[llvm] [LV] fix logical error in trunc cost (PR #91136)

[Florian Hahn via llvm-commits]

================
@@ -7123,16 +7117,15 @@ LoopVectorizationCostModel::getInstructionCost(Instruction *I, ElementCount VF,
       // "zext i8 %1 to i32" becomes "zext i8 %1 to i16".
       //
       // Calculate the modified src and dest types.
-      Type *MinVecTy = VectorTy;
       if (Opcode == Instruction::Trunc) {
-        SrcVecTy = smallestIntegerVectorType(SrcVecTy, MinVecTy);
+        SrcVecTy = smallestIntegerVectorType(SrcVecTy, VectorTy);
         VectorTy =
-            largestIntegerVectorType(ToVectorTy(I->getType(), VF), MinVecTy);
+            smallestIntegerVectorType(ToVectorTy(I->getType(), VF), VectorTy);
----------------
fhahn wrote:

Is this now a no-op as `VectorTy` should also use `I`'s return type?

Also looking at the code above, is it actually computing the narrowed types anywhere?

Is this missing something like the below?
```
RetTy = IntegerType::get(RetTy->getContext(), MinBWs[I]);
```

https://github.com/llvm/llvm-project/pull/91136