[PATCH] D49507: [Power9] Add __float128 support in the backend for bitcast to a i128

[Nemanja Ivanovic via Phabricator via llvm-commits]

nemanjai added inline comments.


================
Comment at: lib/Target/PowerPC/PPCISelLowering.cpp:14314
+  const SDValue &Op0 = N->getOperand(0);
+  if (Op0.getValueType() == MVT::i128 && N->getValueType(0) == MVT::i64) {
+    bool BigEndian = DCI.DAG.getDataLayout().isBigEndian();
----------------
This logic is fairly deeply nested. Can you convert some of the conditions to early exits. This one certainly looks to fit the bill.


================
Comment at: lib/Target/PowerPC/PPCISelLowering.cpp:14331
+      if (Op0.getOperand(0).getValueType() == MVT::i128 &&
+          ConstNode && ConstNode->getZExtValue() == 64 &&
+          Op0.getOperand(0).getNode()->getOpcode() == ISD::BITCAST) {
----------------
For clarity, please do the null pointer check first.


https://reviews.llvm.org/D49507