[PATCH] D47383: [AMDGPU] Avoid using divergent value in mubuf addr64 descriptor

[Matt Arsenault via Phabricator via llvm-commits]

arsenm added a comment.

The test seems pretty lacking.  There are a lot more combinations of divergent and non-divergent changed here that aren't tested



================
Comment at: lib/Target/AMDGPU/AMDGPUISelDAGToDAG.cpp:1002
+  const SDValue Ops0[] = {
+    DAG.getTargetConstant(AMDGPU::SGPR_64RegClassID, DL, MVT::i32),
+    SDValue(DAG.getMachineNode(AMDGPU::S_MOV_B32, DL, MVT::i32, ValLo), 0),
----------------
SReg_64_XEXEC


Repository:
  rL LLVM

https://reviews.llvm.org/D47383