[llvm] [TargetLowering] Lower ldexp into target supported instructions (PR #67552)

[Matt Arsenault via llvm-commits]

================
@@ -57389,3 +57396,40 @@ Align X86TargetLowering::getPrefLoopAlignment(MachineLoop *ML) const {
     return Align(1ULL << ExperimentalPrefInnermostLoopAlignment);
   return TargetLowering::getPrefLoopAlignment();
 }
+
+SDValue X86TargetLowering::LowerFLDEXP(SDValue Op, SelectionDAG &DAG) const {
+  SDValue X = Op.getOperand(0);
+  EVT XScalarTy = X.getValueType();
+  SDValue Exp = Op.getOperand(1);
+
+  SDLoc DL(Op);
+  EVT XVT, ExpVT;
+  SDValue IID;
+  switch (Op.getValueSizeInBits()) {
----------------
arsenm wrote:

Probably should switch over the actual type, or the scalar type 

https://github.com/llvm/llvm-project/pull/67552