[llvm] AMDGPU: Add tonearest and towardzero roundings for intrinsic llvm.fptrunc.round (PR #104486)

[Matt Arsenault via llvm-commits]

================
@@ -6671,10 +6655,34 @@ SDValue SITargetLowering::getFPExtOrFPRound(SelectionDAG &DAG,
                 DAG.getTargetConstant(0, DL, MVT::i32));
 }
 
+SDValue SITargetLowering::lowerFPTRUNC_ROUND(SDValue Op,
+                                             SelectionDAG &DAG) const {
+  if (Op.getOperand(0)->getValueType(0) != MVT::f32)
+    return SDValue();
+
+  // Only support towardzero, tonearest, upward and downward.
+  int RoundMode = Op.getConstantOperandVal(1);
+  if (RoundMode < (int)RoundingMode::TowardZero ||
+      RoundMode > (int)RoundingMode::TowardNegative)
----------------
arsenm wrote:

Same, do exact equality checks for the supported modes 

https://github.com/llvm/llvm-project/pull/104486