Fix calculation of scalarization overhead in BasicTTIImpl::getCmpSelInstrCost

[jordy.potman.lists]

Summary

The arguments to the getScalarizationOverhead call in BasicTTIImpl::getCmpSelInstrCost are inconsistent with the comment above it. The comment seems to be correct so change the arguments to match the comment.

[jordy.potman.lists]

I noticed this for an out-of-tree target and I don't really know how I can easily add a test for this for an in-tree target.

include/llvm/CodeGen/BasicTTIImpl.h

if (ValTy->isVectorTy()) {
       unsigned Num = ValTy->getVectorNumElements();
       if (CondTy)
         CondTy = CondTy->getScalarType();
       unsigned Cost = static_cast<T *>(this)->getCmpSelInstrCost(
           Opcode, ValTy->getScalarType(), CondTy);
 
       // Return the cost of multiple scalar invocation plus the cost of
       // inserting and extracting the values.
-      return getScalarizationOverhead(ValTy, true, false) + Num * Cost;
+      return getScalarizationOverhead(ValTy, true, true) + Num * Cost;
     }
 
     // Unknown scalar opcode.
     return 1;
   }
 
   unsigned getVectorInstrCost(unsigned Opcode, Type *Val, unsigned Index) {
     std::pair<unsigned, MVT> LT =