[llvm] [LV] Vectorize FMax via OrderedFCmpSelect w/o fast-math flags. (PR #146711)

[via llvm-commits]

================
@@ -1227,6 +1236,7 @@ unsigned RecurrenceDescriptor::getOpcode(RecurKind Kind) {
   case RecurKind::UMin:
     return Instruction::ICmp;
   case RecurKind::FMax:
+  case RecurKind::OrderedFCmpSelect:
----------------
ayalz wrote:

Placed between FMax and FMin?

https://github.com/llvm/llvm-project/pull/146711