[PATCH] D20546: AMDGPU: Handle flat in getMemOpBaseRegImmOfs

[Matt Arsenault via llvm-commits]

arsenm created this revision.
arsenm added a reviewer: tstellarAMD.
arsenm added a subscriber: llvm-commits.
Herald added subscribers: kzhuravl, arsenm.

It can still report the base register, and the uses give up when it fails

http://reviews.llvm.org/D20546

Files:
  lib/Target/AMDGPU/SIInstrInfo.cpp

Index: lib/Target/AMDGPU/SIInstrInfo.cpp
===================================================================
--- lib/Target/AMDGPU/SIInstrInfo.cpp
+++ lib/Target/AMDGPU/SIInstrInfo.cpp
@@ -287,6 +287,13 @@
     return true;
   }
 
+  if (isFLAT(*LdSt)) {
+    const MachineOperand *AddrReg = getNamedOperand(*LdSt, AMDGPU::OpName::addr);
+    BaseReg = AddrReg->getReg();
+    Offset = 0;
+    return true;
+  }
+
   return false;
 }
 


-------------- next part --------------
A non-text attachment was scrubbed...
Name: D20546.58176.patch
Type: text/x-patch
Size: 436 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20160524/bdef1988/attachment.bin>