[llvm] [AMDGPU] Add support for llvm.lround and llvm.lrint intrinsics lowering. (PR #96817)

[Matt Arsenault via llvm-commits]

================
@@ -2493,6 +2502,14 @@ SDValue AMDGPUTargetLowering::LowerFRINT(SDValue Op, SelectionDAG &DAG) const {
   return DAG.getNode(ISD::FROUNDEVEN, SDLoc(Op), VT, Arg);
 }
 
+SDValue AMDGPUTargetLowering::LowerLRINT(SDValue Op, SelectionDAG &DAG) const {
+  auto ResVT = Op.getValueType();
+  auto Arg = Op.getOperand(0u);
+  auto ArgVT = Arg.getValueType();
+  SDValue RoundNode = DAG.getNode(ISD::FROUNDEVEN, SDLoc(Op), ArgVT, Arg);
+  return DAG.getNode(ISD::FP_TO_SINT, SDLoc(Op), ResVT, RoundNode);
----------------
arsenm wrote:

The lrint complexity seems related to FE_INEXACT, which we don't care about for the non-strictfp case 

https://github.com/llvm/llvm-project/pull/96817