llvm/lib/Bytecode/Reader/InstructionReader.cpp
Chris Lattner 0b16ae209a - Rename Instruction::First*Op to *OpsBegin, and Num*Ops to *OpsEnd to
reflect the fact that it's a range being defined.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@4147 91177308-0d34-0410-b5e6-96231b3b80d8
2002-10-13 19:39:16 +00:00

472 lines
15 KiB
C++

//===- ReadInst.cpp - Code to read an instruction from bytecode -----------===//
//
// This file defines the mechanism to read an instruction from a bytecode
// stream.
//
// Note that this library should be as fast as possible, reentrant, and
// threadsafe!!
//
// TODO: Change from getValue(Raw.Arg1) etc, to getArg(Raw, 1)
// Make it check type, so that casts are checked.
//
//===----------------------------------------------------------------------===//
#include "ReaderInternals.h"
#include "llvm/iTerminators.h"
#include "llvm/iMemory.h"
#include "llvm/iPHINode.h"
#include "llvm/iOther.h"
using std::vector;
using std::cerr;
bool BytecodeParser::ParseRawInst(const uchar *&Buf, const uchar *EndBuf,
RawInst &Result) {
unsigned Op, Typ;
if (read(Buf, EndBuf, Op)) return true;
// bits Instruction format: Common to all formats
// --------------------------
// 01-00: Opcode type, fixed to 1.
// 07-02: Opcode
Result.NumOperands = (Op >> 0) & 03;
Result.Opcode = (Op >> 2) & 63;
switch (Result.NumOperands) {
case 1:
// bits Instruction format:
// --------------------------
// 19-08: Resulting type plane
// 31-20: Operand #1 (if set to (2^12-1), then zero operands)
//
Result.Ty = getType((Op >> 8) & 4095);
Result.Arg1 = (Op >> 20) & 4095;
if (Result.Arg1 == 4095) // Handle special encoding for 0 operands...
Result.NumOperands = 0;
break;
case 2:
// bits Instruction format:
// --------------------------
// 15-08: Resulting type plane
// 23-16: Operand #1
// 31-24: Operand #2
//
Result.Ty = getType((Op >> 8) & 255);
Result.Arg1 = (Op >> 16) & 255;
Result.Arg2 = (Op >> 24) & 255;
break;
case 3:
// bits Instruction format:
// --------------------------
// 13-08: Resulting type plane
// 19-14: Operand #1
// 25-20: Operand #2
// 31-26: Operand #3
//
Result.Ty = getType((Op >> 8) & 63);
Result.Arg1 = (Op >> 14) & 63;
Result.Arg2 = (Op >> 20) & 63;
Result.Arg3 = (Op >> 26) & 63;
break;
case 0:
Buf -= 4; // Hrm, try this again...
if (read_vbr(Buf, EndBuf, Result.Opcode)) return true;
Result.Opcode >>= 2;
if (read_vbr(Buf, EndBuf, Typ)) return true;
Result.Ty = getType(Typ);
if (Result.Ty == 0) return true;
if (read_vbr(Buf, EndBuf, Result.NumOperands)) return true;
switch (Result.NumOperands) {
case 0:
cerr << "Zero Arg instr found!\n";
return true; // This encoding is invalid!
case 1:
if (read_vbr(Buf, EndBuf, Result.Arg1)) return true;
break;
case 2:
if (read_vbr(Buf, EndBuf, Result.Arg1) ||
read_vbr(Buf, EndBuf, Result.Arg2)) return true;
break;
case 3:
if (read_vbr(Buf, EndBuf, Result.Arg1) ||
read_vbr(Buf, EndBuf, Result.Arg2) ||
read_vbr(Buf, EndBuf, Result.Arg3)) return true;
break;
default:
if (read_vbr(Buf, EndBuf, Result.Arg1) ||
read_vbr(Buf, EndBuf, Result.Arg2)) return true;
// Allocate a vector to hold arguments 3, 4, 5, 6 ...
Result.VarArgs = new vector<unsigned>(Result.NumOperands-2);
for (unsigned a = 0; a < Result.NumOperands-2; a++)
if (read_vbr(Buf, EndBuf, (*Result.VarArgs)[a])) return true;
break;
}
if (align32(Buf, EndBuf)) return true;
break;
}
#if 0
cerr << "NO: " << Result.NumOperands << " opcode: " << Result.Opcode
<< " Ty: " << Result.Ty->getDescription() << " arg1: " << Result.Arg1
<< " arg2: " << Result.Arg2 << " arg3: " << Result.Arg3 << "\n";
#endif
return false;
}
bool BytecodeParser::ParseInstruction(const uchar *&Buf, const uchar *EndBuf,
Instruction *&Res,
BasicBlock *BB /*HACK*/) {
RawInst Raw;
if (ParseRawInst(Buf, EndBuf, Raw))
return true;
if (Raw.Opcode >= Instruction::BinaryOpsBegin &&
Raw.Opcode < Instruction::BinaryOpsEnd && Raw.NumOperands == 2) {
Res = BinaryOperator::create((Instruction::BinaryOps)Raw.Opcode,
getValue(Raw.Ty, Raw.Arg1),
getValue(Raw.Ty, Raw.Arg2));
return false;
}
Value *V;
switch (Raw.Opcode) {
case Instruction::Cast: {
V = getValue(Raw.Ty, Raw.Arg1);
const Type *Ty = getType(Raw.Arg2);
if (V == 0 || Ty == 0) { cerr << "Invalid cast!\n"; return true; }
Res = new CastInst(V, Ty);
return false;
}
case Instruction::PHINode: {
PHINode *PN = new PHINode(Raw.Ty);
switch (Raw.NumOperands) {
case 0:
case 1:
case 3: cerr << "Invalid phi node encountered!\n";
delete PN;
return true;
case 2: PN->addIncoming(getValue(Raw.Ty, Raw.Arg1),
cast<BasicBlock>(getValue(Type::LabelTy,Raw.Arg2)));
break;
default:
PN->addIncoming(getValue(Raw.Ty, Raw.Arg1),
cast<BasicBlock>(getValue(Type::LabelTy, Raw.Arg2)));
if (Raw.VarArgs->size() & 1) {
cerr << "PHI Node with ODD number of arguments!\n";
delete PN;
return true;
} else {
vector<unsigned> &args = *Raw.VarArgs;
for (unsigned i = 0; i < args.size(); i+=2)
PN->addIncoming(getValue(Raw.Ty, args[i]),
cast<BasicBlock>(getValue(Type::LabelTy, args[i+1])));
}
delete Raw.VarArgs;
break;
}
Res = PN;
return false;
}
case Instruction::Shl:
case Instruction::Shr:
Res = new ShiftInst((Instruction::OtherOps)Raw.Opcode,
getValue(Raw.Ty, Raw.Arg1),
getValue(Type::UByteTy, Raw.Arg2));
return false;
case Instruction::Ret:
if (Raw.NumOperands == 0) {
Res = new ReturnInst(); return false;
} else if (Raw.NumOperands == 1) {
Res = new ReturnInst(getValue(Raw.Ty, Raw.Arg1)); return false;
}
break;
case Instruction::Br:
if (Raw.NumOperands == 1) {
Res = new BranchInst(cast<BasicBlock>(getValue(Type::LabelTy, Raw.Arg1)));
return false;
} else if (Raw.NumOperands == 3) {
Res = new BranchInst(cast<BasicBlock>(getValue(Type::LabelTy, Raw.Arg1)),
cast<BasicBlock>(getValue(Type::LabelTy, Raw.Arg2)),
getValue(Type::BoolTy , Raw.Arg3));
return false;
}
break;
case Instruction::Switch: {
SwitchInst *I =
new SwitchInst(getValue(Raw.Ty, Raw.Arg1),
cast<BasicBlock>(getValue(Type::LabelTy, Raw.Arg2)));
Res = I;
if (Raw.NumOperands < 3) return false; // No destinations? Wierd.
if (Raw.NumOperands == 3 || Raw.VarArgs->size() & 1) {
cerr << "Switch statement with odd number of arguments!\n";
delete I;
return true;
}
vector<unsigned> &args = *Raw.VarArgs;
for (unsigned i = 0; i < args.size(); i += 2)
I->dest_push_back(cast<Constant>(getValue(Raw.Ty, args[i])),
cast<BasicBlock>(getValue(Type::LabelTy, args[i+1])));
delete Raw.VarArgs;
return false;
}
case Instruction::Call: {
Value *M = getValue(Raw.Ty, Raw.Arg1);
if (M == 0) return true;
// Check to make sure we have a pointer to method type
const PointerType *PTy = dyn_cast<PointerType>(M->getType());
if (PTy == 0) return true;
const FunctionType *MTy = dyn_cast<FunctionType>(PTy->getElementType());
if (MTy == 0) return true;
vector<Value *> Params;
const FunctionType::ParamTypes &PL = MTy->getParamTypes();
if (!MTy->isVarArg()) {
FunctionType::ParamTypes::const_iterator It = PL.begin();
switch (Raw.NumOperands) {
case 0: cerr << "Invalid call instruction encountered!\n";
return true;
case 1: break;
case 2: Params.push_back(getValue(*It++, Raw.Arg2)); break;
case 3: Params.push_back(getValue(*It++, Raw.Arg2));
if (It == PL.end()) return true;
Params.push_back(getValue(*It++, Raw.Arg3)); break;
default:
Params.push_back(getValue(*It++, Raw.Arg2));
{
vector<unsigned> &args = *Raw.VarArgs;
for (unsigned i = 0; i < args.size(); i++) {
if (It == PL.end()) return true;
// TODO: Check getValue for null!
Params.push_back(getValue(*It++, args[i]));
}
}
delete Raw.VarArgs;
}
if (It != PL.end()) return true;
} else {
if (Raw.NumOperands > 2) {
vector<unsigned> &args = *Raw.VarArgs;
if (args.size() < 1) return true;
if ((args.size() & 1) != 0)
return true; // Must be pairs of type/value
for (unsigned i = 0; i < args.size(); i+=2) {
const Type *Ty = getType(args[i]);
if (Ty == 0)
return true;
Value *V = getValue(Ty, args[i+1]);
if (V == 0) return true;
Params.push_back(V);
}
delete Raw.VarArgs;
}
}
Res = new CallInst(M, Params);
return false;
}
case Instruction::Invoke: {
Value *M = getValue(Raw.Ty, Raw.Arg1);
if (M == 0) return true;
// Check to make sure we have a pointer to method type
const PointerType *PTy = dyn_cast<PointerType>(M->getType());
if (PTy == 0) return true;
const FunctionType *MTy = dyn_cast<FunctionType>(PTy->getElementType());
if (MTy == 0) return true;
vector<Value *> Params;
const FunctionType::ParamTypes &PL = MTy->getParamTypes();
vector<unsigned> &args = *Raw.VarArgs;
BasicBlock *Normal, *Except;
if (!MTy->isVarArg()) {
if (Raw.NumOperands < 3) return true;
Normal = cast<BasicBlock>(getValue(Type::LabelTy, Raw.Arg2));
Except = cast<BasicBlock>(getValue(Type::LabelTy, args[0]));
FunctionType::ParamTypes::const_iterator It = PL.begin();
for (unsigned i = 1; i < args.size(); i++) {
if (It == PL.end()) return true;
// TODO: Check getValue for null!
Params.push_back(getValue(*It++, args[i]));
}
if (It != PL.end()) return true;
} else {
if (args.size() < 4) return true;
Normal = cast<BasicBlock>(getValue(Type::LabelTy, args[0]));
Except = cast<BasicBlock>(getValue(Type::LabelTy, args[2]));
if ((args.size() & 1) != 0)
return true; // Must be pairs of type/value
for (unsigned i = 4; i < args.size(); i+=2) {
// TODO: Check getValue for null!
Params.push_back(getValue(getType(args[i]), args[i+1]));
}
}
delete Raw.VarArgs;
Res = new InvokeInst(M, Normal, Except, Params);
return false;
}
case Instruction::Malloc:
if (Raw.NumOperands > 2) return true;
V = Raw.NumOperands ? getValue(Type::UIntTy, Raw.Arg1) : 0;
if (const PointerType *PTy = dyn_cast<PointerType>(Raw.Ty))
Res = new MallocInst(PTy->getElementType(), V);
else
return true;
return false;
case Instruction::Alloca:
if (Raw.NumOperands > 2) return true;
V = Raw.NumOperands ? getValue(Type::UIntTy, Raw.Arg1) : 0;
if (const PointerType *PTy = dyn_cast<PointerType>(Raw.Ty))
Res = new AllocaInst(PTy->getElementType(), V);
else
return true;
return false;
case Instruction::Free:
V = getValue(Raw.Ty, Raw.Arg1);
if (!isa<PointerType>(V->getType())) return true;
Res = new FreeInst(V);
return false;
case Instruction::Load:
case Instruction::GetElementPtr: {
vector<Value*> Idx;
if (!isa<PointerType>(Raw.Ty)) return true;
const CompositeType *TopTy = dyn_cast<CompositeType>(Raw.Ty);
switch (Raw.NumOperands) {
case 0: cerr << "Invalid load encountered!\n"; return true;
case 1: break;
case 2:
if (!TopTy) return true;
Idx.push_back(V = getValue(TopTy->getIndexType(), Raw.Arg2));
if (!V) return true;
break;
case 3: {
if (!TopTy) return true;
Idx.push_back(V = getValue(TopTy->getIndexType(), Raw.Arg2));
if (!V) return true;
const Type *ETy = GetElementPtrInst::getIndexedType(TopTy, Idx, true);
const CompositeType *ElTy = dyn_cast_or_null<CompositeType>(ETy);
if (!ElTy) return true;
Idx.push_back(V = getValue(ElTy->getIndexType(), Raw.Arg3));
if (!V) return true;
break;
}
default:
if (!TopTy) return true;
Idx.push_back(V = getValue(TopTy->getIndexType(), Raw.Arg2));
if (!V) return true;
vector<unsigned> &args = *Raw.VarArgs;
for (unsigned i = 0, E = args.size(); i != E; ++i) {
const Type *ETy = GetElementPtrInst::getIndexedType(Raw.Ty, Idx, true);
const CompositeType *ElTy = dyn_cast_or_null<CompositeType>(ETy);
if (!ElTy) return true;
Idx.push_back(V = getValue(ElTy->getIndexType(), args[i]));
if (!V) return true;
}
delete Raw.VarArgs;
break;
}
if (Raw.Opcode == Instruction::Load) {
Value *Src = getValue(Raw.Ty, Raw.Arg1);
if (!Idx.empty()) {
cerr << "WARNING: Bytecode contains load instruction with indices. "
<< "Replacing with getelementptr/load pair\n";
assert(GetElementPtrInst::getIndexedType(Raw.Ty, Idx) &&
"Bad indices for Load!");
Src = new GetElementPtrInst(Src, Idx);
// FIXME: Remove this compatibility code and the BB parameter to this
// method.
BB->getInstList().push_back(cast<Instruction>(Src));
}
Res = new LoadInst(Src);
} else if (Raw.Opcode == Instruction::GetElementPtr)
Res = new GetElementPtrInst(getValue(Raw.Ty, Raw.Arg1), Idx);
else
abort();
return false;
}
case Instruction::Store: {
vector<Value*> Idx;
if (!isa<PointerType>(Raw.Ty)) return true;
const CompositeType *TopTy = dyn_cast<CompositeType>(Raw.Ty);
switch (Raw.NumOperands) {
case 0:
case 1: cerr << "Invalid store encountered!\n"; return true;
case 2: break;
case 3:
if (!TopTy) return true;
Idx.push_back(V = getValue(TopTy->getIndexType(), Raw.Arg3));
if (!V) return true;
break;
default:
vector<unsigned> &args = *Raw.VarArgs;
const CompositeType *ElTy = TopTy;
unsigned i, E;
for (i = 0, E = args.size(); ElTy && i != E; ++i) {
Idx.push_back(V = getValue(ElTy->getIndexType(), args[i]));
if (!V) return true;
const Type *ETy = GetElementPtrInst::getIndexedType(Raw.Ty, Idx, true);
ElTy = dyn_cast_or_null<CompositeType>(ETy);
}
if (i != E)
return true; // didn't use up all of the indices!
delete Raw.VarArgs;
break;
}
Value *Ptr = getValue(Raw.Ty, Raw.Arg2);
if (!Idx.empty()) {
cerr << "WARNING: Bytecode contains load instruction with indices. "
<< "Replacing with getelementptr/load pair\n";
const Type *ElType = GetElementPtrInst::getIndexedType(Raw.Ty, Idx);
if (ElType == 0) return true;
Ptr = new GetElementPtrInst(Ptr, Idx);
// FIXME: Remove this compatibility code and the BB parameter to this
// method.
BB->getInstList().push_back(cast<Instruction>(Ptr));
}
const Type *ValTy = cast<PointerType>(Ptr->getType())->getElementType();
Res = new StoreInst(getValue(ValTy, Raw.Arg1), Ptr);
return false;
}
} // end switch(Raw.Opcode)
cerr << "Unrecognized instruction! " << Raw.Opcode
<< " ADDR = 0x" << (void*)Buf << "\n";
return true;
}