[PATCH] D68729: [AMDGPU] Fixed dpp combine of VOP1

[Stanislav Mekhanoshin via Phabricator via llvm-commits]

rampitec created this revision.
rampitec added reviewers: vpykhtin, kzhuravl, arsenm.
Herald added subscribers: MaskRay, kbarton, hiraditya, t-tye, tpr, dstuttard, yaxunl, nhaehnle, wdng, jvesely, nemanjai.
Herald added a project: LLVM.

If original instruction did not have source modifiers they were
not added to the new DPP instruction as well, even if needed.


https://reviews.llvm.org/D68729

Files:
  llvm/lib/Target/AMDGPU/GCNDPPCombine.cpp
  llvm/test/CodeGen/AMDGPU/dpp_combine.mir


Index: llvm/test/CodeGen/AMDGPU/dpp_combine.mir
===================================================================
--- llvm/test/CodeGen/AMDGPU/dpp_combine.mir
+++ llvm/test/CodeGen/AMDGPU/dpp_combine.mir
@@ -526,3 +526,14 @@
     %3:vreg_64 = REG_SEQUENCE %2, %subreg.sub0 ; %3.sub1 is undef
     %4:vgpr_32 = V_MOV_B32_dpp %3.sub1, %1, 1, 15, 15, 1, implicit $exec
     %5:vgpr_32 = V_ADD_U32_e32 %4, %0.sub1, implicit $exec
+...
+
+# CHECK-LABEL: name: dpp_vop1
+# CHECK: %3:vgpr_32 = V_CEIL_F32_dpp %1:vgpr_32, 0, undef %2:vgpr_32, 1, 15, 15, 1, implicit $exec
+name: dpp_vop1
+tracksRegLiveness: true
+body: |
+  bb.0:
+    %2:vgpr_32 = V_MOV_B32_dpp undef %1:vgpr_32, undef %0:vgpr_32, 1, 15, 15, 1, implicit $exec
+    %3:vgpr_32 = V_CEIL_F32_e32 %2, implicit $exec
+...
Index: llvm/lib/Target/AMDGPU/GCNDPPCombine.cpp
===================================================================
--- llvm/lib/Target/AMDGPU/GCNDPPCombine.cpp
+++ llvm/lib/Target/AMDGPU/GCNDPPCombine.cpp
@@ -195,6 +195,10 @@
       assert(0LL == (Mod0->getImm() & ~(SISrcMods::ABS | SISrcMods::NEG)));
       DPPInst.addImm(Mod0->getImm());
       ++NumOperands;
+    } else if (AMDGPU::getNamedOperandIdx(DPPOp,
+                   AMDGPU::OpName::src0_modifiers) != -1) {
+      DPPInst.addImm(0);
+      ++NumOperands;
     }
     auto *Src0 = TII->getNamedOperand(MovMI, AMDGPU::OpName::src0);
     assert(Src0);
@@ -214,6 +218,10 @@
       assert(0LL == (Mod1->getImm() & ~(SISrcMods::ABS | SISrcMods::NEG)));
       DPPInst.addImm(Mod1->getImm());
       ++NumOperands;
+    } else if (AMDGPU::getNamedOperandIdx(DPPOp,
+                   AMDGPU::OpName::src1_modifiers) != -1) {
+      DPPInst.addImm(0);
+      ++NumOperands;
     }
     if (auto *Src1 = TII->getNamedOperand(OrigMI, AMDGPU::OpName::src1)) {
       if (!TII->isOperandLegal(*DPPInst.getInstr(), NumOperands, Src1)) {


-------------- next part --------------
A non-text attachment was scrubbed...
Name: D68729.224158.patch
Type: text/x-patch
Size: 1874 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20191009/102f9ad3/attachment-0001.bin>