[llvm] [X86] Respect denormal mode in f32-to-bf16 conversions (PR #221052)

[Simon Pilgrim via llvm-commits]

================
@@ -22831,6 +22831,57 @@ SDValue X86TargetLowering::LowerFP_EXTEND(SDValue Op, SelectionDAG &DAG) const {
   return DAG.getNode(X86ISD::VFPEXT, DL, VT, Res);
 }
 
+static bool hasCVTNEPS2BF16(const X86Subtarget &Subtarget) {
----------------
RKSimon wrote:

or at the very least merge this into canUseCVTNEPS2BF16

https://github.com/llvm/llvm-project/pull/221052