[PATCH] D63731: [AMDGPU] Prevent VGPR copies from moving across the EXEC mask definitions

[Stanislav Mekhanoshin via Phabricator via llvm-commits]

rampitec added inline comments.


================
Comment at: lib/Target/AMDGPU/SIFoldOperands.cpp:504
+    MachineInstr::mop_iterator ImpOpE = UseMI->implicit_operands().end();
+    while (ImpOpI != ImpOpE) {
+      MachineInstr::mop_iterator Tmp = ImpOpI;
----------------
I believe the intent was to remove only duplicate operands, not all implicit operands at all.


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D63731/new/

https://reviews.llvm.org/D63731