[PATCH] D69275: Add constrained int->FP intrinsics

[Craig Topper via Phabricator via llvm-commits]

craig.topper added inline comments.


================
Comment at: llvm/lib/CodeGen/SelectionDAG/LegalizeVectorOps.cpp:1209
+        DAG.getNode(ISD::STRICT_SINT_TO_FP, DL, {Op.getValueType(), MVT::Other},
+                    {Src, HI});
+    fHI = DAG.getNode(ISD::STRICT_FMUL, DL, {Op.getValueType(), MVT::Other},
----------------
Src here should be Op.getOperand(0) I think. I think that's causing your MULPD crash.


================
Comment at: llvm/lib/CodeGen/SelectionDAG/LegalizeVectorOps.cpp:1362
+
   EVT VT = Op.getValueType();
   EVT EltVT = VT.getVectorElementType();
----------------
Can this code be replaced with DAG.UnrollVectorOp now? Or should we use this code for strict instead of DAG.UnrollVectorOp for ExpandUINT_TO_FLOAT?


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D69275/new/

https://reviews.llvm.org/D69275