[llvm] expandFMINIMUMNUM_FMAXIMUMNUM: Improve compare between zeros (PR #140193)

[Nikita Popov via llvm-commits]

================
@@ -8704,13 +8705,27 @@ SDValue TargetLowering::expandFMINIMUMNUM_FMAXIMUMNUM(SDNode *Node,
       DAG.getTargetConstant(IsMax ? fcPosZero : fcNegZero, DL, MVT::i32);
   SDValue IsZero = DAG.getSetCC(DL, CCVT, MinMax,
                                 DAG.getConstantFP(0.0, DL, VT), ISD::SETEQ);
-  SDValue LCmp = DAG.getSelect(
-      DL, VT, DAG.getNode(ISD::IS_FPCLASS, DL, CCVT, LHS, TestZero), LHS,
+  unsigned BitSize = VT.getScalarSizeInBits();
+  EVT IntVT = EVT::getIntegerVT(*DAG.getContext(), BitSize);
+  EVT FloatVT = EVT::getFloatingPointVT(32);
+  if (VT.isVector()) {
+    IntVT =
+        EVT::getVectorVT(*DAG.getContext(), IntVT, VT.getVectorElementCount());
+    FloatVT = EVT::getVectorVT(*DAG.getContext(), FloatVT,
+                               VT.getVectorElementCount());
+  }
+  SDValue LHSTrunc = LHS;
+  if (!isOperationLegal(ISD::BITCAST, IntVT) &&
+      !isOperationLegal(ISD::IS_FPCLASS, VT)) {
+    LHSTrunc = DAG.getNode(ISD::FP_ROUND, DL, FloatVT, LHS,
+                           DAG.getIntPtrConstant(0, DL, /*isTarget=*/true));
----------------
nikic wrote:

So this assumes that for all cases where the FP type is legal and the int type is not, that `float` is a legal type and that it either has legal IS_FPCLASS or i32 is also legal. Do these assumptions hold in practice? I think it's reasonable to expect float to be a legal type if there are any legal FP types, but I'm less sure about the rest. I think we have targets where f32 is legal but i32 is not.

Could you please add tests for https://github.com/llvm/llvm-project/issues/139380, https://github.com/llvm/llvm-project/issues/139381 and https://github.com/llvm/llvm-project/issues/140445 (just for minimumnum instead of minimum)? I *think* all of those are going to be fine with this patch, but I'd like to be sure.

https://github.com/llvm/llvm-project/pull/140193