[llvm] [LV]: Improve accuracy of calculating remaining iterations of MainLoopVF (PR #156723)

[David Sherwood via llvm-commits]

================
@@ -4371,11 +4371,30 @@ VectorizationFactor LoopVectorizationPlanner::selectEpilogueVectorizationFactor(
   const SCEV *TC =
       vputils::getSCEVExprForVPValue(getPlanFor(MainLoopVF).getTripCount(), SE);
   assert(!isa<SCEVCouldNotCompute>(TC) && "Trip count SCEV must be computable");
-  RemainingIterations =
-      SE.getURemExpr(TC, SE.getElementCount(TCType, MainLoopVF * IC));
+
+  // Calculate runtime estimated value for TC/MainLoopVF only when they are in
+  // different numerical space.
+  // TODO: This is a workaround until SCEV can evaluate vscale-based
+  // expressions.
+  if (match(TC, m_scev_Mul(m_SCEV(), m_SCEVVScale())) &&
+      MainLoopVF.isScalable()) {
----------------
david-arm wrote:

It might be a bit clearer if you handle all cases where the numerical space is the same here, i.e.

```
  bool ScalableTC = match(TC, m_scev_Mul(m_SCEV(), m_SCEVVScale()));
  if (ScalableTC == MainLoopVF.isScalable())
    RemainingIterations =
        SE.getURemExpr(TC, SE.getElementCount(TCType, MainLoopVF * IC));
  else {
    if (ScalableTC) {
      ...
```

https://github.com/llvm/llvm-project/pull/156723