[PATCH] D143713: [ARM] Fix Chain/Glue Bug in PerformVMOVhrCombine

[Dave Green via Phabricator via llvm-commits]

dmgreen added a comment.

Sounds OK if you can clear up the tests. It might be worth making sure there is a test with multiple uses too.



================
Comment at: llvm/lib/Target/ARM/ARMISelLowering.cpp:15006
+  //     t2: f32,ch1,gl1? = CopyFromReg ch, Register:f32 %0, gl?
+  //   t5: i32 = bitcast t2 t18: f16 = ARMISD::VMOVhr t5
   // t18: f16 = ARMISD::VMOVhr t5
----------------
Can remove the extra `t18: f16 = ARMISD::VMOVhr t5`?


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D143713/new/

https://reviews.llvm.org/D143713