[PATCH] D48366: [X86] Adding a check against i64 inputs in combineScalarToVector

[Craig Topper via Phabricator via llvm-commits]

craig.topper added inline comments.


================
Comment at: llvm/lib/Target/X86/X86ISelLowering.cpp:39127
         if (Mask.getOpcode() == ISD::TRUNCATE &&
-            Mask.getOperand(0).getValueType() != MVT::i16)
+            Mask.getOperand(0).getValueType() != MVT::i16 &&
+            Mask.getOperand(0).getValueType() != MVT::i64)
----------------
You should probably just check for i8 and i32. I'm not sure anything prevents illegal types from getting here. Which might get legalized into i64 or i16 and be just as broken.


================
Comment at: llvm/lib/Target/X86/X86ISelLowering.cpp:39142
         if (Mask.getOpcode() == ISD::TRUNCATE &&
-            Mask.getOperand(0).getValueType() != MVT::i16)
+            Mask.getOperand(0).getValueType() != MVT::i16 &&
+            Mask.getOperand(0).getValueType() != MVT::i64)
----------------
Same here


Repository:
  rL LLVM

https://reviews.llvm.org/D48366