[llvm] [LoopVectorize] Enhance Vectorization decisions for predicate tail-folded loops with low trip counts (PR #69588)

[Florian Hahn via llvm-commits]

fhahn wrote:

> > It looks like test coverage for the new functionality is missing at the moment?
> 
> That's a question I would like to ask. Since this change doesn't add new functionality, just changes the decision made in the process, what would be a good test for it?

I might be missing something, but IIUC the patch may take different decisions in the presence of PGO data now, so there's some new behavior that should be testable?

https://github.com/llvm/llvm-project/pull/69588