Bug in ExecutionEngine::getConstantValue() -- Instruction::PtrToInt

Project Infrastructure LLVM Dev List Archives
1

I think I might have found a bug in the exection engine’s ‘constants folding’.

Basically APInt’s parameters are the wrong way round.

Line 577 for lib/ExecutionEngine/ExecutionEngine.cpp :-

case Instruction::PtrToInt: {
GenericValue GV = getConstantValue(Op0);
uint32_t PtrWidth = TD->getPointerSizeInBits();
GV.IntVal = APInt(PtrWidth, uintptr_t(GV.PointerVal));
return GV;
}

Aaron

2

Aaron Gray wrote:

I think I might have found a bug in the exection engine’s ‘constants folding’.

Basically APInt’s parameters are the wrong way round.

Line 577 for lib/ExecutionEngine/ExecutionEngine.cpp :-

case Instruction::PtrToInt: {
GenericValue GV = getConstantValue(Op0);
uint32_t PtrWidth = TD->getPointerSizeInBits();
GV.IntVal = APInt(PtrWidth, uintptr_t(GV.PointerVal));
return GV;
}

Aaron

That looks to be the right order to me:

APInt(unsigned numBits, uint64_t val, bool isSigned=false)

John.

3

The code matches:

APInt(unsigned numBits, uint64_t val, bool isSigned = false)

which is correct.

-Chris

4

2009/7/25 John McCall <rjmccall@apple.com>

Aaron Gray wrote:

I think I might have found a bug in the exection engine’s ‘constants folding’.

Basically APInt’s parameters are the wrong way round.

Line 577 for lib/ExecutionEngine/ExecutionEngine.cpp :-

case Instruction::PtrToInt: {
GenericValue GV = getConstantValue(Op0);
uint32_t PtrWidth = TD->getPointerSizeInBits();
GV.IntVal = APInt(PtrWidth, uintptr_t(GV.PointerVal));
return GV;
}

Aaron

That looks to be the right order to me:

APInt(unsigned numBits, uint64_t val, bool isSigned=false)

Ah I was reading the private version !

Whoopse !

Aaron