[llvm] [LV] Prefer fixed VFs for epilogue vectorization (PR #211814)

[David Sherwood via llvm-commits]

================
@@ -706,6 +706,21 @@ bool LoopVectorizationPlanner::isMoreProfitable(const VectorizationFactor &A,
         !B.Width.isScalar())
       return true;
 
+  // Prefer fixed VFs for epilogue loops (unless the fixed cost is much more
+  // expensive -- >= 2x). There are some extra costs for using scalable vectors
----------------
david-arm wrote:

I'm not sure about adding a hidden, magic tuning factor here. At the very least, such a tuning factor needs modelling properly with a tunable cost (i.e. one that can be controlled via the command line).

https://github.com/llvm/llvm-project/pull/211814