[PATCH] D27057: AMDGPU: Fix adding extra implicit def of register

[Matt Arsenault via llvm-commits]

arsenm updated this revision to Diff 79135.
arsenm added a comment.

Also for readlane


https://reviews.llvm.org/D27057

Files:
  lib/Target/AMDGPU/SIRegisterInfo.cpp


Index: lib/Target/AMDGPU/SIRegisterInfo.cpp
===================================================================
--- lib/Target/AMDGPU/SIRegisterInfo.cpp
+++ lib/Target/AMDGPU/SIRegisterInfo.cpp
@@ -476,16 +476,18 @@
       = MF->getMachineMemOperand(PInfo, MMO->getFlags(),
                                  EltSize, MinAlign(Align, EltSize * i));
 
-    BuildMI(*MBB, MI, DL, Desc)
+    auto MIB = BuildMI(*MBB, MI, DL, Desc)
       .addReg(SubReg, getDefRegState(!IsStore))
       .addReg(ScratchRsrcReg)
       .addReg(SOffset, SOffsetRegState)
       .addImm(Offset)
       .addImm(0) // glc
       .addImm(0) // slc
       .addImm(0) // tfe
-      .addMemOperand(NewMMO)
-      .addReg(ValueReg, RegState::Implicit | SrcDstRegState);
+      .addMemOperand(NewMMO);
+
+    if (NumSubRegs > 1)
+      MIB.addReg(ValueReg, RegState::Implicit | SrcDstRegState);
   }
 
   if (RanOutOfSGPRs) {
@@ -660,26 +662,31 @@
           .addReg(MFI->getScratchWaveOffsetReg());
       }
 
-      BuildMI(*MBB, MI, DL, TII->get(AMDGPU::S_BUFFER_LOAD_DWORD_SGPR), SubReg)
+      auto MIB =
+        BuildMI(*MBB, MI, DL, TII->get(AMDGPU::S_BUFFER_LOAD_DWORD_SGPR), SubReg)
         .addReg(MFI->getScratchRSrcReg()) // sbase
         .addReg(OffsetReg)                // soff
         .addImm(0)                        // glc
-        .addMemOperand(MMO)
-        .addReg(MI->getOperand(0).getReg(), RegState::ImplicitDefine);
+        .addMemOperand(MMO);
+
+      if (NumSubRegs > 1)
+        MIB.addReg(MI->getOperand(0).getReg(), RegState::ImplicitDefine);
 
       continue;
     }
 
     SIMachineFunctionInfo::SpilledReg Spill
       = MFI->getSpilledReg(MF, Index, i);
 
     if (Spill.hasReg()) {
-      BuildMI(*MBB, MI, DL,
-              TII->getMCOpcodeFromPseudo(AMDGPU::V_READLANE_B32),
-              SubReg)
+      auto MIB =
+        BuildMI(*MBB, MI, DL, TII->getMCOpcodeFromPseudo(AMDGPU::V_READLANE_B32),
+                SubReg)
         .addReg(Spill.VGPR)
-        .addImm(Spill.Lane)
-        .addReg(MI->getOperand(0).getReg(), RegState::ImplicitDefine);
+        .addImm(Spill.Lane);
+
+      if (NumSubRegs > 1)
+        MIB.addReg(MI->getOperand(0).getReg(), RegState::ImplicitDefine);
     } else {
       // Restore SGPR from a stack slot.
       // FIXME: We should use S_LOAD_DWORD here for VI.
@@ -699,9 +706,13 @@
         .addReg(MFI->getScratchWaveOffsetReg()) // soffset
         .addImm(i * 4)                          // offset
         .addMemOperand(MMO);
-      BuildMI(*MBB, MI, DL, TII->get(AMDGPU::V_READFIRSTLANE_B32), SubReg)
-        .addReg(TmpReg, RegState::Kill)
-        .addReg(MI->getOperand(0).getReg(), RegState::ImplicitDefine);
+
+      auto MIB =
+        BuildMI(*MBB, MI, DL, TII->get(AMDGPU::V_READFIRSTLANE_B32), SubReg)
+        .addReg(TmpReg, RegState::Kill);
+
+      if (NumSubRegs > 1)
+        MIB.addReg(MI->getOperand(0).getReg(), RegState::ImplicitDefine);
     }
   }
 


-------------- next part --------------
A non-text attachment was scrubbed...
Name: D27057.79135.patch
Type: text/x-patch
Size: 2933 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20161123/e3bdd381/attachment.bin>