[llvm] r312640 - [AMDGPU] Fix shouldClusterMemOps to process flat loads

[Stanislav Mekhanoshin via llvm-commits]

Author: rampitec
Date: Wed Sep  6 08:31:30 2017
New Revision: 312640

URL: http://llvm.org/viewvc/llvm-project?rev=312640&view=rev
Log:
[AMDGPU] Fix shouldClusterMemOps to process flat loads

Flat loads do not have vdata operand but have vdst instead.

Differential Revision: https://reviews.llvm.org/D37502

Added:
    llvm/trunk/test/CodeGen/AMDGPU/cluster-flat-loads.mir
Modified:
    llvm/trunk/lib/Target/AMDGPU/SIInstrInfo.cpp

Modified: llvm/trunk/lib/Target/AMDGPU/SIInstrInfo.cpp
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/Target/AMDGPU/SIInstrInfo.cpp?rev=312640&r1=312639&r2=312640&view=diff
==============================================================================
--- llvm/trunk/lib/Target/AMDGPU/SIInstrInfo.cpp (original)
+++ llvm/trunk/lib/Target/AMDGPU/SIInstrInfo.cpp Wed Sep  6 08:31:30 2017
@@ -366,7 +366,11 @@ bool SIInstrInfo::shouldClusterMemOps(Ma
       (isMTBUF(FirstLdSt) && isMTBUF(SecondLdSt)) ||
       (isFLAT(FirstLdSt) && isFLAT(SecondLdSt))) {
     FirstDst = getNamedOperand(FirstLdSt, AMDGPU::OpName::vdata);
+    if (!FirstDst)
+      FirstDst = getNamedOperand(FirstLdSt, AMDGPU::OpName::vdst);
     SecondDst = getNamedOperand(SecondLdSt, AMDGPU::OpName::vdata);
+    if (!SecondDst)
+      SecondDst = getNamedOperand(SecondLdSt, AMDGPU::OpName::vdst);
   } else if (isSMRD(FirstLdSt) && isSMRD(SecondLdSt)) {
     FirstDst = getNamedOperand(FirstLdSt, AMDGPU::OpName::sdst);
     SecondDst = getNamedOperand(SecondLdSt, AMDGPU::OpName::sdst);

Added: llvm/trunk/test/CodeGen/AMDGPU/cluster-flat-loads.mir
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/test/CodeGen/AMDGPU/cluster-flat-loads.mir?rev=312640&view=auto
==============================================================================
--- llvm/trunk/test/CodeGen/AMDGPU/cluster-flat-loads.mir (added)
+++ llvm/trunk/test/CodeGen/AMDGPU/cluster-flat-loads.mir Wed Sep  6 08:31:30 2017
@@ -0,0 +1,20 @@
+# RUN: llc -march=amdgcn -mcpu=gfx900 -run-pass machine-scheduler %s -o - | FileCheck -check-prefix=GCN %s
+
+# GCN-LABEL: name: cluster_flat_loads
+# GCN:      FLAT_LOAD_DWORD %0, 0
+# GCN-NEXT: FLAT_LOAD_DWORD %0, 4
+# GCN-NEXT: V_ADD_F32_e64
+name: cluster_flat_loads
+tracksRegLiveness: true
+registers:
+  - { id: 0, class: vreg_64 }
+  - { id: 1, class: vgpr_32 }
+  - { id: 2, class: vgpr_32 }
+  - { id: 3, class: vgpr_32 }
+body:             |
+  bb.0:
+    %0 = IMPLICIT_DEF
+    %1 = FLAT_LOAD_DWORD %0, 0, 0, 0, implicit %exec, implicit %flat_scr :: (load 4)
+    %2 = V_ADD_F32_e64 0, killed %1, 0, 1, 0, 0, implicit %exec
+    %3 = FLAT_LOAD_DWORD %0, 4, 0, 0, implicit %exec, implicit %flat_scr :: (load 4)
+...