = true;
break;
}
case Instruction::NEG_INT:
case Instruction::NOT_INT:
work_line_->CheckUnaryOp(dec_insn, reg_types_.Integer(), reg_types_.Integer());
break;
case Instruction::NEG_LONG:
case Instruction::NOT_LONG:
work_line_->CheckUnaryOp(dec_insn, reg_types_.Long(), reg_types_.Long());
break;
case Instruction::NEG_FLOAT:
work_line_->CheckUnaryOp(dec_insn, reg_types_.Float(), reg_types_.Float());
break;
case Instruction::NEG_DOUBLE:
work_line_->CheckUnaryOp(dec_insn, reg_types_.Double(), reg_types_.Double());
break;
case Instruction::INT_TO_LONG:
work_line_->CheckUnaryOp(dec_insn, reg_types_.Long(), reg_types_.Integer());
break;
case Instruction::INT_TO_FLOAT:
work_line_->CheckUnaryOp(dec_insn, reg_types_.Float(), reg_types_.Integer());
break;
case Instruction::INT_TO_DOUBLE:
work_line_->CheckUnaryOp(dec_insn, reg_types_.Double(), reg_types_.Integer());
break;
case Instruction::LONG_TO_INT:
work_line_->CheckUnaryOp(dec_insn, reg_types_.Integer(), reg_types_.Long());
break;
case Instruction::LONG_TO_FLOAT:
work_line_->CheckUnaryOp(dec_insn, reg_types_.Float(), reg_types_.Long());
break;
case Instruction::LONG_TO_DOUBLE:
work_line_->CheckUnaryOp(dec_insn, reg_types_.Double(), reg_types_.Long());
break;
case Instruction::FLOAT_TO_INT:
work_line_->CheckUnaryOp(dec_insn, reg_types_.Integer(), reg_types_.Float());
break;
case Instruction::FLOAT_TO_LONG:
work_line_->CheckUnaryOp(dec_insn, reg_types_.Long(), reg_types_.Float());
break;
case Instruction::FLOAT_TO_DOUBLE:
work_line_->CheckUnaryOp(dec_insn, reg_types_.Double(), reg_types_.Float());
break;
case Instruction::DOUBLE_TO_INT:
work_line_->CheckUnaryOp(dec_insn, reg_types_.Integer(), reg_types_.Double());
break;
case Instruction::DOUBLE_TO_LONG:
work_line_->CheckUnaryOp(dec_insn, reg_types_.Long(), reg_types_.Double());
break;
case Instruction::DOUBLE_TO_FLOAT:
work_line_->CheckUnaryOp(dec_insn, reg_types_.Float(), reg_types_.Double());
break;
case Instruction::INT_TO_BYTE:
work_line_->CheckUnaryOp(dec_insn, reg_types_.Byte(), reg_types_.Integer());
break;
case Instruction::INT_TO_CHAR:
work_line_->CheckUnaryOp(dec_insn, reg_types_.Char(), reg_types_.Integer());
break;
case Instruction::INT_TO_SHORT:
work_line_->CheckUnaryOp(dec_insn, reg_types_.Short(), reg_types_.Integer());
break;
case Instruction::ADD_INT:
case Instruction::SUB_INT:
case Instruction::MUL_INT:
case Instruction::REM_INT:
case Instruction::DIV_INT:
case Instruction::SHL_INT:
case Instruction::SHR_INT:
case Instruction::USHR_INT:
work_line_->CheckBinaryOp(dec_insn, reg_types_.Integer(), reg_types_.Integer(), reg_types_.Integer(), false);
break;
case Instruction::AND_INT:
case Instruction::OR_INT:
case Instruction::XOR_INT:
work_line_->CheckBinaryOp(dec_insn, reg_types_.Integer(), reg_types_.Integer(), reg_types_.Integer(), true);
break;
case Instruction::ADD_LONG:
case Instruction::SUB_LONG:
case Instruction::MUL_LONG:
case Instruction::DIV_LONG:
case Instruction::REM_LONG:
case Instruction::AND_LONG:
case Instruction::OR_LONG:
case Instruction::XOR_LONG:
work_line_->CheckBinaryOp(dec_insn, reg_types_.Long(), reg_types_.Long(), reg_types_.Long(), false);
break;
case Instruction::SHL_LONG:
case Instruction::SHR_LONG:
case Instruction::USHR_LONG:
/* shift distance is Int, making these different from other binary operations */
work_line_->CheckBinaryOp(dec_insn, reg_types_.Long(), reg_types_.Long(), reg_types_.Integer(), false);
break;
case Instruction::ADD_FLOAT:
case Instruction::SUB_FLOAT:
case Instruction::MUL_FLOAT:
case Instruction::DIV_FLOAT:
case Instruction::REM_FLOAT:
work_line_->CheckBinaryOp(dec_insn, reg_types_.Float(), reg_types_.Float(), reg_types_.Float(), false);
break;
case Instruction::ADD_DOUBLE:
case Instruction::SUB_DOUBLE:
case Instruction::MUL_DOUBLE:
case Instruction::DIV_DOUBLE:
case Instruction::REM_DOUBLE:
work_line_->CheckBinaryOp(dec_insn, reg_types_.Double(), reg_types_.Double(), reg_types_.Double(), false);
break;
case Instruction::ADD_INT_2ADDR:
case Instruction::SUB_INT_2ADDR:
case Instruction::MUL_INT_2ADDR:
case Instruction::REM_INT_2ADDR:
case Instruction::SHL_INT_2ADDR:
case Instruction::SHR_INT_2ADDR:
case Instruction::USHR_INT_2ADDR:
work_line_->CheckBinaryOp2addr(dec_insn, reg_types_.Integer(), reg_types_.Integer(), reg_types_.Integer(), false);
break;
case Instruction::AND_INT_2ADDR:
case Instruction::OR_INT_2ADDR:
case Instruction::XOR_INT_2ADDR:
work_line_->CheckBinaryOp2addr(dec_insn, reg_types_.Integer(), reg_types_.Integer(), reg_types_.Integer(), true);
break;
case Instruction::DIV_INT_2ADDR:
work_line_->CheckBinaryOp2addr(dec_insn, reg_types_.Integer(), reg_types_.Integer(), reg_types_.Integer(), false);
break;
case Instruction::ADD_LONG_2ADDR:
case Instruction::SUB_LONG_2ADDR:
case Instruction::MUL_LONG_2ADDR:
case Instruction::DIV_LONG_2ADDR:
case Instruction::REM_LONG_2ADDR:
case Instruction::AND_LONG_2ADDR:
case Instruction::OR_LONG_2ADDR:
case Instruction::XOR_LONG_2ADDR:
work_line_->CheckBinaryOp2addr(dec_insn, reg_types_.Long(), reg_types_.Long(), reg_types_.Long(), false);
break;
case Instruction::SHL_LONG_2ADDR:
case Instruction::SHR_LONG_2ADDR:
case Instruction::USHR_LONG_2ADDR:
work_line_->CheckBinaryOp2addr(dec_insn, reg_types_.Long(), reg_types_.Long(), reg_types_.Integer(), false);
break;
case Instruction::ADD_FLOAT_2ADDR:
case Instruction::SUB_FLOAT_2ADDR:
case Instruction::MUL_FLOAT_2ADDR:
case Instruction::DIV_FLOAT_2ADDR:
case Instruction::REM_FLOAT_2ADDR:
work_line_->CheckBinaryOp2addr(dec_insn, reg_types_.Float(), reg_types_.Float(), reg_types_.Float(), false);
break;
case Instruction::ADD_DOUBLE_2ADDR:
case Instruction::SUB_DOUBLE_2ADDR:
case Instruction::MUL_DOUBLE_2ADDR:
case Instruction::DIV_DOUBLE_2ADDR:
case Instruction::REM_DOUBLE_2ADDR:
work_line_->CheckBinaryOp2addr(dec_insn, reg_types_.Double(), reg_types_.Double(), reg_types_.Double(), false);
break;
case Instruction::ADD_INT_LIT16:
case Instruction::RSUB_INT:
case Instruction::MUL_INT_LIT16:
case Instruction::DIV_INT_LIT16:
case Instruction::REM_INT_LIT16:
work_line_->CheckLiteralOp(dec_insn, reg_types_.Integer(), reg_types_.Integer(), false);
break;
case Instruction::AND_INT_LIT16:
case Instruction::OR_INT_LIT16:
case Instruction::XOR_INT_LIT16:
work_line_->CheckLiteralOp(dec_insn, reg_types_.Integer(), reg_types_.Integer(), true);
break;
case Instruction::ADD_INT_LIT8:
case Instruction::RSUB_INT_LIT8:
case Instruction::MUL_INT_LIT8:
case Instruction::DIV_INT_LIT8:
case Instruction::REM_INT_LIT8:
case Instruction::SHL_INT_LIT8:
case Instruction::SHR_INT_LIT8:
case Instruction::USHR_INT_LIT8:
work_line_->CheckLiteralOp(dec_insn, reg_types_.Integer(), reg_types_.Integer(), false);
break;
case Instruction::AND_INT_LIT8:
case Instruction::OR_INT_LIT8:
case Instruction::XOR_INT_LIT8:
work_line_->CheckLiteralOp(dec_insn, reg_types_.Integer(), reg_types_.Integer(), true);
break;
/*
* This falls into the general category of "optimized" instructions,
* which don't generally appear during verification. Because it's
* inserted in the course of verification, we can expect to see it here.
*/
case Instruction::THROW_VERIFICATION_ERROR:
break;
/* These should never appear during verification. */
case Instruction::UNUSED_EE:
case Instruction::UNUSED_EF:
case Instruction::UNUSED_F2:
case Instruction::UNUSED_F3:
case Instruction::UNUSED_F4:
case Instruction::UNUSED_F5:
case Instruction::UNUSED_F6:
case Instruction::UNUSED_F7:
case Instruction::UNUSED_F8:
case Instruction::UNUSED_F9:
case Instruction::UNUSED_FA:
case Instruction::UNUSED_FB:
case Instruction::UNUSED_F0:
case Instruction::UNUSED_F1:
case Instruction::UNUSED_E3:
case Instruction::UNUSED_E8:
case Instruction::UNUSED_E7:
case Instruction::UNUSED_E4:
case Instruction::UNUSED_E9:
case Instruction::UNUSED_FC:
case Instruction::UNUSED_E5:
case Instruction::UNUSED_EA:
case Instruction::UNUSED_FD:
case Instruction::UNUSED_E6:
case Instruction::UNUSED_EB:
case Instruction::UNUSED_FE:
case Instruction::UNUSED_3E:
case Instruction::UNUSED_3F:
case Instruction::UNUSED_40:
case Instruction::UNUSED_41:
case Instruction::UNUSED_42:
case Instruction::UNUSED_43:
case Instruction::UNUSED_73:
case Instruction::UNUSED_79:
case Instruction::UNUSED_7A:
case Instruction::UNUSED_EC:
case Instruction::UNUSED_FF:
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "Unexpected opcode " << inst->DumpString(dex_file_);
break;
/*
* DO NOT add a "default" clause here. Without it the compiler will
* complain if an instruction is missing (which is desirable).
*/
} // end - switch (dec_insn.opcode)
if (have_pending_hard_failure_) {
if (!Runtime::Current()->IsStarted()) {
/* When compiling, check that the last failure is a hard failure */
CHECK_EQ(failures_[failures_.size() - 1], VERIFY_ERROR_BAD_CLASS_HARD);
}
/* immediate failure, reject class */
info_messages_ << "Rejecting opcode " << inst->DumpString(dex_file_);
return false;
} else if (have_pending_rewrite_failure_) {
/* replace opcode and continue on */
std::string append("Replacing opcode ");
append += inst->DumpString(dex_file_);
AppendToLastFailMessage(append);
ReplaceFailingInstruction();
/* IMPORTANT: method->insns may have been changed */
insns = code_item_->insns_ + work_insn_idx_;
/* continue on as if we just handled a throw-verification-error */
opcode_flags = Instruction::kThrow;
}
/*
* If we didn't just set the result register, clear it out. This ensures that you can only use
* "move-result" immediately after the result is set. (We could check this statically, but it's
* not expensive and it makes our debugging output cleaner.)
*/
if (!just_set_result) {
work_line_->SetResultTypeToUnknown();
}
/* Handle "continue". Tag the next consecutive instruction. */
if ((opcode_flags & Instruction::kContinue) != 0) {
uint32_t next_insn_idx = work_insn_idx_ + CurrentInsnFlags()->GetLengthInCodeUnits();
if (next_insn_idx >= code_item_->insns_size_in_code_units_) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "Execution can walk off end of code area";
return false;
}
// The only way to get to a move-exception instruction is to get thrown there. Make sure the
// next instruction isn't one.
if (!CheckNotMoveException(code_item_->insns_, next_insn_idx)) {
return false;
}
RegisterLine* next_line = reg_table_.GetLine(next_insn_idx);
if (next_line != NULL) {
// Merge registers into what we have for the next instruction, and set the "changed" flag if
// needed.
if (!UpdateRegisters(next_insn_idx, work_line_.get())) {
return false;
}
} else {
/*
* We're not recording register data for the next instruction, so we don't know what the prior
* state was. We have to assume that something has changed and re-evaluate it.
*/
insn_flags_[next_insn_idx].SetChanged();
}
}
/*
* Handle "branch". Tag the branch target.
*
* NOTE: instructions like Instruction::EQZ provide information about the
* state of the register when the branch is taken or not taken. For example,
* somebody could get a reference field, check it for zero, and if the
* branch is taken immediately store that register in a boolean field
* since the value is known to be zero. We do not currently account for
* that, and will reject the code.
*
* TODO: avoid re-fetching the branch target
*/
if ((opcode_flags & Instruction::kBranch) != 0) {
bool isConditional, selfOkay;
if (!GetBranchOffset(work_insn_idx_, &branch_target, &isConditional, &selfOkay)) {
/* should never happen after static verification */
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "bad branch";
return false;
}
DCHECK_EQ(isConditional, (opcode_flags & Instruction::kContinue) != 0);
if (!CheckNotMoveException(code_item_->insns_, work_insn_idx_ + branch_target)) {
return false;
}
/* update branch target, set "changed" if appropriate */
if (!UpdateRegisters(work_insn_idx_ + branch_target, work_line_.get())) {
return false;
}
}
/*
* Handle "switch". Tag all possible branch targets.
*
* We've already verified that the table is structurally sound, so we
* just need to walk through and tag the targets.
*/
if ((opcode_flags & Instruction::kSwitch) != 0) {
int offset_to_switch = insns[1] | (((int32_t) insns[2]) << 16);
const uint16_t* switch_insns = insns + offset_to_switch;
int switch_count = switch_insns[1];
int offset_to_targets, targ;
if ((*insns & 0xff) == Instruction::PACKED_SWITCH) {
/* 0 = sig, 1 = count, 2/3 = first key */
offset_to_targets = 4;
} else {
/* 0 = sig, 1 = count, 2..count * 2 = keys */
DCHECK((*insns & 0xff) == Instruction::SPARSE_SWITCH);
offset_to_targets = 2 + 2 * switch_count;
}
/* verify each switch target */
for (targ = 0; targ < switch_count; targ++) {
int offset;
uint32_t abs_offset;
/* offsets are 32-bit, and only partly endian-swapped */
offset = switch_insns[offset_to_targets + targ * 2] |
(((int32_t) switch_insns[offset_to_targets + targ * 2 + 1]) << 16);
abs_offset = work_insn_idx_ + offset;
DCHECK_LT(abs_offset, code_item_->insns_size_in_code_units_);
if (!CheckNotMoveException(code_item_->insns_, abs_offset)) {
return false;
}
if (!UpdateRegisters(abs_offset, work_line_.get()))
return false;
}
}
/*
* Handle instructions that can throw and that are sitting in a "try" block. (If they're not in a
* "try" block when they throw, control transfers out of the method.)
*/
if ((opcode_flags & Instruction::kThrow) != 0 && insn_flags_[work_insn_idx_].IsInTry()) {
bool within_catch_all = false;
CatchHandlerIterator iterator(*code_item_, work_insn_idx_);
for (; iterator.HasNext(); iterator.Next()) {
if (iterator.GetHandlerTypeIndex() == DexFile::kDexNoIndex16) {
within_catch_all = true;
}
/*
* Merge registers into the "catch" block. We want to use the "savedRegs" rather than
* "work_regs", because at runtime the exception will be thrown before the instruction
* modifies any registers.
*/
if (!UpdateRegisters(iterator.GetHandlerAddress(), saved_line_.get())) {
return false;
}
}
/*
* If the monitor stack depth is nonzero, there must be a "catch all" handler for this
* instruction. This does apply to monitor-exit because of async exception handling.
*/
if (work_line_->MonitorStackDepth() > 0 && !within_catch_all) {
/*
* The state in work_line reflects the post-execution state. If the current instruction is a
* monitor-enter and the monitor stack was empty, we don't need a catch-all (if it throws,
* it will do so before grabbing the lock).
*/
if (dec_insn.opcode != Instruction::MONITOR_ENTER || work_line_->MonitorStackDepth() != 1) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD)
<< "expected to be within a catch-all for an instruction where a monitor is held";
return false;
}
}
}
/* If we're returning from the method, make sure monitor stack is empty. */
if ((opcode_flags & Instruction::kReturn) != 0) {
if (!work_line_->VerifyMonitorStackEmpty()) {
return false;
}
}
/*
* Update start_guess. Advance to the next instruction of that's
* possible, otherwise use the branch target if one was found. If
* neither of those exists we're in a return or throw; leave start_guess
* alone and let the caller sort it out.
*/
if ((opcode_flags & Instruction::kContinue) != 0) {
*start_guess = work_insn_idx_ + insn_flags_[work_insn_idx_].GetLengthInCodeUnits();
} else if ((opcode_flags & Instruction::kBranch) != 0) {
/* we're still okay if branch_target is zero */
*start_guess = work_insn_idx_ + branch_target;
}
DCHECK_LT(*start_guess, code_item_->insns_size_in_code_units_);
DCHECK(insn_flags_[*start_guess].IsOpcode());
return true;
}
const RegType& MethodVerifier::ResolveClassAndCheckAccess(uint32_t class_idx) {
const char* descriptor = dex_file_->StringByTypeIdx(class_idx);
const RegType& referrer = GetDeclaringClass();
Class* klass = dex_cache_->GetResolvedType(class_idx);
const RegType& result =
klass != NULL ? reg_types_.FromClass(klass)
: reg_types_.FromDescriptor(class_loader_, descriptor);
if (result.IsConflict()) {
Fail(VERIFY_ERROR_BAD_CLASS_SOFT) << "accessing broken descriptor '" << descriptor
<< "' in " << referrer;
return result;
}
if (klass == NULL && !result.IsUnresolvedTypes()) {
dex_cache_->SetResolvedType(class_idx, result.GetClass());
}
// Check if access is allowed. Unresolved types use xxxWithAccessCheck to
// check at runtime if access is allowed and so pass here.
if (!result.IsUnresolvedTypes() && !referrer.IsUnresolvedTypes() && !referrer.CanAccess(result)) {
Fail(VERIFY_ERROR_ACCESS_CLASS) << "illegal class access: '"
<< referrer << "' -> '" << result << "'";
}
return result;
}
const RegType& MethodVerifier::GetCaughtExceptionType() {
const RegType* common_super = NULL;
if (code_item_->tries_size_ != 0) {
const byte* handlers_ptr = DexFile::GetCatchHandlerData(*code_item_, 0);
uint32_t handlers_size = DecodeUnsignedLeb128(&handlers_ptr);
for (uint32_t i = 0; i < handlers_size; i++) {
CatchHandlerIterator iterator(handlers_ptr);
for (; iterator.HasNext(); iterator.Next()) {
if (iterator.GetHandlerAddress() == (uint32_t) work_insn_idx_) {
if (iterator.GetHandlerTypeIndex() == DexFile::kDexNoIndex16) {
common_super = &reg_types_.JavaLangThrowable();
} else {
const RegType& exception = ResolveClassAndCheckAccess(iterator.GetHandlerTypeIndex());
if (common_super == NULL) {
// Unconditionally assign for the first handler. We don't assert this is a Throwable
// as that is caught at runtime
common_super = &exception;
} else if (!reg_types_.JavaLangThrowable().IsAssignableFrom(exception)) {
// We don't know enough about the type and the common path merge will result in
// Conflict. Fail here knowing the correct thing can be done at runtime.
Fail(VERIFY_ERROR_BAD_CLASS_SOFT) << "unexpected non-exception class " << exception;
return reg_types_.Conflict();
} else if (common_super->Equals(exception)) {
// odd case, but nothing to do
} else {
common_super = &common_super->Merge(exception, &reg_types_);
CHECK(reg_types_.JavaLangThrowable().IsAssignableFrom(*common_super));
}
}
}
}
handlers_ptr = iterator.EndDataPointer();
}
}
if (common_super == NULL) {
/* no catch blocks, or no catches with classes we can find */
Fail(VERIFY_ERROR_BAD_CLASS_SOFT) << "unable to find exception handler";
return reg_types_.Conflict();
}
return *common_super;
}
Method* MethodVerifier::ResolveMethodAndCheckAccess(uint32_t dex_method_idx, MethodType method_type) {
const DexFile::MethodId& method_id = dex_file_->GetMethodId(dex_method_idx);
const RegType& klass_type = ResolveClassAndCheckAccess(method_id.class_idx_);
if (klass_type.IsConflict()) {
std::string append(" in attempt to access method ");
append += dex_file_->GetMethodName(method_id);
AppendToLastFailMessage(append);
return NULL;
}
if (klass_type.IsUnresolvedTypes()) {
return NULL; // Can't resolve Class so no more to do here
}
Class* klass = klass_type.GetClass();
const RegType& referrer = GetDeclaringClass();
Method* res_method = dex_cache_->GetResolvedMethod(dex_method_idx);
if (res_method == NULL) {
const char* name = dex_file_->GetMethodName(method_id);
std::string signature(dex_file_->CreateMethodSignature(method_id.proto_idx_, NULL));
if (method_type == METHOD_DIRECT || method_type == METHOD_STATIC) {
res_method = klass->FindDirectMethod(name, signature);
} else if (method_type == METHOD_INTERFACE) {
res_method = klass->FindInterfaceMethod(name, signature);
} else {
res_method = klass->FindVirtualMethod(name, signature);
}
if (res_method != NULL) {
dex_cache_->SetResolvedMethod(dex_method_idx, res_method);
} else {
// If a virtual or interface method wasn't found with the expected type, look in
// the direct methods. This can happen when the wrong invoke type is used or when
// a class has changed, and will be flagged as an error in later checks.
if (method_type == METHOD_INTERFACE || method_type == METHOD_VIRTUAL) {
res_method = klass->FindDirectMethod(name, signature);
}
if (res_method == NULL) {
Fail(VERIFY_ERROR_NO_METHOD) << "couldn't find method "
<< PrettyDescriptor(klass) << "." << name
<< " " << signature;
return NULL;
}
}
}
// Make sure calls to constructors are "direct". There are additional restrictions but we don't
// enforce them here.
if (res_method->IsConstructor() && method_type != METHOD_DIRECT) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "rejecting non-direct call to constructor "
<< PrettyMethod(res_method);
return NULL;
}
// Disallow any calls to class initializers.
if (MethodHelper(res_method).IsClassInitializer()) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "rejecting call to class initializer "
<< PrettyMethod(res_method);
return NULL;
}
// Check if access is allowed.
if (!referrer.CanAccessMember(res_method->GetDeclaringClass(), res_method->GetAccessFlags())) {
Fail(VERIFY_ERROR_ACCESS_METHOD) << "illegal method access (call " << PrettyMethod(res_method)
<< " from " << referrer << ")";
return res_method;
}
// Check that invoke-virtual and invoke-super are not used on private methods of the same class.
if (res_method->IsPrivate() && method_type == METHOD_VIRTUAL) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "invoke-super/virtual can't be used on private method "
<< PrettyMethod(res_method);
return NULL;
}
// Check that interface methods match interface classes.
if (klass->IsInterface() && method_type != METHOD_INTERFACE) {
Fail(VERIFY_ERROR_CLASS_CHANGE) << "non-interface method " << PrettyMethod(res_method)
<< " is in an interface class " << PrettyClass(klass);
return NULL;
} else if (!klass->IsInterface() && method_type == METHOD_INTERFACE) {
Fail(VERIFY_ERROR_CLASS_CHANGE) << "interface method " << PrettyMethod(res_method)
<< " is in a non-interface class " << PrettyClass(klass);
return NULL;
}
// See if the method type implied by the invoke instruction matches the access flags for the
// target method.
if ((method_type == METHOD_DIRECT && !res_method->IsDirect()) ||
(method_type == METHOD_STATIC && !res_method->IsStatic()) ||
((method_type == METHOD_VIRTUAL || method_type == METHOD_INTERFACE) && res_method->IsDirect())
) {
Fail(VERIFY_ERROR_CLASS_CHANGE) << "invoke type does not match method type of "
<< PrettyMethod(res_method);
return NULL;
}
return res_method;
}
Method* MethodVerifier::VerifyInvocationArgs(const DecodedInstruction& dec_insn,
MethodType method_type, bool is_range, bool is_super) {
// Resolve the method. This could be an abstract or concrete method depending on what sort of call
// we're making.
Method* res_method = ResolveMethodAndCheckAccess(dec_insn.vB, method_type);
if (res_method == NULL) { // error or class is unresolved
return NULL;
}
// If we're using invoke-super(method), make sure that the executing method's class' superclass
// has a vtable entry for the target method.
if (is_super) {
DCHECK(method_type == METHOD_VIRTUAL);
const RegType& super = GetDeclaringClass().GetSuperClass(&reg_types_);
if (super.IsConflict()) { // unknown super class
Fail(VERIFY_ERROR_NO_METHOD) << "unknown super class in invoke-super from "
<< PrettyMethod(method_idx_, *dex_file_)
<< " to super " << PrettyMethod(res_method);
return NULL;
}
Class* super_klass = super.GetClass();
if (res_method->GetMethodIndex() >= super_klass->GetVTable()->GetLength()) {
MethodHelper mh(res_method);
Fail(VERIFY_ERROR_NO_METHOD) << "invalid invoke-super from "
<< PrettyMethod(method_idx_, *dex_file_)
<< " to super " << super
<< "." << mh.GetName()
<< mh.GetSignature();
return NULL;
}
}
// We use vAA as our expected arg count, rather than res_method->insSize, because we need to
// match the call to the signature. Also, we might might be calling through an abstract method
// definition (which doesn't have register count values).
size_t expected_args = dec_insn.vA;
/* caught by static verifier */
DCHECK(is_range || expected_args <= 5);
if (expected_args > code_item_->outs_size_) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "invalid argument count (" << expected_args
<< ") exceeds outsSize (" << code_item_->outs_size_ << ")";
return NULL;
}
/*
* Check the "this" argument, which must be an instance of the class that declared the method.
* For an interface class, we don't do the full interface merge (see JoinClass), so we can't do a
* rigorous check here (which is okay since we have to do it at runtime).
*/
size_t actual_args = 0;
if (!res_method->IsStatic()) {
const RegType& actual_arg_type = work_line_->GetInvocationThis(dec_insn);
if (actual_arg_type.IsConflict()) { // GetInvocationThis failed.
return NULL;
}
if (actual_arg_type.IsUninitializedReference() && !res_method->IsConstructor()) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "'this' arg must be initialized";
return NULL;
}
if (method_type != METHOD_INTERFACE && !actual_arg_type.IsZero()) {
const RegType& res_method_class = reg_types_.FromClass(res_method->GetDeclaringClass());
if (!res_method_class.IsAssignableFrom(actual_arg_type)) {
Fail(VERIFY_ERROR_BAD_CLASS_SOFT) << "'this' argument '" << actual_arg_type
<< "' not instance of '" << res_method_class << "'";
return NULL;
}
}
actual_args++;
}
/*
* Process the target method's signature. This signature may or may not
* have been verified, so we can't assume it's properly formed.
*/
MethodHelper mh(res_method);
const DexFile::TypeList* params = mh.GetParameterTypeList();
size_t params_size = params == NULL ? 0 : params->Size();
for (size_t param_index = 0; param_index < params_size; param_index++) {
if (actual_args >= expected_args) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "Rejecting invalid call to '" << PrettyMethod(res_method)
<< "'. Expected " << expected_args << " arguments, processing argument " << actual_args
<< " (where longs/doubles count twice).";
return NULL;
}
const char* descriptor =
mh.GetTypeDescriptorFromTypeIdx(params->GetTypeItem(param_index).type_idx_);
if (descriptor == NULL) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "Rejecting invocation of " << PrettyMethod(res_method)
<< " missing signature component";
return NULL;
}
const RegType& reg_type = reg_types_.FromDescriptor(class_loader_, descriptor);
uint32_t get_reg = is_range ? dec_insn.vC + actual_args : dec_insn.arg[actual_args];
if (!work_line_->VerifyRegisterType(get_reg, reg_type)) {
return res_method;
}
actual_args = reg_type.IsLongOrDoubleTypes() ? actual_args + 2 : actual_args + 1;
}
if (actual_args != expected_args) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "Rejecting invocation of " << PrettyMethod(res_method)
<< " expected " << expected_args << " arguments, found " << actual_args;
return NULL;
} else {
return res_method;
}
}
void MethodVerifier::VerifyNewArray(const DecodedInstruction& dec_insn, bool is_filled,
bool is_range) {
const RegType& res_type = ResolveClassAndCheckAccess(is_filled ? dec_insn.vB : dec_insn.vC);
if (res_type.IsConflict()) { // bad class
DCHECK_NE(failures_.size(), 0U);
} else {
// TODO: check Compiler::CanAccessTypeWithoutChecks returns false when res_type is unresolved
if (!res_type.IsArrayTypes()) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "new-array on non-array class " << res_type;
} else if (!is_filled) {
/* make sure "size" register is valid type */
work_line_->VerifyRegisterType(dec_insn.vB, reg_types_.Integer());
/* set register type to array class */
work_line_->SetRegisterType(dec_insn.vA, res_type);
} else {
// Verify each register. If "arg_count" is bad, VerifyRegisterType() will run off the end of
// the list and fail. It's legal, if silly, for arg_count to be zero.
const RegType& expected_type = reg_types_.GetComponentType(res_type, class_loader_);
uint32_t arg_count = dec_insn.vA;
for (size_t ui = 0; ui < arg_count; ui++) {
uint32_t get_reg = is_range ? dec_insn.vC + ui : dec_insn.arg[ui];
if (!work_line_->VerifyRegisterType(get_reg, expected_type)) {
work_line_->SetResultRegisterType(reg_types_.Conflict());
return;
}
}
// filled-array result goes into "result" register
work_line_->SetResultRegisterType(res_type);
}
}
}
void MethodVerifier::VerifyAGet(const DecodedInstruction& dec_insn,
const RegType& insn_type, bool is_primitive) {
const RegType& index_type = work_line_->GetRegisterType(dec_insn.vC);
if (!index_type.IsArrayIndexTypes()) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "Invalid reg type for array index (" << index_type << ")";
} else {
const RegType& array_type = work_line_->GetRegisterType(dec_insn.vB);
if (array_type.IsZero()) {
// Null array class; this code path will fail at runtime. Infer a merge-able type from the
// instruction type. TODO: have a proper notion of bottom here.
if (!is_primitive || insn_type.IsCategory1Types()) {
// Reference or category 1
work_line_->SetRegisterType(dec_insn.vA, reg_types_.Zero());
} else {
// Category 2
work_line_->SetRegisterType(dec_insn.vA, reg_types_.ConstLo());
}
} else if (!array_type.IsArrayTypes()) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "not array type " << array_type << " with aget";
} else {
/* verify the class */
const RegType& component_type = reg_types_.GetComponentType(array_type, class_loader_);
if (!component_type.IsReferenceTypes() && !is_primitive) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "primitive array type " << array_type
<< " source for aget-object";
} else if (component_type.IsNonZeroReferenceTypes() && is_primitive) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "reference array type " << array_type
<< " source for category 1 aget";
} else if (is_primitive && !insn_type.Equals(component_type) &&
!((insn_type.IsInteger() && component_type.IsFloat()) ||
(insn_type.IsLong() && component_type.IsDouble()))) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "array type " << array_type
<< " incompatible with aget of type " << insn_type;
} else {
// Use knowledge of the field type which is stronger than the type inferred from the
// instruction, which can't differentiate object types and ints from floats, longs from
// doubles.
work_line_->SetRegisterType(dec_insn.vA, component_type);
}
}
}
}
void MethodVerifier::VerifyAPut(const DecodedInstruction& dec_insn,
const RegType& insn_type, bool is_primitive) {
const RegType& index_type = work_line_->GetRegisterType(dec_insn.vC);
if (!index_type.IsArrayIndexTypes()) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "Invalid reg type for array index (" << index_type << ")";
} else {
const RegType& array_type = work_line_->GetRegisterType(dec_insn.vB);
if (array_type.IsZero()) {
// Null array type; this code path will fail at runtime. Infer a merge-able type from the
// instruction type.
} else if (!array_type.IsArrayTypes()) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "not array type " << array_type << " with aput";
} else {
/* verify the class */
const RegType& component_type = reg_types_.GetComponentType(array_type, class_loader_);
if (!component_type.IsReferenceTypes() && !is_primitive) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "primitive array type " << array_type
<< " source for aput-object";
} else if (component_type.IsNonZeroReferenceTypes() && is_primitive) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "reference array type " << array_type
<< " source for category 1 aput";
} else if (is_primitive && !insn_type.Equals(component_type) &&
!((insn_type.IsInteger() && component_type.IsFloat()) ||
(insn_type.IsLong() && component_type.IsDouble()))) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "array type " << array_type
<< " incompatible with aput of type " << insn_type;
} else {
// The instruction agrees with the type of array, confirm the value to be stored does too
// Note: we use the instruction type (rather than the component type) for aput-object as
// incompatible classes will be caught at runtime as an array store exception
work_line_->VerifyRegisterType(dec_insn.vA, is_primitive ? component_type : insn_type);
}
}
}
}
Field* MethodVerifier::GetStaticField(int field_idx) {
const DexFile::FieldId& field_id = dex_file_->GetFieldId(field_idx);
// Check access to class
const RegType& klass_type = ResolveClassAndCheckAccess(field_id.class_idx_);
if (klass_type.IsConflict()) { // bad class
AppendToLastFailMessage(StringPrintf(" in attempt to access static field %d (%s) in %s",
field_idx, dex_file_->GetFieldName(field_id),
dex_file_->GetFieldDeclaringClassDescriptor(field_id)));
return NULL;
}
if (klass_type.IsUnresolvedTypes()) {
return NULL; // Can't resolve Class so no more to do here, will do checking at runtime.
}
Field* field = Runtime::Current()->GetClassLinker()->ResolveFieldJLS(*dex_file_, field_idx,
dex_cache_, class_loader_);
if (field == NULL) {
LOG(INFO) << "unable to resolve static field " << field_idx << " ("
<< dex_file_->GetFieldName(field_id) << ") in "
<< dex_file_->GetFieldDeclaringClassDescriptor(field_id);
DCHECK(Thread::Current()->IsExceptionPending());
Thread::Current()->ClearException();
return NULL;
} else if (!GetDeclaringClass().CanAccessMember(field->GetDeclaringClass(),
field->GetAccessFlags())) {
Fail(VERIFY_ERROR_ACCESS_FIELD) << "cannot access static field " << PrettyField(field)
<< " from " << GetDeclaringClass();
return NULL;
} else if (!field->IsStatic()) {
Fail(VERIFY_ERROR_CLASS_CHANGE) << "expected field " << PrettyField(field) << " to be static";
return NULL;
} else {
return field;
}
}
Field* MethodVerifier::GetInstanceField(const RegType& obj_type, int field_idx) {
const DexFile::FieldId& field_id = dex_file_->GetFieldId(field_idx);
// Check access to class
const RegType& klass_type = ResolveClassAndCheckAccess(field_id.class_idx_);
if (klass_type.IsConflict()) {
AppendToLastFailMessage(StringPrintf(" in attempt to access instance field %d (%s) in %s",
field_idx, dex_file_->GetFieldName(field_id),
dex_file_->GetFieldDeclaringClassDescriptor(field_id)));
return NULL;
}
if (klass_type.IsUnresolvedTypes()) {
return NULL; // Can't resolve Class so no more to do here
}
Field* field = Runtime::Current()->GetClassLinker()->ResolveFieldJLS(*dex_file_, field_idx,
dex_cache_, class_loader_);
if (field == NULL) {
LOG(INFO) << "unable to resolve instance field " << field_idx << " ("
<< dex_file_->GetFieldName(field_id) << ") in "
<< dex_file_->GetFieldDeclaringClassDescriptor(field_id);
DCHECK(Thread::Current()->IsExceptionPending());
Thread::Current()->ClearException();
return NULL;
} else if (!GetDeclaringClass().CanAccessMember(field->GetDeclaringClass(),
field->GetAccessFlags())) {
Fail(VERIFY_ERROR_ACCESS_FIELD) << "cannot access instance field " << PrettyField(field)
<< " from " << GetDeclaringClass();
return NULL;
} else if (field->IsStatic()) {
Fail(VERIFY_ERROR_CLASS_CHANGE) << "expected field " << PrettyField(field)
<< " to not be static";
return NULL;
} else if (obj_type.IsZero()) {
// Cannot infer and check type, however, access will cause null pointer exception
return field;
} else {
const RegType& field_klass = reg_types_.FromClass(field->GetDeclaringClass());
if (obj_type.IsUninitializedTypes() &&
(!IsConstructor() || GetDeclaringClass().Equals(obj_type) ||
!field_klass.Equals(GetDeclaringClass()))) {
// Field accesses through uninitialized references are only allowable for constructors where
// the field is declared in this class
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "cannot access instance field " << PrettyField(field)
<< " of a not fully initialized object within the context of "
<< PrettyMethod(method_idx_, *dex_file_);
return NULL;
} else if (!field_klass.IsAssignableFrom(obj_type)) {
// Trying to access C1.field1 using reference of type C2, which is neither C1 or a sub-class
// of C1. For resolution to occur the declared class of the field must be compatible with
// obj_type, we've discovered this wasn't so, so report the field didn't exist.
Fail(VERIFY_ERROR_NO_FIELD) << "cannot access instance field " << PrettyField(field)
<< " from object of type " << obj_type;
return NULL;
} else {
return field;
}
}
}
void MethodVerifier::VerifyISGet(const DecodedInstruction& dec_insn,
const RegType& insn_type, bool is_primitive, bool is_static) {
uint32_t field_idx = is_static ? dec_insn.vB : dec_insn.vC;
Field* field;
if (is_static) {
field = GetStaticField(field_idx);
} else {
const RegType& object_type = work_line_->GetRegisterType(dec_insn.vB);
field = GetInstanceField(object_type, field_idx);
}
const char* descriptor;
ClassLoader* loader;
if (field != NULL) {
descriptor = FieldHelper(field).GetTypeDescriptor();
loader = field->GetDeclaringClass()->GetClassLoader();
} else {
const DexFile::FieldId& field_id = dex_file_->GetFieldId(field_idx);
descriptor = dex_file_->GetFieldTypeDescriptor(field_id);
loader = class_loader_;
}
const RegType& field_type = reg_types_.FromDescriptor(loader, descriptor);
if (is_primitive) {
if (field_type.Equals(insn_type) ||
(field_type.IsFloat() && insn_type.IsIntegralTypes()) ||
(field_type.IsDouble() && insn_type.IsLongTypes())) {
// expected that read is of the correct primitive type or that int reads are reading
// floats or long reads are reading doubles
} else {
// This is a global failure rather than a class change failure as the instructions and
// the descriptors for the type should have been consistent within the same file at
// compile time
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "expected field " << PrettyField(field)
<< " to be of type '" << insn_type
<< "' but found type '" << field_type << "' in get";
return;
}
} else {
if (!insn_type.IsAssignableFrom(field_type)) {
Fail(VERIFY_ERROR_BAD_CLASS_SOFT) << "expected field " << PrettyField(field)
<< " to be compatible with type '" << insn_type
<< "' but found type '" << field_type
<< "' in get-object";
work_line_->SetRegisterType(dec_insn.vA, reg_types_.Conflict());
return;
}
}
work_line_->SetRegisterType(dec_insn.vA, field_type);
}
void MethodVerifier::VerifyISPut(const DecodedInstruction& dec_insn,
const RegType& insn_type, bool is_primitive, bool is_static) {
uint32_t field_idx = is_static ? dec_insn.vB : dec_insn.vC;
Field* field;
if (is_static) {
field = GetStaticField(field_idx);
} else {
const RegType& object_type = work_line_->GetRegisterType(dec_insn.vB);
field = GetInstanceField(object_type, field_idx);
}
const char* descriptor;
ClassLoader* loader;
if (field != NULL) {
descriptor = FieldHelper(field).GetTypeDescriptor();
loader = field->GetDeclaringClass()->GetClassLoader();
} else {
const DexFile::FieldId& field_id = dex_file_->GetFieldId(field_idx);
descriptor = dex_file_->GetFieldTypeDescriptor(field_id);
loader = class_loader_;
}
const RegType& field_type = reg_types_.FromDescriptor(loader, descriptor);
if (field != NULL) {
if (field->IsFinal() && field->GetDeclaringClass() != GetDeclaringClass().GetClass()) {
Fail(VERIFY_ERROR_ACCESS_FIELD) << "cannot modify final field " << PrettyField(field)
<< " from other class " << GetDeclaringClass();
return;
}
}
if (is_primitive) {
// Primitive field assignability rules are weaker than regular assignability rules
bool instruction_compatible;
bool value_compatible;
const RegType& value_type = work_line_->GetRegisterType(dec_insn.vA);
if (field_type.IsIntegralTypes()) {
instruction_compatible = insn_type.IsIntegralTypes();
value_compatible = value_type.IsIntegralTypes();
} else if (field_type.IsFloat()) {
instruction_compatible = insn_type.IsInteger(); // no [is]put-float, so expect [is]put-int
value_compatible = value_type.IsFloatTypes();
} else if (field_type.IsLong()) {
instruction_compatible = insn_type.IsLong();
value_compatible = value_type.IsLongTypes();
} else if (field_type.IsDouble()) {
instruction_compatible = insn_type.IsLong(); // no [is]put-double, so expect [is]put-long
value_compatible = value_type.IsDoubleTypes();
} else {
instruction_compatible = false; // reference field with primitive store
value_compatible = false; // unused
}
if (!instruction_compatible) {
// This is a global failure rather than a class change failure as the instructions and
// the descriptors for the type should have been consistent within the same file at
// compile time
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "expected field " << PrettyField(field)
<< " to be of type '" << insn_type
<< "' but found type '" << field_type
<< "' in put";
return;
}
if (!value_compatible) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "unexpected value in v" << dec_insn.vA
<< " of type " << value_type
<< " but expected " << field_type
<< " for store to " << PrettyField(field) << " in put";
return;
}
} else {
if (!insn_type.IsAssignableFrom(field_type)) {
Fail(VERIFY_ERROR_BAD_CLASS_SOFT) << "expected field " << PrettyField(field)
<< " to be compatible with type '" << insn_type
<< "' but found type '" << field_type
<< "' in put-object";
return;
}
work_line_->VerifyRegisterType(dec_insn.vA, field_type);
}
}
bool MethodVerifier::CheckNotMoveException(const uint16_t* insns, int insn_idx) {
if ((insns[insn_idx] & 0xff) == Instruction::MOVE_EXCEPTION) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "invalid use of move-exception";
return false;
}
return true;
}
void MethodVerifier::ReplaceFailingInstruction() {
// Pop the failure and clear the need for rewriting.
size_t failure_number = failures_.size();
CHECK_NE(failure_number, 0U);
DCHECK_EQ(failure_messages_.size(), failure_number);
std::ostringstream* failure_message = failure_messages_[0];
VerifyError failure = failures_[0];
failures_.clear();
failure_messages_.clear();
have_pending_rewrite_failure_ = false;
if (Runtime::Current()->IsStarted()) {
LOG(ERROR) << "Verification attempting to replace instructions at runtime in "
<< PrettyMethod(method_idx_, *dex_file_) << " " << failure_message->str();
return;
}
CHECK_EQ(dex_file_->GetPermissions() & PROT_WRITE, PROT_WRITE); // Dex file needs to be writable.
const Instruction* inst = Instruction::At(code_item_->insns_ + work_insn_idx_);
DCHECK(inst->IsThrow()) << "Expected instruction that will throw " << inst->Name();
VerifyErrorRefType ref_type;
switch (inst->Opcode()) {
case Instruction::CONST_CLASS: // insn[1] == class ref, 2 code units (4 bytes)
case Instruction::CHECK_CAST:
case Instruction::INSTANCE_OF:
case Instruction::NEW_INSTANCE:
case Instruction::NEW_ARRAY:
case Instruction::FILLED_NEW_ARRAY: // insn[1] == class ref, 3 code units (6 bytes)
case Instruction::FILLED_NEW_ARRAY_RANGE:
ref_type = VERIFY_ERROR_REF_CLASS;
break;
case Instruction::IGET: // insn[1] == field ref, 2 code units (4 bytes)
case Instruction::IGET_BOOLEAN:
case Instruction::IGET_BYTE:
case Instruction::IGET_CHAR:
case Instruction::IGET_SHORT:
case Instruction::IGET_WIDE:
case Instruction::IGET_OBJECT:
case Instruction::IPUT:
case Instruction::IPUT_BOOLEAN:
case Instruction::IPUT_BYTE:
case Instruction::IPUT_CHAR:
case Instruction::IPUT_SHORT:
case Instruction::IPUT_WIDE:
case Instruction::IPUT_OBJECT:
case Instruction::SGET:
case Instruction::SGET_BOOLEAN:
case Instruction::SGET_BYTE:
case Instruction::SGET_CHAR:
case Instruction::SGET_SHORT:
case Instruction::SGET_WIDE:
case Instruction::SGET_OBJECT:
case Instruction::SPUT:
case Instruction::SPUT_BOOLEAN:
case Instruction::SPUT_BYTE:
case Instruction::SPUT_CHAR:
case Instruction::SPUT_SHORT:
case Instruction::SPUT_WIDE:
case Instruction::SPUT_OBJECT:
ref_type = VERIFY_ERROR_REF_FIELD;
break;
case Instruction::INVOKE_VIRTUAL: // insn[1] == method ref, 3 code units (6 bytes)
case Instruction::INVOKE_VIRTUAL_RANGE:
case Instruction::INVOKE_SUPER:
case Instruction::INVOKE_SUPER_RANGE:
case Instruction::INVOKE_DIRECT:
case Instruction::INVOKE_DIRECT_RANGE:
case Instruction::INVOKE_STATIC:
case Instruction::INVOKE_STATIC_RANGE:
case Instruction::INVOKE_INTERFACE:
case Instruction::INVOKE_INTERFACE_RANGE:
ref_type = VERIFY_ERROR_REF_METHOD;
break;
default:
LOG(FATAL) << "Error: verifier asked to replace instruction " << inst->DumpString(dex_file_);
return;
}
uint16_t* insns = const_cast<uint16_t*>(code_item_->insns_);
// THROW_VERIFICATION_ERROR is a 2 code unit instruction. We shouldn't be rewriting a 1 code unit
// instruction, so assert it.
size_t width = inst->SizeInCodeUnits();
CHECK_GT(width, 1u);
// If the instruction is larger than 2 code units, rewrite subsequent code unit sized chunks with
// NOPs
for (size_t i = 2; i < width; i++) {
insns[work_insn_idx_ + i] = Instruction::NOP;
}
// Encode the opcode, with the failure code in the high byte
uint16_t new_instruction = Instruction::THROW_VERIFICATION_ERROR |
(failure << 8) | // AA - component
(ref_type << (8 + kVerifyErrorRefTypeShift));
insns[work_insn_idx_] = new_instruction;
// The 2nd code unit (higher in memory) with the reference in, comes from the instruction we
// rewrote, so nothing to do here.
LOG(INFO) << "Verification error, replacing instructions in "
<< PrettyMethod(method_idx_, *dex_file_) << " "
<< failure_message->str();
if (gDebugVerify) {
std::cout << "\n" << info_messages_.str();
Dump(std::cout);
}
}
bool MethodVerifier::UpdateRegisters(uint32_t next_insn, const RegisterLine* merge_line) {
bool changed = true;
RegisterLine* target_line = reg_table_.GetLine(next_insn);
if (!insn_flags_[next_insn].IsVisitedOrChanged()) {
/*
* We haven't processed this instruction before, and we haven't touched the registers here, so
* there's nothing to "merge". Copy the registers over and mark it as changed. (This is the
* only way a register can transition out of "unknown", so this is not just an optimization.)
*/
target_line->CopyFromLine(merge_line);
} else {
UniquePtr<RegisterLine> copy(gDebugVerify ? new RegisterLine(target_line->NumRegs(), this) : NULL);
if (gDebugVerify) {
copy->CopyFromLine(target_line);
}
changed = target_line->MergeRegisters(merge_line);
if (have_pending_hard_failure_) {
return false;
}
if (gDebugVerify && changed) {
LogVerifyInfo() << "Merging at [" << reinterpret_cast<void*>(work_insn_idx_) << "]"
<< " to [" << reinterpret_cast<void*>(next_insn) << "]: " << "\n"
<< *copy.get() << " MERGE\n"
<< *merge_line << " ==\n"
<< *target_line << "\n";
}
}
if (changed) {
insn_flags_[next_insn].SetChanged();
}
return true;
}
InsnFlags* MethodVerifier::CurrentInsnFlags() {
return &insn_flags_[work_insn_idx_];
}
const RegType& MethodVerifier::GetMethodReturnType() {
const DexFile::MethodId& method_id = dex_file_->GetMethodId(method_idx_);
const DexFile::ProtoId& proto_id = dex_file_->GetMethodPrototype(method_id);
uint16_t return_type_idx = proto_id.return_type_idx_;
const char* descriptor = dex_file_->GetTypeDescriptor(dex_file_->GetTypeId(return_type_idx));
return reg_types_.FromDescriptor(class_loader_, descriptor);
}
const RegType& MethodVerifier::GetDeclaringClass() {
if (foo_method_ != NULL) {
return reg_types_.FromClass(foo_method_->GetDeclaringClass());
} else {
const DexFile::MethodId& method_id = dex_file_->GetMethodId(method_idx_);
const char* descriptor = dex_file_->GetTypeDescriptor(dex_file_->GetTypeId(method_id.class_idx_));
return reg_types_.FromDescriptor(class_loader_, descriptor);
}
}
void MethodVerifier::ComputeGcMapSizes(size_t* gc_points, size_t* ref_bitmap_bits,
size_t* log2_max_gc_pc) {
size_t local_gc_points = 0;
size_t max_insn = 0;
size_t max_ref_reg = -1;
for (size_t i = 0; i < code_item_->insns_size_in_code_units_; i++) {
if (insn_flags_[i].IsGcPoint()) {
local_gc_points++;
max_insn = i;
RegisterLine* line = reg_table_.GetLine(i);
max_ref_reg = line->GetMaxNonZeroReferenceReg(max_ref_reg);
}
}
*gc_points = local_gc_points;
*ref_bitmap_bits = max_ref_reg + 1; // if max register is 0 we need 1 bit to encode (ie +1)
size_t i = 0;
while ((1U << i) <= max_insn) {
i++;
}
*log2_max_gc_pc = i;
}
const std::vector<uint8_t>* MethodVerifier::GenerateGcMap() {
size_t num_entries, ref_bitmap_bits, pc_bits;
ComputeGcMapSizes(&num_entries, &ref_bitmap_bits, &pc_bits);
// There's a single byte to encode the size of each bitmap
if (ref_bitmap_bits >= (8 /* bits per byte */ * 8192 /* 13-bit size */ )) {
// TODO: either a better GC map format or per method failures
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "Cannot encode GC map for method with "
<< ref_bitmap_bits << " registers";
return NULL;
}
size_t ref_bitmap_bytes = (ref_bitmap_bits + 7) / 8;
// There are 2 bytes to encode the number of entries
if (num_entries >= 65536) {
// TODO: either a better GC map format or per method failures
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "Cannot encode GC map for method with "
<< num_entries << " entries";
return NULL;
}
size_t pc_bytes;
RegisterMapFormat format;
if (pc_bits <= 8) {
format = kRegMapFormatCompact8;
pc_bytes = 1;
} else if (pc_bits <= 16) {
format = kRegMapFormatCompact16;
pc_bytes = 2;
} else {
// TODO: either a better GC map format or per method failures
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "Cannot encode GC map for method with "
<< (1 << pc_bits) << " instructions (number is rounded up to nearest power of 2)";
return NULL;
}
size_t table_size = ((pc_bytes + ref_bitmap_bytes) * num_entries) + 4;
std::vector<uint8_t>* table = new std::vector<uint8_t>;
if (table == NULL) {
Fail(VERIFY_ERROR_BAD_CLASS_HARD) << "Failed to encode GC map (size=" << table_size << ")";
return NULL;
}
// Write table header
table->push_back(format | ((ref_bitmap_bytes >> PcToReferenceMap::kRegMapFormatShift) &
~PcToReferenceMap::kRegMapFormatMask));
table->push_back(ref_bitmap_bytes & 0xFF);
table->push_back(num_entries & 0xFF);
table->push_back((num_entries >> 8) & 0xFF);
// Write table data
for (size_t i = 0; i < code_item_->insns_size_in_code_units_; i++) {
if (insn_flags_[i].IsGcPoint()) {
table->push_back(i & 0xFF);
if (pc_bytes == 2) {
table->push_back((i >> 8) & 0xFF);
}
RegisterLine* line = reg_table_.GetLine(i);
line->WriteReferenceBitMap(*table, ref_bitmap_bytes);
}
}
DCHECK_EQ(table->size(), table_size);
return table;
}
void MethodVerifier::VerifyGcMap(const std::vector<uint8_t>& data) {
// Check that for every GC point there is a map entry, there aren't entries for non-GC points,
// that the table data is well formed and all references are marked (or not) in the bitmap
PcToReferenceMap map(&data[0], data.size());
size_t map_index = 0;
for (size_t i = 0; i < code_item_->insns_size_in_code_units_; i++) {
const uint8_t* reg_bitmap = map.FindBitMap(i, false);
if (insn_flags_[i].IsGcPoint()) {
CHECK_LT(map_index, map.NumEntries());
CHECK_EQ(map.GetPC(map_index), i);
CHECK_EQ(map.GetBitMap(map_index), reg_bitmap);
map_index++;
RegisterLine* line = reg_table_.GetLine(i);
for (size_t j = 0; j < code_item_->registers_size_; j++) {
if (line->GetRegisterType(j).IsNonZeroReferenceTypes()) {
CHECK_LT(j / 8, map.RegWidth());
CHECK_EQ((reg_bitmap[j / 8] >> (j % 8)) & 1, 1);
} else if ((j / 8) < map.RegWidth()) {
CHECK_EQ((reg_bitmap[j / 8] >> (j % 8)) & 1, 0);
} else {
// If a register doesn't contain a reference then the bitmap may be shorter than the line
}
}
} else {
CHECK(reg_bitmap == NULL);
}
}
}
void MethodVerifier::SetGcMap(Compiler::MethodReference ref, const std::vector<uint8_t>& gc_map) {
MutexLock mu(*gc_maps_lock_);
GcMapTable::iterator it = gc_maps_->find(ref);
if (it != gc_maps_->end()) {
delete it->second;
gc_maps_->erase(it);
}
gc_maps_->Put(ref, &gc_map);
CHECK(GetGcMap(ref) != NULL);
}
const std::vector<uint8_t>* MethodVerifier::GetGcMap(Compiler::MethodReference ref) {
MutexLock mu(*gc_maps_lock_);
GcMapTable::const_iterator it = gc_maps_->find(ref);
if (it == gc_maps_->end()) {
return NULL;
}
CHECK(it->second != NULL);
return it->second;
}
Mutex* MethodVerifier::gc_maps_lock_ = NULL;
MethodVerifier::GcMapTable* MethodVerifier::gc_maps_ = NULL;
Mutex* MethodVerifier::rejected_classes_lock_ = NULL;
MethodVerifier::RejectedClassesTable* MethodVerifier::rejected_classes_ = NULL;
#if defined(ART_USE_LLVM_COMPILER) || defined(ART_USE_GREENLAND_COMPILER)
Mutex* MethodVerifier::inferred_reg_category_maps_lock_ = NULL;
MethodVerifier::InferredRegCategoryMapTable* MethodVerifier::inferred_reg_category_maps_ = NULL;
#endif
void MethodVerifier::Init() {
gc_maps_lock_ = new Mutex("verifier GC maps lock");
{
MutexLock mu(*gc_maps_lock_);
gc_maps_ = new MethodVerifier::GcMapTable;
}
rejected_classes_lock_ = new Mutex("verifier rejected classes lock");
{
MutexLock mu(*rejected_classes_lock_);
rejected_classes_ = new MethodVerifier::RejectedClassesTable;
}
#if defined(ART_USE_LLVM_COMPILER) || defined(ART_USE_GREENLAND_COMPILER)
inferred_reg_category_maps_lock_ = new Mutex("verifier GC maps lock");
{
MutexLock mu(*inferred_reg_category_maps_lock_);
inferred_reg_category_maps_ = new MethodVerifier::InferredRegCategoryMapTable;
}
#endif
}
void MethodVerifier::Shutdown() {
{
MutexLock mu(*gc_maps_lock_);
STLDeleteValues(gc_maps_);
delete gc_maps_;
gc_maps_ = NULL;
}
delete gc_maps_lock_;
gc_maps_lock_ = NULL;
{
MutexLock mu(*rejected_classes_lock_);
delete rejected_classes_;
rejected_classes_ = NULL;
}
delete rejected_classes_lock_;
rejected_classes_lock_ = NULL;
#if defined(ART_USE_LLVM_COMPILER) || defined(ART_USE_GREENLAND_COMPILER)
{
MutexLock mu(*inferred_reg_category_maps_lock_);
STLDeleteValues(inferred_reg_category_maps_);
delete inferred_reg_category_maps_;
inferred_reg_category_maps_ = NULL;
}
delete inferred_reg_category_maps_lock_;
inferred_reg_category_maps_lock_ = NULL;
#endif
}
void MethodVerifier::AddRejectedClass(Compiler::ClassReference ref) {
MutexLock mu(*rejected_classes_lock_);
rejected_classes_->insert(ref);
CHECK(IsClassRejected(ref));
}
bool MethodVerifier::IsClassRejected(Compiler::ClassReference ref) {
MutexLock mu(*rejected_classes_lock_);
return (rejected_classes_->find(ref) != rejected_classes_->end());
}
#if defined(ART_USE_LLVM_COMPILER) || defined(ART_USE_GREENLAND_COMPILER)
const InferredRegCategoryMap* MethodVerifier::GenerateInferredRegCategoryMap() {
uint32_t insns_size = code_item_->insns_size_in_code_units_;
uint16_t regs_size = code_item_->registers_size_;
UniquePtr<InferredRegCategoryMap> table(
new InferredRegCategoryMap(insns_size, regs_size));
for (size_t i = 0; i < insns_size; ++i) {
if (RegisterLine* line = reg_table_.GetLine(i)) {
const Instruction* inst = Instruction::At(code_item_->insns_ + i);
// GC points
if (inst->IsBranch() || inst->IsInvoke()) {
for (size_t r = 0; r < regs_size; ++r) {
const RegType &rt = line->GetRegisterType(r);
if (rt.IsNonZeroReferenceTypes()) {
table->SetRegCanBeObject(r);
}
}
}
/* We only use InferredRegCategoryMap in one case */
if (inst->IsBranch()) {
for (size_t r = 0; r < regs_size; ++r) {
const RegType &rt = line->GetRegisterType(r);
if (rt.IsZero()) {
table->SetRegCategory(i, r, kRegZero);
} else if (rt.IsCategory1Types()) {
table->SetRegCategory(i, r, kRegCat1nr);
} else if (rt.IsCategory2Types()) {
table->SetRegCategory(i, r, kRegCat2);
} else if (rt.IsReferenceTypes()) {
table->SetRegCategory(i, r, kRegObject);
} else {
table->SetRegCategory(i, r, kRegUnknown);
}
}
}
}
}
return table.release();
}
void MethodVerifier::SetInferredRegCategoryMap(Compiler::MethodReference ref,
const InferredRegCategoryMap& inferred_reg_category_map) {
MutexLock mu(*inferred_reg_category_maps_lock_);
const InferredRegCategoryMap* existing_inferred_reg_category_map = GetInferredRegCategoryMap(ref);
if (existing_inferred_reg_category_map == NULL) {
inferred_reg_category_maps_->Put(ref, &inferred_reg_category_map);
} else {
CHECK(*existing_inferred_reg_category_map == inferred_reg_category_map);
delete &inferred_reg_category_map;
}
CHECK(GetInferredRegCategoryMap(ref) != NULL);
}
const InferredRegCategoryMap*
MethodVerifier::GetInferredRegCategoryMap(Compiler::MethodReference ref) {
MutexLock mu(*inferred_reg_category_maps_lock_);
InferredRegCategoryMapTable::const_iterator it =
inferred_reg_category_maps_->find(ref);
if (it == inferred_reg_category_maps_->end()) {
return NULL;
}
CHECK(it->second != NULL);
return it->second;
}
#endif
} // namespace verifier
} // namespace art