[llvm] [NFC][LV]: Differentiate between Pred and Unpred masked operations. (PR #169509)

[David Sherwood via llvm-commits]

================
@@ -705,9 +707,15 @@ class LoopVectorizationLegality {
   AssumptionCache *AC;
 
   /// While vectorizing these instructions we have to generate a
-  /// call to the appropriate masked intrinsic or drop them in case of
-  /// conditional assumes.
-  SmallPtrSet<const Instruction *, 8> MaskedOp;
+  /// call to the appropriate masked intrinsic or drop them.
+  /// To differentiate between needing masked op because there is a conditional
----------------
david-arm wrote:

nit: Perhaps this might be a little bit clearer if we emphasise the difference between control flow present in the source code, and control flow introduced by the vectoriser? For example, something like

```
  /// In order to differentiate between control flow introduced at the source
  /// level and that introduced by the loop vectoriser when tail-folding we
  /// keep two lists:
```

What do you think?

https://github.com/llvm/llvm-project/pull/169509