[PATCH] Fix vselect when getSetCCResultType returns a different type from the operands

[Michael Liao]

test case?

- michael

On Fri, 2013-05-03 at 20:42 -0700, Matt Arsenault wrote:
> Hi nadav,
> 
> http://llvm-reviews.chandlerc.com/D743
> 
> Files:
>   lib/CodeGen/SelectionDAG/LegalizeVectorOps.cpp
> 
> Index: lib/CodeGen/SelectionDAG/LegalizeVectorOps.cpp
> ===================================================================
> --- lib/CodeGen/SelectionDAG/LegalizeVectorOps.cpp
> +++ lib/CodeGen/SelectionDAG/LegalizeVectorOps.cpp
> @@ -652,13 +652,14 @@
>  SDValue VectorLegalizer::ExpandVSELECT(SDValue Op) {
>    // Implement VSELECT in terms of XOR, AND, OR
>    // on platforms which do not support blend natively.
> -  EVT VT =  Op.getOperand(0).getValueType();
>    DebugLoc DL = Op.getDebugLoc();
>  
>    SDValue Mask = Op.getOperand(0);
>    SDValue Op1 = Op.getOperand(1);
>    SDValue Op2 = Op.getOperand(2);
>  
> +  EVT VT = Mask.getValueType();
> +
>    // If we can't even use the basic vector operations of
>    // AND,OR,XOR, we will have to scalarize the op.
>    // Notice that the operation may be 'promoted' which means that it is
> @@ -673,8 +674,22 @@
>        TargetLowering::ZeroOrNegativeOneBooleanContent)
>      return DAG.UnrollVectorOp(Op.getNode());
>  
> -  assert(VT.getSizeInBits() == Op1.getValueType().getSizeInBits()
> -         && "Invalid mask size");
> +  // If the mask and the type do not match and Select is legal,
> +  // go ahead and unroll the vector op.
> +  // This occurs when getSetCCResultType returns something that
> +  // is different than the operand types. For example,
> +  // v4i8 = select v4i32, v4i8, v4i8.
> +  if (Op1.getValueType() != VT &&
> +      TLI.getTypeAction(*DAG.getContext(), VT.getScalarType())
> +        == TargetLowering::TypeLegal &&
> +      TLI.getOperationAction(ISD::SELECT, VT.getScalarType())
> +        == TargetLowering::Legal) {
> +    return DAG.UnrollVectorOp(Op.getNode());
> +  }
> +
> +  assert(VT.getSizeInBits() == Op.getOperand(1).getValueType().getSizeInBits()
> +         && "Invalid mask size and ISD::SELECT is not Legal!");
> +
>    // Bitcast the operands to be the same type as the mask.
>    // This is needed when we select between FP types because
>    // the mask is a vector of integers.
> _______________________________________________
> llvm-commits mailing list
> llvm-commits at cs.uiuc.edu
> http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits